Thanks for adding your opinion!

I am only starting to learn Ancient Greek, restored Attic. I am quite specific about Attic and keen with best language pronunciation as possible. To you, Ancient Greek professionals, I'd like to submit an idea. A major hurdle in proper language pronunciation acquisition is insufficient access to native speech. This has always been an issue with dead languages (though I like to think that Greek isn't really dead, nor even Latin, to the extent that various linguistic continuities can be found). So here goes my proposal... Why not take advantage of AI and LLMs, all developments, since some years ago already, to have a software learn and produce a natural sounding Attic Greek, etc. (any language of interest). Such a software could be refined several times, until reaching a version where most experts will agree they are absolutely or near absolutely happy with. Such a product could then be used by restored Attic Greek learners all over the world, listening to endless hours of ancient literature, thereby being able to acquire, if so they wish, a natural sounding Attic Greek. We could partially ressucitate the language and experience the joy of speaking it with near-native ease and pronunciation. At least for those who care this much about pronunciation. Another application would be making use of recorded native voices, from the purest speakers of each language today at risk of extinction or accelerated degradation (of which there are thousands), so we can teach AI how to speak it, for the benefit of concerned populations. Such a tool could be used to stop and even reverse language degradation and loss in a multitude of places around the world, Europe included. As a native Catalan speaker, I feel despair at how my language is fast becoming a radically impoverished and transvestite language, where not only words and expressions are lost by the thousands, every decade, but even the phonetics and syntax etc. are quickly being corrupted. Soon, the Spanish political framework and demographics, combined, will make this evolution unstoppable and irreversible. The only way we could prevent such catastrophic change would be some political miracle and/or a technological one, with AI.

Did you ever get ideas about real languages with real pitch accent in action pronounced by native speakers ? 1) Finnish 2) Japanese 3) Classical Arabic I include also Sanskrit because some speakers can show this caracteristic very clearly into the traditional recitation.

@@cattubuttas4749 , whatever the phonetics of a language, if it is alive, native speaker recordings can be used to train AI systems. Whether it is tonal, has pitch accent, clicks, or whatever. Sanskrit is alive, but I have no idea how today's real life phonetics are regarded in terms of a standard language. Around 15,000 people in India claim to speak it as their native first language, but how are their pronunciation and other language features regarded? They might be regarded as dialectal or modern, relative to some classical standard. Many more people learn Sanskrit as a second language. But again, how is their pronunciation (or prosody in general) regarded? No idea. Language experts should have a debate, in each case (for each language), about what standard(s) of language it is desirable for an AI system to learn/develop. Both classical and modern standards seem desirable for some languages. Even dialectical standards. Why not go for everything while we can? Everything is valuable. I have just started learning Sanskrit, by the way, together with Ancient Greek. I am still learning the alphabet. 😅

@@nomcognom2414 for the "live" languages there is no problem in finding recordings but for Sanskrit I use ASSIMIL recordings which are great for getting an idea. They have free samples of chapter 1, chapter 50 and chapter 100 of the book. Assimil courses recordings for Latin and Ancient Greek are less valuable in particular the Latin ones are very bad, Greek ones barely acceptable.

Thank you!

Interesting, I hadn't heard of the claim that the pitch was a one-fifth jump! That is pretty high. Though still plausible, at least in theory - if I sound out the Mandarin tones to myself, I can hear that in isolation at least, the high first tone and the low third tone are maybe even a bit more than a fifth apart in pitch. In rapid speech the tones do all kinds of interesting things to each other and probably don't sound as far apart as when someone is demonstrating pitch for pedagogical purposes. If "a musical fifth" was what the grammarians wrote down, then maybe when the Ancient Greek teachers were speaking slowly and emphatically to students it was a fifth interval, but it could have been a bit flatter in rapid speech.

@@foundinantiquityancientgre8010 I believe Dionysius of Halicarnassas wrote that the pitch accent was confined to an interval of about a fifth. But its assumed he meant that it was never more than that, not that it was always such a jump.

@@foundinantiquityancientgre8010not to discount the possibility of it being as large as a 5th but in my intuition as someone with basic mandarin and fluent Japanese; tonal languages like mandarin with 4 or more tones have a larger range of frequencies going on than pitch accent languages (based on an admitted sample size of 1 of each lol). In any case i suspect the highest tone and lowest tones of mandarin have around twice the distance between them (maybe more) than the high and low of Japanese. I’m interested in hearing from people from other pitch accent languages.

So actually Japanese is, like Greek, only able to have one accented mora per word. This mora can be higher than surrounding mora, or it can simply be marked with a downstep. The preceding mora aren't really high at a phonological level, but they tend to approach or equal the accented mora in pitch, since it's the downstep that marks the accent, not absolute pitch. It's quite likely that Greek worked the same way given the musical evidence. That is, the accented mora doesn't necessarily need to be higher than the preceding mora, as long as its followed by a downstep. The only phenomenon definitely present in Japanese and probably not present in Greek is that there tends to be a noticeable rise in pitch between the first two morae of the word if the accent is later in the word, which results in transcriptions like the one you gave, 'wàtáshí'. But this is really more noticeable in emphasized words - otherwise, it's really just the downstep that's mandatory. Greek very well may have been exactly the same. As for the idea that the difference in pitch should be of a particular musical interval, this can probably be discounted.

