Alignment

2025-07-15 19:57
[syndicated profile] languagelog_feed

Posted by Mark Liberman

In today's email there was a message from AAAI 2026 that included a "Call for the Special Track on AI Alignment""

AAAI-26 is pleased to announce a special track focused on AI Alignment. This track recognizes that as we begin to build more and more capable AI systems, it becomes crucial to ensure that the goals and actions of such systems are aligned with human values. To accomplish this, we need to understand the risks of these systems and research methods to mitigate these risks. The track covers many different aspects of AI Alignment, including but not limited to the following topics:

  • Value alignment and reward modeling: How do we accurately model a diverse set of human preferences, and ensure that AI systems are aligned to these same preferences?
  • Scalable oversight and control: How can we effectively supervise, monitor and control increasingly capable AI systems? How do we ensure that such systems behave according to predefined safety considerations?
  • Robustness and security: How do we create AI systems that work well in new or adversarial environments, including scenarios where a malicious actor is intentionally attempting to misuse the system?
  • Interpretability: How can we understand and explain the operations of AI models to a diverse set of stakeholders in a transparent and methodical manner?
  • Governance: How do we put in place policies and regulations that manage the development and deployment of AI models to ensure broad societal benefits and fairly distributed societal risks?
  • Superintelligence: How can we control and monitor systems that may, in some respects, surpass human intelligence and capabilities?
  • Evaluation: How can we evaluate the safety of models and the effectiveness of various alignment techniques, including both technical and human-centered approaches?
  • Participation: How can we actively engage impacted individuals and communities in shaping the set of values to which AI systems align?

This reminded me of my participation a few months ago in the advisory committee for "ARIA: Aligning Research to Impact Autism", which was one of the four initiatives of the "Coalition for Aligning Science".

Alignment, like journey, is an old word that has been finding new meanings and broader uses over the past few decades. I suspect a role for Dungeons & Dragons, which has been impacting broader culture in many ways since the 1970s:

In the Dungeons & Dragons (D&D) fantasy role-playing game, alignment is a categorization of the ethical and moral perspective of player characters, non-player characters, and creatures.

Most versions of the game feature a system in which players make two choices for characters. One is the character's views on "law" versus "chaos", the other on "good" versus "evil". The two axes, along with "neutral" in the middle, allow for nine alignments in combination. […]

The original version of D&D (1974) allowed players to choose among three alignments when creating a character: lawful, implying honor and respect for society's rules; chaotic, implying rebelliousness and individualism; and neutral, seeking a balance between the extremes.

In 1976, Gary Gygax published an article title "The Meaning of Law and Chaos in Dungeons and Dragons and Their Relationships to Good and Evil" in The Strategic Review Volume 2, issue 1, that introduced a second axis of good, implying altruism and respect for life, versus evil, implying selfishness and no respect for life. The 1977 release of the Dungeons & Dragons Basic Set incorporated this model. As with the law-versus-chaos axis, a neutral position exists between the extremes. Characters and creatures could be lawful and evil at the same time (such as a tyrant), or chaotic but good (such as Robin Hood).

For some metaphorical extensions, see "Alignment charts and other low-dimensional visualizations", 1/7/2020.

A quick scan of Google Research results shows a steady increase in references including the word alignment, though 2014 or so. (I've includes counts for the word results to check for general corpus-size increases).

  YEARS   ALIGNMENT RESULTS  RATIO
1970-1974   19000   200000   10.53
1975-1979   31700   350000   11.04
1980-1984   56900   355000    6.24 
1985-1989  119999   305000    2.54 
1990-1994  207000   362000    1.75 
1995-1999  363000   546000    1.50
2000-2004  644000   799000    1.24 
2005-2009 1080000   856000    0.79 
2010-2014 1220000   760000    0.62 
2015-2019 1200000  1260000    1.05 
2020-2024  967000  1800000    1.86

And a graphical version:

It would be interesting to track the evolution of alignment and aligning usages across the decades in various cultural areas.

[syndicated profile] scalziwhatever_feed

Posted by Athena Scalzi

When author Josh Rountree’s story just wasn’t quite working, he decided to change his perspective. Literally. Travel back in time in the Big Idea for his newest novel, The Unkillable Frank Lightning, and see how switching things up narratively ended up being the solution to his problem.

JOSH ROUNTREE:

Well, I’m knee deep in monsters now, aren’t I?

A lot of my Big Ideas these days seem to involve them. For a while now I’ve been working on a series of monster stories set in long ago Texas.  I’ve tackled werewolves and snake-headed harvest gods.  Vengeful mermaids and The Creature from the Black Lagoon. Now I’ve worked my way up to one of the greatest monsters of all time.

Frankenstein!  (Cue lightning strike.) 

Or Frankenstein’s monster, I guess.

My Big Idea for The Unkillable Frank Lightning was to tell a version of the Frankenstein story, set in the Old West. I had plenty I wanted to say about death.  How breaking the fundamental laws of nature to bring a person back to life would be a really bad idea, with consequences that would reach far beyond the reanimated corpse and those responsible for his resurrection. Frank Lightning is not the only character in this story who has cheated death, and all carry that around like a weight on their souls. And I wanted to say a lot about the mythology of the Old West. How wild west shows and Hollywood movies have sold us an often sanitized version of the period, that centers the wrong heroes.

But also? I wanted to see my monster go on a rampage. I wanted to see what would happen when an unkillable person found himself in a gunfight. I wanted black magic and secret occult orders and townspeople with torches.

I wanted my monster to tell us his story.

But of course, the Big Idea doesn’t always unfold the way a writer expects it to. And the character you think is going to be at the center of it all isn’t always the voice that comes alive and demands to be heard. I worked several months, trying to tell this story through the monster’s point of view, and eventually realized it just wasn’t working.  I tried alternating points of view, trying to tell the story through the eyes of various characters. But one voice, that of Catherine Coldbridge, spoke louder than all the rest. And I realized she was my protagonist. She was the character to tell this story.

Catherine is my “mad scientist” in this tale. My Dr. Frankenstein stand-in.  She’s a doctor in the 1870’s, and she’s a member of an occult order called the Three Rose Temple.  Catherine is an orphan, and when she loses her husband too, it causes her to make one terrible decision that will haunt her for decades.

Catherine is terribly flawed, and desperate to make amends. She is determined and practical and willing to forgive anyone but herself. She is an exceptionally strong woman who has, for a time, given up on her life and let the world ruin her. But as she beings to tell her story, Catherine is finally beginning to emerge from that sorry state, and planning to take control again. Catherine is endlessly fascinating to me, and as soon as she started telling the tale, it poured out of her, and it poured out of me, and I knew we were in this together.

Catherine Coldbridge is not our typical pulp western hero. But who needs more cowboys in white hats? Who needs another hard man with a thousand-yard stare to ride in and save the day?

Catherine is so much more than that.


The Unkillable Frank Lightning: Amazon|Barnes & Noble|Bookshop|Powell’s

Author socials: Website|Bluesky|Facebook|Instagram

Read an excerpt.

Spinach smorgasbord

2025-07-15 10:00
[syndicated profile] languagelog_feed

Posted by Victor Mair

I want to thank Jonathan Silk (comment here) for pushing Popeye to further heights and deeper depths in our understanding of his favorite vegetable.  We're not "finiched" with spinach yet.

Now it's getting very interesting and confusing (Armenian is creeping in):

palak

English

Etymology

From Hindi पालक (pālak), from Sanskrit पालक्या (pālakyā).

Noun

palak (uncountable)

    1. (India, cooking) Spinach or similar greens (including Amaranthus species and Chenopodium album).

Turkish

Etymology

Borrowed from Armenian բալախ (balax), dialectal փալախ (pʻalax).

Noun

palak (definite accusative palağı, plural palaklar)

    1. (dialectal, Artvin) leaf
    2. (dialectal, Ahlat) a tender soft grass that grows in wet places
    3. (dialectal, Artvin) short grass that grows again after being mown
    4. (dialectal, Divriği) crop sown early that remains short and does not form ears
    5. (dialectal, Çemişgezek, Ağın, Şanlıurfa, Ankara) dry grass
    6. (dialectal, Ahlat) type of grass eaten by animals
    7. (dialectal, Ardanuç) time of crop to form ears
    8. (dialectal, Ovacık) dry grass

Related terms

References

    • palaḫ (II)”, in Türkiye'de halk ağzından derleme sözlüğü [Compilation Dictionary of Popular Speech in Turkey] (in Turkish), volume 9, Ankara: Türk Dil Kurumu, 1977, page 3382a
    • palak (IV), (V)”, in Türkiye'de halk ağzından derleme sözlüğü [Compilation Dictionary of Popular Speech in Turkey] (in Turkish), volume 9, Ankara: Türk Dil Kurumu, 1977, page 3382b
    • Dankoff, Robert (1995) Armenian Loanwords in Turkish (Turcologica; 21), Wiesbaden: Harrassowitz Verlag, § 41, page 24
    • Bläsing, Uwe (1992) Armenisches Lehngut im Türkeitürkischen am Beispiel von Hemşin (Dutch Studies in Armenian Language and Literature; 2) (in German), Amsterdam and Atlanta: Rodopi, § 101, pages 64–65

(Wiktionary)

Armenian

spanakh սպանախ, but we also have to contend with balax բալախ < Mid. Arm. balax բալախ < Old Arm. balax բալախ (common glasswort [Salicornia europaea]), which we have cited from Wiktionary several times above, without any indication of where it comes from.  Surely, though, it must be cognate with Hindi पालक (pālak) < Sanskrit पालक्या (pālakyā).  So how / when did it pass between Sanskrit and Armenian?

