EARLY TRAGEDY AND EPIC
Browning has recalled the story of how Greek war captives taken at
Syracuse in the Peloponnesian war earned their release by reciting
snatches from the plays of Euripides. It was a century and a half after
that siege that the Romans came to Sicily in the First Punic War, and
the city was still interested in the old drama, indeed was now taking
its part in producing tragedies. One of the last of the dramatists, one
of the so-called “Pleiad,” was a Syracusan of Hiero’s time, and King
Hiero was himself so devoted to the drama that he even built a theater
for Agyrion, a petty village on the border of his small kingdom. We have
noticed how the Roman youth who campaigned year after year in Sicily
learned something of the arts of civilization and on their return home
created a demand for the things they had come to enjoy while abroad.
The year after the victorious troops returned from Sicily, Livius, a
schoolmaster of Greek origin, staged a translation of a Greek tragedy
as a supplement to the annual chariot race. This production marks the
beginning of Rome’s education in letters. There must be some close
connection between this homecoming of the army, and the performance of
Livius’ play, for the change in character of a great religious festival
could not have been suggested by a freedman. The magistrates responsible
for the performance were senators and the senate had of course requested
the play. In all likelihood it was also the senate that invited King
Hiero of Syracuse to Rome to see the games; for he, if any one, would
have been asked to supply some actor to help stage the first play, and
it was only appropriate that he should come to inaugurate the new era of
culture.
From that time on plays were produced every year. Five years after the
first performance, Naevius, who had served in the Sicilian campaigns
(and had perhaps learned Greek there), began to help in the work of
adapting Greek plays for the Roman stage. Only brief fragments of those
early plays have survived and in reviewing the list of titles we might
wonder at the enthusiasm they reveal for plays shaped on the old Greek
mythology. But the predominance of titles derived from the Trojan cycle
explains this enthusiasm. It was in Sicily that the Roman soldiers had
learned the Greek story of how Rome had been founded by Trojan refugees.
The stories of Hector, of the Trojan horse, Achilles, Ajax, Iphigenia,
and the rest were therefore not without personal interest in the barbaric
city. The unlettered shoemakers, smiths, and carpenters at Rome, men
whose modern equals could hardly be expected to sit patiently through a
performance of Gilbert Murray’s _Trojan Women_, eagerly listened to the
half-comprehended lines of Livius’ translation. They had been told that
these were the stories of their long-lost ancestors.
Livius is merely a name, which is unfortunate, since we know that
he deserved well of Roman civilization. Naevius is less shadowy, a
personality whose creative work left an impress on such powerful men
as Cicero and Vergil two centuries later. He wrote not only plays, but
an epic, condensing Rome’s history in an annalistic poem, the climax
of which was the great victory over Carthage in which he had had a
share. From the sixty scattered lines of this epic rescued by late
lexicographers we do not quite find the justification for Vergil’s high
regard. There is no poetry in them. But grammarians pick their lines
to illustrate linguistic usage and not for effective phrasing. Even
Shakespeare becomes prose if judged by the citations found in Webster.
However, for the preservation of the metrical schemes employed by Naevius
we are grateful. Though he had used a large variety of Greek meters for
his drama, he did not in his epic. Here he preserves the native Saturnian
line that had been used in religious songs, and apparently in ballads.
That he did not adopt a standard Greek meter for his epic, as he did for
his tragedies and comedies, is proof enough that the old native narrative
verse was fully established in a well-known body of poetry which we have
lost.
In many respects this verse resembles the old English line that relies
upon alliteration and rhythmic ictuses which balance each other in the
two severed parts of the line:
In a sómer séson whan sóft was the sónne.
But the Saturnian had six ictuses instead of four, and as Latin verse was
more aware of its quantities and less of its word stress than English
the ictuses, while somewhat regardful of word accent, were more attentive
to quantity. Finally, since alliteration is more effective when the ictus
falls on the first syllable, and since the Latin accent had to a large
extent shifted away from the first syllable by the time of Naevius,
the use of alliteration was somewhat less frequent in Naevius than in
_Beowulf_. In Vergil’s day the effect of this verse must have been
somewhat like that of Langland’s poems upon the Elizabethans. The shift
of the Latin word accent toward the penult was already destroying the
effectiveness of the verse even when Naevius wrote; and the break of the
line in the center rendered it ineffective for sustained narration. Its
halting movement may be somewhat inadequately illustrated by a paraphrase
of Naevius’ own epitaph:
If death of any mortal sadden hearts immortal,
The heavenly Muses surely Naevius’ death bemoan;
For after he departed to the shades of Orcus
The voice of Rome is silent music is forgotten.
Ennius abandoned the line, and it was eventually doomed, just as the
Anglo-Saxon meters in England began to disappear when the richer rhythms
of French poetry came to be appreciated.
It was Naevius also who broke away from Greek subjects in the drama,
though with what success we cannot say. He made what we may call a
“chronicle play” of the Romulus legend which disregarded the conventional
unities, and he also wrote a pageantry play to commemorate the heroic
single combat of Marcellus with a Celtic chieftain. He is therefore among
the first to stage contemporary drama and to disregard the restrictions
of time and place. That he made the same innovations in comedies like
his _Hariolus_ is probable but cannot be proved from the few lines that
remain.
An independent creator he was and might have carried progress far had
not so large a part of his activity fallen in the restraining period
of the Second Punic War. His end was in character. Accustomed to speak
his mind freely in his comedies, he vigorously supported the Fabian
policy when it was unpopular, and after the group supporting Scipio,
which demanded a more aggressive conduct of the war, came into power, he
continued his sarcastic criticism of the Scipionic group. Rome had always
tolerated free speech, but even at Rome patience was short in war time.
War censorship discovered an old law which, with a little imaginative
interpretation, could be stretched to cover the case of this satirist.[1]
Only one line has survived of the satiric comedy which referred to
the fact that Metellus, a friend of Scipio’s, had taken advantage of
a fortuitous circumstance to stand for the consulship. He was elected
through no desert of his own. The point of the line—Fato Metelli Romae
fiunt consoles—rests on a _double entendre_, because _fato_ may be
construed either as ablative or as dative, while _Romae_ may be genitive
or locative. The line therefore may mean either:
“The Metelli became consuls at Rome by chance,”
which is hardly a flattering remark, or what is even less flattering:
“The Metelli became consuls to Rome’s sorrow.”
