The Historical Linguistics of Uto-Aztecan Agriculture

William L. Merrill

Anthropological Linguistics, Volume 54, Number 3, Fall 2012, pp. 203-260

(Article)
Published by University of Nebraska Press
DOI: 10.1353/anl.2012.0017

For additional information about this article

http://muse.jhu.edu/journals/anl/summary/v054/54.3.merrill.html

Access provided by Bristol University (14 Aug 2013 20:55 GMT)

The Historical Linguistics of Uto-Aztecan Agriculture

WILLIAM L. MERRILL
Smithsonian Institution
Abstract. The Uto-Aztecan language family figures prominently in research
on early agriculture in western North America. A central issue is the role that
the members of the Proto-Uto-Aztecan speech community might have played
in the diffusion of maize agriculture from Mesoamerica to the southwestern
United States. Key to addressing this issue is determining whether an agricultural lexicon can be reconstructed for Proto-Uto-Aztecan, but despite several
comparative studies of the agricultural lexica of the Uto-Aztecan languages,
consensus remains elusive. A detailed reanalysis of these lexica indicates
that an agriculture-related vocabulary can be reconstructed only for Proto
Southern Uto-Aztecan, supporting the conclusion that maize agriculture
entered the Uto-Aztecan world after the division of the Proto-Uto-Aztecan
speech community into southern and northern branches. Additional lexical
and biogeographical data suggest that the ProtoSouthern Uto-Aztecan
speech community was located near the modern Arizona-Sonora border when
its members began cultivating maize, a development that may have occurred
around four thousand years ago, when the earliest evidence of maize agriculture appears in the archaeological record of the North American Southwest.

1. Introduction.
The transition from food collecting to food production
began on the North American continent some ten thousand years ago with the
domestication of the pepo squash (Cucurbita pepo), followed at about fourthousand-year intervals first by the domestication of maize (Zea mays) and
then the common bean (Phaseolus vulgaris) (Smith 1997a, 2001a; Kaplan and
Lynch 1999; Piperno 2011; Brown 2006, 2010a). The earliest evidence for the
cultivation of these plants comes from archeological sites in southern and central Mexico. Data from sites in northeastern Mexico and the southwestern
United States indicate that the northward diffusion of these tropical cultigens
took place separately and gradually over the course of several millennia. Pepo
squash is first documented in northeastern Mexico, just south of the Tropic of
Cancer, around 6300 BP, with maize appearing around 4400 BP and the common bean around 1300 BP (Smith 1997b:37374; Kaplan and Lynch 1999:
269).1 The earliest archaeological records of these domesticated plants north of
the Tropic of Cancer come from Arizona and New Mexico. Multiple radiocarbon
dates on samples of maize indicate that it was present in the American Southwest at least by 4100 BP, while pepo squash arrived at approximately 3150 BP
and common beans around 2300 BP (Merrill et al. 2009: table S3).
The archaeological sites that document the inception of agriculture in the
southwestern United States are located more than a thousand kilometers
203

204

ANTHROPOLOGICAL LINGUISTICS

54 NO. 3

north of the Tropic of Cancer and about double that distance from the early
agricultural sites in southern and central Mexico. The routes and timing of the
diffusion of domesticated plants across the intervening area remain a mystery
because no archaeological sites from the period when this diffusion would have
occurred have been excavated there. However, a number of scholars have speculated that the ancestors of speakers of Uto-Aztecan languages may have
been involved in the process, based primarily on the fact that at European
contact these languages were spoken from Mesoamerica to what is today the
western United States (Matson 1991:31920; Bellwood 1993, 2001; Fowler
1994:453; Hill 2001a, 2001b, 2002a; Carpenter, Snchez de Carpenter, and
Mabry 2001; Carpenter, Snchez, and Villalpando 2002, 2005; Diamond and
Bellwood 2003; Bellwood and Oxenham 2008; LeBlanc 2008; Mabry,
Carpenter, and Sanchez 2008; Wilcox et al. 2008) (see map 1).

Map 1. The distribution of the Uto-Aztecan subfamilies at initial European contacts.

2012

WILLIAM L. MERRILL

205

Evaluating the role that Uto-Aztecans might have played in the diffusion
of agriculture from Mesoamerica to the Southwest requires as an initial step
establishing the place of agriculture in Uto-Aztecan cultural history. Several
scholars have addressed the issue of whether members of the Proto-UtoAztecan (PUA) speech community were farmers by applying the methods of
historical linguistics to an analysis of the agriculture-related vocabularies
documented for the Uto-Aztecan languages. The most significant studies are
by Romney (1957), Miller (1966), Fowler (1994), and Hill (2001b), but no consensus has been reached: Romney and Hill concluded that an agricultural lexicon could be reconstructed for PUA, while Miller and Fowler concluded that it
could not.2
The majority of recent research on the cultural history of Uto-Aztecan agriculture has been undertaken by Hill, who presented her results in a series of
studies published from 2001 on (see References). Basing her interpretations on
a broader set of terms than had been considered previously, Hill identified a
subset of these terms, all with maize-related meanings in some Uto-Aztecan
languages, as reflexes of PUA etyma and concluded that it is highly likely
that maize cultivation was present in the PUA community (2001b:922). She
(2001b:91617, 2012:65) further proposed that this ancestral community was
located in the northwest quadrant of Mesomerica when its members first
adopted agriculture. Building upon Bellwoods (1997, 2001) perspectives linking the dispersal of Uto-Aztecan languages to the expansion of early agricultural populations, Hill (2001b:913, 2002b, 2006) argued that the northward
migration of Uto-Aztecan farmers was responsible for both the introduction of
maize agriculture to the southwestern United States and the formation of a
chain of Uto-Aztecan dialects and languages that extended between Mesoamerica and the American Southwest.3
Despite the insights that Hills contributions provide, the debate continues on whether the members of the PUA speech community were farmers.
Campbell (2002) and Kaufman and Justeson (2009) have questioned Hills
reconstructions of some etyma in her proposed PUA maize vocabulary, as well
as her identification of agriculture-related meanings as their original referents. Colleagues and I (Merrill et al. 2009, 2010; cf. Hill 2010 and Brown
2010b) have reiterated these concerns and have challenged her postulation of
migrating farmers, Uto-Aztecan or not, as the mechanism for the diffusion of
maize from Mesoamerica to the Southwest.
The purpose of this article is to present the results of my analysis of the
historical relationships among the Uto-Aztecan words that have been considered in previous studies and to offer my perspectives on the place of agriculture in Uto-Aztecan cultural history. In the next section, I provide a brief
overview of the Uto-Aztecan language family and the distribution of farming
and foraging strategies among speakers of its languages. In section 3, I argue
that an agricultural lexicon definitely can be reconstructed for ProtoSouthern

206

ANTHROPOLOGICAL LINGUISTICS

54 NO. 3

Uto-Aztecan (PSUA), discussing in separate subsections each of the eight

etyma that constitute this lexicon, as well as a few other PSUA etyma whose
reflexes have maize-related referents in at least some Southern Uto-Aztecan
(SUA) languages. Although maize is the only cultigen for which a PSUA etymon is reconstructible, SUA terms for cucurbits and the common bean and the
insights that they offer into the development of the agricultural lexica of the
SUA languages are considered in sections 4 and 5.
In section 6, I compare the various agriculture-related lexica recorded for
the Northern Uto-Aztecan (NUA) languages, interpreting their diversity as an
indication of the absence of a ProtoNorthern Uto-Aztecan (PNUA) agricultural lexicon. In section 7, I focus on one NUA language, Hopi. Several words in
the Hopi agricultural lexicon clearly are cognate with words in the SUA languages, but cognates are not attested in any other NUA language. I consider
both linguistic and historical evidence to evaluate the hypothesis that some
ancestors of the modern Hopi originated in Mexico and spoke a language or
languages affiliated with the southern branch of the language family.
The data and analyses presented in first seven sections of the essay support the conclusion that members of the ProtoSouthern Uto-Aztecan speech
community were the first Uto-Aztecan farmers. I suggest in section 8 that
when they initially adopted maize agriculture, PSUA speakers were living in
the modern Arizona-Sonora borderlands, near where some of the oldest maize
in the American Southwest has been recovered, and that the dispersal of the
SUA languages probably began from there, making this area a likely candidate
for the SUA homeland. I conclude the essay by offering four generalizations
about the place of farming in Uto-Aztecan cultural history and by proposing
that some of the speakers of NUA languages may have shifted between
foraging and mixed foraging-farming strategies at different points in their
histories.4
2. The Uto-Aztecan language family. The Uto-Aztecan language family
comprises thirty languages organized into two major branches, Northern UtoAztecan and Southern Uto-Aztecan (see table 1 and map 1) (Miller 1983b;
Campbell 1997:13338; Caballero 2011; Merrill forthcoming).5 The NUA subfamiliesNumic, Tubatulabal, Takic, and Hopiare all located in the western
United States. The northernmost of the SUA subfamilies, Tepiman, is found in
both the southwestern United States and Mexico. The Taracahitan subfamily
is situated in northwestern Mexico, the Corachol subfamily in western Mexico,
and the Aztecan subfamily in western, central, and southern Mexico, with outliers in Central America. The Tubar language, spoken in northwestern Mexico
until the early twentieth century, is treated here as the sole member of a fifth
subfamily of the SUA branch, but it is poorly documented and its classification
is, in Stubbss words, enigmatic (Stubbs 2003:6; cf. Stubbs 2000, Hill 2011).

Table 1. The Uto-Aztecan Language Family

NORTHERN UTO-AZTECAN
Numic
Western Numic
Northern Paiute
Mono
Central Numic
Timbisha Shoshone
Shoshone
Comanche
Southern Numic
Kawaiisu
Colorado River Numic (Southern Paiute, Chemehuevi, Ute)
Tubatulabal
Hopi
Takic
Cupan
Cahuilla
Cupeo
Luiseo
Gabrielino-Fernandeo
Serrano
Kitanemuk
Serrano
SOUTHERN UTO-AZTECAN
Tepiman
Upper Pima (Tohono Oodham, Akimel Oodham, and other dialects)
Lower Pima (Nvome, Yepachi Pima, and other dialects)
Northern Tepehuan
Southern Tepehuan
Taracahitan
Cahitan
Yaqui-Mayo
patan
Eudeve
pata
Tarahumaran
Rarmuri
Warih
Tubar
Corachol
Cora
Huichol
Aztecan
Pochutec
General Aztecan
Nahuatl
Pipil

208

ANTHROPOLOGICAL LINGUISTICS

54 NO. 3

At the time of initial European contacts, Uto-Aztecan societies varied dramatically in subsistence strategies, settlement patterns, and levels of social
complexity.6 At one extreme were the small-scale, highly mobile egalitarian
bands of Numic speakers who relied exclusively on the wild resources of the
Great Basin for their survival. At the other were the urbanized state societies
of Aztecan speakers in Mesoamerica, who practiced various forms of intensive
agriculture that supported populations estimated to have numbered in the
millions. In between were foraging and mixed foraging-farming societies that
included both sedentary foragers and mobile agriculturalists and ranged in
population size from hundreds to hundreds of thousands of people. The distribution of these diverse foraging and farming strategies tended to coincide with
the two principal branches of the language family: all of the Southern UtoAztecans were farmers while the majority of the Northern Uto-Aztecans were
foragers.
3. The Southern Uto-Aztecan agricultural lexicon. The integration of
farming into preexisting foraging economies obviously involves the creation
or borrowing of terminology for the plants that are cultivated, the practices
that are associated with their cultivation and processing, and related items
of material culture. There is no evidence that the agricultural lexicon of any
Southern Uto-Aztecan language was borrowed entirely from another language,
Uto-Aztecan or otherwise. Instead, the lexicon of each language combines reflexes of Proto-Uto-Aztecan and ProtoSouthern Uto-Aztecan etyma with loanwords from other Uto-Aztecan languages or external sources, as well as innovations attested only in that language.
Table 2. The ProtoSouthern Uto-Aztecan Agricultural Lexicon
PSUA ETYMA
1. *suhunu maize (generic)
2. *sita immature maize ear
3. *hora ~ *hori to shell maize
4. *saki parched maize kernels
5. *tma tamale
6. *ca to plant, to sow
7. *wasa field for cultivation
8. *wika planting stick

TEP

TRC

CRC

AZT

SET

+
+
+
+
+
+

+
+
+
+
+
+
+
+

+
+
+
+
+
+
+

+
+
+
+
+

#17
#15
#3
#14

#4
#21
#22

NOTE:

TEP = Tepiman; TRC = Taracahitan; CRC = Corachol; AZT = Aztecan; + =

cognate present; = cognate absent.

An agricultural lexicon comprising eight etyma can be reconstructed for

ProtoSouthern Uto-Aztecan. These terms appear in table 2, which also indicates the presence or absence of reflexes in four of the five SUA subfamilies
and the numbers of the associated cognate sets in appendix 2 (for *tma, see
section 3.5). The Tubar subfamily is not included in the table because Tubar

2012

WILLIAM L. MERRILL

209

words were recorded for only three of the referents (maize, parched maize
kernels, to plant, to sow) and these words are not cognate with the reflexes of
the PSUA etyma having these referents in the other SUA languages.
The first etymon in table 2 is a generic label for maize. The next designates one stage in the development of the maize ear. It is followed by three
etyma related to the processing of maize for consumption. The final three
etyma are associated with agriculture in general rather than with maize cultivation specifically. Each of these etyma is discussed in a separate subsection
(3.13.8). The following five subsections (3.93.13) are devoted to PSUA etyma
whose reconstructed referents extend beyond the domain of agriculture but
have reflexes that are associated with maize. The final subsection (3.14) explores the possibility of reconstructing PSUA *kopi as an etymon that originally labeled a wild plant but whose reflexes acquired maize-related meanings in
some SUA languages.
3.1. *suhunu maize.
The generic label for maize in PSUA usually is
reconstructed as *sunu (Miller 1966:96; Fowler 1994:44953; Hill 2004:6568),
but the reconstruction of *suhunu is indicated by the identical vowel sequences
in the initial syllables of the Rarmuri and Tepiman cognates, the occurrence
of high tone on both vowels in this sequence in the Northern Tepehuan cognate, and the medial u in the River Warih cognate, as shown in (1).
(1) RR suun
UP hu
LP hun
NT nui
ST hun
WR-R suun

No Corachol or Tubar reflexes of *suhunu are reported in the available

sources. The generic terms for maize in these languages are Cora yuuri,
Huichol ik, and Tubar kot. Given that Huichol u reflects *o, the Huichol and
Tubar terms could be related, with metathesis having occurred in one language or the other (see section 3.13). A reflex of *suhunu also is lacking in
Yaqui but is attested in Mayo snnu, the meaning of which has shifted to
maize field. The generic terms for maize in both Yaqui and Mayo reflect
PSUA *paci seed (see section 3.10).
Similar words exist in two NUA languages, Hopi and the Gosiute dialect of
Western Shoshone, but they appear to reflect *suu rather than *suhuu
(NUA regularly corresponds with SUA n).7 The Hopi term is soow (PUA
**u > Hopi o), which labels the giant sandreed (Calamovilfa gigantea), a tall
wild grass whose reedlike stems are used by the Hopis as a raw material
(Whiting 1966:65). The third syllable, w, probably derives from the PUA augmentative suffix **w, suggesting that the Hopi word should be glossed as big

210

ANTHROPOLOGICAL LINGUISTICS

54 NO. 3

soo or tall soo. However, it is impossible to identify what this soo might
have been because soo, without the suffix, is not attested in the extensive
literature on the Hopi language and ethnobotany (Hopi Dictionary Project
1998; Whiting 1966).
The Gosiute term su ~ suno was recorded by the early ethnobotanist
Ralph Chamberlin (1911:5253) as alternate forms of the label for Atriplex
confertifolia, commonly known as shadscale and spiny saltbrush. Linguistic
research indicates that PUA ** is reflected in Gosiute and other Western
Shoshone dialects as the consonant cluster [g] and the geminate [nn],
which are in free variation, and further that ** is the only PUA consonant
to have such alternating reflexes (Miller 1972:16; Merrill forthcoming).8 Presumably Chamberlin simply failed to note the phonetic details, but his recording of both and n in the Gosiute word confirms that it is cognate with the
Hopi term.
Chamberlin reported that this and other species of Atriplex were one
of the most important sources of seed food (1911:52) for the Gosiute, and
Steward (1938:22) documented the dietary significance of the seeds of another
saltbrush species, Atriplex argentea, among other Western Shoshone bands
located in north central Nevada. Steward recorded sunu, suuna, and snuu
as the labels for this plant in different bands and commented that Atriplex
argentea probably was the species of Atriplex that was frequently sown broadcast by their members. Although [g] ~ [nn] does not appear in any of
Stewards terms, the phonological similarities and shared referent of Atriplex
suggest that they are cognate with Gosiute su ~ suno.9
Because different plants are labeled by the Hopi and Numic cognates, the
original referent of PNUA *suu cannot be determined. As a member of the
grass family, the giant sandreed more closely resembles maize than saltbrush,
a member of the chenopod family, but both saltbrush and maize are valued
sources of seed food. In either case, the similarity of PNUA *suu and PSUA
*suhunu raises the possibility that PSUA speakers adopted the name for a
wild plant as their term for maize (Campbell 2002:5253; Hill 2004).
3.2. *sita immature maize ear.
The reconstruction of PSUA *sita
immature maize ear is based on cognates in the Taracahitan and Corachol
subfamilies, for example, Rarmuri sit and Huichol ~ita. No term for
immature maize ear was recorded for Tubar, and only one Tepiman term that
specifically designates the maize ear in its early stages of development is
attested: the Nvome word tutunopa tender maize ear before it forms kernels,
which is not a reflex of *sita and contrasts with tunibo mature fresh maize
ear.
I suspect that *sita may also have existed in Proto-Aztecan, but the
evidence is equivocal. The Proto-General Aztecan term for immature maize
ear can be reconstructed as *iiloo, the first syllable of which is the expected

2012

WILLIAM L. MERRILL

211

reflex of PSUA *si. However, the second syllable *loo cannot derive from
*ta. The loo element is attested in the Aztecan languages in a large number
of terms for plants and animals, including two others associated with the
maize ear specifically, *eeloo mature fresh maize ear and *ooloo maize
cob.10 Dakin (2001a:10711) suggests that *loo is attached to morphemes
that designate features possessed by the entity being labeled and thus is a
derivational suffix conveying the sense of possession of feature.
3.3. *hora ~ *hori to shell maize. Reflexes of the etymon *hora ~ *hori
to shell maize are attested in the Taracahitan, Corachol, and Aztecan subfamilies. An initial *h is reconstructed based on Eudeve horan. Reflexes of
PSUA *h were lost in the ancestral languages of the Tepiman, Corachol, and
Aztecan subfamilies while ancestral Taracahitan and Tubar apparently retained *h. Although h often disappears in Rarmuri and Sierra Warih and
sometimes in the other Taracahitan languages, it also is encountered as a
regular correspondence in these languages, as well as Tubar.
The original referent of PSUA *hora ~ *hori probably was to shell maize,
which in some SUA languages was extended to include the shelling of the seeds
of other plants. For example, the cognate in Classical Nahuatl, ooya, is glossed
to shell something (corn, peas, etc.), but the term for shelled and dried maize
kernels is tlaoolli. The first syllable, tla, is an indefinite object prefix that can
be translated as something. The second syllable is the verb stem oo to
shell, which is followed by the nominalizing suffix l and the absolutive suffix
li (Campbell and Langacker 1978:#33). The word means literally something
that is shelled, but that something in this case is always maize.
3.4. *saki parched maize kernels. PSUA *saki parched maize kernels is
reflected in words in languages belonging to all SUA subfamilies except Tubar,
where the equivalent concept is labeled with kumalt, probably derived from
the verb kumi to eat small or ground up things (see section 3.13). That *saki
also was a deverbal noun is suggested by the Cora reflex ari toasted
maize, derived from the verb ae to toast maize by the addition of the
nominalizing suffix ri.11 However, in some other SUA languages, the nouns
and verbs are homophones or differ only in showing final nominal or verbal
markers, as in (2).
(2) RR sak to parch maize kernels; sak parched maize kernels
ED sakn to toast maize; sakt toasted maize

The nominal and verbal reflexes of *saki in all the SUA languages are
invariably associated with maize, but in the Tepiman subfamily they also are
linked to the parching of other grains and seeds. For the Akimel Oodham, Rea
notes, Many kinds of seeds were prepared by being parched with live coals

