Language Log

It's a word I've barely heard of, though I did write a post about it two years ago ("A fancy way to say 'fancy'" [9/22/24] — with lengthy, learned discussion in the comments).  Yet this morning it turns up at the top of "America's most misspelled words in 2026" (5/22/26)!

"Words as Data: How Data Journalists Tell Stories about Documents and Text." Bradshaw, Paul. Online Journalism Blog, May 14, 2026. 

"Discovering Regularity and Mechanisms of Word Sense Acquisition in Childhood." Li, Jiangtian et al. Proceedings of the National Academy of Sciences 123, no. 20 (May 11, 2026): e2525788123. 

Below is a guest post by Randoh Sallihall:

Analysis of Google search data for 2026 reveals the most misspelled words for each U.S. state and America.

National Spelling Bee will be held from May 26 to May 28. The research is well timed.

America's most misspelled words:

1. Bougie – 134 400 searches.
2. Favorite – 128 400 searches.
3. Through – 127 200 searches.
4. Business – 123 600 searches.
5. Tomorrow – 121 200 searches.
6. Because – 106 800 searches.
7. Definitely – 104 400 searches.
8. Beautiful  – 102 000 searches.
9. Niece – 100 800 searches.
10. Separate – 98 400 searches.

In "Analysis of prosodic timing in reading" (4/5/2026), I suggested that inter-word timings in fluent reading can give a surprisingly clear picture of prosodic phrasing, despite the many other effects on word durations in speech.

That post looked at data from the Speech Accent Archive, which involves reading a short and somewhat weird passage. Since then, I've explored readings of a variety of other texts, so far only in English. The results continue to look promising.

Today's xkcd:

Painstaking research by Christian Benz, Ewa Dutkewicz, et al.

Significance

Below is a guest post by Matt Rips (and Google Gemini).

In a discussion of linguistic matters, an LLM began using a word, an adjective, repeatedly, to describe a concept. The word does not exist. There were multiple instances, such as:

One possible explanation for the tendency of quantified plural event nominals to resist direct actorive eventuality recovery is that many such nominals encode inherently multi-party relational events.

This word does not show up in OED, Google, etc. It is not an English word. Google AI suggests: “if you are studying historical texts or phonetics, you might find it useful to know that "actorive" is an archaic or reconstructed variant that sometimes appears in early phonetic research or mistranscribed Latin."

Listening a couple of days ago to the radio program Exploring Music, I learned something about English morphology. The episode broadcast on 5/18/2026 has the title "What else ya got?", and host Bill McGlaughlin introduces it this way:

The second open-access monograph in the Publications of the Philological Society has just come out, in two volumes: Gary D. German, Benjamin Franklin: Orthoepist and Phonetician —

Vol. 1: Language, Literacy and Social Mobility in Franklin’s World.
Vol. 2: Colonial American Voices and London Norms: Franklin’s Quest for an Orthographic Reform.

Today during our graduation party, I was chatting with our M.A. students from China.  The question of anxiety came up — anxiety about getting a job and / or keeping one.

I said, yes, I noticed that some of your classmates are worrywarts, and told them that we have many other colloquial words for such folk (fussbudgets, fussbuttons, nervous nellies, and so on) in English.  They were fascinated by these quixotic terms, so I asked them if they had any similar words in Chinese.  Since they couldn't readily offer any, I invented one on the spot:  zhāojíguǐ 著急鬼 (lit., "anxious ghost").  Of course, we all knew that wasn't an authentic sinicism, and I myself wasn't satisfied with it, so I pushed them a little harder:  "Are you sure you don't have such  term in your online messaging and the like?" Finally, one of them volunteered that she uses "green banana" in the same sense as "worrywart", and several others chimed in that they did too.

• "What Is Laughter?" Almeida, Abilio. Humanities and Social Sciences Communications (May 9, 2026). https://www.nature.com/articles/s41599-026-07432-4.

• "A Sentence Is Worth a Thousand Pictures: Can Large Language Models Understand Hum4n L4ngu4ge and the W0rld behind W0rds?" Leivada, Evelina et al. Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences 384, no. 2320 (May 14, 2026): 20250008. https://royalsocietypublishing.org/rsta/article/384/2320/20250008/481681/A-sentence-is-worth-a-thousand-pictures-can-large.

From the renowned cartoonist, Rebel Pepper

"What Happens When We Lose a Language?" by Sophia Smith Galer, The Guardian (5/12/26)

The Ubykhs are people I've known about for half a century, but now I wish that I knew them better.

We are lucky to know anything at all about the Ubykh language. In the 1800s, tens of thousands of people spoke it on the Black Sea coast. When Russia conquered the region, the Ubykhs resisted until they were forced into exile in the Ottoman empire. Transported thousands of miles by a traumatised community now scattered across Turkey, Ubykh survived until 1992 when its last fluent speaker died. It was one of at least 244 languages that has become extinct since 1950, and soon – unless anything changes – my grandmother’s language will have joined them.