The Metelli apparently thought Naevius meant to suggest both, which
is likely enough, and they succeeded in having him imprisoned, and
eventually banished. He seems to have found a home in Carthage, the land
of the enemy against whom he had once fought.
These two dramatists, for reasons which must be discussed later,
increased the use of musical accompaniment in tragedy and comedy. In
Euripides the body of the drama had been in recited trimeters. The
choral parts were of course sung to the accompaniment of rhythmical
movements called dance, and there was also music when the actors engaged
in dialogue with the chorus, as well as in some of their monologues. But
the musical element had been reduced very much during the century that
followed Euripides when the drama had gradually dispensed with the chorus
even in the staging of Euripides. Rome was then too primitive to provide
the twelve or more trained singers and dancers that even the later
Greeks had found beyond their resources. Livius indeed had experienced
such difficulty in securing actors with good voices that he himself took
the leading rôle, and, not adequately gifted for the singing parts, he
tried, we are told, the inartistic device (not unknown on our comic
stage) of placing a singer beside the musician to carry the melody of
the lyric parts while he acted and presumably recited the lines of the
songs. That was, of course, only a temporary makeshift, but it shows how
difficult it was to provide satisfactory artists at the time. It would
seem then that since these early writers found it impossible to produce
choruses adequately because these required the elaborate training of
many singers in intricate musical compositions, they compensated as best
they could by increasing the number of monodies in their plays, writing
them in a few well-defined meters, such as the septenarii, cretics, and
bacchiacs,[2] which were not too difficult to learn. Thus it was that
Roman tragedy became even more like modern opera than the tragedy of the
Greek stage had been. These early men of Rome, who mean so little to us,
had developed a form which was capable of carrying on the work of Greek
tragedy on a primitive stage, and capable also of growing into a richer
drama as soon as the resources of the small city should permit. They made
the drama possible for Rome.
As a composer of tragedies and epic verse, Ennius succeeded Naevius,
but, writing in an era of enthusiastic philhellenism, he came near
yielding too much to a great foreign influence. Had he not been a man
of remarkable poetic powers his example might well have quenched the
spirit of Rome in the rising literature. One of the first Latin authors
that we would ask the excavators of Herculaneum to restore to us is
Ennius. No single work of his has survived. Of his twenty-five or more
tragedies we possess only about four hundred lines; of the eighteen books
of _Annals_ a little over five hundred complete lines; of his satires,
his _Euhemerus_ and his _Epicharmus_, not enough fragments have survived
to give us a very clear idea of the scope of each. All in all we have
about three per cent of his work in scraps, but here at least there are
several connected passages cited in appreciation of something else than
the grammatical usages they illustrate.
Ennius too, like Naevius, told Rome’s story in verse. One’s impulse is
to discount the accuracy of any history that employs artifice. And one
must grant of course that a poet will select his incidents with a view to
their dramatic values and picturesqueness. One must remember, however,
that poetry had a serious place in all early literature for the reason
that, before the day of much writing, all teachable things, even history
and philosophy, were put into verse for mnemonic purposes. The works of
Solon and Heracleitus would not have contained different matter if they
had been put into prose. In Ennius’ day many national histories that
purported to be accurate were composed in verse. And Ennius probably did
not permit himself to include fictive incidents in his _Annals_, nor
has he been proved incorrect at any point.[3] Cicero cited him for the
gist of the famous speech of Appius Claudius, and added as a matter of
indifference that the original of the speech was in existence. Apparently
Ennius’ summary was accurate enough so that it was not necessary to refer
to the original text.
The influence of his _Annals_ was in its field comparable to that of
Homer. From Ennius all schoolboys got their first impressions of what
Rome’s great heroes had accomplished. He was unsurpassed as a painter
of character. With a few telling strokes he revealed the essential
traits of those strong, bold, tireless heroes who made the old Republic
irresistible in power, magnificent in tradition, and a saving inspiration
in the days of decadence. He was near to these men, and it was as he
saw them that they lived on in memory and still live on. He made Roman
character memorable in the two lines on Fabius Maximus:
Unus homo nobis cunctando restituit rem:
Noenum rumores ponebat ante salutem.
and in the single line on Curius:
Quem nemo ferro potuit superare nec auro.
His epic was an exposition of the text he himself devised so effectively:
Moribus antiquis res stat Romana virisque:
And it was Ennius who more than any one else kept Roman society upon that
foundation.
We happen to be able to test his influence by what he did with the
portrait of Pyrrhus. Only a generation before Ennius was born this
picturesque enemy of Rome had had a friendly alliance with the Messapian
tribe to which Ennius himself belonged. The poet, therefore, had heard
much about the king. Pyrrhus, in fact, had some very sympathetic traits
of character, a remarkable chivalry, and a certain sense of honor and
loyalty such as is often found in the chieftains of primitive folk. These
qualities stand out in the characterization of him that Ennius has left
us; and these are the outstanding traits that we find in all the later
Roman references to Pyrrhus. That Ennius should have responded to these
qualities is not strange, but that all the rest of the Romans should thus
enthusiastically have lauded an enemy who nearly wrecked Rome is less to
be expected. The explanation is of course that what Ennius wrote colored
the historic conceptions of all who followed. This becomes evident when
we read Plutarch, a Greek, and his biography of Pyrrhus. When drawing
upon Roman sources for the Italian campaign of the king, Plutarch paints
the same picture as Ennius, but when he draws upon Greek authors in
describing the Greek campaigns, he reveals the fact that the knightly
hero of the Roman historians had a less charming side which certain close
observers at home were well aware of. Like all historians Ennius had
his enthusiasms, and he had such power of portraiture that not a trait
blurred.
