double τ

Here you can discuss all things Ancient Greek. Use this board to ask questions about grammar, discuss learning strategies, get help with a difficult passage of Greek, and more.
Post Reply
Hammurabi
Textkit Neophyte
Posts: 70
Joined: Mon Mar 20, 2006 6:40 am
Location: Paris, France

double τ

Post by Hammurabi »

Hi guys!

before anything I would like to say thank you because you are really helping me in this learning adventure:

now, I have a question about pronunciation: I've seen that many times the double τ was changed in the modern greek for a double sigma,

θάλαττη /// θάλασση
γλώττα ... γλώσσα

just to mention some of them, but I would like to know what it the pronunciation on classic greek for this double consonant, :oops:
I tried to find out in this forum in previous threaths but I could not find anything, also in the books... nothing...

thank you very much in advance for your help :D :shock:

annis
Textkit Zealot
Posts: 3399
Joined: Fri Jan 03, 2003 4:55 pm
Location: Madison, WI, USA
Contact:

Re: double τ

Post by annis »

Hammurabi wrote:now, I have a question about pronunciation: I've seen that many times the double τ was changed in the modern greek for a double sigma,

θάλαττη /// θάλασση
γλώττα ... γλώσσα
This is a dialect difference. Homer and Herodotus would have used the double sigma, too. The double tau is characteristic of Attic. But Koine, upon which modern Greek is based, used the double sigma.

As for the pronunciation, probably as written. The variation may go back to something like the ch in "church", or perhaps some /ts/ variation, and there are even signs of this in some dialects.
William S. Annis — http://www.aoidoi.org/http://www.scholiastae.org/
τίς πατέρ' αἰνήσει εἰ μὴ κακοδαίμονες υἱοί;

ThomasGR
Textkit Enthusiast
Posts: 444
Joined: Mon Aug 23, 2004 8:49 pm

Post by ThomasGR »

I was always curious to know why people came with writing some consonants double. In the above example, is /t/ or /s/ spoken double? It's somehow difficult to do this. There are also other examples like "gramma" or "alla", but I don't know any example with double /k/ or /χ/.

Chris Weimer
Textkit Enthusiast
Posts: 563
Joined: Mon Feb 13, 2006 9:34 am

Post by Chris Weimer »

It's from a loss of the /j/ in Greek.

annis
Textkit Zealot
Posts: 3399
Joined: Fri Jan 03, 2003 4:55 pm
Location: Madison, WI, USA
Contact:

Post by annis »


William S. Annis — http://www.aoidoi.org/http://www.scholiastae.org/
τίς πατέρ' αἰνήσει εἰ μὴ κακοδαίμονες υἱοί;

annis
Textkit Zealot
Posts: 3399
Joined: Fri Jan 03, 2003 4:55 pm
Location: Madison, WI, USA
Contact:

Re: double τ

Post by annis »

annis wrote: But Koine, upon which modern Greek is based, used the double sigma.
Um. That's misleading. "But Koine, from which modern Greek is descended" is better.
William S. Annis — http://www.aoidoi.org/http://www.scholiastae.org/
τίς πατέρ' αἰνήσει εἰ μὴ κακοδαίμονες υἱοί;

ThomasGR
Textkit Enthusiast
Posts: 444
Joined: Mon Aug 23, 2004 8:49 pm

Post by ThomasGR »



annis
Textkit Zealot
Posts: 3399
Joined: Fri Jan 03, 2003 4:55 pm
Location: Madison, WI, USA
Contact:

Post by annis »

ThomasGR wrote:The examples for Latin and Italian, I am afraid are the results of grammatical rules and the intervention of grammarians.
ὀτοτοτοῖ τοτοῖ!

