ROMAN ORTHOEPY:
A PLEA
FOR THE RESTORATION
OF THE
TRUE SYSTEM
OF
LATIN PRONUNCIATION.
BY
JOHN F. RICHARDSON,
MSOP1E8B0 OF TUE LATIN LANGUAGE AND LITERATURE IN THI
UNIVERSITY OF ROCHESTER.
NEW  YORK:
S IIE LDON It: CO Mp ANY,
115 NASSAU STREET.
1859.




Entered, according to Act of Congress, in the year 1859, by
SHELDON & COMPANY,
In the Clerk's Office of the District Court for the Southern District of
New York
ELECTROTYPF) BY                               PRINTED BY
THOMAS B. SMITHI & SON                     PUDNEY & RUSSELL,
82 & 84 Beekmai-str-et.                       79 John-street.




PRE FACE.
THE author of this little Manual was, from 1838
to 1850, professor of the Latin language and literature in MIadison University. During the latter part
of that period, he became dissatisfied with the
English mode of pronouncing Latin, and convinced
that a change, at least so far as to. substitute the
German for the English vowel sounds, would be,
intrinsically, a great improvement. But the fancied
inconveniences of a change-inconveniences which
experience has taught him were greatly exaggerated-together with a strong repugnance to the
notoriety of singularity, presented obstacles which
his conviction of its inherent advantages was then
unable to surmount.
Immediately after the organization of this University, in the autumn of 1850, his attention was
again called to this question, and he was induced to
institute a new and still more careful examination
of the whole subject of Latin pronunciation. The
result of that examination was a thorough conviction, not only of the possibility, but of the decided




iv                 PREFACE.
expealency, of restoring the Roman method of
pronouncing the language. He submitted to the
Faculty of the University the most important considerations which had led him to that conclusion,
and obtained their approval and sanction of the
proposed change. But again his courage was unequal to a total reform. By way of compromise,
and as a tentative, he adopted the Roman sounds
of the vowels and diphthongs, and of the consonants
J, s, and T, but shrank from the greatest and, perhaps, the most important change to the Roman
sounds of c, G, and QV. Where the English sounds
of these letters differed from the Roman, he was
in the habit of suggesting to his higher classes, for
the purpose of explaining some point in etymology,
prosody or comparative philology, &c., the Roman
pronunciation of the word in question; and the
almost invariable response which he received fiom
his pupils was the inquiry, " Why then do you not
pronounce it so?" This question it was easy to
evade but not to answer satisfactorily, either to his
more intelligent pupils or to himself.
In the latter part of 1851, there appeared in the
New York Recorder, a review of Professor Haldeman's able treatise on " Latin Pronunciation," fiom
the- pen of the late lamented Robert Kelly, LL. D.,
of that city, one of the founders and most influenial managers of this University, and a man of the




PREFACE.                    V
highest style of classical scholarship, earnestly and
ably advocating the general adoption, in our literary institutions, of the Roman method of pronouncing Latin. Delighted with the suggestions
of this reviewer, which were in such full accordance
with my own convictions, and greatly strengthened
in the soundness of those conclusions by his cogent
reasonings and illustrations, I determined to introduce the true system entire to my own classes without waiting any longer for the cooperation of other
institutions. In this position the author has had no
occasion for self-condemnation. He has met, as he
expected, with. some opposition and a little reproach; but he has been more than satisfied with
the working of his system, and he has been sustained and cheered by the cordial sympathy and
cooperation of his pupils.
The following Manual has been prepared in compliance with the requests of several literary friends
and classical teachers in other institutions, who have
called on the author for a description of the Roman
system, and his reasons for using it.
It is proper to say that, as the work is designed
for the use of learners as well as teachers of
Latin, the author has, in many cases, made the
arguments and illustrations more full than he would
have deemed necessary or appropriate had he been'riting exclusively for the latter ciass of readers




PREFACE.
In the preparation of the Manual, the author
would acknowledge his indebtedness to the work of
Charles Kraitsir, M.D., entitled " Glossology," published in 1852, for some important facts and proofs
in relation to the early English-Latin pronunciation,
and for several of the extracts fiom  the Latin
grammarians which are found in the appendix of
this work.
With these explanations, the Manual is now submitted to the public, in the hope that it may contribute to the general abandonment of the diverse,
conflicting, unscientific and unscholarly methods
now in use, and the substitution of that one which
is demonstrably correct and alone correct.




ROMAN  ORTHOEPY.
IT is a very common opinion that there are but
two widely prevalent systems of Latin pronunciation, the English and the Continental, and that
these differ only in the sounds they respectively
assign to the vowels.  A large majority of the
modern Latin  grammars and elementary works
published in our language for the use of English
and American students, have given expression to
this opinion, and many of them have laid down
rules for the sounds of the vowels in each of the
two systems; and then, assuming that there is but
one common mode of sounding the consonants and
diphthongs, they simply offer the student his choice
between these two vowel systems as his only alternative in respect of Latin pronunciation.  They
either give him or leave him to understand that,
according as he connects with these common sounds,he one or the other of the two vowel systems, he




8              ROMAN ORTHOEPY.
will have either the English or the Continental system, pure and entire.
Now what are the facts in respect to the different modes of pronouncing Latin?  In the first
place, there are two distinct and very different
English systems of Latin pronunciation, the earlier
and the later. In the second place, it Is an entire
misnomer to speak of the " Continental Method" of
pronouncing Latin. There is, in fact, no common
Continental system, but there are several Continental systems of Latin pronunciation, e. g., German, Italian, French, Spanish. These four agree,
to be sure, substantially in regard to the vowels;
but, in other important points, they differ decidedly
both from the English and from each other, most
of the diphthongs and some of the most important
consonants being sounded differently in all five.
The idea, therefore, that he who combines the German vowel sounds with the English diphthongal and
consonant sounds has the Continental system or
any Continental system of Latin pronunciation, is
simply absurd.
Of the six different systems of Latin pronunciation, then, prevailing in western Europe and our
own country, five are strictly national. Their differences find at once an origin and an explanation




ROMAN ORTHOEPY.                 9
in the fact that the scholars of each nation have
followed, in their pronunciation of the Latin, the
analogies of their own vernacular.  In this way,
while making sure of mutual disagreement, all
have departed more or less from the true Roman
method, and the whole subject has been involved
in  uncertainty  and  confusion.  Meanwhile the
pseudo-Continental system, destitute alike of historical dignity and scientific accuracy, and lacking
even the poor support of national prejudice and
pride, is powerless to mediate and compose these
differences. Although it undoubtedly avoids some
of the grossest absurdities peculiar to the English
system, it lacks the elements which commmad respect, and can never establish a claim to universal
adoption and use.
Under these circumstances, it is not strange that
the expediency of attempting a restoration of the
Roman pronunciation should, for many years, have
engaged the attention of prominent Latin scholars
in all countries. So numerous, so constant, and so
serious are the practical evils connected with the
want of a uniform standard, and still more, with the
falsity of the methods in actual use, that the question of restoring the true pronunciation is continually forcing itself upon the mind of the earnest
1*




10                ROMAN  ORTHOEPY.
Latinist as one of vital importance to the dignity,
the value and the progress of Latin philology.*
These evils manifest themselves most fully where
* Justus Lipsius, in his " DIALOGUS DE RECTA PRONUNCIATIONE LING-UAE LATINAE," expresses himself upon the false
pronunciation of the consonant c, in this forcible and felicitous
manner:-" PUDET NON TAM ERRORiS QVAM PERTINACIAE, QVIA
CORRIPI PATIUNTUR AT NON CORRIGI, ET TENENT OMNES QVOD
DEFENDAT NEMO. ITALI, HISPANI, GERMANI, GALLI, BRITANNI,
IN HOC PECCATO; A QVA GENTE INITIUM EMENDANDI?  AUDEAT
ENIM UNA ALIQYA ET OMNES AUDIENT."
The Rt. Hon. W. E. Gladstone, D.C.L., M. P. for the University of Oxford, at the close of the Prolegomena to Homer and
the Homeric Age, employs the following language:
" Finally, though sharing the dissatisfaction of others at the
established preference given among us to the Latin names of
deities originally Greek, and at some part of our orthography for
Greek names, I have thought it best to adhere in general to the
common custom  and only to deviate from it where a special
object was in view. I fear that diversity and even confusion are
more likely to arise, than any benefit, from efforts at reform
made by individuals, and without the advantage either of authority or of a clear principle, as a groundwork for general consent. I am here disposed to say,'ovK gayaOov iro2vicotpaviy;'
and again with Wordsworth,
"'Me this unchartered freedom tires;'
yet I should gladly see the day when, under-the authority of
scholars, and especially of those who bear rule in places of education, improvement might be effected not only in the points
above mentioned, but in our solitary and barbarous method of
pronouncing both the Greek and the Latin language. In this
one respect the European world may still with justice describe
the English at least as the'PENITUS TOTO DIVISOS ORBE BRITANNOS.'"




ROMAN ORTHOEPY.               11
the present English system of Latin pronunciation
obtains-a system which deviates most widely from
the Roman, and is most decidedly at war with the
structure and genius of the Latin language; and
here especially, the restoration of the vernacular
pronunciation becomes a matter of urgent practical
moment.
I have spoken of the present English system of
Latin pronunciation in distinction from an earlier
and much purer method which formerly prevailed
in England, and which accorded substantially with
the Roman.  Although the English is now the
most irregular and confused of all the alphabets of
Europe, yet no modern tongue entered upon its
career as a written language, with better prospects
of securing a harmonious system of orthoepy than
the Anglo-Saxon. To this language the Roman alphabet was very skillfully adapted. All the sounds
which the two languages had in common were
represented by characters taken fiom the Roman
alphabet, while those which were wanting in the
Latin were indicated by characters newly invented,
or borrowed from' other alphabets. C and G, c. g.,
which are now so frequently employed respectively
as lingual and dental spirants (the former having
the sound of s sharp, and the latter, that of J, a




12             ROMA1N ORTHOEPY.
sound unknown alike to the Anglo-Saxon and the
Latin), were in all Saxon, as well as in all Latin
words, employed as guttural mutes, representing
the hard and soft K-spunds.  The unequaled irregularities of English ortho6py are attributable,
not to the lack of sagacity and good judgment on
the part of our Anglo-Saxon ancestors, but to the
undue influence of the Norman conquerors of England over its language as well as over its laws and
customs. But a substantially correct Latin pronunciation had been established in England long before
the Norman conquest; and while, after that event,
the higher classes in that country, in the spirit of
undiscriminating subserviency to the superior civilization and refinement of their conquerors, imitated
alike, and with equal readiness, their errors and
their virtues, and thus sacrificed some of the best
peculiarities of their own language to the irregularities of the French, this pure Latin pronunciation
maintained its ground until a comparatively recent
period.  Not until about the middle of the last
century was the process of its corruption fully completed. Nor, meanwhile, did it escape the sharp
censure and earnest protests of many noble English
scholars, who attempted, at different periods, either
to arrest the progress of this corruption, or to re



ROMAN ORTHOEPY.                13
form the pronunciation when corrupted. Thle Latin
grammars formerly used in England contain many
cautions against falling into the vulgar errors. One
of these, first published in the reign of Henry the
Eighth, lays the following injunction upon the Latin
teacher:
"ANTE OMNIA DETERRENDI SUNT PUERI AB IIS
VITIIS QVAE NOSTRO VULGO PAENE PROPRIA ESSE
VIDENTUR."
At this period of English history very great attention was paid to the study of the Latin. The
elementary works employed in the schools abound
in warnings, not only against grave errors of pronunciation, but also against minor faults of articulation. The teacher is urged to see to it that the
pupil shall not merely preserve the pure sounds of
the letters but that he shall utter those sounds ORE
ROTUNDO. Philology, it is true, was not then understood or cultivated as a science, and no special
value was set on the scientific importance of preserving the sounds of the language unchanged; but
the superiority of euphony to cacophony, of harmony to discord, was fully appreciated, and thus on
principles of good taste, though not of philological
science, the scholars of that and of the two succeeding centuries sought to preserve unimpaired the




14               ROMAN ORTHOEPY.
purity and beauty of Latin orthoepy.  The most
distinguished men of those times took a lively interest in the preservation of this system.  In a
letter addressed by Cardinal Wolsey to the masters
of his school at Ipswich, he exhorts the teachers to
use great diligence in forming the tender lips of
their pupils to an elegant and correct Latin pronunciation:
" QVORUMI* OS TENERUUM  FORMAARE PRAECIPUA
CURA VOBIS SIT, UTPOTE QVI ET APERTISSIMA  ET
ELEGANTISSIMA  VOCIS PRONUNCIATIONE, TRADITA
ELEMENTA PROFERANT."
In like manner, the learned men of the seventeenth  and  eighteenth  centuries  endeavored  to
maintain the pure sounds of the language, and remonstrated against the growing corruption of Latin
pronunciation, urging, among other considerations,
that unless this process of deterioration was arrested, the language would no longer serve as a
medium  of oral communication between English
* Since, in the classical period of the Latin language, the
character v employed uniformly after Q, became in that position,
entirely silent, and.was thus merely a part of a compound character representing a single consonant sound, viz., that of K, and
the modern rounded character u, if introduced at all into Latin,
should be employed only for the v VOCALIS of the Romans, I
have thought it proper invariably to retain the v after Q.




ROMAN ORTHOEPY.                15
and Continental scholars. Prominent among those
who thus censured the increasing barbarity of the
English mode of pronouncing Latin, stands the
majestic and classical Milton. In the letter which
he addressed to Mr. Hartlib on the subject of education he lays down among the first rules for the exercises of the pupils in his model school the following
direction in respect to their instruction in Latin:
"Their speech is to be fashioned to a distinct
and clear pronunciation, as near as possible to the
Italian, especially in vowels. For we Englishmen,
being far northerly, do not open our mouths in the
cold air wide enough to grace a southern tongue,
but are observed by all other nations to speak exceeding close and inward; so that to smatter Latin
with an English tongue, is as ill a hearing as LawFrench."
The closeness of which he here complains in the
English method of pronouncing Latin, was most
apparent, undoubtedly, in the sound of the vowels
to which he expressly alludes, and particularly in
the sound of the open or back vowels; and it is
wortiy of notice how fully this precept and remonstrance harmonize with the " APERTISSIMA vOCIS
PRONUNCIATIO," recommended by Cardinal Wolsey
to the teachers of his own school.