He was also fair. Pyrrhus got his meed of praise, but the opponents of
Pyrrhus, Fabricius and Appius Claudius, were characterized with equal
sympathy. Of his own contemporaries, Fabius the Slow-goer was effectively
portrayed as we have seen, and Cato “in caelum tollitur,” as Cicero
affirms, although Scipio Africanus, who was bitterly opposed by these
conservatives, became, as he deserved to be, the outstanding hero of the
book.
It was entirely appropriate that, for his heroic narrative, Ennius
borrowed the dactylic hexameter of the Greeks, but it was after all a
daring thing to do, since meters seldom transplant with success. However,
Naevius’ use of the native Saturnian had demonstrated its inability
to carry heroic narrative. Imagine _Paradise Lost_ crammed into the
primitive English rhythms of Langland! The dactylic hexameter was in
Greek regularly associated with the epic. It had one disadvantage in
its requirement of a larger proportion of short syllables than normal
Latin writing contained, but that was overcome by simply permitting more
spondees than Homeric custom had enjoyed. This resulted in a reduction
of tempo which after all suited Roman military movement. There was
another difficulty which was more serious. While Greek verse needed to
give little attention to word accent, the Latin word accent was jealous
of attention. With the relative fixity of the accent, it was impossible
to write Latin dactyls based both upon quantity and word-accent. Ennius
nevertheless ventured upon an experiment. That he had a very delicate
ear for the demands of the Latin language is proved by the careful
adjustments in his dramatic senarii, where adjustments were not easy to
make. He would not have foisted impossible dactyls upon Rome. The fact
that he wrote quantitative dactyls and continued to write them, and that
his _Annals_ lived for centuries is proof that he did not overstep the
bounds of good taste. The explanation of his success is probably that
the word stress in the Latin of his day was so moderate that a conflict
with the ictus was not fatal to aesthetic pleasure if only it fell
upon long syllables, and also that during the forty years of dramatic
performances at Rome, the ears of the audiences had become trained under
the influence of music to disregard such conflicts in the many lyric
rhythms, including dactyls. By his sensible modification of the Homeric
line, Ennius created as great a resource for Latin poetry as Chaucer did
for English poetry, and shaped for Vergil’s use “the stateliest measure
ever moulded by the lips of man.”
Ennius began to write tragedies about 200 B.C. at the very time when
philhellenism was at its height. Being a man of wide culture who knew
his Greek well he readily responded to the general demand for things
Greek. Though he produced one play (_The Rape of the Sabine Women_) on a
Roman theme, and a pageantry play called the _Ambracia_ to commemorate
the victory of a friend during the war with Aetolia, he seems to have
striven chiefly to reproduce on the Roman stage the effects of Euripides’
tragedies. And now that the restraints of poverty had become somewhat
relaxed, and the drama had continued long enough to foster a certain
amount of skilful talent for its interpretation, he was freer to present
his tragedies more nearly in the old Greek manner. It has accordingly
been plausibly conjectured[4] that it was Ennius who reintroduced the
chorus so that the Greek plays might be given without cutting. There
is no reason for supposing that the choral song in the _Thyestes_
(written in bacchiacs) or the one in the _Medea_ (octonarii) or the one
in the _Iphigenia_ (septenarii) were recited by a single singer. It is
clear from the fragments that in several of his plays, notably in the
_Achilles_, the _Eumenides_, the _Hector_, and the _Hecuba_, choral
groups were actually participants in the plays as they had been in the
Greek originals. And since in the plays of his successor, Accius,
it can be demonstrated that a chorus sang, we ought to accept the
reasonable interpretation of the Ennian fragments and attribute to this
philhellenist the importation of choral song into Roman tragedy. Ennius,
however, deferred to Roman taste so far as the rhythms were concerned.
He adhered largely to the lyric meters which Livius and Naevius had
popularized, and seldom attempted to employ the more intricate systems of
the Greeks.[5] That Ennius was as successful in his tragedies as in his
epic is adequately proved by the fact that many of his plays were still
being produced a century after his death and were avidly read by men like
Cicero.
Pacuvius, the nephew and successor of Ennius, did not write many plays.
From the little that remains of his work we should judge that he
preferred themes somewhat off the beaten track and that in choosing plays
that contained heterodox discussions of ethical themes, he, too, felt the
influence of the new Greek learning and kept in mind the interests of
the intellectualist at Rome. The grammarians have also noticed the fact
that his lyric meters paid more attention to Latin word stress than those
of his predecessors.[6] They cite particularly his care in composing
anapaests with caesuras in such a way that long initial syllables fell
under the ictus. These anapaests in fact read like dactyls with an
anacrusis of two shorts at the beginning. This innovation decidedly
proves that the poet had a precise ear and desired to attain harmonious
effects. His successors showed that they appreciated his innovation,
but they occasionally used the old turbid lines to express emotional
excitement.
The most successful of the writers of tragedy was Accius, a poet who
spanned the era between the Gracchans and the Social War. We have
fragments of more than forty tragedies from his busy pen, and many of his
plays were re-staged in Cicero’s day. He was the favorite of the great
actors, Aesopus and Roscius. He did not depart far from the customs laid
down by Ennius in respect to meters, music, and chorus, but the fact
that he freely readapted the Greek plays which furnished themes to his
predecessors can only mean that he used the same liberty in giving his
own interpretation to old plots that Euripides had used in treating anew
the myths that had been staged by Aeschylus and Sophocles. We happen
to know from the remarks of Terence that convention did not permit the
staging of more than one paraphrase of any given Greek play. When,
therefore, Accius writes plays upon familiar themes we must assume that
he is offering something essentially original in his interpretation of
the old plot. In fact we find good evidence of his original treatment
in the fragments. So, for instance, in his _Antigone_ he changed the
personnel of the chorus (as Ennius had done in the _Iphigenia_), which
implies that the purpose of the play was altered. It is also clear that
Accius made free to disregard the conventional unities of place and time,
for in the _Brutus_ there are scenes laid in Gabii, in Ardea, and in
Rome.
All these dramatists apparently altered their originals freely in order
to make the story and its meaning more plausible to a Roman audience.
