languagehat.com

Karin Kaneko’s Japan Times article ChatGPT, Bing, Bard and DeepL: Which one offers the best Japanese-to-English translation? provides some detailed comparisons:

As a reporter in Japan working in both English and Japanese, I was keen to test these tools and see which might be best for my line of work, and whether they offer benefits for Japan Times readers. […] Specifically, we wanted to see which one of the AI-assisted tools best understands context in Japanese, a language in which sentences sometimes lack a subject, and accurately translates it into natural English.

Kanako Takahara, a senior bilingual Japan Times editor, compared Japanese-to-English translations from ChatGPT-4, Bing and Bard, along with DeepL, an AI-based machine translation service, using Japanese text in three different categories:

• Literature: The opening lines of “Snow Country,” authored by Nobel laureate Yasunari Kawabata
• Lyrics: Japan’s national anthem “Kimigayo”
• Speech: Japanese superstar Shohei Ohtani’s speech to his teammates just before the World Baseball Classic final against the United States in March

She scored on a scale of 1 to 5, with 5 being the highest based on accuracy, how natural the English was and whether the translation reflected the context. […]

[Read more…]

I liked very much these passages from Hugh Trevor-Roper’s “Apologia transfugae” quoted at Laudator Temporis Acti:

From our present position there is unquestiably something very arrogant in the old claims of classical humanism to be the necessary centre of our studies. The supposition that the seeds at least of all knowledge, all wisdom, all philosophy, are contained in the brief experience of a chosen people is no more defensible in secular studies than in theology. Why should two ancient cities, minuscule by the standards of our provincial towns, be the repositories of truth, the sources of civilization, any more than a fanatical semitic hill tribe in Palestine? Such a concept, however it may be sophisticated, is repugnant to those who believe in progress; and it had to be defeated before the idea of progress could be denizened in European thought.

[…]

I suppose I really ought to modify my irrational antipathy to Wilamowitz. He did. after all, affect my life. It was because of him that I learned German. Brought up, as I was, in the extraordinary and indeed, it now seems to me the ludicrous belief that one could not be a good classical scholar unless one read the works of this Prussian high-priest of the subject, I obediently set myself, as a first year undergraduate, to master his rebarbative language. After spending the best part of vacations in Germany learning it, I returned to Oxford somewhat disillusioned. With undergraduate confidence, I decided that Wilamowitz, whose works I had now read (or perhaps only tasted) was a fraud and that Nazi Germany was not only very disagreeable itself but also a menace to the world. I therefore used up the linguistic expertise which I had so mistakenly acquired to read Hitler’s Mein Kampf, which was not then available in translation. I found it very rewarding, in a certain sense, and the experience has had some influence on my later career. Since war I have occasionally dipped into Mein Kampf again. I have not found any occasion to reread Wilamowitz. My view of him, however erroneous, was fixed in 1934, and I saw him now only as a salutary warning. He symbolised to me the barrenness of a purely literary and philological approach to the classics, and indeed to literature in general, and the absurd pretentiousness of assuming that so narrow an approach can have any wider meaning. He warned me to turn away from line of study which, within those limits, led nowhere. […]

[Read more…]

Hannah Jane Parkinson reports for the Guardian on something that has become vital to modern communication:

There have been plenty of, shall we say, unusual or eye-raising legal decisions around technology. […] But the thing that caught my eye last week was a Canadian court decision that ruled a thumbs-up emoji is legally permissible as contract assent. There are more examples of emojis finding their way before the bench. In 2014, a Michigan court tried a defamation case involving a stuck-out tongue emoticon (rendered as :-p). In Ohio, a judgment in a harassment case queried what, exactly, the rat emoji meant in that context.

This is because emojis – as many unfortunates have discovered (often gen X parents, but that I, a millennial in her early 30s, am increasingly, devastatingly, discovering) – do not always have clearcut meanings. This is true of all language of course – and emojis are a type of language, despite what the likes of John Humphrys et al have sneered in the past. A thumbs-up emoji, to take the example from the Canadian case, can, just as in offline life, be used sarcastically. (This was noted in the court ruling.) In some regions such as in the Middle East a thumbs-up can be offensive. […]

Particularly interesting is the way in which emojis are approved, which is by an industry body called the Unicode Consortium; I like to imagine its members as a bunch of normal-looking suits sitting around a table – except they all have giant yellow heads and love hearts for eyes. The emoji-accepting process is not simple – it’s basically the emoji equivalent of winning a place at Oxbridge. This is why the subjects of the “two dancing girls” emoji look so pleased. It can take two years for submitted emoji proposals to be accepted and completed, and there is often a public clamour for new additions. […]

The language we use in tech, then, is just as important as that coming from our mouths, or our hands, or any other form of IRL communication. And, as with those forms, a lot can be said about a person’s age or social demographic in their use of emojis. Gen Z thinks the crying-with-laughter emoji terribly uncool, and more often uses the skull emoji (as in, “I’m dead”, when finding something amusing or out there). I’m not sure if we’ve moved on from using the painting nails emoji as a sort of jovial smugness to signify one’s own achievement or moment of aptitude, but I hope not. I think, by now, we all know what the aubergine emoji means – and it isn’t an aubergine. […]

Never got around to reading Moby-Dick? Well, it’s since been translated into emoji. As have the lyrics to a number of pop songs. To not know the latest emoji use or etiquette has almost become a modern form of illiteracy.