​@@mikem9001 I'm not sure we really know what the ancient Greek intervals were. When in Greece I saw the music notation of the Ode to Apollo written on a series of stones and no one as yet has been able to decipher it. In any case, our span of a 5th is based on the Well-Tempered Clavier of JS Bach and still exists today in modern music. But Greek intervals could be smaller than our 12 steps of the chromatic scale. Some music from the Middle East uses quarter-tones, and if the Greeks used a system such as that, their chromatic scale might have as many as 27 steps, making a leap of a 5th resembling a span of 2 1/2 steps in our system. It's all highly speculative but it's a possible explanation that would avoid the extraordinarily large, cumbersome leap of our 5th.

Modern Greek normally do not distinct length of the vowels, but as a native speaker if I observe it while I read, most likely pitch rules fall into place.

Yes, I think Modern Greek has retained a kind of vestigial pitch accent even if it isn't necessarily done consciously. Certainly it has a very flowing, sing-song quality to it.

Thanks! 😃

Glad it was helpful!

Thank you!

χάριν σοι έχω!

As a German speaker the aspirated allophones are pretty natural to me. It's the unaspirated ones that I have trouble with, contrary to what is often said.

@@firstaidsack I'm an English-speaker and I'd say the same. That's why I mentioned Spanish. My Spanish is rudimentary but I am generally quite good at the mimickry of accents. The first time I ever heard Spanish my reaction was "wow, that's quite hard to follow, as they are swallowing the consonants". As in most contemporary languages native to Europe and spoken in Europe, there is no allophone issue: a Spanish person could aspirate all their consonants and it wouldn't cause misunderstandings. But I would put money on Ancient Greek's unaspirated consontants having sounded quite Spanish.

Thank you, that is a pretty neat idea! I know there are some performers who do a good job of recreating plausible musical settings for Homer's epic poetry. It's really hard to get into the right headspace for doing that though - Ancient Greek music is fundamentally different from Western harmony, and it's like speaking in a different musical language.

@@foundinantiquityancientgre8010 Plus, we will pay you in obols, so don't get too excited about the money.

I had a similar experience!

Thank you! I'm glad this helps provide a practical way to recreate pitch accent! It is still plausible to speak Ancient Greek with a fifth interval rise, but I suspect that in rapid speech it would get flatter. At least, the song melodies in Ancient Greek seem pretty happy to frequently use intervals smaller than a fifth when rising to an acute accent.

Dionysius of Halicarnassus - he probably meant that it was no more than a fifth, not that it was always that far.

I stated this above, but their system of what a fifth is could be different than ours.

@@foundinantiquityancientgre8010 You consistently use exactly a perfect fifth as the interval between the two pitches in this video.

I consider this a well deserved comment.

No but contemporary Hebrew had plenty of glottal stuff going on that was even more exotic. Nothing 21st century about it.

Thank you! With άνθρωπός εστιν, the word following άνθρωπος is 'εστιν' which is an 'enclitic' or basically a word which usually lacks an accent. Enclitics do weird things to the accents around them because they kind of combine with the previous word to make a super long word, and that sometimes means the preceding word gets a second accent on it.

What she said! I would dare to add, that I find it helps to think of the last part of the word as separated off and joined with the following enclitic. So it becomes "anthrop osestin". You are probably already used to reading "de" as almost joined with the word it follows, so this is really just the reverse process. I don't claim to have any great knowledge about this, but I've just found it makes the two raised pitch accents fit well together.

Thank you for your kind words! I do agree, a well-executed Erasmian is not very different at all in practice to a well-executed Lucian or Attic pronunciation. It's nice to hear you're using pitch accent with your Erasmian.

It’s because some of the ”diphthongs” were actually /aj/ instead of /ai/ 😊

It isn't.

Learn the only real pronouciation which is The Greek of native speakers all others are artificial and hypothetical. Greek language is a living language and still the same language and not other like Latin and Italian which is different languages.

@@user-vs8vf1ij9x nonsense, it hath changed greatly over time

@@strongbutthead the only nonsense here is to try to pronounce a language with an accent that never exist and native speakers don't do

@@user-vs8vf1ij9x Trying to pronounce classical Greek in a modern Greek way is artificial and hypothetical. Be guided by science, rather than by unthinking prejudice.

@@mikem9001 What science you ape, we have zero ways of knowing how they sounded like. Best we can do is pronounce them by how we speak today.

μα η ιστορική προφορά της ελληνικής ήταν μελωδική όπως αυτή που έχει σήμερα η νορβηγική ή ιαπωνική.