212

ANTHROPOLOGICAL LINGUISTICS

54 NO. 3

(haak), then ground into flour (chui) (1997:69), mentioning that amaranths,
chia (Salvia columbariae), and wheat following its introduction by Europeans
were among the plants whose seeds were processed in this fashion.
Evidence from the NUA languages Tubatulabal and Luiseo indicate that
**saki can be reconstructed for PUA, forming part of a suite of verbs in both
NUA and SUA languages that begin with *sa and denote boiling, melting, and
parching (Stubbs 2011:#266, #267, #524). The Tubatulabal verb aag(t) ~
aak to roast it may have designated to roast in general, but it clearly
applied to the roasting of wild seeds. Erminie Voegelin recorded the deverbal
noun aagit as the term for parching tray, used for parching small seeds,
such as chia, with live embers (1938:31). An association of reflexes of *saki
with tray parching also is seen in the Tohono Oodham verb haak ~ haaki ~
hahak to roast grain with coals in a basket.
The Luiseo cognate is ~axi grain, wheat.12 The final syllable i is a
deverbalizer, and sax to toast is anticipated but not attested as the source
verb. Harrington collected ~a~a as the Luiseo verb to toast, but this word
was not encountered in subsequent research by Elliott (1999:830, 1015), who
recorded only walki ~ wlki to toast seeds, wheat.
3.5. *tma tamale.
The consumption of maize in a form resembling a
tamale by members of the PSUA speech community is suggested by the fact
that reflexes of PSUA *tma tamale are attested in all SUA subfamilies
except Tubar, for which no term for tamale was recorded.13 It is likely that
*tma is a deverbal noun, deriving originally from a verb that denoted a particular way of preparing maize, probably by roasting or baking small cakes of
ground maize under ashes or in pit ovens. In most cases, terms for pit-roasting
in Tepiman, Taracahitan, and Corachol languages closely resemble the reflexes of *tma. These verbs are listed in (3), along with the reflexes of *tma if they
are attested.
(3) TO uama to roast in ashes; mait a tortilla
NV tuamaha to pit-roast things other than agave, like squash or pumpkins
(calabazas); tuamahi something pit-roasted; tumaita cake (?) (attested in
vivac tumaita, glossed as pan de piciete, which perhaps can be translated as
tobacco cake)
ED temson pit-roast; cf. Op temi to make bread or tortillas.
WR-S wehtemena to pit-roast food; temi tortilla
CR tima to pit-roast; tema tamal

Assuming that these verbs are cognate, evidence from Nvome suggests
that the PSUA form of which they are reflexes was composed of two morphemes: NV maha to pit-roast agave contrasts with tuamaha to pit-roast
things other than agave. The glottal stop in presumed cognates from the three
NUA languages in (4) also may indicate that two separate morphemes were
involved.14

2012

WILLIAM L. MERRILL

213

(4) KW tma ~ tuma to roast, bake

SP tma to roast under ashes
KT t to roast; tac roasting pit

However, the significance of the initial t and the comparable elements in the
SUA verbs (ua, tua, and te) is unknown.
Although some of the reflexes of PSUA *tma label tortilla or both
tamale and tortilla, tamale is assumed to be its original referent because
this sense is encountered in all languages in the Aztecan and Corachol subfamilies and also in Rarmuri in the Taracahitan subfamily.15 In addition,
in SUA languages where distinct terms for tamale and tortilla exist and
neither is clearly a loanword, such as those in (5), the terms for tamale consistently derive from *tma, while those for tortilla lack cognates in the other
SUA languages.
(5) CR tema tamale; hamuii tortilla
HC tem tamale of beans and salt; paap tortilla
NA-CL tamalli tamale; tlakalli tortilla16

The Nahuatl term for tortilla, tlakalli or in some dialects takalli, does
appear in Tepiman and Taracahitan languages, such as those in (6), but it
presumably was borrowed during the Spanish colonial period directly from
Nahuatl speakers who were involved in the colonization of the northern portions of New Spain (West 1949:4952; Griffen 1969:134; Cramaussel 1998:24
25, 33).
(6) PYP taskori; tmit tortilla; nohica tamale
NT(R) taskali
YQ tahkaim; nhim tamale
MY tahkari; nhhim tamales
ED taskari
WR-R takari ~ tahkari
TBR tasekalit ~ tasikalit

It seems that the Nahuatl loanwords replaced the reflexes of *tma in all of
these languages except the Yepachi dialect of Lower Pima, in which terms
derived from both sources are attested, taskori and tmit, both glossed as
tortilla. Also, excluding modern loans of Spanish tamal, terms for tamale
are not reported for any of these languages except Yepachi Pima, Yaqui, and
Mayo. These words clearly do not reflect PSUA *tma, but they may derive
from a distinct PUA verb meaning to roast, to bake that is attested only in
NUA languages, for example, Northern Paiute noho/17 to prepare in earthen
oven on ashes, to roast, to bake (for additional NUA cognates, see Stubbs
2011:#523).

214

ANTHROPOLOGICAL LINGUISTICS

54 NO. 3

3.6. *ca to plant, to sow. Reflexes of the verb *ca to plant, to sow are
attested in the Tepiman, Taracahitan, and Corachol subfamilies. The expected
reflex in Tubar is eca or ica but sa is attested instead. Perhaps the Tubar form
is the result of interaction with Tepiman speakers. The shift of PSUA *c to *s
occurred in Proto-Tepiman (Bascom 1965:13) and is seen in the Tepiman reflexes of *ca, for example, Yepachi Pima sa. Reflexes of *ca are absent in the
Aztecan languages, where the verb to plant, reconstructed for Proto-Aztecan
as **tooka (Dakin 1982:#288), also means to bury. That the semantic scope of
*tooka was expanded from to bury to include to plant is suggested by the
Rarmuri cognate t, which designates to bury only.
PSUA *ca is identical to the form of this verb that can be reconstructed for
Proto-Uto-Aztecan. PUA *ca shifted to *ya in ProtoNorthern Uto-Aztecan as
part of general shift of medial *c to *y (Manaster Ramer 1992). Reflexes of
PNUA *ya are encountered in Hopi and most Numic languages but are absent
in Tubatulabal and the Takic languages (see section 6.4). PUA **ca probably
was originally associated with the broadcast sowing of wild seeds rather than
the cultivation of domesticated crops (Fowler 1972a:221).18 This interpretation
is supported by the fact that in Northern Paiute and Western Shoshone, nouns
derived from **ca label various species of Chenopodium whose seeds were
valued as food and broadcast sown (Chamberlin 1911:55; Steward 1938:23;
Liljeblad, Fowler, and Powell 2012:8485) (see appendix 2, set 4).
3.7. *wasa field for cultivation. Cognates in the Tepiman, Taracahitan,
and Corachol subfamiliesfor example, those in (7)indicate the reconstruction of *wasa as the PSUA label for field for cultivation.
(7) PYP gaha (from *wasa)
RR was
HC waa

The g and h in the Yepachi Pima cognate are the expected reflexes of PSUA *w
and *s. No Tubar word for this referent was recorded. The Aztecan languages
lack cognates, labeling this referent with terms derived from the ProtoAztecan innovation *miil (Campbell and Langacker 1978:#36). Similar NUA
words with field for cultivation as their referent are Hopi paasa and Chemehuevi pasa, but SUA w : NUA p is not a regular correspondence (see section
6.4).
3.8. *wika planting stick.
The final PSUA etymon in table 2 is *wika
planting stick. No term for this concept is reported for Tubar, but reflexes are
attested in the other four SUA subfamilies. Although most indicate that *wika
should be reconstructed for PSUA, there are anomalies in the Taracahitan
reflexes in (8).19

2012

WILLIAM L. MERRILL

215

(8) MY wika
YQ-AZ wiiki
WR-S wka
RR(H) wka

Medial glottal stops in Yaqui or Mayo terms that reflect PSUA etyma usually
are attested in the Sierra Warih cognates, and the initial stress reported for
Sierra Warih and one dialect of Rarmuri is unexpected. A possible explanation is that the initial syllable of the antecedent form of these cognates was
reduplicated as *wiwka. The medial glottal stop in Mayo and Yaqui could indicate the loss of medial w and the Warih and Rarmuri cognates could result
from the loss of the initial syllable and the retention of the antecedent stress
placement.
The most intriguing aspect of this cognate set is that Hopi wiika, which
labels ancient wooden hoe rather than planting stick, clearly is a reflex of
PSUA *wika. Hopi is the only NUA language in which a cognate for the SUA
reflexes of PSUA *wika is attested, and Hopi wiika does not appear to be a
loan from any of the Tepiman languages, the SUA languages located in closest
proximity to northeastern Arizona where the modern Hopi live. Although PUA
**wika could be reconstructed based on the Hopi and SUA cognates, a consideration of both linguistic and historical evidence raises the alternative possibility, which I explore in section 7, that some ancestors of the modern Hopi
were speakers of a SUA language or languages.
3.9. *murayawa inflorescence.
Maize tassel is the referent of the
reflexes of the PSUA etymon *murayawa in all of the SUA subfamilies except
Tubar, for which a reflex was not recorded. However, most designate the
flowering or fruiting heads of other plants as well, suggesting that the PSUA
etymon originally labeled the inflorescences of grasses and other kinds of wild
plants that did not resemble blossoms and then was extended to the maize
tassel following the introduction of this cultigen.
The reconstruction of the PSUA etymon as a polysyllabic is based on the
reflexes documented for two Tepiman languages, Upper Pima mud, aag and
Nvome muzadaga. These words show the shift of *y to *d and *w to *g that
occurred in Proto-Tepiman and the shift of Proto-Tepiman *r and *d to Upper
Pima d, and (Bascom 1965). Phonological and morphological changes in the
reflexes of this etymon also occurred in the ancestral languages of the other
SUA subfamilies. The final two syllables were lost in the Proto-Taracahitan
reflex *mura, while Proto-Corachol *may shows the loss of the final syllable
and medial *r, as well as the shift of *u to Proto-Corachol * (the final * is
unexpected). The interpretation of the changes that resulted in the ProtoAztecan reflex *miyawa is complicated by the fact that PSUA *r sometimes
but not always is replaced by Proto-Aztecan *y and the reflexes of both PSUA
*r and *y can also be lost.20

216

ANTHROPOLOGICAL LINGUISTICS

54 NO. 3

3.10. *paci seed. All SUA subfamilies have reflexes of the PSUA etymon
*paci seed except Tepiman, where the word for seed is reconstructed for
Proto-Tepiman as *kai (Bascom 1965:#93). However, in most Taracahitan
languages, reflexes of *paci or words derived from these reflexes have acquired
associations with maize or squash, as in (9).21
(9) YQ bi maize
MY bti maize
ED bacit squash seed
OP(P) vai maize with formed kernels
WR-S ihpac mature fresh maize ear22
RR pa mature fresh maize ear; ba squash; bara squash seed

Rarmuri bara squash seed could be a reflex of PSUA *pacira, which, as

shown in (10), is attested as the antecedent form for the words for seed in
River Warih, Yaqui, Mayo, and Tubar.
(10) WR-R pahcra
YQ bia
MY bia
TBR wacirn

The function of the final syllable ra, reduced to a in Yaqui and Mayo through
the common r-deletion process, is unknown but presumably it is a suffix.
Suffixes with the form ra have a variety of grammatical functions in all four
languages, which are not entirely understood and in fact differ among these
languages (Miller 1996:24959; Dedrick and Casad 1999:12425, 136; Lionnet
1978:32). Given that cognates are attested only in Tubar and some Taracahitan languages, *pacira may not be a PSUA etymon at all, but rather an
innovation in one of these languages that diffused to the other three.
Hill (2001b:920, 2012:58) proposes that the Tubatulabal and Hopi words
presented in (11) are cognate with the reflexes of PSUA *paci.
(11) TB pacaah ~ apacaah to hull; pacaahil hulled pine nuts (glosses are Hills)
HP paacama hominy

She (2012:58) reconstructs **paci ~ **paca as the antecedent PUA etymon,

to which she assigns the gloss ear of corn, kernel of corn, but her analysis can
be questioned on both phonological and semantic grounds. The reconstruction
of the medial glottal stop is required to account for the retention of medial
c in the NUA words because *c should otherwise have shifted to *y in
PNUA (see section 3.6), but it is not attested in any of the SUA reflexes of
*paci, including Sierra Warih, which tends to retain preconsonantal glottal
stops in reflexes of PUA etyma. The lengthened vowel in the first syllable of
Hopi paacama also is problematical. 23 In reflexes of PUA etyma, Mayo

2012

WILLIAM L. MERRILL

217

normally has a long vowel in initial syllables where Hopi does, but a lengthened vowel is absent in both Mayo bti maize and Mayo bia seed. In addition, the most likely referent of PSUA *paci is seed, with the maize-related
referents that are associated exclusively with Taracahitan reflexes being secondary developments.
3.11. *sona body, stalk. Reflexes of this PSUA etymon are attested in the
Tepiman, Taracahitan, and Tubar subfamilies but absent in the Corachol and
Aztecan subfamilies.
The Proto-Tepiman reflex is *hona, showing the expected shift of PSUA *s
to PTEP *h. As illustrated in (12), all Tepiman reflexes of PTEP *hona include
body among their referents.
(12) TO hon the body (excluding the head)
PYP hona the body, stalk, trunk of a plant
NT(R) honna the body

Reflexes of PTEP *hona also serve as the stems of terms for rib(s) in
several Tepiman languages. As seen in (13), the same morphology is encountered in Yaqui and Mayo words for ribs, but the Yaqui-Mayo reflex of PSUA
*sona is sna, which shows vowel harmonization.
(13) PYP honamar
TO hoonma (from hohonma)
YQ snaim
MY snaarim

An association between reflexes of *sona and the maize plant, presented in

(14), is encountered only in the Taracahitan languages and perhaps Tubar.
The Tubar and Mayo cognates are glossed only as caa, which can be translated as cane or stalk; no word for maize stalk specifically is attested in
either language. Vowel harmonization also has occurred in the Tubar, Eudeve,
Sierra Warih, and Rarmuri reflexes, but in this case from *sona to sono.
(14) YQ snaba corn husk
YQ-AZ sana sugarcane24
MY snaba corn husk; snna cane or stalk
ED son corn husk or maize leaf (the gloss in the original source is hoja de
maz)
RR son maize stalk
WR-S son corn stubble
TBR sono cane or stalk25

Hill (2012:58) identifies Hopi s corncob as cognate with the SUA

reflexes of *sona, on the basis of which she reconstructs PUA **sono parts of
the maize plant not eaten by human beings. However, the medial *n in

218

ANTHROPOLOGICAL LINGUISTICS

54 NO. 3

PSUA *sona and the medial in Hopi so corncob both reflect PUA
**, not **n (Kaufman and Justeson 2009:225; Merrill forthcoming). The
only possible NUA cognate that I have encountered for Hopi s is Luiseo
~ea bedrock. All segments of these words except the final vowels are regular
correspondences and, if the words are cognate, they reflect PNUA *sooa.26
It is unlikely that PSUA *sona reflects PUA **sooa. The expected Mayo
reflex of PUA **sooa is sona or, with vowel harmonization, sana, but snna
is attested instead. The geminate nn indicates the absence of vowel length in
the first syllable of the antecedent PSUA etymon.
If such irregular correspondences are ignored and PUA **soa or **sooa
is reconstructed, then vowel lengthening in PNUA or vowel shortening in
PSUA must have occurred. If either secondary development took place, the
divergent referents of the Hopi and Luiseo words, as well as the diversity of
referents of the SUA reflexes of PSUA *sona, suggest that the original referent
of the PUA etymon was something on the order of foundation or supporting
structure.
The Tepiman referents of body and rib(s) are consistent with these concepts, as are the referents of stalk, maize stalk, cane, and stubble in the
other SUA languages. Only the Yaqui-Mayo referents of corn husk and the
Eudeve referent corn husk or maize leaf seem out of place. The Luiseo referent bedrock also fits with the concept of foundation, and the Hopi referent
corncob is understandable in light of the association of maize cob with stalk
or trunk in other SUA and NUA languages, such as those in (15).
(15) NV vaoka maize cob, maize stalk
CM haniwoora maize cob

The Comanche word combines hani maize with woora, which is identical to
Timbisha Shoshone [woora] tree trunk and, except for the absence of the
glottal stop, Goshiute [woora] tree trunk, waist. These cognates, from the
three subdivisions of Central Numic, indicate that *woota tree trunk can be
reconstructed for ProtoCentral Numic (Stubbs 2011:#2157).27
Hill (2012:58) offers a different analysis of woora in Comanche hani
woora. She regards it as a reflex of a PUA etymon that she reconstructs as
*ora ~ *ori, to which she assigns the referent ear of corn, corn cob. In addition to the Comanche word, she lists six words, three from NUA languages and
three from SUA languages, as definite cognates that support this reconstruction. These words are presented in (16) with the referents reported in the
original sources.
(16) KW onoci hooked stick used to pull down pinyon cones
TSH onnocci pine cone hook
HP qaa maize, dry husked ear of maize
WR-S won maize cob
RR oon ~ koon maize cob; RR(H) kon maize cob
NA-CL oolootl maize cob

2012

WILLIAM L. MERRILL

219

These words do not constitute a valid cognate set. The Kawaiisu and Timbisha words are cognate with one another but with none of the other terms.
The Hopi word is a reflex of PNUA *kaao pine cone (see section 6.1). Dakin
(1982:#60, #229a) interprets the Nahuatl word as deriving from PUA **oho ~
**o bone. The analysis of the Warih and Rarmuri words is a bit more complicated, but, as seen in (17), the River Warih cognate, also with the referent
maize cob, shows features that allow the reconstruction of their antecedent
form as *ona or *oona.
(17) WR-R hon maize cob
WR-S won maize cob
RR oon ~ koon maize cob; RR(H) kon maize cob

The three words in (17) obviously are cognate, but the River Warih
cognate indicates that the initial consonants in the Sierra Warih and Rarmuri cognates are epenthetic. A consonant, usually k or w, often occurs in
Rarmuri words that have initial vowels in their first syllables, and alternate
forms like oon ~ koon are common. Also, if the antecedent form had initial
*w, w would be expected in the River Warih cognate. PSUA word-initial *w
is reflected in both Sierra and River Warih as w, which is not lost in their
reflexes of PSUA or PUA etyma with word-initial *w (Merrill 2007). The
initial h in River Warih hon is the reflex of PSUA word-initial *, which
is reconstructed before first-syllable vowels.28
The reconstruction of *ona or *oona reflects the fact that either form
is possible because River Warih sometimes inserts V to create V1V1
sequences.29 As seen in table 3, such insertion has occurred in its word for
salt, hoon, which reflects PSUA *oona.
Table 3. Words for Salt, Maize Cob, and Bone
salt

maize cob

bone

PSUA
River Warih
Sierra Warih
Rarmuri
Eudeve
Sonoran Yaqui
Mayo
Timbisha Shoshone

*oona
hoon
won
on ~ kon
ont
ona
ona
oapi

hon
won
oon ~ koon
nhro
nao
nago

Kawaiisu

owavi

*ho
howa
o
o
hga
ta
tta
cuhmipph ~
cuhnipph
ohov

Also seen in table 3 is initial h in the Eudeve and River Warih words for
bone but not the Eudeve word for salt. These correspondences indicate that
initial *h should not be reconstructed in the etymon reflected in the Warih
and Rarmuri words for maize cob, and thus eliminate PSUA *ho bone as

220

ANTHROPOLOGICAL LINGUISTICS

54 NO. 3

their antecedent form. Further, these Warih and Rarmuri words cannot be
cognate with the Timbisha and Kawaiisu words for pine cone hook because
n in Warih and Rarmuri regularly corresponds with Timbisha and
Kawaiisu w following reflexes of PUA **o. This correspondence is documented in the words for salt in these languages, as seen in table 3.
3.12. *tusi something ground up.
The PSUA noun *tusi something
ground up derives from the verb *tusa to grind. Both the nominal and verbal
etyma are reflected in all five SUA subfamilies, and cognates in all NUA subfamilies indicate that PUA **tusi and **tusa should be reconstructed (appendix 2, sets 19 and 20).
As seen in (18), the deverbal nouns in a few SUA languages are associated
primarily if not exclusively with maize.
(18) ED tust ground parched maize,
WR-R tus ~ tuus ground parched maize, maize dough
PP titi dough, corn dough

In most SUA languages, however, the nominal forms have the general sense of
something that is ground up, with more specific, maize-related senses created
by prefixing a morpheme linked to maize. These morphemes vary considerably,
as illustrated by the four examples in (19), all glossed in the sources as pinole
(ground parched maize).30
(19) MY sk tssi
TBR matust
CR(V) matsi
RR kobrusi