The _Medea_ of Ennius reveals many changes of this kind. For instance,
the Latin author felt that he must prepare the audience early in the
play for the gruesome death of the children,[7] a detail unnecessary
in Euripides, who wrote for an audience that knew the plot. This kind
of thing must have occurred frequently. Again, Ennius had to alter
Medea’s long monologue, since before a Roman audience accustomed to
seeing a matron in public, there was no point in making her apologize
for appearing outside of the palace.[8] Ennius has here been needlessly
accused of misunderstanding the Greek original! Ennius knew his Greek; he
had learned it at school in Tarentum. His alterations were introduced to
suit the psychology of his own audience. Similar changes are numerous and
need not be dwelt upon.
The alteration of the very purport of the plays is of more importance to
us. For instance, Atreus, the old Greek tyrant of primitive brutality,
was calculated to offend Roman taste. It is apparent from the fragments
of Accius that it was the sufferings of Thyestes rather than the daring
of Atreus which received sympathetic attention—a fact not surprising in
a city where the word _rex_ was feared and hated. Euripides’ story of
Andromeda had a matter-of-fact plot in which Andromeda’s father begged
Perseus to slay the dragon and to rescue his daughter. This plot followed
the myth and was expected in Athens. But not so at Rome. In Accius’ play
Perseus is rather the chivalrous knight; he rescues the lady first and
then pleads for her hand. Similarly, in the _Clytemestra_ of Accius one
also finds a very modern note, for Accius suggests that if Agamemnon’s
inconstancy could be excused because of his long separation from his
wife, Clytemestra might possibly have the benefit of the same argument.
In the _Andromache_ of Ennius and the _Astyanax_ of Accius there is
an intense note of sorrow for the child of Hector and Andromache that
reminds one of Vergil’s lines in the third book of the Aeneid. This is
a Roman strain deriving from the Romans’ claim to be descended from
the Trojans. In the _Phoenissae_ of Accius the motivation of the whole
play is changed by representing Eteocles breaking a command rather than
a personal pledge. In the _Eurysaces_ of Accius we have a slightly
different reason for the use of Roman motive. This play was re-staged by
the great actor Aesopus when Cicero was in exile, because of its picture
of the unjust banishment of Telamon. The Roman audience appreciated the
possible allusion to Cicero’s suffering and cheered Aesopus’ lines to
the echo. Accius may well have written it originally and introduced the
changes in order to influence his audience and obtain the recall of some
political exile like Popilius, about 130 B.C. The lines have a genuine
Roman ring.
In our own day when every dramatist is compelled to create a new plot it
is easy to underestimate the originality of men like the Greek Euripides,
the Roman Accius, the French Racine, the English Shakespeare, who all in
varying degrees were satisfied to use old plots, even old plays, and to
give all their attention to a personal and original interpretation of the
inner meaning of a familiar story and of the motives that impelled the
characters. We may illustrate the old method of procedure by examining
Seneca’s _Medea_, since here we have a complete Latin play which shows
what even an uninspired Roman dramatist might do by way of re-reading an
ancient legend. Medea in the old unvarnished myth of the barbaric age
was apparently a bundle of natural passions, a savage creature gifted
with superhuman powers. Jason owed her his life, but since a Greek prince
could hardly wed a barbarian and make her his queen, he might reasonably,
according to Greek standards, abandon her when his “higher” duties to
state and position demanded it. In a rage of jealous hate, the creature
might then wreak her vengeance upon Jason and Jason’s children. Such
action was quite comprehensible to the semibarbarous age that shaped the
myth, but not to the more humane Athenians of Euripides’ day. The Greek
dramatist, accordingly, had offered a new explanation of the problem. In
his version Jason has disregarded the higher demands of humanity for a
selfish passion or a more selfish ambition. Medea, the woman, has been
infinitely wronged, and in her helplessness—it is not all jealousy and
hate—she slays her children to save them from a worse fate. But to the
Roman even this interpretation seems impossible, and the character of
Jason least comprehensible of all. A Roman nobleman could not so abandon
his sons, and the woman, if she was indeed human, could not slay her
children either in hate or in love. Seneca, therefore, while keeping
the main plot, seeks a new explanation for the woman’s act. Medea is
again painted as the barbaric witch that she was before Euripides
transformed her. Jason marries Creusa for the sake of his children—a
wholly comprehensible act to a Roman of Nero’s day—and the uncontrollable
Medea is driven into a rage that does not hesitate to commit murder.
But, however jealous she might have been, Seneca feels that she could
not have laid hands upon her own offspring. Yet the tale said that she
did. Seneca’s solution of the dilemma is simple. Woe has driven Medea
insane and the ghost of her brother hovers before her, a symbol of that
insanity. Accordingly, it is in a fit of madness that she does the deed.
In Seneca, as in Euripides, the action follows the ancient myth, but the
interpretation of that myth varies with the author, and in both cases
this reinterpretation is not so much an invention of the dramatist as
a reflection of the changed point of view of the society of his time.
The moderns have, of course, felt the same need for a re-reading of
the story as the widely differing versions of Grillparzer and Catulle
Mendéz demonstrate. This is but one simple illustration of how the
Roman dramatists could re-stage old myths and yet constantly invite the
audience to something new. The emphasis upon the interpretation rather
than upon the plot is precisely the same as it was in the days of Racine
and Shakespeare.