16             ROMAN ORTHOEPY.
Robert Ainsworth, Thesaur. Ling. Lat., Lond.,
1746, in a very valuable preface on Latin pronunciation, which has been improperly, if not dishonestly omitted in the American school editions, says,
" With much reluctance, I remark that foreigners
hold us little better than barbarians in many parts
of pronunciation."  He particularly censures the
neglect of the quantity of vowels, and the "depraved sound" of c and G before AE, OE, E, I. " The
irregular and uncertain pronunciation of these letters
proves often a great discouragement to those who
desire to learn our tongue; and this, together with
our different sounds of the vowels, makes our Latin,
though much purer generally (meaning, in structure,) than theirs, almost as unintelligible as our
English.... This I leave to the consideration
and redress of the learned schoolmasters of this
kingdom, as well deserving it..... To say NASHIO,
instead of NATIO, T as in TILL, is absurd, if we did
not submit to a barbarous prescription."
Philipps, a preceptor to some of the princes of
the royal family, a man of superior classical attainments, and very familiar also with many modern
tongues, in his " Method of Teaching Languages,"
published in 1750, complains of the very faulty and
unpleasant manner in which Englishmen pronounce




ROMAN ORTHOEPY.              17
Latin. Hea describes his mode of teaching this language to a youth placed in his charge, and tells us
that he took " special care" to wean him from his
awkward manner of pronouncing.
"He gave me a great deal of trouble for some
months on this head; so that I had much ado to
persuade him to open his mouth; for he pronounced
the vowels very badly, especially the A and E; for
instead of Aio, he pronounced EMO; and when he
pronounced Eiro, to buy, he called it Iro; and instead of IMO, yes, he said AIMo."  He speaks of
this inelegant mode of pronouncing Latin as of a
thing common, yet not universal. " Many gentlemen in England," is his pertinent observation,
" still speak Latin like men, ORE ROTUNDO."
Dr. Foster (Essay on Accent and Quantity) complains of the "violence done to the quantity of the
ancient languages by the English pronunciation, and
that, though an attachment be professed to it, yet
this very quantity they do all (most of them without knowing it,) most grossly corrupt."
Mitford (Inquiry into the Principles of Harmony
in Language, published at the close of the eighteenth
century,) points out the absurdity of introducing
into Latin the eccentric pronunciation of the English; he represents its incompatibility with the




18              ROMAN ORTHOEPY.
true quantity of syllables, and proposes the restoration of the ancient sounds of the vowels, as in
Italian.
But in spite of these warnings, remonstrances,
and regrets of prominent English scholars, this corruption of Latin pronunciation went on, and was
finally and fully effected about the middle of the
last century. Since that time, this method of pronouncing Latin has been reduced to a grave system,
and is formally inculcated in nearly all the manuals
of Latin instruction prepared in our language for
the use of English and American students. No
one pretends to claim for it any scientific or historical basis. It is justified solely on grounds of prescription, convenience and nationality, and yet it
is enforced by precept and example in a very large
majority of the literary institutions of this country
and of England.
The young student opens his Latin grammar, or
comes to his Latin teacher, and is informed, at the
outset, that, the Roman system of Latin pronunciation being in a great measure lost, each of the
different modern nations follows, in its pronunciation of this language, the analogies of its own.
This is all the information that is usually afforded
him  on the subject.  Occasionally, within a few




ROMAN ORTHOEPY.               19
years past, there has appeared some elementary
work which has expressed a preference for the socalled "Continental Method," and, that the student
may take his choice between the two systems, the
author proceeds to give the system of vowel sounds
supposed to prevail on the Continent, and especially
in Germany. But not one of these works, so far as
my observation extends, has given the real distinction between the long and short sounds of the German vowels; and thus a system of vowel sounds
which nowhere exists on the Continent of Europe,
is to be combined with the English  consonant
sounds to form what is called the "Continental
Method" of Latin pronunciation, but which can be
justly denominated " Continental" only because it
has, by these means, obtained a partial foothold on
the continent of North America.
This, I think, is a fair statement of the amount
of light, if light it may be called, which our English
manuals of Latin instruction shed on the subject of
Latin pronunciation. Not a hint is given that the
English pronunciation of Latin deviates fiom the
Roman far more than that of the other nations of
Europe; not an inkling is vouchsafed of the clear
and abundant proofs by which the correct sounds
of the Roman letters have been established; not a




20             ROMAN ORTHOEPY.
gleam of light is thrown on the intimate and most
interesting connection between the sounds which
enter into and compose a language and its pervading genius, the structure and harmony of its verse
and its affinities in the great family of speech. On
all these points an absolute silence is maintained.
The pupil is informed that, in the absence of a-ny
other guide, he is at full liberty to conform his
Latin pronunciation to what is termed "English
analogy," but, with the exception of those falselyexhibited Continental vowel sounds, he has no light
and no liberty given him to do any thing else; and,
as if the more certainly to insure an absolute uniformity in error, this English analogy is expounded
in a set of rules which require the constant perversion of the pure sounds and violation of the true
quantities of the Latin language.
These charges I hold myself bound to make good
before I close the present discussion. The proofs,
however, can be more clearly presented and their
force will be more readily appreciated after a careful consideration of the true Roman system, to
which I now proceed. But, says one, the Latin
ceased some twelve or thirteen hundred years since
to be a living language, and in these circumstances,
can the Roman pronunciation of it be satisfactorily




ROMAN ORTHOEPY.                21
ascertained?  Every one who has properly investigated the subject will answer this question in the
affirmative. Latin pronunciation with the Romans,
was perfectly simple and invariable. The writings
of the ancient Latin grammarians contain the most
elaborate discussions in respect to the sounds of the
letters; every variation even of the vowel sounds
is noticed and the exact position of the organs in
the enunciation of each letter is described as accurately as the nature of the subject admits. The
entire silence of these authors with regard to so remarkable an irregularity in the notation ot the Latin
language as the use of one character to denote
totally distinct sounds, would be, in itself, sufficient
proof that no such irregularity existed. In addition
to this explicit testimony of those whose statements
are authoritative, the incidental proofs found in
other writings, and especially in the language itself-in its obvious laws of vowel, diphthongal and
consonant changes-are innumerable and entirely
harmonious both with each other and with the
statements of the Latin grammarians and rhetoricians.  Schneider, in his "Elementarlehre der
Lateinischen Sprache," gives the results with great
minuteness, and with references to fifty ancient
authors. The object of this Manual will be accom



22             ROMAN ORTHOEPY.
plished by a succinct statement of those results
accompanied, as we advance, by such illustrations
and proofs as the limits of our discussion will
admit.
The exceeding simplicity and regularity of the
Roman system render it, fortunately, an easy thing
to exhibit, and of course easy to teach and to learn;
and its beautiful consistency and completeness once
clearly apprehended, carry with them so much of
the force of an internal demonstration as to need
comparatively little corroboration from outward
and incidental proofs.
A.
SOUNDS OF THE VOWELS.
It is a distinguishing and excellent feature of this
system that each alphabetical character, with some
few and slight exceptions, represents but one elementary sound.
Each of the vowel letters, however, like the
a, t, and v, of the Greek system, stands for both
the short and the long sound of the vowel. But
these sounds, while differing in quantity, do not
differ radically in quality. Unquestionably a slight
modification of the quality of the sound is some



ROMAN ORTHOElPY.               23
times made necessary by the change of quantity,
but this, in no case, amounts to a radical change.
The long and short sounds of each vowel are, in
quality, substantially identical.
The vowels (A, E, O, o U,) are arranged in the
alphabet in a natural series, which is determined
by the position of the organs in sounding them.
There is a certain point in a scale extending from
the throat to the lips, where each sound is articulated —A is formed in the throat; E in the back part
of the mouth; I near the teeth; o between the
teeth and the lips; u in the lips. Again, in uttering
the sound A, the lips are opened widest; in E, they
approach each other slightly; in I, they approach
still nearer, so that they are closed at the corners,
and the mouth has now its widest lateral extension;
in o, they are rounded; and in u, they are both
rounded and protruded. See Appendix, note A.
The following may be given as a scheme of the
vowel sounds:A sounds like      the Eng. ah,  as in DA-BAM.
A   "    "         "   "   ah,    "  NA-BAM.
B,  c" "a  in    "         f ate,   "  VE-NI-O.
"    " ea  "  "   "   fane,  "  VE-NI.
I "      iee "  "   "   fleet,  "  VI-DE-O.




24             ROMAN ORTHOEPY.
i sounds like ee in the Eng. flee    "  VI-DI.
"                " o   "  "  "  note,  FO-VE-O.
"    "o  "  "  tne,     FO-VI.
u "    " 00 "  "  "  boot,  "  FU-GI-O.
u        " o""00          moon, "  FU-GI.
Y, used only in Greek derivatives, sounds nearly
like the French u.
We would here request the reader to keep in
mind an important difference between the Roman
and the English principle of syllabication. On the
Roman system, a single consonant standing between
two vowels, belongs to the latter; and any two or
more consonants, which may together begin a word,
belong, when standing between two vowels, to the
latter vowel. Hence the syllables of NOBILIS are
not NOB-I-LIS, but -N-BI-LIS; of COGNOSCO, not cognos-co, but co-gno-sco, because we have, e. g., such
words as GNA-RUS and SCI-PI-o. These rules, however, do not apply to the case of compound words,
where the syllabication is determined by the component parts, e. 9., the syllables of INTERERAM, are
not IN-TE-RE-RAM3, but IN-TER-E-RAM.




ROMAN ORTHOEPY.               25
B.
SOUNDS OF THE DIPHTHONGS.
As each of the five Roman vowel letters represents but one radical sound, varying only in quantity, so, in conformity with this strict and beautiful
simplicity of the vowel system, the sound of each
of the two elements composing the Latin diphthong was distinctly heard, and each diphthong
preserved uniformly the same sound. The only
difference, e. g., between AI as a diphthong and as
a dissyllable, was, that in the former the two sounds
were uttered with one emission of the breath
through the voice-passage, and in the latter, with
two such emissions.
Of the above Latin vowels, A is the most full and
open; next to A, in fullness and openness, come E
and o; I and u are closer articulations, allied respectively to E and o, and are either semi-vowels,
or, passing fiomn the vowel to the consonant sound,
are semi-consonants.  Now  in every diphthong,
whatever be the initial sound, the lips seem most
naturally either to be extended laterally so as to be
closed at the corners and drawn near to each other
at the centre, or to be protruded and rounded, until
2




2!5            ROMAN ORTHOEPY.
the i or u sound respectively is reached as the vanish of the diphthong, and in accordance with this
natural tendency of the human voice in combining
vowel sounds, the Latin diphthong consists properly of an open vowel followed by a semi-vowel, as
follows, viz.:
\A            -AI, AU;
E   ~<X)  ]EI, EU;
U
0o             0I, OU.
These were in fact the original diphthongs, the
exact counterpart of which is seen in the Greek
system, where the A, E, and o-sounds are each followed by iota and upsilo,.
Of these, however, the EI and EU are obviously
difficult combinations.  The E, being articulated
near the middle of the voice-passage, and being
also a weaker sound than A or o, is not easily, and
therefore not naturally, employed as the initial
sound of a diphthong, on which sound, of course,
the principal stress of the voice will generally rest,
and which, therefore, needs to be a strong vowel.
Accordingly, we find  that those  combinations,
though occurring not unfrequently in earlier times,




ROMAN ORTHOEPY.               27
were gradually changed into simple vowels, until
the former remained only in the two interjections
HEI, EIA, and also in DEIN when contracted into one
syllable, as is usual in poetry; and the latter, only
in HEUS, HIEU, EIIEU, CEU, SEU, NEU, NEUTER, and
NEUTIQVAM. Even in these hard combinations the
Romans adhered firmly to the principle of sound
ing distinctly the two elements of the diphthong, so
that the word NEUTER, e. g., was by some regarded
as properly a trissyllable, although the poets use it
as a dissyllable. With the above few exceptions, we
have, then, simply the two strongest vowels A and
o followed by the two semi-vowels, i and u, thus:A    i      AI, AU;
U
0           OI, OU.
These were the most common original diphthongs.
But subsequently, as the Latin  grammarians
state, and the monuments of the language clearly
show, the AI and oI came to be written usually AE
and OE, the sound remaining essentially unchanged.
This change of character must be regarded as a
mere orthographic corruption, growing out of the
close resemblance between these two vowel sounds;




28             ROMAN ORTHOEPY.
a resemblance which might easily pass for identity
in the vanish of a diphthong as we know that it did
sometimes in a short unaccented syllable. Cf. CAEREMONIA and CAERMAONIA, INTELLEGO and INTELLIGO.
Indeed, as a simple matter of fact, no orthographic
change occurs more frequently in the early monuments of language than that between E and o, I and
u respectively. But though the character E was
substituted for I, yet the voice would not naturally
stop with or vanish on the more open E sound, but
would advance to the closely allied and more slender sound which is articulated with the extreme
lateral extension of the lips. It was a natural result, then, that when E took the place, it also took
the sound of i after A and o. In the early period
of the language, for example, we find such forms as
AIDILIS, QVAISTOR, QVAIRERE, etc., instead of AEDILIS,
QVAESTOR, QVAEEREE, etc. It is contrary to all analogy and all probability to suppose that the popular
pronunciation of such words was changed with the
written character. Again, we find that the Romans
expressed the Greek Al and Olby AE and OE. And
to make the demonstration perfect, the ancient
grammarians agree in the statement that the AE and
OE took the sound as well as the place of AI and oI.
See Appendix, note B.




ROMAN ORTHOEPY.                29
The diphthong ou is found only in the early language. In consequence, no doubt, of the slight
difference between the o and u sounds, the diphthong was supplanted by u, which of course, was
regularly long.  Thus the ancient JOUS, JOUSTUS,
JOUSTITIA, became JU-s, JUJSTUS, JUSTITIA. So JOUBEO, JOUBERE, JOUSSI, JOUSSUM, became JUBEO,
JUBERE, JUSSI, JUSSUM. In the imperfect tenses of
this verb, the u is, by exception, short.
TUi, though generally treated as such in our Latin grammars, is certainly not a diphthong, the ear
itself being judge. Pronounced with the stress of
voice upon the first vowel, or equally distributed
between the two, it is manifestly a dissyllable
oo-ee; or, if the stress be thrown upon the second
(which is never the case with a proper diphthong)
it becomes a monosyllable = we; the u, in this case,
being necessarily regarded as a semi-consonant.
We give, then, the following scheme of the
diphthongs in common use. Their pronunciation
is determined at once and infallibly, by the well
established sounds of the component elements.
AE (=AI) sounds like ay the Eng. adv. of affirmation.
AU         "' i  OW in  "  now.
OE (=OI)  "          " Oi "   "   Stoic, or,
oe "   "   co-equal, if pro



30             ROMAN ORTHOEPY.
nounced rapidly, the former as a monosyllable, and
the latter as a dissyllable. It is nearly, but not
quite equivalent to oi in the English coil.
This view of the Roman vowel and diphthongal
systems will be found to throw much light upon
many points of the language; e. g., upon the question of the quantity of vowels in compounds and
derivatives, and upon the entire subject of the
composition and derivation of words. From  the
universal principle and practice of giving, in the
sound of each diphthong, the two elementary sounds
composing it, we can readily see how a diphthong,
in a simple or primitive word, will often pass, in the
compound or derivative, into the long sound of one
of the two component elements, or into a long vowel
closely related to one of the two sounds. Thus
AE (= AI) passes into A or i, as, e. g., AULAEBUS or
AULABUS becomes AULABUS or AULIBUS; firom AD and
QVAERO we have ACQVIRO; fiom OB and CAEDO, oCCIDO, etc. So also oi and OE are changed into 5 or i,
and often into u, in consequence of the close approximation of the sounds o and u. Hence, fiom
CON and EO, IRE, we have COITUS, then COITUS, then
(without change of sound) COETUS, a coming together, a meeting, an assembly. Hence, we find in




ROMAN ORTHOEPY.               31
the early language LOEBER, and LOEBERTAS for LIBER
and LIBERTAS, and at once see the reason of the long
i. Comp. FolKog, Foivo9, and vicus, ViNUM; so also
we find anciently MOERUS, PLOIRUME, OINIVERSEI,
OINOS, COMOINIS, COIRO, ARE, etc., for MURUS, PLURIME, 7UNIVERSI, UNUS, COMMUNIS, CURO, ARE, etc.
Hence fiom PROVIDENS we have first (the v dropping out, as often when standing between two
vowels), PROIDENS, then PRUDENS; from  JOVISGLANS, first JOVIGLANS, then JOIGLANS, then JUGLANS;
from CON and VINCTUS, COVINCTUS, COINCTTJS, CUNCTUS, " all together;" fiom MOVITO, ARE, fi'eq. fiom
MOVEO, ERE, we have IMOITO, then ~MUTO, ARE, to
move frequently, to change; fiom NOVIPER, NUPER;
so from POENUS arose PUNICUS, for which we find
anciently both POENICUS and PoINICUS.  Comp.
POENA (rrotvi) and PUNIO; MOENIA and rMUNIO, IRE;
so also AU often passes into fr, or its cognate o.
Hence, CAUSA, INCUSO; FRAUDO, DEFRUDO; CLAUDO,
RECLUDO, etc.; so fi'om SUB and FAUCES, the throat,
SUFF6CO, to stop up the throat, to svffocate; so
PLOSTRUM  and LOTUS, together with PLAUSTRUM
and LAUTUS, CODEX, and CAUDEX, etc., etc. See
Appendix, notes C and D.
To the student accustomed only to the English
sounds of the vowels, and in the habit of giving to