Mayo sk comes from saki parched maize kernels, with sk tssi literally
meaning ground parched maize kernels. Tubar ma and Cora ma may
represent the initial syllables of their terms for metate, which reflect PUA
**mata (appendix 2, set 9). The source of Rarmuri kob is discussed in
section 3.14.
3.13. *kumi to nibble, to chew. Reflexes of this PSUA verb are encountered in all SUA subfamilies except Aztecan.31 The glosses assigned to these
reflexes suggest that the PSUA etymon designated a form of eating that involved nibbling or chewing foods that were small in size and hard or crunchy
(see appendix 2, set 8). Examples given of these foods include fresh, parched, or
dried maize kernels and other grains, as well as squash seeds, hard fruits, and
pieces of candy.
Although *kumi probably did not refer to the consumption of maize exclusively, a specific association with maize is found in nouns derived from reflexes
of *kumi in the Tubar and Tepiman subfamilies. Lionnet (1978:59) identifies

2012

WILLIAM L. MERRILL

221

Tubar kumalt parched maize kernels as deriving from the verb kumi to eat
small or ground up things.32 In the majority of Tepiman languages, a related
deverbal noun labels maize cob. In Tohono Oodham, the form of this noun is
kuumikud, , which Mathiot (1973, 2:15) glosses as something on which one
chews. She indicates that corn cob (without kernels) is created by the addition of the term for maize (huukuumikud, ), which corresponds to the form
recorded for Akimel Oodham by Rea (1997:352). Other sources on the Tepiman languages report that maize cob is labeled by kuumikud, and related
forms alone, without the maize term (Saxton, Saxton, and Enos 1983:35;
Valias Coalla 2000:198).
The verb *kumi to nibble, to chew appears to be a PSUA innovation. I do
not reconstruct **kumi as a PUA etymon because verbs cognate with the reflexes of PSUA *kumi are not attested in any of the NUA languages. However,
several scholars have noted that some NUA societies cultivated a variety of
maize labeled with terms that closely resemble *kumi (Voegelin, Voegelin, and
Hale 1962:#88; Fowler 1994:454 n. 5; Hill 2001b:918, 922, 2012:58) (see section
6.1)
Several additional maize-related words in SUA languages include ku in
their stems. This element represents the second syllable of the Corachol words
for mature fresh maize ear: Cora ikri and Huichol hiikri (Corachol is the
reflex of PSUA *u). In Warih and Rarmuri, similar terms label roasted corn
on the cob: ihkusri in Sierra Warih, kusri and ksari in Rarmuri. The
Rarmuri nouns derive from the verb ksa to pit-roast, mainly corn on the
cob, which contrasts with the verb mih, used primarily in reference to the pitroasting of agave (Stubbs 2011:#4, #324). The Warih cognate for the latter is
mahin to bury, to cook something in the ground, but a distinct verb denoting the pit-roasting of maize ears is not reported for Warih.
The fact that languages belonging to two SUA subfamilies have maizerelated words that share the element ku raises the possibility that this element existed with the same association in PSUA. Given its presence in NUA
words for both wild and cultivated plants that yield edible seeds, it can be speculated that **ku may have been a PUA stem linked to the concept of valued
seed plant (see section 6.1).
3.14. *kopi.
The only maize-related word in a SUA language that could
reflect PSUA *kopi is the Rarmuri term for ground parched maize kernels.
This word was recorded in the eighteenth century as kuvrusi (Steffel 1809:
356) and in the twentieth as kobrusi ~ kobsi (Brambila 1976:257). It can be
analyzed as combining kobi parched maize and rusi something that is
ground up, the latter reflecting PSUA *tusi (see section 3.12).
The element kobi is not attested in other Rarmuri words, but the second
syllable of Huichol ik maize reflects an antecedent *ko and ko is the first
syllable of Tubar kot maize, maize kernel. An apparent cognate is Akimel

222

ANTHROPOLOGICAL LINGUISTICS

54 NO. 3

Oodham kovi, which probably labeled the domesticated Chenopodium

berlandieri spp. nuttalliae, cultivated in the southwestern United States by
around 1000 BP but no longer an Akimel Oodham crop (Rea 1997:29798;
Gasser and Kwiatkowski 1991).
NUA cognates may exist in the names for various wild species of
Chenopodium valued as sources of edible seeds and greens, for example,
Southern Paiute kov, identified as the label for C. fremontii (see appendix 2,
set 5). The NUA words, attested in both Numic and Takic languages, support
reconstructing PNUA *ko with Chenopodium as its referent. If Akimel
Oodham kovi is not a loanword, PUA **ko perhaps can also be reconstructed
with the same referent, which later acquired maize-related meanings in some
of the SUA languages.
4. Squash, pumpkins, and gourds in Southern Uto-Aztecan. Generic
terms for domesticated squash or domesticated gourd cannot be reconstructed for PSUA, but PSUA *hari wild squash probably can be reconstructed.
Identifying the cognates that support this reconstruction is difficult because of
the permutations that the reflexes of this etymon underwent during the diversification of the PSUA languages and the derivation of new terms from these
reflexes to provide labels for domesticated squash, pumpkins, and gourds and
for utensils made from them. Further complicating the analysis is the fact that
the reflexes of PSUA initial *h are lost in most SUA languages and the
reflexes of PSUA medial *r are either lost or undergo sound shifts in some.
To illustrate these phonological and semantic changes, I provide in (20)
(22) a sample of the reflexes of PSUA *hari, organized into three groups by
phonological similarity. The complete set of cognates is found in appendix 2
(set 1).
(20) *hari > haari ~ ari ~ ara ~ azabi
WR-R haari gourd canteen, gourd dipper
RR ari gourd, bottle gourd
PYP(R) ara wild squash
NV azabi wild squash
(21) *hari > halipa ~ alapa ~ laba
TBR halipat a kind of gourd, gourd dipper
WR-S alapa gourd, gourd dipper
RR laba gourd dipper cut breadthwise
(22) *hari > *hariw > harawe ~ ayaawi ~ ayo
WR-R harawe squash or pumpkin
YQ ayaawim squash, pumpkin (Cucurbita moschata)
NA-CL ayotli squash, pumpkin, gourd

2012

WILLIAM L. MERRILL

223

The *w that appears in (22) is interpreted as reflecting the PUA augmentative suffix **w, suggesting that the term for domesticated cucurbit
may have literally meant big wild squash (see section 6.2). This set shows
three phonological changes that occur in both Yaqui-Mayo and the Aztecan
languages: optional r-deletion, optional y-insertion to separate the resulting
vowel cluster, and h-deletion, which is optional in Yaqui-Mayo, but a sound
change that occurred in Proto-Aztecan (Dakin 1982:6567). The w suffix is
absent in the Nahuatl cognate, but its former presence is indicated by the shift
in the sound of the preceding vowel (presumably *a) to o and the retention
of the glottal stop at the morpheme boundary (Kaufman 1981:22526; Dakin
2001a:108).
The Tepiman, Taracahitan, Tubar, and Aztecan subfamilies are represented in (20)(22). I have encountered only one possible Corachol reflex of
*hari, Huichol ari, identified by Grimes (1980:272) as the name for a domesticated squash variety (Cucurbita pepo var. ovifera) and by Kindl (2000:37) as
the name for the domesticated bottle gourd (Lagenaria siceraria). Huichol ari
is cognate with Tubar huali a kind of gourd used as a canteen, which show
regular correspondences in all segments, including the loss in Huichol of the
initial h. The first syllable, reflecting *hu, suggests that the original etymon
was a compound, but it is unexplained.
No generic term for domesticated squash can be reconstructed for the
Taracahitan or Aztecan subfamilies. The generic term for domesticated
squash in Corachol is *suci, presumably a loan from an Aztecan language. It
shows regular correspondences with the terms for flower in the Aztecan languages, reconstructed for Proto-Aztecan as *ooi (Campbell and Langacker
1978:#63). Tubar vipt ~ wipt is glossed in the source only as calabaza
(squash or pumpkin) and could have labeled a specific variety of calabaza
rather than serving as a general label.
Bascom (1965:#311) reconstructs *imai as the Proto-Tepiman generic for
squash based on cognates in Northern and Southern Tepehuan and Lower
Pima. The Upper Pimans have a different term, haal, which probably derives
from PSUA *hari but entered Upper Piman as a loan from another SUA language, like Eudeve or River Warih, that usually retained h as the reflex of
initial *h (see section 3.3). The identification of haal as a loanword is based
on the fact that initial h in the Tepiman languages is the reflex of PSUA *s,
not *h, which shifted to the glottal stop in Proto-Tepiman (Bascom 1965:13).
If this interpretation is correct, Upper Piman retained a term derived from
PSUA *hari to label wild squash while borrowing another word derived from
the same PSUA etymon as the generic for domesticated squash and
pumpkin.
5. The common bean in Southern Uto-Aztecan languages. The third
major domesticate cultivated by the members of SUA societies is the common

224

ANTHROPOLOGICAL LINGUISTICS

54 NO. 3

bean (Phaseolus vulgaris). The diversity of generic terms for bean(s) indicates
that this cultigen entered the agricultural complexes of the SUA societies after
the emergence of the five subfamilies.
The Taracahitan labels for different varieties of the common bean all
reflect *muni (see appendix 2, set 11). This etymon can be identified as a loanword from a non-Uto-Aztecan source that presumably was integrated into the
Taracahitan agricultural lexicon while the Proto-Taracahitan speech community was still intact. Similar terms for beans are attested in several different
languages families in North America, as well as in the NUA languages Hopi
and Colorado River Numic (Wolff 1950:175; Hill 2001b:92324; Rankin 2006:
57172). The Colorado River Numic terms likely derive from a term having the
form muri, which in Hopi has become mori (see section 6.3). These terms do not
display the regular correspondence of NUA * with SUA *n and thus cannot be reflexes of a PUA etymon.
In the Corachol languages, the labels for bean are Cora muhume and
Huichol muume. Although these words and Taracahitan *muni share the initial syllable mu, separate introductions of the common bean to Corachol and
Taracahitan speakers are likely. These terms cannot be reflexes of the same
PSUA etymon because Corachol u regularly corresponds with Taracahitan o.
This correspondence reflects a sequence of sound changes in which a shift of
PSUA *u to Proto-Corachol * was followed by a shift of PSUA *o to ProtoCorachol *u. These sound changes presumably had already taken place before
Corachol speakers acquired the common bean.
The Proto-Tepiman generic label for bean(s) appears to have been *bavi,
which may have originally labeled the tepary bean (Phaseolus acutifolius)
(Nabhan and Felger 1978; Rea 1997:32125; Muoz et al. 2006). Reflexes of
*bavi are attested in all four Tepiman divisions (Bascom 1965:#4a). In Tohono
Oodham, this reflex labels tepary bean specifically while muuni is the generic term for bean(s) (Saxton, Saxton, and Enos 1983:43). In the Yepachi dialect
of Lower Pima, both bavi and mna (presumably derived from muuni) are
glossed as the generic bean(s) (Shaul 1994:319, 332). No terms resembling
muuni are attested in any of the other Tepiman languages, suggesting that the
Tohono Oodham and Yepachi Pima terms represent separate loans from Taracahitan sources that occurred after the breakup of the Proto-Tepiman speech
community.
Like PTEP *bavi, the terms for common bean(s) in the Tubar and Aztecan
subfamilies appear to be innovations. Tubar vupust may be related to PSUA
*pusi eye, but it is not derived from the Tubar term for eye, recorded as
tulr ~ tilr. The Proto-Aztecan term for common bean likely was *e or
*ee, based on Classical Nahuatl etl and Pipil eet, but this form can be reconstructed only for Proto-General Aztecan. No term for bean was recorded
for the other branch of the subfamily, Pochutec, before it became extinct (Boas
1917).

2012

WILLIAM L. MERRILL

225

6. The Northern Uto-Aztecan agricultural lexica.

The NUA societies
whose members engaged in farming included the Hopi, the Cahuilla, the
Timbisha Shoshone, some Western Shoshone bands, one Southern Ute band,
and most Southern Paiute bands, including the Chemehuevi (Bradfield 1971;
Lawton and Bean 1968; Steward 1938, 1941; Stewart 1942; Fowler and Fowler
1981; Stewart 1968).
Agriculture was a major component of the Hopi economic strategy (Forde
1931; Hack 1942). It also appears to have been significant among Southern
Paiutes living in the Virgin River drainage of southwestern Utah and southeastern Nevada, who are reported to have cultivated maize and other crops in
irrigated fields (Stoffle and Dobyns 1983:5055; Fowler 1995:11012). Farming definitely was secondary to foraging elsewhere and appears to have been
entirely absent among all other NUA societies (Bean 1978; Bean and Shipek
1978; Bean and Smith 1978a, 1978b, 1978c; Blackburn and Bean 1978; Smith
1978; Zigmond 1978; Fowler 1986; Kelly and Fowler 1986:37071).
One possible exception is the Comanches. They are known to have acquired agricultural products through trading and raiding but are not reported
to have farmed themselves (Kavanagh 2001:88991). Nonetheless, as Hill
(2002a:338) notes, the Comanche agricultural lexicon is sufficiently extensive
to suggest that, before European contact, farming may have formed part of
their economic strategy. The Comanches acquired horses in the early eighteenth century and may have abandoned farming to become specialized bison
hunters and horse pastoralists like several other post-European contact Plains
societies (Shimkin 1986:517; Oliver 1962).
The Hopi agricultural lexicon is exceptionally well documented, while
those recorded for other NUA languages are full of gaps. Nonetheless, sufficient data exist to conclude that an agricultural lexicon cannot be reconstructed for PNUA. Instead, as seen in the comparison of the agricultural lexica of
Hopi, Southern Paiute, Comanche, and Cahuilla presented in table 4, each of
these languages has a distinct agricultural lexicon that appears to have developed independently of the others, although some borrowing has occurred.
Table 4. Northern Uto-Aztecan Agriculture-Related Words

1. maize (generic)
2. pumpkin or squash
3. beans
4. to shell maize
5. parched maize kernels
6. tortilla
7. to sow, to plant
8. field for cultivation
9. planting or digging stick
NOTE: = no data.

HOPI

SOUTHERN PAIUTE COMANCHE

CAHUILLA

qaa
pata
mori
hmi
ktki
piqaviki
ya
paasa
sooya

hawiv, kumi
paraara
muri

a
pasa
poroc

mays
nehwet
xuul
ilay

sawi
we
pawisisual

haniibi
naksi
pihuraa

kukmep

tahnaar

226

ANTHROPOLOGICAL LINGUISTICS

54 NO. 3

6.1. Maize in Northern Uto-Aztecan languages. The generic labels for

maize presented in table 4 include four different native terms and Cahuilla
mays, derived from Spanish maz.
The Hopi term qaa maize, dry husked ear of maize is cognate with
terms for pine cone in Southern Paiute (kao) and Kitanemuk (ka), and
pine cone presumably was its original referent. Hill (2002a:338, 2008:161) reports that in Hopi the possessed form qaaat has the secondary meaning of
green cone of pine or literally its corn ear. This association is reversed in the
Chemehuevi term for maize cob, haw ka (Lawlor 1995:523), the literal
gloss of which is maize pine cone.
The two Southern Paiute terms for maize appear to have been the names
for distinct varieties of maize. Kelly (1964:39) reports that in the Kaibab dialect of Southern Paiute, hawiv labeled an earlier short-eared variety while
kumi labeled an introduced, long-eared variety. Both terms also are recorded
for the Shivwits dialect, but other Southern Paiute dialects apparently included only one term or the other, hawiv in the more westerly dialects (Moapa
and Chemehuevi) and kumi in the more easterly San Juan dialect, as well as in
Southern Ute (Fowler and Fowler 1981:134, 136).33
No data exist regarding the source of the introduced maize variety or its
label kumi. The obvious similarities to PSUA *kumi to nibble, to chew suggest diffusion from the south, perhaps from the Upper Pima (Fowler 1994:454
n. 5). Except for Tubar kumalt parched maize kernels, the Tepiman terms for
maize cob, reconstructed for Proto-Tepiman as *kumikur, are the only SUA
nouns with a maize-related meaning that are phonologically similar to Southern Paiute kumi (see section 3.13). Diffusion from the east or southeast also is
a possibility. A similar term is found in the Comanche word for popcorn,
kuhmitoai, for which Robinson and Armagost (1990:30) offer the literal
translation of heated turns inside out.
The word kumi also could be a Southern Paiute innovation. It has the same
three initial segments as kumut, the Southern Paiute label for cultivated and
wild amaranth species (Amaranthus caudatus, A. palmeri) (Bunte and
Franklin 1987:25, 28). The presumed Hopi cognate of kumut is komo (from
*kumu), the name for a cultivated amaranth (Amaranthus cruentus). Hopi
komo may have been the source of the term kokoma (from *kukuma) dark red,
almost purple maize, the link between them being their shared use as a red
food coloring (Whiting 1966:15; Hopi Dictionary Project 1998:146, 148).
These words provide additional support for the possibility, mentioned in
section 3.13, that PUA **ku served as a stem in the creation of names for valued seed plants. Other examples are the names for various species of Mentzelia
(blazing star), labeled kuu in the San Juan dialect of Southern Paiute (see
appendix 2, set 7). Zigmond (1941:21223) and Steward (1933:243, 1938:104)
report that Mentzelia seeds were an important staple for several Numicspeaking societies and were broadcast sown by some Central Numic bands in
the Great Basin.

2012

WILLIAM L. MERRILL

227

The final generic term for maize in table 4 is Comanche haniibi. The only
related words attested in other Numic languages are Northern Shoshone
haniib and Northern Paiute hanibi. This uneven distribution suggests that it
was a loanword, either from one of these Numic languages to the others or
from an external source.
Hill (2002a:336) notes a similarity between Comanche hani, the combining form of haniibi, and the Hopi word haani maize flour ground to the
desired consistency, but whether a loan occurred and, if so, in which direction
cannot be determined. The possibility exists that both words derive from the
same maize-related morpheme. The initial syllable ha also is encountered in
Southern Paiute hawiv maize, and Fowler (1994:#1.11) reports that terms
for corn, grain and parched corn in Zia, a Keresan language, have ha as
their initial syllable.
Hill (2002a:336) also points out a resemblance between Hopi hooma ceremonial corn meal and Comanche homopi powder, flour. Because Hopi o
reflects PNUA *u and Comanche o reflects PNUA *o, the two terms cannot be
cognates, but they could be the result of a loan, perhaps from Central Numic
into Hopi. Comanche homopi reflects ProtoCentral Numic *hoopi powder,
flour. Medial m is the expected Comanche reflex of * in the context o__o,
as seen in the Comanche term for lungs, soomo, which derives from *soo.
6.2. Squash and pumpkin in Northern Uto-Aztecan languages. The
generic label for squash and pumpkin in Hopi is pata, although Whiting
(1966:93) reported that this term labeled a single species, the domesticated
Cucurbita moschata. The label combines two morphemes, pa water and
taa thing(s) in a container, suggesting considerable antiquity for the use of
cucurbits as water containers. The Hopi word for wild squash (Cucurbita
foetidissima) is msiptaa, a compound of msi food packet and ptaa, one of
the combining forms of pata (Hopi Dictionary Project 1998:257). The bottle
gourd (Lagenaria siceraria) is labeled with a completely different term, tawiya,
for which I have found no cognates in other Uto-Aztecan languages.34
It is unclear if the first Southern Paiute term for squash or pumpkin in
table 4 is a generic term or labeled a specific species or variety of squash or
pumpkin, but there is no question that it is a Hopi loan. The forms attested in
different Southern Paiute dialects are [paraara] and [paraara] (Fowler and
Fowler 1981:136). The phonemic representation of the first is /patawata/.
This word corresponds to the Hopi word pataawta, which can be glossed as
water is inside.
A second Southern Paiute term for squash or pumpkin, reported only
from the San Juan band, was recorded as naxrs by Bunte and Franklin (1987:
28) and as nagtis by Kelly (1964:170). These alternate forms suggest that the
antecedent form was *naktis, which resembles Comanche naksi squash. A
loan may be involved but the original source of the term is unclear.

