Yesterday’s Pi Day got me to thinking…
I’ve always pronounced it (and heard it pronounced) as if it rhymes with “high.” But in the absence of recordings, for all I know the Ancient Greeks could have pronounced it “pee” or “pih” or “Larry.”
Do linguists (cunning or otherwise) have any means of determining how the ancients pronounced their languages? Or are we just taking our best guess?
For written languages, one key is rhyming poetry.
As for “pi”, I really can’t say. However, continental vowels are (at least these days) rarely long like “pie”.
This.
You’ll notice that things like the Canterbury tales, when read out in a modern American accent, don’t exactly always rhyme. Since it seems obvious that the intent was to have an ongoing rhyming scheme, it stands to reason that it once rhymed.
I heard it read once, something like:
*Wan aprilluh with hees show-res soot
whaht the rots off Mahrch haith peerc-ed too the rut * (ryhmes with “soot”)
etc.
Phi is another Greek letter often used in mathematical contexts, and I’ve heard it pronounced both as “fie” and as “fee.” One site discussing the pronunciation mentions in passing that “in Greek the letter we call PI is also pronounced PEE.” I can understand why we don’t pronounce it that way, but apparently, Modern Greeks do pronounce it “pee.” (And FWIW the Greeks didn’t use the letter pi to stand for the number 3.14159…, but that’s irrelevant to your question.)
As others have pointed out, rhyming poetry is a big clue to pronunciation—but do we have rhyming poetry from ancient Greece? And even so, that would only tell us which words sounded like each other, not what they sounded like in an absolute sense.
In Ancient Greek, phi would sound the most like English “pee,” because we aspirate our initial Ps. The letter pi would sound like “spee” without the S. Besides that (in my accent, at least) the English ee sound isn’t a pure vowel like it should be in Greek. I don’t know about the length of the vowel.
You’re aware that “pee” and ““spee” without the S” sound exactly the same, right?
No, they don’t. When English speakers say “spee,” they do not aspirate the P. When they say “pee,” they do.
It’s one of those subtle things that pretty much only trained linguists can hear (or native speakers of languages that differentiate between aspirated and non-aspirated). 
Touch your fingertips to your lips and say “pee” then “spee.” You will feel a puff of air with the “pee” but not the “spee.” Unless you’re a native of India. In their accent, they omit the puff of air, so words starting with “p” almost sound like they start with “b” instead (to an American.)
Whan that Aprille with his shoures soote,
The droghte of March hath perced to the roote,
We had to read it aloud in class, in the original. It does rhyme, and those words are pronounced “sota” and “rota”.
Found this:
1 Wah-n that Ah-prill with hiss shō-urs sō-tuh
2 Thuh drog-tuh uh’ March hath pair-cèd to the rō-tuh,
I felt a puff of air with each… American native.
Sure. That’s how we know that “tea” once rhymed with “bay” not “bee.” Alexander Pope was too much a perfectionist to include that as an off-rhyme, and he rhymed it that way. (Actually, “obey,” but you get the point.)
Actually, doesn’t it still rhyme over there? (“Will ye bey wahntin’ a cuppa tay?”) Or am I confusing English and Irish accents?
Is the non-aspirated sound something closer to what we might recognize as a “b”?
“Sparring” (aspirated) vs. “Sbarro” (not), for example?
Only to native English speakers. It’s something that speakers of many other languages can differentiate easily.
In a sense. It’s the reason that “Beijing” has its current spelling. There isn’t an actual ** phoneme in Mandarin Chinese. It’s actually an unaspirated [p].
Mad Magazine did a parody of this once, mid-1960’s or so.
Wan thot Aprille swithen potrzebie
Tha’ bergrid prilly gives one heebie-jeebie
You’ll still get a burst of air with unaspirated consonants, but it’s not quite as strong as aspirated ones. /p/ is plosive, so by definition some air must be expelled.
English major here - I agree entirely with this.
Doesn’t this imply that unaspirated vs. aspirated /p/ are phonemically distinct* in Mandarin?
*i.e., you might have two completely distinct words between which the only difference is whether the /p/ is aspirated.