Japanese

Nathan Hopson:

Wikipedia gives the following for the etymology of ほうれん草
 
ホウレンソウ(菠薐草)の由来は、中国の唐代に「頗稜(ホリン)国」(現在のネパール)から伝えられたことによる[6]。後に改字して「菠薐(ホリン)」となり、日本では転訛して「ホウレン」となった[7][8]。「ホウレン」の語源は、「菠薐」の唐音とされ[6]、「法蓮草」は当て字とされる。
 
"The etymology of hōrensō 菠薐草 is from the Táng dynasty-era name Horin 頗稜(ホリン)国 (Nepal). The characters later changed to 菠薐(ホリン), which came to be pronounced hōren (not horin) in Japan… [The alternative]  法蓮草 is phonetic assignation."
 
The entry for 頗稜 includes this:
 
Compare modern Nepali पालुङ्गो (pāluṅgo, “spinach”), Assamese পালেং (paleṅ), Bengali পালং (paloṅ, “spinach (Spinacia oleracea)”). Possibly the source of 菠菜 (bōcài).

The following two Chinese blogs provide much interesting information and food for thought.

Wáng Guóliáng 王國良 (5/2/16) emphasizes the pentagonal cross section of the spinach stem to account for the lîng / ren 薐 syllable / morpheme in the Taiwanese and Japanese words for the plant.

 He was preceded in some of his ideas by Susan Plant Kingdom Blogspot (2/4/16), such as that 菠薐 was fancifully transcribed in Teochew / Chaozhou and other Southern Min topolects as bue-lóng 飛龍 ("flying dragon").

Korean

From Bob Ramsey:

sigeumchi 시금치

Korean word for 'spinach': it's 시금치. However, I confess I hadn't really thought about what the origin of the word was, so I immediately went to some reliable Korean lexical sources, and they all repeat what you'll find in a Wikipedia search, namely, that the word was borrowed from Chinese 赤根菜 'red-root vegetable', adding that it's probably borrowed from the Early Mandarin form of the word. It seems it was first attested in Korea in a 1517 Middle Korean text, where the form was written sikunchoy (transcribed in Martin's Yale Romanization).

There are still a lot of loose ends / fibers, so we may have to come back for a second / third helping later on.  For now, though:

Tilt Forums

 

Selected readings

[Thanks to Bob Ramsey

[syndicated profile] languagelog_feed

Posted by Victor Mair

This research investigates the semantic change and conceptual metaphor of the Thai word prèet (/เปรต/), which originates from the Pali-Sanskrit term meaning “departed.” The primary objective is to explore how the term’s meaning has shifted in contemporary Thai society, where it is now used pejoratively to criticize behaviors such as excessive greed, gluttony, immorality, and social deviance. Data for this study are drawn from both historical texts, particularly the Traibhumi Phra Ruang (a prominent Thai Buddhist text from the 14th-century Sukhothai period), and modern Thai linguistic usage. The analysis employs conceptual metaphor theory, focusing on metaphors like SOCIAL DEVIANCE IS MONSTROSITY, MORAL FAILURE IS DEGRADATION, GREED IS HUNGER, and SPIRITUAL LIMINALITY IS MONSTROSITY. to understand how these shifts reflect changing cultural and societal values. Additionally, Impoliteness Theory is applied to examine how prèet functions as a linguistic tool for social critique. Findings show that the semantic evolution of prèet reveals an intricate relationship between language, culture, and metaphor, as it transitions from a religious concept to a vehicle for social commentary. The implications of this study highlight the dynamic nature of language in reflecting societal shifts.

The socioeconomic background of people and how they use standard forms of language are not independent, as demonstrated in various sociolinguistic studies. However, the extent to which these correlations may be influenced by the mixing of people from different socioeconomic classes remains relatively unexplored from a quantitative perspective. In this work we leverage geotagged tweets and transferable computational methods to map deviations from standard English across eight UK metropolitan areas. We combine these data with high-resolution income maps to assign a proxy socioeconomic indicator to home-located users. Strikingly, we find a consistent pattern suggesting that the more different socioeconomic classes mix, the less interdependent the frequency of their departures from standard grammar and their income become. Further, we propose an agent-based model of linguistic variety adoption that sheds light on the mechanisms that produce the observations seen in the data.

  • "Re-Examining Second Language Acquisition of English Reflexives: New Evidence for Lexical Learning Driven Process and against First Language Transfer." Zeng, Li et al. Humanities and Social Sciences Communications 12, no. 1 (July 9, 2025): 1063. https://www.nature.com/articles/s41599-025-05466-8.

This study re-examines second language (L2) acquisition of English reflexives by testing 98 first language (L1)-Chinese learners of L2 English with different proficiency levels and 12 native English speakers as controls. Using a truth-value judgment task, we systematically tapped the learners’ judgments of various types of antecedents including long-distance objects. The results show that L2 English learners’ errors in referring English reflexives to long-distance antecedents cannot be due to L1 transfer of Chinese reflexive referential pattern. Instead, these errors align with those documented in the literature on native English children’s acquisition of reflexives. Moreover, as L1-Chinese learners’ English proficiency improved, most of them unlearned the errors, and performed similarly to native English adult controls. This developmental trajectory recapitulates the pattern seen in native English children’s acquisition of reflexives. These findings cast doubt on the view of L1 Chinese transfer and provide support for the Lexical Learning Hypothesis.

  • "Metaphor Interpretation in Jordanian Arabic, Emirati Arabic and Classical Arabic: Artificial Intelligence vs. Humans." Zibin, Aseel et al. Humanities and Social Sciences Communications 12, no. 1 (July 1, 2025): 942. https://www.nature.com/articles/s41599-025-05282-0.

This study examines how well humans, both Jordanians and Emiratis, and four AI tools—ChatGPT-4, ChatGPT-3.5, Google Gemini, and Ask PDF—can understand metaphors in Classical Arabic (CA) and its everyday forms in Jordanian Arabic (JA) and Emirati Arabic (EA). We tested fifty participants from Jordan and the UAE on their grasp of various colloquial and CA metaphorical expressions. Two distinct tests were employed, each comprising 40 items. Test 1 was administered to Jordanian participants and included 20 metaphorical expressions in Jordanian Arabic and 20 metaphorical expressions in Classical Arabic. Similarly, Test 2 was administered to Emirati participants and contained 20 expressions in Emirati Arabic and 20 expressions in Classical Arabic. The Mann–Whitney U test was employed to evaluate differences in accuracy and interpretation between AI tools and human participants from both regions in the contexts of colloquial and Classical Arabic. The results showed that participants from Jordan had a better understanding than the AI tools, likely due to their strong cultural background. In contrast, the Emirati participants performed similarly to the AI. The AI tools were more effective at interpreting CA metaphors compared to Emirati participants; AI tools are typically trained on diverse datasets and that usually leads to strong performance in interpreting formal or Classical Arabic expressions. These findings emphasize the need for improvements in AI models to boost their language processing abilities, as they often miss the cultural aspects required for accurately interpreting figurative language. This study adds to the ongoing discussion about AI and language interpretation, revealing both the potential and the obstacles AI faces when dealing with culturally rich and context-sensitive language.

Religions, topolects, language learning, AI — linguistics is exciting and ever changing, never boring.

[Thanks to Edward M "Ted" McClure]

[syndicated profile] scalziwhatever_feed

Posted by Athena Scalzi

I’ve been making more charcuterie boards than usual lately, and I’d like to think practice makes fairly decent, so I’d like to show y’all some of my recent spreads I’ve done for gatherings and parties and whatnot. I usually post them on Instagram and Bluesky, but just in case you missed them, you can get your fix right here and now!

For the 4th of July I was in Texas, and my friend hosted a party, for which I volunteered to supply some snackage for. Here’s the charcuterie board:

A large wooden serving board covered with meats and cheeses. There's a salami rose, a river of prosciutto, cubed Munster, triangular sliced chimichurri Gouda, rustically crumbled Kerrygold aged cheddar, a log of fig and honey goat cheese, a wheel of Brie, a small bowl of Castelvetrano olives, candied pecans, cherries, and small mounds of honeycomb scattered around, plus a tiny jar of Mike's Hot Honey.

For this board, I used prosciutto, salami, Munster, Kerrygold aged cheddar, Brie, chimichurri gouda, fig and honey goat cheese, candied pecans, Castelvetrano olives, Mike’s Hot Honey, Honeycomb, and cherries. Everything on there except the pecans I picked up at H.E.B.

This was the rest of what I served:

The charcuterie board from the previous photo is the main focus of this shot of the entire spread of food on a large butcher block counter. There's also a bowl of salsa and guacamole alongside a bowl of tortilla chips. There's a plate of watermelon, feta, and mint salad, plus a plate full of Caprese skewers. There's also a serving board of crackers on the far side to accompany the charcuterie board.

While the salsa and guac I bought pre-made from H.E.B., I did assemble the watermelon, feta, and mint salad and drizzled it with honey, and put together the caprese skewers with balsamic glaze. I honestly think this turned out really well! I was very happy with my summery salad and light bites.

Just a few days ago I got my AppyHour Box (which I have regrettably not been doing posts over lately!) and decided to make a little board for my dad and his friend that was visiting from out of town.

A long and narrow stone serving board filled to the brim with meats and cheeses. There's crumbled Togarashi cheese, a river of coppa, crumbled aged gouda, sliced smoked goat cheese, and a mound of prosciutto. There's handfuls of dried cherries around, as well as two mini jam jars that I forgot to take the lids off of for the shot.

