cters
in Plautus use better Latin than the freedmen in Petronius. The
illiterate freedmen in Petronius speak very differently from the freemen
in his story. Sometimes a particular occupation materially affects the
speech of those who pursue it. All of us know something of the linguistic
eccentricities of the London cabman, the Parisian thief, or the American
hobo. This particular influence cannot be estimated so well for Latin
because we lack sufficient material, but some progress has been made in
detecting the peculiarities of Latin of the nursery, the camp, and the
sea.

Of course a spoken language is never uniform throughout a given area.
Dialectal differences are sure to develop. A man from Indiana and another
from Maine will be sure to notice each other's peculiarities. Even the
railway, the newspaper, and the public school will never entirely
obliterate the old differences or prevent new ones from springing up.
Without these agencies which do so much to promote uniformity to-day,
Italy and the rest of the Empire must have shown greater dialectal
differences than we observe in American English or in British English
even.

For the sake of bringing out clearly some of the points of difference
between vulgar and formal Latin we have used certain illustrations, like
_caballus_, where the two forms of speech were radically opposed to each
other, but of course they did not constitute two different languages, and
that which they had in common was far greater than the element peculiar to
each, or, to put it in another way, they in large measure overlapped each
other. Perhaps we are in a position now to characterize colloquial Latin
and to define it as the language which was used in conversation throughout
the Empire with the innumerable variations which time and place gave it,
which in its most highly refined form, as spoken in literary circles at
Rome in the classical period, approached indefinitely near its ideal,
literary Latin, which in its most unconventional phase was the rude speech
of the rabble, or the "sermo inconditus" of the ancients. The facts which
have just been mentioned may be illustrated by the accompanying diagrams.

[Illustration: Fig. I]

[Illustration: Fig. II]

[Illustration: Fig. III]

[Illustration: Fig. IV]

In Fig. I the heavy-lined ellipse represents the formal diction of Cicero,
the dotted line ellipse his conversational vocabulary. They overlap each
other through a great part of their