I remember when I first became aware of emoji, I sneered; now I use them frequently and shamelessly in my messages. So thanks, Trevor, for the link, and happy Emoji Day to all!

We’ve discussed Jhumpa Lahiri’s switch to writing in Italian before, and I’m freshly impressed every time I read something by her (I really should acquire one of her books instead of depending on the New Yorker to feed stories to me); I’m shoehorning her latest, “P’s Parties” (translated by Todd Portnowitz; archived), into LH because language is something of a plot point:

They came from different countries, for work or for love, for a change of scenery, or for some other mysterious reason. They were a nomadic population that piqued my interest—prototypes, perhaps, for one of my future stories, the kind of people I’d have the chance to meet and casually observe only at P’s house. In no time at all they’d manage to visit nearly all parts of our country, tackling the smaller towns on the weekends, skiing our mountains in February, and swimming in our crystalline seas in July. They’d pick up a decent smattering of our language, adapt to the food, forgive the daily chaos.

[…]

My memories of the past five or so parties had blurred together. Each year was different, and each year, for the most part, was the same. I made the same small talk I’d forget a minute later, I practiced my two rusty but still passable foreign languages, which I’d always brush up on a bit.

[…]

The woman spoke in a strange mix of her language and ours, but it was easy enough to follow.

[…]

Because of the girlfriend, we never spoke to each other in Italian. He gushed about the multiethnic neighborhood where they lived, where they’d go out every night of the week to eat food from seven different countries. His answers to my questions were polite but brief. We conversed in a language I struggled to keep up with, a sensation that I enjoyed at P’s house but that here, with my own son, felt frustrating and artificial.

But really, I’m just hoping to entice people to read it; it reminds me of Virginia Woolf, and I don’t have much higher praise than that.

I have to apologize up front, because those of you who aren’t subscribers won’t be able to read Gabriel Gianordoli and Robert Ito’s NY Times article in the interactive way it’s meant to be read, seeing the original Japanese pages morph into the translated ones and selected areas highlighted; the archived version has only the text. But that text is interesting enough I figure it’s worth passing along:

Open a Japanese comic in a bookstore — say, the latest edition of “Dragon Ball Super” — and you’re likely to find a note saying, “Stop! You’re reading the wrong way.” That’s because manga begins where all Japanese books do: on what, to Western readers, would be the last page. The text then moves right to left. Try reading manga in the Western way and you’re likely to spoil a great ending.

Since manga was first introduced to the U.S. in the 1980s, American companies have wrestled with how to adapt the genre for their readers. It requires taking into account not only art and visual concepts that are unique to Japanese, but also an entirely different system of reading.

Today manga is enormously popular in the U.S. and is published in something close to its original form: in black and white, on inexpensive paper stock, to be read in the Japanese style. But this wasn’t always the case.

[Read more…]

I am experiencing a lexical/botanical crisis, and I hope the Hattery can help. I’ve long been vaguely familiar with the Russian word шиповник, defined by my trusty Oxford dictionary as ‘wild rose’; I knew it mainly as the name of a Silver Age publisher, and thought of it as “some kind of plant.” But having just run across it in a novel I’m reading, I thought I’d better refresh my memory, and when I googled it the page said prominently “Rose / Plant.” “It’s not just a rose, is it?” thought I, but when I went to the Russian Wikipedia page I discovered it was the equivalent of the English one for Rose, and said “Запрос «Rosa» перенаправляется сюда” [The query “Rosa” redirects here]. So it is just a rose? But checking the Russian corpus got me citations like “Изредка, там же, растет и шиповник ― дикая роза; чаще всего это роза собачья” [Occasionally, there is also a шиповник, the wild rose, growing there; most often it is the dog rose]. So it’s a special variety of rose? But then why is the Wikipedia article called Шиповник and not Роза? And what is a “wild rose” anyway? That is not an English lexical item (at least it’s not in any of my dictionaries, even the OED), it’s just a collocation — a rose that happens to be wild. Can anyone who knows more about botany than I disentangle this for me? And can my Russian-speaking readers tell me how they think of шиповник and роза? Are they pretty much the same thing, used in different contexts, or distinct plants?

Exciting news from Phys.org:

The Kushan Empire in Central Asia was one of the most influential states of the ancient world. A research team at the University of Cologne’s Department of Linguistics has now deciphered a writing system that sheds new light on its history.

A team of early career researchers at the University of Cologne has succeeded in decoding a script that has been puzzling scholars for more than 70 years: the so-called “unknown Kushan script.” Over a period of several years, Svenja Bonmann, Jakob Halfmann and Natalie Korobzow examined photographs of inscriptions found in caves as well as characters on bowls and clay pots from various Central Asian countries in order to put the pieces of the puzzle together.