Thomas, is there any linguistic quirk you will not attribute to the mischief of perverted grammarians? ;)

Because the pronunciation of 'octo' became 'oto', and to keep the historical records, you write it as 'otto', but I don't think one speaks any double /t/.
In the case of Italian you would be very much mistaken. Double consonants are distinctly pronounced. Nono and nonno are different words, with different meanings, differently pronounced.
The same goes for 'ekklesia', it's an invetions of grammarians to remind people that it is actually coming from two words 'ex' and 'kalw'>'exkalw'>ekkalw'>'ekklese'>'ekklesia'. But before the grammarians, did they spoke it as double /k/?
Yes. It is metrically guaranteed. Doubled consonants were pronounced doubly in ancient Greek.
Where people of the 5th century aware that some clusters /ts/ ot /tS/ are reduced to 'tt' or 'ss', or that any sound /j/ is missing in 'alla'?
Not that I'm aware of.
William S. Annis — http://www.aoidoi.org/http://www.scholiastae.org/
τίς πατέρ' αἰνήσει εἰ μὴ κακοδαίμονες υἱοί;

ThomasGR
Textkit Enthusiast
Posts: 444
Joined: Mon Aug 23, 2004 8:49 pm

Post by ThomasGR »

Maybe it's due to my stubbornness, but I have great incapability to pronounce a double /t/ or /k/ without putting a (glottal?) stop in-between. Though I can imagine a double /n/, /m/ only as lasting longer than usually.

Chris Weimer
Textkit Enthusiast
Posts: 563
Joined: Mon Feb 13, 2006 9:34 am

Post by Chris Weimer »

Did not thalatta and glotta become doubled consonants from the loss of the /j/ in Greek? I don't believe that was the case with either gamma or ekklhsia though, nor was it with the Italian examples. But the premise is there - a loss of a consonant to a doubled consonant. octo = otto.

And of course it's pronounced doubly, in Italian, in Greek, even in English. If not two full consonants, you at leat do one consonant aspirated, no?

ThomasGR
Textkit Enthusiast
Posts: 444
Joined: Mon Aug 23, 2004 8:49 pm

Post by ThomasGR »

I am convinced that the grammarians dictated the reasons for writing some consonants double without taking into consideration the real pronunciation. I am thinking of German words like "tag" (spoken "tak" or "ta/ch/" (/ch/ as in scottish Loch)), zug (=tsu/ch/), but they write it with g because in the plural form and other declensions its spoken as /g/, and we have "tage" (=tage) or "zuege" (=tsuege). Another example is "wald" (valt), "waelder" (velder). In the example for "wald" notice also the quasiumlaut "ae". I could say the same for Greek if there weren't some cases like "glotta" or "thalatta", were almost from beginning these words were written with double consonants. For modern speakers it's a tongue twister to speak /t/ twice without using a stop or aspiration in-between, but than the aspiration will be clearly auditive and will distort the euphonics of the word.

annis
Textkit Zealot
Posts: 3399
Joined: Fri Jan 03, 2003 4:55 pm
Location: Madison, WI, USA
Contact:

Post by annis »

ThomasGR wrote:Maybe it's due to my stubbornness, but I have great incapability to pronounce a double /t/ or /k/ without putting a (glottal?) stop in-between. Though I can imagine a double /n/, /m/ only as lasting longer than usually.
That's exactly right for doubled /n/ and /m/, and is how other doubled stops work.

The doubled consonants are not rearticulated! Here, in fat alpaca the /t/ is single. In fat tom cat the /t/ is doubled, at least at a normal speaking pace.
William S. Annis — http://www.aoidoi.org/http://www.scholiastae.org/
τίς πατέρ' αἰνήσει εἰ μὴ κακοδαίμονες υἱοί;

Bardo de Saldo
Textkit Enthusiast
Posts: 392
Joined: Tue Mar 15, 2005 11:34 pm
Location: Newer Mexico

Post by Bardo de Saldo »

Couldn't they be doubled just to show that the preceding vowel is long, just like English does (inversely) with words like mite and mitten?

Isn't Achilles spelled with one or two lambdas just to suit the meter?

annis
Textkit Zealot
Posts: 3399
Joined: Fri Jan 03, 2003 4:55 pm
Location: Madison, WI, USA
Contact:

Post by annis »

Bardo de Saldo wrote:Couldn't they be doubled just to show that the preceding vowel is long, just like English does (inversely) with words like mite and mitten?
Greek already had ways to do that: ἀγγέλλω, ἀγγελῶ, ἤγγειλα.
Isn't Achilles spelled with one or two lambdas just to suit the meter?
That's a Homeric licence to get a difficult name into the verses, which later Greeks and grumpy Roman poets alike scolded him for.
William S. Annis — http://www.aoidoi.org/http://www.scholiastae.org/
τίς πατέρ' αἰνήσει εἰ μὴ κακοδαίμονες υἱοί;

