Category: Uncategorized

Yes, and there’s a linguistic pragmatics set of principles at work there, over and above the inherent limitations of language pointed out by Daniel Bamberger : see Daniel Bamberger’s answer to Is there such a thing as “taking things too literally”?

The Cooperative principles defined by Grice are a way of making sense of how people don’t take things literally. The underlying understanding, when you’re talking with someone, is that your interlocutor is not being an arsehole, and is not talking to you just to troll you. You assume that what they are telling you makes sense and is relevant. So if their literal meaning comes across as trolling, you try to think up figurative and indirect meanings, which make what they’re saying make sense.

This kind of second guessing of literal meaning underpins humour, figurative language, metaphor, literature, wit, allusion—all the potent stuff in language. The fact that the meaning is indirect in such expressions, and has to be teased out by listeners assuming that you are not trolling them, is a big part of their potency.

And of course doing that teasing out of indirect meaning requires a large amount of emotional intelligence and social context—which notoriously puts autistic people at a disadvantage. But yes, there is a societal expectation that you will use Gricean principles to make sense of figurative language, and if you fail to do so, you are taking things too literally for that social norm.

Very broad question. I’m going to give you one recommendation:

Understanding Language Change by April M. S. McMahon. 1994.

Magnificent, and goes into a lot more of the mechanics of language change, informed by sociolinguistics, than the older treatments.

The linguistics textbooks will tell you that the unfortunate case of Genie (feral child) demonstrates that puberty delimits a window of opportunity for language acquisition, past which full language acquisition is not possible. Genie learnt enough English for the first time at 13 to communicate, but her English was never grammatical.

There are plenty of horrors surrounding Genie’s case, and quite apart from the monstrous lapses in ethics around how she was handled, any scientific conclusions gleaned from her are problematic.

Genie was cut off from any linguistic or social input, because her father was convinced she was brain damaged. If she wasn’t before, his abuse guaranteed that she was after. If she was handicapped before, which we have no way of knowing, her ability to acquire language may well have been compromised anyway.

On the other hand, she seems to have developed an adequate command of sign language (itself a political football between linguists and psychologists, while they were belly flopping on their professional ethics all around her). Sign language was the psychologists’ idea, not the linguists’, so we don’t really know how much better her sign language was than her spoken language. In any case, once she was released from the lab to endure further physical and sexual abuse, her command of sign language seems to have been impaired as well.

What have we actually learnt about first language acquisition from Genie’s case? Probably not as much as people like to think. And hopefully, we will never have the opportunity to repeat such an experiment again.

Tikhon Jelvis has just followed me, and I don’t want to annoy him by liking Go. 🙂

And I do like Go. But treating Golang as a general all-purpose language is silly and hype-y.

Golang is a low level, strictly typed language. It is almost as pleasant as a low level language can get: a lot of syntactic sugar has gone in to hide pointers and types from the developer, and to backdoor void * through the empty interface. (The interface is the one thing about Golang that I still bristle at.)

But the low level stuff is all still there. And it makes more sense to deal with low level stuff in applications where you need to, than in naturally high-level or prototype-y code. It makes more sense in context where it is useful to have a compiled binary.

I’m using Go in messaging infrastructure. I’m happy to, it makes sense to use a low level language there, and I’d rather cut Golang code than C code. (25 years of cutting C code does not make it any more congenial.) But I’d hesitate to use Golang in natural language processing.

(I find Python much more abhorrent than Golang, but in natural language processing, the library (NLTK) is king. If only Ruby had gotten there first.)

Performance, from what I gather, got better: my CTO was very enthused about the recent upgrade to v1.8 in the compiler, and you don’t care about compiler upgrades in more mature languages. The maturity nevertheless has come along, it seems, compared to earlier versions. The code is still low-level looking, and not at all as readable as Ruby or *shudder* Python; but because of the sugar, it is easier to write and to read than C.

The proper answer is Kutluk Ozguven’s: Kutluk Ozguven’s answer to Why do some Greek surnames end with “oğlu” which means “son of” in Turkish?

Turkish Republic did not enforce surnames to its population before 1934. Turks had patronymous names like in Arabic countries or Iceland.

However Greeks and Armenians used family surnames of their choice. Unlike post-nationalist myths Greek Orthodox and Muslim populations were closer and more dependant to each other […]

Population exchange between Greece and Turkey was from 1923–1933. […]

Since Oglou was a sign of Turkish migration who were scorned upon in their arrival, many might have changed to more mainland Greek surmames.

Some didn’t bother.

Indeed: –oglou is a patronymic suffix specific to the descendants of refugees in Greece from Asia Minor; I’m not aware of any serious traditional usage within Greece in the 19th century. As Kutluk pointed out, Christians took up Turkish surnames in Turkey before Muslims did.

Often, that surname suffix was dropped by the arrivals in Greece, in favour of something more Hellenic. And nothing is more Hellenic than the Ancient Greek patronymic –ides. (Because of how memes happen, –ides also supplanted the Greek Pontic patronymic –ant[is], as in Ypsilanti[s].)

So if you see a surname ending in –ides, chances are the bearer is descended from Asia Minor. (Or Cypriot, where –ides also came into vogue. And these really are matters of vogue: in Crete –akis is universal as a surname suffix, and it was unknown before the 19th century.)

The surname Σαλπιγκτίδης Salpingtides, for example, is quite Hellenic, and rather challenging to romanise (you’ll usually see it as Salpigktidis in English.) It’s Ancient Greek for “bugler-son”—and it’s a transparent Hellenisation of Borazancıoğlu.

Some refugees refused to switch their surnames. The father of my coauthor George Baloglou was a refugee from Sille, near Konya. He kept his surname, which is Turkish for “honey-son”. Most of his extended family switched it to the Hellenised Melidis.

Marc Okrand, who invented the language, was a consultant on all the TOS Star Trek movies. He made sure all the Klingon spoken was canonical, and if the actors flubbed their lines, he retconned them.

Okrand was not involved with the Klingon used on the TV series. As a result, the TV series featured words like wIjjup for “my friend”, where –wIj is supposed to be a possessive suffix. And that’s when the script writers bothered to use the Klingon Dictionary at all.

Be hesitant with TNG for any Klingon utterances you hear that are one word long. Don’t take seriously anything more than one word long.

SVO in Standard Greek. The linguist Erma Vasiliou has argued in her PhD that it’s VSO in Cypriot. http://arrow.latrobe.edu.au:8080…“

There are Greek Communities in all capital cities of Australia, but the largest communities by a wide margin are in Sydney and Melbourne, and Melbourne is renowned as the main Greek Community.

SBS, the national multicultural broadcaster, has been putting out Greek programming on tv and radio for decades. Radio station 3XY, a rock music mainstay in baby boomer times, has been a Greek only radio station for close to two decades now in Melbourne. There is no homegrown Greek TV channel, but satellite TV and cable TV do a brisk trade in rebroadcasting content from Greece. Antenna has the cable TV channel monopoly.

Because Quora may have greater confidence in what its bots can do than most users have, but even Quora does not yet claim it has automated utilitarian ethics.

No need for concern though. We have been assured that all Anonymous questions are being carefully vetted before publication now.

😐

I’m so sorry I didn’t notice this A2A.

Quora has a time limit on blog posts. I got blocked for ten blog posts in half an hour.

Other answers on Quora asserted that the block was temporary. After 4 or 5 days, my block was not lifted, and I asked for Tatiana’s intervention. Tatiana lifted my block, and said it was a bug. When I asked for some assurance that it would not happen again, she repeated it was a bug.

So. Bug.

Let us know how you go!