Dirk Pijpops

Ik heb een open positie voor een doctoraatsonderzoeker: uantwerpen.be/nl/jobs/vacatu…. We gaan samen variatie in de Nederlandse syntaxis onder de loep nemen met corpusonderzoek en experimenten. Als je iemand kent die mogelijks geïnteresseerd is, laat het hen dan zeker weten!

Just got back from an amazing #ICLC16. Held one talk on lectal contamination and another one on redundancy with @evaeva_z, and talked to a lot of great people! For lectal contamination, see: orbi.uliege.be/bitstream/2268… and for redundancy: glossa-journal.org/article/id/876…

I'm happy to announce that the Centre for Linguistics at ULiège has two open PhD-positions in Dutch diachronic linguistics and typology, as part of our collaborative project "Grammar from Space", funded by the ULiège research council. Come work with us! lilith.uliege.be/cms/c_9979357/…

Leuk stukje door Marc van Oostendorp over een recent artikel van @evaeva_z en mij

Out now! Is language redundant, or is it organized efficiently? Check out my new joint paper with Eva Zehentner in Glossa! Available in Open Access: glossa-journal.org/article/id/876…

The article is now available in open access!

We know that language variation should be studied at low levels of schematicity. But how can we do that in a scalable way? Check out my new paper with Dirk Speelman and @antalvdb! doi.org/10.1515/lingva…

Today in 'Linguistics', open-access, aop: our statistically-driven study on differences in writing style between men and women, how this has changed in the course of the 20th century, and a discussion on why these differences emerge in the first place: doi.org/10.1515/ling-2…

Zaterdag vindt het Herfstcolloquium van de KZM plaats in Gent, met lezingen over taalkunde, letterkunde, geschiedenis en klassieke studies en een plenaire lezing door niemand minder dan dr. Dana Louagie! Iedereen is welkom, het hele programma vind je hier: tinyurl.com/2cca3bn5

In this #onlinefirst paper, @DirkPijpops shows how agent-based simulation can be used as a complementary technique to corpus research in the study of language variation doi.org/10.1075/ijcl.2… #languagevariation #corpuslinguistics

How can you combine linguistic corpus research with agent-based simulation and why should you do it? Find out in my new paper in IJCL: lilith.uliege.be/cms/c_8676280/…

@DirkPijpops moreover, we observe an impact of ‘complexity’ (length/weight of recipient) on probability of redundant marking: the longer the constituent, the more likely we are to get multiple strategy use > we take this to confirm both hypotheses & a balance between efficiency and robustness

@DirkPijpops corpus data from Present Day Dutch and English corpora shows us that redundant marking, i.e. using more than one disambiguation strategy in a single utterance, is the default, but only to a limit: double marking is frequent, but triple/ quadruple marking rare(r) across languages

@DirkPijpops we test 2 opposing hypotheses: (A) on a production efficiency account, redundant (costly) strategy use should be dispreferred (cf. popular trade-off scenarios) vs (B) on a robustness account, redundancy is nice: it aids transmission in noisy channels & complex environments

@DirkPijpops we choose this example since it allows us to ignore a further common disambiguation strategy, viz. semantic biases (subjects tend to be animate, objects inanimate etc.), with both agents and recipients being very similar in semantic-pragmatic terms, & focus on formal strategies

@DirkPijpops we look at strategies for distinguishing agents and recipients in Dutch and English ditransitives, i.e. (i) prepositional marking (gave a book to the students), (ii) formal differentiation (subject vs object pronouns), (iii) subject-verb agreement, and (iv) constituent order

redundancy: do dutch and english speakers like it? we investigate redundant use of argument disambiguation strategies as a prime test case of competing motivations, specifically production efficiency vs transmission robustness #linguistweets #TW0830 @DirkPijpops

We are looking for a postdoc to work with distributional models on lectometry in 2022! All info here: linguistlist.org/issues/32.3516/

Op woensdag 8 december komt Pascal Verbeken op uitnodiging van de vakgroep Nederlands naar Luik om een lezing te geven, getiteld 'Arm Wallonië – Arm Vlaanderen'. De lezing is vrij toegankelijk voor eenieder, maar inschrijving is wel verplicht: lilith.uliege.be/cms/c_7482703/…