228

ANTHROPOLOGICAL LINGUISTICS

54 NO. 3

Cahuilla nehwet pumpkin reflects an antecedent *nhwt and appears to

be a Cahuilla innovation, derived from the Cahuilla word nekhi wild squash
(Cucurbita foetidissima).35 It combines the stem neh wild squash with the
augmentative suffix wet, literally meaning big wild squash. The same etymology is proposed in section 4 for the generic word for squash, pumpkin in
some SUA languages, also formed with reflexes of the PUA augmentative
suffix **w but with an unrelated stem.
6.3. The Northern Uto-Aztecan bean vocabulary. The terms for beans
in the four lexica represented in table 4 derive from sources external to the
Uto-Aztecan language family. The Comanche and Cahuilla labels are distinct
transformations of the Spanish word frijol bean(s), while the original source
of Hopi mori and Colorado River Numic muri is unknown (see section 5). The o
in the Hopi word may indicate that it was introduced before the general shift of
PNUA *u to Hopi o, but it could also have been introduced after this shift
occurred. Because o is the only back rounded vowel in Hopi, Hopi speakers may
simply have integrated the loanword muri as mori, just as they replaced an
original u by o in the word moola mule, from Spanish mula.
A medial r also is found in the Yuman term for bean, which has the basic
form of mark (Jel 1978:8386), but the difference in the vowels in the initial
syllable precludes concluding that the Hopis and Southern Paiutes acquired
the common bean and their terms for it from Yuman speakers. Jel (1978:
8687) identified the Yuman term as a likely loan from Hopi. She eliminated
Colorado River Numic as a possible source because she assumed that the
medial consonant in the Colorado River Numic term was [t] rather than [r],
based on the phonemic form /muutii/ presented by Miller (1967:#29) rather
than [muurii], the phonetic realization recorded by Sapir (1931:574), which
presumably represents the form of the word when it was loaned into Colorado
River Numic. If the Yuman term was a loan from a Northern Uto-Aztecan
language, the Chemehuevi dialect of Southern Paiute is the most likely source
because the Chemehuevis were in close contact with the Yuman-speaking
Mohaves. It is possible, however, that the loan occurred in the opposite direction. Chemehuevi terms for pumpkin, muskmelon, cotton and wheat are
all Mohave loans (Stewart 1968; Fowler and Fowler 1981:13637).
6.4. Other agricultural terms. The absence of documented terms in Southern Paiute, Comanche, and Cahuilla for many of the remaining six referents in
table 4 obviously precludes an evaluation of the relationships among them.
However, the forms that are attested in these languages are unrelated to the
Hopi terms with the same referents, with two exceptions.
The first involves the Hopi and Southern Paiute words for to plant, to
sow. These terms reflect PNUA *ya and, as discussed in section 3.6, derive
ultimately from PUA **ca, which likely had the broadcast sowing of wild

2012

WILLIAM L. MERRILL

229

seeds as its original referent. In addition to Hopi and Southern Paiute, reflexes
of PNUA *ya with the meaning to plant or to sow are attested in Southern
Ute and Kawaiisu (Southern Numic) and in Timbisha Shoshone (Central
Numic). Nouns derived from the reflexes of *ya but not the verbal reflexes
themselves are attested in Western Shoshoni (Central Numic) and Northern
Paiute (Western Numic), where they label Chenopodium species that were
broadcast sown (Steward 1938:23; Kelly and Fowler 1986:371) (see appendix 2,
set 4).
The absence of reflexes of PNUA *ya in Comanche and Cahuilla is unexpected. Because other Central Numic languages retain reflexes, the Comanche
reflex may have been lost late in the diversification of the Central Numic subdivision. In contrast, terms deriving from **ca are not recorded for any of the
Takic languages, suggesting that the loss of the reflex predated the emergence
of Cahuilla as a distinct language.
The second exception is paasa or pasa, shared by Hopi and Chemehuevi as
the term for field for cultivation. Lowie (1924:200) recorded the similar word
pass. as the Shivwits Southern Paiute label for an implement used to dig
irrigation ditches. The initial pa in this word and its association with irrigation suggests that the initial syllables in the Hopi and Chemehuevi words for
field for cultivation are the combining forms of their terms for water, paa in
Hopi, pa in Chemehuevi. The same morpheme may appear in the equivalent
Cahuilla term pawisisual, glossed as place where you can plant things (Bean
and Saubel 1972:206). Seiler and Hioki (1979:139, 143) recorded pa water
and paw to get water as the combining forms of these Cahuilla words, but
pawisisual does not appear in their work.
The relationship between the Hopi and Chemehuevi terms cannot be
determined, in part because terms for field for cultivation are not reported for
any other Southern Paiute dialects. However, if a loan was involved, a loan
from Hopi to Chemehuevi is suggested by the existence of an alternative
Southern Numic term for the same referent, derived from PNUA *ya to plant,
to sow and recorded in Southern Ute as ap and in Kawaiisu as ap.
7. A Hopi link to Southern Uto-Aztecan. In section 3.8, I noted that Hopi
is the only NUA language with a cognate for the SUA terms reflecting PSUA
*wika planting stick. Drawing on Stubbss (2011) extensive compilation of
Uto-Aztecan cognate and resemblant sets, I have compiled a list of ten other
words from SUA languages for which possible cognates exist only in Hopi or, if
cognates or resemblants are attested in other NUA languages, only the SUA
and Hopi words share referents.
All eleven words are presented in table 5. The first five have cognates in
more than one SUA subfamily, which allow PSUA etyma to be reconstructed. I
include these etyma and the Hopi cognates in table 5, together with the numbers of the sets in appendix 2 where the cognates in the SUA languages are
listed. PSUA etyma cannot be reconstructed for the other six words. For them,

230

ANTHROPOLOGICAL LINGUISTICS

54 NO. 3

I present in table 5 the SUA words that most closely resemble the Hopi words.
The numbers preceded by S- correspond to the numbers of the sets compiled
by Stubbs (2011) in which these words appear. (His set numbers for the first
six items in table 5 appear with those sets in appendix 2.) Stubbs does not have
cognate or resemblant sets for the words in items 6 and 8, but he (2011:402)
provides a list of the words for wolf in all of the Uto-Aztecan subfamilies.
Table 5. Hopi and Southern Uto-Aztecan Cognates and Resemblants

1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.

HOPI FORM

SUA FORMS

wiika wooden hoe

maalama to clear an area for
a new field
yooyoki to be raining
yaqa nose
karo parrot
kew wolf
kapi(at) neck
paallqanq Water Serpent,
the deity of water
pika(at) immature maize ear

*wika planting stick (#22)

*mawe to prepare land for
cultivation (#10)
*yuki to rain (#24)
*yaka nose (#23)
*haro macaw (#2)
NA-CL ketlaatli wolf
TBR kar neck (S-1510)
RR walluwi malevolent water deity

yoowi(at) corn silk

(from *yuuwi)
pca fuzzy caterpillar or worm

CR ikri, HC hiikri mature fresh maize

ear (from *pikuri) (S-545)
CR yuuri maize, mature maize ear
(from *yoori) (S-549)
NA-CL pooootl silk-cotton tree (S-557)

Definite cognates in NUA languages other than Hopi exist only for set 4,
nose. Reflexes of *yaka are attested in Tubatulabal yahaawit ~ yahaawil
summit, point and in Southern Numic words for side, edge (Stubbs 2011:
#1546), but only the Hopi and SUA words share the referent nose. Takic
words for snow, which reflect Proto-Takic *yuyi, perhaps are related to the
Hopi and SUA words that reflect *yuku to rain, but again the referents are
different (Stubbs 2011:#1763, #2076).
Four of the sets in table 5 are associated with agriculture or maize. Set 1,
planting stick, has cognates in all the SUA subfamilies except Tubar, while
cognates for set 2, to prepare land for cultivation, are attested in the Taracahitan, Corachol, and Aztecan subfamilies. For the two maize-related sets
(sets 9 and 10), only Hopi and Corachol words have been found. The Hopi and
Cora correspondences in set 10 are irregular, but the initial syllable of the
postulated antecedent form for each is the form attested in the other language.
The similarities between the Hopi and Cora words in set 5, macaw, are
intriguing. Reflexes of PSUA *haro are found in all five SUA subfamilies, with
the initial *h lost in most and replaced by another consonant in some, for example, Tubar walo (see appendix 2, set 2). Only in Hopi and Cora does an initial k appear instead: HP karo and CR kara. The final a in the Cora word is
due to vowel harmonization, and the Hopi word shows the palatalization of k

2012

WILLIAM L. MERRILL

231

expected before a and e (Hopi Dictionary Project 1998:866). In neither language is k the reflex of *h. This shared irregularity makes the two words exact
phonological cognates with one another but not with the other SUA words for
macaw.
Given the geographical proximity of Hopi and Tepiman speakers, the loan
of SUA words from Tepiman to Hopi cannot be discounted, but in the case of
the three words in (23), the loans would have had to have taken place before
PSUA *w and *y shifted to Proto-Tepiman *g and *d (Bascom 1965: 13).36
(23) *wika planting stick > PTEP *giika
*yuki to rain > PTEP *duuki
*yaka nose > PTEP *daaka

Shaul and Hill estimate that these Proto-Tepiman phonological innovations

occurred early in the first millennium A.D. (1998:380), indicating considerable antiquity for the loans if the Hopi words are from Tepiman.
An alternative possibility consistent with modern Hopi perspectives is that
some of their ancestors spoke a language or languages affiliated with the
southern branch of the Uto-Aztecan language family.37 Ferguson and ColwellChanthaphonh report that Some Hopis suggest the Tsuu (Snake Dance) and
Powamuy (Bean Dance) ceremonies have linguistic associations with languages spoken in the southern Uto-Aztecan area (2006:115). In addition, according to Hopi history (summarized by Ferguson and Colwell-Chanthaphonh
[2006:95149]), their ancestors included some people who originated far to the
south and later migrated northward to a place or region named Palatkapi.
After residing there for awhile, they resumed their migrations until they
reached the Hopi mesas in northeastern Arizona.
A growing body of archaeological and ethnographic evidence indicates that
some Hopi ancestors participated in the cultural traditions like Hohokam and
Salado that developed in the first and second millennia AD south of the
Mogollon Rim, in central and southern Arizona (Ferguson and ColwellChanthaphonh 2006:12048; Teague 1993; Webster and Lomaomvaya 2004;
Hays-Gilpen 2008:7476). This area may correspond to the location of
Palatkapi, and it may have been during this period in their history that they
acquired the wooden hoe labeled with the reflex of PSUA *wika planting
stick. (The shift in referent to wooden hoe also occurred in Akimel Oodham;
see appendix 2, set 22.) Remains of these hoes were recovered from the ruins of
Casa Grande, a major Hohokam center constructed around AD 1300 (Fewkes
1912:146; Crown 1991:15052).38
Following their arrival at the Hopi mesas, the Hopis southern ancestors
introduced new agricultural and religious practices, including some associated
particularly with water and rain. Given these associations, the fact that Hopi
cognates or resemblants exist for four sets in table 5 related to agriculture (sets
1, 2, 9, 10), as well as set 3, to rain, appears less than coincidental. Moreover,

232

ANTHROPOLOGICAL LINGUISTICS

54 NO. 3

the similarities between the Hopi and Rarmuri words for water deity suggests some interaction in the past between speakers of these languages.39 Hopi
ceremonies linked to the Hopi deity paallqanqw are also linked to the Water
Clan, one of the Hopi clans that migrated from the south. Two other southern
clans are the Parrot Clan, karam, and the Young Corn Clan, pikasam.
The stem of the name of the first clan is the combining form of the Hopi reflex
of the PSUA etymon that I have glossed as macaw (set 5). The stem of the
second is the combining form of the word for which Corachol resemblants exist
(set 9).
Cora cognates or resemblants are attested for five of the eleven sets in
table 5 (sets 1, 2, 5, 9, and 10). If the Hopis southern ancestors originated in
the interior of Mexico, they might have been in contact with the ancestral
Cora, and their language or languages could have been related to those of the
Corachol subfamily.40 The Coras have lived in west-central Mexico from the
time of European contact and presumably were located in the same area for
centuries before (Weigand 1992:18288; cf. Weigand and Garca de Weigand
2000). The northward migrations of the Hopis southern ancestors might have
begun from wherever the ancestral Coras were living at the time of their departure, either where the Coras currently reside or somewhere between there
and the southwestern United States. Such a long-distance migration is not entirely far-fetched given the fact that speakers of Tepiman languages expanded
across the same area, from southern Arizona to near the current Cora territory, at some point prior to European contact (Shaul and Hill 1998:38889;
Wilcox et al. 2008). In this scenario, the ancestral Hopi would have come in
contact with speakers of languages belonging to at least some of the other SUA
subfamilies.
On the other hand, except for the agriculture-related terms discussed here,
there is little evidence of SUA inflence in the Hopi agricultural lexicon. The
Hopi and PSUA lexica are quite different, as demonstrated in table 6, where
the PSUA terms presented in table 2 and the Hopi terms with the same referents are juxtaposed. The only Hopi word in this table that might be of SUA
origin is pika immature maize ear, which, as noted above, could be cognate
with Corachol words for mature fresh maize ear.
Table 6. ProtoSouthern Uto-Aztecan and Hopi Agricultural Lexica
REFERENT
1. maize (generic)
2. immature maize ear
3. to shell maize
4. parched maize kernels
5. tamale
6. to plant, to sow
7. field for cultivation
8. planting stick

PSUA

HOPI

*suhunu
*sita
*hora ~ *hori
*saki
*tma
*ca
*wasa
*wika

qaa
pika
hmi
ktki
somiviki
ya
paasa
sooya

2012

WILLIAM L. MERRILL

233

8. Agriculture and the ProtoSouthern Uto-Aztecan homeland. The

linguistic evidence presented in section 3 indicates that the PSUA speech community was intact when its members adopted maize agriculture. Evaluating
the degree to which they could have been involved in the initial diffusion of
agriculture and specifically maize agriculture between Mesoamerica and the
southwestern United States depends upon determining their location during
the period when this diffusion occurred. A consideration of additional linguistic
evidence and biogeographical data suggests that the PSUA speech community
most likely was located in northeastern Sonora and southeastern Arizona
when maize agriculture was introduced to the region.41
The identification of this area as the PSUA homeland is based on the geographical distribution of four wild plant and animal taxa labeled with etyma
that are PSUA innovations: wild chile, two kinds of palm, and macaw (see
table 7). The ranges of these taxa overlap in northeastern Sonora and regions
to the south. The wild chile (Capsicum annuum var. glabriusculum) is found in
southern Arizona up to latitude 32 north (Tewksbury et al. 1999:99100;
Kraft 2009). Based on the botanical identifications of the palm species labeled
by terms in SUA languages derived from PSUA *taku and *soyawa (see appendix 2, sets 16 and 18), the original referents of these terms likely were Sabal
uresana and one or more species of Brahea, the ranges of which extend northward to near the thirty-first parallel (Joyal 1995:146; Felger and Joyal 1999:3).
Macaws (Ara militaris) reach the northern limits of their distribution at about
latitude 30 north (CONABIO n.d.; igo-Elas 2000).42
Table 7. ProtoSouthern Uto-Aztecan Innovations in Names for Flora and
Fauna
PROTO-SUA ETYMA
1. *koori wild chile
2. *taku palm
3. *soyawa palm
4. *haro macaw

TEP

TRC

CRC

TBR

AZT

+
+

+
+
+
+

+
+

+
+
+
+

+
+

NOTE:

TEP = Tepiman; TRC = Taracahitan; TBR = Tubar; CRC = Corachol; AZT =

Aztecan; + = cognate present; = cognate absent.

It thus can be argued that the PSUA foraging bands were exploiting the
wild resources of an area that was located between the thirtieth and thirtysecond parallels. Although wild chiles, palms, and macaws occur south of the
thirtieth parallel, a more southerly location for the SUA homeland is unlikely.
No PSUA terms can be reconstructed for any species associated exclusively
with subtopical and tropical zones, including those encountered in the tropical
deciduous forests that occur in Sonora as far north as the twenty-ninth parallel
(Fowler 1983:234, 24546; Hill 2001b:917; Brquez et al. 1999:54, fig. 2.6;
Robichaux and Yetman 2000).

234

ANTHROPOLOGICAL LINGUISTICS

54 NO. 3

If the PSUA speech community had been located farther south, a likely
candidate for a label reconstructible to PSUA would be the silk-cotton tree
(Ceiba spp.), a morphologically quite distinctive and economically useful
genus, distributed in tropical areas from Sonora to South America. However,
the SUA terms for this tree are not cognates, even in SUA languages spoken in
Sonora, where they label Ceiba acuminata; terms for this tree documented for
these languages are shown in (24).
(24) NV aupukama
MY(Y) baoga
ED svur
WR-R wakap (Yetman and Felger 2002:185)

The location proposed for the SUA homeland includes the northern half of
the Serrana region of northeastern Sonora, situated along the upper drainages
of the Ro Sonora and Ro Yaqui and at the time of European contact one of the
most productive agricultural zones in the entire region (Doolittle 1980, 1984a,
1984b, 1988). Given the proximity of this postulated SUA homeland to the
early agricultural sites in southern Arizona, the inception of maize agriculture
among PSUA speakers presumably would have occurred at roughly the same
time, around 4100 BP (Merrill et al. 2009). If so, then it can be hypothesized
that the PSUA speech community was intact at least until around four thousand years ago.
If the PSUA speech community was located north of the twenty-ninth
parallel, it is doubtful that its members would have been responsible for diffusing maize agriculture across the thousand kilometers separating them from
Mesoamerica. It is, of course, reasonable to suppose that they could have
played a role in the introduction of maize agriculture to foraging societies
farther north (Carpenter, Snchez de Carpenter, and Mabry 2001; Carpenter,
Snchez, and Villalpando 2002). In addition, the movement of PSUA farmers
into new areas suitable for farming presumably would have been one of the
factors contributing to the dispersal of the PSUA speech community. Given the
distribution of the SUA languages at the time of European contact, movements
appear to have been primarily to the east, west, and south, where seasonal
rainfall and temperature regimes were more amenable to maize agriculture
than areas to the north. Expansion southward could have created a corrider
through which domesticated cucurbits and beans later diffused northward.
9. Conclusions. Four generalizations about the place of agriculture in UtoAztecan cultural history can be proposed, based on the comparative analysis of
the agricultural lexica of the Uto-Aztecan languages presented here.
First, the members of the Proto-Uto-Aztecan speech community were foragers who engaged in some forms of wild plant husbandry that included the
broadcast sowing of wild seeds. They also developed vocabulary, practices, and

2012

WILLIAM L. MERRILL

235

material culture linked to the procurement and processing of wild plants that
were later applied to domesticated plants and their cultivation by speakers of
both Southern and Northern Uto-Aztecan languages.
Second, prior to the adoption of agriculture by any Uto-Aztecan speakers,
the PUA speech community divided into two separate speech communities,
resulting in the emergence of the first-level daughter languages, Proto
Southern Uto-Aztecan and ProtoNorthern Uto-Aztecan. Following this split,
interaction between the speakers of these intermediate protolanguages was
minimal, and their subsequent engagements with agriculture occurred for the
most part independently of one another.
Third, the ProtoSouthern Uto-Aztecan speech community was intact
when its members adopted maize agriculture. Their dispersal and the diversification of PSUA into distinct dialects and languages began before the introduction of domesticated cucurbits and was well advanced by the time that they
integrated domesticated beans into their crop complexes. The interaction of
Tepiman and Taracahitan speakers after the emergence of the ancestral languages of the SUA subfamilies is indicated by loans of agriculture-related
terms between them. Proto-Corachol and Proto-Aztecan speakers appear not
to have formed part of this interaction sphere.
Finally, the members of the ProtoNorthern Uto-Aztecan speech community were foragers, not farmers, and foraging continued to be the sole or primary
component of the economic strategies of most Northern Uto-Aztecan societies.
The shift from foraging to a mixed foraging-farming strategy occurred late in
the diversification of the NUA branch of the language family and involved only
some NUA societies.
The third generalization, regarding the cultural history of maize, cucurbits, and beans among the Southern Uto-Aztecans, corresponds to the perspective advocated by Hill (2001a:346), but the fourth, regarding the cultural history of agriculture among the Northern Uto-Aztecans, is quite different from
the position that she has advocated. She (2001b:91617, 2012:65) proposes
that the members of the PUA speech community were farmers located within
Mesoamerica. As demand for new arable land increased, some began spreading
northward, leading to the emergence of a separate PNUA speech community
whose members eventually reached the southwestern United States, where
they introduced maize agriculture. To account for the absence of evidence for
an agricultural lexicon reconstructible to PNUA, she (2001b:927, 2002a) suggests that reflexes of PNUA etyma related to agriculture existed in all of the
ancestral languages of the NUA subfamilies but either were never recorded or
were lost when some NUA societies ceased farming to rely exclusively on foraging for their survival.
Lexical loss or the failure to record agriculture-related words in some NUA
languages definitely must have occurred in the case of the reflexes of the PUA
verb to plant, to sow. This etymon can be reconstructed to PUA as *ca, but
reflexes of it are not attested in Tubatulabal or the Takic languages, nor in a

236

ANTHROPOLOGICAL LINGUISTICS

54 NO. 3

few of the Numic languages (see section 6.4). It also is likely that some of the
NUA societies documented ethnohistorically or ethnographically as fulltime
foragers practiced some farming earlier in their histories. However, I do not
interpret the available evidence as supporting the conclusion that the members of the ancestral PNUA speech community as a whole were farmers.
From my perspective, the first NUA farmers likely were speakers of ancestral Hopi who could have integrated farming into their foraging economies
during the period, roughly 30002150 BP, when maize agriculture was initially spreading across the American Southwest and into adjacent areas of the
northern Colorado Plateau (Wilde and Newman 1989:714; Lyneis 1995:2078;
Madsen and Simms 1998:293; Matson 2002; Kohler and Glaude 2008:8283).
Later, speakers of other ancestral NUA languages could have developed mixed
foraging-farming strategies, but some never adopted maize agriculture because they were located in areas, like southern California, where reliable wild
food resources were abundant, or farther north, where local environmental
conditions rendered maize production unreliable.
I also suspect that climatic fluctuations in western North America were
responsible for shifts between foraging and mixed foraging-farming strategies
that likely occurred on multiple occasions during the history of the region.
Although a number of factors have been proposed to account for the disappearance of farming by around AD 1250 (700 BP) in areas of the Great Basin
and Colorado Plateau associated with the Fremont archaeological tradition,
increasing aridity during the maize growing season must have been involved
(Madsen and Simms 1998:31320). Similarly, decreasing temperatures during
the Little Ice Age, dated for western North America to roughly AD 14001850
(550100 BP), may been responsible for subsequent abandonment of farming
in the same general area (Koerper, Killingley, and Taylor 1985; Matthews and
Briffa 2005; Stine 2004:5354).
In fact, I interpret reports from the nineteenth and early twentieth centuries that some Numic speakers who lived in the Great Basin and on the northern Colorado Plateau had recently begun small-scale farming as evidence
that they were actually resuming a mixed foraging-farming strategy that they
had abandoned during the Little Ice Age (Steward 1938:122, 137; Kelly and
Fowler 1986:371; Fowler and Fowler 1981:13238). The ensuing warming
trend that began around AD 1800 (150 BP) allowed Numic farmer-foragers to
expand northward into areas of the Great Basin that may have been farmed
prior to AD 1400 (550 BP), but this expansion (or reexpansion) was cut short by
the arrival of Euro-American settlers, who appropriated the best and in some
cases only arable lands (Matthews and Briffa 2005:23; Stoffle and Dobyns
1983:49; Stoffle and Zedeo 2001). A mixed foraging-farming strategy persisted, however, in a few Numic communities located farther to the south,
where ethnographers like Isabel Kelly (1964; Fowler 1995) were able to
observe its pursuit in the 1930s.

