Now out, "A Simple Explanation for Harmonic Word Order".- we show that a core typological pattern in syntax can be explained as a consequence of word-class frequencies, so it does not need to be posited as an independent principle.

Thanks to Lothar Sebastian Krapp for adding a formal mathematical description! .I also got incredibly generous input along the way from Charles Kemp and Jeff Good. but not sure if I ever want to spend so long on one article again?? 😬.

We treat syntax as an evolving, inherently variable phenomenon. We also argue that word classes are better treated as gradient categories, rather than fully discrete. This approach fits natural language data better than rigid rule-based models.

To show this, we use an explicit model of word order, where each phrase is ordered by "replication with modification" of previous phrases. With minimal .assumptions, this shows that the "head" category (most frequent word class in the phrase) spontaneously gravitates to an edge.

I am also open to applicants whose degree is not in linguistics, but still have relevant analytic skills or academic background, and can demonstrate a good understanding of linguistic structures and linguistic diversity.

For the PhDs, an interest in Papuan or Australian languages would be great, but other regions also possible. For the PosDoc, data science skills esp spatial and/or phylogenetic analysis are crucial.

I'm hiring! 2 x PhDs and 1 x PostDoc, to work with me at Uni Zurich on a project investigating how semantic patterns differ by geographic region. Linguistic typology, data science, spatial analysis, phylogenetic methods, potential fieldwork. REPOST 🙏

if this make no sense, don't worry, just read the paper :).

Main finding: Parallel harmonic word order can be explained as a frequency effect. Given a simple word-order that samples previous phrases and approximately replicates their order, the most frequent word class gravitates to the phrase edge by simple probabilistic processes.

"A simple explanation for harmonic word order".😎 new paper with Lothar Sebastian Krapp. Formal modelling of word order as a dynamic system helped us understand some fundamental mathematical patterns!.

RT @jewishcouncilAU: Thanks to John Lyons for sharing the newspaper ad of 500+ Australian Jews calling on Australia to say NO to Trump and….

"the extraordinary callousness, bordering on racial contempt, in Germany over the fate of Palestinians".

RT @weijiexu_97: Glad to share that our paper (with @rljfutrell) is now on JML. In this paper, we proposed strategic memory allocation as a….

Anindilyakwa has amazing long words (Leeding 1989: 68). Here is a random selection of basic vocab: nɛɲcarŋaʎiʎa ‘boy’.ɛŋkəparŋʷarŋʷa ‘heavy’.al ̪uŋkᵚuwaruwaʎa ‘shade’.jinmamuwa ‘egg’.-wɛrəŋɛkpuɻakəna ‘comfort’. this is *without* the polysynthetic add-ons.

Epenthesis volume is out! Our contribution is on Anindilyakwa, an Australian Aboriginal lang where about half the vowels are kinda epenthetic. We take an info-theoretic approach: since epenthetic segments are lexically uninformative, we expect they can be freely omitted.

RT @John_Attridge: Reading the google books preview with exaggerated nonchalance so it can't sense which parts to omit.

"Disentangling contact and inheritance in lexical semantics" .📣LAST CALL for abstracts for this workshop, proposed for SLE 2025 . in Bordeaux!🍷.

RT @LangSciPress: We accepted a new series "Open Text Collections" and its first book "Speaking the map: Komnzo texts" by Christian Döhler.….

RT @weGotlieb: If you are interested in topics at the intersection of NLP, Cognitive Science, and Linguistics, please consider applying to….