Another Look at Siouan and Iroquoian

Abstract

THAT there is a remote relationship between and speaking peoples of North America is not a new hypothesis. Robert G. Latham implied something of sort in 1845 when he noted affinities between certain languages which included the Iroquois tongues, Cherokee, and Catawba (Latham 1846:44). Catawba's place in family was later established. Lewis H. Morgan, probably first systematic observer to take a long and objective look at both groups, wrote that the Iroquois were, as there are reasons for believing, an early offshoot, and one of advanced bands of great Dakota stock, who first made their way eastward to valley of St. Lawrence ... and afterwards into lake region of Central New York . . . (Morgan 1871:150-151). Considerably less was implied by Edward Sapir's inclusion of both language families in his Hokan-Siouan stock (Sapir 1929), since he gave no more reason to believe that and were more closely related to each other than either was to, say, Keresan or Hokan. Instead, he grouped most closely with Yuchi (see subsequently Haas 1951), and these two then with Muskogian and Natchez to form a so-called eastern group of stock (see subsequently Haas 1960). was grouped most closely with Caddoan. But in any case HokanSiouan as a whole was among more tenuous of Sapir's hunches. Shortly after Sapir's publication, an article titled Siouan and Iroquoian appeared in International Journal of American Linguistics, written by one Louis Allen (Allen 1931). Purporting to give evidence of a relationship between two language families, this article has been given little serious attention. It is partly that data and methodology were rather disorganized, but impact of an opening statement by author must also be considered: writer's persistence in attempt (to establish a relationship) was due, perhaps, more than to anything else, to evident conviction with which a Wyandot Indian assured him that when he went to an Indian school in Indiana, along with boys from several Indian stocks, he felt by far most at home with Sioux, as they seemed to be same sort of people as Wyandots, and words of their language seemed much less unfamiliar to him than those of other languages. The apparent irrelevance of this observation must have suggested to many readers that little credence need be given to article as a whole. In what follows I am going to try to vindicate Allen's conclusion, if not his motivation. What I will present was come upon by accident, as a result of using Winnebago and Dakota informants in field methods courses at Georgetown University and University of California. Similarities between these two languages and language Seneca became in-