32             ROMAN ORTHOEPY.
the broad, open sounds of AE and OE the slender
sound of the Roman I, and to the AU the simple
sound of a in fall, these changes seem  unnatural
and almost unaccountable. He can not see how AE
and OE, having that slender sound, should pass, the
one into i, and the other into o or v, or why AU
should go into 5 or t. But he who is familiar with
the true sounds of the Roman vowels and diphthongs will understand these changes the moment
they strike his ear.
Again, the whole subject of quantity becomes,
on this system, as simple as that one and one make
two. The merest child can understand it. The
distinction between long and short syllables is not
nominal but real-not capricious and arbitrary but
stable and rational. And a child who has been
made familiar with the simple Roman system of
vowel and diphthongal sounds will comprehend
this distinction as readily as he will the difference
between two and one. Indeed, precisely that is
the distinction and the whole of it. It is well
known that the Romans anciently expressed the
long  vowel sounds by the use of two vowel
characters, and Quintilian informs us that this
custom  continued until after the time of Accius.
On this system, then, we may dispense with the




ROMAN ORTHOEPY.               33
tedious, often conflicting and ever perplexing chapters on quantity with their interminable rules and
exceptions, all made necessary, as we shall more
fully show hereafter, by the attempt to reconcile
things which are in themselves irreconcilable, viz.,
English vowel and diphthongal sounds and Latin
quantities. Only let the pupil be taught from the
first to give the proper sounds of the vowels and
diphthongs, in measure as well as in tone (and they
can be given in proper measure only with the
Roman tone) and he will no more mistake the true
quantity or accent of a word than the Romans
themselves did. Let him only be taught to pronounce the language correctly, and he will intuitively deduce from the language itself its general
rules of quantity. IHe will have that fine sense of
quantity which will enable him to catch the rhythm
of Latin poetry at once, and, as Horace says, to
" mark the legitimate sounds with his fingers and
ear." Mr. Walker gravely talks about " diphthongs
to the eye," and the English system, as will be fully
shown hereafter, actually converts all the common
Roman diphthongs into these "ocular diphthongs,"
which are really no diphthongs at all; that is, they
are not two different vowel sounds in one syllable,
but simply two vowel characters with one vowel
2*




34             ROMAN ORTHOEPY.
sound. But Horace knows of no such diphthongs;
and Cicero says, " OMNIUM LONGITUDINUM ET BREVITATUM IN SONIS SICUT ACUTARUM GRAVIUMQVE VOCUM  JUDICIUM, NATURA  IN AURIBUS  NOSTRIS
COLLOCAVIT."
On the other hand, to teach the student, from
his first entrance upon the study of Latin, the
English system of pronunciation; to get him thoroughly habituated to this false method, and then to
attempt, by lodging in his brain some verbal rules
of quantity and prosody at war often with each
other and  commonly with his pronunciation, to
make him appreciate and observe the rhythm  of
Latin poetry, is like keeping a child in a rude
society, where all the laws of a pure and finished
language are habitually violated, until his habits of
thought and speech are completely formed and confirmed, and then expecting him, by virtue of his
committing to memory the common rules of grammar and rhetoric, to talk at once with grammatical
and rhetorical correctness and elegance.
Again, this view  of the vowels and diphthongs
will also explain how the four vowel declensions,
commonly called the first, second, fourth and fifth
declensions, sprang fiom the one original declension
which has been retained most fully in what is now




ROMAN ORTHOEPY.                 35
commonly called the third, but which should be
called the consonant declension in distinction from
the other four. The case endings of this original
declension are as follows, viz.:
SINGULAR.                   PLURAL.
Nom.         s, or M, E*S.
Gen.         IS,                 ERUM.
Dat.         I,                  EBUS or IBUS.
Ace.         M,                  ES.
Voc.         like nom.           Es.
Abl.         E,                  EBUS or IBUS.
When these endings, all of them, except the nominative, accusative, and vocative singular, commencing with or consisting of a vowel, were attached to
roots ending in a vowel, there arose by syncopation, contraction and elision, various changes which
produced the four varieties of declension properly
termed the A, E, o and u declensions, according as
the roots end in those vowels respectively. Hence,
e. g., FAMILIA, FAMILIA-IS, FAMILIA-I, FAMILIA-M,
FAMILIA; FAMILIA-E.  In the  genitive we have
sometimes a contraction of the AI into A, giving
us FAMILIAS, as in PATERFAIMILIAS; generally the s
is dropped, leaving AY, as in the trissyllables still
* s generally for personal, and M for impersonal nouns.




36             ROMAN ORTHOEPY.
found in the poets, AULAI, AQvai, etc.; at other
times, the AI are united, giving us AULAI, a dissyllable; and finally the I is supplanted by E without change of sound. Thus we get the first or A
declension as follows, viz.:
SINGULAR.
Originally.                          Finally.
AULA,                               AULA.
AULA-IS, AULAS ALA, AUL ULAI, or    AULAE.
AULA-I,                             AULAE.
AULA-M,                             AULAM.
AULA,                               AULA.
AULA-E,                             AULA.
PLURAL.
AULA-ES,           (s, dropped)      AULAE.
AULA-ERUM,         (contracted)      AULARUM.
AULA-EBUS or IBUS, (con. and sync.)  AULIS.
AULA-ES,          (contracted)       AULAS.
AULA-ES,          (s, dropped)      AULAE.
AULA-EBUS or IBUS, (con. and sync.)  AULIS.
The dative and ablative plural are first contracted
into AULABUS or AULIBUS (both which forms are still
occasionally retained,) and the latter is then syncopated and still further contracted into AULIS, just as




ROMAN ORTHOEPY.                3 
QvIBJs goes occasionally into QVIS, and as HIBUS, the
old dative and ablative plural of HIc, has passed
into HIS.
We see at once, from  the above, why the ultimate of the ablative singular is long, though final A
in words declined is elsewhere short-why the accusative plural AS is long, while the accusative singular AM is short-why the penult of the genitive
plural is long, and why the ultimate of the dative
and ablative plural is long, though final is, is generally short, etc.  In the same way, we might
explain, with equally satisfactory results, the development both of the other vowel declensions fiom
the original declension, and also of the three vowel
conjugations (with roots ending respecively in A, E,
and I,) as distinguished from the third or consonant
conjugation; but it is unnecessary here to go
through with the details.
Now to the Roman-Latinist all these changes
are simple and natural, while the English-Latinist
without going out of his system, could never even
discover them.




38             ROMAN ORTHOEPY.
C.
SOUNDS OF THE CONSONANTS.
The main differences between the Roman and
English systems in sounding the consonants are
limited to the letters c, G, Qv, J, s, and T. I need
not, therefore, speak particularly of the other consonants.
The original Roman alphabet of sixteen letters,
as borrowed fiom the Greeks, did not contain the
letter G; but the C occupied the place of the Greek
r, and corresponded to it nearly in form and precisely in sound, as the K did to the Greek kappa.
Hence, in the earlier times, we find uniformly
LECIONES, MACISTRATOS, PUCNANDO, etc., instead of
LEGIONES, MAGISTRATUS, PUGNANDO, etc. See Aristarchus, De Arte Gram., i., 18, SUB FIN.
The C, however, was early hardened into the sound
of K, and, by degrees, actually supplanted the K in
all the words of the language except KALENDAE and
the proper names KAEso and KARTHAGO. It became
necessary therefore to employ a new character to
represent the sound of y; and for this purpose the
G was formed by appending the mark (t) to the C.
Thenceforth, so long as the Latin remained a living
tongue, the simple C was the sign of the hard




ROMAN ORTHOEPY.                39
K-sound (= Greek K), and the modified C or G was
the sign of the soft K-sound (= Greek y). Nor is
there the least evidence that they were ever employed for any other purpose.
In the golden age of the Latin, the Greeks and
Romans, who were perfectly familiar with each
other's language, regarded the c and the Greek c,
the G and the Greek y as entirely equivalent. When
the Romans wrote Greek words in Latin characters,
they used c to represent ic before the vowels E and I,
as well as before the other vowels and the liquids;
as CECROPS, KKpbo; CILICIA, KtUldia. The Greeks,
on the other hand, used their Ki for the Roman c in
all cases; since in writing Latin names they write
the syllables, CE, cI, with K and not with f; as, KtLKpov, CICERO;  liZtrov, SCIPIO.  Suidas, speaking
of the c worn on the shoes of Roman Senators, calls
it Tr Pwo/ai'lKv cKad7ra. See Aristarchus, DE ARTE
GRAM., LIB. i., c. xvii. The Romans wrote repvlov,
GERYON; yTya/, GIGAS. The Greeks wrote VIRGILIus, OV7tpyL)to. A passage in St. Augustine furnislhes positive evidence that the Greek y, and the
Roman G, were pronounced alike even before the
vowel E.  He says, "CurI DICO LEGE, IN HIS
DUABUS SYLLABIS, ALIUD GRAECUS, ALIUD LATINUS




40            ROMAN ORTHOEPY.
INTELLIGIT," thus clearly showing that the Latin
LEGE and the Greek XTye had the same sound.
The fact, too, that the union of either c or G with
s always produces x, shows that both are gutturals,
as, e. g., PAX from root PAC, and nominative ending
s; REX from BEG and s. See App., Note. I.
Again, we have in syncopated words proof that
c and G invariably retained their guttural sound; as
in the supines of DOCERE and FACERE. If the c had
the sound of s, the supines would have been pronounced DOSETUM, FASiTUM  and the syncopated
forms would naturally have been DOSTUM, FASTUTM,
not DOCTUM, FACTUM. If the G of LEGI had the sound
of the English J, the supine LEGITUM must have been
pronounced LEJITUM, and the syncopated form could
not have been LECTUM.
Again, we have TRICESIMUS and TRIGESIJMUS from
TRIGINTA; QVADRINGENTI for QVA)RINCENTI; DECI-.Mus and DECUMUS from DECEM. Now unless the c
and G were invariably gutturals before E and i as
well as before the ether vowels, such orthographical changes could not have taken place. We must
suppose that the c in DECIMUS, DECUMUS and DECEM
had one uniform sound corresponding to that of the
K in 6eKa.
On this point, also, the testimony of the Latin




ROMAN ORTHOEPY.             41
grammarians is full and conclusive. In the discussion of the question whether the K was to be numbered among the letters of the Roman alphabet the
Latin writers make the most explicit declarations
of the perfect identity of the c and K in regard to
the sound they represented. Quintilian, on the
ground of this identity, condemns the use of K in
writing Latin words.
"NAM K QVIDEM IN NULLIS VERBIS UTENDUM
PUTO.. (HOC EO NON OMISI QVOD QVIDAM EAM,
QVOTIES A SEQVATUR, NECESSARIAM CREDUNT;) CUII
SIT C LITERA, QVAE AD OMNES VOCALES VIM SUAM
PERFERAT." LIB. i., Vii. 10.
Priscian agrees with Quintilian in regarding the
K as superfluous.
" II ENIM ET Q QVAMVIS FIGURA ET NOMINE VIDEANTUR ALIQVAME HABERE DIFFERENTIAM, CUM C
TAMEN EANDE3M TAM IN SONO VOCUMI QVAMI IN METRO,
CONTINENT POTESTATEM.  ET K QVIDEM PENITUS
SUPERVACUA EST; NULLA ENIM RATIO VIDETUR, CUR,
A SEQUENTE, K SCRIBI DEBEAT." LIB. i., C. iv.
And again:
"QVAM3VIS IN VARIA FIGURA ET VARIO NOMINE
SINT K ET Q ET C TAMEN, QVIA UNAM VIM HABENT
TAM IN METRO QVA3 IN SONO, PRO UNA LITERA ACCIPI DEBENT." LIB. i., C. iv.




42             ROMAN ORTHOEPY.
"K  SUPERVACUA EST, UT SUPRA DIXIMUS.; QVAE
QVAIVIS SCRIBATUR, NULLAM ALIAM  VIM  HABET
QVAM C." LIB. i., c. viii. See further, Appendix,
notes E and F.
We may be sure then that the Roman c and G
corresponded precisely and uniformly with the
Greek K and y.
Q appears only before v, and Qv only before a
Latin vowel. In the classical period of the language, Qv as a compound character, represents invariably a simple consonant sound; never creates
position; is often interchanged with c; and had,
according to the statements of the Latin grammarians, precisely the same sound with the c and K.
In addition to the authorities already quoted, see
Aristarchus, DE ARTE GRAM., LIB. i., CAP. xviii.,
who states explicitly that the c, K and Q have the
same sound, and that the v after Q is silent. Cicero
makes a pun, in which COQVE, the voc. of COQVUs,
a cooc, and the adverb QVOQVE, also, must have
been assumed to be sounded alike. QVOTIDIE and
COTIDIE, QVU3M and CUfM, COQVI and coci, COLLIQVIAS
and COLLICIAS were different modes of spelling the
same words, both in common use at the same
period of time, and must therefore have been pronounced alike. The Romans also frequently wrote




ROMAN ORTHOEPY.                43
SEQVUTUS and SECUTUS, LOQVUTUS and LOCUTUS;
from QVATIO we have CONCUTIO, where v represents
the A and c the QV of the simple verb; and fiom
RELINQVO we have RELI(N)CTUS, RELICTUS; again the
QV, like any other guttural, combined with s forms x,
e. g., COQvjO, COQviERE, coxii, COC[TUM; again, all
adjectives whose roots end in a vowel are compared
by the use of MAGIS and  AXIMLE, but those whose
roots end in QV are not so compared, e. g., IDONEJUS,
MAGIS IDONEIUS, MAXIME IDONEjUS, but ANTIQVjUS,
ANTIQVIIOR, ANTIQVIISSIMUS. Hence in the triliteral
word QVA, QVE, QVI, QVO, QVU, the QV has always the
simple sounds of'c or K. In the genitive and dative
singular of QVIS or QVI, viz., cUJUS or cuius and cui,
the u represents not the u of the nominative, but the
o which is found in the old orthography QvoIus  and
QVOI, and the c represents the Qv of the nominative.
The substitution of c for QV in this word took place
in Quintilian's time. He tells us that, in his youth,
the dative was still written QvoI.  The genitive
therefore is to be pronounced koo-yoos, and the
dative either as a dissyllable koo-ee, or more commonly as a monosyllable in which tile u passes over
into a consonant sound resembling our w, kwee.
But the nominative is kee. The English system,
witl characteristic perversity, reverses this rule and




44             ROMAN ORTI-IOEPY.
mispronounces both the QV of the nominative and
cu of the dative, to say nothing of its false sound of
the vowel. Upon such general grounds, we miglht
be sura of the Roman pronunciation of the genitive
and dative of this pronoun; and I was in the habit
myself of pronouncing the nominative kee and the
dative kwee for many years before I noticed an
incidental remark of Priscian, which shows very
clearly how the Romans pronounced them. After
stating that c, K and Q were identical in sound, and
that K was therefore superfluous, he adds that the
Q as well as the K would be superfluous, but that it
seems to distinguish words in which the v was
silent —as in QvI —from those in which it was pronounced, as in cuI. L. i., c. iv. That the real difference between the pronunciation of QVI and cuI
was anciently observed in England, is proved by
the directions given by Beda in his treatise on
orthography.  "Q  LITERA TUNC RECTE PONITUR
CUM ILLI STATIM U  LITERA ET ALIA QVAELIBET
PLURESVE VOCALES CONIUNCTAE FUERINT ITA UT
UNA SYLLABA FIAT; CAETERA PER C SCRIBUNTUR.
QvI SYLLABA PER QVI SCRIBITUR; SI DIVIDITUR, PER
cuI SCRIBENDA EST."  So HUIC the dative of H1c, is
either hoo-eec or. hweec. LO-Qvor, SE-QVOR are
dissyllables, but they were sometimes written LO-CU



ROMAN  ORTHOEPY.                     45
OR, SE-Cu-OR, when  they  were trissyllables, firom
which  we have LO-CU-I-TUS, SE-CU-I-TUS, and  contracted  LO-CU-TUS, SE-CU-T-S.   There is not the
slightest doubt, then, that in the classical times, QV
represents the simple guttural or K-sound.*   See
Appendix, note G.
* In regard to the characters c, K, QV, it is natural to inquire,
first, Why did the Romans, the notation of whose language is
in general so simple and regular, employ these three characters
to represent one and the same sound (== Greek K)? and second,
How are we to account for the fact that thev after Q is invariably silent and without significance in the classical period of
the language?
The first question admits of a ready answer. In the Latin, as
in other ancient languages, e. g., the Hebrew and the Greek,
several of the letters appear to have had a syllabic power. One
character, was used to denote the syllable KA, another KO,
and so on. These two characters were called in Hebrew kaph
and koph, and in Greek, kappa and koppa, the names of these
letters thus indicating their syllabic nature and use. After they
lost this syllabic power, they were still employed as signs of the
same initial consonant sounds, but were restricted, for a time,
each to a connection with the vowel of the syllable it originally
represented. The Greek kappa, e. g., after it ceased to be syllabic in power, was nevertheless used, for a time, only before the
vowel a, and koppa only before the o-sound; but eventually,
one of them becoming thus superfluous, the koppa disappeared
from the alphabet, and the kappa was employed before all
vowels. So in Latin, K originally denoted the syllable KA, and,
after it ceased to be syllabic, was never used before any vowel
but A; so Q was used only before v, and c before E, I and o. The
name of the a, viz., KA (a as in father), and of the Q, viz., Qv,
(- koo) clearly indicates the nature and use of each. And our
names of these and other letters are relics of this same usage,
only we have corrupted the vowel sound. We call k, ka (a as




