A Short Historical Latin Grammar: Chapter I: Alphabet

The Greeks of Chalcis in Euboea had at an early time founded colonies at Cumae and other spots on the coast of Campania –colonies which took a leading place among those mercantile centres of Southern Italy which played so great a part in Italian civilisation. It was from these Chalcidic neighbours of theirs that the Latins first learnt the art of writing.

The connexion of the Latin with the Greek Alphabet is seen plainly enough, even when we look at the later forms of the two, the forms which we are in the habit of using in editions of the classical authors:

But when we go farther back and compare the forms of the letters on the earliest Latin inscriptions with those on the earliest Euboean inscriptions, we see that the two alphabets are one and the same:

the fact being that the Latins took over bodily from the Greeks of Campania the written signs by which these Greeks expressed the various words and sounds of their language. Now it is clear that the sounds of the Latin language were not in every case the same as the sounds formed by Greek lips, so that a Greek alphabet could not be a perfectly suitable means of expressing Latin words and syllables; and, in fact, we find that it took a great number of years before the borrowed alphabet was altered and improved to the satisfaction of the borrowers.

One of the first alterations made was the disuse of the signs for the three Greek aspirates, θ, φ, χ. These were written in the Chalcidic alphabet Θ (⊕), Φ (ↀ), Ψ (like -V- but trident-shaped) [1], and expressed sounds unknown to the Latins, (1) the sound of t followed by an h-sound (something like our th in "ant-heap"), (2) the sound of p followed by an h-sound (cf. our "up-hill"), (3) the sound of k followed by an h-sound (cf. our "ink-horn"). Since these sounds never occurred in any Latin word, the three signs were never used in writing Latin, and dropped out of the Latin alphabet [2]. This then was the first departure of the borrowed alphabet from its original.

[1]: The use of X for x, Ψ for ch, and the order X, Φ, Ψ are features of what are called the "Western" Greek Alphabets, as opposed to the Ionian, &c.

[2]: Though they were not used to indicate sounds, they were put to another function, that of numerical signs. Θ was employed as the sign of 100, and became in time adapted in form to the initial of centum and written C; Φ was made the sign of 1000, and came to be written ↀ, ∞, and finally M, like the initial of mille; while D, the half of Φ, became the sign of 500, just as V, 5, is the upper half of X, 10; Ψ indicated 50 and came to be written ⊥ and finally L.

Again the k- and g-sounds of the Greeks had not the same sharp distinction as the corresponding sounds had from Roman lips, so that when a Greek pronounced his letter κ, he often seemed to a Roman to be pronouncing a g-sound [3] and not a k-sound. This led to the use of the third letter of the borrowed alphabet, written sometimes circularly, C, sometimes with an angle, <, in a promiscuous fashion, now for a g-sound, now for a k-sound, so that the written signs AC or A< might express either the syllable ag or the syllable ac. This encroachment of the sign C or < on the sphere of the sign K was aided by the custom of writing K in two parts I<, with the upright part I separate from the angular part <, a custom which led to K being regarded as an awkward double symbol, of which < (apparently the right-hand half of the symbol) was a short and convenient expression; and the result of all this was that the letter K practically went out of use, C or < being retained in the double capacity of a k-sign and a g-sign [4].

[3]: Greek κόμμι, gum, was written by the Romans gummi; Greek κυβερνῶ is in Roman spelling guberno.

[4]: Thus on a very old inscription, called the Dvenos inscription, we have VIR<O (virgo) and <OSMIS (cōmis) side by side.

Another double sign was reduced in the same way, namely FH. This FH was a letter, or rather a letter-group, which expressed in early Greek words a sound that came nearest to the sound of early Latin f (ch. ii. § 6 n.), and was at first used by the Latins as a sign for this sound. In time it too was simplified by the dropping of one of its parts, and was written without the H as a single sign, F. Thus fec- of fecit would be first written FHEK (or FHE<), then FEK. The sign for f became in this way identical with the sixth letter of the borrowed alphabet, and opened the way to a further change. For this sixth letter, the Digamma, was used in the Chalcidic alphabet to express the w-sound heard in the beginning of such early Greek words as Ϝοῖνος, the same as in the Latin vinum or the English wine. After a while the sound was dropt by the Greeks and became quite unfamiliar to them, so that we hear of Greek visitors to Italy finding the greatest difficulty in pronouncing Latin words like vinum, veni. The nearest they could come to the proper pronunciation was the use of the vowel u instead of the consonantal w-sound, u-i-num, u-e-ni. It was perhaps through Greek influence that the Romans came to use the sign of the vowel u, V, for this w-consonant as well as for the vowel, so that for a time V and F were employed promiscuously to express the w-sound, in precisely the same way as we found K and C or < to have been used promiscuously for the k-sound. The result in both cases was the same. The sign that had encroached on the province of the other sign ousted its rival; and for the future F held exclusively the function of representing the f-sound, while the w-sound was expressed by V, the same sign as did duty for the vowel u.