How far the Roman dramatists were indebted to predecessors for their very
striking employment of song is still a moot problem. Leo,[9] following
a suggestion of Crusius, held that the Plautine cantica followed the
manner of the contemporary music-hall lyrics of Greece as illustrated by
the then recently discovered “Grenfell song.” This theory was rejected by
Fraenkel[10] because he found no vital similarity between the Grenfell
fragment and the Plautine cantica. In his view the Roman predecessors of
Plautus—Livius and Naevius—who paraphrased both tragedy and comedy, had
probably developed the cantica in tragedy from Euripidean models and then
employed them in comedy as well. This theory has a certain plausibility
but cannot yet be tested because the cogent examples of cantica in
tragedy must be drawn from Ennius, who was not a predecessor but rather a
tardy contemporary of Plautus. The view of Leo has received some little
support from a brief and peculiar mime-fragment of the British Museum
recently published by Milne.[11] However this fragment is so late that
it may represent post-Plautine developments, and therefore cannot be
pressed into decisive service. It must also be added that recent studies
tend to show that Greek New Comedy of the time of Menander had not wholly
given up the use of strophic lyrics,[12] and that the Plautine and Ennian
cantica themselves seem to have retained not a few traces of strophic
structure.[13]
Without attempting to solve a problem for which too many of the
quantities are still unknown, I would only wish to suggest the need of
considering the practical factors of Roman experience and of Roman
exigencies when we try to explain the Roman trend toward an operatic
form. In the first place it is well to keep in mind that Naevius, who
dominated the Roman theater for thirty years of its formative period,
had campaigned in Sicily long enough to become the first annalist of
the Punic war. Practically every city of Sicily where Roman troops
were stationed had a theater, and in the days of Hiero the demand for
dramatic entertainment in Sicily was so vigorous that new theaters were
being built. We still have evidence[14] of Hellenistic theaters at
Syracuse, Tauromenium, Segesta, Tyndaris, Akrae, Catania, and Agyrion.
It is agreed that the Greek tragedies and comedies that were then being
produced—the plays that Naevius probably saw—were generally devoid of
choruses. The elaborate choruses of the tragedies had fallen away, partly
because of the cost of staging them, and partly doubtless because new
musical fashions had grown impatient of the somewhat academic formalism
of the strophic songs.[15] In comedy, considerations of the expense
and a desire for scope and freedom in choosing theme and form in song
worked toward the same end. There can be little doubt that in Sicily
Naevius saw performances of post-classical tragedies and comedies, not to
mention music-hall performances of mimes and farces, that gave him good
suggestions as to how the plays of Euripides could be staged without a
chorus, and how a paraphrase of a Menandrian comedy that had lost its
_entr’acte_ songs could be turned into something like light opera. And a
genius as inventive and independent as Naevius would soon break through
the limitations of the Roman stage and shape, with the help of such
suggestions, a performance suited to Roman needs.
But even if the Sicilian performances offered suggestions of how to
stage comedies and tragedies without choruses it seems to have been
the Romans who made the old classics conform to the new method and
in doing so greatly enlarged rather than diminished the scope of the
musical accompaniment. The second reason for this increase in songs
seems to me, therefore, to lie in the need for music to help carry the
new meters which dramatic writing demanded. Latin had been as poor in
meters as early English was later. The chief drudge of all work had been
the Saturnian verse, a form unfit for either sustained narrative or for
realistic dialogue. Its line was slow and reflective. It had been used
for ritual song, for funeral elegy, for lullabies, for gnomic poetry, and
apparently also for lampoons; but it was as unfit for the drama as Ennius
had found it to be for epic narration. There was also apparently a lively
marching verse, the quadratus, the meter with which we are familiar from
the trochaic tetrameters of the Greeks and from the lines of Tennyson’s
_Locksley Hall_:
With the standards of the peoples plunging through the thunder storm.
At least critics are now ready to accept the remark of Horace that lines
like
Rex erit qui recte faciet, qui non faciet, non erit,
were sung in the days of old Camillus. Whenever we happen to have a
fragment of a soldier’s song quoted in Latin it is in this quick step:
Ecce Caesar nunc triumphat, qui subegit Gallias.
That meter had possibilities in the drama, and it was very freely used,
though it doubtless had to be weaned away from its boisterous military
associations. For rapid action and excitement it served well. It appears
that early tragedy felt that it belonged to music and used it in lyric
passages, in recitative chants as well as in dramatic speeches. Naevius
was very fond of it.
Tragedy, however, needed an easy line of moderate length for its ordinary
dialogues, and several meters in different moods to carry the monologues,
songs, and emotional dialogues. For these Livius and Naevius, as we have
noticed, had taken over and adapted a large number of Greek verse-forms.
Now the adaptation of a foreign meter is a very serious matter. It
took English poetry hundreds of years to merge French and old English
rhythms, as it took France centuries to find a satisfactory adaptation
of the medieval Latin systems. The labor of reshaping Greek meters for
use in Latin was all the more difficult at the time because the Latin
language happened to be just then at a critical point in its accentual
development. The Greek word-accent had but very slight stress, so that
quantity was permitted to determine verse-rhythm. In Latin, also, the
quantity of the vowel and the syllable was still the dominant element
at this time, indeed determined the position of the word accent, and
was responsible for the penultimate accent rule that prevailed in most
words during the century in which Naevius wrote. Latin must have been
nearly as precise in the observance of longs and shorts as Greek. But
the difficulty was that the stress of the word accent had also been a
marked factor in Latin pronunciation for some time. Now in forming or
introducing new rhythms the Latin poets would have to choose either
stress or quantity as the decisive element on which to build and force
the other element to comply. This is a choice that very few languages
have imposed upon their poets. In English there was of course no such
decision necessary since our accent remained a strong stress while our
syllabic quantities, in the mingling of Germanic and French, became
so completely confused that the values of half of them are hardly
determinable by ear. This difference between Latin and English has not
always been given due weight. When, for instance, the late Poet Laureate
of England assumed that the quantitative meters of Ennius and Vergil
resemble in effect the quantitative meters that he composed in English,
he disregarded the vital difference between the two languages.[16] While
in Latin quantities were readily distinguished even by the rabble, a fact
that is shown by the emergence of the penultimate law before there were
any teachers, in English it requires a laboratory apparatus to decide
what really is the length of certain syllables. On the other hand, stress
is dominant in English and unmistakable in all colloquial speech, whereas
in Latin it was so moderate in the new position it had recently acquired
that for many centuries after Plautus it had very little effect upon
the morphology of the language. Apparently the first Roman poets chose
as wisely as could be expected in determining to base their meters upon
quantity rather than upon word-stress. But in doing so they had to face
a serious dilemma: a stress-accent does not like to be disregarded, and
ultimately (six centuries later) it asserted itself and insisted upon
dominance. The quadratus, or trochaic tetrameter, which apparently grew
up before the Romans knew Greek or grammar, had made a compromise that
satisfied the ear. It looked to quantity as the dominant element, placing
the verse-beat invariably upon a long (or its equivalent), but it by no
means wholly disregarded word accent. In the lines of soldiers’ songs
that survived, it is not often that word accent is slighted more than
once in a line, and Ennius, Naevius, and Plautus in their plays seldom
permitted themselves to neglect it more often than twice in a spoken line.
In “Rex erit qui recte faciet, qui non faciet, non erit,” aside from
the last syllable which of course is hidden in a falling cadence, only
_erit_ at the beginning, an unemphatic word, gets what may be called a
mechanical accent. But this smoothness is natural chiefly in the trochaic
meter and it occurs here because the normal penultimate accent of Latin,
which stresses a long syllable next to the final, is by nature adapted
to a trochaic quantitative rhythm. Obviously an iambic line can take
advantage of all the qualities of the trochaic line if the poet will
so adapt the first word as to secure a trochaic swing in the rest of
the line. Livius was very skilful in adapting the Greek trimeter to the
spirit of the Latin trochaic. He increased the caesuras—that is he freely
cut the iambic foot in two—not for the sake of caesuras but in order that
by cutting iambic feet he could create a trochaic rhythm which would
operate easily with a penultimate accent; he permitted resolved longs
in any position except the last foot, because when the penult is short
the antepenult receives the accent, and a fair coincidence of accent and
ictus is again secured; finally, since there was no way of avoiding a
slight clash in the sixth iambic foot, he frequently tempered the fifth
foot by insisting that when it contained a single word, this word must be
spondaic. That is, by dwelling upon the first syllable of the fifth foot
he reduced the ictus on the second.[17] The result of this exceedingly
delicate modulation of the line by Livius—a modulation revealing an
astonishingly keen ear—was that the dramatic senarius in Latin had a
rhythm in which quantitative and accentual beats usually coincided, and
this rhythm served its purpose in Latin drama quite as effectively as did
the trimeter in Greek. Considering the gentleness of the accent in Latin
we may surmise that Latin dramatic senarii, when thus treated, ran at
least as smoothly as Browning’s blank verse despite the fact that they
had to give heed to accent as well as to quantity.
In teaching the rules of the Latin senarius it is a pedagogical mistake
to compare it with the Greek trimeter as Lindsay does in his brilliant
book, _Early Latin Verse_; indeed I am persuaded that it distorts
historical facts to do so. If Livius was the man who shaped this line for
Latin needs, we must remember that he had reached Rome as a mere child
and had as a youth grown accustomed to the swing of verse pronounced in
the Saturnian and the quadratus meters and that he would not have had
any occasion at Rome to learn to comprehend the amazing precision of
the Menandrian trimeter. And Naevius, the Campanian soldier, must have
had much the same experience. To such men the Greek trimeter could only
have suggested the possibility of writing a six-foot iambic line which
would carry through to the end, with the lightness of the quadratus, the
opening rhythm of the Saturnian. And the rules of the first hemistich
of the Saturnian must have been the determining regulations of the
senarius. Those rules had all to do with the purely Latin problem of
writing quantitative verse that should not overmuch offend the demands of
an accentual stress. Indeed it is fair to say that if Livius had never
seen a Greek trimeter but had undertaken to adapt a six-foot iambic line
on suggestions taken only from the Saturnian and the quadratus, he would
have arrived at precisely what he did. By failing to see this simple
historical sequence we have, from Bentley to the elaborate but misleading
statistics of Klotz, followed Horace in misconceiving the spirit of the
very worthy Latin senarius.
But there was more for the early dramatists to do than to shape a line
suitable for dialogue, for Greek drama had taught these poets that a
great variety of meters must be used to give the mood and tempo of
emotional scenes. The Roman writers of tragedy did not attempt to
reproduce the intricate polymetric and antistrophic Greek songs. However,
they adopted several very effective meters (perhaps also creating some)
which they used for massed effects, such rhythms as the cretic, bacchiac,
anapaestic, glyconic, and the longer iambic and trochaic lines, not to
mention various rarer forms. In a fragment of Ennius quoted by Cicero,
Andromache in distress runs from senarii through a passage of pleading
cretics:
(Quid petam praesidi aut exequar quove nunc etc.)
then through excited narration in excellent alliterative septenarii:
(Fana flamma deflagrata tosti alti stant parietes)
into turgid and wild anapaests:
(Priamo vi vitam evitari etc.).
And Cassandra’s mad scene runs similarly from septenarii through dactylic
tetrameters, trochaic octonarii, and anapaests into iambic octonarii.
The tone of such cretics has been caught fairly well in Tennyson’s _The
Oak_,
All his leaves, fall’n at length,
while the bacchiac rhythm is, if pronounced with care, conveyed by
Arnold’s
Ye storm-winds, of autumn
These brief experiments on the part of English poets, which show an
observance of word-stress and also of quantity, will indicate the
nature of the difficult task which Latin poetry had to face in taking
over meters native to the Greek language, except that the Latin
poet, conversely, must place his verse ictus on a long syllable and
secondarily, if possible, observe the word stress as well. That was a
difficulty with which classical Greek did not have to contend, since its
word accent was musical and could easily be slighted. German and English
poetry—except in learned experiments—has refused to face the double task,
a task which has fortunately never been compulsory.
If we keep these facts in mind I think we may be willing to concede that
the Latin poets of the early time may have called in the extended aid
of the flute and of melody partly in order to obscure the occasionally
inevitable conflict between the word accent and verse ictus. The point
can be illustrated by a simple example. In Tennyson’s song “Blow, bugle,
blow,” the line
And the wild cataract leaps in glory,
which falls unrhythmically in the midst of an iambic system, hides its
confusion when sung in regular three-fourths time. The flute or violin,
unlike any of the percussion instruments, does not convey a stressing
tone, it measures notes and carries a quantitative rhythm readily,
thereby obscuring any word accents that fall irregularly.