This board consisted of a Togarashi cheese, an aged gouda, a smoked goat cheese, dried cherries, coppa, and prosciutto (I think it was a Calabrian Chili prosciutto?). The two jams I forgot to take the lids off of are a caramelized pear and honey spread, and a raspberry hibiscus jam. I thought this was a cute little lunch for my dad and his guest, and I’m glad the enjoyed it.

Finally, this past weekend, I hosted a friend’s baby shower at the church. She said she expected around fifty people to attend, and I can say with confidence I’ve never tried to make a spread for that many people before. I was definitely intimidated, but I was determined to make an approachable spread that would appeal to the masses and not spend hundreds of dollars doing it.

I didn’t capture everything, but here’s the gist of how it turned out:

A long white counter covered in parchment paper with a ton of food on top of it. There's a vegetable assortment consisting of carrots, celery, cherry tomatoes, cucumber, and bell peppers, all surrounding a dish of dill dip. There's a bowl of salsa visible, plus some crackers in the shot, too. At the far end there is a charcuterie section that I will go into detail of in the next photo.

And of course, a close up:

A giant salami rose is the main focus of the charcuterie spread, the middle of which is filled with some prosciutto. It's surrounded by crumbled Asiago, rosemary almonds, cubed jalapeno havarti, grapes, a log fig goat cheese, more prosciutto, cherries, crumbled cranberry cheddar, and pimento stuffed olives. There's also a fig and orange spread, and whole grain mustard.

The spread contained Asiago, jalapeno Havarti, fig goat cheese, cranberry cheddar, smoked cheddar, cherries, grapes, rosemary almonds,  chocolate covered almonds, hard salami, prosciutto, pimento stuffed olives, fig spread, and whole grain mustard.

There was also cucumbers, bell peppers, baby carrots, celery, cherry tomatoes, dill dip, salsa, tortilla chips, garden vegetable entertainment crackers, fig and sesame crisps, honey mustard mini pretzels, rosemary flatbread crackers, coconut macaroons, chocolate covered pretzels, chocolate covered shortbread cookies, and chocolate covered Belgian waffle cookies.

Other than the tortilla chips, whole grain mustard, fig spread, rosemary almonds, and chocolate covered pretzels, I bought everything at Aldi, and despite buying doubles if not three of absolutely everything I listed, my total came out to $220. I was able to make this huge spread and refill it when it got low and feed 50 guests for just over $200. Who knew Aldi was so cool?! I spent about fifty bucks more than that on my spread for the Texas party, and that was only to feed about ten people.

The best thing on the spread from Aldi was the honey mustard mini pretzels, or the chocolate covered waffle cookies. I am definitely going to be stopping at Aldi more often for some surprisingly cheap and yummy treats.

What item looks the best to you? What’s your go-to cheese to serve for entertaining guests? Am I the only one who didn’t realize how neat Aldi was? Let me know in the comments, and have a great day!

-AMS

[syndicated profile] scalziwhatever_feed

Posted by Athena Scalzi

Author Marie Vibbert is back on the blog today with a fresh new novel that’s bigger and badder than ever. Dive into her Hellcats universe once more and see the world from an all new perspective in Andrei and the Hellcats.

MARIE VIBBERT:

Why are all sex robots depicted as miserable? Wouldn’t they just program them to be really horny? Thus, I created the character of Andrei: a sex-positive sex robot who loves his job, and humans. All the humans. When Galactic Hellcats was released in 2021, people responded warmly to this minor character, and when I asked, “What do you want in a sequel?” one of the top responses was, “More Andrei! And let him have sex! Why do you have two sex bots in this book and no one gets past second base?”

Why? Because my dad might read this book! I didn’t say that. I blushed and affirmed I would try harder in the sequel to be as uninhibited as Andrei.

And sequel time has come! My first goal was a book that wouldn’t require reading the previous one, and so Andrei and the Hellcats is from Andrei’s perspective, as a relative outsider. As I drafted, I realized that my little idea about sex robot preferences led into a bigger idea: how do we form our moral preferences? Can robots (or AI) have a conscious choice in their morality? What does that say about morality itself?

I know, that sounds heavy for a book about a sex robot enlisting a space biker gang to rescue his sister from an evil queen, but come drop down this mental rabbit hole with me.

As I fleshed out Andrei’s character to take center stage, I had to confront aspects of his life, personality, and preferences. He likes sex, a lot. Check. So why does he work as a hospitality manager at a space station instead of enjoying an all he can shag buffet at a brothel? Did he have a choice, or was he built for this role? Was he hired or purchased?

All of these questions funnel down into the intersection of consent and capitalism. Andrei cares a lot about consent in sex; he has whole libraries of code for it. He recognizes hesitation, the body language of distress and coercion. “I don’t enjoy inflicting harm,” he laments at one point, “I’m really only comfortable when there are safe words.” Yet he has never examined his own consent to play the role assigned to him, until the evil queen comes along and kidnaps him and his sister. He thinks Queen Jasmine of Ratana is simply roleplaying “Evil Queen and the mindless sexbot” until after their tryst, when she waves her hand and says “deactivate yourself.” Through her words and actions she makes it clear she doesn’t think Andrei is self-aware, and he realizes she felt that way the entire time they were fooling around. This prompts a moral crisis: have other clients of his mistaken play for reality? Does he need to update his most sacred algorithms?

Poor Andrei! What is sapience? What is consent? I wanted to have him follow this rabbit hole to money as a motivation, anti-capitalist gal that I am. To that end, I had this fun idea to have him in conflict with the Hellcats. They are, to put it mildly, uninterested in property rights, especially kleptomaniac Ki. Wouldn’t it be a hoot if, having finally enlisted the gang, he’d interrupt the ensuing caper with, “Excuse me, does that belong to you? Put it back.” 

To quote Ki, “Ugh, it’s like taking my social worker on a heist!”

Nothing makes a plot outline happier than juicy, theme-relevant complications! But I found the right opening for my little gag difficult to find, and part of that was, well, I was already forcing Andrei to confront his programming through every step of the plot. First the evil queen, then I put him on his own on a strange planet with nothing but his keen fashion sense and gift of gab. How can he find where they’ve taken her? How can he begin to save her? He has no local currency and has to contemplate breaking his programmed reverence for property rights to even get to a point where he can start searching in earnest. He wonders why he can’t bring himself to steal what he needs, why he was programmed to be a good little capitalist. “Were we designed to desire things to keep us working? Could we learn to do without designer clothes and porn subscriptions?”

By the time he gets to the Hellcats, he’s ready to hear Ki out when she takes it upon herself to turn him to the lawless side of the force. He has already seen that laws can be unjust; the queen’s law declares him and his sister property! When Ki compares hoarding money to hoarding kisses, Andrei accepts that ownership is not as important as good snogs… or the safety of sapient beings.

As I was writing, I found myself a little envious of Andrei. He can consciously edit his moral programming when he discovers a bias in it, while I make the same gaffs nine or ten thousand times before I learn. Well, so do the Hellcats. They have those relationship arcs to get through, all that learning what and who to prioritize. So while he learns from them, Andrei gets to drop some truth bombs in return. “Darling, I get it. When you and your brother left the factory… I mean, when you were born, your settings were the same…. Then you were sent out into the world, and your programs updated.”

Ninety percent of writing a novel is making decisions. Where could I fit in my little anti-theft Andrei gaff? Well, I couldn’t. Then I realized he still cared about the rule of law. Ah ha! The confrontation now comes not from stealing, but from breaking and entering, and I got to use all the snarky lines I had daydreamed.

Andrei sighs, “Humans take so long to make decisions! How did they ever get around to inventing us?”

(… she wrote, thinking about her own plot outlining.)

That accomplished, I was free to make everything worse! Bwah ha ha. Have to push all those moral lessons harder with some external examples, right? So they all get captured in a forced labor camp. It’s dark. I found myself pausing on a scene where an explosive implant takes out a guy’s arm because sometimes they just go off and thinking, “Um, self? Is this still a lighthearted space romp?” And lo, the moral quandary has come home: the choices I make as an author can reflect my own morality (or lack thereof.) 

I couldn’t just have them blow up their way out of there, leaving the other slaves behind. Crap. I had to go back and change the entire ending heist to reflect my values of collective action. Andrei, I hope, would be proud. 

Andrei gets through the plot, the Hellcats get through the plot, and the author gets through the plot: each a little wiser for it.

The book should be available wherever books are sold. Have your local independent bookstore order it, or your library! 

And we all snarked happily ever after.


Andrei and the Hellcats: Amazon|Barnes & Noble|Powell’s

Marie Vibbert: Website|Facebook|Instagram

[syndicated profile] languagelog_feed

Posted by Victor Mair

Bibliographical cornucopia for linguists, part 1

Since we have such an abundance of interesting articles for this fortnight, I will divide the collection into two parts, and provide each entry with an abstract or paragraph length quotation.

