Quoting Muke Tever <[log in to unmask]>:

> From: "Thomas R. Wier" <[log in to unmask]>
> Subject: Consonant harmony?
> > I was sitting here reading an article about word classes in
> > Winnebago when all of a sudden, more or less out of nowhere, I
> > began to wonder if there were conlangs with consonant harmony.
> > (I suppose some of the examples reminded me of some examples
> > of palatal harmony I'd studied last year.) None of my languages
> > have it, and I've never heard of any.  How about it? Are
> > there any?
>
> The only example that jumps to my head is Greek stops that are brought
> together sharing their voicing type--voiced/voiceless/aspirate, so:
> phth, khth, but not pth, gth.

Actually, what is usually meant by "consonant harmony" is the direct
analogue to vowel harmony: nonlocal assimilation, or assimilation at
a distance.  Gunnar Hansson had this to say about it in his
dissertation: "an important corollary of articulatory locality
is that consonant harmony can *only* involve the shape and orientation
of the tongue tip-blade, since these alone can permeate intervening
vowels and consonants without interfering significantly with their
articulation or acoustic properties.  These are, of course precisely
the parameters which define coronal-specific distinctions such as
dental vs. alveolar vs. postalveolar, apical vs. laminal etc., and the
prediction of the theory is thus that coronal harmony is the only
*possible* type of consonant harmony.... [t]he present study adopts
a simple pretheoretical working definition of consonant harmony as
stated [below]:

  Any assimilatory effect of one consonant on another consonant,
  or assimilatory cooccurrence restriction holding between two
  consonants, where:
  a. the two consonants are separated by a string of segmental
     material consisting of at the very least a vowel; and
  b. intervening segments, in particular vowels, are not audibly
     affected by the assimilating property."

Sibilant harmony (especially between [s] and [S]) is one of the most
common forms of consonant harmony.  Here's an example of that from
Navajo:

   a. [Sili~?]   'my horse'
      [Sita:?]   'my father'
   b. [SitSi~h]  'my nose'
   c. [sits'a:?] 'my basket'
      [sizid]    'my scar'

(where /Si/ is a 1Sg possessive morpheme.) Here, these forms in (a)
show that "the underlying form of the 1Sg possessive prefix is /Si/, and
it surfaces as such also when the following stem contains one of the
[-anterior] sibilants /S/, /tS/, etc.(b)  However, if a [+anterior]
sibilant occurs in the stem, the /S/ of the prefix harmonizes with
it (c), surfacing as [si] rather than the otherwise regular [Si]."

 =========================================================================
Thomas Wier            "I find it useful to meet my subjects personally,
Dept. of Linguistics    because our secret police don't get it right
University of Chicago   half the time." -- octogenarian Sheikh Zayed of
1010 E. 59th Street     Abu Dhabi, to a French reporter.
Chicago, IL 60637