Re: Transitivity marking on verbs.

--- In conlang@yahoogroups.com, Benct Philip Jonsson <bpj@M...> wrote:
>
> I'm musing on whether it would be out of character
> for a natlang to mark the or n-transitivity of a
> verb.  This is of course related to my question on
> the arising of prepositions.
> --
>
> /BP 8^)>
> --
> Benct Philip Jonsson -- melroch at melroch dot se
According to Thomas E. Payne's "Describing Morphology", and also to Joan L.
Bybee's "Morphology: A Study of the Relation between Meaning and Form"
(Typological Studies in Language #9; John Benjamins Publishing Company,
P241.B9 1985), "valence" is the commonest morphology on verbs.  84% of
languages in Bybee's 50-language, one-language-per-"phylum" sample had
valence derivation, and 6% had valence inflection. (Consult her book for
the difference between derivation and inflection, as she used those terms
in her book.)

She considered valence, voice, aspect (not including "perfect"*
or "retrospective"), tense (including "perfect"* or "retrospective"), mood
(including evidentials), number agreement with any one or more argument(s),
person agreement with any one or more argument(s), and gender agreement
with any one or more argument(s).  (Bybee used a somewhat restricted
definition of "mood", yet it was still the semantically-broadest of all of
these categories.  Consult her book for the definition of "mood" she used
in it.)

*(BTW I would have included "perfect" as a mood, rather than a tense; but
at any rate Bybee agrees with me that it is not an aspect.)

In order by fraction of sampled languages which marked them either by
inflection or derivation, either an affix or a stem-change, the categories
were;
90% valence
74% aspect
68% mood
66% number
56% voice (tied with person)
56% person (tied with voice)
50% tense
28% person of a second participant (e.g. object)
16% gender

In order by fraction of sampled languages which marked them by inflection,
the categories were;
68% mood
56% person
54% number
52% aspect
48% tense
28% person of a second participant (e.g. object)
26% voice
16% gender
6%  valence

This next table doesn't come from Bybee; I got these values by subtracting
the values in the table above.  In order by fraction of sampled languages
which marked them by derivation only (not by inflection), the categories
were;
84% valence
30% voice
22% aspect
12% number
2%  tense
0%  mood
0%  person
0%  person of a second participant (e.g. object)
0%  gender

Bybee says the most frequent type of valence morpheme was causatives.

If you are going to mark valence on _every_ verb, then, in your conlang,
valence will be an _inflectional_, rather than a _derivational_, category,
as far as Bybee's use of those terms would go.  (The biggest difference
between "derivation" and "inflection" is that an "inflectional" feature is
obligatory and productive -- there is a way to mark every value of that
feature on any new word in the open class (verbs, in this case) --
while "derivation" is optional and not completely productive -- not every
word in the class (in this case, verbs) has to be marked for the feature,
and some values of the feature cannot be marked on some words in the
class.)  As you can see from above, Bybee found that, while valence-marking
is "nearly universal", _inflection_ for valence is rather rare.

(BTW The semantics expressed by voice-marking the verb may also be
expressed by case-marking the nominals, or by re-ordering the constituents
(a syntactic marking that is not a morphological marking); that, says
Bybee, is why voice-marking of verbs isn't more common than it is.)

Bybee also investigates, for each pair of markings, which one tends to be
closer to the stem in case they occur on the same side of the stem.  She
also investigates which markings tend to be fused with the stem.  This
involved investigating both which features tended to be marked by stem-
changes instead of affixes, and which affixes tended to get changed
depending upon which stem they were being applied to.

In every language derivational markers tend to be closer to the stem than
inflectional markers; and whichever marker is closest to the stem is most
likely to be fused with it.  Within those restrictions, the hierarchy
tended to be valence, voice, aspect, tense, mood, person-and-number.
Person-and-number agreement were often combined in one morpheme.  Person-
agreement is never derivational, it is only inflectional.

(BTW many languages either inflect or derive the verb for number-of-
patient.)

---

Tom H.C. in MI

Reply