The Roman Pronunciation of Latin
F >>
Frances E. Lord >> The Roman Pronunciation of Latin
In the matter of exceptions to the rule that accent does not fall on the
ultimate, we find a somewhat wide divergence of opinion among the
grammarians. Some of them give numerous exceptions, particularly in the
distinguishing of parts of speech, as, for instance, between the same
word used as adverb or preposition, as _ánte_ and _anté_; or between the
same form as occurring in nouns and verbs, as _réges_ and _regés_; and
in final syllables contracted or curtailed, as _finit_ (for _finivit_).
But since on this point the grammarians do not agree among themselves,
either as to number or class of exceptions, or even as to the manner of
making them, we may treat this matter as of no great importance (as in
English, we please ourselves in saying _pérfect_ or _perféct_). And here
it may be said that due attention to the quantity will of itself often
regulate the accent in doubtful cases; as when we say _doce_, if we duly
shorten the o and lengthen the e the effect will be correct, whether the
ear of the grammarian detect accent on the final syllable, or not. For
as Quintilian well says:
Nam ut color oculorum indicio, sapor palati, odor narium dinoscitur, ita
sonus aurium arbitrio subjectus est.
PITCH.
But besides the length of the syllable, and the place and quality of the
accent, another matter claims attention.
In English all that is required is to know the place of the accent,
which is simply distinguished by greater stress of voice. This
peculiarity of our language makes it more difficult for us than for
other peoples to get the Latin accent, which is one of pitch.
In Latin the acute accent means that on the syllable thus accented you
raise the pitch; the grave indicates merely the lower tone; the
circumflex, that the voice is first raised, then depressed, on the same
syllable. To quote again the passage from Priscian:
[Keil. v. in. p. 519.] Acutus namque accentus ideo inventus est quod
acuat sive elevet syllabam; gravis vero eo quod deprimat aut deponet;
circumflexus ideo quod deprimat et acuat.
In conclusion of this part of the work the following anecdotes from
Aulus Gellius are given, as serving to show that to the rules of classic
Roman pronunciation there were exceptions, apparently more or less
arbitrary, some--perhaps many--of which we may not now hope to discover;
and as serving still more usefully to show, by the stress laid upon
points of comparative insignificance, that exceptions were rare, such as
even scholars could afford to disagree upon, and not such as to affect
the general tenor of the language. So that we are encouraged to believe
that, as the English language may be well and even elegantly spoken by
those whose speech still includes scores, if not hundreds, of variations
in pronunciation, in sounds of letters or in accent, so we may hope to
pronounce the Latin with some good degree of satisfaction, whether, for
instance, we say _quiésco_ or _qui'esco_, _ãctito_ or _actito_:
[Aul. Cell. VI. xv.] Amicus noster, homo multi studii atque in bonarum
disciplinarum opere frequens, verbum _quiescit_ usitate e littera
correpta dixit. Alter item amicus homo in doctrinis, quasi in
praestigiis, mirificus, communiumque vocum respuens nimis et
fastidiens, barbare eum dixisse opinatus est; quoniam producere
debuisset, non corripere. Nam _quiescit_ ita oportere dici praedicavit,
ut _calescit_, _nitescit_, _stupescit_, atque alia hujuscemodi multa.
Id etiam addebat, quod _quies_ e producto, non brevi, diceretur. Noster
autem, qua est omnium rerum verecunda mediocritate, ne si Aelii quidem
Cincii et Santrae dicendum ita censuissent obsecuturum sese fuisse ait,
contra perpetuam Latinae linguae consuetudinem. Neque se tam insignite
locuturum, absona aut inaudita ut diceret. Litteras autem super hac re
fecit, item inter haec exercitia quaedam ludicra; et _quiesco_ non esse
his simile quae supra posui, nee a _quiete_ dictum, sed ab eo
_quietem_; Graecaeque vocis [Greek: eschon kai eskon], lonice a verbo
[Greek: escho ischo] et modum et originem verbum illud habere
demonstravit. Rationibusque haud sane frigidis docuit _quiesco_ e
littera longa dici non convenire.
