Thematic

[Koontz John E]

Rory Larson asks:

> Re: the Crow posting, I'm going to have to ask you to explain (maybe on
> the list) what 'thematic' means in Siouan.  In Indo-European, I believe
> 'thematic' vs. 'athematic' refers to having a classificatory stem vowel
> between the root and the grammatical ending, or not.

Thematic does refer to the the e/o stem vowel (mostly) or to the quality
of possessing it, e.g., a "thematic noun" or an "athematic noun."

The term theme is more or less obsolete in general morphology, according
to Trask, and has been replaced with base.  But I think base is used
merely in the sense of "a form that is added to to produce a higher
order form."  For example, in Latin asinus 'donkey' asinu- is the base to
which the inflectional ending -s is added, and within that as- is the base
to which the diminutive -in- is added.  Let me finesse the issue of
whether -u is added to or which the -in-!  I am not really sure what the
consensus is!

Where I think theme differs from base is this:  a theme is a form that is
eligible to receive additional morphemes (so far the same), but there is
also the implication that some or all possible themes are specifically
composed of an inner part - the meaning bearing part - and an outer part -
the part that makes the inner part into a theme, e.g., the thematic vowel
in IE languages.  Clearly this use of theme is somewhat language specific.
It is useful only in morphologies where there is a class of bases that
need a thematic extension of some sort.

Rory referred to the IE thematic vowel as classificatory, and I think
that's an important point in several ways.

First, I think what makes a thematic system appear as such is that there
are always some athematic stems, e.g., Latin has its athematic
consonant-final stems like reg- 'king' (nom sg. rex, i.e., rek-s)
alongside its thematic vowel-final stems like amic-u- (nom. sg. amicus) or
amic-a- (nom. sg. amica, no -s).  So, stems are always classified into
thematic and athematic under a theme sysem.

Second, as in the amic-u-/amic-a- pair above, there is sometimes a concord
or tendency to concord of variant theme-forming suffixes with some
element of the semantics of the underlying stem.  For example, in the
Latin case there is a tendency for -u to occur with masculine and neuter
stems and for -a to occur with feminine stems, though there are examples
like morus 'mulberry' (feminine) and agrigola 'farmer' (masculine) to
contradict, or at leasy complicate, the matter.  So, to some extent the
theme vowels act as gender markers, and this tendency becomes quite
pronounced, of course, in the modern Romance languages.

Not to obscure anything, I think that the tendency of PIE *-aa- (long a,
i.e., perhaps *e + some laryngeal) to mark feminines is considered to
coincide with the introduction of the feminine category in PIE.  But there
are exceptions, for whatever reasons.  Apart from that, *-aa- isn't the
only PIE thematic element that comes to mark feminines.

Finally, while PIE thematics are characterized by the -e/o- thematic
vowel, I think that in many cases the vowel is not a suffix in itself, but
the final vowel of a longer suffix, e.g., -ne/o-, maybe also the -ine/o-
diminutive, -ske/o- inceptive (with verbs), etc.  In many cases where
-e/o- occurs, there is a doublet with -aa-, or at least one could adopt
this expanation with -inu-/-ina-.  These longer suffixes often have some
specific semantic sense, e.g., diminutive.