Of the first six letters of the borrowed alphabet two have thus in course of time come to express quite different sounds from the sounds which they expressed in the original alphabet. C or <, which in the original expressed the g-sound, has come to express the k-sound; F, originally the sign of the w-sound, has become the sign of the f-sound. The seventh letter was likewise affected. Greek Z was the letter for the soft s-sound ("voiced" s) of our "as," etc., a sound in use among the Latins at the time when they borrowed their alphabet. But as centuries went on, this soft s-sound assumed in Roman lips the sound of r (ch. X. §19), so that the Gen. Plur. termination of the First Declension (ch. III. §4), at first pronounced -asom with soft s, and written AZOM, came to be pronounced and written -arom (later -arum); and the sign Z fell out of use in writing. This is a change in the Roman alphabet to which we can give a date. We are told that it was Appius Claudius Caecus, the famous censor of 312 B.C., who was the author of this reform; and Roman tradition preserves a curious story that one of his arguments against the letter was, that the sound which it properly expressed, the soft s-sound, gave the mouth the appearance of the grinning teeth of a skull. In its place in the alphabet he put a new letter, G, which was coming into use as a symbol of the g-sound, and was employed, for example, by Sp. Carvilius Ruga (consul in 293 B.C.) in the writing of his name (RVGA, instead of the older RVCA), just as L. Papirius Crassus, dictator in 339 B.C., was the first of his family to write his name PAPIR- instead of PAPIZ-. This new letter G was nothing but the letter C slightly altered by the addition of a small stroke to show that the g-sound and not the k-sound was meant. The expression of the g-sound had thus a curious history in Latin. First the symbol C was used; then this symbol came to be used also for the k-sound; finally this later use ousted the older use altogether, and a new symbol was devised for the g-sound.

The alphabet of Appius Claudius remained the official Roman alphabet of twenty-one letters: A B C D E F G H I K L M N O P Q R S T V X.

The letters Y and Z are sometimes added, but are not strictly speaking Roman letters. They are nothing but the Greek Upsilon and Zeta, which came to be used in Cicero's time in writing Greek loan-words, e.g. cymba, zona, gaza, where previously u and s (ss) had been used, e.g. cumba, sona (Plaut.) (so classical buxus, cupressus, guberno, massa, &c.), just as we use the Spanish letter -ñ- in writing the Spanish loan-word "cañón," or the French letter ç in writing the French loan-word "façade." They were never used in writing Latin words, except occasionally in a few words which were wrongly believed to be Greek loan-words, e.g. lacryma, inclytus; much as we write "rhyme" instead of "rime," owing to an idea that it represents the Greek ῥυθμός. And though the Latin writers on Grammar usually include them in the Roman alphabet, they are careful to make a distinction between them and the other letters. Similarly th, ph, ch, rh were at the end of the Republic used for the more exact expression of Greek loan-words like Corinthus, Philippus, Achilles, Rhodus, words which were earlier written with t, p, c, r (cf. tus, paenula, a coat, calx, lime, from Gk. θύος, φαινόλης, χάλιξ), and found their way even into genuine Latin words like pulcer, from polio, wrongly referred to Gk. πολύχροος and written pulcher (cf. ch. II. §6).

So conservative were the Romans that the old forms and significations of the letters were often retained in isolated cases. C retained its old signification of the g-sound in C., Cn., the contractions for the proper names Gāius (older Gāvios, connected with gaudeo, gāvisus) and Gnaeus (older Gnaivos, connected with (g)nascor, (g)natus). K was retained in the proper name Kaeso, written shortly K.; and there were old-fashioned people even in Imperial times who in their correspondence always spelt karissime with k, not c (cf. the spellings Kalendae, interkalaris, kalumnia, and other words where the guttural precedes a). The old five-stroked form of M was retained as symbol of the proper name Mānius, and came to be written M', while the ordinary form of the letter was used as symbol of Marcus. Legal phraseology was especially retentive of old forms; thus the -e of jure dicundo, solvendo aere alieno is a relic of the very early usage of denoting the ei-diphthong by E as in the early Greek writing.

Some changes of the original form of writing were permanent, others lasted only for a time. To the first class belongs the practice of writing double a consonant which was pronounced double (ch. ii. § 9), a practice introduced by the poet Ennius. Thus buca was the old spelling of bucca. To the second belong such changes as: (1) writing double a long vowel, (a, e, u, and o?) [5], e.g. PAASTORES, a practice introduced by the poet Accius, but soon dropped; (2) writing EI for long i. This practice came in about the middle of the second century B.C., when the diphthong ei (ch. x. § 11) had come to be pronounced like long i, and remained till the beginning of the Empire; (3) indicating a long vowel (a, e, o, u) by a stroke above, called the apex, e.g. PĀSTORES (first two centuries of the Empire); (4) indicating a long i by the tall form of the letter, e.g. MĪLIA. This tall form was also used to denote the y-sound and also for initial i; (5) indicating a doubled consonant by a curve above, called the sicilicus, e.g. OSA (time of Augustus). Many, or most, of these practices were borrowed from Oscan [6] orthography. Another change introduced, but without much success, by the poet Accius, was borrowed from Greek, viz. the use of g for n before g, c, e.g. aggulus, agcora (cf. Gk. ἄγκυρα). Equally short-lived were the additions to the Roman alphabet by the Emperor Claudius, viz. Ⱶ for the ü-sound of optimus, &c. (ch. ii. § 1); Ↄ for ps; ⅎ for v (Tac. Ann. xi. 14). The byforms of E and F, which found favour with the uneducated classes, viz. II (often confused on inscriptions with I) and ||, are interesting, because they must have originated from writing on wood, where the scratching of horizontal lines was not easy to read.

[5]: II indicated the double y-sound of words like aio, Maia, written by Cicero aiio, Maiia (ch. ii. § 3).

[6]: Oscan is the name given to the language of the Samnites, who lived in Samnium and Campania. So that Campania, the birthplace of the Roman alphabet (§ 1), continued to influence Roman writing for many centuries.