[Aul. Gell. IX. vi.] Ab eo, quod est _ago_ et _egi_, verba sunt quae
appellant grammatici frequentativa, _actito_ et _actitavi_. Haec quosdam
non sane indoctos viros audio ita pronuntiare ut primam in his litteram
corripiant; rationemque dicant, quoniam in verbo principali, quod est
_ago_, prima littera breviter pronuntiatur. Cur igitur ab eo quod est
_edo_ et _ungo_, in quibus verbis prima littera breviter dicitur,
_esito_ et _unctito_, quae sunt eorum frequentativa prima littera longa
promimus? et contra, _dictito_, ab eo verbo quod est _dico_, correpte
dicimus? Num ergo potius _actito_ et _actitavi_ producenda sunt?
quoniam frequentativa ferme omnia eodem modo in prima syllaba dicuntur,
quo participia praeteriti temporis ex iis verbis unde ea profecta sunt
in eadem syllaba pronuntiantur; sicut _lego_, _lectus_, _lectito_
facit; _ungo_, _unctus_, _unctito_; _scribo_, _scriptus_, _scriptito_;
_moneo_, _monitus_, _monito_; _pendeo_, _pensus_, _pensito_; _edo_,
_esus_, _esito_; _dico_, autem, _dictus_, _dictito_ facit; _gero_,
_gestus_, _gestito_; _veho_, _vectus_, _vectito_; _rapio_, _raptus_,
_raptito_; _capio_, _captus_, _captito_; _facio_, _factus_, _factito_.
Sic igitur _actito_ producte in prima syllaba pronuntiandum, quoniam ex
eo fit quod est _ago_ et _actus_.
PART II.
HOW TO USE IT.
The directions now to be given may be fittingly introduced by a few
paragraphs from Professor Munro's pamphlet on the pronunciation of
Latin, already more than once quoted from. He says--and part of this has
been cited before:
"We know exactly how Cicero, or Quintilian did or could spell; we know
the syllable on which they placed the accent of almost every word; and
in almost every case we already follow them in this. I have the
conviction that in their best days philological people took vast pains
to make the writing exactly reproduce the sounding; and that if
Quintilian or Tacitus spelt a word differently from Cicero or Livy, he
also spoke it so far differently. With the same amount of evidence,
direct and indirect, we have for Latin, it would not, I think, be worth
anybody's while to try to recover the pronunciation of French or
English; it might, I think, be worth his while to try to recover that of
German or Italian, in which sound and spelling accord more nearly, and
accent obeys more determinable laws."
"I am convinced," he says in another place, "that the mainstay of an
efficient reform is the adoption essentially of the Italian vowel
system: it combines beauty, firmness and precision in a degree not
equalled by any other system of which I have any knowledge. The little
ragged boys in the streets of Rome and Florence enunciate their vowels
in a style of which princes might be proud."
And again:
"I do not propose that every one should learn Italian in order to learn
Latin. What I would suggest is, that those who know Italian should make
use of their knowledge and should in many points take Italian sounds for
the model to be followed; that those who do not know it should try to
learn from others the sounds required, or such an approxi-mation to them
as may be possible in each case."
We may then sum up the results at which we have arrived in the following
directions:
First of all pay particular attention to the vowel sounds, to make them
full and distinct, taking the Italian model, if you know Italian, and
always observing strictly the quantity.
Pronounce
[long a] as in Italian _fato_ or as final a in aha!
a as in Italian _fatto_; or as initial a in aha! or as in fast (not as
in fat).
[long e] as second e in Italian _fedele_; or as in fête (not fate); or
as in vein.
e as in Italian _fetta_; or as in very.
[long i] as first i in Italian _timide_; or as in caprice,
i as second i in Italian _timide_; or as in capricious.
i or u, where the spelling varies between the two (e.g. _maximus_,
_maxumus_), as in German Müller.
[long o] as first o in Italian _orlo_; or as in more.
o as first o in Italian _rotto_; or as in wholly (not as in holly).
[long u] as in Italian _rumore_; or as in rural.
u as in Italian _ruppe_; or as in puss (not as in fuss).
Let i in vi before d, t, m, r or x, in the first syllable of a word, be
pronounced quite obscurely, somewhat as first i in virgin.
In the matter of diphthongs, be sure to take always the correct
spelling, to begin with, and thus avoid what Munro justly terms "hateful
barbarisms like _coelum_, _coena_, _moestus_." Much time is wasted by
students and bad habits are acquired in not finding, at the outset, the
right spelling of each word and holding to it. This each student must do
for himself, consulting a good dictionary, as editors and editions are
not always to be depended on. Here it is the diphthongs that present the
chief difficulty and call for the greatest care.
In pronouncing diphthongs sound both vowels, but glide so rapidly from
the first to the second as to offer to the ear but a single sound. In
the publication of the Cambridge (Eng.) Philological Society on
"Pronunciation of Latin in the Augustan Period," the following
directions are given:
"The pronunciation of these diphthongs, of which the last three are
extremely rare, is best learnt by first sounding each vowel separately
and then running them together, AE as ah-eh, AU as ah-oo, OE as o-eh, EI
as eh-ee, EU as eh-oo, and UI as oo-ee."