2012

WILLIAM L. MERRILL

Appendix 1: Language Abbreviations and Sources

AK = Akimel Oodham (Rea 1997)
AZT = Aztecan
CA = Cahuilla (Seiler and Hioko 1979)
CA(B) = Cahuilla (Bean and Saubel 1972)
CH = Chemehuevi (Press 1979)
CM = Comanche (Robinson and Armagost 1990)
CP = Cupeo (Hill and Nolasquez 1973)
CR = Cora (McMahon and McMahon 1959)
CR(O) = Cora (Ortega 1860)
CR(P) = Cora (Preuss 1934)
CR(V) = Cora (Valias Coalla 2000)
CRC = Corachol
ED = Eudeve (Pennington 1981)
HC = Huichol (McIntosh and Grimes 1954)
HC(G) = Huichol (Grimes et al. 1981)
HC(GM) = Huichol (Gmez Lpez 1999)
HP = Hopi (Hopi Dictionary Project 1998)
KT = Kitanemuk (Anderton 1988)
KW = Kawaiisu (Zigmond, Booth, and Munro 1991)
LP = Lower Pima (Bascom 1965)
LS = Luiseo (Elliott 1999)
LS(B) = Luiseo (Bright 1968)
MY = Mayo (Collard and Collard 1962)
MY(V) = Mayo (Valias Coalla 2000)
MY(Y) = Mayo (Yetman and Van Devender 2002)
NA-CL = Classical Nahuatl (Karttunen 1992)
NA-CL(M) = Classical Nahuatl (Molina 1970)
NP = Northern Paiute (Liljeblad, Fowler, and Powell 2012)
NT = Northern Tepehuan (Bascom 1965)
NT(R) = Northern Tepehuan (Rinaldini 1994)
NUA = Northern Uto-Aztecan
NV = Nvome (Pennington 1979)
OP = pata (Lombardo 1702)
OP(P) = pata (Pimentel 1863)
PNUA = ProtoNorthern Uto-Aztecan
PO = Pochutec (Boas 1917)
PP = Pipil (Campbell 1985)
PSUA = ProtoSouthern Uto-Aztecan
PTEP = Proto-Tepiman
PUA = Proto-Uto-Aztecan
PYP = Lower Pima (Yepachi dialect) (Shaul 1994)
PYP(R) = Lower Pima (Yepachi dialect) (Rea 1997)
RR = Rarmuri (Brambila 1976)
RR(H) = Rarmuri (Hilton 1959)
SP = Southern Paiute (Kaibab dialect) (Sapir 1931)
SP-K = Southern Paiute (Kaibab dialect) (Kelly 1964)
SP-SJ = Southern Paiute (San Juan dialect) (Kelly 1964)
ST = Southern Tepehuan (Bascom 1965)
SUA = Southern Uto-Aztecan
SUt = Southern Ute (Givn 1979)

237

238

ANTHROPOLOGICAL LINGUISTICS

54 NO. 3

TB = Tubatulabal (Munro and Mace 1995)

TB(EV) = Tubatulabal (Voegelin 1938)
TBR = Tubar (Lionnet 1978)
TEP = Tepiman
TO = Tohono Oodham (Saxton, Saxton, and Enos 1983)
TO(M) = Tohono Oodham (Mathiot 1973)
TRC = Taracahitan
TSH = Timbisha Shoshone (Dayley 1989)
UP = Upper Pima (Bascom 1965)
WR-R = River Warih (Medina Murillo 2002)
WR-S = Sierra Warih (Miller 1996)
WSH-G = Western Shoshone (Gosiute dialect) (Miller 1972)
WSh-G(C) = Western Shoshone (Gosiute dialect) (Chamberlin 1911)
YQ = Yaqui, Sonora (Estrada Fernndez et al. 2004)
YQ-AZ = Yaqui, Arizona (Molina, Valenzuela, and Shaul 1999)

Appendix 2: Cognate Sets

The cognate sets are organized in alphabetical order by the reconstructed etyma,
which should be regarded as approximations. Many regular phonological correspondences among the Uto-Aztecan languages remain unidentified, especially in secondsyllable vowels in reflexes of disyllabic etyma, where sound changes and loss are
common. The numbers with the S- prefix correspond to those in Stubbs (2011), the
most comprehensive compilation of Uto-Aztecan cognate and resemblant sets available.
The words included in each set show expected correspondences in the initial syllable plus the initial segment of the second syllable. Deviations relevant to the analysis
are noted in the comments. The sets do not include cognates from all Uto-Aztecan languages and dialects. The principal source of data for each language is the first source
listed in appendix 1. Data from the other sources are included only when cognates are
not attested in the principal sources or sometimes when the attested forms in them
differ from those of the principal source.
The cognates in each set are organized by subfamilies in the following order: (1)
Numic, (2) Tubatulabal, (3) Hopi, (4) Takic, (5) Tepiman, (6) Taracahitan, (7) Tubar, (8)
Corachol, and (9) Aztecan. Most sets lack cognates from several subfamilies, but the
order and numbering are retained. The abbreviations for all of the languages are found
in appendix 1. Glosses for individual cognates are presented only when they deviate
from the referents assigned to the reconstructed terms. When a cognate has multiple
referents and one corresponds to that of the reconstructed etyma, the other referents
are not included.
1. PSUA *hari wild squash (S-2141). (5) TO ad,awi wild squash (Cucurbita digitata),
haal squash, pumpkin (loanword); NV aari rainy-season squash or pumpkins
(probable loanword), azabi wild squash; PYp(R) ara wild squash; PYP ara small
squash. (6) YQ ayaawim squash, pumpkin (Cucurbita moschata); MY aayau (sg.),
ayawim (pl.) squash, pumpkin; ED ar gourd dipper, haris wild squash; WR-R
haar ~ ar gourd canteen, gourd dipper, halwe ~ harwe ~ alwe ~ arwe
squash, pumpkin, maharwe wild squash, halap ~ alap ~ arap gourd, gourd
dipper; WR-S ar gourd canteen, gourd dipper, alwe pepo squash (Cucurbita pepo),
alap gourd, gourd dipper, alci gourd, gourd dipper; RR ar bottle gourd, aris wild
squash, lab gourd dipper cut breadthwise, lci gourd dipper cut lengthwise. (7) TBR
halipt a kind of gourd, gourd dipper, hualt a kind of gourd used to transport water.
(8) HC ari calabash, bottle gourd. (9) NA-CL ayotli gourd, squash, pumpkin; PP

2012

WILLIAM L. MERRILL

239

ayuh a variety of squash or pumpkin; PO eyut squash or pumpkin. COMMENT:

Pennington (1963:45, 164) identifies Rarmuri ar and aris as the labels for Lagenaria
siceraria and Cucurbita foetidissima respectively.
2. PSUA *haro macaw (S-217). (3) HP karo. (5) TO aad,ho peafowl; NV azo. (6) ED
hro; WR-S wal; RR wal ~ war. (7) TBR wal. (8) CR(P) kara(s). (9) NA-CL(M) alo.
COMMENT: The Classical Nahuatl reflex of *haro clearly served as a generic label.
Molina (1970:4r, 19v, 86r) glossed alo as large parrot, kaw alo as large green parrot,
and iiltik alo as large red parrot.
3. PSUA *hora ~ *hori to shell maize (S-552). (6) ED hran; WR-S ola; RR or ~ or
to shell (generic). (8) HC urika. (9) NA-CL ooya to shell (generic); PP uuya to shell
(generic), tauuya to shell maize; PO teyul maize. COMMENT: Nvome hoza to harvest
maize closely resembles Eudeve hran, but h in the Tepiman languages is the reflex of
PSUA *s, not *h. If Nvome hoza and its Tohono Oodham cognate oood,a ~ od, ~ oo to
harvest, to gather fruit are linked to this set, they likely are loans.
4. PUA **ca to plant, to sow (S-1635). (1) NP api pigweed; TSH ah; WSH-G appih
pigweed (?); KW a; SP a; CH ga; SUT ay. (3) HP ya. (5) TO a. NV sa; PYP sa;
NT i; ST i. (6) YQ ea; MY ea; ED eca; WR-S eca; RR i. (8) CR(O) aca; HC ecarika.
5. PNUA **ko chenopod (S-1655). (1) WSH-G kokax; KW koov; SP-K kov. (4) CA(B)
kiawet; CP qit; LS(B) qet; KT kokt. COMMENT: The referents of the Cupeo and
Kitanemuk terms are given as Chenopodium spp. The species of Chenopodium
associated with the other terms are: Kawaiisu: C. album, C. humile; Southern Paiute
(Kaibab): C. fremontii; Cahuilla: C. californicum, C. humile, C. fremontii, C. murale;
Luiseo: C. album.
6. PSUA *kokori (wild) chile (S-1597). (5) TO kookol a chili pod; chili powder; TO
aal kookol ~ uus kookol wild chile; NV kokori; PYP kokoli ~ kookil; NT kkoli;
ST kokol. (6) YQ kokoi; WR-S kokri; RR kor. (7) TBR kokl. (8) CR kuukuri; HC
kuukuri. COMMENTS: (a) Except in Tohono Oodham, these cognates apparently serve
as generic labels for both wild and domesticated chiles. (b) Classical Nahuatl illi is an
innovation and the source of the word chile. (c) Classical Nahuatl kokoaa to be sick, to
hurt; to hurt someone is a reflex of PSUA *kooko to be sick, to hurt from which the
label for chile is derived. Takic cognates are attested for the verb but not for the words
for chile (Stubbs 2011:#1597).
7. Proto-Numic *kuha blazing star (Mentzelia spp.); (1) NP kuha blazing star seeds;
TSH kuha blazing star (Mentzelia spp.), kuhwa blazing star seeds; WSH-G(C) kuhwa
Mentzelia spp.; KW kuuv; SP-SJ kuu. (2) TB(EV) kuul. COMMENTS: (a) Vowel harmonization and the loss of the medial h could account for the Kawaiisu, Southern
Paiute, and Tubatulabal words, but they also could reflect an etymon or etyma distinct
from *kuha. (b) Zigmond (1941:213), Kelly (1964:42, 153, 179), and Lawlor (1995:483
85) provide additional labels for Mentzelia species from other Southern Paiute dialects.
(c) The Hopi label for Mentzelia spp. is the innovation sililitaqa, formed from the root
sili to crackle apparently in reference to the rattling sound made by the ripe seed pod
(Whiting 1966:85; Hopi Dictionary Project 1998:502). The Havasupai word sel, which
labels Mentzelia albicaulis (Smith 1973), presumably is a Hopi loan.
8. PSUA *kumi to chew, to crunch, to nibble (S-777). (5) TO kuum to chew, to crunch;
TO(M) kuum to eat, chew on something that comes in little pieces, such as corn,
popcorn, and pieces of candy; PYP kuum to chew. (6) YQ kume to chew; MY kume
to chew; WR-S kumi to chew something hard and crunchy like parched maize
kernels or squash seeds; RR kum to eat maize kernels (parched, fresh, or dried). (7)
TBR kumi to eat small or ground up things, to eat maize. (8) HC(G) kmi to nibble.
COMMENT: The medial glottal stop in the Warih reflex and the correspondence of

240

ANTHROPOLOGICAL LINGUISTICS

54 NO. 3

identical vowel sequences in the Tepiman, Yaqui, and Mayo reflexes suggest that the
PSUA etymon should be reconstructed as *kuhumi.
9. PUA **mata metate (S-1082). (1) NP mata; KW maraci; SP maraci; SUT maraci. (2)
TB manaal. (3) HP mata. (4) CA malal; CP malal; LS malaal. (5) TO maud, ; PYP
maatur; NT mauturai; ST mattur. (6) YQ mta; MY mtta; ED matt; WR-S maht; RR
mat. (7) TBR matt. (8) CR maata; HC maataa. (9) NA-CL metlatl; PP metat; PO mot.
COMMENTS: (a) The medial r in the Southern Numic reflexes is the result of lenition of
*t. (b) Tubatulabal l rather than n would be expected.
10. PSUA *mawe to prepare land for cultivation (S-1639). (3) HP maalama. (6) YQ
mohte; MY magohte; ED mwan to plow; WR-S mawe; RR maw. (8) CR maire; HC
imayaari. (9) NA-CL sakamoaa; PP meewa. COMMENTS: (a) Hopi l is the expected
reflex of *w between low vowels (Voegelin, Voegelin, and Hale 1962:53). (b) The
medial w has been lost in the Yaqui, Cora, Huichol, and Classical Nahuatl reflexes,
and the Pipil reflex shows vowel metathesis.
11. mu bean (loanword) (S-131). (1) SP muurii; SP-SJ muruis. (3) HP mori. (5) TO
muuni; PYP mna. (6) YQ mni; MY munim beans; ED mun; WR-S mun; RR mun. (8)
CR muhume; HC muume. COMMENTS: (a) In the PYP reflex, uu has shifted to the
unrounded . (b) The identical vowel sequences seen in words from four of the five
subfamilies suggest that the first two syllables of the loanword from which they derive
may have been muhu, as attested in the Cora word.
12. PSUA *murayawa inflorescence (S-536). (5) TO mud,aag; NV muzadaga; PYP
murat maize spike. (6) YQ ma wheat tassel; MY mouga to produce spike(s); ED
murt; WR-S mul maize tassel; RR mur. (8) HC imaye to produce spike(s). (9) NACL miyaawatl maize tassel and flower.
13. PSUA *paci seed (S-1916). (6) YQ bi maize; MY bti maize; ED suvci; OP(P)
vai maize with formed kernels; WR-S pahc; RR pa mature fresh maize ear. (7) TBR
wacirn. (8) CR haci; HC haci. (9) NA-CL atli; PO at. COMMENTS: (a) Words that derive
from reflexes of *paci are discussed in section 3.10. (b) The pata reflex is attested in
Pimentel (1863:311), glossed as el maiz ya granado. I have been unable to find this
gloss in Lombardo (1702), Pimentels principal source of pata lexical items. Lombardo
(1702:151v) does include vait, glossed as squash seeds or a similar thing (las pepitas
de las calabazas o cosa semejante), essentially the same gloss assigned to the Eudeve
cognate, bacit squash seed. This gloss does not appear anywhere in Pimentels work,
suggesting that he may have altered the original gloss. However, Pimentels vai lacks
the final t, presumably the absolutive suffix, and Pimentel appears to include the
absolutive suffix and other word-final consonants if they are attested in Lombardos
work. Perhaps two different words are involved. Valias Coalla (2000:19798) glosses
pata paci as both seed and mature fresh maize ear.
14. PSUA *saki parched maize kernels (S-524). (5) TO haaki; PYP haahaki; NT ki;
ST haak. (6) YQ-AZ saakim; MY saki; ED sakt; WR-S sak; RR sak. (8) HC ~aki. (9) NACL iiskitl; PP iiseki to toast. COMMENT: The Tepiman terms label parched grains in
general.
15. PSUA *sita immature maize ear (S-538). (6) YQ-AZ sita; MY(V) sitawa; ED stven
to sprout an ear of maize; WR-S sit corn silk; RR sit. (8) CR(O) sitati; HC ~iita. (9)
NA-CL iilootl; PP iilut. COMMENTS: (a) Miller (1993:150) glosses Sierra Warih sit as
immature maize ear. (b) Tubar solt mature fresh maize ear could be a Nahuatl
loanword if vowel metathesis occurred.
16. PSUA *soyawa a kind of palm (S-1607). (6) WR-S sa Sabal uresana; RR sow. (7)
TBR saywt. (9) NA-CL sooyaatl. COMMENTS: (a) The PSUA etymon is reconstructed as

2012

WILLIAM L. MERRILL

241

trisyllabic based primarily on Tubar saywat, which presumably lost a second-syllable

vowel. (b) The reworking that is evidenced in all the reflexes includes vowel harmonization and the loss and metathesis of both vowels and consonants.
17. PSUA *suhunu maize (generic) (S-535). (5) TO huuni; UP huu; NV hunu; PYP
huun ~ huuno; NT nui; ST huun. (6) MY snnu maize field; ED sunt; WR-S sun;
WR-R suun; RR suun; RR(H) sun. (9) NA-CL sintli ~ sentli dried maize ear; PP
sinti maize, dried maize ear; PO son dried maize ear. COMMENT: Hill (2005:2, 2008:
164, 2012:58) indicates that the Gabrielino term for tortilla is ~oaaxey, which she
analyzes as ~o corn + aaxe put in mouth + y nonpossessed noun suffix, citing
John P. Harringtons unpublished fieldnotes as her source for the Gabrielino data (Hill
2008:159 n. 6). I did not encounter this word in my review of the microfilm of these
fieldnotes. The notes do, however, include a similar word that Harrington transcribed
and glossed as aahaj, bread (Harrington 1981: reel 102, frame 0672). This word
may derive from Proto-Takic *sawa (Stubbs 2011:#266c) to make tortillas or bread,
with the velar nasal resulting from a shift of *w to that Munro (1973) reports for
some words in Luiseo, another Takic language.
18. PSUA *taku a kind of palm (S-1606). (5) TO takui soaptree or soapweed yucca
plant (Yucca elata). (6) Y Q tko Washingtonia robusta; MY(Y) taako Brahea
aculeata, Sabal uresana; ED takt; WR-S tahk Brahea aculeata; RR rak. (7) TBR
takt. (8) CR tak; HC(Gm) tak. COMMENT: The shift in the referent of Tohono Oodham
takui presumably occurred because the palm taxa labeled with reflexes of *taku in
other SUA languages do not occur in the Upper Piman area, and the leaves of at least
some of these taxa and Yucca elata are used in weaving (Rea 1997:28485; cf. Joyal
1996a, 1996b).
19. PUA **tusa to grind (S-1081). (1) NP tusu; TSH tusu; CM tusur; KW tusu; SP tusu;
SUT tsui. (2) TB tusut. (3) HP tosta. (4) CA tus; LS tuu~ to crumble; KT tuh. (5) TO
ua ~ ua ~ uai; NV tuha; PYP tuia. (6) YQ tuse; MY tuse. ED tusn; WR-S tusu;
RR rus. (7) TBR tus. (8) CR tsh; HC t~iya. (9) NA-CL tesi; PP tisi; PO toso.
20. PUA **tusi something ground up. (1) NP nadussup meal, hanibinnadussupe
ground maize kernels; CM tusup pulverized or grated object; CM hanitusup ground
corn, cornmeal; CH tusup flour, something ground up; SUT tsup flour. (3) HP toosi
roasted sweet corn that is dried and ground to a fine texture. (4) CA tus something
ground up. (5) TO ui flour, ground food, pollen; NV tuhi ~ tusi anything ground up,
ground parched maize; PYP tui meal, flour; NT ti flour, tapi maize flour; ST
tui flour, tuip maize flour. (6) YQ saktsi ground parched maize; MY sk tssi
ground parched maize, tusi dough; ED tust ground parched maize; WR-R tus ~
tuus ground parched maize, maize dough; RR rus finely ground grain, kobrusi ~
kobsi ground parched maize. (7) TBR matust ground parched maize. (8) CR(V)
matsi ground parched maize. (9) NA-CL tetli flour, meal; PP titi dough, maize
dough; Po tot dough.
21. PSUA *wasa field for cultivation (S-1636). (5) TO gagka a clearing, gagkat to
clear land; NV gaga; PYP gaha. (6) YQ wasa; ED gast; WR-S was; RR was. (8) CR
wastii planted (adj.); HC waa. COMMENTS: (a) The initial syllable in the Tohono
Oodham form, gag, apparently involves reduplication, as seen in the Nvome reflex.
The Yepachi Pima cognate shows the expected Tepiman reflexes g and h of PSUA *w
and *s. (b) Grimes (1980:272) reports that Huichol waa also designates maize plant.
22. PSUA *wika planting stick (S-672). (3) HP wiika ancient wooden hoe. (5) AK giiki
wooden hoe; TO giiki; NV gika; PYP giika plow (n.); NT gikai; ST giik. (6) MY wika;
WR-S wka; RR wik; RR(H) wka. (8) CR(O) vikati; HC(G) wika. (9) NA-CL(M) wiktli.

