Livius Andronicus, Naevius, and Plautus in the third century before our era show the language as first used for literary purposes, and with them the breach between the spoken and written tongues begins. So far as Livius Andronicus, the Father of Latin literature, is concerned, allowance should be made without doubt for his lack of poetic inspiration and skill, and for the fact that his principal work was a translation, but even making this allowance the crude character of his Latin is apparent, and it is very clear that literary Latin underwent a complete transformation between his time and that of Horace and Virgil. Now, the significant thing in this connection is the fact that this transformation was largely brought about under an external influence, which affected the Latin of the common people only indirectly and in small measure. Perhaps the circumstances in which literary Latin was placed have never been repeated in history. At the very outset it was brought under the sway of a highly developed literary tongue, and all the writers who subsequently used it earnestly strove to model it after Greek. Livius Andronicus, Ennius, Accius, and Pacuvius were all of Greek origin and familiar with Greek. They, as well as Plautus and Terence, translated and adapted Greek epics, tragedies, and comedies. Several of the early writers, like Accius and Lucilius, interested themselves in grammatical subjects, and did their best to introduce system and regularity into their literary medium. Now, Greek was a highly inflected, synthetical, regular, and logical medium of literary expression, and it was inevitable that these qualities should be introduced into Latin. But this influence affected the spoken language very little, as we have already noticed. Its effect upon the speech of the common people would be slight, because of the absence of the common school which does so much to-day to hold together the spoken and written languages.
The development then of preliterary Latin under the influence of this systematizing, synthetical influence gave rise to literary Latin, while its independent growth more nearly in accordance with its original genius produced colloquial Latin. Consequently, we are not surprised to find that the people's speech retained in a larger measure than literary Latin did those qualities which we noticed in preliterary Latin. Those characteristics are, in fact, to be expected in conversation. When a man sets down his thoughts on paper he expresses himself with care and with a certain reserve in his statements, and he usually has in mind exactly what he wants to say. But in speaking he is not under this constraint. He is likely to express himself in a tautological, careless, or even illogical fashion. He rarely thinks out to the end what he has in mind, but loosely adds clauses or sentences, as new ideas occur to him.
We have just been thinking mainly about the relation of words to one another in a sentence. In the treatment of individual words, written and spoken Latin developed along different lines. In English we make little distinction between the quantity of vowels, but in Latin of course a given vowel was either long or short, and literary tradition became so fixed in this matter that the professional poets of the Augustan age do not tolerate any deviation from it. There are indications, however, that the common people did not observe the rules of quantity in their integrity. We can readily understand why that may have been the case. The comparative carelessness, which is characteristic of conversation, affects our pronunciation of words. When there is a stress accent, as there was in Latin, this is especially liable to be the case. We know in English how much the unaccented syllables suffer in a long word like "laboratory." In Latin the long unaccented vowels and the final syllable, which was never protected by the accent, were peculiarly likely to lose their full value. As a result, in conversational Latin certain final consonants tended to drop away, and probably the long vowel following a short one was regularly shortened when the accent fell on the short syllable, or on the syllable which followed the long one. Some scholars go so far as to maintain that in course of time all distinction in quantity in the unaccented vowels was lost in popular Latin. Sometimes the influence of the accent led to the excision of the vowel in the syllable which followed it. Probus, a grammarian of the fourth century of our era, in what we might call a "Guide to Good Usage"[20] or "One Hundred Words Mispronounced," warns his readers against masclus and anglus for masculus and angulus. This is the same popular tendency which we see illustrated in "lab'ratory."
The quality of vowels as well as their quantity changed. The obscuring of certain vowel sounds in ordinary or careless conversation in this country in such words as "Latun" and "Amurican" is a phenomenon which is familiar enough. In fact a large number of our vowel sounds seem to have degenerated into a grunt. Latin was affected in a somewhat similar way, although not to the same extent as present-day English. Both the ancient grammarians in their warnings and the Romance languages bear evidence to this effect.
We noticed above that the final consonant was exposed to danger by the fact that the syllable containing it was never protected by the accent. It is also true that there was a tendency to do away with any difficult combination of consonants. We recall in English the current pronunciations, "February," and "Calwell" for Caldwell. The average Roman in the same way was inclined to follow the line of least resistance. Sometimes, as in the two English examples just given, he avoided a difficult combination of consonants by dropping one of them. This method he followed in saying santus for sanctus, and scriserunt for scripserunt, just as in vulgar English one now and then hears "slep" and "kep" for the more difficult "slept" and "kept." Sometimes he lightened the pronunciation by metathesis, as he did when he pronounced interpretor as interpertor. A third device was to insert a vowel, as illiterate English-speaking people do in the pronunciations "ellum" and "Henery." In this way, for instance, the Roman avoided the difficult combinations-mn- and-chn-by saying mina and techina for the historically correct mna and techna. Another method of surmounting the difficulty was to assimilate one of the two consonants to the other. This is a favorite practice of the shop-girl, over which the newspapers make merry in their phonetical reproductions of supposed conversations heard from behind the counter. Adopting the same easy way of speaking, the uneducated Roman sometimes said isse for ipse, and scritus for scriptus. To pass to another point of difference, the laws determining the incidence of the accent were very firmly established in literary Latin. The accent must fall on the penult, if it was long, otherwise on the antepenult of the word. But in popular Latin there were certain classes of words in whose case these principles were not observed.
The very nature of the accent probably differed in the two forms of speech. In preliterary Latin the stress was undoubtedly a marked feature of the accent, and this continued to be the case in the popular speech throughout the entire history of the language, but, as I have tried to prove in another paper,[21] in formal Latin the stress became very slight, and the pitch grew to be the characteristic feature of the accent. Consequently, when Virgil read a passage of the AEneid to Augustus and Livia the effect on the ear of the comparatively unstressed language, with the rhythmical rise and fall of the pitch, would have been very different from that made by the conversation of the average man, with the accented syllables more clearly marked by a stress.