Show more

the other day I was barefoot and wanted to tell gf that the floor was cold, and I could feel my brain trying and failing to find the right word, like "no no not just 'cold' as in ambient, what's the term for touch-cold".

English (the lang I was using) has no lexical distinction between ambient-cold and touch-cold. neither does my mother language. it's from my L3, Japanese. somehow my mind was trying to operate under this division of reality, and frustratedly trying to hold the distinction.

covid-19, latinoamérica, lenguas originarias 

«Qhepakuy wasiykipi:
Iniciativas de la sociedad civil en quechua, náhuatl o mapuche sustituyen a los Gobiernos de Latinoamérica que no informan en lenguas originarias sobre la covid-19»

I hope somebody wants a book chapter in English about Japanese Tōhoku dialects cause I’m supposed to write one now 🙈

science sillypost 

every time when you finish inputting new data and rerun your models

(source + extra panel: )

linguistics confession 

[ə] [ɘ] are both confusing sounds and confusing symbols

voice training, game of thrones 

@elilla tagging myself for #conlang #highValyrian above ^

my next voice training conlang(-adjacent) goal is "Ph’nglui mglw’nafh Cthulhu R’lyeh wgah’nagl fhtagn", which I chose interpret as roughly [ˈp̩ʰ.ʔŋluːj mglɯ.ˈʔnɑfh ˈχtl̥ʊ.ɬuː r̥̩.ˈʔl̥jɛh ɯˈgɑh.ʔnɑgl̥ f̩h.ˈtɑŋ̚]. but it's taking more time than I thought 😇

Show thread


I literally cannot wait to dig into this!! It's what I've been wanting!

lockdown games, academia 

Headaches because hard:
- Linear algebra
- Discrete math
- Probabilistic algorithms 🤯
Headaches because boring and dumb and shouldn't exist:
- Management
- Software engineering

Done for kicks:
- Typography.
- High Middle Ages palæography.
- Writing as visual art.
- Old Tupi.
- Chinese proverbial wisdom.

Show thread

lockdown games, academia 

on my social media: list 5 subjects you enjoyed in college, 5 which were headaches, and 5 you did out of curiosity.

Enjoyed/stayed with me:
- Literary analysis under Marxist-historicist lens.
- Paul Valéry’s lit theory and genetic criticism.
- Writing systems.
- Wright's approach to oral Latin and Romance reconstruction.
- Barthes.

The common thread is, for all of these I had excellent, knowledgeable teachers whose enthusiasm was contagious.

me on day N: I wish I did more stuff with .

me on day N+2: get an email from Kurdistán Solidaridad regarding the «Defendamos la vida en Rojava en tiempo de Covid-19» initiative, asking me questions regarding my signature, all in Spanish.

> Dedico este trabajo a la flor que inició su traslado al resplandor del universo, o al fondo del océano, -¿cómo saberlo?- cuando la cueva que guarda el corazón del maíz me mostraba su belleza.

I think this might be the most beautiful PhD thesis dedication I've ever seen. (« Des-ordenando el género / ¿Des-centrando la nación? El Zapatismo de las mujeres indígenas y sus consecuencias », Márgara Millán, )

Show thread

is close enough to Portuguese – we're a dialect, really, don't @ me – that you can just grab a book and read it until you get used to the few hundred false friends. It's the perfect condition for Krashenian language learning, but somehow I never sat down to do it.

I'm thinking I'm gonna pour publications on my brain until it can Spanish, ~and~ I'll learn more from successful anticapitalism in the process!

"this meeting has reached a maximum of 100 people" 😢

guess online broadcasting isn't that great.

oh well still plenty of resources to brush up on and become xwebûn! I like this linguist-friendly intro:

Show thread

A timely video class is taking place every Monday! Def gonna check this out.

@melissaboiko A while ago I thought that at some point I should learn an ergative language, just to get some actual feeling of it.

And then I read that most sign languages are considered showing ergativity in that the handshape usually represents either intransitive subject or transitive object, but not transitive subject.

And I was OMG I've already studied TWO ergative languages without knowing it! :D

learning :

girl, ergative languages are such a trip :D

and how do they manage to juggle accusative alignment in the present tense and ergativity in the past :D

(3/3) This could provide a quantifiable measure of historical signal in a set of data – and back up linguistic arguments like "tone classes are historically stable and resist standardisation in Japanese".

So far I couldn't find a ready-made PTP tool tho 🤔 R pkg "fangorn" has MPT inference at least, but it will take some work for me to massage our tone data into a format digestible by packages that are expecting DNA sequences…

Show thread

(2/?) you note down the length of the MPT. then you scramble your data in random order, infer the MPT for *that*, and measure it. and do it again say 10k times.

now you have a distribution of minimum trees for random data. now if the MPT of your actual data looks like those, your data is indistinguishable from random data. but if your MPT is a lot smaller, your data is not random but "tree-like" – likely to have arisen by common descent from a single ancestor, with inherited "mutations".

Show thread

(1/?) Permutation Tail Probability (PTP) test is one of the many tools that I think may be valuable for .

it’s a neat idea! for a set of observed data, you calculate the most parsimonious tree (MPT) that could have generated it—kind of a fantasy perfect history where every change arose by descent via modification. in reality there prob. were some duplicated changes (homoplasy). but if your data came from a single ancestor, it would need at least this many changes.

have a nice Sunday! today I bring to you the intensely white male energy of this outraged footnote from my 19th century German textbook (Worman's « Complete German Grammar »).

Show more
Scholar Social

The social network of the future: No ads, no corporate surveillance, ethical design, and decentralization! Own your data with Mastodon!