I'll try to duplicate my previous lost reply. It's beginning to look like I 
did a better job the first time (except for things I just thought of this 
time), justifying my complaint. But, Tom, your message deserves a reply.

On Wed, 31 Aug 2005 16:55:02 -0000, tomhchappell <[log in to unmask]> 
wrote:
>
> --- In [log in to unmask], Jeffrey Jones <jsjonesmiami@Y...> wrote:
> > Does anybody know the precise definition of "actant"?
>
> According to "Oxford Concise Dictionary of Linguistics" by P.H. Matthews
> (Oxford Paperback Reference, Oxford University Press 1997,
> LoC P29.M34 1997, ISBN 0-19-280008-6 (pbk.)),
>
> | /actants/ used by L. Tesnie`re, and thence occasionally in English,
> | for the elements in a clause that identify the participants in a
> | process, etc. referred to by a verb.  Thus, in French or English,
> | a subject, direct object, and indirect object.

Thanks, Tom! This helps a lot. It looks like what you thought was right. 
Although it means that "actants" wasn't really the term I was looking for. 

> I would read this as:  "... participatants in a process [or] etc.,
> referred to by a verb."; and "in English, subject, primary object, and
> secondary object.", since to me it seems the preponderance of credible
> evidence favors Modern English having the Object of Monotransitive
> clauses align with the Recipient rather than with the Theme of
> Ditransitive clauses.
>
> This would mean that, in those languages spoken in and near Georgia
> and the Caucasus in which some forms of some verbs regularly
> distinguish between the instigator of an event and its ultimate
> executor, the Instigator and the Executor would be actants; splitting
> between them the role usually referred to as Agent in most languages
> (but only when the verb splits that role; otherwise there's just the
> Agent).
>
> In the same sprachbund several languages have some forms of
> some verbs distinguish "version", that is, "to whom(?) the action is
> oriented/directed".  (In some of them, this is simply a choice among
> the other actants; or a choice of person (1st, 2nd, or 3rd); or a
> combination of the preceding.) In case the verb has a principal
> beneficiary or maleficiary, that main bene- or male-ficiary is
> the "person" to "whom" the action is oriented.  Thus a portrait
> painted on commission, vs. a portrait painted as "art for art's sake",
> in one of these languages, will be painted with two different
> "versions" of the verb "paint", IIUC.  However, even when
> there is no beneficiary nor maleficiary, the verb may still have a
> version; "I'll run" is oriented me-ward, since I speak only of my
> affect on myself alone; while "I'll run this bundt cake over to Aunt
> Mabel's" is directed either cake-ward or Mabel-ward.

Interesting. I've been using "version" with its ordinary meaning in 
linguistic contexts, and now I'll have to be careful to not cause confusion.

> At least one of these languages does both of the above, and so some of
> its verbs have some forms that have five (5) actants; the instigator,
> the executor, the theme, the recipient, and the main beneficiary.
>
> That's assuming I understood everything correctly.
> My wife is not a member of this group, so it is safe for me to say
> the following:  I could be wrong.  (She's not supposed to know I
> know that)
>
> > Also, while I'm in question mode, what's the proper linguistic term
> > for, not the person/number affixes themselves, but the kind of thing
> > they represent, collectively.
>
> Three suggestions:  accidents (or accidence), inflections, desinences.
>
> Matthews's "Oxford Concise Dictionary of Linguistics" says of
>
> | /accidents/ Ancient term for a variable property of words
> | belonging to a specific part of speech.  Accidents included
> | categories of inflection:  e.g. number and case as variable
> | features of nouns.  They also included any other feature that
> | might vary:  e.g. the 'quality' of nouns
> | (lit. their 'what-sort-ness') was an accident initially
> | distinguishing proper nouns from common nouns.
> |     Later used especially of categories of inflection:  hence
> | '/accidence/' is in effect an older term for
> | inflectional morphology.

Thanks again; "accidents" seems to be the term, since it refers to 
_properties_.

> | /desinence/ An older term for an inflectional ending.  E.g. /-s/
> | in /books/ is the plural desinence.
>
> | /inflection/ Any form or change of form which distinguishes
> | different grammatical forms of the same lexical unit.
> | E.g. plural /books/ is distinguished from singular /book/ by the
> | inflection /-s/, which is by that token a plural inflection.
> |    The term originally meant 'modification' (lit. 'bending'):
> | thus /book/ is modified, by addition of /-s/, to /books/.
>
> Although I like "accidents", I think it might be "inflections".
>
> I have heard several different definitions of the difference
> between"inflection" and "derivation", and some pairs of them are
> compatible with each other.
>
> Among them are these:
>
> "Derivation" is what you do to a "root" to get a "stem";
> "inflection" is what you do to a "stem".