@@velvetcroc9827 1⁰ με τον όρο "ιστορική προφορά" αναφερόμαστε σ' αυτή που μιλάμε και σήμερα, ενώ με τους όρους (artificial, reconstructive, hypothetical) αναφερόμαστε στις τεχνίτες (ψεύτικες) προφορές που προσπαθούν να αποδώσουν υποθετικά την προφορά της αρχαίας, πράγμα αδύνατο διότι δεν υπήρχε μαγνητόφωνο εκείνη την εποχή. 2⁰ για την νορβηγική δεν γνωρίζω αλλά τουλάχιστον για ιαπωνικά έχω να πω ότι δεν έχουν καμία προσωδία. 3⁰ δυστυχώς όταν προσπαθούμε να μιμηθούμε άκριτα κάτι που δεν ξέρουμε, μοιάζουμε περισσότερο με μίμους. και αν κάποιος αρχαίος Έλληνας μας άκουγε μάλλον θα γέλαγε. Αλλά εκτός των άλλων είναι και ασέβεια προς την ελληνική γλώσσα και τους Έλληνες να μιλάει κάποιος τα ελληνικά έτσι ακατάληπτα, ειναι το ίδιο με το να διαβάζει κάποιος το Σέξπιρ και αντί π.χ. για made (μέιντ) να διαβάζει (μαδε)

@@user-vs8vf1ij9x Πάτα Japanese pitch accent για να καταλαβαίνεις τι εννοούμε με τον όρο μελωδική προφορά. Εσύ για ποιό λόγο νομίζεις ότι εφευρέθηκε το πολυτονικό σύστημα; Έτσι για ομορφιά; Και δεν χρειαζόμαστε κανένα μαγνητόφωνο. Υπάρχουν πάμπολλες αποδείξεις από την γραμματεία (έχουμε μέχρι και αρχαίους γραμματικούς που μας εξηγούν οι ίδιοι κάποια πράγματα), τις επιγραφές καθώς και την ίδια την επιστήμη της γλωσσολογίας.

@@velvetcroc9827 (παρακαλώ διάβασε το όλο πριν απαντήσεις, αλλα και το προηγούμενο γιατί συμπλήρωνα κατι ενω απάντησες) Δεν είπα οτι τα αρχαία ελληνικά δεν είχαν προσωδία αλλα τα ιαπωνικά (άλλο το pitch accent δλδ τονισμός άλλο προσωδία)(εν πάση περιπτώσει δεν είναι το θέμα προς συζήτηση). Ουτε είπα οτι η προφορά έχει μείνει ατόφια τρείς χιλιάδες χρόνια. Είπα πως όλες αυτές η προφορές εκκλίνουν της πραγματικότητας. Τώρα όσον αφορά τις γλωσσολογικές αποδείξεις οι περισσότερες καταρρίπτουν της ερασμιακές προφορές (συν ο ίδιος ο Έρασμος) παραδείγματα: (Γαία αλλα γεωγραφία, γεωλογία κτλ) (εικόν>icon λατ. Ουρανός>Uranus λατ. Βο-ος συναίρεση βους αλλα δωρικός βως) (ω,ουF>j>∅) οι δίφθογγοι φαίνεται να προκύπτουν μόνο απο την ένωση με το F. όσο για τους αρχαίους συγγραφείς ο Θουκυδίδης αναφέρει ότι τον λιμό τον μπέρδεψαν με τον λοιμό αρα δεν υπάρχει αμφιβολία ότι τουλάχιστον το οι είχε μονοφθογγιστεί . Για να ειμαι ακριβοδίκαιος ναι φαίνεται πως υπάρχει κάποια βάση ως προς τα υ και η αλλα δεν είμαστε σίγουροι για την ακριβή φωνητική τούς αξία το (η) ηταν μάλλον ενδιάμεσο ε και ι ή το υ μάλλον ενδιάμεσο κλειστό ο και ι ή ο και ου. Και πρέπει να θυμόμαστε ότι οι αρχαίοι δεν είχαν εναν τρόπο προφοράς, αλλα οτι αυτή μπορούσε να διαφέρει απο λέξη σε λέξη, πόλη σε πολη και άτομο σε άτομο. Λοιπόν όπως είναι εμφανές η γλωσσολογία δεν μπορεί να μας διασαφήνισει την ακριβή φωνητική αξία αλλά μόνο τις διαφοροποιήσεις που υπάρχουν μεταξύ τούς. Οπότε η φράση (δεν είχαν κασετόφωνο) δεν ειναι ειρωνική (btw irony αγγ.) Αλλα όντως δεν έχουμε τρόπο να ξέρουμε την ακριβή προφορά, και αρα κάθε μοντέλο πλην της ιστορικής είναι επισφαλείς.

@@user-vs8vf1ij9x Η προφορά των αγγλικών της εποχής της Σέξπιρ διέφερε αρκετά από τη σημερινή βρετανική και άμα ψάξεις θα βρεις απόπειρες ανάγνωσης των έργων με την τότε προφορά. Τώρα αν αυτό θες να το πεις ανούσιο πες το άλλα ασέβεια δεν είναι.

That's Byzantine Greek, stupid. Not Ancient Greek.