It is my belief that when the drama came into Rome and found the language
just at the point where the quantitative principle was having its
conflict for dominance with the accentual factor, a moment when the task
of shaping adequate rhythms for new forms would be very difficult, it did
the natural thing, accepted quantity as dominant, attempted at the same
time to observe the word stress, and then hid occasional discrepancies by
using song and recitative freely. And this, it seems to me, is one of the
reasons why Roman tragedy was the more willing to go in the direction of
modern opera.
If a recent theory concerning French verse be true, we may find there
an instructive parallel. It has been suggested that when medieval Latin
verse floundered between quantity and accent, early French verse, unable
to find usable quantitative distinctions and hampered by a monotonous
word accent, hesitated for a dominant principle, and allowed the singing
line with its counted notes to assume control. Whether or not this is
the reason, at any rate the French lyric emerged with its isosyllabic
lines and fluid ictus, and in so far provides a partial parallel to what
happened in Latin verse.
It is not improbable that, if the Romans had come in contact with
culture a century later than they did, so that the Latin accent might
have affected colloquial morphology unhindered by literature and
sophistication for another century, native poetry might have abandoned
its quantitative basis and frankly accepted word accent as the most vital
factor of its rhythm. It would perhaps have been a liberating influence
had this happened. As it was, by their use of music and by their
reasonable compromise with Greek meters, the early poets accustomed the
Roman ear to slight the claims of accent, and Ennius was able to compose
spoken lines in hexameters which almost entirely followed the dictates
of quantity. Once completely naturalized, this method was no longer
questioned, and Lucretius, Horace, and Vergil—except at line ends—could
safely disregard the word accents. It was the musical part of the drama
that had naturalized such principles of rhythm.
After Accius the writing of tragedy fell off as rapidly at Rome as it had
in Greece after the conquests of Alexander. How is this to be explained?
Why did not England produce great tragedies after the successes of the
Elizabethan stage, or France for a long time after the classical period,
or why did not America during the two centuries of play-writing before
1900 beget a single great dramatist? Recently there was published a list
of the American plays copyrighted in Washington between 1870 and 1920; it
contains over 60,000 titles. How many of these have become a part of the
world’s literature? Probably not one in 10,000. Can we explain why?
It is not well to be dogmatic in discussing the reasons for such a
phenomenon as the decline of tragedy at Rome, but we may be permitted
perhaps to repeat some conjectures. We have already remarked[18] that
the second century B.C. was a period of striking social changes, of a
decrease in the middle class native stock and a very remarkable increase
in the slave population, and from this slave population there grew up at
Rome the new generation of proletariat citizens that had to be amused at
festival seasons. It was a population that was probably as intelligent
as the old, but it had hitherto been brought up in slavery and in the
devotion to material advancement that slavery implies. These new Romans
could hardly be expected to concern themselves with the quality of the
entertainment provided, with civic ideals and artistic standards. In
Cicero’s day the games at festivals were more frequently gladiatorial
shows and wild beast hunts. To freedmen and freedmen’s sons these
seemed to provide what Aristotle called tragic purgation somewhat more
effectively than did representations of the _Medea_, _Orestes_, and
_Oedipus_. It is apparent that if society was to continue in its course
of degeneration the exacting tragedy of the old type was doomed.
Nevertheless, the old plays were being revived by men who were interested
in high standards, and when a famous actor played a part he would draw
large audiences. Aesopus and Roscius, the best actors of Cicero’s day,
were in great demand and both grew rich at their profession. Though
references to dramatic performances in Cicero’s day are casual, we hear
of not a few. We know, for instance, that there were reproductions of
Ennius’ plays a century after his death, and we find in the list his
_Andromache_, _Telamo_, _Thyestes_, the _Alcumeo_, the _Iphigenia_ and
the _Hector_. Of Pacuvius’ plays Cicero had seen the _Antiope_, the
_Iliona_, and a play about Orestes which he describes as a favorite of
the gallery. Accius was even more popular. Aesopus produced his _Atreus_
repeatedly. His _Eurysaces_ was given in 57 B.C., the _Clytemestra_ in
55, and the _Tereus_ in 44 after the authorities had suppressed the
_Brutus_ because of its political significance. And there were many more.
This success of the old plays—artificial though it may have been in some
instances—shows that respectable audiences could still be reckoned on so
long as the Republic lasted, and that the plays were attractive enough
to justify the aediles in presenting them. With the Empire, however, the
decline was rapid; the populace found the tragedies tedious, and when in
Horace’s day a popular actor discovered a way of cutting the plays and
presenting the more effective scenes in pantomime, with a lavish amount
of music and a gorgeous setting, legitimate tragedy gave way to something
resembling a Russian ballet. Old tragedies were cut and adapted for this
new kind of presentation and new ones were written that consisted chiefly
of scenarios and monologues. Even closet plays, like Seneca’s, were
shaped into a succession of recitations in the hope that they might sell
to the new industry. Literary tragedy, however, had come to its end at
Rome.
This process of decay was natural enough and was only to be expected,
given the changes in Rome’s society and with them the decline of Roman
ideals. But it is still somewhat of a riddle why at Rome as well as at
Athens good playwrights ceased to write a hundred years before tragedy
ceased to attract respectable audiences. It would seem as if the art of
writing plays lost its stimulus even before the plays themselves ceased
to please. The reason for this may well be that tragedy kept too long to
its convention of interpreting sacred myths. The themes were outworn, and
each myth had had every human interest exploited by the time that several
writers had given it their several interpretations.
Today it would seem quite the obvious thing to have dramatized fictitious
experience, even as comedy had long ago learned to do. But a moment’s
reflection will show that to assume that this might have been done
involves an anachronism. Greece did not take this step after Euripides,
for Agathon’s experiment was not followed, nor France for some time after
the classical period, nor England after the Elizabethan successes, and
conditions at Rome in the days of Accius were in many respects analogous
to those in the countries named. Though the dramatic instinct seems
always to be presumable, the drama depends upon social conditions and
must draw its life from that which society provides. Its evolution has
accordingly been a fairly consistent story. Early tragedy assumes the
rôle of interpreting the most sacred and time-honored of a nation’s
stories. The sufferings, thoughts, emotions of the great—heroes,
demigods, and kings—are worthy of presentation, and these alone. At
first the tale must not be altered, it must be told as nearly as possible
in the way that tradition has hallowed. As time goes on, however, and
men have changed, the tale thus told will seem inconsonant with human
nature; then the dramatist may re-tell it, suppressing what has grown
obsolete, emphasizing the elements that still seem true to experience.
