On Wed, 4 Oct 2000, jesse stephen bangs wrote:

> > Yes. But it does require the user to be aware of the derivational
> > history of a word--something which won't always be true. Last year on
> > an exam I asked students to parse complex words into their component
> > morphs. One of the words was 'actively'. I was very surprised to see
> > how many students did not see the root 'act' in the word. They did
> > find 'active', but 'act' was, for all intents and purposes, invisible
> > to them. Just because a linguist is able to demonstrate a particular
> > morphological structure, it doesn't mean that linguistically innocent
> > speakers will see the same structure. So organizing a dictionary
> > according to roots will have problems.
>
> Yes, but remember that this is supposed to be for an *agglutinative*
> language, not English.  English derivational morphology is heavily
> influenced by Latin and many, many roots which are historically analyzable
> and obvious to linguists are no longer seen as roots by a native
> speaker.  The example you gave makes it seem that your students no longer
> think of 'act' as the root of 'active', probably because -ive is only
> productive in the scientific vocabulary, and the word 'active' is not seen
> as scientific.  In an agglutinating lang, on the other hand, the root
> relationships are generally much more obvious and regularly occuring, and
> any L1 speaker or competent L2 speaker should be able to recognize them.

Absolutely. But remember the -hood example. -hood is a native English
suffix and it attaches to native English roots without any morpho-
phonemic funny business. Even there, there were exceptions and gaps in
coverage. All I meant to say was that in any *naturalistic* language,
whether agglutinative or inflectional, you will have these kinds of
irregularities; it's just in the nature of derivation. If your
agglutinative project has no exceptions like this, then fine; there's
no reason to include articulated entries like I suggested.

> > I do think that having the affixes in an appendix is a great idea. But
> > having derivational information *only* in such an appendix would
> > create many problems that could be avoided.
>
> I agree with you, but the Yivrindil dictionary will nonetheless *only*
> define roots and derived forms whose meanings have drifted
> unpredictably.  Other derivatives will be listed but not separately
> defined, as the user should be able to arrive at their meaning using the
> appendix of affixes.  Any other solution requires far too much redundancy
> and is really unnecessary.

That seems to be a sensible compromise solution. As a linguist, I
would prefer a dictionary which gives fuller derivational information,
even for predictable cases, but I recognize that as a matter of
practicality it may not be possible or desirable to do so.

Dirk

--
Dirk Elzinga
[log in to unmask]