242

ANTHROPOLOGICAL LINGUISTICS

54 NO. 3

23. PSUA *yaka nose (S-1546). (3) HP yaqa. (5) TO aak; PYP daaka; NT daka; ST
daak. (6) ED dakt; YQ yka; MY ykka; WR-S yahk; RR ak. (9) NA-CL yakatl; PP yak;
PO yeket.
24. PSUA *yuki to rain (S-1763). (3) HP yooyoki to be raining. (5) TO juuk; NT duki
rain (n.). (6) YQ yke; MY ykke; ED dkun; WR-S yuku; Rr uk.

Notes
Acknowledgments. I am grateful to Karen Adams, Cecil Brown, T. J. Ferguson,
Catherine Fowler, Ives Goddard, Robert Hard, Jane Hill, Kenneth Hill, A. C.
MacWilliams, and Brian Stubbs for their comments on earlier versions of this article. I
also thank Francisco Barriga, Lyle Campbell, Karen Dakin, Maggie Dittemore, Zarina
Estrada, Carmen Ezyaguirre, Wanda West, and Rosa Yaez for their assistance in
obtaining copies of several key studies and Marcia Bakry for preparing map 1.
Transcription and graphic conventions. My orthography corresponds in most respects to the Americanist phonetic notation; c represents the voiceless alveolar affricate, the voiceless post-alveolar and alveopalatal affricates, and the voiceless alveopalatal fricative; VV (where V stands for any vowel) denotes both vowel length and
identical vowel sequences. The acute accent indicates high tone in Northern Tepehuan
words and stress elsewhere. Falling tone in Hopi is marked by a grave accent.
I retain the modern technical orthographies developed for each of the languages
considered with a few exceptions. I use ~ to represent a voiceless retroflex sibilant and
(instead of ) for a high, central or back unrounded vowel. In Southern Ute words,
represents the high, central unrounded vowel, and the high back unrounded vowel. I
have adopted z for the Nvome sound represented in the original source with the
digraph rh. For Tohono Oodham, I use d, for the retroflex apico-alveolar stop and for
the lenis apico-dental stop. For Tubar, o represents the phoneme that Lionnet (1978)
interpreted as , u the allophone of u that he interpreted as [o], and l corresponds to the
graphemes l and that represented the sounds that he interpreted as allophones of r.
I use the citation forms of the original sources, but I eliminate initial glottal stops
before vowels except when they are relevant to the analysis. In appendix 2 and often in
the main text, I also omit the hyphen used in some sources to separate noun stems from
absolutive suffixes, which typically mark nouns in a nonpossessed state in the UtoAztecan languages that have retained them. Etyma are marked with ** for Proto-UtoAztecan (PUA) and * for ProtoNorthern Uto-Aztecan (PNUA), ProtoSouthern UtoAztecan (PSUA), and the other intermediary protolanguages.
1. The BP (before present) dates are calculated using 1950 as the point of reference
for the present.
2. Valias Coalla (2000) also is a major contribution that focuses on the Southern
Uto-Aztecan languages.
3. Hill developed this scenario within the framework of the farming and language
dispersal hypothesis, originally proposed by Renfrew (1987, 1992) and initially elaborated primarily by Renfrew and Bellwood (Bellwood 1993, 1997, 2001, 2005; Bellwood
and Renfrew 2002; Diamond and Bellwood 2003; Bellwood and Oxenham 2008). The
farming and language dispersal hypothesis has been examined in a large number of
publications, some supportive of it, others critical. Samples of both perspectives are
found in Renfrew et al. (1988) and Bellwood and Renfrew (2002); Hammarstrm (2010)
offers a global-scale evaluation of the hypothesis. Studies that relate specifically to Hills
proposals for the place of agriculture in Uto-Aztecan cultural history include Wichmann
(2002), Matson (2002), Carpenter, Snchez, and Villalpando (2002), Mabry (2005),

2012

WILLIAM L. MERRILL

243

Mabry, Carpenter, and Sanchez (2008), LeBlanc (2008), Wilcox et al. (2008), Brown
(2010a), Wichmann, Mller, and Velupillai (2010), and Caballero (2011).
4. The abbreviations and sources of data for each of the Uto-Aztecan languages
considered in this article are listed in appendix 1. My transcription and graphic conventions are discussed above, following the acknowledgments. Cognate sets not included in
the main body of the essay are presented in appendix 2. In the introduction to the latter,
I explain the criteria I have used in identifying cognates.
5. A number of additional languages, now extinct, may have belonged to the UtoAztecan language family, but their affiliation cannot be determined because they are
undocumented (Miller 1983a; Campbell 1997:13335).
6. No overview of Uto-Aztecan societies has been prepared. The simplest way to
access basic information on these societies is through the Wikipedia page Uto-Aztecan
languages (http://en.wikipedia.org/wiki/Uto-Aztecan_languages) and associated links.
7. Hill (2005:2, 2008:164, 2012:58) identifies Gabrielino so as cognate with Hopi
soow and the SUA terms for maize (see appendix 2, cognate set 17). For a discussion of
the regular correspondence of NUA and SUA n, see Merrill (forthcoming).
8. The square brackets around [g] and [nn] indicate that these consonant
clusters are the phonetic realizations of phonemic nk and nn.
9. Steward (1933:244, 1938:22) also encountered the use of Atriplex argentea among
the Owens Valley Northern Paiute, reporting that they called this plant sunuva. This
term presumably is the same as sunp reported by Liljeblad, Fowler, and Powell
(2012:470), who identify it as the name for saltbrush (Atriplex rosea). Because Northern
Paiute tends to retain as the reflex of PNUA *, the n in these labels may indicate
that the Western Shoshone labels for Atriplex species recorded by Steward are not
cognate with Hopi soow, unless the Northern Paiute terms are loans from Western
Shoshone.
10. The reconstructions *ooloo and *iiloo are from Dakin (1982:#229, #269).
Terms for immature maize ear and maize cob probably existed in the Proto-Aztecan
agricultural lexicon, but the reconstruction of Proto-Aztecan forms for them is precluded
by the fact that Pochutec words with these referents were not recorded (Boas 1917).
11. Both terms are attested in the Cora vocabulary originally published in 1732
(Ortega 1860:588, 599). Cora is a reflex of PSUA *k. The final e, which reflects *, is
unexpected, although both i and are attested in alternate forms of the Tohono
Oodham cognate hahak ~ haaki to roast grain with coals in a basket (h is the
Tepiman reflex of PSUA *s). The closed first syllable in ae is the result of reduplication of the initial syllable followed by the loss of the second-syllable vowel: *ae >
*aae > ae.
12. In a previous publication, colleagues and I (Merrill et al. 2010:E35) questioned
Hills (2010:E33) identification of the Luiseo term for grain, wheat as cognate with the
SUA reflexes of *saki, basing our view on the assumption that the i formed part of the
Luiseo stem, as presented by Bright (1968:39): ~axi. The correct form, ~axi, is
reported by Elliott (1999:832; cf. Hill 2012:58).
13. The SUA cognates and some possible NUA cognates are presented in this section
rather than in appendix 2. These words also appear in Stubbs (2011:#284, #527; cf. #285).
14. Hopi tma the polished piki stone, the griddle for making piki may be cognate
with the Numic terms in (4) and by extension the PSUA reflexes of *tma. In a 2001
article, Hill (2001b:92122) suggested that the word is cognate but later concluded that it
is not, since Hopi shows no reflex of the glottal stop (Hill 2004:7374). However, Hopi
seldom retains an original preconsonantal glottal stop. In her most recent essay on the
subject, she (2012:58) does not include the cognates in her PUA maize vocabulary, only
the reconstructed form *tma, which she glosses as tortilla, tamale even though none of
the potential NUA cognates has these referents.

244

ANTHROPOLOGICAL LINGUISTICS

54 NO. 3

15. In Rejogochi, the Rarmuri community where I have conducted most of my

research, the word rem labels both tortilla and tamale. Brambila (1976:464) and
Hilton (1959:67) gloss this term only as tortilla.
16. In Pipil, no word for tamale is attested, only a word for meat tamale,
nakatamal, a compound of naka meat and tamal. Pipil takal designates a tortilla
made of younger, tender ears of corn, while tamal is glossed as tortilla.
17. Liljeblad, Fowler, and Powell (2012) adopt the slash (/) to represent fortis
consonant alternation, often called a geminating final feature and represented by a
superscript g g or by straight quotation marks ".
18. Downs (1966) and Winter and Hogan (1986) discuss the sowing of wild seeds and
other ways in which Indigenous people in the Great Basin and on the northern Colorado
Plateau manipulated wild plants to increase their productivity. Doolittle (2000) and
Smith (2001b, 2011) provide general overviews of such practices.
19. The words for planting stick reported for Eudeve and pata, both Taracahitan
languages, are naakt and nt, respectively. They do not reflect PSUA *wika, but they
are cognate with terms in three NUA languages: Tubatulabal nahat cane, Luiseo
naxut walking stick, and Kitanemuk nakat digging stick or any kind of stick.
20. Proto-Aztecan *i reflects both PSUA *u and *i (Campbell and Langacker 1978:
85). Dakin (2001b:32833) and Stubbs (2011:2930) discuss the loss of PSUA *r or its
replacement by y in the SUA languages.
21. Classical Nahuatl atli and Pochutec at show the expected correspondences of
*paci and share the referent seed, but a Pipil cognate is not attested. Instead, the
concept of seed, as well as grain, pit, face, and eye, is conveyed by i:, the Pipil
reflex of PSUA *pusi eye. The same range of meanings is encountered in the cognates of
i: in NUA languages, but the reflexes of *pusi in most SUA languages designate only
eye (Stubbs 2011:#824, #1917).
22. Sierra Warih ihpac derives from pahc seed, pit, which shows the expected
phonological reflexes of PSUA *paci and retains the referent reconstructed for this
etymon. In fact, maize kernels and seeds for planting are two meanings excluded from
the semantic scope of pahc. These referents are conveyed by sun orla (combining sun
maize with the deverbal noun orla, derived from olan to shell corn; see section 3.3)
and ihtri, respectively. The prefix ih is attested in ihtri and ihkusri roasted corn on
the cob, as well as a variety of other words not related to maize, but its significance is
unclear (Miller 1996:27374).
23. Kaufman and Justeson (2009:226) suggest that Hopi paacama might be analyzed
as a compound of paa water and cama removed ashes, ashes being one source of the
lime that is used in the preparation of hominy. The Hopi word also could be related to
Numic verbs for to wash, e.g., Northern Paiute paca ~ baca (Liljeblad, Fowler, and
Powell 2012:363; see Stubbs 2011:#2487). The initial syllable pa ~ ba in the Numic
verbs reflects the PUA etymon for water, as does Hopi paa, the combining form of Hopi
paah water.
24. The Sonoran Yaqui word for sugarcane is yosana, a compound of yi ~ yri
mestizo and sna cane, literally ,mestizo cane. The first word also appears in River
Warih yrihom sugarcane, but the second element in the compound reflects ProtoTaracahitan *oma cane (Lionnet 1985:#15).
25. Tubar [honat] stubble is not cognate. It appears to be loan from a Tepiman
language, most likely Mountain Pima, e.g., Yepachi Pima hona stalk, trunk of a plant,
body.
26. Hill (2012:58) includes several Numic words for hay or grass in the same set as
Hopi s and the reflexes of SUA *sona (see Stubbs 2011:#1061), but these words
reflect Proto-Numic *soni, not PNUA *sooa.

2012

WILLIAM L. MERRILL

245

27. The noncombining form of the Comanche word for tree trunk is owoora, also
without a medial glottal stop. It is possible that haniwoora maize cob is a loan from
another Numic language, but the data are insufficient to reach a conclusion. Timbisha is
an unlikely source because its word for maize is maisi, from Spanish maz. No word for
maize cob is attested in Timbisha, nor in Northern Shoshone or Northern Paiute, the
only Numic languages that have words for maize that resemble Comanche hani maize
(see section 6.1).
28. Sierra Warih and Rarmuri have both lost the glottal stop in word-initial position (Miller 1996:39; Burgess 1970:51; Caballero 2008:65, 8083). I think that this loss
either took place independently or resulted from the interaction of speakers of the two
languages after the split of the Sierra and River dialects. However, the loss could have
occurred in ancestral Rarmuri-Warih, in which case the initial h in the River Warih
form would be epenthetic, but the reconstruction of *ona or *oona would not be
affected.
29. The Yaqui and Mayo words for maize cob, also presented in table 3, reflect antecedent *nawo, which could not be a reflex of either *ona or *oona. It could, however, derive by metathesis from won, the form attested in Sierra Warih. Although an
initial *w cannot be reconstructed for the antecedent form reflected in the Warih and
Rarmuri cognates, the word could have entered Cahitan as a loan from the dialect of
Warih that gave rise to Sierra Warih. Eudeve nhro maize cob could be a loan from
Yaqui-Mayo, but only the initial n and final o are regular correspondences (Stubbs
2011:#540, #546).
30. The Spanish word pinole is a loanword derived from Classical Nahuatl pinolli,
which apparently designated flour made from both maize and chia (Molina 1970:82r).
The Nahuatl noun probably derives from the PUA verb **pia to pulverize, but a reflex
of the verb is not attested in Nahuatl (Stubbs 2011:#1080).
31. Dakin (1982:#174) suggests that Classical Nahuatl kimiin mouse, rat may be
a deverbal noun deriving from *kumi, but a verbal form is not documented for Nahuatl.
32. The exact form of the noun reported in Lionnets work is komat. Lionnet
(1978:1920) analyzes Tubar [o] and [] as allophones of u and r, respectively. I change
the vowel and remove the bar from [] because the phoneme could also be represented as
l. In the Spanish-Tubar vocabulary list, he provides the gloss to eat maize for kumi
(1978:73), which is attested only in its present tense form kuminyt.
33. Sapir (1931:641) gives kummia as the Kaibab Southern Paiute form of this word,
which appears in Southern Ute as kmy (Givn 1979:126). These attestations suggest
that the antecedent form was trisyllabic, perhaps kumiya.
34. An exact phonological match for Hopi tawiya is Pipil tawiyal, but the Pipil word
labels dried shelled maize kernels. It derives from the verb tawiya to shell maize, a
form not attested in Hopi.
35. Bean and Saubel (1972:5758) provide the identification and the Cahuilla name
nekhi, which Seiler and Hioki (1979:127) recorded as nexi.
36. The Proto-Tepiman reconstructions in (23) are from Bascom (1965:#11a, #27,
#42), with two minor differences: he reconstructs stress on the initial syllables of all three
etyma and *gikai ~ *giki as the etymon for planting stick.
37. Cultural similarities between the Hopi and SUA societies have long intrigued
researchers. Recent studies focused on this theme are James (2000), Neurath (2005),
Gutirrez (2006), Secakuku (2006), Hays-Gilpin (2008), and Carot and Hers (2011).
38. A photograph of some of these hoes appears in Fewkes (1912: plate 76 [following
p. 146]). Secakuku (2006) and Carot and Hers (2011; cf. Washburn 1995:2022) propose
that Palatkapi may have been the Mesoamerica metropolis Teotihuacan.
39. I collected the Rarmuri word in the community of Rejogochi, and although I
heard it pronounced on numerous occasions, I never detected a medial glottal stop

246

ANTHROPOLOGICAL LINGUISTICS

54 NO. 3

(Merrill 1988:73). Brambila (1976:583), however, gives the form of this word as
warruwi ~ warruwa, which he glosses as legendary being.
40. For comparative analyses of Hopi, Cora, and Huichol cosmology and ritual, see
Neurath (2005) and Gutirrez (2006).
41. Relying on evidence distinct from that presented here, several scholars have
proposed this same area as a possible location of the homeland of the Southern UtoAztecans and even the Uto-Aztecans as a whole (Romney 1957; Miller 1983b:123; Fowler
1972b:110; Fowler 1983:242; cf. Carpenter, Snchez, and Villalpando 2002). Hill (2012:
65) mentions this possibility for the Proto-Uto-Aztecan speech community, but concludes
that the available evidence indicates a Mesoamerican homeland instead.
42. The scarlet macaw (Ara macao), native to the tropical lowlands of eastern and
southern Mexico and Central and South America, was imported into northern Mexico
and the American Southwest, with the earliest evidence for its presence in the region
dating to around AD 100 (1850 BP) (Somerville, Nelson, and Knudson 2010). Reflexes of
the PSUA *haro macaw may have served as generic term for large parrot in SUA
languages whose speakers were familiar with both species, as was the case in Classical
Nahuatl (see appendix 2, set 2).