Thus:
AE (ah-éh) as in German _näher_; or as EA in pear; or AY in aye (ever);
(not like a* in fate nor like AI in aisle).
AI (ah-ée) as in aye (yes).
AU (ah-óo) as in German _Haus_, with more of the U sound than OU in
house.
EI (eh-ée) nearly as in veil. (In _dein_, _deinde_, the EI is not a
diphthong, but the E, when not forming a distinct syllable, is elided.)
EU (eh-óo) as in Italian _Europa_. (In _neuter_ and _neutiquam_ elide
the E.)
OE (o-éh) nearly like German ö in _Goethe_.
OI is not found in the classical period. (In _proin_, _proinde_, the O
is either elided or forms a distinct syllable. OU in _prout_ is not a
diphthong; the U is either elided or forms a distinct syllable.)
UI (oo-ée) as in cuirass.
In the pronunciation of consonants certain points claim special
attention. And first among these is the sounding of the doubled
consonants. Whoever has heard Italian spoken recognizes one of its
greatest beauties to be the distinctness, yet smoothness, with which its
ll and rr and cc--in short, all its doubled consonants--are pronounced.
No feature of the language is more charming. And one who attempts the
same in Latin and perseveres, with whatever difficulty and pains, will
be amply rewarded in the music of the language.
A good working rule for pronouncing doubled consonants is to hold the
first until ready to pronounce the second: as in the words _we'll lie
till late_, not to be pronounced as _we lie till eight_.
Next in importance, and, in New England at least, first in difficulty,
is the trilling of the r. There can be no approximation to a
satisfactory pronunciation of Latin until this r is acquired; but the
satisfaction in the result when accomplished is well worth all the pains
taken.
Another point to be observed is that the dentals t, d, n, l, require
that the tongue touch the teeth, rather than the palate. Munro says: "d
and t we treat with our usual slovenliness, and force them up to the
roof of our mouth: we should make them real dentals, as no doubt the
Romans made them, and then we shall see how readily _ad at_, _apud
aput_, _illud illut_ and the like interchange." This requires care, but
amply repays the effort.
It is necessary also to remember that n before a guttural is pronounced
as in the same position in English, e.g., in _ancora_ as in anchor; in
_anxius_ as in anxious; in _relinquo_ as in relinquish.
Remember to make n before f or s a mere nasal, having as little
prominence otherwise as possible, and to carefully lengthen the
preceding vowel.
Studiously observe the length of the vowel before the terminations
_gnus_, _gna_, _gnum_.
Remember that the final syllable in m, when not elided, is to be
pronounced as lightly and rapidly as possible, the more lightly and
indistinctly the better.
Remember that s must not be pronounced as z, except where it represents
z in Greek words, as Smyrna (Zmyrna), Smaragdus (Zmaragdus), otherwise
always pronounce as in sis.
Remember in pronouncing v to direct the lower lip toward the upper lip,
avoiding the upper teeth.
In general, in pronouncing the consonants conform to the following
scheme:
b as in blab.
b before s or t, sharpened to p, as _urbs_==_urps_; _obtinuit_==
_optinuit_.
c as sceptic (never as in sceptre).
ch as in chemist (never as in cheer or chivalry).
d as in did, but made more dental than in English.
d final, before a word beginning with a consonant, in particles
especially, often sharpened to t as in tid-bit (tit-bit).
f as in fief, but with more breath than in English.
g as in gig (never as in gin).
gn in terminations _gnus_, _gna_, _gnum_, makes preceding vowel long.
h as in hah!
i (consonant) as in onion.
k as in kink.
l initial and final, as in lull.
l medial, as in lullaby, always more dental than in English.
m initial and medial, as in membrane.
m before q, nasalized.
m final, when not elided, touched lightly and obscurely, somewhat as in
tandem (tandm); or as in the Englishman's pronunciation of Blenheim
(Blenhm), Birmingham (Birminghm).
n initial and final, as in nine.
n medial, as in damnable, always more dental than in English.
n before c, g, q, x, as in concord, anger, sinker, relinquish, anxious,
the tongue not touching the roof of the mouth.
n before f or s, nasal, lengthening the preceding vowel, as in
_renaissance_.
p as in pup.
q as in quick.
r as in roar, but trilled, as in Italian or French. (This is most
important.)
s as in sis (never as in his).
t as in tot, but more dental than in English (never as in motion).
th nearly as in then (never as in thin).
v (u consonant) nearly as in verve, but labial, rather than labio-
dental; like the German w (not like the English w). Make English v as
nearly as may be done without touch-* the lower lip to the upper teeth.
x as in six.
z nearly as dz in adze.