Bert
Textkit Zealot
Posts: 1889
Joined: Sat May 31, 2003 2:28 am
Location: Arthur Ontario Canada

Post by Bert »

annis wrote:
The doubled consonants are not rearticulated! Here, in fat alpaca the /t/ is single. In fat tom cat the /t/ is doubled, at least at a normal speaking pace.
Back in 2003
viewtopic.php?t=635&highlight=trilled+lambda I asked a question concerning trilled Lambdas. Is this what that refers to?

annis
Textkit Zealot
Posts: 3399
Joined: Fri Jan 03, 2003 4:55 pm
Location: Madison, WI, USA
Contact:

Post by annis »

Bert wrote:I asked a question concerning trilled Lambdas. Is this what that refers to?
Bert, I must confess that to this day I have no idea what a "trilled" lambda would be or sound like. I suspect he's using the word in a sense I don't know.
William S. Annis — http://www.aoidoi.org/http://www.scholiastae.org/
τίς πατέρ' αἰνήσει εἰ μὴ κακοδαίμονες υἱοί;

ThomasGR
Textkit Enthusiast
Posts: 444
Joined: Mon Aug 23, 2004 8:49 pm

Post by ThomasGR »

annis wrote: Bert, I must confess that to this day I have no idea what a "trilled" lambda would be or sound like. I suspect he's using the word in a sense I don't know.
I can understand how a 'trilled' /l/ sounds. /l/ is one of the consonants one can "sing", that's keep it sounding as long as one wishes, like /s/ or /m/. Something is not possible to do with /t/ or /g/ and other consonants. Sorry, but I tried to keep /t/ and /k/ sounding double and was not able to do it. Ancient Greeks must have had a fine ear to distinguish the duration of these consonants so as to keep them recorded correctly. I doubt that they were at such an early stage such great linguists and keen to record the sounds perfectly. Now I am thinking more if there was another sound they wanted to write, like a ‘harder’ or ‘louder’ /t/ than usual, where the plosive part is more forced than usual, but without resulting to an aspiration (like in theta) or to a voiced one like in delta. Just my thoughts.

annis
Textkit Zealot
Posts: 3399
Joined: Fri Jan 03, 2003 4:55 pm
Location: Madison, WI, USA
Contact:

Post by annis »

ThomasGR wrote: I doubt that they were at such an early stage such great linguists and keen to record the sounds perfectly.
For those who speak languages with geminate consonants — the Italians, the Japanese — the distinction differentiates words. I doubt they regard it as subtle.

I'll see if I can find some sound files contrasting single from double consonants.
William S. Annis — http://www.aoidoi.org/http://www.scholiastae.org/
τίς πατέρ' αἰνήσει εἰ μὴ κακοδαίμονες υἱοί;

ThomasGR
Textkit Enthusiast
Posts: 444
Joined: Mon Aug 23, 2004 8:49 pm

Post by ThomasGR »

I see it now, that could be a good explanation. Do we have any information that Ancient Greeks were aware of it and used therefore gemination?

annis
Textkit Zealot
Posts: 3399
Joined: Fri Jan 03, 2003 4:55 pm
Location: Madison, WI, USA
Contact:

Post by annis »

ThomasGR wrote:I see it now, that could be a good explanation. Do we have any information that Ancient Greeks were aware of it and used therefore gemination?
That's what the double consonants signify.
William S. Annis — http://www.aoidoi.org/http://www.scholiastae.org/
τίς πατέρ' αἰνήσει εἰ μὴ κακοδαίμονες υἱοί;

swiftnicholas
Textkit Enthusiast
Posts: 408
Joined: Fri Jan 07, 2005 4:04 pm
Location: New York

Post by swiftnicholas »

ThomasGR wrote:Sorry, but I tried to keep /t/ and /k/ sounding double and was not able to do it.
I have a terrible time understanding this stuff without actually hearing it. But would the word "bookkeeper" be an example of a double k sound, as opposed to "book, keeper" with a pause? Or is it becoming a single k sound with an aspiration, like "book-heaper"?