46                 ROMAN  ORTHOEPY.
I, when used as a consonant, or J as it has commonly been written within the last two centuries,
had the sound of our las it has in German.  This
was the natural result of the  hardening  of the
in made,) and not ke which would better accord with be, ce, de,
etc., the names of b, c, d, etc., and we call q, ku, only changing
the vowel sound from oo to u as in tube. The Anglo-Saxons,
undoubtedly, pronounced the ka and ku as the Romans did, and
~the Germans do, - kah and koo.
Now just as in Greek the koppa became superfluous, and disappeared from the alphabet, so in Latin, after c became established as a hard.K-sound, the K became superfluous and virtually
disappeared. KAPUT, KALUMNIA, etc., became CAPUT, CALUMNIA,
etc. The K was retained only in the three words already mentioned, and in certain abbreviations. One mode of abbreviation
common among the Romans, furnishes an illustration of the
original custom of giving to the letters a syllabic power. This
consisted in denoting a syllable by a single consonant whenever
the vowel of this syllable was identical with that by which the
consonant was pronounced in the alphabet.  Thus B stood for
the syllable BE; BNE stood for BENE; so D for DE; and DciMus for
DECIMUS; so Q for QV, and QID, QAE, for QVID, QVAE; so c stood for
the syllable CE; but as the vowel A was contained in the name
of K, this letter was used to denote the syllable cA; thus CRA
would be read CERA; but KRA, CARA. See Appendix, note H.
There remain, then, in ordinary use in Latin, only c and Q or QV,
as the representatives of the hard K-sound.
How then, the second question is, did this v become invariably silent and without significance? This usage arose, I think,
from the singular fact that the QV, unlike either the CE, ci, CO, or
KA, were employed only before a vowel. The v after Q must
originally have been sounded; as a vowel = oo, and as a consonant - w.  QVAM, e. g., must originally have been pronounced as a dissyllable koo-am, and as a monosyllable, which
it ordinarily was, kwam. But when the QV stood before another




ROMAN  ORTHOEPY.                     47
vowel sound I before another vowel.  IAM  may
be either a dissyllable or a monosyllable, according
as the I is used as a vowel or a consonant. It is,
therefore, either EE-AM, or, when the two syllables
v, the two short vowels would naturally unite to form a long
one. Just as AMAABAM, though so written till near the time of
Cicero, was nevertheless a trissyllable, pronounced AMABAM, so
QVvM, which might be a dissyllable, koo-oorn, would ordinarily
be a monosyllable, and pronounced koom rather than kwoomnY
Now, if the Romans could have reduced the written characters
QVVM to QVM, as they did AMAARE to AMARE, DOCEERE to DOCERE,
etc., they would unquestionably have done it, and then, when
the oetters had lost their syllabic power, and c had become established as a hard K-sound, the c would, in all probability, have
entirely supplanted the Q and the Q alone, as it did the K, while
the single v following it, would have retained its sound and significance as a vowel, and the syllables of the-words in which
the c thus took the place of Q would have been unchanged. But
as they could write only QVVA (and not QvM) and this was ordinarily pronounced koom, the attempt to reduce the word to the
written form of a monosyllable would naturally result in the substitution of c for Qv, i. e., cvw, for QVVM. Thus the c would be apparently identical with QV. Then again, as the o and u-sounds were
closely allied, and the two characters frequently interchanged, as
SERVOS and SERVUS, QVOR and QvuR (subsequently, CUR) the QVo,
also, would naturally be sounded as a monosyllable, Ko, not.rwo,
and we should have QVOTIDIE and COTIDIE, two common contem.
poraneous orthographies with one pronunciation.  Now just il
proportion as these orthographic variations, these interchanges of
c and QV became more numerous and frequent, the original
power of the QV would be gradually modified in the popular
conception and pronunciation until it became, in the classical
times, actually equivalent to the simple c before all vowels; so
that QVA, QVE, QVI, would be pronounced like KA, KE, KI, as
well as Qvo and QVU like KO and KU.




48             RO.MAN ORTHOEPY.
are rim into one by a rapid pronunciation, becomes
inevitably YAM, and the compound ETIAM is sometimes a trissyllable, ET-EE-AM, but generally a dissyllable, ET-YAM. So PARIES is either PA-RI-ES (pro.
nounced, PA-REE-ACE) or PAR-YES (pronounced PARYACE); PARIETE is either PA-RI-E-TE or PAR-YE-TE;
POMPEIUS is either POM-PI-I-US or POMI-PE-YUS, etc.
S is always a sharp sibilant like the Greek I, and
is sounded as in the English sin. The Latin MUrs
then corresponds in sound precisely with the Greek
tovaa.  The Romans, by the way, uniformly represent the Greek ov by their v, and the Greeks the
Latin v by their ov. CATf'LITS, in which, as Klotz
states, the penult vowel was so short as searcery to
be heard, was translated thus into Greek, Kar)wovg.
T always preserves its pure sound, RATIO=RATEE-O; ARTIUM = AR-TEE-00O; JUSTITIA = YOOSTEE-TEE-A; etc., etc.
F and H, I may remark in passing, were originally mere marks of aspiration, the former for the
spiritus lenis (the  Eolic digamma F) the latter
for the spiritus asper. The form of the latter resulted from the union of the two Grecian breathing
marks F and j. They were afterwards treated as
letters, and together with G introduced between E
and I. The originally soft F was, however, hard



ROMAN ORTHOEPY.               49
ened, and passed over almost into the sound of the
Greek I. Hence FABITS = — dflog', q)vyO- = FUGA;
while the breathing previously indicated by F was
subsequently expressed by V when used as a consonant. The H besides its use in the beginning of
aspirated syllables, and between two vowels in the
middle of a word (e. g., VEUHEENS), was employed
also after the consonants c, P, T, R, to aspirate
them. This usage appears originally in Grecian
words where CH, PH, TH, RH, answer to the Greek
x,,, tp.
With these remarks on the consonants whose
Roman pronunciation differed from the English, we
may proceed to the
DIVISION OF THE CONSONANTS.
They are divided first into simple and double
consonants.
The simple consonants are distinguished,
1.) According to the different organs which are
particularly active in their formation; as
Throat-letters, gutturals;'G, c (QV,) CH, R, H
Tongue-letters, linguals; D, T, TH, L,N, J, S.
Lip-letters, labials; B, P, PH, F, M, V.
2.) According to the nature of the action of the.R




50              ROMAN ORTHOEPY.
organs in their formation, or according to certain
degrees of articulation, as
a.) Miutes, i. e., such as are formed by the strongest
effort of the vocal organs, or, are most perfectly
articulated.  These are either,
Soft   (mediae), G, D, iB,
HIard  (tenues); c (QV), i, P;
Rough (aspiratae); CH, TH, PH, F.
b.) Semi-vowels, i. e., consonants which, in respect
to the action of the vocal organs, approach most
nearly to the vowels.  These again are subdivided
into
a.) Liquids; i. e., those which readily flow or
unite with the mutes; L, 3, N, B.
3.) Spirants, those which  are formed by the
breath or voice most slightly interrupted; n, J, s, v.
By the breath, H, s.
{ "  "  voice,  J, v.
In accordance with these distinctions we give the
following tabular view of the simple consonants,
exhibiting the two grounds of classification. The
mutes belonging to the three organs, the throat,
the tongue and the lips are called respectively the
K-, T-, and P-sounds.




ROMAN ORTHOEPY.                  51
LETTERS OF THE SAME ORGAN.
GUTTURALS.   LINGUALS.    LABIALS.
1. MUTE.     (K-Sound.)   (T-Sound.)   (P-Sound.)
A) soft         G           D            B
B) hard       C (QV)      T           P
C) rough      CH          TH          PI, F
2. LIQUIDS.    R         L, N        M
8. SPIRANTS.    H         J, S 
The hard and soft mutes are sometimes properly
distinguished as surds and sonants, the term surd
being applied to those letters whose articulation is
preceded by an entire interruption of the voice;
and the term  sonant to those which are accompanied by an  audible murmuring of the voice.
Thus c, T, P, are surds; G, D, B, sonants.
The double consonants are,
x from cs, GS, QVS, rarely HS;
z from DS or SD.
The former is the only genuine Latin double consonant. The latter belongs properly to no Latin
word, but is found only in Greek words and in
some proper nouns borrowed from  foreign languages, e. g., ZONA, ZAMA.




52             ROMAN ORTHOEPY.
The aspirates cH, TH, PH are originally foreign to
the language. They appear, therefore, chiefly in
words borrowed from  the Greek, although even
many of these were deprived of the aspiration.
Only in a few common nouns, as PULCHER, SEPULCHRUM, the cH is used instead of c. It occurs also
in a few proper nouns, as GRAccHUS. TI occurs
also in some proper nouns, as CETHEGUS, SPINTHER, OTHO, CARTHAGO.  Elsewhere it was not
used in native words.  F was frequently written
instead of PH, even in nouns derived fiom  the
Greek. PHASELUS, PHALERAE, DELPHINUS were frequently written FASELUS, FALERAE, DELFINUS, etc.
Instead of TRIUMPHUS (Opltaftzog) the original orthography was TRIUMPUS.
The aspirated R (nR) is retained in words from
the Greek in which p stood at the beginning of a
word  or after another p, and hence we have
RHYTHMUS, RHETOR, PARRHASIUS.  Also several
words of barbarian origin were written with RH, as
RHENUS, RHODANUS, RHEDA,
F, in the best times, corresponded more or less
closely to the Greek 0, as is evident from  Quintilian 12, 10, 29. It is strictly therefore the only
aspirated mute among the sounds of the Latin
language.




ROMAN ORTHOEPY.               53
Such, in its admirable simplicity, symmetry and
completeness, was the system according to which
the Romans themselves pronounced their noble
language: a system ascertained and established not
only by its interior harmony with all the parts of its
structure, and all the facts of its historical development, but also, in every important particular, by
the testimony of its own grammarians, and the
concurrent opinions of the best Latin scholars of
modern times. With this system before us, we are
prepared to justify the twofold charge which we
have brought against the present English method
of Latin pronunciation: (1.) that it thoroughly
perverts and vitiates its elementary sounds, thereby
destroying its euphony and greatly obscuring the
facts and principles of its derivation and composition; and (2.) that it absolutely compels the violation of its quantities, and renders impossible a right
reading of its verse.
And first, how utterly at variance the English
method is with that above detailed must be obvious
at a glance to every observer. It gives to every
vowel, and also to many of the consonants, sounds
which are radically different, not only from those of
the Romans, but also fiom  each other; and to
every diphthong, a sound not merely different from




54             ROMAN ORTHOEPY.
that of the Romans, but entirely incompatible with
the fundamental principle of the Roman diphthongal system. A character which, in one position, represents the short sound of one vowel, in
another position, may represent the long sound of
another vowel. Thus in SAGA, a mantle, the first A
is pronounced as E long, the second as A short; in
DECEM, the first E is pronounced as I long, the second
as E short; in cicl, the long and short vowels are
both made long with the sound of the diphthong
AI; but in the adjective cicINus, the first I, which
is still long, is made short with the entirely different
sound of I in pin; in NOVI, the long o has its proper
sound, but in its derivative, NOBILIS, the long o is
made short, with a different sound wholly unknown
to the Latin language; in FUO, FUI, the u is made
long; in FUR, the f is made short, and the sound
radically changed.
Again, this system gives to simple vowels the
sound of diphthongs, and to diphthongs that of
simple vowels and even of short vowels, though a
diphthong is always long and always admitted to
be so by the English-Latinist himself. The I of
DICO is pronounced AI, the AE of CAEDO, I; in the
compound PRAECIDO the diphthong and the simple
vowel I are made to change places; the diphthong




ROMAN ORTHOEPY.               55
in CAESAR has the sound of i long; in CAESARIS,
that of E short. Not one of the Latin diphthongs
is, on the English system, pronounced as a diphthong, except EI (which occurs in only three or
four words) and that is mis-pronounced. AE and
OE have the simple vowel sound of the English e,
either long or short, as in mete or met; AU, that
of the simple a in fall; ETu that of u in tube. So
much for the perversion of the pure sounds of the
Latin vowels and diphthongs. Not one of them
has escaped the contaminating touch of the English
system.
But the system does, perhaps, even greater violence to some of the consonant sounds. C, always
with the Romans, a hard guttural mute, which is
one of the most important consonants in a language,
often becomes a mere lingual spirant, either hard
or soft, i. e., it becomes either the hard or soft
sibilant, s or z.  G, always a soft guttural, is
changed into a lingual or dento-lingual, representing one of the most unpleasant sounds in our language, and one which is wholly foreign to the Latin,
that of J, or DG. So, again, the J consonant (English y) is made to represent the same harsh sound.
In JUNGIo, JUNGIERE, JUNxlI, JUNCITUMf, pronounced
Latine, there is not a disagreeable sound, and each




56             ROMAN ORTHOEPY.
letter represents here, as elsewhere, but one sound;
but it is impossible to imagine a more unpleasant
combination of sounds than is heard when those
words are pronounced Anglice.  So again, the
sound required to be given to si and TI when followed by a vowel, as in RATIO, OFFENSIO, pronounced RASHEO, OFFENSHEO, is wholly alien fiom
the Latin, and offensive to every refined ear. In
view of these corruptions of Latin vowels, diphthongs and consonants, it must be admitted, I think,
by every candid mind, that the English mode of
Latin pronunciation has forced into that language a
vast number of harsh and discordant sounds which
greatly mar its really beautiful and euphonious
character.
Nor does this system do violence merely to the
euphony of the language. The changes which we
make in the sounds of the consonants, especially of
the gutturals and of J, introduce a great number of
irregularities and anomalies of pronunciation, which,
besides being wholly foreign to the language, obscure the relation of derivatives to each other and
to their primitive; of compound to simple terms;
of different cases of the same noun, or tenses of the
same verb to each other and to their common root.
These changes are sometimes so wanton and absurd




ROMAN ORTHOEPY.                57
that nothing but the force of habit, established
from  our infancy in the use of our vernacular
tongue, could reconcile us to them.  For instance,
in verbs which are reduplicated in the perfect, we
form the reduplication by using a consonant sound
totally different fiom that of the initial letter of the
verb-root, and also change the sound of that letter
itself. From  CA-NO, instead of C.-CI-NI (KE-KI-NI)
we have SES-I-NI; from CA-DO, instead of c.-ci-DI
(KE-KI-DI) we have sES-I-DI; from CAE-DO, instead of
CE-CI-DI (KE-KI-DI) we have SES-I-DI.  But perhaps
the absurdity of such changes is nowhere more
apparent than in many compound verbs where the
final consonant of the preposition is assimilated to
the initial of the verb, as in ACCIpio, OCCIDO and
occiDo.  How  beautifully and clearly does the
pronunciation, AKKIPIO, 6KKIDO and OKKIDO point
out the elements AD and cXPIO, OB and CADO, and
OB and CAEDO! But, on the English system, after
we have changed the sound of the verb's initial
from a guttural to the sharp sibilant s, we do not
allow the preposition, either to recover its original
final letter, or, by way of assimilation, to assume a
sibilant. We say neither AD-SIPIO, nor AS-SIPIO,
but AK-SIPIO; neither OB-SIDO, nor OS-SIDO, but
OK-SIDO, etc.
3*




58             ROMAN ORTHOEPY.
Again, by this practice of capriciously changing
the sounds of the consonants according tq the
vowel following them, a slight inflectional or other
variation in the spelling of a word, the elision or
transposition of a single letter are enough to make
an entirely new word of it.  The cases of AMICUS
pronounced Anglice, without stopping  here to
notice the radical changes of vowel sounds, are
as follows:
AMIKUS,                      AMISI,
AMISI,                       AMIKORUM,
AMIKO,                       AMISIS,
AMIKUM,                      AMIKOS,
AMISE,                       AMISI,
AMIKO.                       AMISIS.
What a regular and beautiful variety! Nor have
we variety merely, but confusion. How is it possible, for example, to distinguish the genitive singular, and nominative and vocative plural, from the
perfect of the verb AMITTO?  Who, again, would
ever imagine that FAC and DIC had come, by elision,
fromn FASE and DISE? or NEC fiom NEKWE?  Who
would suppose CERTUS, and CRETUS, as we pronounce them, to be different forms of the same participle? For it must be borne in mind that such
elisions and metatheses are part of the history of




ROMAN ORTiHOEPY.              59
the living language, occurring not in the use of
the learned or on the written page, but in the
mouth,of the people. Such illustrations of the
confision created by these capricious and radical
changes of consonant sounds might be multiplied
AD LIBITUM. But those already furnished are sufficient for our present purpose.
Again, this corruption of Latin pronunciation
has erected a high wall of partition between the
Latin and its kindred languages-a wall which
very few Latin students, accustomed only to the
English method, have ever attempted, and fewer
still have been able, to scale-a wall which entirely
shuts out from the view of the exclusive English
pronouncer a large majority of the ethnical affinities of the Latin language. What mere English
Latinist would be likely to discover the relation,
e. g., between ACOETIS (pronounced a-see-tis) and
the Greekj acolTlg? between CERCURUS (sur-cu-rus)
and KcpKOpvpo? between cicI (sai-sai) and  cti?
between CICINus (sis-i.nzus) and KitLvo? between
COENA (see-na) and Kolvog? between CELLA (sella)
and the German KELLER? between CISTA (sista)
and EKISTE? between CICER (sai-ser) and KIcHE?
between  CINC-TUM  (sink-turn) and the English
kink? between  TUM-ULUS and tomb? etc., etc.