Doubled consonants to be pronounced each distinctly, by holding the
first until ready to pronounce the second.
As Professor Ellis well puts it: "No relaxation of the organs, no puff
of wind or grunt of voice should intervene between the two parts of a
doubled consonant, which should more resemble separated parts of one
articulation than two separate articulations."
"Duplication of consonants is consequently regarded simply as the
energetic utterance of a single consonant."
ELISION.
Professor Ellis believes that the m was always omitted in speaking and
the following consonant pronounced as if doubled (_quorum pars_ as
_quoruppars_). Final m at the end of a sentence he thinks was not heard
at all. Where a vowel followed he thinks that the m was not heard, the
vowel before being slurred on to the initial vowel of the following
word.
The Cambridge (Eng.) Philological Society, however, takes the view that
"final vowels (or diphthongs) when followed by vowels (or diphthongs)
were not cut off, but lightly run on to the following word, as in
Italian. But if the vowel was the same the effect was that of a single
sound."
Professor Munro says:
"In respect of elision I would only say that, by comparing Plautus with
Ovid, we may see how much the elaborate cultivation of the language had
tended to a more distinct sounding of final syllables; and that but for
Virgil's powerful influence the elision of long vowels would have almost
ceased. Clearly we must not altogether pass over the elided vowel or
syllable in m, except perhaps in the case of e* in common words, _que_,
_neque_, and the like."
This view, held by the Cambridge Philological Society and by Professor
Munro, is the one generally accepted; the practice recommended by them
is the one generally in use, and that which seems safe and suitable to
follow. That is: Do not altogether pass over the elided vowel or
syllable in m, except in cases of very close connection, in compound
words or phrases, or when the final and initial vowel are the same, or
in the case of e* final in common words, as _que_, _neque_, and the
like; but let the final vowel run lightly on to the following vowel as
in Italian, and touch lightly and obscurely the final syllable in m. The
o or e of _proin_, _proinde_, _prout_, _dein_, _deinde_, _neuter_,
_neutiquam_, when not forming a distinct syllable, are to be treated as
cases of elision between two words.
QUANTITY.
In the pronunciation of Latin the observance of quantity and of pitch
are the two most difficult points of attainment; and they are the
crucial test of good reading.
The observance of quantity is no less important in prose than in verse.
A little reflection will convince the dullest mind that the Romans did
not pronounce a word one way in prose and another in verse, that we have
not in poetry and prose two languages. Cicero and Quintilian both enjoin
a due admixture of long and short syllables in prose as well as verse;
and any one who takes delight in reading Latin will heartily agree with
Professor Munro when he says: "For myself, by observing quantity, I seem
to feel more keenly the beauty of Cicero's style and Livy's, as well as
Virgil's and Horace's."
Therefore until one feels at home with the quantities, let him observe
the rule of beating time in reading, to make sure that the long
syllables get twice the time of the short ones. In this way he will soon
have the pronunciation of each word correctly fixed in mind, and will
not be obliged to think of his quantities in verse more than in prose. A
long step has been taken in the enjoyment of Latin poetry when the
reader does not have to be thinking of the 'feet.'
Young students particularly should be especially careful in the final
syllable of the verse. Since, so far as the measure is concerned, there
is no difference there between the long and the short syllable, the
reader is apt to be careless as to the length of the syllable itself,
and to make all final syllables long, even to the mispronouncing of the
word, thereby both making a false quantity and otherwise injuring the
effect of the verse, by importing into it a monotony foreign to the
original. Does not Cicero himself say that a short syllable at the end
of the verse is as if you ' stood (came to a stand), but a long one as
if you ' sat down'?
It is, in fact, in the pronouncing of final syllables everywhere that
the most serious and persistent faults are found, bus for bus being one
of the worst and most common cases. How much of the teacher's time might
be spared, for better things, if he did not have to correct bus into
bus!
The disposition to neglect the double and doubled consonants is another
serious fault, as well as the slovenly pronunciation of two consonants,
where the reader fails to give the time necessary to speak each
distinctly, making false quantity and mispronunciation at the same time.
In general, if two symbols are written we are to infer that two sounds
were intended. The only exception to this is in the case of a few words
where the spelling varies, as casso or caso. In such cases we may
suppose that the doubled consonant was only designed to indicate length.