A fundamental question in word learning is how, given only evidence about what objects a word has previously referred to, children are able to generalize to the correct class. How does a learner end up knowing that “poodle” only picks out a specific subset of dogs rather than the broader class and vice versa? Numerous phenomena have been identified in guiding learner behavior such as the “suspicious coincidence effect” (SCE)—that an increase in the sample size of training objects facilitates more narrow (subordinate) word meanings. While SCE seems to support a class of models based in statistical inference, such rational behavior is, in fact, consistent with a range of algorithmic processes. Notably, the broadness of semantic generalizations is further affected by the temporal manner in which objects are presented—either simultaneously or sequentially. First, I evaluate the experimental evidence on the factors influencing generalization in word learning. A reanalysis of existing data demonstrates that both the number of training objects and their presentation-timing independently affect learning. This independent effect has been obscured by prior literature’s focus on possible interactions between the two. Second, I present a computational model for learning that accounts for both sets of phenomena in a unified way. The Naïve Generalization Model (NGM) offers an explanation of word learning phenomena grounded in category formation. Under the NGM, learning is local and incremental, without the need to perform a global optimization over pre-specified hypotheses. This computational model is tested against human behavior on seven different experimental conditions for word learning, varying over presentation-timing, number, and hierarchical relation between training items. Looking both at qualitative parameter-independent behavior and quantitative parameter-tuned output, these results support the NGM and suggest that rational learning behavior may arise from local, mechanistic processes rather than global statistical inference.

A crucial feature of language is the ability to communicate cognitive goals to a specific audience, i.e. goal-directed intentionality. Core criteria for this ability include (i) audience directedness: signalling in the presence of an attentive audience, (ii) persistence: continuing signalling until goals are met, and (iii) elaboration: using new signals following communicative failure. While intentional use has been demonstrated in individual gestures in some non-primates, primates—in particular apes—show this ability across many gestures. But is goal-directed intentionality across many gestures restricted to primates? We explored whether savannah elephants use many gestures with goal-directed intentionality. We presented semi-captive elephants with desired and non-desired items, recording their communicative attempts when an experimenter met, partially met or failed to meet their goal of getting the desired item. Elephants used 38 gesture types almost exclusively when a visually attentive experimenter was present, demonstrating audience directedness. They persisted in gesturing more when their goal was partially as compared with fully met but showed no difference in persistence when the goal was met or not met. Elephants elaborated their gesturing when their goal was not met. We find goal-directed intentionality across many elephant gestures and reveal that elephants, like apes, assess the communicative effectiveness of their gesturing.

The extensive vocal repertoire of mountain chickadees has yet to be fully documented. There are five basic categories of call types:

    • Contact calls: communicate identity, sort of like a name, and location.
    • “Chick-a-dee” calls: coordinate flock movement and communicate a variety of complex information about the environment, from food availability to predator presence and type.
    • Alarm calls: alert others of the presence of a predator.
    • Begging calls: used by chicks or females to elicit feeding behavior from males.
    • Gargle calls: advertise dominance over other individuals in a flock, primarily used by males.

“Chick-a-dee” calls contain several elements resembling the basic elements of human grammar. Essentially, the various sounds a chickadee utters mean different things, similar to words in human languages. And the way that a chickadee combines these sounds changes the meaning. Word order matters, just like grammar matters in human language. If a chickadee were to phrase its calls in the wrong note order, the call would no longer convey the same meaning, even if composed of the same elements.

The author distinguishes between the two large categories of songs and calls.  A video is included; in it you can hear the author distinguish and mimic different types of bird talk

In 2015, the editors of the Oxford English Dictionary went all-in on the still-novel phenomenon of emoji. That year, the guardians of the venerable OED named the FACE WITH TEARS OF JOY emoji ()—now, as then, the world’s most popular emoji—as Word of the Year, beating out such zingers as “ad blocker,” “Brexit,” “lumbersexual,” “on fleek,” and “sharing economy.” For emoji to be blessed in this way by the OED was remarkable enough, but it also invited a question: if was a word, did that make emoji a language?

This morning I stood out on my stoop and listened to a flock of crows conversing.  After about 5 minutes, I could distinguish a variety of different caws and calls.  Some were soft and subdued, almost like whispers, others were excited and raucous.  I was convinced that, if I listened to them for half a day and observed their behavior in relation to the caws and calls, I would be able to figure out what they were communicating to each other.

Then I sat down at my computer and wrote some messages to friends.  It has become my custom to follow my signature with the emoji for a snail, which happens to be my logo, and has been for many decades.  I don't know if it will come through in WordPress, but I'll give it a try:  VHM .  That is pronounced "snail / wōniú / ghongha / ciplēkirā / katatsumuri / etc., etc., etc."

[Thanks to Edward M "Ted" McClure] 

Recursive summarization

2025-07-14 10:20
[syndicated profile] languagelog_feed

Posted by Mark Liberman

Today's SMBC:

The mouseover title: "I saw an article that said it was a 3 minute read then offered an AI summary, and I believe it may be included in an eventual epitaph for civilization."

The Aftercomic:

Of course it's not just news…

So far, none of the AI helpers offer emotionally indicative grunting noises. At least not as far as I know, though maybe Grok is on it?

Accurate graphical summarizations would actually sometimes be useful. Current AI poster generation is about posters to advertise an event, not the sort of figures that would be useful as a summary of complex material, e.g. here:

[syndicated profile] languagelog_feed

Posted by Mark Liberman

Joel Becker et al., "Measuring the Impact of Early-2025 AI on Experienced Open-Source Developer Productivity", METR 7/10/2025:

Despite widespread adoption, the impact of AI tools on software development in the wild remains understudied. We conduct a randomized controlled trial (RCT) to understand how AI tools at the February–June 2025 frontier affect the productivity of experienced open-source developers. 16 developers with moderate AI experience complete 246 tasks in mature projects on which they have an average of 5 years of prior experience. Each task is randomly assigned to allow or disallow usage of early-2025 AI tools. When AI tools are allowed, developers primarily use Cursor Pro, a popular code editor, and Claude 3.5/3.7 Sonnet. Before starting tasks, developers forecast that allowing AI will reduce completion time by 24%. After completing the study, developers estimate that allowing AI reduced completion time by 20%. Surprisingly, we find that allowing AI actually increases completion time by 19%—AI tooling slowed developers down. This slowdown also contradicts predictions from experts in economics (39% shorter) and ML (38% shorter). To understand this result, we collect and evaluate evidence for 20 properties of our setting that a priori could contribute to the observed slowdown effect—for example, the size and quality standards of projects, or prior developer experience with AI tooling. Although the influence of experimental artifacts cannot be entirely ruled out, the robustness of the slowdown effect across our analyses suggests it is unlikely to primarily be a function of our experimental design.


(See also this version…)

A graph of their results:

This Swedish thesis confirmed those survey results, but did not test actual development time — the METR results show that users' opinions about productivity are by no means always accurate. Of course those METR results were based on Claude 3.5 — Claude 4 might be different. Or might not be. And maybe making coders feel good is worth a 19% productivity decline…

(Here's someone who's really enthusiastic about using Claude Code — I assume the latest version — but again, it's opinion and not productivity measurement.)

Articles like "Generative AI is Turning Publishing Into a Swamp of Slop" (7/10/2025) suggest that LLMs are enhancing "productivity" in certain corners of the publishing industry. So it would be interesting to understand (beyond the obvious reasons) why coding is different, and what the implications are for other applications.

The METR discussion includes some attempts to "very roughly gesture at some salient important differences", which would apply in other fields. My own concern, based on considerable experience, is that the motives of the administrators (and consultants) responsible for tool choice are pretty clearly not always aligned with productivity improvements. Or user satisfaction, for that matter…

Asterisk the Gaul

2025-07-13 10:49
[syndicated profile] languagelog_feed

Posted by Victor Mair

A learned friend recently sent me a draft composition on medieval Chinese history in which he referred to "*" as an "asterix".  This reminded me that ten years ago I wrote a post, "The many pronunciations of '*'" (12/17/15), on this subject and we had a lengthy, vigorous discussion about it.

Given that lately we've been talking a lot about Celts, Galatians, and so on, I think it is appropriate to write another post on Asterix the Gaul, that famous French comic book character, and how he got his name.  Also inspired / prompted by Chris Button's latest comment.

I often hear "*" pronounced "asterix" or "asterick", and so on (e.g., "astrisk" [two syllables], esp. in rapid speech).  It's hard even for me to pronounce "*" or type the symbol those ways, so ingrained is the pronunciation "as-ter-isk".

First, a little refresher course on "*", how / when it came about, how it is written, how it is pronounced, and what it signifies:

The asterisk (/ˈæstərɪsk/ *), from Late Latin asteriscus, from Ancient Greek ἀστερίσκος, asteriskos, "little star", is a typographical symbol. It is so called because it resembles a conventional image of a heraldic star.

Computer scientists and mathematicians often vocalize it as star (as, for example, in the A* search algorithm or C*-algebra). An asterisk is usually five- or six-pointed in print and six- or eight-pointed when handwritten, though more complex forms exist. Its most common use is to call out a footnote. It is also often used to censor offensive words.

The asterisk was already in use as a symbol in ice age cave paintings. There is also a two-thousand-year-old character used by Aristarchus of Samothrace called the asteriskos, , which he used when proofreading Homeric poetry to mark lines that were duplicated. Origen is known to have also used the asteriskos to mark missing Hebrew lines from his Hexapla. The asterisk evolved in shape over time, but its meaning as a symbol used to correct defects remained.

In the Middle Ages, the asterisk was used to emphasize a particular part of text, often linking those parts of the text to a marginal comment. However, an asterisk was not always used.

One hypothesis to the origin of the asterisk is that it stems from the 5000-year-old Sumerian character dingir , though this hypothesis seems to only be based on visual appearance.

(Wikipedia)

Now, on to how Asterix the Gaul and the other characters in the comic got their names.