60             ROMAN ORTHOEPY.
The Greek, which is so closely allied to the Latin,
but which has suffered far less than the Latin from
these corruptions, becomes to our English-Latinist
a very strange language, rather than a sister of the
same family. lie has learned, e. g., to pronounce
the Latin OCEANUS, o-shee-a-nu or still more erroneously, o-shee-d-nus; he meets the Greek word
Rcedvog and thinks it quite an alien, but the Roman
Latinist finds the two words pronounced precisely
alike, (except a slight difference between the o and
u,) and perceives at once their identity; the English Latinist is staggered at the difference between
CILICIA (Sai-lish-y-a) and KILIKRA, but the Roman
Latinist sees no difference at all. Just so in hundreds of other instances which might be named.
But this point will be more fully illustrated in
another place.
Having thus substantiated, as I think, the statement that the English system of Latin pronunciation thoroughly perverts the pure sounds of the
language, and having shown some of the pernicious
effects of that perversion in introducing harshness,
irregularity and confusion, into a language which,
in itself, is eminently euphonious, symmetrical and
philosophical, I proceed to the remaining charge,
viz., that this system not merely tolerates, but




ROMAN ORTIIOEPY.              61
enjoins the violation of the true quantities of the
language, and, consequently, of the just measure of
Latin verse. We concede to the English system
that it makes this stern demand of the sacrifice of
Latin quantity "not willingly, but of necessity."
Such is its nature, it can not do otherwise. It possesses an inherent and ineradicable incompatibility
with the proper Latin quantities, and with the
whole theory of Roman verse. It may be thought
a bold and arrogant statement, and yet we will
have both the courage and the frankness to make
it, that no man can read, on the English method,
any ten consecutive lines of Latin poetry of any
considerable length, say dactylic hexameters, without violating Latin quantity at least as many as
twenty-five or thirty times, and frequently twice or
thrice thirty times. We sometimes meet with the
striking inconsistency of an earnest stickler for the
English method, who plumes himself upon his
achievements and skill in this matter of Latin
quantity, and manifests a self-righteous horror of
what he calls " false quantity," but who nevertheless unconsciously makes one or more false quantities in nearly every Latin word he utters. The
fault as well as the inconsistency belong not so
much to the individual as to the system, a system




62             ROMAN ORTHOEPY.
which, at the same time, both forces its practitioner
into error, and blinds him to its existence. To this
point it becomes necessary for us to give special
attention, in order to set forth in its true light the
immeasurable superiority of the Roman over the
English system of Latin pronunciation. By "false
quantity" the pronouncer of Latin on the English
system, or the English pronouncer, as I will briefly
call him, means simply "false accentuation" of a
word from  non-observance of the quantity of its
penultimate syllable. The quantity of the other
syllables he is indifferent to, and, by the necessities
of his system, is perpetually violating. Only place
the accezt on the right syllable of a word, and our
adept English pronouncer will pronounce you " all
right" in the matter of quantity, though you give,
as on the English system you may, a false quantity
to every syllable in it. Take, e. g., the word FRIGIDIS. The penult being short, the accent is on the
antepenult. But the ultimate and antepenultimate
vowels are long; yet our English-Latinist doesn't
say fraije-dais but fr-j-e-dis, making  the first
and last syllables short instead of long, and thus
robbing the word of two fifths of its quantity;
in FABULIS, though he give the right accent, he




ROMAN ORTHOEPY.                 63
makes the two long syllables short and the short
one long.
This disregard of quantity, except so far as quantity determines the place of the accent, the EnglishLatinist constantly exhibits.  The Latin Grammar
of Andrews and Stoddard, which is, perhaps, most
commonly used in this country, contains the following: "To pronounce Latin words correctly, it is
necessary to ascertain the quantities of their last
two syllables only; and the rules for the quantities
of final syllables would be unnecessary, but for the
occasional addition of enclitics. As these are generally monosyllables, and, for the purpose of accentuation, are considered as parts of the words to
which they are annexed, they cause the final syllable
of the original word to become the penult of the
compound.  It is necessary, therefore, to learn the
quantities of those final syllables only which end in
a vowel." This quotation gives us the pith of the
whole doctrine of the English system on the subject
of Latin quantity. The quantity of the penultimate
syllable is ordinarily all we need to know in order
" to pronounce Lcatin words correctly."  Now the
truth is, that to read Latin, and especially Latin
poetry, correctly, we must give each syllable of
each word its just measure of sound.  Cicero in



64              ROMAN ORTHOEPY.
forms us, that the Romans, even in their common
discourse, were very careful to mark the quantity
of every syllable. And think we that the poet was
indifferent to the quantity of a single syllable introduced into his verse?  Is it not the regulated
succession of long and short syllables, and not, as
in English, of accented and unaccented syllables,
which constitutes the rhythm of Latin poetry?
Cicero tells us, DE ORAT., L. iii., c. 48, that the
Roman poet was so strictly controlled by the laws
of measure that not a syllable in his verse might be,
by the smallest breath even, longer or shorter than
was proper.  Even if the poet had been disposed
to admit the slightest violation of the proper measure, the people, as Cicero states in the same book,
c. 50, would not allow him  to do it; he says that
those who were entirely ignorant of the rules of
versification had so nice and intuitive a sense of the
true syllabic measures, that they would raise a
clamor against the dramatic poet, if he made a
single syllable either longer or shorter than was;ust. How then can we give the real rhythm of a
Latin verse without observing the true quantity of
each syllable of it? It is obviously impossible.
We have, then, two impossible things. First, it is
impossible for the English-Latinist not to violate




ROMAN ORTHOEPY.               65
Latin quantity, perpetually; and second, it is impossible to preserve Latin rhythm without observing
Latin quantity, uniformly.
Before proceeding further with this point, we
would request the reader to bear in mind that, on
the English system of Latin pronunciation, A long
sounds like a in made, A short, like a in mat; E
long, like e in mete, E short, like e in met; I long,
like i in pine, I short, like i in pin; o long, like o
in note, o short, like o in not; u long, like u in tube,
u short, like u in tub. Now it will be found, upon
examination, that it is only in words having the
penult vowel long, and followed by a single consonant, that we are sure, on this system, of having
the quantity of even one syllable correctly given;
and that, in all other words, whether dissyllables
with short penult, or polysyllables with accented
antepenults, every syllable may have a false quantity. TU-BIS, e. g., the dative plural of TU-BA, is
pronounced TU-BiS; RE-GU-LIS, dative of R:E-GU-LA,
is pronounced REG-U-LIS. In words of two syllables,
the English rule lengthens the penult vowel when
followed by a single consonant, whatever be its real
quantity; and in words of more than two syllables
having the antepenult accented, it shortens the
accented syllable, whatever its quantity, except




66             ROMAN ORTHOEPY.
(lst) in the case of u, and (2d) in those words in
which the vowel of the penult is followed by another vowel.  In these, the rule lengthens the
accented antepenult, whatever its quantity, unless
the vowel of the accented syllable be i, which is
invariably shortened, whatever its quantity. In an
accented antepenultimate syllable, then, the I is
always short, and the u always long, whatever be
their real quantity. In accordance, then, with the
above rules, PATER, MERUS, [MISER, MOLA, NURUS,
become PATER, MERUS, [MISER, MOLA,. NURUS: FABULOR, REGULA, DILIGO, NOBILIS, become FABULOR,
REGULA, DILIGO, NOBILIS; but DUBITO  becomes
DUBITO: MANEO, SEDEO, FOVEO, F-UGO, become
MAINEO, SEDEO, FOVEO, FUGIO; but FRIGEO becomes
FRIGEO.
No one, it is presumed, who is familiar with the
English method of pronouncing, will deny the justness of this representation of the rules and practice
of the English system. But to remove every possible doubt on this point, let us hear Mr. Walker
himself, whose rules for pronouncing Greek and
Latin proper names have been generally adopted
in this country and in England, and whom we cheerfully concede to be the highest authority for the




ROMAN ORTHOEPY.                67
mispronunciation of Greek and Latin words the
world has yet produced.
"Every accented antepenultimate but v, even
when followed by one consonant only, is, in our
pronunciation of Latin as well as of English, short;
thus FABULA, S:EPARO, DILIGO, NOBILIS, CUCUMIS, have
the first syllables pronounced as in the English
words, capital, celebrate, simony, solitude, luculent,
in direct opposition to the Latin quantity, which
makes every antepenultimate vowel in all these
words but the last, long; and this we pronounce
long, though short in Latin.  But if a semi-consonant diphthong succeed, then every such vowel is
sounded long but I in our pronunciation of both languages; and EUGANEUS, EUGENIA, FILIUS, FOLIUM,
DUBIA, have the vowel in the  antepenultimate
syllable pronounced  exactly as in the  English
words satiate, menial, delirious, notorious, penurious; though they are all short in Latin but the I,
which we pronounce short, though in the Latin it
is long."
In all these  accented  antepenults, then, Mr.
Walker exp2ressly requires every long vowel to be
shortened and every short vowel to be lengthened.
As another illustration of the inherent and insuperable difficulties under which the English system




68             ROMAN ORTHOEPY.
labors in respect to the giving of Latin quantity,
take the verb FUGIO in its four principal parts;
FUGIO, FUGERE, FUGI, FUGITUM. If now the attempt be made to distinguish the long and short
sounds of the u, pronouncing the former as in tube,
and the latter as in tub, then according as the G is
made hard or soft, we shall have either FUG-1-O,
FUG-E-RE, F;-JI,  FU G-I-TUM, or FUDG-I-O, FUDG-E-RE,
FU-JI, FUDG-I-TUM. But this is a little too fudg-y
even for our English pronouncer, and so, in perfect
consistency with himself and his intelligent guides,
he concludes to violate the quantity in three cases
out of the four.
The proper quantity of final syllables and monosyllables is rarely given. The latter are generally
pronounced short except where they are short,
e. g., 5s, the mouth, is made short. NIL, coming
by syncopation and contraction firom NiHmL, is made
short, but the short first vowel of the full form is
changed into a diphthong, naikil; thus the word,
at one time, is robbed of half its quantity, and, at
another, has one half more than its due. On the
other hand, QVE, YE, NE, TE, etc., are made long.
Es, however, whether it be a final syllable or a
monosyllable, is always long, whatever its real
quantity.




ROMAN ORTHOEPY.               69
Again, this system of capriciously changing both
the quantities and the sounds of the vowels according to their position, becomes most ludicrously conspicuous in the inflectional variations of words. In
the declensions of nouns, the radical vowel, which
has and should receive the same quantity and sound
throughout, is changed from short to long or long
to short and bacl again with startling rapidity and
frequency, the sound also varying with the quantity. Os, the mouth, goes thus: os, ORIS, ORI, 5s,
OS, ORE, etc.; MILES, goes thus: MILES, MILITIS,
MIL1TI, MILITE, M MILES, MILITE, etc.: the plural of
MITIS, mild, goes thus: MITES, MITES, MITIA; MITIUM,
MITIUM, MITUM; MITIBUS, MTIBUS, MITIBUS; MITES,
MITES, MITIA, etc., pronounced MAI-TEZ, MAI-TEZ,
MISH-Y-A; MISH-Y-UM, MISH-Y-UM, MISH-Y-UM; MIT-YBUS, M3T-Y-BUS, MIT-Y-BUS; MAI-TRZ, MRI-TEZ, IMAI- SH-Y-A,
etc. So also in the variations of verbs. Where
the quantity and sound of the radical vowel are
really unchanged, they are constantly changing in
our pronunciation; and on the other hand, where
the quantity is actually changed, distinguishing contracted and uncontracted forms or indicating difference of tense or action, we neglect the distinction,
and make the different quantities alike long or alike
short. Sometimes we even reverse the facts, making




70             ROMAN ORTHOEPY.
the short long, and the long short, Thus in the
imperfect tenses, LEGO, LEGEBAM, the short E of the
present is made long; and in the perfect tenses,
LGi, LEGERAM, the long E of the pluperfect is
made short. We make no distinction between the
E of PES and that of PEDIS. In BOS, BOVIS, the o
of the nominative, which has become long by contraction, is made short, while the short o of the
genitive is made long. In VENIO, VENIRE, VENI,
etc., we make no distinction between the short E
of the present and the long E of the perfect. And
still further; as, on the one hand, we lengthen the
short E in most of the imperfect tenses, so, on the
other, we shorten the long E in all the perfect
tenses except the dissyllabic VENI.;- Such are some of the many ways in which the
English system of Latin pronunciation necessarily
violates the true quantities of the language, and is,
therefore, fatal to the proper rhythm of Latin verse.
These multitudinous and glaring corruptions, which,
against many earnest protests, gradually invaded
and finally supplanted the older and purer English
method of pronouncing Latin, are now, as we have
seen, systematically inculcated in our Latin grammars and other elementary works. But no ingenuity of man can harmonize the parts of this sys



ROMAN ORTHEOPY.                71
tem. The rules laid down for the sounds of the
letters are often, and of necessity, in direct conflict
with those given by the same authority for the
quantity of syllables, so that, if the pupil gives the
prescribed sound he will violate the quantity, or if
he gives the quantity, he will violate the rule for
the sound.  Numerous illustrations of this fact
might be drawn from  the Latin grammars most
commonly used in our schools and colleges. One
or two will be sufficient for our purpose.
" I is long in the first syllable of a word, the
second of which is accented: 1st. When it stands
alone before a consonant.  2d. When it ends a
syllable before a vowel."
The example adduced in illustration of the first
part of this rule is IDONEUS, in which the initial I is
short. With the second, viz., I is long when it ends
a syllable before a vowel, compare the first general
rule of quantity, "A vowel before another vowel is
short."
Again, " When a syllable ends with a consonant,
it has always the short English sound."
The student who reads this brief and comprehensive rule, and understands the meaning of words,
must consider the whole question for ever settled
with regard to the quantity of a syllable ending in




72             ROMAN ORTHOEPY.
a consonant. But what is his surprise when, in the
chapter on Quantity, he observes a long succession
of rules with numerous exceptions on this very subject of the quantity of syllables ending in a consonant. And his wonder and perplexity are still
further increased when he reads the following rule
of quantity: "A vowel naturally short before two
consonants is long."
Now what is the actual effect of this corrupted
system  of pronunciation  upon the English and
American method of reading Latin verse?  It is
obvious that the real quantity, as indicated and
determined by the sounds, is entirely disregarded.
The line is simply divided into the proper number
of feet, and the foot into the proper number of
syllables. These syllables are assumed to be long
or short, according to the exigencies of the case,
though  they may be sounded of the  opposite
quantity, and then the rhythm is secured by giving
the ICTUS to the proper syllable of the foot. This
system of scanning, then, though professedly resting on quantity, is really, like the English, based on
accentuation. For example, take the 10th and 11th
lines of the first book of the XEneid:
INSIGrLNEM PIEITATE VTIRUM, TOT AIDIRE LAIBORES,
IMPPULEIRIT. TiNITAENE ANIIMiS COEILESTIBUSIIRAE?




