iiinova — where your digital self keeps evolving

Mal Shah @compositiomality.bsky.social · 4h

It’s a classic “bracketing paradox” example in linguistics! The semantics treats it as [[ancient philosoph] er] but prosodically it’s ((ancient) (philosopher))

020

Reposted by Mal Shah

victor @notnaughtknot.bsky.social · 20h

great, this bullshit correlates with some other bullshit we don’t understand

0152

Reposted by Mal Shah

talking lion (hauntological) @talkinglion.bsky.social · 02/10/2026

I miss when we were on Twitter before it was evil and someone would post something like, "humanities professors only want one thing and it's discussing" and everyone was pleasantly diverted

0175

Reposted by Mal Shah

Tibor Tauzin @tibortauzin.bsky.social · 30/09/2026

The current non-replication of Marcus et al. (1999) may reflect this confound. DOI:10.1111/desc.70259 We have empirical data on infants' ability to differentiate between repetition of the same sound sequence vs variable sequences. It induces different oculomotor and brain responses in them.

133

Mal Shah @compositiomality.bsky.social · 30/09/2026

At some point in my first year I said I couldn’t remember the argument in a paper and my advisors replied “That’s good evidence that it didn’t make any sense; good papers are easy to remember” and I try to carry this with me

030

Mal Shah @compositiomality.bsky.social · 30/09/2026

And of course I notice a typo now :,)

020

Mal Shah @compositiomality.bsky.social · 30/09/2026

Accepted to Linguistic Inquiry! I argue on syntactic grounds “some of the apples” is really “some apples of the apples” with NP-ellipsis. This explains which quantifiers and which modifiers (adjectives, prepositional phrases, relative clauses) can’t appear. Preprint: ling.auf.net/lingbuzz/009...

ling.auf.net

44512

Reposted by Mal Shah

noam @noamchompers.bsky.social · 23/09/2026

‘Sure Ironman, your suit has always worked… before’—Metainducer, the main Phase 5 villain

2344

Mal Shah @compositiomality.bsky.social · 23/09/2026

The only weird thing here is that “enough” never occurs with “much” and it always inverts with the adjective

010

Mal Shah @compositiomality.bsky.social · 23/09/2026

I think it’s not different to “more”, “as much”, etc is (1) This bag of flour is more than we need/enough for the recipe/as much as we need I think the fact that they adjoin to any category makes me comfortable enough to say they’re all heads of DegP

130

Reposted by Mal Shah

noam @noamchompers.bsky.social · 19/09/2026

sometimes i give people feedback on job market materials, and thought it might be useful to do a thread of some common feedback i find myself giving. just one guy's opinion (and one thing that can be frustrating about producing and getting feedback on these documents is that informed people *will*

16913

Reposted by Mal Shah

Alex von Tunzelmann @alexvont.bsky.social · 18/09/2026

Well done to the picture editor who captioned this delightful snap of a black pig swimming off the west coast of Greece “swine-dark sea” www.theguardian.com/environment/...

Swine-dark sea ... Mediterranean black pigs swim at a beach on the private island of Atokos, in the lonian Sea off the west coast of Greece. The island, which is uninhabited by humans, is owned by a shipping magnate
Photograph: Aris Messinis/AFP/Getty Images

72545501

Mal Shah @compositiomality.bsky.social · 17/09/2026

Word that’s like “yearning” but describes the relationship you have to a syllabus you write for a course you are unlikely to ever teach

080

Mal Shah @compositiomality.bsky.social · 09/09/2026

Blog post: On "Scope" I got curious about the history of the term in semantics/logic, and found some surprising things about talk of the scope of quantifiers vs. the scope of variables. It's brief, and I'm no historian of logic, so if anyone knows more, please share! www.malshah.com/blog/on-scope

061

Mal Shah @compositiomality.bsky.social · 30/08/2026

PJ Harvey Pulp Garbage Sleater-Kinney Slowdive Mogwai Keane Heavenly The Swirlies The Killers

120

Reposted by Mal Shah

Schnitza Fars @schnitzafars.bsky.social · 19/08/2026

not dyslexic, but also every time Kayne is cited I think it's Kanye

171

Reposted by Mal Shah

Jake Quilty-Dunn @quiltydunn.bsky.social · 19/08/2026

nominative determinism is bullsh-

1425

Reposted by Mal Shah

Nvidie @dervine7.bsky.social · 18/08/2026

can confirm, as an academic I read* hella books** per year *download **PDFs

0142

Reposted by Mal Shah

David Poeppel @davidpoeppel.bsky.social · 18/08/2026

Conceptual Form: The Hidden Structure of Commonsense Concepts is out @mitpress.bsky.social This is just great cognitive science! If you like concepts, words, AI, neuroscience (like me)...Sandeep has a nuanced theory. Finally I know what DOG is all about :-) Congrats mitpress.mit.edu/978026205313...

