Re: "Semantic domain", an definition of terms!

Paul Zellmer (
Fri, 22 Nov 1996 00:19:28 -0800

> Paul Evans wrote;
> > A quick question about a term being used frequently on the list of late.
> > My 13 year ago Greek classes provided me with an intro. to Koine, and I
> > have tried to sharpen my skills by further study. This list, with the
> > opportunity to ask, what are undoubtedly, elementary questions, has been
> > invaluable. I assume that by the term "Semantic Domain," which is the
> > peculiar approach of Louw & Nida according to the posts, one means the
> > spectrum or range of meanings a particular word may have.
> >

Carlton Winbery wrote:
> I would say that the "Semantic Domain" is an area of meaning such as
> "emotion" or "feelings." Then is listed under the domain the words that
> signify the related meanings.

Paul, et al.,

Semantic domain is a linguistic term that recognizes that languages
divide up the spectrums of concepts differently from other languages.
If you use Ken's example of run vs. walk, crawl, etc. (I am going to
assume that we each got a copy of that posting), the English word "run"
occupies a specific set of concepts in the realm of motion using the
legs. Native speakers develop these world views while they are learning
the language as children, and so Carlton's description of "feeling" has
some merit. The way that linguists determine semantic domains is by
using a word in different contexts, asking for native speakers to give
feedback as to whether the word is appropriate in that situation.

Sometimes the domain of a word is broader in one language than it's
"definition" is in another. For example, in English we can describe a
horse as a mare, a stallion, a colt, a gelding, etc. In a language in
the Filipines in which I minister, there is only one word--the generic
term for horse. Sometimes the domain of a word is much smaller, e.g.,
the number of words for the types of snow in Eskimo dialects with no
single word (as I have been led to believe) which is generic for the
entire spectrum of snow. Then there are times when the domain in the
language covers parts of two or more domains in another language.
Again, taking an example out of the Filipino language that I work with,
"orange" is split between "yellow" and "red".

The problem that we run into with Koine is that there are no longer
native speakers. All we can do is look at how the words are used in the
extant documents, the usage of parallel words in related languages, and
the etymology and then guess. Louw and Nida, with their interest in
assisting translation work in minority languages, tried to help
reconstruct the conceptual spectrums. Using this work, the translator
of a minority language can (hopefully) better know what questions to ask
the native checkers to make sure that the correct word in the target
language was chosen. A lot of their work is based on the concept of a
"meaning-based" translation vs. a "literal" translation.

Paul, it's not all subjective and guess-work, but I guess it could be
considered a bit of an art to define semantic domains.

Paul Zellmer
Southern Methodist Missions