ROMAN ORTHOEPY.                 73
In the first foot of the first line, the two long syllables are pronounced short; in the second, the long is
made short and the two shorts, long; in the third,
the two shorts are made long, the short I in VIRUM as
well as in PIETATE, being also changed to a diphthong; in the fourth, the long is made short; in
the fifth, the two shorts are made long: in the first
foot of the second, the long is made short, and the
two shorts, long; in the second, the two longs are
made short; in the fourth, the first long is made
short; in the fifth, the long is made short.  In the
first eleven lines this English method requires the
violation of Latin quantity, as indicated by its own
rules and sounds, about seventy times. The rhythm
is made to depend solely upon raising the voice in
the ARSIS, and lowering it in the THESIS of the foot.
Closely connected with this, and a natural consequence of it, is the almost universal and yet utterly
abominable practice of separating the feet instead
of the words by slight CAESURAS. Each word does
not stand by itself as it should, and as it does in
prose, but the syllables which constitute the foot,
though forming parts of different words, stand by
themselves.  Whenever, therefore, the foot ends in
the middle of a word, as it frequently will in all
4




74            ROMAN ORTHOEPY.
melodious verse, the parts of the word are improperly separated by a CAESURA.
I claim  now to have fully substantiated the
charges brought against the present English system
of Latin pronunciation, and shown it to be liable
to very grave objections. We have also seen that
the so-called " Continental" method is entitled
neither to the authority implied in that name, nor
to any respect based on grounds of intrinsic worth;
and we may safely take it for granted that the
scholars of England and America will not servilely
copy the usage of any single nation on the Continent, as France or Germany, Italy or Spain. Finally,
we have found the original Roman system of pronunciation fully developed by the researches of
the learned, and established by complete and satisfactory evidence. And the question naturally arises,
can this system, in theory so beautiful and so authentic,
be revived in practice? This question has sometimes been answered negatively by persons who
maintain that the man who has, in substance, the
true theory of the Romans in pronouncing their
language, and attempts to carry it out in his own
pronunciation of the Latin, may, in many instances,
depart fiom  the actual practice of the Romans.
Possibly he may; it is hardly possible, we know,




ROMAN ORTHOEPY.               75
for a living language to maintain its original form
in all respects unchanged through successive centuries. But yet, in view of the exceeding simplicity
and symmetry of the Roman system, he has, I
think, a better reason to suppose that he does not,
than anybody else has to suppose that he does.
And besides, whether in the application of the
theory, he does or does not vary, occasionally, from
the actual usage of the Romans, is a question nearly
as immaterial as it is manifestly undebatable.  Of
course, no man can know or reasonably pretend
to know, whether his pronunciation of the Latin
corresponds precisely with that of Cicero or Casar,
but with the well authenticated and simple theory
of the Romans before him, he may be reasonably
assured that his pronunciation is substantially identical with theirs. He may not, for instance, be
sure whether he gives to F the same degree of
aspiration  that the  Romans did; whether he
sounds the v consonant, or the rare and difficult
combinations EI and Eu, precisely as they did;
but he may be certain that he sounds all the
vowels and all the common diphthongs as they
did; that he makes the same consonants gzttturals,
linguals or labials; mutes, liquids or spirants;
the same mutes, soft, hard or rough which the




76             ROMAN ORTHOEPY.
Romans did; he may thus know that he has the
full idea on which the entire Roman system was
based, and may be reasonably assured that his
actual practice is substantially identical with the
Roman.
But, admitting the possibility of restoring the
true pronunciation, can not all the advantages of
its restoration be secured without the inconvenience
and singularity of a change?  Why not, it has
sometimes been asked, teach the Roman pronunciation in theory for purposes of comparative philology, but follow the English system in practice, to
avoid being singular? In this position we have, I
think, two unwarrantable assumptions.  1st, that a
mere theoretical knowledge of the Roman pronunciation is all we need for the successful study of the
relations of the Latin to its sister languages, and
2d, that the study of Comparative Philology is the
only thing to be promoted by our knowledge of
Roman orthoepy.  Admitting, for the moment,
the truth of the second assumption, viz.: that Comparative Philology is the only thing to be affected
by our knowledge or ignorance of Roman orthoepy
-is it true, we ask, that a merely theoretical knowledge of the Roman pronunciation is amply sufficient for all purposes of comparative philology?




ROOMAN ORTHOEPY.               77
It would seem to me that the mere statement of
this question is amply sufficient for the refutation of
this assumption.  The mere knowledge of the
Roman theory of pronunciation without the habitual
practice of that pronunciation is, to my mind and
in my experience, of comparatively little utility in
respect to the study of the foreign affinities of the
Latin language. I maintain that the Roman pronunciation becomes valuable to the student of the
character and relations of the Latin just in proportion to his familiarity with it. It is the constant
speaking, and reading, and study of the language,
on the Roman system of orthoepy-it is this accustoming of the ear to the Roman sounds, and this
alone, I had almost said, which sheds a continuous
and ever increasing light on its foreign relations;
and opens successively new and attractive vistas
into the wide field of comparative philology. It
must be so from the nature of the case. The affinities between the words of sister languages lie especially in their radical sounds, i. e., in the sounds of
their root-letters. While these sounds may be substantially identical-sounds which the sister nations
had in common long before they had a written
language, and even when they were yet, in fact,
but one nation-the alphabetical representations of




78             ROMAN ORTHOEPY.
those sounds adopted by the separated members of
the one original family after the time of their separation, may be, and in fact are, widely diverse.
tIow different, for example, the a2ppearance of the
Saxon tomb and the Latin TUMULUS; of the Saxon
kink and the Latin CINCTUM; and how  different their sounds, also, on the English method of
pronouncing Latin. But, on the Roman method,
the two Saxon words have, respectively, the precise
sounds of the roots of the two Latin words, TUMulus and cINc-tum.  Now which of the two, I
ask, will the more readily discover this affinity, the
man who, though informed of the Roman theory,
practices the English; who habitually gives to
those Latin words their foreign, false, unrelated
sounds, or, the man to whose ear the true Roman
and related sounds are perfectly familiar?  The
question needs no formal answer. To study with
advantage, then, the relations of the Latin to other
languages, the real sounds of the language must be
familiar to the ear; and they never can become so
except by their being familiar on the tongue.
But is there any truth in the second of the above
assumptions, viz., that comparative philology is the
only thing to be affected by our knowledge or ignorance of Roman orthoepy? Very far from it. An




1RO1MAN ORTHOEPY.               79
auricular familiarity with the Roman system  of
pronunciation, I deem to be quite as valuable an aid
in the study of the language itself, as in that of comparative philology, and, therefore, of far greater
practical value, inasmuch as few persons will have
leisure to carry their investigations into the field of
Comparative Philology, while thousands are engaged in the study of the Latin, and interested in
having their knowledge of that language as accurate and complete as the researches of the learned
can make it. If I may be allowed to refer to my
own experience, I would say that I am  filly conscious of having made more real progress in the
study of the language itself, as well as of its affinities, during the first five years in which I practiced
the Roman pronunciation, than I did by the harder
study of the preceding ten years in the use of the
English system; although then not ignorant of the
principal points of difference between the Roman
and English methods of pronouncing Latin. And
from  the very nature of things it could not be
otherwise.  In the study of every language, it is a
matter of vast importance t    v  to have the right pronunciation, the trite vocal sounds of the language,
not merely in the head, but habitually and familiarly on the tongue.  Truth here, as everywhere,




80              ROMAN ORTHOEPY.
is, in itself, an argument sufficient to outweigh a
mountain of objections.  Nor do I see how a professed teacher of Latin can answer it to his conscience or sense of honor, deliberatety and elaborately
to train his pupils in a way which, by his own confession, misrepresents the form of the language and
will as certainly belie its spirit. For the true
vernacular pronunciation of a language is necessarily homogeneous with its structure-it is part and
parcel of the genzius of the language. Having been
developed and established simultaneously with the
growth and maturity of the language itself, they
must, of necessity, be so thoroughly harmonized
and adjusted to each other, that they can not be
severed without great practical disadvantage and
loss. I think it may be laid down as a universal
truth, that no student of any language can possibly,
by any ingenuity, discover and comprehend the real
structure and genius of the language as fully and
clearly on a foreign and false, as on the true, vernacular system of orthoepy.  I am persuaded fiom the
experience of twenty-four years in teaching Latin,
seventeen on the English and seven on the Roman
system, that I can teach the important principles of
the language far more successfully with the true
than with the false pronunciation. I have given




ROMAN ORTHOEPY.                 81
the two systems a fair trial with no interest but to
ascertain the truth; and I not merely think but I
know that, by the daily use of the true pronunciation, I can secure, on the part of the student, a
much more intelligent and lively interest in questions pertaining to the etymology of the language;
to its various inflectional forms and laws; to its
quantities; and, above all, to its metrical system
and its relations to kindred languages. The only
and the all-satisfactory explanation of these facts is
simply this, that over all the important questions
connected with the structure and ethnical relations
of the Latin language, the English method of pronouncing it, so far as it differs fiom  the Roman,
spreads a cloud which the student must himself
clear away, or with difficulty penetrate, in order to
any considerable success in his studies, while on
these same questions the Roman  pronunciation
throws a strong light.  The false pronunciation
necessarily darkens and obstructs the student's
pathway; the true one illuminates and clears it.
The man who, in the use of the former, has achieved
much in Latin philology, would have achieved
much more in the use of the latter.  What he
has accomplished has been done in despite of adverse influences.  He has been rowing his Latin
4*




82             ROJMAN ORTHOEPY.
boat ADVERSO FLUMIArXE instead of FLUMINE SECUNDO.
As a preparation for a proper observance of
Latin quantity and metre, the position which I am
controverting, viz., teach the Roman system  of
pronunciation, but practice the English, is just
about as rational as it would be to expect that, by
teaching a child the theory of virtue while you
habituate him to the practice of vice, he will ultimately become a virtuous man. It is not mere
theoretical principles, but well established moral
habits which control the life. The wise man does
not say, " Teach the child" (theoretically) "the
way he should go and when he is old," etc., but
" train up the child."  Form him to right habits.
The same geieral principle is involved in this matter of Latin quantity and measure. Accustom your
pupil to the English system of pronunciation and,
no matter what you teach him, you are forming him
to wrong habits. You are " training him up" in the
way he should not go, through the noble measures
of Horace and Virgil, Ovid and Juvenal. On this
point, I shall have a word or two more to say byand-by. But of this one thing let every teacher of
English-Latin be assured. He can not continue to
impose upon his pupils a system of pronunciation




ROMAN ORTHOEPY.                83
so entirely incongruous with that which belongs to
the language, and is really and essentially part of
itself, and save them from the confusion and error
which that system necessarily involves. We may,
if we choose, through indolence or timidity, refuse
to abandon the worse for the better method; but
let us do it with our eyes open. Let us make up
our minds clearly on the only alternative that is
before us, either to forego the advantages attending
on the truth or fiankly to abandon and boldly to
repudiate the falsehood.
The most important reasons for adopting the
latter branch of this alternative have already been
foreshadowed in the preceding pages, and some of
them have been so fully developed as, perhaps, to
need no further illustration. I shall content myself,
therefore, in conclusion, with a recapitulation of the
main points of the argument, with such expansion and such additional considerations as may seem
requisite.
I advocate, then, the adoption of the system of
Latin pronunciation above described, in lieu of
every other system, and especially of the English:
1. Because it is the true.Roman pronunciation,
and, therefore, in full harmony with the entire
structure and genius of the language. This fact




84             1ROMAN ORTHOEPY.
alone is, in my judgment, a sufficient reason for the
course proposed.
2. Because it is more simple, regular and philosophical than the English system. Each alphabetical
character represents but one articulate sound; and
these sounds, whether vowel or consonant, are
easily classified or combined in accordance with
clear, definite, and invariable-phonetic laws.
3. Because it is more euphonious and beautiful
than the English system. Let any man of candor
and good literary taste go through, for example,
with the forms of REGO, REGERE, JUNGO, JUNGERE,
first on the Roman plan, giving the E, u, and G their
appropriate sounds throughout, and then on the
English plan, changing frequently and radically the
sounds of the vowels, and also that of the G from a
guttural to a lingual spirant and back again; let
him decline, on the two plans, AMICUS, ROGUS, the
plural of MITIS, FELIX, etc.; let him pronounce, on
the two plans, such words as JUSTITIA, AUDACIA,
CILICIA, GALATIA, INIMICITIA, INSCITIA, OCEANUS,
OFFICIUM, ROSACEUM, OTIUM, NEGOTIUM, INITIUM,
etc., etc.; let him, I say, in these and hundreds of
other like cases, carefully compare the two systems
in respect of their euphony, and then defend the
English system if he can. It is not uncommon for




ROMAN ORTHOEPY.                85
persons who have been accustomed only to the English sounds, and whose classical scholarship is not
sufficient to enable them to appreciate the intrinsic
merits and defects of the two systems, to attempt
to cast ridicule upon the strange words revealed by
the Roman pronunciation; such, e. g., as those in
which c and G occur before E, I, Y, AE, and OE.
Ridicule is the most available weapon in a weak
cause. But as well might a rude Thracian have
laughed at the polished discourse of the sage of
the Athenian Academy; as well might a driveling,
reeling inebriate, meeting a sober and upright man
and fancying him to be the stammering staggerer,
sneer at his really clear speech and steady gait, as an
English-Latinist cast ridicule upon the pronunciation
of a Roman-Latinist. The Roman system of Latin
pronunciation must be pronounced, by every competent judge, to be intrinsically far more euphonious
and dignified than the English.
4. Because it always distinguishes words of different orthography and signification by their sounds,
while the English system very frequently does not;
as, e. P., CENSEO, CENSIO, SENTIO; CENSUS and
SENSUS; CERVUS and SERvUS; CAEDO, CEDO and
SEDO; CICER and SISER; CEU and SEv; CELLA and
SELLA; CITUS and SITUS; scis and sis; SCIN (for




86             ROMAN ORTHOEPY.
SCISNE) f1and SIN; AMICI and AMISI; CIICiLUS and
SURCULUS; etc., etc. Now  when we make such
radical changes as these, it is not a mere change in
the pronunciation of a word but an actual substitution of one word for another. When we say that
SENSUS means a valuation (cENsus) we state a falsehood; SERVrU  signifies not a stag (cERvus) but
a slave; SELLA, not a room (CELLA) but a chair;
SEDO, not I smite (CAEDO), nor I yield(cEDo) but I
soothe; etc., etc.
5. Because it diffuses universal light while the
English system spreads more or less darkness over
the etymology of the language, and over the entire
subject of vowel, diphthongal and consonant combinations and changes.  This point has already
perhaps been sufficiently illustrated. I will only
say here that the instances, in which the one system
illuminates and the other obscures questions respecting the derivation, composition or orthographical variation of words, occur on every page of
every Roman author.
6. Because it not merely throws much light on
the subject of Latin versification, but is'the only
system  on which Latin poetry can be correctly
read. That this statement is true in regard to
Latin quantity has already been shown. It is also




ROMAN ORTHOEPY.                 87
true in two or three other respects.  Whenever,
for example, the letter c is left at the end of a word,
after a final vowel or final n with a preceding vowel
has been cut off before a word commencing with a
vowel, the c, on the English system of pronunciation, is often improperly transferred fiom  its own
to the following word, or changes its sound fiom in
to s or s to i; but on the Roman system, the
c always remains with its own word, and its power
is unchanged.
So again, QV, when similarly left at the end of a
word, must, on the English system, be connected in
syllabication with the following word. And thus
the QV of one word unite with the first syllable of
the next to form  either another word or what is
no word at all. Every man who reads Latin poetry
with the English pronunciation always commits
this barbarism. He can not do otherwise. Because
the sounds of KW represented by QV on the English
system, can not be uttered at the end of a syllable,
and must, therefore, in all such cases, be severed
firm  their own word and connected with the following. Take, e. g., the following dactylic hexameter from Juvenal:
" SED QVQVAMQA IN MAGNIS OPIBUS PLUMAQVE PATERNA."
The final M of the conjunction with the vowel




88             ROMAN ORTHOEPY.
preceding are cut off, and on the English system
we must read
SEDI QVAMll QV' IN1 MAAGIjNIs, etc.
The hearer woull, of course, take the second
word to be QVAM (kwamn) and the third, QVIN (kwin.)
But, on the Roman system, the QVAMQV' sounds like
KAMK; the final QV, then, is sounded with its own
word; the preposition IN stands properly by itself;
and the Roman hearer would know at once from
the sound KAMK that the word was QVAMQV.AM with
the AM cut off by ecthlipsis. Take again Virgil's
" MULT A QVOQVE ET BELLO PASSUS, DUM CONDERET URBEM."
On the English system, we necessarily divide as
follows, viz.:
MnLjTAl QVOllQv' ETI BELjIILO  PiASJSUS, etc.
In this reading, the third word QVET (kwet) is
just no word at all. But on the Roman system,
we divide thus:
MuLITAX QVOQV[II ET BELjIIL. P1ASlISUS, etc.
(K6K)
Here again each word stands properly by itself,
and the hearer knows that the second and third
words are not iwo and KWET, but QVOQVE (KOKE)
and ET. The QVOQV' remains short because the QV
never form position.




