Subscribe to Blog via Email
Join 327 other subscribersJuly 2025 M T W T F S S 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31
How can a software engineer get into computational linguistics?
- You need programming chops, though nothing too flash and algorithmic.
- You need to be across regexes.
- You need to pick up some linguistics, but honestly, not as much as you might think. You certainly don’t need formal syntax or phonology. You will need to know what morphology is, especially if you’ll be working on languages other than English.
- You will inevitably end up getting into some stats and stochastic work. The NLP that works best is statistical, not rule-based.
- You can pick up a lot from Natural Language Toolkit. Python these days is the premier language for NLP, and the NLTK is the major reason why.
- Look for patterns; don’t be too prescriptive; know when close enough is good enough.
What is Ferdinand de Saussure’s linearity principle?
See e.g. http://personal.bgsu.edu/~dcalle… : Principle II: The Linear Nature of the Signifier
The linearity principle is Saussure’s statement that, because linguistic signifiers are sounds (spoken words), they are intrinsically sequential (“linear”). They cannot be perceived simultaneously, the way visual signs are: they must be perceived one after the other, as a sequence in time. That principle is also carried over to writen words, as a visual representation of spoken words.
What are the top 5 best Greek Songs of all time?
I’m going to give one for each decade from the 30s through the 70s. I’m going to put up, not necessarily my favourite songs, but the songs I think have had the greatest cultural impact.
1935. Φραγκοσυριανή (Frangosyriani): Catholic Girl from Syros. Lyrics: Markos Vamvakaris. Music: Markos Vamvakaris.
Markos was the master of the Peiraeus tradition of rebetiko, which switched from an Anatolian, plaintive setting for songs about hashish and swag, to a Greek, jaunty setting for songs about hashish and swag.
Frangosyriani marks the beginning of the end of the tradition. Wikipedia says it was written in ’35, but it already sounds like it was written in ’36, when the Metaxas dictatorship censored both the lyrics and—more lastingly—the scales of rebetiko. It’s just a list of scenic locations in Syros where Markos would take his fellow Catholic girlfriend. Its music lacks the bite of what made Markos great.
But this is his lasting legacy, a tune that gets everyone swaying, a tune suffused with elegance and romance—and jauntiness.
Here’s the best known, 1960 recording.
I have a swelling, a flame, inside my heart
as if you’ve cast a spell on me, sweet Catholic girl from Syros.
I’ll come meet you down by the beach.
I’d like to fill you with caresses and kisses.
I’ll take you on a trip to Finikas and Parakopi,
to Galissa and Della Grazia, even if I get a heart attack.
At Pateli, at Nichori, a fine time at Alithini,
and romancing at Piskopio, my sweet Catholic girl from Syros.
1948. Συννεφιασμένη Κυριακή (Synnefiasmeni Kyriaki): Cloudy Sunday. Lyrics: Alekos Gouveris or Vasilis Tsitsanis. Music: Vassilis Tsitsanis.
If Markos was master of the Peiraeus blues, Tsitsanis was master of what came next: the transformation of rebetiko into a genre palatable to the masses, cleaned up, and with virtuoso flourishes.
Cloudy Sunday is the song that, at least when I was young, every Greek knew by heart, and was guaranteed to bring a tear to every Greek’s eye. It has a dignified, stately melancholy to it, a proud swelling of the heart.
Tsitsanis let it be understood that the song was written in 1942, while Athens all around him was starving. That was no doubt a big part of why the song became so well loved. The prevalent theory is that the lyrics were written in 1946 by a friend of his, because his soccer team had lost that Sunday.
In truth, that doesn’t diminish the song a bit. The song stands just fine on its own. Here’s the definitive performance by the great Stelios Kazantzidis.
stixoi.info: Συννεφιασμένη Κυριακή
Cloudy Sunday, you’re like my heart,
that’s always overcast—Oh Christ and Virgin Mary.
When I see you all rainy, I have not a moment’s peace.
You blacken my life, and I sigh bitterly.
You’re a day just like the day I lost my joy.
Cloudy Sunday, you make my heart bleed.
1958. Δυο πόρτες έχει η ζωή (Dio Portes Echi i Zoi): Life has two doors. Lyrics: Eftichia Papagianopoulos. Music: Stelios Kazantzidis.
Kazantzidis also wrote songs, and this is perhaps one of his greatest.
Kazantzidis is perhaps an acquired taste. In my youth, I dismissed him as boorish, mawkish, too Oriental. I matured, I learned there’s a place for that in life. I learned that pain in life deserves wallowing in music. And after I took part in a drunken singalong to this, I could never dismiss him again.
I’m living my last night tonight.
And those who have embittered me so,
now that I am leaving life behind,
I forgive them all.
Everything is but a lie,
a breath, a sigh.
