[Lingtyp] Motion verbs and actionality classes

Wed Sep 4 13:52:29 UTC 2024

Dear Sergey, dear all,
the fact that certain tense-aspect grams license a present time interpretation with states and a non-present (past or future) interpretation with events is not uncommon across languages. This phenomenon has sometimes been called the present perfective paradox [PPP] (Malchukov 2009; De Wit 2017), based on the assumption that such tense-aspect grams come with perfective semantics. As it happens, I am conducting a cross-linguistic study on the PPP for my dissertation.
Crucially, I acknowledge that Sergey was talking about situations ongoing at reference time instead of situations ongoing at speech time, i.e., about relative rather than absolute time reference. I still think the Modern Western Aramaic pattern is so strikingly similar to the things that I have been investigating that I decided to gloss over this (admittedly, non-trivial) difference in the remainder of this message. I can see that this might not have been appropriate, but I hope that some of the things written below could still be relevant.
Examples for other gram types that manifest the PPP are iamitive particles, which are quite widespread in Southeast Asia (e.g., Olsson 2013), and the so-called factative verb forms that can be found in many Bantu languages (e.g., Welmers 1973). I believe the isiXhosa present tense that Stefan Savić described earlier in this thread could be considered to be a factative tense-aspect form as well.
For some grams, the semantic distinction ‘triggering’ the PPP is (a)telicity instead of (non-) stativity, such that both states and atelic events receive a present-time interpretation, while telic events receive a non-present (mostly past) interpretation. In Modern Western Aramaic, however, the situation appears to be reversed: telic intransitive events seem to pattern with states, setting them apart from other event verbs. I find this surprising for two reasons: 1. the opposition between (necessarily atelic) states and telic events vs. atelic events, and 2. the fact that valency seems to play a crucial role here. So far, I haven’t found languages where (in)transitivity on top of actionality influences the PPP. In summary, I cannot explain the pattern that you found, Sergey, but I can tell you that it looks like a unique subtype of an otherwise relatively common phenomenon. I hope this helps a bit.
 …obviously, this is all based on the assumption that I understood your observations correctly, so please let me know if I got anything wrong. In particular, I would like to point out that your examples for intransitive telic verbs that do not pattern with states (“fall”, “pounce” and “die”) could all be characterized as punctual – while all examples for verbs that are subject to the PPP, in my words, are clearly durative or could at least be conceptualized as durative situations. This might explain their different behaviour. So, the crucial verb class we are dealing with here seems to be intransitive motion accomplishments. Would you agree? And, going a step further, would you say the following statement is an accurate description of the phenomenon you found in Modern Western Aramaic (yet again, glossing over the difference between relative and absolute time reference):
The denominal perfective verb form licenses present-time reference with states and intransitive motion accomplishments, but licenses past-time reference with all other non-stative verbs (irrespective of valency, telicity, durativity etc.).
…?
Many thanks for bringing this topic to my attention. Let me know if you would like to have any additional information on the PPP.
Best wishes,
Tom Koss (PhD student at University of Antwerp)
________________________________
From: Lingtyp <lingtyp-bounces at listserv.linguistlist.org> on behalf of Stefan Savić via Lingtyp <lingtyp at listserv.linguistlist.org>
Sent: Tuesday, September 3, 2024 12:45 PM
To: Mark Donohue <mhdonohue at gmail.com>; Sergey Loesov <sergeloesov at gmail.com>
Cc: LINGTYP at LISTSERV.LINGUISTLIST.ORG <lingtyp at listserv.linguistlist.org>
Subject: Re: [Lingtyp] Motion verbs and actionality classes

CAUTION: This email originated from outside of the organization. Do not click links or open attachments unless you recognize the sender and know the content is safe.

Hello,

In isiXhosa there are quite a few verbs for which the (recent) past marker refers to the moment of speech (sometimes referred to as 'inchoative' verbs in three literature), albeit I don't remember motion verbs per se. These include:

- ukubamba 'catch'
- ukuphatha 'hold, carry'
- lala 'sleep'
- hlala 'sit'
- ukulamba 'get hungry'
- ukulahleka 'get lost'
- ukufaneleka 'be suitable'
- ukulunga 'be good'
- phila 'be well/fine'
- ukuqina 'become firm/hard'
- ukubukela 'watch'
- ukumamela 'listen'

The list of such verbs is, of course, much longer. Uku- is the infinitive augment + prefix.

In the present tense these verbs often refer to the onset of the activity/position, e.g. in present habituals. Some of these verbs can also be used as non-inchoative verbs, e.g. ukuhlala with the present tense means to 'reside, live', whereas ukuphila with the same tense means 'to live, be alive'.
As a non-inchoative verb, ukuhamba normally means 'walk, travel', but it can also denote the act of leaving ("onset" of walking, travelling) but unlike the inchoative verbs from the list above the activity of walking/travelling is not restricted to the present tense when used with the recent past (it can also refer to an event of walking/travelling entirely located in the past).

Best,
Stefan

Sent from Outlook for Android<https://aka.ms/AAb9ysg>
________________________________
From: Lingtyp <lingtyp-bounces at listserv.linguistlist.org> on behalf of Mark Donohue via Lingtyp <lingtyp at listserv.linguistlist.org>
Sent: Tuesday, September 3, 2024 2:50:34 AM
To: Sergey Loesov <sergeloesov at gmail.com>
Cc: LINGTYP at LISTSERV.LINGUISTLIST.ORG <lingtyp at listserv.linguistlist.org>
Subject: Re: [Lingtyp] Motion verbs and actionality classes

Hi,

Back in 1973 Andy Pawley noted the odd behaviour of directed motion verbs in Oceanic, observing that they can simultaneously be viewed as agentive (the moving person initiates the action) and patientive (the moving person undergoes a change of location). For this reason, they can be interpreted differently by different languages, including their behaviour with respect to aspectual classes, or what have you.

-Mark

On Tue, 3 Sept 2024 at 06:13, Sergey Loesov via Lingtyp <lingtyp at listserv.linguistlist.org<mailto:lingtyp at listserv.linguistlist.org>> wrote:

Dear colleagues,

In Modern Western Aramaic, an endangered language spoken in Syrian Qalamoun mountains, verbs of telic intransitive motion (those for ‘enter’, ‘go out of/leave’, ‘go up/ascend’, ‘go down/descend’, ‘go past’ [German ‘vorbeigehen]’) use their denominal Resultative/Perfective form to encode the respective events in progress at the reference time, quite like static verbs of perception (‘see’, ‘hear’) or body posture (‘lie’, ‘stand’, ‘sit’). Various other dynamic intransitives (e.g., the verbs for ‘fall’ or ‘pounce’, and also ‘die’) use, quite expectedly, the denominal Imperfective for progressive situations at the reference time, while their Perfective encodes Resultative, Perfect, (past time) Evidential, etc.

I observe a similar picture in Levantine Arabic dialects, in particular the Damascene. I wonder what is so special about basic verbs of intransitive telic motion that they behave like statives?  Have you ever seen something similar?

Thank you very much,

Sergey

_______________________________________________
Lingtyp mailing list
Lingtyp at listserv.linguistlist.org<mailto:Lingtyp at listserv.linguistlist.org>
https://listserv.linguistlist.org/cgi-bin/mailman/listinfo/lingtyp
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.linguistlist.org/pipermail/lingtyp/attachments/20240904/a7e0b9cb/attachment.htm>