All the fictional characters in Asterix have names which are puns on their roles or personalities, and which follow certain patterns specific to nationality. Certain rules are followed (most of the time) such as Gauls (and their neighbours) having an "-ix" suffix for the men and ending in "-a" for the women; for example, Chief Vitalstatistix (so called due to his portly stature) and his wife Impedimenta (often at odds with the chief). The male Roman names end in "-us", echoing Latin nominative male singular form, as in Gluteus Maximus, a muscle-bound athlete whose name is literally the butt of the joke. Gothic names (present-day Germany) end in "-ic", after Gothic chiefs such as Alaric and Theoderic; for example Rhetoric the interpreter. Greek names end in "-os" or "-es"; for example, Thermos the restaurateur. British names usually end in "-ax" or "-os" and are often puns on the taxation associated with the later United Kingdom; examples include Mykingdomforanos, a British tribal chieftain, Valuaddedtax the druid, and Selectivemploymentax the mercenary. Names of Normans end with "-af", for example Nescaf or Cenotaf. Egyptian characters often end in -is, such as the architects Edifis and Artifis, and the scribe Exlibris. Indic names, apart from the only Indic female characters Orinjade and Lemuhnade, exhibit considerable variation; examples include Watziznehm, Watzit, Owzat, and Howdoo. Other nationalities are treated to pidgin translations from their language, like Huevos y Bacon, a Spanish chieftain (whose name, meaning eggs and bacon, is often guidebook Spanish for tourists), or literary and other popular media references, like Dubbelosix (a sly reference to James Bond's codename "007").[68]

Most of these jokes, and hence the names of the characters, are specific to the translation; for example, the druid named Getafix in English translation – "get a fix", referring to the character's role in dispensing the magic potion – is Panoramix in the original French and Miraculix in German.[69] Even so, occasionally the wordplay has been preserved: Obelix's dog, known in the original French as Idéfix (from idée fixe, a "fixed idea" or obsession), is called Dogmatix in English, which not only renders the original meaning strikingly closely ("dogmatic") but in fact adds another layer of wordplay with the syllable "Dog-" at the beginning of the name.

The name Asterix, French Astérix, comes from astérisque, meaning "asterisk", which is the typographical symbol * indicating a footnote, from the Greek word ἀστήρ (aster), meaning a "star". His name is usually left unchanged in translations, aside from accents and the use of local alphabets. For example, in Esperanto, Polish, Slovene, Latvian, and Turkish it is Asteriks (in Turkish he was first named Bücür meaning "shorty", but the name was then standardised). Two exceptions include Icelandic, in which he is known as Ástríkur ("Rich of love"), and Sinhala, where he is known as සූර පප්පා (Soora Pappa), which can be interpreted as "Hero". The name Obelix (Obélix) may refer to "obelisk", a stone column from ancient Egypt (and hence his large size and strength and his task of carrying around menhirs), but also to another typographical symbol, the obelisk or obelus ().

For explanations of some of the other names, see List of Asterix characters.

The translators who created all of these different versions in so many languages are to be commended for maintaining the humorous spirit of the onomastics in the original.  "Vive 'Astérix le Gaulois'!"

 

Selected readings

 

[syndicated profile] scalziwhatever_feed

Posted by John Scalzi

So, if I want to buy a 100 Grand candy bar (or an Almond Joy, or a Heath Bar, or whatever), and I go to the candy aisle in the local gas station, or the one in the IGA, the candy bar will cost about two bucks, give or take. But! If I go to the local dollar store to buy the same brand of candy, packaged as four to six “fun sized” individually wrapped bars, each of these packages are $1.25, or $5 if I buy five. The aggregate amount of candy by weight is pretty much the same as if I bought the full-sized candy bar, but because they are packaged differently, it costs half as much. Does this make sense? No! It does not! I mean yes, I understand that one is positioned as an impulse buy and one is not, I do understand the psychology of the supermarket. I get it, truly I do. But it still boggles my mind.

Likewise if I go to the store to buy a 12-pack of Coke Zero, it will be somewhere between $8 and $10, but if Krissy goes to Kroger on a particular day and shows them her Kroger card (or whatever) then she can get the “buy two, get three free” discount, which again means the actual cost of the 12-pack is 40% of what its usual cost is. It works similarly for lots of other things, including things that are not junk food or drink.

“Congratulations, Scalzi, you’ve discovered coupons,” I hear you say. Look, I’m not saying any of this is new. But it does seem to me the variance in pricing is more significant now than it was before. I’m not exactly what you would call a price-sensitive individual these days, but I still finally broke down and got myself a CVS card because the difference in cost between having that card, and not, was high enough that my brain rebelled against needlessly spending that much more.

(Yes, I’m aware that CVS, Kroger, et al are data farming what I purchase. As a practical matter, I don’t really care if CVS learns I’m buying Doritos; they were tracking the UPC when I went to check out anyway. And as a general matter I’m not purchasing anything in a supermarket or pharmacy that I want to hide from data crunching.)

I know this is a bit of an aimless rant, but I think what I’m really getting at is that the answer to the question “why are things so expensive right now” really is “because fuck you, that’s why.” That candy bar quite evidently doesn’t need to be $2; that 12-pack of soda doesn’t need to be $8, and there are a lot of people who can’t afford the clearly arbitrary high prices that things have, who have to pay them anyway. It’s annoying for me, but for someone else it might mean skipping a meal or two, or more, here and there. It doesn’t seem fair, and it doesn’t seem right.

— JS

Steele v. Monboddo

2025-07-12 16:40
[syndicated profile] languagelog_feed

Posted by Mark Liberman

In "AI win of the week" I explored the inter-personal dimensions of Rousseau's 1754 contention that "there is neither rhythm nor melody in French music, because the language is not capable of them". In the comments, AntC objected that "But, but. Rousseau wrote an opera, in French, to his own Libretto. audio + full score available on Youtube".

For now, I have only two comments on this. First, trolls are often happy to abandon consistency in the service of pwning their audience. And second, the 1754 edition of Rousseau's screed, published two years after the debut of his opera, goes into considerable detail about how he painfully transferred the musicality of Italian prosody to the composition and performance of a work with French lyrics.

But rather than diving further into Rousseau's argument about the relative musicality of different languages' prosody, the point of today's post is to note its resonance with another mid-18th century prosodic dispute, namely Joshua Steele's refutation of James Burnett's claim that English prosody gives its syllables "nothing better than the music of a drum, in which we perceive no difference except that of louder or softer, according as the instrument is more or less forcibly struck".

My connection with this argument began in 1973, when I was trying to learn something about English intonation. The (very small) relevant section of the stacks in MIT's library happened to have (a facsimile edition of) Steele's 1775 work, An Essay Towards Establishing the Melody and Measure of Speech to be Expressed and Perpetuated by Peculiar Symbols. I read it carefully and learned a lot.

One of the first things I learned was Steele's motivation for the enterprise. His description starts this way:

[M]y learned and honoured friend Sir John Pringle, President of the Royal Society, desired me to give him, in writing, my opinion on the musical part of a very curious and ingenious work lately published at Edinburgh, on The Origin and Progress of Language, which I should find principally in part II. book ii, chap. 4. and 5. wherein several propositions, denying that our language has either the melody of modulation, or the rhythmus of quantity, gave occasion to the following systematic attempt to prove the contrary.

At that point I paid no further attention to Monboddo's  "very curious and ingenious work", partly because I was convinced by Steele's arguments against it, and partly because the library didn't have a copy of (any of the six volumes) of  the work in question.  But digital facsimiles are now easy to come by, and so I've taken a look at the stuff that led Pringle to question Steele, and led Steele to write his essay.

The author was James Burnett, Lord Monboddo (see Language Hat's post on the interpretation of the name), and the six volumes of The Origin and Progress of Language were published between 1774 and 1792. Volume II was part of the initial 1774 publication,  giving Steele only a year to prepare and publish his response a year later.

The relevant pages of Monboddo's work are here, if you really want to slog through them. The critical passage comes at the end:

But what do we mean then when we speak so much of accent in English, and dispute whether a word is right or wrong accented? My answer is, That we have, no doubt, accents in English, and syllabical accents too: but they are of a quite different kind from the antient accents ; for there is no change of the tone in them; but the voice is only raised more, so as to be louder upon one syllable than another. Our accents therefore fall under the first member of the division of sound, which I made in the beginning of this chapter, namely, the distinction of louder, and softer, or lower.

That there is truly no other difference, is a matter of fact, that must be determined by musicians. Now I appeal to them, whether they can perceive any difference of tone betwixt the accented and unaccented syllables of any word; and if there be none, then is the music of our language in this respect nothing better than the music of a drum, in which we perceive no difference except that of louder or softer, according as the instrument is more or less forcibly struck.

Of course Monboddo is also wrong that drum sounds can differ only in loudness and not in frequency content — watch and listen here for a refutation, or here for another (and more linguistically relevant) one.

But in fairness to Lord Monboddo, his "English accents are like drum beats" claim is not quite so idiotically tone-deaf as it seems, since he makes two other claims earlier that fuzzify it somewhat. One is the idea that English pitch changes exist, but only as "the tones of passion or sentiment":

As to accents in English, Mr Foster, from a partiality, very excusable, to his country, and its language, would fain persuade us, that in English there are accents such as in Greek and Latin. But to me it is evident that there are none such; by which I mean that we have no accents upon syllables, which are musical tones, differing in acuteness or gravity. For though, no doubt, there are changes of voice in our speaking from acute to grave, and vice versa, of which a musician could mark the intervals, these changes are not upon syllables, but upon words or sentences. And they are the tones of passion or sentiment, which, as I observed, are to be distinguished from the accents we are speaking of.

