Re: How many verbs?
From: | Apollo Hogan <apollo@...> |
Date: | Tuesday, July 13, 2004, 5:57 |
On Mon, 12 Jul 2004, Gary Shannon wrote:
> This is just a random thought, but I was wondering if
> there were any natlangs with this feature that
> occurred to me for use in a conlang. The number of
> verbs in the language would be very small, covering
> only the most basic form of a particular action and
> the details would be provided by an adverb-like word
> that narrowed down the meaning of the verb.
>
> For example, the language would not have the verbs "to
> run" or "to walk", but would use adverbs with the verb
> "to move" giving, for example: "to move 'runly'" or
> "to move 'walkly'". These adverbs could then be
> applied to other verbs to create idioms unique to the
> language like: (using English words)
>
> This boy act-he excitedly too-much did, and speak-he
> runly did. Speak-I requestly did, that speak-he
> walkly do.
>
> By using the adverbs "runly" and "walkly" in different
> contexts their meanings would somewhat broader and
> more versatile, thus keeping the number of such
> special adverbs reasonably small as well.
>
> So I can't help but wonder how small a set of verbs
> would constitute a useful set. I'm not thinking of
> minimalism for its own sake, but to create as many
> verbs as would be useful without violating the spirit
> of not creating a new verb when a verb + adverb would
> do the job. A mere 10 verbs and 10 adverbs would give
> 100 verb + adverb combinations possibly taking over
> for 100 separate English verbs, or possibly creating
> novel meanings not expressible in English without some
> circumlocution.
You might be interested in Kalam, a language from the
New Guinea Highlands described by Andrew Pawley in
"A language which defies description by ordinary means"
in _The_Role_Of_Theory_In_Language_Description_.
I quote:
In this language verb stems are a closed set,
consisting of about 100 members. Speakers rely
heavily on a small subset of these, termed here
"generic verbs": 15 generic verbs account for 89
percent of all verb occurrences in text; 35 generic
verbs account for 98.6 percent of all verb tokens.
There is extensive verb serialization in Kalam, e.g. a
conventional expression translating roughly as 'to massage' is:
pk wk d ap tan d ap yap g-
strike rub hold come ascend hold come descend do
Other examples from the paper:
b ak am mon p-wk d ap ay-a-k
man that go wood hit-break get come put-3SG-PAST
'The man fetched some firewood'
b tw d-y mon tb lak-p
man axe hold-SS.PRIOR wood cut split-PERF.3SG
'The man split the wood with an axe'
b mon-day d yok-e-k, waty at am-b,
man stick-piece hold displace-DS.PRIOR-3SG fence top go-PERF.3SG
wog-mgan yow-p
garden-in fall-PERF.3SG
'The man threw a stick over the fence into the garden'
kab añañ ap yap pk-e-k pag-p ok
rock glass come fall strike-DS.PRIOR-3SG break-PERF-3SG that
'The stone fell on the glass and it broke'
and so on. A few sample generic verbs:
ag- make a sound: emit, resound, speak, etc.
d- control, constrain: hold, obtain, get, touch, handle, catch, seize,
stop, cease, complete something
yok- force something away from its usual or secure position: remove,
displace, dislodge, dislocate, throw, propel, get rid of something,
put something out of place, send away, banish, etc.
A very interesting language (and a good article to read if you can get your
hands on it.)
--Apollo
Reply