Chris Weimer
Textkit Enthusiast
Posts: 563
Joined: Mon Feb 13, 2006 9:34 am

Post by Chris Weimer »

It sounds like it to me, Nick. Book-keep-(h)er is how I pronounce it, and there is a difference between that and boo-keeper, however subtle.

anthropostis
Textkit Neophyte
Posts: 8
Joined: Sun Apr 16, 2006 2:44 am

Post by anthropostis »

The ancient Greek grammarians were perfectly aware of geminate consonants and their pronunciation--the already cited example of a double consonant creating a long/heavy syllable in metrical texts is discussed by, for instance, Dionyisius Thrax.

Two other matters briefly: ThomasGR talks about having to insert a glottal stop in geminate consonants. I doubt there's actually a glottal stop there, but there is a sort of pause of articulation, which is exactly what one does in careful articulation of geminate stops.

In terms of "trilled" lambda, I'm not familiar with the original discussion, but I would guess that someone was talking about the distinction between "light" and "dark" l, which are simply allophones in English (generally, in standard English light pre-vocalic, dark post-vocalis). Greek lambda was, as far as anyone knows, always light.

annis
Textkit Zealot
Posts: 3399
Joined: Fri Jan 03, 2003 4:55 pm
Location: Madison, WI, USA
Contact:

Post by annis »

anthropostis wrote:In terms of "trilled" lambda, I'm not familiar with the original discussion, but I would guess that someone was talking about the distinction between "light" and "dark" l,
Given the context, I doubt it.
which are simply allophones in English (generally, in standard English light pre-vocalic, dark post-vocalis). Greek lambda was, as far as anyone knows, always light.
Nope. Or not in every dialect, at least. In some, lambda in some positions became υ, which is a more usual development of the dark /l/.
William S. Annis — http://www.aoidoi.org/http://www.scholiastae.org/
τίς πατέρ' αἰνήσει εἰ μὴ κακοδαίμονες υἱοί;

anthropostis
Textkit Neophyte
Posts: 8
Joined: Sun Apr 16, 2006 2:44 am

Post by anthropostis »

Well, that teaches me to jump into conversations of which I don't know the original context!

On your second matter, interesting. Which dialects and where? I don't doubt the outcome of dark l in some dialects, but I'm not familiar with them.

More randomly: how does one quote a previous post on these forums?[/i]

annis
Textkit Zealot
Posts: 3399
Joined: Fri Jan 03, 2003 4:55 pm
Location: Madison, WI, USA
Contact:

Post by annis »

anthropostis wrote:On your second matter, interesting. Which dialects and where? I don't doubt the outcome of dark l in some dialects, but I'm not familiar with them.
From Palmer's The Greek Language, p.209, "A velar coloring before a consonant is suggested by the occasional substitution of υ in Cretan (ἀδευπιαί for ἀδελφιαί)."
More randomly: how does one quote a previous post on these forums?
At the upper right of each post is a "quote" button which will drop you into a reply form with the entire thing in quote tags.
William S. Annis — http://www.aoidoi.org/http://www.scholiastae.org/
τίς πατέρ' αἰνήσει εἰ μὴ κακοδαίμονες υἱοί;

anthropostis
Textkit Neophyte
Posts: 8
Joined: Sun Apr 16, 2006 2:44 am

Post by anthropostis »

From Palmer's The Greek Language, p.209, "A velar coloring before a
consonant is suggested by the occasional substitution of υ in Cretan (ἀδευπιαί for ἀδελφιαί)."
Forgive me for forgetting the inscriptional evidence, though one should point out that this is limited to Cretan, and while the Latin grammarians are, for example, reasonably clear on the two different outcomes of /l/, the Greek grammarians do not so distinguish, and there seems to be no reason to assume a "velar coloring" in other dialects based on their evidence or the outcome of lambda in modern Greek.
At the upper right of each post is a "quote" button which will drop you into a reply form with the entire thing in quote tags.
Many thanks! Amazing how one misses the obvious.

annis
Textkit Zealot
Posts: 3399
Joined: Fri Jan 03, 2003 4:55 pm
Location: Madison, WI, USA
Contact:

