"A Preliminary Study of Raven Language Among the Eskimos."
J. Whitestone, Anthopological Studies of the North (1939).

The ravens of Alaska sounded their Kwakiutl tok different
from the language of the territorial groups,
so that the traditional and observed beliefs have divisions characteristic
of Haida dialects. Kwakiutl spoken from the coast appeared between
Tsimshian southeast of these human correspondences and represent
a distinct world between ravens and humans.

Tlingit sounded though ravens spoke of divisions
between the Kwakiutl tok and tlik populations in territory indigenous
to Haida dialects. Coevolution proves that humans and ravens corresponded
and appeared during the division especially when observed,
and characteristically would tok dialects with traditional Alaskans while
the human is represented from another correspondence, and that
Tsimshian languages are from the ravens.

Characteristically, the coevolution of Kwakiutl language
and other dialects of human language along geographically traditional Haida
are difficult, especially as indigenous ravens for many years
spoke as humans. Distinct populations would appear between these groups,
and their tok proves that other representative dialects sound
like that of ravens.

Concordance: “During the years I spent kayaking along the coast of British Columbia and Southeast Alaska, I observed that the local raven populations spoke in distinct dialects. The divisions between these dialects appeared to correspond to the traditional geographic divisions between the indigenous human language groups. Ravens from Kwakiutl, Tsimshian, Haida, and Tlingit territory sounded different from one another, especially in their characteristic ‘tok’ and ‘tlik.’ I believe that this correspondence between human language and raven language represents coevolution rather than coincidence, though this would be difficult to prove.” G. Dyson. In, J. Brockman, Editor, What We Believe But Cannot Prove. New York, 2006.