ROMAN ORTHOEPY.                89
So, a few lines below, we read, on the English
system:.... STU}DI|jIS{ QV' ASlIPERiRjIMAll BELILI;
but on the Roman system,.... STU'lDIlISQV'j AljSPLRljBIMAjl BELILI. I
(STU|mD1|IISK)
So again, in Horace, we read the following lines
on the Roman system, thus:
NON| EIGETI MnAURljISI JAlcU|LIsl| NEQV'I A4RCU;ll|
(NEK)
but on the English system,
NON NJ IGETI MAulIRIls| JACIU[LISII NEIQV'!AR[CJ.
If there were any truth in spirit rapping, no
man, in reading this passage, would utter that word
QVARCU or NEQVATRCU (neckwarcu), without getting,
instanter, a terrible rap fiom the spirit of my fiiend
Horace; but as there is not, I feel myself in duty
bound to do a little honest rapping for him. The
instances in which QV is thus left at the end of a
word by the cutting off of a final vowel or final AM,
EM, Im, OMI, or Uw, befbre a word commencing with
a vowel, are innumerable, and on the English system, therefore, we are forced, in the reading of
Latin poetry, to the perpetration of innumerable
barbarisms.




90              ROMAN ORTHOEPY.
7. Because it facilitates immensely the study of
comparative philology, especially when prosecuted
with a view to the illustration of the Latin language.
In tracing the affinities among different languages, we must examine especially the roots of
words, and these roots in respect of their elementary souncls, not of their different modes of alphabetical representation; and in order to pursue this
path of inquiry with safety and success, we must
know accurately and familiarly the affinities of articulate sounds and their natural and legitimate
changes and interchanges.  The habit of judging
of words by the eye rather than the ear-a habit
engendered among cultivated nations by the use of
a written language-greatly obstructs that quick
perception of the affinities of sounds which has
been observed to exist in rude, unlettered tribes.
This habit of comparing words by their outward
appearance, and not by the sounds of their radical
elements, often becomes so strong and fixed that
the substitution of one character for another of the
same class or organ, a mere vowel change, or a
transposition of the root-letters, is sufficient to hide
from our view the most perfect affinities.  And further, the obscurity and confusion in which the ir



ROMAN ORTHOEPY.                91
regularity of the English alphabet tends to involve
all our ideas respecting the nature and relations
of articulate sounds-an irregularity which, on the
one hand, fiequently either assigns to one and the
same character elementary sounds uttered by different organs or with different degrees of articulation, or, on the other, employs a variety of characters to represent one and the same sound-present
to the English and American student still greater
obstacles to the successful study of comparative
philology. The only remedy for both these evils
-the obtuseness of our perception of affinities and
the vagueness and confusion of our ideas of affinity-is to teach the English alphabet in such a way
as to give an early and familiar acquaintance with
the nature and legitimate mutations of articulate
sounds; and the true Roman system  of notation
furnishes the best possible basis on which to impart
this instruction and effect this remedy.  If we were
taught, fiom our first entrance upon the study of
Latin, to classify the letters, especially the consonants, according to their different organs of utterance or degrees of articulation, and were uniformly
accustomed to a close observation of the relations
of letters and to the examination and comparison
of words by their radical sounds and not by their




92             ROMAN ORTHOEPY.
mere outward forms, these interchanges of sounds
of the same class, so frequent in different dialects
of the same language and different languages of the
same family would no longer escape or surprise us.
We should then readily discover, where we should
otherwise naturally overlook, such affinities as the
following, e. g., Latin GeLiDus, German KaLT,
English CoLD; Greek BPeM(o, Latin FReMo;
Greek'PiPog, Latin FRiGus; Greek PrrvvFcu, Latin
FRanGo (root FRaG), English, BReaK; Latin
ViGilo,  German   WTaCIen,  English  WVaKe,
WatCH; Latin DoCeo, Anglo-Saxon TaeCan, English, TeaCH;* Latin N~ox (root, NoCT), German
NaCHT, English NiGHT; Latin TRaho, English
DRaw, DRag; German DuRST, English THiRST;
Latin PaNDo, English BeND; Latin GeRo, En
glish CaRry; Latin FeRo, English BeaR; etc., etc
Now it is obvious that a very large proportion
of the affinities existing between the Latin and the
other members of the Indo-European family of languages are discoverable only on the Roman system
of pronunciation, since the affinities themselves ex.
ist only between the Roman sounds of the rootletters and the radical sounds of the corresponding
* The Anglo Saxon c has here, as elsewhere, suffered the
same corruption in English which the Roman c has in Italian.




ROMAN ORTHOEPY.                93
words in the cognate tongues. It follows, therefore, that, in the study of the foreign relations of
the Latin language and of comparative philology in
general, the man to whose ears the Roman paonunciation is entirely familiar, has immensely the advantage of him  who is unfamiliar with it. And
such familiarity will be acquired, on the part of the
English or American student, only by the constant
practice of the Roman system.
Instances in illustration of the affinities between
Latin words, and words of similar sound and import in the kindred languages-affinities which are
apparent on the Roman pronunciation, but obscured by the English-might be adduced almost
AD INFINITUM. Take, e. g., the following Latin
and Greek words: ACESIS and daeao'S, a kind of
borax; ACOETIS and a/cotv', a concubine; ACIES and
adct, dac, a point or edge of any thing; CEDRUS
and Kedpo', a cedar tree; CEDRIS and tedptL, the
fruit of the cedar; CELAENAE and KeOcLvat; CELLO
(from which ANTECELLO, etc.) and KiE^2, to move,
put in quick motion; akin to which are, CELER,
quick; CELES, a race-horse; CELOX, a fast-sailing
yacht, and IKcf, quick, and also (sc. "iTrTro~) a racehorse, and (sc. vavg) a little boat, pinnace or yacht;
CERA and Kqrpbo, wax; CENTRUM and Kevrpov; cE



94              ROMAN ORTHOEPY.
RASUS and cpadaog, a cherry-tree; CERCCTRUS and
KepKovpog, a light sailing-vessel; CIBRIUAiM and ctc5ptov, a drinking-cup; cici and KiCl, the castoroil plant; cicLNus and Kilivog; CIRCAETUM and Ktpacdov, a mandrake; CISTA and KiaoTl, a box or chest;
CITHARA and KtOcdpa; CIRRHA and Kippa; CIRCE and
Kiplcq1; CIRCAEUS and Ktptcato, of or belonging to
CIRCE; CIRCINUS and citpicvog, a pair of compasses;
CETE and  I7rT], a whale; COELUMa and Icoltof (English, hollow); CINYPS and Kivvib; CINEAS and
Ktveag; CELTAE and K/l-Tat; CILICIA and KtLcfia;
CINAEDUS and Kivatdog, an unchaste person; CITTIUM and KirTL'ov; GEMO and yeGOt, to groan; GENESIS and yTveat; GERRAE and yeTpa; GETAE and
rETal; GYPSUM  and yvpbog; GYRUS and yTpog;
GYTHIIUMJ  and rvIOtov; SCILLA, or SQVILLA, and
atigAa, a sea-onion; SCIURUS and aiciovpog, a squirrel; QVE and ial; CESTUS and  ~cearTO, a girdle;
CENSUS and Icvaog, a tax; CINCINNUS and citctvvog,
a ringlet of hair; CIccus and icioS, the fruitcore; CIO and irio, to move, etc.; CIRCus  and
iipKog, etc., etc.: or the following Latin and German, or Greek words: AG-ER and ack-er, a field;
AQV-A and ach, water, as in.Biberach; CELLA and
keller, a room; coQvus or cocus and koch, a
cook; VACILLARE and woackeln, to waver, totter;




ROMAN ORTHOEPY.                95
MURus and mauer;  ufs and nmaus; SUG-ERE and
saug-en, to suck; CISTA and kciste; CICElz and kicher,
a chick-pea; COENA (IoZvoo, common) and koena,
which, in some parts of Silesia, is the word for the
common meal; CARCER, Greek Kidpapov, German
kerker, Welsh carcar, a prison; CAE SAR and KJaiser;
CEVA and kuh, English cow; NATURA and nattr;
COELUM  (ioliXov) and hohl, hohle, hollowo; VICIA
and wicke, a vetch; CEIASUS and kirsche; JUG-UM
and joch, a yoke; AC-IEs and eck-e, a point, corner; ACER (FAcER) and wacker, fierce, valiant;
VIGILO and wack-en; FAC-IO and mach-en; HESTERN-Us and gestern, etc., etc.: or the followinoLatin and Saxon, or Welsh words: CINC-TUM3 and
kink; TUMI-ULUS and tomb; PUp-pIS and poop, the
stern of a ship; CISTA and Welsh cist; CRoc-IO
and croak; SCINC-US and skink, a kind of lizard;
ALC-IS and elk; GEN-US and kin, kind, as a noun;
etc., etc.
8. Because it greatly augments the incidental
and collateral advantages connected with the study
of the Latin language. The value of a thorough
knowledge of the Latin to the student of general
philology is universally conceded, and can hardly
be over-estimated. Constituting, as it does, the
basis of all the languages of south-western Europe,




96             ROMAN ORTHOEPY.
and being, at the same time, in its essential elements, most intimately allied to the Teutonic and
Celtic families-so intimately that some distinguished scholars have been disposed to consider it
a daughter rather than a sister of the German, and
others, to regard it as simply a dialectical development'of the Celtic language; and others still,
as Reinhold Klotz, have more recently, and with
more correctness, considered it to be, in its original and essential elements, Celto-Germanic, but
developed under a controlling Grecian spirit and
influence-sustaining, I say, these intimate relations
to the Celtic, Teutonic, and Romance languages, it
forms the connecting link between them; while
many of these again, Celtic, Teutonic, Latin, and
its daughter, the Norman-French, have each contributed largely to the formation of our own language; so that a correct knowledge of the Latin
becomes indispensable to the full understanding of
either the English or any of the south-western European tongues.  It is obvious, then, that the use
of the Roman system  of pronunciation-a system
which is itself an integral and vital part of the
genius of the language, and as such absolutely essential to the full comprehension of its structure
and spirit, and which is also closely allied to the




ROMAN ORTHOEPY.               97
vocal systems of the cognate languages —must
greatly enhance all the incidental and collateral
advantages to be derived from the study of the
Latin, and this in two ways; 1st, indirectly, by imparting a far more correct and intimate knowledge
of the Latin itself; and, 2d, directly, by preparing
the way, especially to the English and American
student, whose vernacular orthoepy's so alien from
that of the kindred tongues, for the more ready
and thorough acquisition of the languages of modern Europe. Were we to study the Latin on the
Roman system of orthoepy, it would be found that
the shortest road to a thorough knowledge of either
of the more cultivated continental languages would
not only commence with and run through the Latin,
but, even after leaving the Latin, would be greatly
smoothed and facilitated, to its very termination,
by a familiar acquaintance with the true pronunciation of the Latin.
9. Because it is the only system on which a uniform pronunciation can ever be secured even in our
own country, to say nothing of uniformity among
the different modern civilized nations.  In our
country there is, at the present time, a great diversity of systems.  "A very few," says that accomplished Latin scholar, the late Robert Kelly,
5




98             ROMAN ORTHOEPY.
LL.D., in the article above mentioned, " employ
the continental vowel sounds, but let the diphthongs and certain consonants suffer. Some give
the true sound of the letter A, but give E the
sound of I, I the sound of the English i in pine, (E
the sound of E, etc. Others are still more barbarian, giving A  the sound of E, AE  and (E the
sound of I, with other atrocious aberrations from
the true standard. The ears of persons trained in
one of these schools are offended with the pronunciation given by the others, and would be offended
in like manner if the true pronunciation were given.
Indeed, we have heard of instances where the professor has assailed with ridicule pupils who came to
him with a -Rornan pronunciation."  In immediate
connection with this he asks, " Why can we not,
with one consent, abandon our erroneous modes of
pronouncing Latin, and adopt an uniform and correct pronunciation?" And then, after giving the
Roman scheme, he adds: "iNow, what can be desired or imagined simpler than this whole scheme?...... The sounds are all familiar, and can be
learned with the utmost facility. The unlearning
of the various false methods will not be found a difficult undertaldng by any one who will set about it.
The most inveterate barbarian, by reading aloud




ROMAN ORTHOEPY.               99
for an hour or two a day, will in a fortnight metamorphose himself into a veritable' old Roman.' It
will probably be more difficult to persuade scholars
to adopt the hard sounds of c and G where they
now make them soft, than to assent to any other
change required. Some words will strike them
ludicrously. They will recalcitrate against restoring to the great Roman orator his good name, of
which he has been so long and so unreasonably
deprived. But there can be nothing intrinsically
ridiculous or undignified in the pronunciation which
classic antiquity uttered.'Tis better to give to
Scipio and Cicero the names by which they were
called when in tie flesh, and which they invested
with immortal glory-far better all these changes
-than, as Kraitzer hath it,'to turn the pompous
senate of Rome into a mass of hissing serpents.'
The true pronunciation will, as a matter of course,
become to us infinitely more harmonious, dignified
and expressive, than the dissonant utterances by
which we have so long caricatured the noble language.....   Of what importance is the retention of the false methods that within two centuries
have invaded the language?  There is nothing
about them  which entitles them  to respect, and
every American scholar who visits continental Eu



100             ROMAN ORTHOEPY.
rope and attempts to speak or read Latin, is perfectly ashamed of his pronunciation.......  The
irregularity of pronunciation that prevails, the general want of confidence, and the conviction resting
oftentimes, upon the mind of the teacher, of the
falseness of the system  he practices, have a most
unfavorable effect on the whole style of reading
Latin, and not unfrequently upon a rigid observance of quantity.  The facile and correct reading
aloud of the classic writers, is not only a beautiful
and rare accomplishment, but exercises an important influence upon the taste and upon the appreciation of nameless graces and beauties of style."...
"The philological bearings of a correct pronunciation involve considerations of paramount importance in connection with this question. The argument fiom philology will press with irresistible force
upon the mind of the scholar who will examine the
subject carefully."
These eminently scholar-like views and suggestions I have  preferred  to quote consecutively,
though not all bearing specifically upon the present
point, viz., the impossibility of restoring or securing in our country an entire uniformity of Latin
pronunciation on any other than the Roman system.