References
Anderton, Alice J.
1988
The Language of the Kitanemuks of California. Ph.D. diss., University of
California, Los Angeles.
Bascom, Burton W.
1965
Proto-Tepiman. Ph.D. diss., University of Washington.
Bean, Lowell John
1978
Cahuilla. In Handbook of North American Indians. Vol. 8: California,
edited by Robert F. Heizer, 57587. Washington, D.C.: Smithsonian
Institution.
Bean, Lowell John, and Katherine S. Saubel
1972
Temalpakh: Cahuilla Indian Knowledge and Usage of Plants. Banning,
Calif.: Malki Museum Press.
Bean, Lowell John, and Florence C. Shipek
1978
Luiseo. In Handbook of North American Indians. Vol. 8: California,
edited by Robert F. Heizer, 55063. Washington, D.C.: Smithsonian
Institution.
Bean, Lowell John, and Charles R. Smith
1978a
Gabrielino. In Handbook of North American Indians. Vol. 8: California,
edited by Robert F. Heizer, 53849. Washington, D.C.: Smithsonian
Institution.
1978b
Serrano. In Handbook of North American Indians. Vol. 8: California,
edited by Robert F. Heizer, 57074. Washington, D.C.: Smithsonian
Institution.
1978c
Cupeo. In Handbook of North American Indians. Vol. 8: California,
edited by Robert F. Heizer, 58891. Washington, D.C.: Smithsonian
Institution.
Bellwood, Peter
1993
An Archaeologists View of Language Macrofamily Relationships. Bulletin
of the Indo-Pacific Prehistory Association 13:4660.
1997
Prehistoric Cultural Explanations for Widespread Language Families. In
Archaeology and Linguistics: Aboriginal Australia in Global Perspective,

2012

WILLIAM L. MERRILL

247

edited by Patrick McConvell and Nicholas Evans, 12334. Melbourne:

Oxford University Press.
2001
Early Agriculturalist Population Diasporas? Farming, Languages, and
Genes. Annual Review of Anthropology 30:181207.
2005
First Farmers: The Origins of Agricultural Societies. Oxford: Blackwell.
Bellwood, Peter, and Marc Oxenham
2008
The Expansions of Farming Societies and the Role of the Neolithic Demographic Transition. In The Neolithic Demographic Transition and Its
Consequences, edited by Jean-Pierre Bocquet-Appel and Ofer Bar-Yosef,
1334. Dordrecht: Springer.
Bellwood, Peter, and Colin Renfrew, eds.
2002
Examining the Farming/Language Dispersal Hypothesis. Cambridge:
McDonald Institute for Archaeological Research, University of Cambridge.
Blackburn, Thomas C., and Lowell John Bean
1978
Kitanemuk. In Handbook of North American Indians. Vol. 8: California,
edited by Robert F. Heizer, 56469. Washington, D.C.: Smithsonian
Institution.
Boas, Franz
1917
El dialecto mexicano de Pochutla, Oaxaca. International Journal of
American Linguistics 1:944.
Bradfield, Richard M.
1971
Changing Patterns of Hopi Agriculture. Royal Anthropological Institute
Occasional Paper 30. London: Royal Anthropological Institute.
Brambila, David
1976
Diccionario rarmuricastellano (tarahumar). Mexico City: La Obra
Nacional de la Buena Prensa.
Bright, William
1968
A Luiseo Dictionary. University of California Publications in Linguistics
51. Berkeley: University of California Press.
Brown, Cecil H.
2006
Prehistoric Chronology of the Common Bean in the New World: The Linguistic Evidence. American Anthropologist 108(3):50716.
2010a
Development of Agriculture in Prehistoric Mesoamerica: The Linguistic
Evidence. In Pre-Columbian Foodways: Interdisciplinary Approaches to
Food, Culture, and Markets in Ancient Mesoamerica, edited by John E.
Staller and Michael Carrasco, 71107. New York: Springer.
2010b
Lack of Linguistic Support for Proto-Uto-Aztecan at 8900 BP. Proceedings
of the National Academy of Sciences, U.S.A. 107:E34.
Bunte, Pamela A., and Robert J. Franklin
1987
From the Sands to the Mountain: Change and Persistence in a Southern
Paiute Community. Lincoln: University of Nebraska Press.
Burgess, Don H.
1970
Tarahumara Phonology (Rocoroibo Dialect). University of Texas at El
Paso, Studies in Language and Linguistics, 19691970:4566.
Brquez, Alberto, Angelina Martnez-Yrzar, Richard S. Felger, and David Yetman
1999
Vegetation and Habitat Diversity at the Southern Edge of the Sonoran
Desert. In Ecology of Sonoran Desert Plants and Plant Communities,
edited by Robert H. Robichaux, 3667. Tucson: University of Arizona
Press.

248

ANTHROPOLOGICAL LINGUISTICS

54 NO. 3

Caballero, Gabriela
2008
Choguita Rarmuri (Tarahumara) Phonology and Morphology. Ph.D.
diss., University of California, Berkeley.
2011
Behind the Mexican Mountains: Recent Developments and New Directions in Research on Uto-Aztecan Languages. Language and Linguistics
Compass 5(7):485504.
Campbell, Lyle
1985
The Pipil Language of El Salvador. Berlin: Mouton.
1997
American Indian Languages: The Historical Linguistics of Native America. Oxford: Oxford University Press.
2002
What Drives Linguistic Diversification and Language Spread? In Examining the Farming/Language Dispersal Hypothesis, edited by Peter
Bellwood and Colin Renfrew, 4963. Cambridge: MacDonald Institute for
Archaeological Research, University of Cambridge.
Campbell, Lyle, and Ronald W. Langacker
1978
Proto-Uto-Aztecan Vowels: Part III. International Journal of American
Linguistics 44(4):26279.
Carot, Patricia, and Marie-Areti Hers
2011
De Teotihuacan al can de Chaco: nueva perspectiva sobre las relaciones
entre Mesoamrica y el suroeste de los Estados Unidos. Anales del Instituto de Investigacines Estticas 33(98):553.
Carpenter, John P., Guadalupe Snchez de Carpenter, and Jonathan B. Mabry
2001
La arqueologa de los grupos yutoaztecas tempranos. In Avances y
balances de lenguas yutoaztecas: homenaje a Wick R. Miller, edited by
Jos L. Moctezuma Zamarrn and Jane H. Hill, 35973. Mexico City:
Instituto Nacional de Antropologa e Historia.
Carpenter, John P., Guadalupe Snchez, and Maria Elisa Villalpando
2002
Of Maize and Migration: Mode and Tempo in the Diffusion of Zea mays in
Northwest Mexico and the American Southwest. In Traditions, Transitions, and Technologies, edited by Sarah H. Schlanger, 24558. Boulder:
University Press of Colorado.
2005
The Late Archaic/Early Agricultural Period in Sonora, Mexico. In The
Late Archaic Across the Borderlands: From Foraging to Farming, edited
by Bradley J. Vierra, 1340. Austin: University of Texas Press.
Chamberlin, Ralph V.
1911
The Ethno-Botany of the Gosiute Indians. Proceedings of the Academy of
Natural Sciences of Philadelphia 63(1):2499.
Collard, Howard, and Elizabeth S. Collard
1962
Castellano-mayo, mayo-castellano. Mexico City: Instituto Lingstico de
Verano.
CONABIO [Comisin Nacional para el Conocimiento y Uso de la Biodiversidad ]
n.d.
AVESMX.NET: La red de conocimiento sobre las aves de Mxico.
http://avesmx.conabio.gob.mx/.
Cramaussel, Chantal
1998
Sistema de riego y espacio habitado: la lenta y azarosa gnesis de un
pueblo rural. In Historia y arte en un pueblo rural: San Bartolom (hoy
Valle de Allende, Chihuahua), edited by Clara Bargellini, 1789. Mexico
City: Instituto de Investigaciones Estticas, Universidad Nacional Autnoma de Mxico.
Crown, Patricia L.
1991
The Hohokam: Current Views of Prehistory and the Regional System. In
Chaco and Hohokam: Prehistoric Regional Systems in the American

2012

WILLIAM L. MERRILL

249

Southwest, edited by Patricia L. Crown and W. James Judge, 13557.

Santa Fe: School of American Research Press.
Dakin, Karen
1982
La evolucin fonolgica del proto-nhuatl. Mexico City: Universidad
Nacional Autnoma de Mxico.
2001a
Animals and Vegetables: Uto-Aztecan Noun Derivation, Semantic Classification, and Cultural History. In Historical Linguistics 1999: Selected
Papers from the Fourteenth International Conference on Historical Linguistics, Vancouver, 913 August 1999, edited by Laurel J. Brinton,
10517. Amsterdam: John Benjamins.
2001b
Isoglosas e innovaciones yutoaztecas. In Avances y balances de lenguas
yutoaztecas: homenaje a Wick R. Miller, edited by Jos Luis Moctezuma
Zamarrn and Jane H. Hill, 31343. Mexico City: Instituto Nacional de
Antropologa e Historia.
Dayley, Jon P.
1989
Tmpisa (Panamint) Shoshone Dictionary. University of California Publications in Linguistics 116. Berkeley: University of California Press.
Dedrick, John M., and Eugene H. Casad
1999
Sonora Yaqui Language Structures. Tucson: University of Arizona Press.
Diamond, Jared, and Peter Bellwood
2003
Farmers and Their Languages: The First Expansions. Science 300:597
603.
Doolittle, William E.
1980
Aboriginal Agricultural Development in the Valley of Sonora, Mexico.
Geographical Review 70(3):32842.
1984a
Settlements and the Development of Statelets in Sonora, Mexico.
Journal of Field Archaeology 11(1):1324.
1984b
Cabeza de Vacas Land of Maize: An Assessment of Its Agriculture. Journal of Historical Geography 10(3):24662.
1988
Pre-Hispanic Occupance in the Valley of Sonora, Mexico: Archaeological
Confirmation of Early Spanish Reports. University of Arizona Anthropological Papers 48. Tucson: University of Arizona Press.
2000
Cultivated Landscapes of Native North America. Oxford: Oxford University Press.
Downs, James F.
1966
The Significance of Environmental Manipulation in Great Basin Cultural
Development. In The Current Status of Anthropological Research in the
Great Basin, 1964, edited by Warren L. dAzevedo, 3956. Reno: Desert
Research Institute.
Elliott, Eric B.
1999
Dictionary of Rincn Luiseo. Ph.D. diss., University of California, San
Diego.
Estrada Fernndez, Zarina, Crescencio Buitimea Valenzuela, Adriana Elizabeth
Gurrola Camacho, Mara Elena Castillo Celaya, and Anabela Carln Flores
2004
Diccionario yaqui-espaol y textos: obra de preservacin lingstica.
Mexico City: Plaza y Valds; Hermosillo: Universidad de Sonora.
Felger, Richard S., and Elaine Joyal
1999
The Palms (Arecaceae) of Sonora, Mexico. Aliso 18(1):118.
Ferguson, T. J., and Chip Colwell-Chanthaphonh
2006
History Is In The Land: Multivocal Tribal Traditions in Arizonas San
Pedro Valley. Tucson: University of Arizona Press.

250

ANTHROPOLOGICAL LINGUISTICS

54 NO. 3

Fewkes, Jesse W.
1912
Casa Grande, Arizona. In Twenty-Eighth Annual Report of the Bureau of
American Ethnology for the Years 19061907, 25179. Washington, D.C.:
Government Printing Office.
Forde, C. Daryll
1931
Hopi Agriculture and Land Ownership. The Journal of the Royal Anthropological Institute of Great Britain and Ireland 61:357405.
Fowler, Catherine S.
1972a
Comparative Numic Ethnobiology. Ph.D. diss., University of Pittsburgh.
1972b
Some Ecological Clues to Proto-Numic Homelands. In Great Basin Cultural Ecology: A Symposium, edited by Don D. Fowler, 10521. Reno:
Desert Research Institute.
1983
Some Lexical Clues to Uto-Aztecan Prehistory. International Journal of
American Linguistics 49(3):22457.
1986
Subsistence. In Handbook of North American Indians. Vol. 11: Great
Basin, edited by Warren L. dAzevedo, 6497. Washington, D.C.: Smithsonian Institution.
1994
Corn, Beans, and Squash: Some Linguistic Perspectives from UtoAztecan. In Corn and Culture in the Prehistoric New World, edited by
Sissel Johannessen and Christine Ann Hastorf, 44567. Boulder, Colo.:
Westview Press.
1995
Some Notes on Ethnographic Subsistence Systems in Mojavean Environments in the Great Basin. Journal of Ethnobiology 15(1):99117.
Fowler, Catherine S., and Don D. Fowler
1981
The Southern Paiute: A.D. 14001776. In The Protohistoric Period in the
North American Southwest, A.D. 14501700, edited by David R. Wilcox
and W. Bruce Masse, 12962. Arizona State University Anthropological
Papers 24. [Tempe]: Arizona State University.
Gasser, Robert E., and Scott M. Kwiatkowski
1991
Regional Signatures of Hohokam Plant Use. Kiva 56(3):20726.
Givn, Talmy, ed.
1979
Ute Dictionary (Preliminary Edition). Ignacio, Colo.: Printed and distributed by Ute Press, the Southern Ute Tribe.
Gmez Lpez, Paula
1999
Huichol de San Andrs Cohamiata, Jalisco. Mexico City: El Colegio de
Mxico.
Griffen, William B.
1969
Culture Change and Shifting Populations in Central Northern Mexico.
University of Arizona Anthropological Papers 13. Tucson: University of
Arizona Press.
Grimes, Joseph E.
1980
Huichol Life Form Classification II: Plants. Anthropological Linguistics
22(6):26474.
Grimes, Joseph E., Pedro de la Cruz Avila, Jos Carrillo Vicente, Filiberto Daz, Romn
Daz, Antonio de la Rosa, and Toribio Rentera
1981
El huichol: apuntes sobre el lxico. Ithaca, N.Y.: Department of Modern
Languages and Linguistics, Cornell University.
Gutirrez, Arturo
2006
Mitologa y ritualidad: un acercamiento comparativo a los sistemas
religiosos hopi, huichol y cora. In Las vas del noroeste I: una macroregin
indgena americana, edited by Carlo Bonfiglioli, Arturo Gutirrez del
Angel, and Mara Eugenia Olavarra, 17187. Mexico City: Instituto de

2012

WILLIAM L. MERRILL

251

Investigaciones Antropolgicas, Universidad Nacional Autnoma de

Mxico.
Hack, John T.
1942
The Changing Physical Environment of the Hopi Indians of Arizona.
Papers of the Peabody Museum of American Archaeology and Ethnography, Vol. 35, No. 1. Cambridge, Mass.: Peabody Museum, Harvard
University.
Hammarstrm, Harald
2010
A Full-Scale Test of the Language Farming Dispersal Hypothesis. Diachronica 27(2):197213.
Harrington, John P.
1981
The John Peabody Harrington Papers. National Anthropological
Archives, Smithsonian Institution, Microfilm Edition, part 3, reels 102
105: Gabrielino. White Plains, N.Y.: Kraus International Publications.
Hays-Gilpin, Kelly
2008
All Roads Lead to Hopi. In Las vas del noroeste II: propuesta para una
perspectiva sistmica e interdisciplinaria, edited by Carlo Bonfiglioli,
Arturo Gutirrez, Marie-Areti Hers, and Mara Eugenia Olavarra, 6582.
Mexico City: Instituto de Investigaciones Antropolgicas, Universidad
Nacional Autnoma de Mxico.
Hill, Jane H.
2001a
Dating the Break-up of Southern Uto-Aztecan. In Avances y balances de
lenguas yutoaztecas: homenaje a Wick R. Miller, edited by Jos L.
Moctezuma Zamarrn and Jane H. Hill, 34557. Mexico City: Instituto
Nacional de Antropologa e Historia.
2001b
Proto-Uto-Aztecan: A Community of Cultivators in Central Mexico? American Anthropologist 103(4):91334.
2002a
Proto-Uto-Aztecan Cultivation and the Northern Devolution. In Examining the Farming/Language Dispersal Hypothesis, edited by Peter
Bellwood and Colin Renfrew, 33140. Cambridge: MacDonald Institute
for Archaeological Research, University of Cambridge.
2002b
Toward a Linguistic Prehistory of the Southwest: Azteco-Tanoan and
the Arrival of Maize Cultivation. Journal of Anthropological Research
58(4):45775.
2004
The Non-agricultural Origins of the Proto-Uto-Aztecan Vocabulary of
Maize Use. In Memorias del VII Encuentro Internacional de Lingstica
en el Noroeste, Vol. 2, edited by Mara del Carmen Mora Leyva and Rosa
Mara Ortiz Ciscomani, 5981. Hermosillo: Universidad de Sonora.
2005
A Grammar of Cupeo. University of California Publications in Linguistics 136. Berkeley: University of California Press.
2006
The Historical Linguistics of Maize Cultivation in Mesoamerica and North
America. In Histories of Maize: Multidisciplinary Approaches to the Prehistory, Linguistics, Biogeography, Domestication, and Evolution of
Maize, edited by John E. Staller, Robert H. Tykot, and Bruce F. Benz,
63145. Amsterdam: Elsevier Academic Press.
2008
Northern Uto-Aztecan and Kiowa-Tanoan: Evidence of Contact Between
the Proto-Languages? International Journal of American Linguistics
74(2):15588.
2010
New Evidence for a Mesoamerican Homeland for Proto-Uto-Aztecan. Proceedings of the National Academy of Sciences, U.S.A. 107:E33.
2011
Subgrouping in Uto-Aztecan. Language Dynamics and Change 1(2):241
78.

252
2012

ANTHROPOLOGICAL LINGUISTICS

54 NO. 3

Proto-Uto-Aztecan as a Mesoamerican Language. Ancient Mesoamerica

23(1):5768.
Hill, Jane H., and Rosinda Nolasquez
1973
Muluwetam: The First People: Cupeo Oral History and Language.
Banning, Calif.: Malki Museum Press.
Hilton, K. Simon
1959
Vocabulario tarahumara. Mexico City: Instituto Lingstico de Verano.
Hopi Dictionary Project
1998
Hopi Dictionary / Hopikwa Lavytutuveni: A Hopi-English Dictionary of
the Third Mesa Dialect. Tucson: University of Arizona Press.
igo Elas, Eduardo E.
2000
Guacamaya verde. In Las aves de Mxico en peligro de extincin, edited
by Gerardo Ceballos and Laura Mrquez Valdelamar, 21315. Mexico
City: Instituto de Ecologa, Universidad Nacional Autnoma de Mxico;
CONABIO; Fondo de Cultura Econmica.
James, Susan E.
2000
Some Aspects of the Aztec Religion in the Hopi Kachina Cult. Journal of
the Southwest 42(4):897926.
Jel, Judith
1978
The Yuman Word for Bean as a Clue to Prehistory. Journal of California
Anthropology, Papers in Linguistics 1:7792.
Joyal, Elaine
1995
An Ethnoecology of Sabal uresana Trealease (Arecaceae) in Sonora,
Mexico. Ph.D. diss., Arizona State University.
1996a
The Use of Sabal uresana (Arecaceae) and Other Palms in Sonora,
Mexico. Economic Botany 50(4):42945.
1996b
The Palm Has Its Time: An Ethnoecology of Sabal uresana in Sonora,
Mexico. Economic Botany 50(4):44662.
Kaplan, Lawrence, and Thomas F. Lynch
1999
Phaseolus (Fabaceae) in Archaeology: AMS Radiocarbon Dates and Their
Significance for Pre-Colombian Agriculture. Economic Botany 53(3):261
72.
Karttunen, Frances E.
1992
An Analytical Dictionary of Nahuatl. Norman: University of Oklahoma
Press.
Kaufman, Terrence
1981
Comparative Uto-Aztecan Phonology. MS.
Kaufman, Terrence, and John Justeson
2009
Historical Linguistics and Pre-Columbian Mesoamerica. Ancient Mesoamerica 20(2):22131.
Kavanagh, Thomas W.
2001
Comanche. In Handbook of North American Indians. Vol. 13: Plains,
edited by Raymond J. DeMallie, 886906. Washington, D.C.: Smithsonian
Institution.
Kelly, Isabel T.
1964
Southern Paiute Ethnography. University of Utah Anthropological
Papers 69. Salt Lake City: University of Utah Press.
Kelly, Isabel T., and Catherine S. Fowler
1986
Southern Paiute. In Handbook of North American Indians. Vol. 11: Great
Basin, edited by Warren L. dAzevedo, 36897. Washington, D.C.: Smithsonian Institution.

