Journal of the Southwest

1. Introduction

Linguistic data have great potential for contributing in unique ways to our understanding of prehistory, both regionally and globally. In the American Southwest, there are several notable examples of linguistic research contributing to archaeological reconstructions of cultural history (e.g., Davis 1959; Hill 2002, 2008; Merrill et al. 2009; Ortman, 2012; Shaul 2014; McNeil and Shaul 2018; Ortman and McNeil 2018). In particular, linguistic data have played an important role in the study of Puebloan prehistory. For example, Davis (1959) used the proportion of shared cognates determined through analysis of sound correspondences, phonetic similarity, and semantic affinity to estimate the historical relationships within the Kiowa-Tanoan and Keresan language families, and to generate glottochronological estimates of language divergence (cf. also Hale and Harris 1979). More recently, following Hale (1967) and others (Trager 1942; Davis 1979, 1989), Ortman (2012) examined sound correspondences for vowels and consonants among Kiowa-Tanoan languages. Using the comparative method, Ortman identifies phonological innovations shared among sets of languages. He then uses the pattern of shared innovations to estimate historical relationships among Kiowa-Tanoan [End Page 255] languages, and the sequence of protolanguage divergence. Ortman (2012) also uses shared cognates to reconstruct protolanguage terms for a large set of items including plants, animals, cultigens, and items of material culture that are represented and well dated in the archaeological record. These reconstructed terms were used to date protolanguages within the language family, and to locate their geographic homeland based on current and historical distributions of plant and animal species (also see Ortman and McNeil 2018).

In the present research we examine the historical relationships among Kiowa-Tanoan languages using tree-like diagrams generated from phylogenetic analysis of phonological data, as well as measures of linguistic dissimilarity derived from lexical data. We estimate the timing of branching events marking linguistic divergence within the language tree by employing an alternative to glottochronology. Following Ortman (2012), we also estimate the timing of language divergence within the Kiowa-Tanoan language family through a qualitative analysis of shared cognates using a modified "words and things" approach. We also use lexical and cognate data to delineate the Kiowa-Tanoan and Tanoan homelands. Our intention in this paper is to contribute to the study of Kiowa-Tanoan prehistory by presenting chronological and geographic contexts based on linguistic data that will inform archaeological inquiry.

The eleven tables appear together after the text of this paper.

2. The Kiowa-Tanoan Language Family

The Kiowa-Tanoan language family consists of four primary, undisputed branches with languages that are still spoken today: Kiowa (K), Tewa, Tiwa, and Towa (To). Linguists further divide Tewa into two substantially distinct varieties, Arizona Tewa (AT) and Rio Grande Tewa (RGT), the latter comprising five extant dialects,¹ and Tiwa into three main varieties, Taos Northern Tiwa (Ta), Picuris Northern Tiwa (Pi), and Southern Tiwa (ST), with the last consisting of three extant dialects.² Linguistic specialists therefore tend to treat the family as consisting of seven extant and welldocumented languages, two of which have notable dialect variation (cf. Sutton 2014; Kroskrity 1993; Trager 1967), although much of the literature nominally considers only the four main branches as distinctive, with all variation within Tewa and Tiwa being treated as "dialects." The distinction is one of degree of detail, not of kind, but should be noted, particularly in consideration of time depth of linguistic diversification. [End Page 256]

The hyphenated name of the language family reflects both a culturalgeographic distinction and the history of scholarship. The Kiowa language is spoken by the Kiowa people, who historically migrated through the Central and Southern Plains and currently reside in Oklahoma. The Tanoan branches, Tewa, Tiwa, and Towa, are currently spoken across eleven Pueblo communities in the Rio Grande Valley of New Mexico, as well as one community in Arizona and one in El Paso, Texas (figure 1). Another Tanoan language, Piro, which is no longer spoken, was minimally documented by surveyors in the late 1800s in a few place names, two overlapping word lists (approximately...