Like a flower, a hand
will cut us one dawn.
Where I am going, tears and pain have no purchase.
Suffering and sorrow
will stay behind in life,
and I will leave alone.
Life has two doors: I opened one and went in.
I took a stroll one morning,
and by the time sunset came
I left by the other.
1962. Ένα Δειλινό (Ena dilino): One evening. Lyrics: Mikis Theodorakis. Music: Mikis Theodorakis.
Theodorakis is a defining figure in Greek music and Greek politics; his star has been tarnished, but he expressed a generation, and that does not change, even if the generation has become disillusioned since.
This was likely his greatest song, where he transformed the bouzouki song into a a not-so traditional lament for the dead. The song was one of many Theodorakis used in his play on the reccent Greek Civil War—each and every one a hit.
And none more transcendental in its keening than this.
One evening, they bound you to the cross.
They nailed your hands, they nailed my insides.
They bound your eyes, they bound my soul.
One evening, they broke me in two.
They stole my sight, they took my touch,
My hearing remains, to listen to you, my child.
One evening, like the golden eagle,
swoop over the sea, swoop over the fields.
Make the mountains flower, and the people rejoice.
1974. Τα λόγια και τα χρόνια (Ta logia ke ta chronia): The Lost Words and Years. Lyrics: Manos Eleftheriou. Music: Yannis Markopoulos.
I’ve already posted about this song, and its cultural resonance, in Nick Nicholas’ answer to What are your favourite lyrics?
Yes, there’s wild applause whenever the altered lyrics alluding to the Athens Polytechnic uprising are sung (“Friday the Killer’s night”). That kept happening throughout the 70s.
What English words appear to be derived from Latin, but aren’t?
In a roundabout way: syllabus is ultimately derived from a garbling of the obscure Greek word sittyba, which got mangled progressively in manuscripts and then print editions of Cicero, and reinterpreted from its original meaning “title slip”.
Can someone write in their language using it’s grammatical structure while still using English words?
(Modern Greek > English)
If it is possible! You hear there, “It can someone to write in the language theirs using the grammar theirs but English words?” Hey not you us quit? For what you us passed, for revue? Not will I sit to you make theatre the how I speak, so you to break slab! Elsewhere these! Not they have slaughtered! Hey shoo!
Well I never! The nerve, “Can someone write in their language using their grammar but English words?” Why don’t you leave me alone! What do you take me for, a performing monkey? I’m not going to sit around making a show of how I speak for your amusement! Tell it to someone else! It ain’t happening. Get lost!
How many words does the Greek language have?
I wrote an extensive set of blog posts in 2009 under Ἡλληνιστεύκοντος (read them backwards), trying to deal with this question with a fixed(ish) corpus, that I was responsible for lemmatising: the TLG. It has a whole lot about the distinction between word tokens (individual instances of words), wordforms, and lemmata (dictionary words).
It starts with several posts about how pointless this question is. Which noone seems to pay attention to.
The count of lemmata for the Corpus in the TLG (ancient and mediaeval literature) plus PHI (inscriptions) was 214,000 in 2009. By the time I was terminated from the TLG in 2016, I had gotten recognition up to 240,000 lemmata.
For the strictly classical corpus, up to the 4th century BC, it was 66,000.
If we add Modern Greek and Modern Greek dialect, it’ll be more. I’ve seen a guess by Christophoros Charalambakis, director of the Historical Dictionary of Modern Greek (dialect dictionary) at the Academy of Athens, of 600,000. I think that’s implausible. Given Zipf, I think 350,000 to 400,000 for all periods of Greek is plausible.
OED has something like 600,000 for English.
Why do the English say “leftenant” and the Americans say “lootenant” when the spelling of “lieutenant” indicates a pronunciation like “lyewtenant”?
The American “lootenant” is easy: it’s a general rule of American English that [ju] after alveolar consonants is reduced to [u]: news, tune = nooz, toon. In British English, they are nyooz, tyoon. (And there is variation within American English.)
The lack of a French pronunciation is also regular: French ieu is rendered in English as -(y)oo. Thus in lieu of; adieu.
The oddity is the left- pronunciation. This is what the Oxford English Dictionary says:
The origin of the β type of forms (which survives in the usual British pronunciation, though the spelling represents the α type) is difficult to explain. The hypothesis of a mere misinterpretation of the graphic form (u read as v), at first sight plausible, does not accord with the facts.
I.e. that people misread lieutenant as lievtenant.
In view of the rare Old French form luef for lieu (with which compare especially the 15th cent. Scots forms luf– , lufftenand above) it seems likely that the labial glide at the end of Old French lieu as the first element of a compound was sometimes apprehended by English-speakers as a v or f.