2012

WILLIAM L. MERRILL

253

Kindl, Olivia S.
2000
The Huichol Gourd Bowl as Microcosm. Journal of the Southwest 42(1):
3760.
Kohler, Timothy A., and Matt Glaude
2008
The Nature and Timing of the Neolithic Demographic Transition in the
North American Southwest. In The Neolithic Demographic Transition
and Its Consequences, edited by Jean-Pierre Bocquet-Appel and Ofer BarYosef, 81105. [Dordrecht]: Springer.
Koerper, Henry C., Henry C. Killingley, and R. E. Taylor
1985
The Little Ice Age and Coastal Southern California Human Economy.
Journal of California and Great Basin Anthropology 7(1):99103.
Kraft, Kraig H.
2009
The Domestication of the Chile Pepper, Capsicum annuum: Genetic,
Ecological, and Anthropogenic Patterns of Genetic Diversity. Ph.D. diss.,
University of California, Davis.
Lawlor, Elizabeth J.
1995
Archaeological Site-Formation Processes Affecting Plant Remains in the
Mojave Desert. Ph.D. diss., University of California, Riverside.
Lawton, Harry W., and Lowell J. Bean
1968
A Preliminary Reconstruction of Aboriginal Agricultural Technology
among the Cahuilla. Indian Historian 1(5):1824, 29.
LeBlanc, Steven A.
2008
The Case for an Early Farmer Migration into the Greater American
Southwest. In Archaeology without Borders: Contact, Commerce, and
Change in the U.S. Southwest and Northwestern Mexico, edited by Laurie
D. Webster and Maxine E. McBrinn, 10742. Boulder: University Press of
Colorado.
Liljeblad, Sven S., Catherine S. Fowler, and Glenda Powell
2012
The Northern Paiute-Bannock Dictionary. Salt Lake City: University of
Utah Press.
Lionnet, Andrs
1978
El idioma tubar y los tubares. Mexico City: Universidad Iberoamericana.
1985
Relaciones internas de la rama sonorense. Amerindia 10:2658.
Lombardo, Natal
1702
Arte de la lengua teguima vulgarmente llamada pata. Mexico City:
Miguel de Ribera.
Lowie, Robert H.
1924
Notes on Shoshonean Ethnography. Anthropological Papers of the American Museum of Natural History 20(3):185314. New York: American
Museum Press.
Lyneis, Margaret M.
1995
The Virgin Anasazi: Far Western Pueblos. Journal of World Prehistory
9(2):199241.
Mabry, Jonathan B.
2005
Changing Knowledge and Ideas about the First Farmers in Southeastern
Arizona. In The Late Archaic across the Borderlands: From Foraging to
Farming, edited by Bradley J. Vierra, 4183. Austin: University of Texas
Press.
Mabry, Jonathan B., John P. Carpenter, and Guadalupe Sanchez
2008
Archaeological Models of Early Uto-Aztecan Prehistory in the ArizonaSonora Borderlands. In Archaeology without Borders: Contact, Commerce, and Change in the U.S. Southwest and Northwestern Mexico,

254

ANTHROPOLOGICAL LINGUISTICS

54 NO. 3

edited by Laurie D. Webster and Maxine E. McBrinn, 15583. Boulder:

University Press of Colorado.
McIntosh, John B., and Joseph E. Grimes
1954
Niuqui Iquisicayari (Vocabulario huichol-castellano, castellano-huichol).
Mexico City: Instituto Lingstico de Verano.
McMahon, Ambrosio, and Maria Aitn de McMahon
1959
Vocabulario cora. Mexico City: Instituto Lingstico de Verano.
Madsen, David B., and Steven R. Simms
1998
The Fremont Complex: A Behavioral Perspective. Journal of World Prehistory 12(3):255336.
Manaster Ramer, Alexis
1992
A Northern Uto-Aztecan Sound Law: *c y. International Journal of
American Linguistics 58(3):25168.
Mathiot, Madeleine
1973
A Dictionary of Papago Usage. 2 vols. Bloomington: Indiana University.
Matson, Richard G.
1991
The Origins of Southwestern Agriculture. Tucson: University of Arizona
Press.
2002
The Spread of Maize Agriculture into the U.S. Southwest. In Examining
the Farming/Language Dispersal Hypothesis, edited by Peter Bellwood
and Colin Renfrew, 34156. Cambridge: MacDonald Institute for Archaeological Research, University of Cambridge.
Matthews, John A., and Keith R. Briffa
2005
The Little Ice Age: Re-Evaluation of an Evolving Concept. Geografiska
Annaler, Series A, Physical Geography 87(1):1736.
Medina Murillo, Ana A.
2002
Diccionario morfolgico: formacin de palabras en el guarijo. M.A. thesis,
Universidad de Sonora.
Merrill, William L.
1988
Rarmuri Souls: Knowledge and Social Process in Northern Mexico.
Washington, D.C.: Smithsonian Institution Press.
2007
La obra lingstica del padre Matthus Steffel, S.J. In Desde los confines
de los imperios ibricos: los jesuitas de habla alemana en las misiones
americanas, edited by Karl Kohut and Mara Cristina Torales Pacheco,
40939. Madrid: Iberoamericana; Frankfurt: Vervuert Verlag.
forthcoming The Genetic Unity of Southern Uto-Aztecan. Language Dynamics and
Change.
Merrill, William L., Robert J. Hard, Jonathan B. Mabry, Gayle J. Fritz, Karen R.
Adams, John R. Roney, and A. C. MacWilliams
2009
The Diffusion of Maize to the Southwestern United States and Its Impact.
Proceedings of the National Academy of Sciences, U.S.A. 106:2101926.
2010
Reply to Hill and Brown: Maize and Uto-Aztecan Cultural History. Proceedings of the National Academy of Sciences, U.S.A. 107:E35E36.
Miller, Wick R.
1966
Anthropological Linguistics in the Great Basin. In The Current Status of
Anthropological Research in the Great Basin, 1964, edited by Warren L.
dAzevedo, 75112. Reno: Desert Research Institute.
1967
Uto-Aztecan Cognate Sets. University of California Publications in Linguistics 48. Berkeley: University of California Press.
1972
Newe Natekwinappeh: Shoshoni Stories and Dictionary. University of
Utah Anthropological Papers 94. Salt Lake City: University of Utah Press.
1983a
A Note on Extinct Languages of Northwest Mexico of Supposed Uto-

2012

WILLIAM L. MERRILL

255

Aztecan Affiliation. International Journal of American Linguistics 49(3):

32834.
1983b
Uto-Aztecan Languages. In Handbook of North American Indians. Vol. 10:
Southwest, edited by Alfonso Ortiz, 11324. Washington, D.C.: Smithsonian Institution.
1993
Guarijo de Arechuyvo, Chihuahua. Mexico City: El Colegio de Mxico.
1996
Guarijo: gramtica, textos y vocabulario. Mexico City: Universidad
Nacional Autnoma de Mxico.
Molina, Alonso de
1970
Vocabulario en lengua castellana y mexicana y mexicana y castellana.
Edited by Miguel Len-Portilla. Mexico City: Editorial Porra. (Facsimile
edition of 1571 original.)
Molina, Felipe S., Herminia Valenzuela, and David L. Shaul
1999
Yoeme-English, English-Yoeme Dictionary, with a Comprehensive Grammar of Yoeme Language. New York: Hippocrene Books.
Munro, Pamela
1973
Proto-Uto-Aztecan *w: One Source for Luiseo . International Journal of
American Linguistics 39(3):13536.
Munro, Pamela, and William E. Mace
1995
A New Tbatulabal Dictionary. MS.
Muoz, L. Carmenza, Myriam C. Duque, Daniel G. Debouck, and Matthew W. Blair
2006
Taxonomy of Tepary Bean and Wild Relatives as Determined by Amplified
Fragment Length Polymorphism (AFLP) Markers. Crop Science 46(4):
174454.
Nabhan, Gary P., and Richard S. Felger
1978
Teparies in Southwestern North America. Economic Botany 32(1):319.
Neurath, Johannes
2005
Cosmogonic Myths, Ritual Groups, and Initiation: Toward a New Comparative Ethnology of the Gran Nayar and the Southwest of the U.S.
Journal of the Southwest 47(4):571614.
Oliver, Symmes
1962
Ecology and Cultural Continuity as Contributing Factors in the Social
Organization of the Plains Indians. University of California Publications
in American Archaeology and Ethnology 48(1):190. Berkeley: University
of California Press.
Ortega, Jos de
1860
Vocabulario en lengua castellana y cora. Boletn de la Sociedad Mexicana
de Geografa y Estadstica, 1a. poca, 8:561605. (Originally published in
1732.)
Pennington, Campbell W.
1963
The Tarahumar of Mexico: Their Environment and Material Culture. Salt
Lake City: University of Utah Press.
Pennington, Campbell W., ed.
1979
The Pima Bajo of Central Sonora, Mexico. Vol. 2: Vocabulario en la
Lengua Nvome. Salt Lake City: University of Utah Press.
1981
Arte y vocabulario de la lengua dohema, heve o eudeva. Mexico City: Universidad Nacional Autnoma de Mxico.
Pimentel, Francisco
1863
Vocabulario manual de la lengua pata. Boletn de la Sociedad Mexicana
de Geografa y Estadstica, 1a poca 10:287313.
Piperno, Dolores R.
2011
The Origins of Plant Cultivation and Domestication in the New World

256

ANTHROPOLOGICAL LINGUISTICS

54 NO. 3

Tropics: Patterns, Process, and New Developments. Current Anthropology 52(S4):S453S470.

Press, Margaret L.
1979
Chemehuevi: A Grammar and Lexicon. University of California Publications in Linguistics 92. Berkeley: University of California Press.
Preuss, Konrad Theodor
1934
Wrterbuch Deutsch-Cora. International Journal of American Linguistics
8(2):81102.
Rankin, Robert L.
2006
Siouan Tribal Contacts and Dispersions Evidenced in the Terminology
for Maize and Other Cultigens. In Histories of Maize: Multidisciplinary
Approaches to the Prehistory, Linguistics, Biogeography, Domestication,
and Evolution of Maize, edited by John E. Staller, Robert H. Tykot, and
Bruce F. Benz, 56375. Amsterdam: Elsevier Academic Press.
Rea, Amadeo M.
1997
At the Deserts Green Edge: An Ethnobotany of the Gila River Pima.
Tucson: University of Arizona Press.
Renfrew, Colin
1987
Archaeology and Language: The Puzzle of Indo-European Origins.
London: Jonathan Cape.
1992
Archaeology, Genetics and Linguistic Diversity. Man 27(3):44578.
Renfrew, Colin, David W. Anthony, Bernard Wailes, Philip Baldi, Graeme Barker,
Robert Coleman, Marija Gimbutas, Even Neustupny, and Andrew Sherratt
1988
Archaeology and Language: The Puzzle of Indo-European Origins: A CA
Book Review. Current Anthropology 29(3):43768.
Rinaldini, Benito
1994
Arte de la lengua tepeguana, con vocabulario, confessionario, y catechismo. Edited by Javier Guerrero Romero. Mexico City: Consejo Nacional
para la Cultura y las Artes; [Durango]: Gobierno del Estado de Durango.
(Facsimile edition of 1743 original.)
Robichaux, Robert H., and David A. Yetman, eds.
2000
The Tropical Deciduous Forest of Alamos: Biodiversity of a Threatened
Ecosystem in Mexico. Tucson: University of Arizona Press.
Robinson, Lila W., and James Armagost
1990
Comanche Dictionary and Grammar. Dallas: Summer Institute of Linguistics; [Arlington]: University of Texas at Arlington.
Romney, A. Kimball
1957
The Genetic Model and Uto-Aztecan Time Perspective. Davidson Journal
of Anthropology 3(2):3541.
Sapir, Edward
1931
Southern Paiute Dictionary. Proceedings of the American Academy of
Arts and Sciences 65(3):539730.
Saxton, Dean, Lucille Saxton, and Susie Enos
1983
Dictionary Tohono Oodham/Pima to English, English to Tohono
Oodham/Pima. 2d ed., rev. and expanded. Edited by R. L. Cherry.
Tucson: University of Arizona Press.
Secakuku, Ferrell H.
2006
Hopi and Quetzalcoatl: Is There a Connection? M.A. thesis, Northern
Arizona University.
Seiler, Hansjakob, and Kojiro Hioki
1979
Cahuilla Dictionary. Banning, Calif.: Malki Museum Press.

2012

WILLIAM L. MERRILL

257

Shaul, David L.
1994
A Sketch of the Structure of Oob Nook (Mountain Pima). Anthropological
Linguistics 36(3):277365.
Shaul, David L., and Jane H. Hill
1998
Tepimans, Yumans, and Other Hohokam. American Antiquity 63(3):375
96.
Shimkin, Demitri B.
1986
The Introduction of the Horse. In Handbook of North American Indians.
Vol. 11: Great Basin, edited by Warren L. dAzevedo, 51724. Washington, D.C.: Smithsonian Institution.
Smith, Bruce D.
1997a
The Initial Domestication of Cucurbita pepo in the Americas 10,000 Years
Ago. Science 276:93234.
1997b
Reconsidering the Ocampo Caves and the Era of Incipient Cultivation in
Mesoamerica. Latin American Antiquity 8(4):34283.
2001a
Documenting Plant Domestication: The Consilience of Biological and
Archaeological Approaches. Proceedings of the National Academy of
Sciences, U.S.A. 98:132426.
2001b
Low-Level Food Production. Journal of Archaeological Research 9(1):143.
2011
General Patterns of Niche Construction and the Management of Wild
Plant and Animal Resources by Small-scale Pre-industrial Societies.
Philosophical Transactions of the Royal Society B: Biological Sciences
366(1566):83648.
Smith, Charlene G.
1973
Sel, a Major Vegetal Component of the Aboriginal Hualapai Diet.
Plateau 45(3):10210.
Smith, Charles R.
1978
Tubatulabal. In Handbook of North American Indians. Vol. 8: California,
edited by Robert F. Heizer, 43745. Washington, D.C.: Smithsonian
Institution.
Somerville, Andrew D., Ben A. Nelson, and Kelly J. Knudson
2010
Isotopic Investigation of Pre-Hispanic Macaw Breeding in Northwest
Mexico. Journal of Anthropological Archaeology 29(1):12535.
Steffel, Matthus
1809
Tarahumarisches Wrterbuch, nebst einigen Nachrichten von den Sitten
und Gebruchen der Tarahumaren, in Neu-Biscaya, in der Audiencia
Guadalaxara im Vice-Knigreiche Alt-Mexico, oder Neu-Spanien. In
Nachrichten von verschiedenen Lndern des Spanischen Amerika, aus
eigenhndigen Aufstzen einiger Missionare der Gesellschaft Jesu, Vol. 1,
edited by Christoph Gottlieb von Murr, 293374. Halle: Joh. Christian
Hendel.
Steward, Julian H.
1933
Ethnography of the Owens Valley Paiute. University of California Publications in American Archaeology and Ethnology 33(3), 233350. Berkeley:
University of California Press.
1938
Basin-Plateau Aboriginal Sociopolitical Groups. Bureau of American
Ethnology Bulletin 120. Washington, D.C.: Government Printing Office.
1941
Cultural Element Distributions, XIII: Nevada Shoshone. University of
California Anthropological Records 4(2), 209360. Berkeley: University of
California Press.
Stewart, Kenneth M.
1968
A Brief History of the Chemehuevi Indians. Kiva 34(1):927.

258

ANTHROPOLOGICAL LINGUISTICS

54 NO. 3

Stewart, Omer C.
1942
Culture Element Distributions XVIII: UteSouthern Paiute. University of
California Anthropological Records 6(4), 231356. Berkeley: University of
California Press.
Stine, Scott
2004
Climate Change in Wildland Management: Taking the Long View. In
Proceedings of the Sierra Nevada Science Symposium: Science for
Management and Conservation, edited by Dennis D. Murphy and Peter A.
Stine, 5155. Albany, Calif.: U.S. Department of Agriculture, Forest Service, Pacific Southwest Research Station.
Stoffle, Richard W., and Henry F. Dobyns
1983
Nvagant: Nevada Indians Comment on the Intermountain Power Project. Reno: Nevada State Office of the Bureau of Land Management.
Stoffle, Richard W., and M. Nieves Zedeo
2001
Historical Memory and Ethnographic Perspectives on the Southern
Paiute Homeland. Journal of California and Great Basin Anthropology
23(2):22948.
Stubbs, Brian D.
2000
The Comparative Value of Tubar in Uto-Aztecan. In Uto-Aztecan: Structural, Temporal, and Geographic Perspectives: Papers in Memory of Wick
R. Miller by the Friends of Uto-Aztecan, edited by Eugene H. Casad and
Thomas L. Willett, 35769. Hermosillo: Universidad de Sonora.
2003
New Sets Yield New Perspectives for Uto-Aztecan Reconstructions. In
Studies in Uto-Aztecan, edited by Luis M. Barragan and Jason D.
Haugen, 120. MIT Working Papers in Endangered and Less Familiar
Languages 5. Cambridge, Mass.: Department of Linguistics, Massachusetts Insitute of Technology.
2011
Uto-Aztecan: A Comparative Vocabulary. Flower Mound, Texas: Shumway Family History Services; Blanding, Utah: Rocky Mountain Books and
Productions.
Teague, Lynn S.
1993
Prehistory and the Traditions of the OOdham and Hopi. Kiva 58(4):435
54.
Tewksbury, Joshua J., Gary P. Nabhan, Donald Norman, Humberto Suzn, John
Tuxill, and Jim Donovan
1999
In Situ Conservation of Wild Chiles and Their Biotic Associates. Conservation Biology 13(1):98107.
Valias Coalla, Leopoldo
2000
Lo que la lingstica yutoazteca podra aportar en la reconstruccin
histrica del norte de Mxico. In Nmadas y sedentarios en el norte de
Mxico: homenaje a Beatriz Braniff, edited by Marie-Areti Hers, Jos L.
Mirafuentes, Mara de los Dolores Soto, and Miguel Vallebueno,175205.
Mexico City: Universidad Nacional Autnoma de Mxico.
Voegelin, C. F., F. M. Voegelin, and Kenneth Hale
1962
Typological and Comparative Grammer of Uto-Aztecan I: Phonology.
Indiana University Publications in Anthropology and Linguistics Memoir
17. Supplement to International Journal of American Linguistics 28(1).
Baltimore: Waverly Press.
Voegelin, Erminie W.
1938
Tbatulabal Ethnography. University of California Anthropological Records 2(1), 184. Berkeley: University of California Press.

2012

WILLIAM L. MERRILL

259

Washburn, Dorothy K.
1995
Living in Balance: The Universe of the Hopi, Zuni, Navajo, and Apache.
Philadelphia: University of Pennsylvania Museum of Archaeology and
Anthropology.
Webster, Laurie D., and Micah Lomaomvaya
2004
Textiles, Baskets, and Hopi Cultural Identity. In Identity, Feasting, and
the Archaeology of the Greater Southwest, edited by Barbara J. Mills,
7492. Boulder: University Press of Colorado.
Weigand, Phil C.
1992
Ensayos sobre el Gran Nayar: entre coras, huicholes y tepehuanos. Mexico
City: Centro de Estudios Mexicanos y Centroamericanos de la Embajada
de Francia en Mxico and Instituto Nacional Indigenista; Zamora: El
Colegio de Michoacn.
Weigand, Phil C., and Acelia Garca de Weigand
2000
Huichol Society before the Arrival of the Spanish. Journal of the Southwest 42(1):1236.
West, Robert C.
1949
The Mining Community in Northern New Spain: The Parral Mining District. Ibero-Americana 30. Berkeley: University of California Press.
Whiting, Alfred F.
1966
Ethnobotany of the Hopi. Flagstaff: Museum of Northern Arizona.
Wichmann, Sren
2002
Contextualizing Proto-languages, Homelands and Distant Genetic Relationship: Some Reflections on the Comparative Method from a Mesoamerican Perspective. In Examining the Farming/Language Dispersal
Hypothesis, edited by Peter Bellwood and Colin Renfrew, 32129. Cambridge: MacDonald Institute for Archaeological Research, University of
Cambridge.
Wichmann, Sren, Andr Mller, and Viveka Velupillai
2010
Homelands of the Worlds Language Families: A Quantitative Approach.
Diachronica 27(2):24776.
Wilcox, David R., Phil C. Weigand, J. Scott Wood, and Jerry B. Howard
2008
Ancient Cultural Interplay of the American Southwest in the Mexican
Northwest. Journal of the Southwest 50(2):103206.
Wilde, James D., and Deborah E. Newman
1989
Late Archaic Corn in the Eastern Great Basin. American Anthropologist
91(3):71220.
Winter, Joseph C., and Patrick F. Hogan
1986
Plant Husbandry in the Great Basin and Adjacent Northern Colorado
Plateau. In Anthropology of the Desert West: Essays in Honor of Jesse D.
Jennings, edited by Carol J. Condie and Don D. Fowler, 11744. University of Utah Anthropological Papers 110. Salt Lake City: University of
Utah Press.
Wolff, Hans
1950
Comparative Siouan III. International Journal of American Linguistics
16(4):16878.
Yetman, David, and Richard Felger
2002
Ethnoflora of the Guarijos. In The Guarijos of the Sierra Madre: Hidden
People of Northwestern Mexico, by David Yetman, 174230. Albuquerque:
University of New Mexico Press.

260

ANTHROPOLOGICAL LINGUISTICS

54 NO. 3

Yetman, David, and Thomas R. Van Devender

2002
Mayo Ethnobotany: Land, History, and Traditional Knowledge in Northwest Mexico. Berkeley: University of California Press.
Zigmond, Maurice L.
1941
Ethnobotanical Studies among California and Great Basin Shoshoneans.
Ph.D. diss., Yale University.
1978
Kawaiisu. In Handbook of North American Indians. Vol. 8: California,
edited by Robert F. Heizer, 398411. Washington, D.C.: Smithsonian
Institution.
Zigmond, Maurice L., Curtis G. Booth, and Pamela Munro
1991
Kawaiisu: A Grammar and Dictionary with Texts. University of California
Publications in Linguistics 119. Berkeley: University of California Press.