Language Log

Or maybe "initialism lengths"? Wiktionary defines initialism as "a term formed from the initial letters of several words or parts of words, which is itself pronounced letter by letter"; while some (fussy) people argue that the term acronym should be reserved for words like laser (= "Light Amplification by Stimulated Emission of Radiation") or NATO (= "North Atlantic Treaty Organization").

Acronyms/Initialisms are (mostly) words, under any reasonable definition. But this category has the special property that most items have multiple specific and distinct senses, generally known to small groups and/or used in very special circumstances.

For example, American linguists know that LSA stands for "The Linguistic Society of America" — but the LSA didn't act in time to lock up https://lsa.org, which belongs to the "Louisiana Sheriffs' Association". And Acronym Finder gives 123 interpretations for LSA, including the linguists but (curiously) not the sheriffs.

Sylvain Farrel is a student nurse from Indonesia.  He came to America four years ago and speaks perfect English.  I asked him how that is possible, how did he learn English so quickly?

Sylvain said that he studied English during his elementary and middle school education.  His national language is Bahasa (Indonesia), i.e., Indonesian.

By ethnic heritage, Sylvain is Chinese, Hokkien / Fujian on one side, and I think Hakka on the other side, but I'm not sure.

Following up on "Meadow writing", I thought it might be interesting to look at LOTR-associated word frequencies, using the the "weighted log-odds-ratio, informative dirichlet prior" algorithm Monroe, Colaresi, and Quinn 2009, "Fightin' Words", as discussed in seven previous LLOG posts. In particular, I thought I'd compare The Fellowship of the Ring to 16 of Charles Dickens' works.

At BMR, the first thing the doctors, nurses, and techs ask patients when they interview them is "Do you feel any pain?"  And they want you to quantify it on a scale of 1-3-5 / small-medium-big.

What is pain?  Physical, mental?

I tend to think of it rather as Sanskrit duḥkha (/ˈduːkə/ दुःख) than as English "pain", because the former is more all encompassing (corporeally, spiritually) than the latter, which I feel is more physical.

Following up on the DP's April Fools "AI-yatollah" article, an Ayatollah pun from Admiral James Stavridis, USN, Ret.:

[image or embed]

— Admiral James Stavridis, USN, Ret. (@admiralstav.bsky.social) April 8, 2026 at 7:57 PM

Today's xkcd:

Mouseover title: "I think I accidentally installed an Overton window in my bedroom. A few months ago, the sun wasn't in my face in the morning, but now it is."

ICYMI: Wikipedia on "Overton Window".

More comically interesting: the menu of "Mode" choices now routinely displayed below the cartoon:

From Mark Dow:

From "Everyday Politics in Russia", The Eurasian Knot 4/6/2026:

The podcast starts with a message from listener Amanda, who has been reading all of Dostoevsky for a workshop in Russia. In addressing the podcast's host Sean Guillory, she says (starting at 4:21.5):

I sympathize with you, Sean, that you just couldn't get into him,
but I've personally never felt that way about Dostoevsky.
I remember trying to read the Lord of the Rings series,
and I couldn't stand it.
I couldn't stand ten pages describing a meadow.
And ever since them I've thought of fiction writing in terms of
meadow-writing and non-meadow-writing.
No wonder I love Dostoevsky —
he has nothing whatsoever to say about meadows.

The battlelines are being drawn ever clearer.  On one side are those who believe that it's all right to use AI to help with the preparation of an (academic) article, essay, or paper.  On the other side are those who think that the utilization of AI is impermissible for such purposes.  As soon as they discern the use of AI in writing a composition, they will dismiss it out of hand.  Use of AI extends to the collection and organization of material to be included in what is being written.

Readers who are sensitive to the stylistics of AI writing can even detect it in punctuation preferences, rhetorical tone, lexical propensities, and so forth.

There are even commercially available "AI detectors", e.g.:  "Pangram can detect AI-generated text even after it has been 'humanized,' or processed by tools that attempt to evade AI detection, ensuring reliable detection."

The most recent SMBC:

Today's xkcd:

Study of 1,700 languages reveals surprising hidden patterns
Languages may seem wildly different, but new research shows they follow surprisingly consistent—and deeply human—rules.

Science News, Max Planck Society (4/5/26)

Summary

A massive new analysis of over 1,700 languages shows that some long-debated “universal” grammar rules are actually real. By using cutting-edge evolutionary methods, researchers found that languages tend to evolve in predictable ways rather than randomly. Key patterns—like word order and grammatical structure—keep reappearing across the globe. The results suggest shared human thinking and communication pressures shape how all languages develop.

This post documents one small step in a larger plan for improved evaluation of prosody in reading. It compares word-level timing in a large number of recordings, from the Speech Accent Archive at GMU, of 3038 people reading the 69-word "Please call Stella" passage. 661 of these people are native speakers of English, with accents from all over the anglophone world, while the remaining 2377 readers have native languages from Afrikaans to Zulu. The reading and speaking level of those non-native readers varies a lot, and many of them have problems in decoding or pronunciation that affect their timing.

Automated analysis of such problems should be useful in foreign-language teaching. And similar analyses might help in early reading instruction for students in anglophone classrooms, whatever their native language.

Let's start with a quick comparison of word-level timing in the 661 native English speakers; the 85 native French speakers; the 99 native Korean speakers; and the 82 native russian speakers.

I calculated word-level time points for those 927 speakers, using a forced-alignment system originally developed many years ago with Jiahong Yuan — a summary of the technology and a few of its application can be found here (open-access version). Here's the output for speaker english1 — note that the segment ID sp means "silent pause".

The key conclusions:

1. Time between word onsets gives a good picture of phrase structure, despite the many other effects on timing;
2. Individual non-native readers, aside from being overall a bit slower, usually show lengthened inter-word intervals in unexpected places, due to decoding or pronunciation problems.