This one's too simple -- what if you have multiple derivations and/or 
inflections?

> "Derivation" creates, from a word, a new word that is a different
> part-of-speech than the original word;
> "Inflection" creates, from a word, a new word that is the same
> part-of-speech as the original word.

I'm getting the idea that the distinction is language-dependent.

> "Derivation" creates, from a word, a new word that is notably
> different in meaning from the original word;
> "Inflection" creates, from a word, a new word that is notably
> similar in meaning as the original word.
>
> "Derivation" is one or more of:  not transparent (it isn't
> obvious what the relationship is between the original word and
> the derived word), or not predictable (the same derivation
> process doesn't produce the same relationship in meaning when
> applied to different original words), or not productive (the
> derivation process cannot be used on new words);
>
> "Inflection" is one or more of:  transparent (it is obvious what
> the relationship is between the original word and the inflected
> word), or predictable (the same inflection process (almost) always
> produces (almost) the same relationship in meaning when applied to
> different original words), or productive (the inflection process
> can be used on new words).

These last couple ... I can't remember even the gist of what I said.

> I think all of the above descriptions of "inflection" apply to
> such things as personal affixes, impersonal affixes, construct
> state, and combined person/number affixes.
>
> > I'd like my grammatical explanations to be clear. Thanks to anyone
> > who even tries to answer.
>
> Well, I gave it a try; you're welcome.
> I don't feel confident "inflections" is the right answer.
> Would someone else on the list say?

I don't think "inflections" is the term, either. It also refers to actual 
affixes or changes in form, rather than properties. I hope some other chime 
in; so far just you, me, taliesin, and Rene (I think -- I'm offline doing 
_this_ reply with a text editor with frequent saves to disk).

> Incidentally, what's wrong with "desinences" except that it applies
> only to endings?

It also seems to refer to the actual endings rather than the property, e.g. 
|s| instead of pluralization.

I have some comments on the 2nd part.
-------------------------------------------------------------------
>
> While we are on this subject:
>
> What /are/ all the accidents of Verbs?
> Aside from concord with actants in Person, Number, and Gender:
> Verbs can have:
> Tense, Aspect, Mood, Voice, Version;
> and what else?
> Is Aktionsart ever an inflection, or is it only an accident?

Aktionsart is stuff like process phases, iteratative, and habitual, isn't 
it? If so, I've used these as inflections (in the definition 4 sense) 
in 'Yemls:
 {NrM: XbSEL-f.}  "Norma was beginning to sing."
where {Xb} is the "beginning to" morpheme.

What about proximate/obviative agreement? I think Central Algonquian has 
that somewheres. I haven't tried that, but am considering marking a 
proximal/distal distinction for 3rd person in my new project.

> What /are/ all the accidents of Nouns?
> Gender, of course, which is usually not an inflection of the noun,
> but rather governs the concord of other words with the noun;
> Case, Number -- Am I leaving anything out?
> Where does State fit in?
> (Definiteness, Referentiality, Specificity, Construct State)

I can think of things that a conlanger might use, but I'm not sure if 
they're covered by referentiality. Exactly what _is_ referentiality? Does 
it include proximal vs. distal?

BTW, I don't think Construct State belongs in the same list as the other 3 
since Construct State is when the noun or adjective acts like a _head_ in 
relation to some word(s) while still being a dependent, and this marked 
morphologically.

> What /are/ all the accidents of Adjectives?
> Aside from Grade or Degree of Comparison
> (e.g. positive, comparative, superlative), do adjectives have any
> other accidents other than concord with their head noun in case,
> gender, and number?
>
> What /are/ all the accidents of Pronouns?
> Obviously Person is the main one; also Case, Number, and Gender;
> but there must be more.

Any accidents used for agreeing with nouns should be OK.

(I moved the following quote to put pronouns with adjectives)

> Do Adpositions, Conjunctions, and/or Interjections ever have accidents?

Adpositions are marked for actants in some languages. This results from 
pronouns getting tacked on, then fusing. Also, some prepositions are 
participles in origin, which are inflected.

> Thanks, like Jeff says, to anyone who even attempts to answer any
> of the above questions.
>
> Tom H.C. in MI
> =========================================================================

I think that's it ....

Jeff