Meaning, the unfamiliar French liøtenã, with a breathy pronunciation, could have been misheard in England as lyeftenant, a spelling present alongside leuetenant since the Middle Ages.
Possibly some of the forms may be due to association with leave n.1 or lief adj.
i.e. Folk etymology.
In 1793 Walker gives the actual pronunciations as /lɛv-//lɪvˈtɛnənt/, but expresses the hope that ‘the regular sound, lewtenant’ will in time become current. In England this pronunciation /ljuːˈtɛnənt/ is almost unknown. A newspaper quot. of 1893 in Funk’s Standard Dict. Eng. Lang. says that /lɛfˈtɛnənt/ is in the U.S. ‘almost confined to the retired list of the navy’.
So the old, predictable pronunciation had died out in England; the spelling survived, and it may be that either the old pronunciation survived in the US, or was revived as a spelling pronunciation.
What do contemporary Greeks think of Lord Byron?
To add to the others (which is why you must upvote the others):
Greeks revere Byron (to the point of Βύρων[ας] Viron(as) being a name they give their kids), because he was a prominent foreign supporter of the Greek War of Independence.
What contemporary Greeks do NOT know is that Byron was a Romantic poet, and that being a Romantic poet is why he was in Greece to begin with. He could have been an ironworker, and they would have still revered him for being this guy:
As opposed to being this guy:
Is Albanian a creole language?
*tosses head back chuckling*
Ah, I know where this question comes from. I did a drive-by shooting in a comment thread, saying “no, Albanian is not a creole”.
Fair enough that I should be asked why.
A creole in linguistics is not just a language that you think sounds mixed. It has a specific meaning. Pidgins arise as people speaking different languages try to communicate, without knowing each others’ languages very well. Creoles are languages that have developed out of a pidgin, as children learn it as their native language.
Now, the thing about pidgins that is truly characteristic is not that that they have a mixed vocabulary. They can in theory, because they are common improvised languages for people who speak different languages. But in fact usually they don’t: most modern pidgins arose in situations of slavery or near-slavery, and the vocabulary is mostly the colonialists’ vocabulary. (And often baby-talk at that, because the colonialists really did talk down to their slaves.)
What is characteristic about pidgins is that they have extremely simple, stripped down grammars. They have minimal inflection and rigid syntax. After all, they are the language you speak when you don’t have a language in common: so you end up stripping language to its basics.
Creoles (as linguists define them) have a lot more grammar than pidgins; and the grammar starts having the quirks and peculiarities and exceptions that we are used to in most human languages. But even so, compared to their source languages, they remain on the simple side grammatically. Compare Haitian Creole with French, for example.
Now, you will hear speculation that English is a creole. But that’s not because of the mixture of French and Germanic vocabulary after 1066. That was a socially stratified situation, nothing creole about it. If English is a creole (which I’m not convinced by), it happened earlier, when the Vikings came to town: the lexicon of the Norsemen and the Anglo-Saxons was the same, the inflections were different, so they just dropped the inflections. Which amounts to dropping the grammar, like pidgins do.
OK. Albanian has a recognisably Albanian core vocabulary. It also has a substantial amount of Latin vocabulary, and some Greek. In fact, the amount of non-Albanian vocabulary in Albanian is remarkable among Indo-European languages—just as it is for Armenian. So, it has a mixed vocabulary. Like English does.
But that doesn’t make it a creole. That just means intense language contact, and high social prestige of Latin and Greek—which is also how English got all its French and Latin vocabulary. If you’re looking for evidence of a creole, you want to check for a minimal grammar, with almost no inflection.
That’s not what you get in Albanian. What you get is all the inflections and complexity you’d expect of a relatively archaic Indo-European language. Hence, no creole.
There’s only one language in the Balkans for which there’s been speculation of creolisation: Tsakonian. The guy who speculated on it was Dirk Hesseling, who was an expert on both creoles and Tsakonian—and got a town meeting in Leonidion to vote a protest against him. Hesseling had invented a hammer, so everything to him seemed a nail; but Tsakonian has clearly undergone some sort of inflectional meltdown in its history. No sign of that in Albanian.
Do Australians cringe when non-native English speakers attempt to learn the Australian accent?
What Christine Leigh Langtree said: Vote #1 Christine Leigh Langtree’s answer to Do Australians cringe when non-native English speakers attempt to learn the Australian accent?
I’ll add that most dialect speakers dislike their accent being mimicked, not just Australians; I know I resented the hell out of 1960s Greek comedies’ bad imitation of Cretan. If you don’t get it right, it does come across as mockery.
BUT for non-native speakers… we make a lot of allowances, because we know it’s not mockery. The accent will gradually come naturally anyway. And the less self-conscious you are about it, the more readily it will happen. We have met IRL, Miguel, and your accent is still mostly US/Philippines—but not completely!