And he adds

[T]here is another difference betwixt our accents and the antient, that ours neither are, nor can, by their nature, be subjected to any rule ; whereas the antient, as we have seen, are governed by rules, and make part of their grammatical art.

Anyhow, Steele took it on himself not only to show that English had "melody and measure", but also to provide "peculiar symbols" for expressing it. The pages where he introduces his notation are here.  His instrumental analysis method — using a bass viol — is better than any other one that would be available for the following couple of hundred years:

Along with some other notational inventions, the result is transcriptions like this:

And his conclusions:

1st, That the sound or melody of speecb is not monotonous, or confined like the found of a drum, to exhibit no other changes than those of loud or soft.

2dly, That the changes of voice from acute to grave, and vice versa, do not proceed by pointed degrees coinciding with
the divisions of the chromatico-diatonic scale; but by gradations that seem infinitely smaller (which we call slides); and though altogether of a great extent, are yet too rapid (for inexperienced ears) to be distinctly sub-divided; consequently they must be submitted to some other genus of music than either the diatonic or chromatic.

3dly, That these changes are made, not only upon words and upon sentences, but upon syllables and monosyllables. Also,

4thly, and lastly, That in our changes on syllables or monofyllables, the voice slides, at least, through as great an extent as the Greeks allowed to their accents; that is, through a fifth, more or less.

Not having access to a bass viol, I followed Steele's example using a computer program for a PDP-9, allowing figures like this one from my 1975 dissertation:

Though for presentational purposes, Ivan Sag used a kazoo in presentations like this one:

 

Tâigael, part 2

2025-07-11 22:05
[syndicated profile] languagelog_feed

Posted by Victor Mair

[This is a guest post by Chau Wu]

The Presbyterian Church in Taiwan (TPC) was first planted by British missionaries in Tainan, which later expanded to all southern parts of Taiwan, constituting the present Southern Synod of TPC. The most important pioneer among them was the Scottish missionary Rev. Thomas Barclay who worked in Taiwan-Fu (the present Tainan).  He was born in Glasgow, and matriculated at the University of Glasgow. While there, he studied under Sir William Thomson, later Lord Kelvin [according to Wikipedia]. The celebrated Lord Kelvin reminds me of the absolute zero degree in physical chemistry and the electric cable equation as the underpinning of the Transatlantic cable as well as the conduction of electric impulses along nerve fibers. 

The majority of Taiwanese Christians are Presbyterian. Those who are aware of the church history always associate the solid establishment of TPC with Barclay, and his name shows up in Barclay Memorial Park and Barclay Memorial Church in Tainan City. The following picture shows the church name on the lintel above the door of the church in POJ: Tâi-Lâm Tang-M̂ng Pa-Khek-Lé Kì-Liām Kàu-Hōe (Tainan East Gate Barclay Memorial Church). You see there are no Sinographs such as the expected 台南東門巴克禮紀念教會 on it!  Only the Romanized Taiwanese script.  Since Barclay was a Scot, I would say this title on the church lintel may qualify as a good example of Tâigael.

(From Wikipedia)

Interestingly, the present Northern Synod of TPC had its beginning in 60 churches established by Rev. George Mackay, a Scottish-Canadian who graduated with a theological degree from Princeton Theological Seminary and received postgraduate training in theology at Edinburgh University. His parents, George Mackay and Helen Sutherland, emigrated from Sutherlandshire of Scotland. So, here we have another case of Tâigael linkage, albeit by a generation removed from Scotland. On a personal note, my grandfather was the second earliest of his many students and was among the first 5 Taiwanese of northern Taiwan to receive baptism from him at Tamsui. The Presbyterian church at Tamsui has a marble plaque recording this event.

Perhaps to those of us among Language Log readers who are interested in Taiwanese linguistics, one of the most important figures is the Scottish missionary Rev. William Campbell.

William Campbell (ChinesePe̍h-ōe-jīKam Ûi-lîm) (1841–1921) was a Scottish Presbyterian missionary to Formosa (Qing Taiwan). He wrote extensively on topics related to Taiwan and was also responsible for founding the island's first school for the blind.[3] Interested in the early history of the island (particularly the Dutch era), his knowledge of the time was such that he was called "without doubt the greatest authority on this subject living".[4] He was probably the first European to see Sun-Moon Lake,[5] which he named Lake Candidius in honour of the seventeenth century Dutch missionary George Candidius.

Campbell arrived in Qing-era Taiwan in 1871 to begin his mission in southern Taiwan, being stationed in Taiwan-fu, the capital of Taiwan Prefecture (modern-day Tainan) and serving both Han Chinese and Taiwanese aborigines in the area.[6] He was a contemporary of Thomas BarclayJames Laidlaw Maxwell and George Leslie Mackay, who were all engaged in missionary work in Taiwan.

(source)

A major contribution of his, “A Dictionary of the Amoy Vernacular Spoken throughout the Prefectures of Chin-chiu, Chiang-chiu and Formosa (Taiwan) 廈門音新字典” stands out as the Tâigael example par excellence. This dictionary, compiled by a Scot, is ironically for me a Taiwanese, the go-to handbook for studying Taiwanese historical linguistics.

Tâigael works wonders in the spread of gospel to Taiwan.

 

Selected readings

[syndicated profile] languagelog_feed

Posted by Victor Mair

As announced in the title of the first post on this subject, my aim is to understand where the Galatians originated and how / why they migrated to where they were when Apostle Paul wrote his epistle to them.  Since I was apparently insufficiently clear about both of those purposes in part 1, in this follow-up post I will provide additional scholarly material.  Inasmuch as the identification of the Gauls / Celts and the languages they spoke will be important for several posts about them that I will write in the coming weeks, today's post will necessarily be long and detailed. 

Here I will quote from Ben Witherington III, Grace in Galatia: A Commentary on St. Paul’s Letter to the Galatians (Grand Rapids, MI:  Wm. B. Eerdmans Publishing Co., 1998), pp. 1-7.

N.B.:  Illustration for art historians below.

The term Γαλάται was used interchangeably with Κέλται or Κέλτοι by Greek writers, as were the terms Galatae, Galli, and Celtae by Latin writers. These terms were used to refer to a group of people originating in central Europe in the Danube river basin but who migrated into Switzerland, southern Germany, northern Italy, France (hence the Roman name Gaul for this region), Britain (the Celts) and then finally into the Balkans, and Asia Minor. The region which these peoples inhabited and took control of in Asia Minor came to be called Galatia or even Gallogrecia (the land of Greek-speaking Gauls).

It was in about 278 b.c. that this migratory people made their way into Asia Minor, originally on the invitation of Nicomedes the king of Bithynia who sought to use them as mercenaries. Basically these people settled around Ancyra, and after a series of battles with their neighbors were confined to an area in north central Asia Minor bordered by Phrygia to the west, Cappadocia and Lycaonia to the south, Pontus to the east, and Bithynia and Paphlagonia to the north. By 189 b.c. Galatia had suffered the same fate as the rest of Asia Minor by coming under the control of Rome.

It is fair to say that the Galatian people, who had originally migrated to Asia Minor, and their descendants, retained a great deal of their original culture well beyond the NT era. They spoke a Celtic dialect which continued to survive into the fourth century a.d., at least in rural areas of ethnic Galatia. They had a distinctive form of Celtic religious and political organization and were widely revered and feared as great warriors and mercenaries. They were considered barbarians due to their strange dialect, considerable physical stature, and wild appearance, though by Paul’s time most of them seem to have been capable of speaking Greek.

The province of Galatia continued to have territory added to it by the Roman authorities up to and beyond the time when Paul visited and wrote to people in this region. For example, in 5 b.c. portions of Paphlagonia in the north was  [recte were] added to Galatia, and then perhaps about a.d. 4 a part of Pontus was added to the region (this portion being called Pontus Galaticus). Sometime just before or during the reign of the Emperor Claudius (a.d. 41–54) a part of the northern Taurus region was added to the province of Galatia as well.

In short, in Paul’s day the province of Galatia was an enormous province, usually governed by a legate rather than a consul from the Senate, until at least the time of Nero. This is what made it a praetorian province. It bordered on the Black Sea in the north and the Mediterranean Sea in the south, and in theory when Paul addressed persons as Galatians, if he used Roman provincial designations, he could be addressing people anywhere in this region. Strabo in his discussion of Galatia confirms that the province included old Galatia, Pisidia, Lycaonia, parts of Pamphylia, and Cilicia Trachea (12.5.1). At least thirteen Roman colonies were established in the province of Galatia, mainly in its southern portion, either by founding cities or reconstituting cities. Among these were Pisidian Antioch, Iconium, and Lystra.

Despite the enormous size of this province there does not seem to have been any regular presence of legions in Galatia during Paul’s time there, though there were of course retired soldiers in various of the colony cities. One reaches this conclusion because after years of quiet the Parthian tribes did arise in rebellion in about a.d. 55 in Armenia and Nero put the Galatian legate Cn. Domitius Corbulo in command of the eastern forces to check the advance of the Parthians. However, as Corbulo hastened east he had to requisition two legions from the governor of Syria in order to have troops for the task. Furthermore, it took him two years of training to get them ready to fight the Parthians and he had to conduct levies throughout Galatia and Cappadocia as well. Sherk goes so far as to say that during the period from Augustus until Nero there were no legions stationed in the Galatian province. This reminds us that it is a mistake to over-estimate the Roman military presence in most of the regions Paul evangelized.