On 1 March 2023, they first announced their partial decipherment of the unknown Kushan script at an online conference of the Academy of Sciences of the Republic of Tajikistan. Currently, about 60% of the characters can be read, and the group is working to decipher the remaining characters. A detailed description of the decipherment has now been published in the journal Transactions of the Philological Society under the title “A Partial Decipherment of the Unknown Kushan Script.” […]

[Read more…]

I’ve been following the adventures of Nick Nicholas on Facebook as he traveled through Greece and Cyprus, and I’m pleased to report he’s posted some linguistic observations at Ἡλληνιστεύκοντος (to quote myself from a few years ago: have I mentioned how happy I am that Ἡλληνιστεύκοντος is back?):

I’m not going to make this a course on Cypriot, and I’m not going to explain the technical terminology thoroughly; at least not yet. I’m tired, and I just want to capture the things that struck me about Cypriot after a week of hearing it all around me, even if in a potentially attenuated-for-“Penpushers” form.

(“Penpushers”, καλαμαράες, being speakers of Standard Greek.)

Phonology: my family were certainly doing most of the familiar Cypriot processes: double consonants, dissimilation of fricative + fricative (e.g. vɣ > fk, fx > fk), palatoalveolar allophony. They did not seem to be doing a whole lot of fricative + yod dissimilation (e.g. ðj > θc). x > θ I caught only a couple of times. As I’d been warned by Tsimplakou, lots of dropping of intervocalic ɣ, intermittent for ð (though ɣajðurin > ɣaurin “donkey” and koruðes > korues “girls” was regular), none for v.

Something I didn’t know about beforehand: ʎ > j. So I heard /palja/ [paʎa] rendered as [paja] several times.

[Read more…]

Nothing spectacular, but a nice story from Mike Cummings at Yale News:

Turning the pages of a manuscript copy of the Maʿrifetnāme, an 18th-century encyclopedia authored by the Ottoman scholar and Sufi poet İbrāhīm Ḥaḳḳī Efendi, can lead readers to seventh heaven and the depths of hell. A copy of the beautifully illuminated manuscript — one of just a handful from the 18th century known to exist — is housed at the Beinecke Rare Book and Manuscript Library, part of the Yale Library. It features a detailed illustration of the Islamic conception of Judgment Day, including the seven layers of paradise and perdition, the scales used to weigh people’s deeds, and the book in which their sins and virtues are recorded. A cauldron of tar awaits those cast into hell. A lote tree marks the upper boundary of heaven.

Brightly colored world maps, diagrams, and charts join the rendering of the afterlife within the manuscript’s pages. The Ottoman-Turkish text explores a vast range of subjects, including astronomy, biology, physics, faith, mathematics, and mysticism. The manuscript is an intellectual feast for scholars and bibliophiles. But until now, few knew how to find it.

The Maʿrifetnāme is one of 567 Ottoman-Turkish objects — spanning the mid-15th century to the early 20th century, including hundreds of manuscripts, a scroll calendar, imperial orders, and the passport of an Ottoman pasha — being cataloged so that researchers can easily find and access them in the library’s collections.

[Read more…]

I won’t get around to reading the bulk of Yuri Leving’s Keys to the “Gift” (see this post) for a while yet, but I couldn’t resist gobbling up the introductory material, and I had to pass along this passage from the section “How to Use This Book” (freely available at JSTOR). After explaining that he had initially resisted The Gift before devouring it with increasing pleasure on his honeymoon (on a kibbutz), he writes:

As often happens, I hesitated for a long time to analyze my feelings rationally and examine the source of my delight under any sort of intellectual magnifying glass. Then, in 1996, Professor Roman Timenchik (my beloved teacher at the Hebrew University) offered for the very first time his graduate seminar entitled “The Russian Nabokov.”

That first semester we only read about twenty-five pages of the opening chapter (the entire novel is over three hundred pages). Usually we looked at several sentences per class, but in the case of some particularly complex constructions, we might spend up to two sessions on a single phrase. Practicing the method of close reading (and our readings were very close indeed!) we brainstormed about the text. We began by discussing a simple understanding of the pragmatic message of each sentence, then moved toward dissecting the syntax, before finally attempting to crack the metatextual codes and track down the implicit literary allusions. I audited the same course the following year and our progress turned out to be even more modest: we managed to get through only the first fifteen pages. By the time I left Israel, I had attended Timenchik’s seminar three times (twice from start to finish and then less regularly in the third year due to other commitments), and our intense discussions almost never duplicated the debates of the previous years, proving to be just as interesting, stimulating, and refreshing.

During the seminars, some of us questioned whether Nabokov could have possibly kept consciously in his mind such a multiplicity of allusions and reminiscences, fusing them in packed images that so deftly entrapped his readers and laying semantically explosive mines in the dense field of his prose. Could our overzealous interpretations lead us to unintentionally presumptuous fallacies? One of the puzzled students, unable to restrain himself, once exclaimed: “But even if half of what we discover here is true, then Nabokov’s mind had to be a kind of computer!”

Timenchik instantly retorted: “Then a computer he was.”

Of course, there’s no way of knowing how much Nabokov (or any other author) kept consciously in mind while writing, but I sure wish I’d been able to take that seminar.