[Lingtyp] A list of 50 basic sentences

[Östen Dahl]

I am a bit hesitant to get into this discussion, since I am afraid of opening too many cans of worms, but since Christian mentioned my earlier work with this methodology, I would like to point out at least one thing: the size of the undertaking, if we understand it as ‘a standard list of "comparative sentence meanings"’, as Martin suggests. The questionnaires from the EUROTYP project that Christian referred to go back to the general questionnaire on tense and aspect that was the basis for Dahl (1985). The latter contained about 200 sentences. The questionnaires from EUROTYP, which each concern one specific tense-aspect category (future, perfect, and progressive) are on average 100 sentences long. I think it is difficult to get adequate results with much shorter questionnaires – there are just too many possibilities that have to be covered, and having just one example of each phenomenon is very risky, you may not at all get what you are looking for. (Also, my EUROTYP contained about ten people working together for five years – obviously, not all the time was spent on the questionnaires, but it gives an idea of how much work it involved.)


  *   Östen

Reference
Dahl, Östen. 1985. Tense and Aspect Systems. Oxford: Blackwell. (https://www2.ling.su.se/staff/oesten/recycled/Tense&aspectsystems.pdf)

Från: Lingtyp <lingtyp-bounces at listserv.linguistlist.org> För Christian Lehmann
Skickat: den 9 maj 2021 11:59
Till: lingtyp at listserv.linguistlist.org
Ämne: Re: [Lingtyp] A list of 50 basic sentences

Just another methodological observation: If the method involves an automatic analysis of morphological glosses, you have to control the glossing method.

For instance, your gloss
1.SG be.1.SG.PRES in DET house
uses DET, supposedly 'Determiner'. This means that some of your glosses (like SG) identify a particular value of a morphological category (number, in this case), while others (like DET instead of DEF) identify an entire category. Your project presupposes that you be consistent in this. See:

https://www.christianlehmann.eu/ling/ling_meth/ling_description/representations/gloss/index.php?open=class_member

Christian
--

Prof. em. Dr. Christian Lehmann
Rudolfstr. 4
99092 Erfurt
Deutschland
Tel.:

+49/361/2113417

E-Post:

christianw_lehmann at arcor.de<mailto:christianw_lehmann at arcor.de>

Web:

https://www.christianlehmann.eu


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.linguistlist.org/pipermail/lingtyp/attachments/20210509/78d7d043/attachment.htm>