“You know, Caesar would have said ‘weni, widi, wici’,” your annoying over-educated friend tells you as you say the famous lines after a night out (or perhapas you fancied yourself clever and said ‘vidi, vici, veni’ you dirty-minded scholar, you); maybe you were the friend offering the correction. In either case, the v-pronouncing culprit may not have been far off the mark.
Let me be clear about one thing first though. In Classical Latin, V (or as I’ll be writing it henceforth, u) could definitely represent [w], and it definitely did so in Caesar’s famous lines. But whether or not Caesar spoke Classical Latin when he was addressing his troops, and whether or not the exact sound he would have articulated was [w], is worth of some consideration.
First, a quick lesson in phonology. There are two aspects that are important for you to consider: i) the mechanisms by which we produce sounds, and ii) the way we ascribe meaning to sound
To visualize the production of sounds, it is helpful to imagine the vocal tract as a long tube. You push air from your lungs through this tube, and you have a mechanism called your ‘vocal folds’ (more commonly known as a vocal cords) that causes this air to vibrate, thus creating sound waves. Once the air has passed the vocal folds, it can be shaped by varying degrees of closure at any point along the tube from the larynx to the lips. A consonant is marked by a complete closure at one point along the vocal tract. If you say the sounds ‘aba’, ‘ada’, ‘aga’, you can feel closures in different areas of the mouth – the lips, the alveolar ridge (that ridge just behind your teeth), and the velum (the back portion of the roof of your mouth).
The important point here is that the difference between [u] and [w] is simply the degree of closure of the articulators. In other words, [w] can be said to be the ‘consonantal’ version of [u] because there is enough constriction at the articulators for it to be considered a consonant.
By comparison, the vowels you pronounced were relatively open and the air flowed freely through your mouth. Vowels are shaped by the position of your tongue in your mouth. If you say “eat” “ate” “at” “ah” “ought” “oops”, you will feel your tongue move from high to low, and from front to back. We will be concerned with what are called ‘high back’ vowels (oops), and velar sounds (aga). One further point that you should notice is that when you say ‘oops’ you push your lips forward. You can compare this to the way you say ‘eat’ with your lips stretched. We call this contrast ’rounded’ versus ‘unrounded’.
There is also a class of sounds called ‘semivowels’. These are consonants that are just opened enough to disturb the airflow passing through your lips. If you say “eye”, your tongue doesn’t quite touch the roof of your mouth (be careful though – if you exaggerate this too much, you are likely to produce an ‘offglide’: ai-ya). [w] is also a semi-vowel. A very cool semi-vowel. It requires two points of articulation – one at the velum, and one at the lips.
If we look a bit closer at this semi-vowel, we’ll notice that it has some similarities to the vowel [u]
We also have to classify how we understand sounds. Sure it is possible to make any number of sounds, but very few sounds of the range of sounds the human vocal tract can produce actually appear in a given language. We call the sounds of a given language its ‘phonemic inventory’.
A phoneme is a meaningful sound. To determine whether a sound is meaningful or not, Linguists attempt to identify minimal pairs. For example, English contrasts /b/ and /p/: ‘bin’ and ‘pin’ are words that differ only in the pronunciation of the first sounds. Furthermore, /b/ an /p/ only differ in on aspect: while they are both pronounced with the same movement of the tongue and lips, /b/ involves vocal fold vibration, while /p/ does not. You can tell this by holding your hand against your throat as you pronounce ‘aba’ and ‘apa’. During the former, you will feel vibration throughout, while during the latter, you will feel a pause in vibration as you pronounce the ‘p’.
There is a further distinction some languages make: aspiration. Aspiration is a puff of air released after the pronunciation of a stop. The /p/ in ‘spin’ is unaspirated, while the /p/ in ‘pin’ is aspirated. If you hold your hand in front of your mouth while you pronounce these two sounds, you will feel a bigger puff of air when you pronounce the /p/ in pin. Even though the manner of articulating these two sounds is different, we consider them to be the same sound, or phoneme. Phonemes are sort of ‘headings’ under which we group similar sounds so that individual variation does not hinder our ability to understand sounds. Since aspirated /p/ and unasiprated /p/ are not contrastive in English, we call aspirated /p/ an ‘allophone’ of a ‘phoneme’ /p/..
What I want to discuss is whether the Latin grapheme (letter) represented two phonemes (/w/ and /u/) or one (/u/). We can determine whether the latter was the case by seeing whether /w/ and /u/ occur in ‘complementary distribution’ (I.e. If [w] only occurs at the beginnings of words and syllables, and if [u] is always in the middle of a syllable, then they cannot possibly contrast with each other. Furthermore, they are phonetically similar enough to be considered consonantal and vocalic variations of the same sound).
Since this post is already quite long, I’ll leave you in what must be thrilling suspense and answer the question with my own interpretation in a few days.
1. It is important to note that there is a difference between the symbols that we use to represent sounds and the sounds we produce. In many cases, multiple sounds are represented by the same symbol. In Latin, V could represent [u], [u:] (long [u]), and [w] (and also [β], but more on that later).