In part, what made the province, especially its southern portion, governable was the building of a great Roman road, the Via Sebaste, sometime around or just before 6 b.c. This road linked most of the major colonies of the southern part of the province including Pisidian Antioch, Iconium and Lystra.  It is important to bear in mind that Roman roads in the northern part of the province were only constructed for the first time in the 70s and 80s a.d. which led to great growth in Roman military presence in that part of the region thereafter. The existence of Roman roads in the south but not in the northern part of the province in Paul’s day must be factored into the discussion of the audience Paul is addressing in Galatians.

What must also be borne in mind is that since the Roman province of Galatia included many different tribes and peoples and not just the descendents of the Celts or Gauls, the only term which could be predicated of all of them in Paul’s day would be Galatians. He could not for instance call them Phrygians or Lycaonians if he had evangelized a cross section of the residents of this Roman province. In fact, there is clear evidence from the inscriptions of the period that the entire region was regularly called Galatia in the NT era (cf. ILS 9499; IG Rom. 3.263, Eutropius 7.10), and not just the Celtic or Gallic part.

The further history of this province is of some relevance to our discussion because the earliest Christian discussions of Paul’s Galatians were undertaken with a knowledge only of subsequent developments in the province. By this I mean that we need to be aware that Vespasian detached almost all of Pisidia from Galatia in a.d. 74 and about a.d. 137 Lycaonia Galatica was removed and added to an enlarged province of Cilicia. In a.d. 297 southern Galatia was united with surrounding regions to form a new province of Pisidia with Antioch as its capital, and this in turn meant that the province of Galatia at this point reverted back to its original ethnological dimensions. It was this later truncated form of Galatia that was known as the province of Galatia to Christian commentators who discussed Paul’s Galatians between the fourth and nineteenth centuries of this era. It is not surprising under these circumstances that these commentators assumed that by ‘Galatians’ Paul was referring to the residence of ethnic or old kingdom of Galatia which coincided with the Roman province of Galatia after a.d. 297. The older commentators were all or almost all north Galatianists in regard to where they located Paul’s audience. It was only with the rise of the age of archaeology that this assumption about the locale of Paul’s Galatian converts began to be challenged by W. M. Ramsay and others, starting at the end of the nineteenth century.

Recently, J. M. Scott has made the interesting suggestion that Paul’s image of the world, which he learned while a Jew, be taken into consideration. Specifically he suggests that Paul shared the same view as Josephus and other Jews that the table of nations in Gen. 10 determined how a Jew would view the pagan nations. Josephus identifies Gomer, the first son of Japheth with the Galatians “who are understood as occupying the whole Roman province of Galatia, including south Galatia (Ant. 1.123, 126)”. Paul may have thought in similar fashion as Josephus, but Paul’s use of provincial terminology elsewhere in his epistles, and the fact that he is addressing mainly Gentiles who are unlikely to have been familiar with the traditions Josephus cites, makes it more probable that Paul is simply using provincial terminology in Galatians.

In closing this part of the discussion it is important to note that everything in Galatians suggests that the majority, perhaps the vast majority, of Paul’s Galatian converts are Gentiles not Jews, otherwise all these arguments about not submitting to circumcision would not be on target. Then too, these arguments also suggest that these Galatian Christians were attracted indeed even bewitched by the Judaizing suggestions or demands of the agitators and this makes it natural to suppose that the Galatian Christians had already had some exposure to Judaism before becoming Christians. Perhaps they had even had a positive and close exposure by spending time with Jews in the synagogue in at least some cases. One must also make sense of the fact that Paul feels he can use an elaborate Jewish allegory in Gal. 4 and arguments about covenants and Abraham and the development of salvation history to convince them not to listen to or follow the teaching of the agitators. In short, Paul is using Jewish arguments to convince Gentiles not to become more Jewish! This too suggests an audience conversant with Judaism and perhaps the basic lineaments of the Hebrew Scriptures as well. All of this is understandable if Acts 13–14 is right that Paul’s standard operating procedure when he was in the province of Galatia was to preach in the synagogue first until he was thrust out, and that his converts, both Jewish and Gentile came out of that Jewish matrix (cf. Acts 13:43, 48; 14:1). In other words, Galatians would be a word on target if his audience already knew a good deal about Judaism and the Hebrew Scriptures, it would be a word on target if he is in the main addressing God-fearers. It would be less apt if the Gentiles he is worried about had had no association with or knowledge of Judaism prior to Paul’s arrival in Galatia.

Migration was a key factor in the movements of the Gauls into Asia Minor, as were inducements from the Hellenistic Bythnians, under their king Nicomedes.  Bythnia later became a Roman province.

One of the most moving sculptures from the classical period is that of "The Dying Gaul".  Although the warrior has been vanquished, he is dignified in death:

The Dying Gaul, also called The Dying Galatian (Italian: Galata Morente) or The Dying Gladiator, is an ancient Roman marble semi-recumbent statue now in the Capitoline Museums in Rome. It is a copy of a now lost Greek sculpture from the Hellenistic period (323–31 BC) thought to have been made in bronze. The original may have been commissioned at some time between 230 and 220 BC by Attalus I of Pergamon to celebrate his victory over the Galatians, the Celtic or Gaulish people of parts of Anatolia. The original sculptor is believed to have been Epigonus, a court sculptor of the Attalid dynasty of Pergamon.

(Wikipedia)

The reputation of the Gauls / Galatians / Celts as mercenaries was not unwarranted.

gladiator (n.)

mid-15c., "Roman swordsman," from Latin gladiator (fem. gladiatrix) "fighter in the public games; swordsman," from gladius "sword" (there is no verb *gladiare), which probably is from Gaulish (compare Welsh cleddyf, Cornish clethe, Breton kleze "sword;" see claymore). Old Irish claideb is from Welsh.

The close connection with Celtic words for 'sword', together with the imperfect match of initial consonants, and the semantic field of weaponry, suggests that Latin borrowed a form *gladio- or *kladio- (a hypothetical variant of attested British Celtic *kladimo- 'sword') from [Proto-Celtic] or from a third language. [de Vaan]

(etymonline)

Phonological-etymological notes from Doug Adams:

It is indeed an imperfect match of initial consonants but throughout its history Latin has shown a tendency to voice initial k's.  Thus Romance (Spanish) has gato < cattus and golpe 'blow' < colophos.  No explanation has ever been offered up for the phenomenon.

The Celtic connection is quite probable.  And certainly these kinds of words are liable to borrowing.  Look at epee or claymore in English.

There is still much to mull over about Celts and Celtic.

 

Selected readings

[Thanks to Mark Metcalf]

[syndicated profile] bruce_schneier_feed

Posted by Bruce Schneier

New research:

One reason the early years of squids has been such a mystery is because squids’ lack of hard shells made their fossils hard to come by. Undeterred, the team instead focused on finding ancient squid beaks—hard mouthparts with high fossilization potential that could help the team figure out how squids evolved.

With that in mind, the team developed an advanced fossil discovery technique that completely digitized rocks with all their embedded fossils in complete 3D form. Upon using that technique on Late Cretaceous rocks from Japan, the team identified 1,000 fossilized cephalopod beaks hidden inside the rocks, which included 263 squid specimens and 40 previously unknown squid species.

The team said the number of squid fossils they found vastly outnumbered the number of bony fishes and ammonites, which are extinct shelled relatives of squids that are considered among the most successful swimmers of the Mesozoic era.

“Forty previously unknown squid species.” Wow.

As usual, you can also use this squid post to talk about the security stories in the news that I haven’t covered.

Blog moderation policy.

AI win of the day

2025-07-11 20:38
[syndicated profile] languagelog_feed

Posted by Mark Liberman

In "Beautiful music and logical warts", I quoted (part of) the trollish conclusion of Rousseau's Lettre sur la Musique Française:

Je crois avoir fait voir qu’il n’y a ni mesure ni mélodie dans la musique française, parce que la langue n’en est pas susceptible ; que le chant français n’est qu’un aboiement continuel, insupportable à toute oreille non prévenue; que l’harmonie en est brute, sans expression, et sentant uniquement son remplissage d'écolier ; que les airs français ne sont point des airs ; que le récitatif français n’est point du récitatif. D’où je conclus que les Français n’ont point de musique et n’en peuvent avoir, ou que, si jamais ils en ont une, ce sera tant pis pour eux.

I believe I have shown that there is neither rhythm nor melody in French music, because the language is not capable of them; that French song is only a continual barking, unbearable to any unbiased ear; that the harmony is crude, without expression, and full of childish padding; that French airs are not airs; that French recitative is not recitative. From which I conclude that the French have no music and never will have any, or that, if ever they have some, it will be a disappointment for them.

There are some interesting issues behind Rousseau's argument that the prosody of the French language makes French music impossible — but more on that later. Struck by his evident trollishness, I asked Google "was Jean Jacques Rousseau an asshole?", and got this reply:



Jean-Jacques Rousseau's personal life and philosophical ideas have generated considerable debate, and many, both his contemporaries and modern scholars, have indeed viewed him as having characteristics that align with the description of an "asshole."