Another, apparent, exception is in the case of a mute followed by a
liquid; but the mute and liquid are regularly sounded as one, and
therefore do not affect the length of the preceding vowel. Sometimes,
however, for the sake of time, the verse requires them to be pronounced
separately. In this case each is to be given distinctly; the mute and
liquid must not coalesce. For it must not be forgotten that, as a rule,
the vowel before a mute followed by a liquid is short, in which case it
must on no account be lengthened. Thus, ordinarily, we say pa-tris, but
the verse may require pat-ris.
Although the vowel before two consonants is generally--short, we find,
in some instances, a long vowel in this position. For example, it would
appear that the vowel of the supine and cognate parts of the verb is
long if the vowel of the present indicative, though short, is followed
by a medial (b, g, d, z), as actus, lectus, from ago, lego.
Let it be remembered in the matter of i consonant between two vowels,
that we have really the force of two ii's, as originally written, one,
vowel, making a diphthong with the preceding, the other, consonant,
introducing the new syllable; and that the same is true of the compounds
of _jacio_, which should be written with a single i but pronounced as
with two, as _obicit (objicit)_.
ACCENT.
The question of accent presents little difficulty as to place, but some
as to quality, and much as to kind. As to quality, it must be remembered
that while the acute accent is found on syllables either short or long
(by nature or position), and on either the penult or the antepenult, the
circumflex is found only on long vowels, and (in words of more than one
syllable) only on the penult, and then only in case the ultima is short.
Thus, _spés_, but _dúx_; _lûna_, but _lún[long a]_; _legâtus_, but
_legáti_. In these examples the length of the syllable is the same and
of course remains the same in inflection, but the quality of the accent
changes. In the one case the voice is both raised and depressed on the
same syllable, in the other it is only raised. As Professor Ellis puts
it: "If the last syllable but one is long, it is spoken with a raised
pitch, which is maintained throughout if its vowel is short, as:
_vént[long o]s_, or if the last syllable is long, as: _f[long a]m[long
a]e_; but sinks immediately if its own vowel is long, and at the same
time the vowel of the last syllable is short, as _fâma_, to be
distinguished from _f[long a]m[long a]_."
But when we come to the question of the _kind_ of accent, we come upon
the most serious matter practically in the pronunciation of Latin, and
this because of a difficulty peculiar to the English speaking peoples.
The English accent is one of _stress_, whereas the Roman is one of
_pitch_.
No one will disagree with Professor Ellis when he "assumes," in his
Quantitative Pronunciation of Latin, "that the Augustan Romans had _no_
force accent, that is, that they did not, as we do, distinguish one
syllable in every word _invariably_ by pronouncing it with greater
force, that is, with greater loudness, than the others, but that the
force varied according to the feeling of the moment, or the beat of the
timekeeper in singing, and was used for purposes of expression; just as
with us, musical pitch is free, that is, just as we may pronounce the
same word with different musical pitches for its different syllables,
and in fact are obliged to vary the musical pitch in interrogations and
replies. The fixity of musical pitch and freedom of degrees of force in
Latin, and the freedom of musical pitch and fixity of degrees of force
in English sharply distinguish the two pronunciations even irrespective
of quantity."
But this pitch accent, while alien to us, is not impossible of
acquisition, and it is essential to any adequate rendering of any Latin
writer, whether of prose or verse. Nor will the attainment be a work of
indefinite time if one pursues with constancy some such course as the
following, recommended by Professor Ellis:
"The place of raised pitch," he says, "must be strictly observed, and
for this purpose the verses had better be first read in a kind of sing-
song, the high pitched syllables being all of one pitch and the low
pitched syllables being all of one pitch also, but about a musical
'fifth' lower than the other, as if the latter were sung to the lowest
note of the fourth string of a violin, and the former were sung to the
lowest note of its third string."
In the foregoing pages an effort has been made to bring together
compactly and to set forth concisely the nature of the 'Roman method' of
pronouncing Latin; the reasons for adopting, and the simplest means of
acquiring it. No attempt has been made at a philosophical or exhaustive
treatment of the subject; but at the same time it is hoped that nothing
unphilosophical has crept in, or anything been omitted, which might have
been given, to render the subject intelligible and enable the
intelligent reader to understand the points and be able to give a reason
for each usage herein recommended.
The main object in view in preparing this little book has been to help
the teachers of Latin in the secondary schools, to furnish them
something not too voluminous, yet as satisfactory as the nature of the
case allows, upon a subject which the present diversity of opinion and
practice has rendered unnecessarily obscure.
To these teachers, then, a word from Professor Ellis may be fitly spoken
in conclusion: