Language Log

China to restrict Tibetan language in region’s college entrance exam
Exclusion of core subject exam stokes fears Beijing is furthering campaign to ‘Sinicise’ region
John Reed, Financial Times (8/6/25)

Cantonese, Uyghur, Mongolian — they're all threatened.  And you can be sure that if China invades and occupies Taiwan, Taiwanese (and all of the aboriginal languages of the island) will be under duress.  What is being done to Cantonese, Uyghur, and Mongolian is the way the CCP deals with the majority languages of its various cultural regions, which together constitute approximately half of China's total land area.  Tibet alone occupies roughly 13% of the total land area of the PRC (Xinjiang is 1/6th [16.6% of the whole of China].  Since seven of Asia's major rivers (the Yangtze, Yellow, Mekong, Salween, Irrawaddy, Brahmaputra, and Indus) originate in Bod, and "The Roof of the World" possesses many other valuable natural and strategic resources, what happens to the native tongue of its inhabitants is no mean matter.

In an update to "Morpho-phonologically AI", I wrote

Ironically, since this puzzle was vocalically inspired by the term "AI" , I'm guessing that current AI systems are not very good at solving (or creating) puzzles like this. I'll give it a try later today.

But it seems that I was wrong.

Will Shortz, "Sunday Puzzle: Artificially Confused", NPR Weekend Edition 8/9/2025:

The theme of today's puzzle is A.I. every answer is a familiar two-word phrase or name in which the first word has a long -A vowel sound and the second word has a long-I vowel sound.

Do you speak Yukon English? These researchers want to hear it
'Linguists know very, very little about what's going on with Englishes in the Canadian North,' researcher says
CBC News · Posted: Aug 10, 2025

If you're not quite sure where Yukon is, it's way up there in northwest Canada, between British Columbia to the south, Alaska to the west, and Northwest Territories to the east.  It's cold, bitterly cold in winter, the coldest place in North America, with the abandoned town of Snag dropping down to −63.0 °C (−81.4 °F) in February, 1947.  Believe it or not, it gets extreme high heat in May and June, with the Mayo Road weather station, located just northwest of Whitehorse, recording a temperature of 36.5 °C (97.7 °F) in June, 2004.

As you might expect, the population of Yukon is sparse, with an estimated total of 47,126 as of 2025.  But now it gets interesting, at least to me.

"OpenAI Brings Back Fan-Favorite GPT-4o After a Massive User Revolt", Gizmodo 8/10/2025:

After a disastrous 72 hours that saw its most loyal users in open revolt, OpenAI is making a major U-turn.

In a series of posts on X (formerly Twitter) Sunday, CEO Sam Altman announced that the company is bringing back its beloved older AI models, including GPT-4o, and dramatically increasing usage limits for paying subscribers, a clear peace offering to a furious customer base.

A PubMed search for the phrase "lived experience" finds 11,139 papers within the past year. And an esperr search shows that the relative frequency of this phrase has been increasing rapidly on PubMed:

It's not just in the fields covered by PubMed — the Social Science Research Network finds the phrase in 1,376 papers within the past year, including titles like "Distant Writing: Literary Production in the Age of Artificial Intelligence", "Civil V. Common Law: The Emperor Has No Clothes", and "The implementation of senior high school in the Philippines: An advantage or disadvantage to students' future opportunities".

Avram Pitch, "Meet President Willian H. Brusen from the great state of Onegon", The Register 8/8/2025:

OpenAI's GPT-5, unveiled on Thursday, is supposed to be the company's flagship model, offering better reasoning and more accurate responses than previous-gen products. But when we asked it to draw maps and timelines, it responded with answers from an alternate dimension.

AI Is Helping Historians With Their Latin
A new tool fills in missing portions of ancient inscriptions from the Roman Empire

By Nidhi Subbaraman Aug. 6, 2025

In recent years, we have encountered many cases of AI assisting (or not) in the decipherment of ancient manuscripts in diverse languages.  See several cases listed in the "Selected readings".  Now it's Latin's turn to benefit from the ministrations of artificial intelligence.

People across the Roman Empire wrote poetry, kept business accounts and described their conquests and ambitions in inscriptions on pots, plaques and walls.

The surviving text gives historians a rare glimpse of life in those times—but most of the objects are broken or worn.

“It’s like trying to solve a gigantic jigsaw puzzle, only there is tens of thousands more pieces to that puzzle, and about 90% of them are missing,” said Thea Sommerschield, a historian at the University of Nottingham.

Now, artificial intelligence is filling in the blanks.

An AI tool designed by Sommerschield and other European scientists can predict the missing text of partially degraded Latin inscriptions made hundreds of years ago and help historians estimate their date and place of origin.

As is usually the case, the Wikipedia article on the Chinese calendar is comprehensive and built on consensus.  It states:

The Chinese calendar, as the name suggests, is a lunisolar calendar created by or commonly used by the Chinese people. While this description is generally accurate, it does not provide a definitive or complete answer. A total of 102 calendars have been officially recorded in classical historical texts. In addition, many more calendars were created privately, with others being built by people who adapted Chinese cultural practices, such as the Koreans, Japanese, Vietnamese, and many others, over the course of a long history.

A Chinese calendar consists of twelve months, each aligned with the phases of the moon, along with an intercalary month inserted as needed to keep the calendar in sync with the seasons. It also features twenty-four solar terms, which track the position of the sun and are closely related to climate patterns. Among these, the winter solstice is the most significant reference point and must occur in the eleventh month of the year. Each month contains either twenty-nine or thirty days. The sexagenary cycle for each day runs continuously over thousands of years and serves as a determining factor to pinpoint a specific day amidst the many variations in the calendar. In addition, there are many other cycles attached to the calendar that determine the appropriateness of particular days, guiding decisions on what is considered auspicious or inauspicious for different types of activities.

From here (at least that's where I saw it):

The chapter on "Calendar and Chronology" in Brill's Encyclopedia of China Online (2009) was authored by Ho Peng Yoke (1926-2014), who was the Director of the Needham Research Institute from 1990-2001.  The first two paragraphs of Ho's chapter begin as follows:

The traditional Chinese calendar is lunisolar, i. e. it is based on both the movement of the moon and on what seems to be the orbit of the sun around the earth. The incommensurability of the lunar synodical period of 29.530587… days and the equinoctial year's 365.2421… days has always been the cause for numerous difficulties with respect to the establishment of a calendar in China. In order to replace the former calendars which after a time had lost their validity, roughly 100 different types of calendars were devised over a period of about 2000 years, many of which were never officially adopted. According to Joseph Needham, the history of calendar making is the consequence of attempts to "make the incompatible compatible."

Below is a guest post by Bob Ladd:

A few days ago I received an editorial decision letter from a journal, which included a request to deal with a few typos. I had begun a sentence with the phrase “In the interests of brevity,” and the editor wanted me to remove the final -s from the word “interests”. Since I know that the editor is not a native speaker of English, my first reaction was to ignore the request, but I thought I should back up my insistence that this was not a typo with some sort of evidence, so I searched for the phrases “in the interests of” and “in the interest of” on Google n-grams. To my surprise, I discovered that both versions of the expression occur, with a roughly 60:40 preference for the version with “interest”, and that this proportion has been roughly stable since the early 20th century. Since Google’s book corpus permits the user to distinguish British and American English, I could also see that the version with “interests” is more common in BrEng and the version with “interest” in AmEng, but that both versions occur in both varieties.