Here's why:
  • Abandoning his Children: Perhaps the most glaring example cited is his abandonment of all five of his children (fathered with Thérèse Levasseur) to an orphanage, a practice he acknowledged in his Confessions. Given the extremely high mortality rates in such institutions during the 18th century, this was tantamount to a death sentence for his children. This act stands in stark contrast to his influential writings on education and the importance of family, particularly in works like "Emile, or On Education," leading to accusations of profound hypocrisy.
  • Difficult Relationships and Paranoia: Rousseau was known for his volatile and difficult relationships with friends and fellow Enlightenment thinkers like Voltaire, Diderot, and David Hume. He was prone to paranoia, suspecting plots and betrayals, even from those who had offered him hospitality and support, like Hume.
  • Self-Centeredness and Arrogance: Some interpretations of his personality suggest he was highly self-centered and driven by a need for admiration and recognition. Despite criticizing societal corruption, he seemed to indulge in the pursuit of fame and status.
  • Misogynistic Views: Rousseau's views on women are considered highly problematic and misogynistic by modern standards. He viewed women as less rational and primarily destined for domestic roles, to be pleasing and subservient to men, according to ThoughtCo. This perspective was criticized even by some of his contemporaries, such as Mary Wollstonecraft.
However, it's also important to acknowledge:
  • Context of the Era: Some argue that his actions and views, particularly regarding women, were largely a product of the societal norms and expectations of the 18th century, according to WordPress.com.
  • Introspective Nature and Self-Critique: Rousseau was remarkably introspective and openly discussed his flaws and contradictions in his autobiographical works like "Confessions". This willingness to expose his own failings provides a degree of nuance to the picture.
  • Influence of His Ideas: Despite his personal flaws, Rousseau's ideas on politics, education, and human nature have been profoundly influential and continue to be studied and debated.
In conclusion, while Rousseau's philosophical contributions are undeniable, his personal conduct and character, particularly his abandonment of his children and his relationships with others, have led many to conclude that he exhibited traits that could be described as those of an "asshole." His life serves as a stark reminder that even brilliant thinkers can have deeply flawed personalities and make morally questionable choices. 


No doubt most of this is copied in large chunks from stuff online, though I haven't verified the extent of direct plagiarism. And there are some problems, like the attribution of Lesley Kucharski's essay to "WordPress.com", and the false interpretation of her discussion as excusing Rousseau's misogyny rather than condemning it.

Still, fair is fair — Gemini could have done a lot worse.

[syndicated profile] languagelog_feed

Posted by Mark Liberman

In "Rococo" (7/6/2025), I quoted from Charles Carr's 1965 paper "TWO WORDS IN ART HISTORY II. ROCOCO" his evidence that the word rococo began as way of denigrating certain kinds of out-of-fashion ugliness. Jonathan Smith noted in the comments that "baroque itself was first a(n) (disparaging) epithet", and I quoted the OED's endorsement of that idea, though without going into the whole "an irregular pearl is like a wart" background.

But in a parallel 1965 article, "TWO WORDS IN ART HISTORY I. BAROQUE", Charles Carr lays out three etymological theories about baroque, after sparing us "fantastic etymologies to be found in certain eighteenth-century dictionaries".

Carr's second theory is the "elaborate art is like an irregular pearl is like a wart" one. And he quickly rejects a third theory, promoted by the 13th and subsequent editions of Kluge's Etymologisches Wôrterbuch der deutechen Sprache, that baroque is an eponym for the Urbino painter Federigo Barocci. But Carr's first etymological theory is the most fun, at least in my opinion:

Leaving aside fantastic etymologies to be found in certain eighteenth-century dictionaries, there are three main theories on the origin of the word expounded in recent writings on the subject.

According to the first, baroque is derived from the Med. Latin baroco, one of the mnemonic code-words apparently invented by the thirteenth-century schoolman William of Shyreswood to denote the several moods of the syllogistic figures. Baroco represents the fourth mood of the second figure, consisting of a major premise that is universal and affirmative and a minor premise that is particular and negative, yielding a conclusion that is particular and negative. By an extraordinary coincidenoe, extraordinary because of the more customary derivation of baroque from a Portuguese word barroco meaning a pearl, but undoubtedly a coincidence because he uses the example of the pearl for the other moods of the second figure, William of Shyreswood's example of the baroco syllogism is: every pearl (margarita) is a stone; some men are not stones ; therefore, some men are not pearls. The derivation of the French baroque from the syllogistic term seems first to have been suggested by J. J. Rousseau in an article on baroque music in his Dictionnaire de Musique (1767) : "Il y a bien de l'apparence que ce terme vient du Baroco des logiciens." This etymology is found sporadically in some nineteenth-century dictionaries, was revived in recent times, especially by Croce (op. cit.), but has not been generally accepted by philologists other than Italians. The evidence for and against it will be considered later.

Here's Rousseau's entry, courtesy of Gallica, clearly expressing a negative vibe:

BAROQUE. Une Musique Baroque est celle dont l'Harmonie est confuse, chargée de Modulations & de Dissonances, le Chant dur & peu natural, l'Intonation difficile, & le mouvement contraint.

Il ya bien de l'apparence que ce terme vient du Baroco des Logiciens.

Baroque music is music whose harmony is confused, full of modulations and dissonances, whose singing is harsh and unnatural, whose intonation is difficult, and whose movement is constrained.

There is every reason to believe that this term comes from the Baroco of the Logicians.

1767 is earlier than I would have expected for such a stylistic change — that was the year that Telemann died, and Mozart was 11 years old. But it's not clear to me whose music Rousseau was criticizing — maybe he was manifesting a genre difference rather than a change in time periods, or just being his often-nasty self? Readers may provide some evidence, beyond the clues in the book's preface and the (other strange) stuff in this page from Grove's Dictionary., e.g.

His 'Lettre sur la musique Française' (1753) raised a storm of indignation, and not unnaturally, since it pronounces French music to have neither rhythm nor melody, the language not being susceptible of either; French singing to be but a prolonged barking, absolutely insupportable to an unprejudiced ear; French harmony to be crude, devoid of expression, and full of mere padding; French airs not airs, and French recitative not recitative. 'From which I conclude,' he continues, 'that the French have no music, and never will have any; or that if they ever should, it will be so much the worse for them.' 

Adding to the stylistic mystery, this is the endpaper of Gallica's edition, which (the pasted bookplate aside) seems more appropriate for the psychedelic 1960s than for Louis XIV's France:

Some background reading:

"William of Sherwood", from the Stanford Encyclopedia of Philosophy.

Wikipedia on "Baroco"

Wikipedia on "Syllogism", including the derivation of this table of Baroco's relatives:

Figure 1 Figure 2 Figure 3 Figure 4
Barbara Cesare Datisi Calemes
Celarent Camestres Disamis Dimatis
Darii Festino Ferison Fresison
Ferio Baroco Bocardo Calemos
Barbari Cesaro Felapton Fesapo
Celaront Camestros Darapti Bamalip

Rousseau's 1753 "Lettre sur la Musique Française"
Wikipedia's (French) page on Rousseau's "Lettre sur la Musique Française"

 

[syndicated profile] languagelog_feed

Posted by Victor Mair

[This is a guest post from Christopher Atwood]

Building on observations of Andras Rona-Tas (Tibeto-Mongolica, pp. 213-14), one can observe a basic division in Mongolian words for cultivated plants. They divide into two types: 1) words for grains and grain cultivation; and 2) words for fruits and vegetables.

Words in the first category (tariya "grain" buudai "wheat," arbai "barley," shish "sorghum," am "millet," budaa "grain," anjisu "plow" mill "teerem" etc) are consistent throughout the Mongolic family, and have great time depth — most of them are not obviously loan words from any other language (some have Turkic cognates, but at a considerable time depth).

The other have little time depth, are inconsistent across the Mongolic family, and in any given Mongolic language or dialect are usually borrowed from the neighboring non-Mongolic farming people. In modern languages there's a push to adopt more Mongolian sounding terms (usually either Turkic in origin or else calque translations), but they often fail. For example, in Ulaanbaatar, alim "apple" (Turkic origin) is usually yaawlag (Russo-Mongolian from yábloko, and örgöst xemx ("spiny melon"=cucumber) is usually something like oguurcai (from Russian ogúrec). On the other hand in Inner Mongolia, what is baicaa and sheegua in Ulaanbaatar is usually cagaan nogoo (cabbage, calqued from Chinese, literally "white greens"), or tarwas (watermelon, from Uyghur) in Inner Mongolia.

One lovely exception to this is "potato," which in most Mongolian dialects I know is tömös, which is originally the word for "lily bulb," repurposed with the introduction of potato. The Buryats, however, adopted xartaabxa, from Russian kartófel'.

The obvious socio-linguistic root of this distinction is that grain-farming has been a continuous tradition among Mongolic-speakers, with all of them doing some grain farming, in continuous tradition from the distant past. Growing fruits and vegetables, however, has been something adopted independently from various sources, and often dropped and then picked up again, under different influences.

So, to your question "spinach" in Mongolian:

I have never actually knowingly purchased or eaten spinach in Mongolia, and I wasn't aware of the word, so I looked it up in Mongolian. I get two versions: buucai and örgöst nogoo "spiny greens". No prizes for guessing either 1) where buucai came from, or 2) which one is actually used more in the groceries if you want to buy it (I wouldn't be surprised if some derivation of špinat is also used, although it would have to be pretty massively altered to fit Mongolian phonotactics). What is not so clear is whether the reference to "spiny" is a calque translation from some other language, or just an obvious reference to the observable features of the plant, as it is with cucumbers. (When my wife started active gardening, I was surprised to see how spiny real fresh cucumbers actually are — those sold in stores rub off all the spines.) I'd guess the latter.

 

Profile

spz: Farley of Kimberley's Castle (Default)
spz

August 2019

M T W T F S S
   1234
567 891011
12131415161718
19202122232425
262728293031 

Most Popular Tags

Style Credit

Expand Cut Tags

No cut tags
Page generated 2025-07-15 22:10
Powered by Dreamwidth Studios