[lg policy] A Linguist Who Worked on 'Avatar' Knows His Na'vi

Harold Schiffman haroldfs at GMAIL.COM
Fri Feb 12 15:16:42 UTC 2010

A Linguist Who Worked on 'Avatar' Knows His Na'vi

By Andrea Fuller

When Paul R. Frommer saw an e-mail message saying that the director
James Cameron sought a linguist to develop a language for his "Project
880," Mr. Frommer jumped at the opportunity. Nearly five years later,
that project has become the world's highest-grossing film, Avatar. Mr.
Frommer, a professor of clinical management commuication at the
University of Southern California's Marshall School of Business who
holds a Ph.D. in linguistics from USC, created the Na'vi language used
by the blue aliens on the planet Pandora in the movie.

Q. What was so appealing about the project to you as a linguist?

A. Doing this kind of work as an academic is not going to advance your
research reputation. It's not going to result in publications in
peer-reviewed journals. But it just may push the world forward in the
way it's turning on young people to the wonders of language.

Q. How did you go about creating the Na'vi language?

A. The first thing is to try to nail down the sound system or, to be
more technical, the phonetics and the phonology. You want to determine
what sounds are in the language, and just as importantly what sounds
are not in the language. ... I also included some interesting
combinations of sounds. You could have a word that begins with "fng."
I excluded certain familiar sounds. There's no "b," "d," or "g" sound.
There's no "sh" sound. I think I came up with a pretty interesting
collection of sounds. Then you start talking about pronunciation
rules. How does one sound change into another under certain
circumstances? Once that's determined, then you can be thinking about
how to build the words, morphology, and how to put them together in
phrases and sentences, which is syntax.

Q. Did you model this on any language?

Related ContentAnd the Academy Award Goes to ... a Computer Scientist
A. Not on any particular language, no. I've studied bits and pieces of
maybe 15 or 16 languages. If you look at Na'vi, you may see this
particular structure reminds you of something in Persian, or something
in Chinese, or something in Hebrew.

Q. You have referred to Klingon, the language of the warrior race in
Star Trek, as the "gold standard" for alien languages. Have you seen
any made-up languages that are particularly bad?

A. The work that people respect the most is work done by linguists who
really know how language operates. When people are just sort of
throwing things together that look alien, you've got kind of a mess.
One thing people like to do is throw apostrophes all over the place.

Q. What was working with James Cameron and the actors on the set like?

A. There were days when I spent 12 or 13 hours on the set. ... I
typically met with [the actors] a couple of weeks before a particular
scene had to be shot to go over the lines, to go over the
pronunciation. I also created some MP3 files, which I sent them by
e-mail. On the set it was my job to go up to an actor—say, between
takes—and say to them, for example, "That line was really good, but
make sure that in this particular word, you're not saying 'e,' you're
saying 'eh.'"

Q. Do you have plans to further develop the Na'vi language?

A. I would love to see it developed. At this point it only has about
1,000 words. If it's going to be used for day-to-day communication, it
needs more than that. I am still adding some words every so often when
I realize that there is an important gap.



 Harold F. Schiffman

Professor Emeritus of
 Dravidian Linguistics and Culture
Dept. of South Asia Studies
University of Pennsylvania
Philadelphia, PA 19104-6305

Phone:  (215) 898-7475
Fax:  (215) 573-2138

Email:  haroldfs at gmail.com


This message came to you by way of the lgpolicy-list mailing list
lgpolicy-list at groups.sas.upenn.edu
To manage your subscription unsubscribe, or arrange digest format: https://groups.sas.upenn.edu/mailman/listinfo/lgpolicy-list

More information about the Lgpolicy-list mailing list