I'm sorry I didn't reply sooner; I've been busy with work stuff that
took up a lot of my mental energy.

There are two different kinds of things which I've conflated, perhaps
erroneously: instrumental prefixes and lexical suffixes. Instrumental
prefixes (so-called) in Numic and several California languages (such as
Kashaya) indicate various body parts, basic implements and natural
forces. They can be used derivationally to extend basic meanings of
roots like 'hit'. For example, Shoshoni has the following verbs formed
from the same stem:

kukkopa   'break from heat'
sekkopa   'break from cold'
kekkopa   'bite in two; break with teeth'
takkopa   'chop with feet'
tsakkopa  'break by pulling apart'
tsikkopa  'cut (rigid object)'
wekkopa   'cut, chop (rigid object)'
mukopa    'break nose'
pikkopa   'break with behind (e.g. by sitting on it)'

The stem -kopa means 'break', and the prefixes indicate means by which
something is broken, or, in the case of _mukopa_, what gets broken.
This is pretty typical stuff. The verb is still transitive; the prefix
does not take the place of an argument.

My lexical suffixes have a bit of this flavor, at least the ones
dealing with body parts, basic implements, and natural forces.

Lexical suffixes proper are a little bit different, but not much. They
are much more specific and can refer to culturally important objects
such as canoes, berries of various kinds, and of standing on the beach
leaning into the wind (to cite a recently heard example from Kwakwala).
They are similar to the instrumental prefixes in that they do not
constitute an argument of the predicate.

As for how these systems came to be ... The Shoshoni example is fairly
transparent. The prefixes were originally nominal stems which were
incorporated into the verb; for example, the word for 'nose' in
Shoshoni is _mupin_ ['muBi]; the instrumental prefix is _mu-_. These
prefixes may have been an argument of the verb or not; I don't know.

The Salish and Wakashan lexical suffixes rarely, if ever, show any
similarity to lexical stems with the same (or similar) meaning. I have
no idea where they came from. But they're pretty handy to have around.


On Saturday, January 24, 2004, at 01:43  PM, David Peterson wrote:

> Dirk wrote:
> <<A lot of languages have derivational affixation which can indicate
> stuff like body parts or basic implements or natural forces (rock,
> stick, wind, sun). My inspiration for these is Uto-Aztecan and
> Salishan. The suffix itself isn't an argument, so this isn't
> incorporation, strictly speaking.>>
> Sorry, I just can't get off this.  It doesn't make sense to me how
> exactly such a system could come to be.  I think I understand how it
> works.  Let me see if I've got this right:
>       n-  sea    <Vk>   kasu   -pte
>       TR- out.of <COLL> remove -EYE
>      He took them out
> This means "He removed his eyes", but more standardly "He eye-removed
> them".  A standard noun-incorporational language, though, would
> probably do something like "He eye-removed himself", right?  So that
> would be the difference.  (What's also interesting: Is this a
> transitive verb that always takes a 3rdplu. object, since it's
> referring to a pair of eyes?  What if he wanted to take somebody
> else's eyes out?)  What's throwing me for a loop is what you wrote
> here:
> <<<<A lot of languages have derivational affixation which can indicate
> stuff like body parts or basic implements or natural forces (rock,
> stick, wind, sun).>>
> A *lot* of languages?  This is the first I've seen.  Do you know how
> these systems arose historically?  And what's the restriction on
> nouns/substantives that can become affixes?  Can you point me to a
> natlang. grammar that has something like this?  (P.S.: I just picked
> up a Siglit grammar [Inuit language], and it has adjective-like
> suffixes, so it's kind of like this.  They're attached to nouns or
> verbs to indicate, for example, "sharpness", "goodness", "bigness",
> etc.)
> -David