Post by annis »

anthropostis wrote:Forgive me for forgetting the inscriptional evidence, though one should point out that this is limited to Cretan, and while the Latin grammarians are, for example, reasonably clear on the two different outcomes of /l/, the Greek grammarians do not so distinguish, and there seems to be no reason to assume a "velar coloring" in other dialects based on their evidence or the outcome of lambda in modern Greek
As I said, the dark /l/ appeared in some dialects at least. Given the examples of other languages where some groups drop /l/ to a glide, while others do not, I'm not prepared to dismiss out of hand the idea that the dark /l/ might have been more widely used. I'm thinking of Old Occitan at the moment, proencal vs. proencau.

In my own pronunciation, I just stick with a clear /l/, though I have to fight the urge to a dark /l/ when the lambda is doubled for some reason. I blame my study of Arabic for that.
William S. Annis — http://www.aoidoi.org/http://www.scholiastae.org/
τίς πατέρ' αἰνήσει εἰ μὴ κακοδαίμονες υἱοί;

anthropostis
Textkit Neophyte
Posts: 8
Joined: Sun Apr 16, 2006 2:44 am

Post by anthropostis »

annis wrote:As I said, the dark /l/ appeared in some dialects at least. Given the examples of other languages where some groups drop /l/ to a glide, while others do not, I'm not prepared to dismiss out of hand the idea that the dark /l/ might have been more widely used.
Well, put in that way, I'm not prepared to dismiss it either. But I don't think that it's much to worry about in Greek.
In my own pronunciation, I just stick with a clear /l/, though I have to fight the urge to a dark /l/ when the lambda is doubled for some reason. I blame my study of Arabic for that.
That's a pretty cool reason to be able to blame that on. In any case, I don't think the precise quality of pronunciation of the lambda is really a problem that ranks very high on the list of issues facing modern learners of ancient Greek.

swiftnicholas
Textkit Enthusiast
Posts: 408
Joined: Fri Jan 07, 2005 4:04 pm
Location: New York

Post by swiftnicholas »

I'm very curious about the terms "light" and "dark" describing sounds. I finally managed to get a copy of Vox Graeca a little while ago, but I don't have it with me now. Can I read about it there? Will it be in the index?

~N

Chris Weimer
Textkit Enthusiast
Posts: 563
Joined: Mon Feb 13, 2006 9:34 am

Post by Chris Weimer »

Light L is the sound in "lullaby" while dark L is the sound in "bull."

IVSTINIANVS
Textkit Neophyte
Posts: 73
Joined: Sun Jun 12, 2005 2:07 am
Location: Somerville, MA (Boston Area)

Post by IVSTINIANVS »

Very enjoyable discussion, and (of course) William Annis explained everything beautifully.

ThomasGR is perfectly correct in that there is a difference between resonants (including nasals and liquids) which are "continuant" and (as he says) can be "sung" for as long as you want, and a stop (or plosive) which cannot be prolongued.

A doubled stop is not "prolongued". It is simply that there are two adjacent stops: one syllable ENDS in a stop (instead of being open), and the next syllable begins in what happens to be the same stop. And in languages such as Japanese and Italian it is an extremely noticeable and important distinction.

The example of "bookkeper" vs. "boo-keeper" illustrates it perfectly. I'm afraid I don't know enough German to think of similar examples that could help ThomasGR understand.

The only other thing I would add to the discussion of double sigma vs. double tau is this: there were some dialects where this sound was written with a special consonant which has (I believe) no Unicode code-point. It looked like a capital T with fairly large serifs on the crossbar. I'll try to find a picture of one and upload it. Some scholars have speculated that this letter (which may have descended from the archaic Greek letter San) was used to write a sound such as the "ch" in "church".

Some scholars have also speculated that in fact the -ss- or -tt- spelling may have been an imperfect attempt to represent this, or a similar sound (e.g., "ch" or the "sh" in "ship"), and that even in Attic Greek, it may not have been pronounced as a double tau.

If so, however, then it is surprising that none of the scholars at the time mentioned the sound, so it is probable that by the classical period at least, -tt- and -ss- were pronounced as written.

That's just my two and a half cents. :-)

Post Reply