0166

Reposted by Mal Shah

Sam Gershman @gershbrain.bsky.social · 18/08/2026

@arthurpr4t.bsky.social has a new preprint with important results on a famous psychophysical law (Weber's law). It isn't, in fact, a law, because it can be broken. A more fundamental principle (efficient coding) shows when and why Weber's law holds true. www.biorxiv.org/content/10.6...

biorxiv.org

Efficient coding makes and breaks Weber's law

Weber's law is a rare quantitative regularity in psychology, yet its origins remain debated. Here we provide causal evidence that it arises from the more fundamental principle of efficient coding. Thi...

413041

Reposted by Mal Shah

Mirjam Knörnschild @berlinbatlab.bsky.social · 18/08/2026

What would it take for AI to decode animal communication? In our new opinion piece, we argue that animal signals should not be understood only by how similar they sound, but by how the receiver perceives them. 🔗 doi.org/10.1016/j.cu...

03010

Mal Shah @compositiomality.bsky.social · 14/08/2026

Turning immediately to the psychology of language users? Fair enough, maybe strategically a bad move. Turning eventually to the psychology of language users? Can’t evaluate the productivity of that in advance of actually looking

010

Mal Shah @compositiomality.bsky.social · 14/08/2026

I go back and forth on this. There was an aspect of this which meant a field got started which could never get started otherwise. But I feel like it never recovered from this neutrality to the point of being allergic to the question of what meanings are (maybe just me being a salty internalist)

120

Mal Shah @compositiomality.bsky.social · 12/08/2026

Yeah, imo incommensurability of this sort is that it’s 100% a pragmatic thing subject to contextual factors etc! (Probably a heterodox view…) This is why I started wondering if people had sat down and asked if there’s general features of comparable properties in a way that semanticists tend not to

110

Mal Shah @compositiomality.bsky.social · 12/08/2026

I know the semantics lit on this but wondering whether the metaphysicians had thoughts about it. Felt related to old questions about predicate reference to me

010

Mal Shah @compositiomality.bsky.social · 12/08/2026

Philosopher friends, any good references on incommensurability per se? Lots of stuff on incommensurability of goods (life as a clarinetist vs lawyer etc) and incommensurability of theories, but like what makes comparisons of two properties weird? (eg ! “Lee is richer than Bo is tall”)

252

Mal Shah @compositiomality.bsky.social · 11/08/2026

Doing the homework for hangout class

020

Mal Shah @compositiomality.bsky.social · 11/08/2026

Having friends in academia has you saying to yourself stuff like “I need to read that paper before we call to catch up tomorrow”

171

Reposted by Mal Shah

Nvidie @dervine7.bsky.social · 08/08/2026

man, it’s crazy that someone who was probably lying about obviously implausible claims and faked data was able to succeed in the social sciences, really shows how unrigorous the field is, you’d never see something like that in physics or biology

1172

Mal Shah @compositiomality.bsky.social · 30/07/2026

Oh cool! we have very very little to say about stuff that isn’t part of the “logical” vocabulary so glad you’re looking into this!

100

Mal Shah @compositiomality.bsky.social · 29/07/2026

I don’t know who this is and i am scared to find out

130

Mal Shah @compositiomality.bsky.social · 29/07/2026

Oh so it’s the same as the offline world

110

Mal Shah @compositiomality.bsky.social · 29/07/2026

(First skeet in a year and i immediately start hawking my wares…)

030

Mal Shah @compositiomality.bsky.social · 29/07/2026

Quantifiers could be associated with different algorithms/hyperintensions from person to person, even if they all agree on truth conditions. Are they, in fact, variant in that way? We find evidence for “no! Everyone represents them the same!” Preprint here: ssrn.com/abstract=719...

ssrn.com

Individual differences reveal underlying similarities in meaning: quantifier-dependent choices in verification procedure are consistent across individuals

Previous literature has investigated how quantificational sentences like 'Most of the dots are blue' are understood, using visual tasks which force pa

3153

Mal Shah @compositiomality.bsky.social · 29/07/2026

Doing my yearly check in to ask, “is this website funny yet” Initial reports more promising than earlier tests

2131

Mal Shah @compositiomality.bsky.social · 21/06/2025

it is affecting the way i speak for sure (i complain about AI a lot more and a lot more annoyedly than i used to)

030

Mal Shah @compositiomality.bsky.social · 01/06/2025

yeah - imo, though i've never gone into the weak islands lit, i feel one of the things we've lost w phases is a way of talking about this sort of thing (or repair by ellipsis, if it exists) when i'm explaining this stuff i always use an ECP/CED violation; i *think* those are horrible in any context

110

Mal Shah @compositiomality.bsky.social · 01/06/2025

would you give such a speech irl??

110

Mal Shah @compositiomality.bsky.social · 28/05/2025

Type Logical Semantics? Well, yeah, I like to think of myself as a logical type of semanticist

030

Reposted by Mal Shah

puuuumpkin lemma @evpok.love · 25/05/2025

glad to be done with work and finally have some free time to work

062

Mal Shah @compositiomality.bsky.social · 22/05/2025

two pages inc refs trees and examples! should be able to find a way to improve it

110

Mal Shah @compositiomality.bsky.social · 22/05/2025

I had a dream that I attended a lecture about linear logics that made me understand them but I woke up and now I'm back to not understanding

120