The Conlang FAQ

isolating languages

adapted from a 13 Sep 96 post by Mark P. Line

Don Blaheta wrote:

I still have a question or two though. One thing I've never quite understood is, what is the functional difference between isolating and non-isolating languages? I mean, just because we _write_ Esperanto like "la viro havas infanoj", does that make it agglutinative?
No, at least not by my definitions. Orthography is irrelevant, because orthographic words do not necessarily parallel phonological words exactly, and my definition of isolation hinges upon the _phonological_ word.

I'm not sure what would be involved in setting up a good definition of the phonological word in Esperanto that would also account orthographic monosyllables. Orthographic words of more than one syllable are no doubt phonological words in Esperanto, because penultimate stress is universal. How the monosyllables are treated would depend on stress and timing.

What if we wrote it "la vir o hav as infan oj"? The isolating particles o, as, and oj would have the obvious effect on the prior word. Would that make it isolating?
Only if each of the orthographic words were also phonological words.

Would there have to also be a slight difference in the speech? Are written and spoken forms even comparable in this respect?

Yes. No. The orthography is irrelevant for the delimitation of phonological words. I suspect that your example has three: /laviro havas infanoj/, in which /la/ is a clitic.

As another example, it seems to me that word boundaries in spoken French are not at all the same as word boundaries in written French.
That is very painfully true.

Words like "se", "je", "la", and "que" lose their last vowel before a vowel-initial word; in the spoken French these words get attached to the following word. So are they agglutinating or isolating?
"Je" generally fills a head slot, not a modifier slot, so it wouldn't normally be counted either way. It's nevertheless an affix, assuming that it never occurs in a phonological word by itself (unlike "moi").

"La" is a clitic. I wouldn't normally count clitics towards isolation or agglutination. If you did count "la", you would say it's inflecting, not agglutinating.

"Se" and "que" might be clitics or they might be affixes, depending on whether or not they can ever occur in a phonological word of their own. If they're affixes, then they're no doubt inflectional affixes of the agglutinating variety.

In my version of English, the word "to" (when it's not attached to the end of "have") gets attached to the following word, eg. "to him" /t@hIm/, "to the house" /t^T@ haUs/, etc. Does that make it an inflecting particle?
It's certainly inflectional in these examples, but not inflecting by my definitions. It doesn't qualify for our typological counting because it's probably a clitic (it occurs by itself when left dangling, for instance). And even if it were an affix, it would be agglutinating, not inflecting.


Return to Conlang-related topics|Back to FAQ page

Copyright © 1997, Jack Durst,
Last updated: 25 April, 1997