_http://www.reddit.com/r/AskHistorians/comments/146zb6/original_pronunciation_of_ceasar/ said:
[–]thatfool 1394 puntos 1 año atrás*
There are many surviving texts, but Latin is a very old language that has been used for thousands of years and its pronunciation has changed significantly during its long history.
Case in point: The Church Latin pronunciation of "Caesar" is more like "Chesar" (Ch like in the word "church", e like in "red"). But that's not how the Romans pronounced it. (Also note that there would be many dialects of Church Latin, too, if Pius X hadn't declared the Roman pronunciation standard.)
Of course not all of the sources we have for pronunciation are learning guides, but there are a few texts on language by scholars. Priscian is probably the most well known Roman gramarrian for his Institutiones Grammaticae ("Grammar Basics").
Another thing is that we render Latin texts completely differently nowadays. It wouldn't have been "Gaius Julius Caesar" during his time, it would have been CAIVS IVLIVS CAESAR. We've changed the C to a G and the first I in IULIUS to a J because we actually have G and J letters now that represent these sounds better. We don't, ever, write UIA, we always spell it "via". The Romans didn't consider U and V to be different letters, but we do now.
(Edit: The G letter existed during Caesar's time, but the name Gaius is older. Caesar's name is recorded with both spellings. I've also changed U's to V's, for some reason I didn't remember to do that even though I wrote that bit about U and V in the same paragraph.)
But as I wrote, we do have classical Roman sources for Latin pronunciation as well as other hints. Let's go over the word Caesar again, just to illustrate what kinds of hints we have. It's not just literal "this is pronounced by forming this shape with your mouth".
C:
Many old Latin inscriptions use it interchangeably with K and Q. (Also representing the G sound at the time, in addition to K.)
Grammarians like Quintilian (1st century) and Priscian (5th) write that they represent the same sound.
Marius Victorinus (4th) writes that X sounds like CS.
Diomedes (4th ... no the other Diomedes) writes that Q before a vowel other than U is a contraction of C and U.
Greek translators always transscribe C as kappa.
Misspellings of Latin words in inscriptions where C is replaced by a different sound (e.g. "paze" instead of "pace") appear much later.
If you spend some time with the letters C, K, Q, you'll find that in very early Latin, the vowel that followed determined which letter was used. For example, "pecunia" would have been spelled "pequnia" because it's followed by a U. So the sounds might actually have been different at some point, but the Romans then went and changed most of these to C's, so it was probably not a very big difference. Think the difference between K and G (which only got its own letter in the Latin alphabet in the 3rd century BC, at which point sources start explaining how it's different).
AE:
In Church Latin, this is actually the E sound (as in "red"), but we can actually see that change happen as AE is replaced in Latin texts by (ae) and (e with comma) in medieval texts. Modern Romance languages are similar. But this started about a thousand years after Caesar.
Pre-classical Latin actually uses AI in places where classical Latin uses AE. It's possible that this similarly corresponds to a change in pronunciation.
Greek translators transscribe it as (alpha iota).
The pronunciation of Latin loan words in Germanic languages is a big hint. The obvious example is "Kaiser", the German word for "emperor" that is directly derived from "Caesar" -> Old High German "Keisar" -> Middle High German "Keiser" -> modern standard German "Kaiser".
Other dipthongs did survive in modern Romance languages, for example "au" still exists in some of them as a diphtong. Many of the above points apply to it as well, for example it was transscribed into Greek as αυ (alpha upsilon).
There actually is a bit of controversy over this. It's generally accepted that AE was pronounced as a dipthong, but interestingly, we do have classical Latin sources for AE being replaced by E in rural Latin (but perhaps this means city Romans didn't do this). E.g. Varro gives "hedus" as a rural variant of "haedus" (child [goat]).
S:
See above about X sounding like CS.
Quintilian writes that a master of speech will not prolong the S sound too much, meaning people must have hissed too long for his taste or he wouldn't be whining about it.
A:
Marius Victorinus very directly writes that you make this sound with your mouth wide open and your tongue not touching the teeth. That's fairly precise.
In pretty much all modern Romance languages (languages directly derived from Latin), A is pronounced in the same way.
Even other languages influenced by Latin pronounce it in the same way (e.g. modern German) or used to (e.g. English until around the 15th century).
R:
There's probably not a lot of controversy around the general idea of this sound, but Marius Victorinus specifies it as a trilled R.
There are some sources that compare it to animal noises such as a dog growling or a cat purring, or use the R sound to represent these noises.
Further reading:
F.E. Lord: "The Roman Pronunciation of Latin" (link to Project Gutenberg).
W.S. Allen: "Vox Latina: A Guide to the Pronunciation of Classical Latin"
W.M. Lindsay: "The Latin Language: An Historical Account of Latin Sounds, Stems, and Flexions"
CORPVS GRAMMATICORVM LATINORVM "Late Latin Grammatical Sources: Full Text Search, Text Archive and Bibliography".
Note: These are not a lot of fun to read if you can't read Latin. Especially not the last one.