ROMAN ORTHOEPY.              101
Uniformity on the modern English system  we
have already lost, and can never regain. The hostility of that system to the structure and genius ot
the Latin language, has forced many institutions,
both in this country and in England, to at least a
partial abandonment of it, i. e., to the substitution,
in part or in full, of the German vowel sounds for
the English. These institutions will be very unlikely to retrace their steps. The intrinsic excellence of the German, or, which is the same thing,
\of the Roman vowel system, the facility with which,
on that system, the true Latin quantities can be observed, and the impossibility of habitually observing them  in the English method, the increasing
practical convenience of the German system arising
from our growing intercourse with the scholars of
the Continent, all will unite to prevent such a return to the English vowel sounds. If these institutions change again, they will be morally certain
to advance still further toward ancient Rome.
But can the scholars of this country and of England ever unite on the falsely so-called "Continental" method of pronouncing Latin?  It is obviously impossible. This combination of the German
vowel sounds with the English diphthongal and
consonant sounds, though far better than the En



102             ROMAN ORTHOEPY.
glish system  entire, is, after all, a mongrel system
-a hybrid-itself the product of an unnatural, uncongenial union-and can never form the basis of a
cordial and lperlanent union among English and
American Latinists.  The true Latin scholars of
our country, the genuine lovers of Roman literature, can never harmonize on a system which subjects the really euphonious and symmetrical language of the Romans to the countless dissonances,
irregularities, and serious practical evils occasioned
by the false sounds of the diphthongs AE and OE; of
c and G before E, I, Y, AE and OE; of Qv,, s and T.
A system  liable to such grave, intrinsic objections,
and wanting the poor outward support of national
prejudice and pride, can never commend itself to
general respect and adoption.
Again, we may safely assume, I think, that the
scholars of this country and of England will rever
servilely exchange their own national system for
that of aniy single nation on the Continent.  They
would rec:-lcitrate more instantaneously and violently agai:st saying with the German, Tsits-ero,
or with the Italian Ch/ich-ero, than with the Roman
Ui-kero (1~ec-.7ero, not Tick.-ero).
We are, then, shut up either to all the serious
disadvantages not only of great error but of great




ROMAN ORTHOEPY.              103
diversity in error, or to the adoption of the true
Roman system  of Latin pronunciation. There is
no other alternative.  On this, as on other subjects,
there can be no intelligent and permanent harmony
of view and practice except on the basis of truth;
and where shall that basis be found except in
the Roman system  of orthoepy?  This, and this
alone, forms a purely scientific and philosophical
ground of uniformity. We must have either a
harmonious union in the truth, or disunion and disharmony in error. The true vernacular pronunciation of the Latin language must ultimately prevail
in this country and in England, or we are certainly
doomed to endless diversity.
And why should we hesitate in regard to this
clear, necessary alternative?  We have nothing to
lose, but every thing to gain, by the cordial adoption of the truth. Aside from our own personal
benefits thereby secured, have we not reason to
hope that, if English and American scholars should
generally adopt the Roman system  of Latin pronunciation, the scholars of the Continent would
soon follow the example?  Might we not expect,
in that case, the prediction of Lipsius to be speedily fulfilled?  " AUDEAT ENIIM UNA ALIQVA (GENS)
ET OMNES AUDIENT."  The German and Italian




104            ROMAN ORTHOEPY.
scholars would have but few changes to make;
chiefly those pertaining to the diphthongs and to
the consonant c. If, then, even in Germany, where
the departures fiom Roman usage are so limited,
the practical adoption of the entire Roman system
is reckoned by some of their best Latin scholars a
great desideratum, how much more, in our country
and in England, where those departures are so numerous and wide, and where there is now such a
diversity of systems, how much more is the general
adoption of the Roman system to be earnestly desired; not simply on the all-sufficient ground of its
intrinsic superiority, theoretical and practical, to
the English system, but also as the only means of
securing uniformity among ourselves; of bringing
us into harmony with the main features of the Ger.
man and Italian systems; and possibly, of bringing
them into full harmony with us and the Roman
system?  " Why," then, in the language of Mr.
Kelly, " why can we not, with one consent, abandon our erroneous modes of pronouncing Latin,
and adopt a uniform and correct pronunciation?"
How glorious a day for the study of this noble language would be -that in which a uniform pronunciation should be secured on the true Roman basis!
0 NOCTES COENAEQVE DEUM! 0 when, it just oc



ROMAN ORTHOEPY.               105
curs to me to ask in passing, when shall we Latin
scholars have a system of pronunciation which will
enable us, with one accord, to utter those words as
IHorace uttered them, instead of giving four false
quantities in those eight syllables, and mispronouncing every syllable but one, as on the English system we not merely do, but must do?  That single
and singularly fortunate exceptional syllable is the
interjectional vowel 0, in which, it would seem, the
English system has the sound right, simply because
it is impossible to get it wrong. In the very next
syllable, that vowel receives, as it often does on the
English method, the sound of a in what or o in
not, a sound unknown alike to the Latin, the Italian and the German languages. How long shall
we continue to inflict a barbarous violence on the
elegant and majestic diction of Cesar and Cicero?
of Virgil and Ovid, Horace and Juvenal?  How
long  shall we continue to  say,'MULTA  KWO
KWET BELLO?' and' NoN EGET MAURIS JACULIS
NEKWVARCU?'" etc., etc., etc.
Now all the evils connected with our false methods of pronouncing Latin admit of a ready and an
effectual remedy. We have but to adopt the perfectly simple and beautiful system of the Romans
in pronouncing their language, to make ourselves.5*




106             ROMAN ORTHOEPY.
familiar with it, "at home" in it; and all these
evils will have vanished. Nor shall we long continue in the careful, thoughtful practice of the Roman pronunciation, without finding ourselves much
more at home in the language itself than we have
been or otherwise could be.
The difficulty of making this change is by no
means so great as might be supposed. As already
suggested in the quotation from  Mr. Kelly, the
practice of reading Latin aloud correctly for an
hour or two a day, will soon create, even in those
long accustomed to a false pronunciation, a thorough
distaste for their old method, and an earnest love
for the new. The true sounds are so harmonious
with each other, and with the entire structure and
genius of the language, as to commend themselves
at once to our enlightened judgment, our chastened
and refined ear, and our cordial good will. As we
become familiar with the true system, we shall love
and prize it. WVe shall appreciate, as we have never
done before, the real majesty of the Latin language;
the true dignity and power of Roman eloquence;
the genuine harmonies, and smoothly-flowing numbers of Roman verse.




APPEND I X.
NOTE A.-PAGE 23.
"The sounds of the long and short vowels, though
elementarily the same, were always distinguished in
length.... The sound of the long vowels was
that of the short vowels doubled." —Scheller's Latin
Grammar.
The author of "Living Latin" (London, 1847)
says: " That the Latin vowels have only one sound
each, long or short, is clear firom  Priscian, who,
when he would enumerate the varieties of sound
which they admit, mentions only those of accent
and aspiration, which are merely varieties of its accidents, not of the sound itself."
A.
In nearly all the cultivated European languages,
ancient or modern, of Celtic, Teutonic or Slavonic
origin, the vowel A has an uniform sound like that
of c in father.  Prominent among these languages
are the Greek, German, Latin and its nmodern derivatives, the Italian, French, Spanish and Portuguese, and our own mixed vernacular. The exceptions in the sound of the A are limited to the English
and French.  The Greek A, to which the Latin A




108                 APPENDIX.
is precisely equivalent in power, Dionysius of Halicarnassus describes as being formed with " the
mouth as much opened as possible."  Penningtol,
on The Pronunciation of Greek, says (page 28):
" It seems clear, from  the description of Dionysius,
that this letter was pronounced as we sound the a
in futher.  The modern Greeks so sound it, as do
most, if not all, of the other nations of Europe."  In
full accordance with the description of Dionysius are
the statements of the Latin authors, Quintilian, Priscian, Terentianus, Victorinus Afer, and Capella, who
speak of the vowel as being uttered RICTU PATULO, HIATU ORIS, etc.
E.
This second Latin vowel is heard, according to
Varro, in the bleating of the sheep, which may,
perhaps, be well represented by the final syllable of
the verb o-BEY. It is, therefore, wholly unlike the
English E.  Its Italian power, which, in its short
and long sounds, corresponds to the E and ffof the
Greeks, is uniformly recognized throughout continental Europe.
I.
The vowel I, Victorinus describes as being made
with the mouth nearly closed.  It was recognized,
by both Greeks and Romans, as entirely identical
with the Greek I, which, as Pennington says (page
36), " was sounded like the e in mete. The lnodern
Greeks so pronounce it; and here, again, the English, in differing from  the modern Greeks, differ




APPENDIX.                109
from all the nations of Europe."  Doctor Webster
says, " The English seat retains the Roman pronunciation of SIT-US, that is, SEETUS."
0.
The character 0 was intended to represent the
position of the lips in forming the sound. Its short
and long sounds correspond entirely to the 0 and
Q of the Greeks, in forming which, Dionysius says,
"the mouth is rounded, and the lips disposed in a
circle, and the breath strikes upon the extremitiy
of the lips."  The sound of o in not, therefore, or
a in what, which is often represented to be like the
short sound of the Greek, Latin and German o, is a
totally different sound, made in a different part of
the voice-passage, and with the lips in a very different position —a sound unknown alike to the Greek,
Latin, German, Anglo-Saxon, Italian, Spanish, Portuguese, etc.
U.
The vowel U was uttered, as described by Capella, with the lips not only rounded, but protruded, and was articulated in the lips. Its sound is
heard in the lowing of kine. Hence the verb MUGIRE, to low, to bellow, is derived from the sound
made by kine in lowing, and represented by the
letters MU, mo-. This vowel, therefore, has invariably the sound of oo, long or short, never that of
the English you. Hence the Greeks, in transferring
Latin proper nouns into their own language, always




110                APPENDIX.
express the Latin U by their diphthong OT; and the
Romans, on the other hand, in translating Greek
proper nouns, always employ their U to represent
the Greek diphthong.
NOTE B.-PAGE 28.
"AI syllabam, cujus secundam  nune E literam
ponimus, varie per A et I efferebant, qvidam serm
per ut Graeci; qvidam singulaliter, tantunm curm in
dativum vel genitivuni casum incidissent."-QVINTIL., 1. i., C. 7.
" Ea qvae nos per AE, antiqvi per AI scriptitaverunt, JULIAI CLAUDIAI. Et nihil obstat, qvominus
hoc ant illo modo scribamus in utroqve numero."VELIUS LONGUS.
" Alpha semper atque Iota, qvem  parant Graeci
sonum, A et E nobis ministrant: sic nos scribirnus."
-TERENTIANUS.
"Apud antiqvos I litera pro E scribebatur, ut
testantur metaplasmi in qvibus est ejus inodi syllabarum deductio, ut PICTAI VESTIS, et AULAI IN MEDIO, pro PICTAE et AULAE."-SCAURUS.
NOTE C.-PAGE 31.
"Transit qvoqve At in o productam more antiqvo, ut LOTUS pro LAUTUS, PLOSTRUM pro PLAUS



APPENDIX.                 111
TRUM, COTES pro CAUTES; SiCUt etiam contra pro o,
AU; ut AUSTRU~M p1' OSTRU3M, AUSCULUM p'ro OSCULUM; frequentissime hoc faciebant antiqvi."PRISCIAN.
NOTE D.-PAGE 31.
"AB OLOES antiqvi dicebant pro AB ILLIS."FESTUS.
" MOERORUi antiqvi pro MU[RORUM; nam veteres
pleraqve, eorum  qvae nos per u dicimus, per OE
diphthongon pronunciabant; et contra PUNIO pro
POENIO, qvod verbum a POENA venit. Hinc est PCNICA REGNA VIDES, cum POENOS ubiqve legerimus."
-SERVIUS.
NOTE E.-PAGE 42.
"Prisci pro ACIPENSER  dicebant AQVIPENSEE,
pro SEXDECIM, SEXDEQV1M. Ex tcapKepa fecernt
febrim qverqveram  et carcerem, qvia scilicet acdpicKE efferebant."-~FESTUS.
" Cum Romani, euntes per Tusciam, interrogarent Agyllinos qvae diceretur civitas, illi, ut pote
Graeci, qvid audirent ignorantes, et optimum ducentes si eos prius salutarent, dixerunt Xalpe; qvam
salutationem  Romani nomen civitatis esse putaverunt et, detracta aspiratione, earn CAERE nomina.
verunt."-JULIUS HYGINUS apud SERVIUM.




112                APPENDIX.
NOTE F.-IPAGE 42.
"Antiqvi per c literae formam  nihilominus G
usurpabant. ACETARE dicebant pro AGITARE; PRODIGIA qvod rrODICAsT futura, permutatione G literae; nam qvae nunc G appellatur, ab antiqvis c
vocabatur, QVINCENTITM per c literam usurpabant
antiqvi."-FEsTUS.
"Pro agro GABINO dicebant CABINO; pro LEGE,
LECE; ACNA pro AGNA. AUCTIO certe ab AUGENDA
dicta est; et numeri cum c habeant, ut DUCENTI,
SEXCENTI, G reliqvi habent ut QVADRINGENTI, NONGENTI."  VICTORINUS.
" CAMELUMr alii dicunt, alii, GAMELUM. NEGOTIIUM dictum est qvia NEC OTIUM."-SCAIRUS.
NOTE G.-PAGE 45.
"Cum is candidatus, qvi coqvi filius habebatur,
coram  Cicerone suffragiur  ab alio peteret, EGO
QVOQVE, Cicero inqvit, TIBI JURE FAVEBO, pro EGO
COQVE."-Q-VINCT.
"Ciceronis dictum refertur in eum, qvi coqvi
filius secum  causas agebat. Tu QVOQVE ADERAS
HUIC CAUSAE; nam veteres COQvus non per c literam, sed per Q scribebant." —DONATUS.
" Cocur nonnulli in utraqve syllaba per Q scribunt, nonnulli et inserta v. In verbo enim COQVERE,
pro QVOQVERE, NISUS censet ubiqve c literam po



APPENDIX.                 113
nendam, tam in nomine qvam in verbo."-VELIUS
LONGUS.
" Apud antiqvos fi'eqventissime loco cu syllabae
QVU ponebatur, et e contrario; ut ARQYUS, COQVIS,
OQVULUS, QVUM, QVURI."-PRISCIAN.
" QVOTIDIE sunt qvi per co, COTIDIE scribant,
qvibus peccare licet desinerent, si scirent inde tractum esse a QVOT DIEBUS, hoc est, OMNIBUS DIEBUS."
-ANNAEUS CORNUTUS.
"RELIQVIAE et RELIQVI per c scribebantur, coTIDIE per c et o dicitur et scribitur, pro Q; qvia
non QVOTIDIE, sed a CONTINENTE DIE dictum  est."
-PAPYRIANUS.
"Licinius Calvus Q litera non est usus. Antiqvi cuMI adverbium scribebant qvatuor literis QVUMr,
sed pronunciabant tamen PERINDE AC SI cuM
scriptum esset."~-VICTORINIS.
NOTE IH.-PAGE 46.
"Antiqvi in connexione.syllabarum ibi tantum
K utebantur ubi A litera subjungenda erat; qvoniam multis vocalibus instantibus, qvoties id verbum scribendum erat, in qvo retinere hae literae
nomen suum possent, singulae pro syllaba scribebantur, tanqvam satis eam ipso nornine explerent,
ut puta DECIMUS, D per se, inde CIMUS; item CERA, c
simplex et RA, et BENE, BEN. Ita et qvoties KANUS,
KARUS scribendum erat, qvia singulis literis primae
syllabae notabantur, K prima ponebatur, qvae suo




114                APPENDIX.
nomine A continebat; qvia si c posuissent, CENUS
et CERUS futurum  erat, non CANUS, et CARUS."ScAURUs. Cf. also ARISTARCHUS, De Arte Gram.,
lib. i., cap. xvii.
NOTE I.-PAGE 40.
"Et nostrarum ultima x, qva tamen carere potuimus, si non qvaesissemus."-QuINT., i., iv.
" Latini voces qvae in x literam incidunt, si in declinatione earam apparebat G, scribebant GS, ut CONJUGS, LEGS."-VICTORINUS.
" X  duplicem  loco c et s, vel G et s postea a
Graecis inventain, assumpsimus, ut DUX, DUCIS; REX,
REGIS."- PRISCIAN.
" Ante Augustum cs vice x."-ISIDORUS.