A Computer-Generated Dictionary of Proto-Algonquian

John Steckley teaches human studies at Hunter College in Toronto.

A protolanguage is a hypothetical language that is proposed as being
ancestral to related known languages. Reconstructing the forms of words
you have neither seen nor heard is a laborious process. John Hewson
begins this book by stating that “[a] dictionary of a protolanguage is
necessarily the product of a generation of scholars, rather than of a
single individual.”

The first of the proto-Algonquian generation was Leonard Bloomfield,
who based his work on four languages: Ojibwa, Cree, Menomini, and Fox.
Next came Hockett and Siebert, who paved the way for George Aubin’s A
Proto-Algonquian Dictionary (1975). Hewson’s major claim for
surpassing Aubin comes from the innovative computer application to
reconstruction that he has been engaged in since the early 1970s. Hewson
has come up with 4066 entries, almost two times Aubin’s 2294.

That Hewson developed these entries almost exclusively from
Bloomfield’s original four languages is both the strength and the
limitation of this work. The strength lies in its thoroughness, a
relatively achievable goal when dealing with a narrow language base. And
its limitation? The Algonquian language family can be divided into
Central (where the four languages come from), Eastern (most closely
related to Central), and the more distant Western. A reconstruction rule
of thumb is that the broader your language base, historically and
geographically, the further back you can go. When the four languages
don’t adequately provide the examples he requires, Hewson occasionally
looks to Eastern languages, most notably Micmac (the Algonquian language
he is most familiar with), to fill the gap. Unlike Aubin, he does not
use examples from any of the Western languages, such as Blackfoot or
Cheyenne. Thus it can be said that Hewson delivers solid Central/Eastern
Algonquian. It is up to the next generation to provide us with
proto-Algonquian that is as well grounded.