A very daring realist will venture to present Telephus in tatters, but
the critics will be upon his heels immediately. For the hero will remind
you of a beggar, and it would be desecration to set mere man upon the
stage made for the demigods. Common man belongs in comedy; you may laugh
at him and with him, but life’s great lessons are illustrated only in
the characters of the great. And that is where Euripides stopped—was
doubtless compelled to stop. And it is nearly where Shakespeare found
the outward boundary of his tragedies. His tragic plots derive from old
Chronicles or from Ancient Rome, or from foreign lands sufficiently
removed from his audience by mists of unknown space to make them suitably
heroic. His tragic characters never represent the men of contemporary
England. They are as real and human as the man of the street, to be sure;
but that is after all not the same thing. Try to imagine the heroines
of Ibsen or Pinero or O’Neill upon the stage of the Globe Theatre in
Shakespeare’s day! The Elizabethan conception of the function of tragedy
makes such heroines unthinkable except in comic rôles.
Realistic tragedy is of course a thing of slow growth, or perhaps we
should say that a nation fits itself slowly for the reception of it.
Comedy paves the way somewhat. When the great may not be laughed at,
it is well that comedy should present the foibles and deformities of
the common man, if it be merely for ridicule. Slaves served the purpose
of comedy for Menander and Plautus, though they were careful not to
compromise the dignity of their art by giving title rôles to such humble
fellows. Yet as a matter of fact the study of mean subjects contributed
directly and very largely to the understanding of the ordinary character
as material for tragedy. Shakespeare’s portraiture of Shylock, for
example, carried him so far that modern critics do not know where comedy
ends and tragedy begins. In the _Andria_, the _Hecyra_, and the _Heauton_
of Terence the emotion shifts more than once from laughter to deep
sympathy. But something more was needed than the dramatist’s study of
the man of the street. Human society must itself change. It is not an
accident that genuine realistic tragedy failed to find its fully accepted
place upon the stage till the nineteenth century, in a word not till
thoroughgoing democracy, by preaching the equality of men, had persuaded
us of the dignity of the mere human being, and through the prose novel
taught the man on the street to concern himself with his fellows as
worthy themes of art. That was a stage of democratic realism which Rome
did not reach while the literary art was still creative. And therein
probably lies the final explanation of the slow failure of Roman tragedy.
FOOTNOTES
[1] _Am. Jour. Phil._, 1927, 105.
[2] Livius and Naevius were both very fond of the septenarii; the iambic
tetrameter appears in the tragic fragments of Naevius once; cretics are
found in the _Equos Trojanus_, and bacchiacs apparently in Naevius’
_Danae_ and in his _Lycurgus_. Fraenkel, _Hermes_ (1927), 357 ff., has
shown that the trochaic septenarius (quadratus) was an old Latin meter.
We need not, however, assume with him that it was derived from the
Greek. As a marching rhythm it is too natural to require explanation.
The assumption of an Indo-European _Urvers_ needs to be exiled from our
books. Song and dance are very old.
[3] See _Cambridge Ancient History_, VII, 644.
[4] See Duckett, _Studies in Ennius_, 56, who revises the views of Leo,
_De Tragoedia Romana_ (Göttingen, 1910).
[5] For a strophic system in Ennius, see Crusius, _Philologus_, Supp.
XXI, 114.
[6] _Gram. Lat. Keil_, VI, 77, 7; Vollmer, _Röm. Metrik_, in Gercke’s
_Einleitung_, I, 8, p. 6; however among the preserved fragments of
Pacuvius there are several anapaests that resemble those of Ennius.
[7] Ennius, ed. Vahlen, _Scaenica_, 272.
[8] See _Am. Jour. Phil._ 1913, 326.
[9] Leo, _Die plautinischen Cantica_ (1897).
[10] Fraenkel, _Plautinisches im Plautus_ (1922), criticized by Immisch,
_Sitz. Heid. Akad._ 1923.
[11] Milne, _Cat. of lit. pap. in British Mus._ 1927 (no. 52); cf. Wuest
and Croenert, _Philol._ 1928, 153 ff.
[12] See Marx’s ed. of _Rudens_, 254 ff.
[13] Crusius, _Die Responsion in den plaut. Cantica_ (1929).
[14] See Bieber, _Denkmäler d. Theaterwesen_ and Bulle, _Abh. Bayer.
Akad._ 1928.
[15] If Horace’s strictures on the new music of the drama in the
_Ars Poet._ 200-15 took a hint from Neoptolemus, we may suppose that
Hellenistic critics had objected to this change.
[16] Robert Bridges, _Ibant Obscuri_. Such hexameters as
They were amid the shadows by night in loneliness obscure
Walking forth i’ the void and vasty dominyon of Ades:
As by an uncertain moonray secretly illumin’d—
do not represent what happened to Latin in Ennius, for the reason that
in Latin pronunciation the quantity was the dominant element controlling
even the accent. In English the reverse is true. Fraenkel, _Iktus und
Akzent_, has recently committed a similar mistake in judgment, influenced
apparently by the high respect that speakers of German must necessarily
have for stress. He has resorted to daring hypotheses in trying to
prove that Plautus always correctly observes a species of stress (see
Sonnenschein in _Class. Quart._, 1929, 81). It is significant that the
French, who feel little stress in their diction, go to the other extreme
and find stress insignificant in Latin. Latin in fact was like neither;
it resembled Hungarian in being primarily quantitative, and in its word
accent had a moderate stress not without a rather noticeable pitch such
as is found in some parts of Sweden.
[17] See Lindsay, _Early Latin Verse_, Leo, _Geschichte Lat. Lit._, p.
68. Fraenkel, _Iktus und Akzent_, seems to me only to have confused the
results that have been summarized with consummate skill and good sense by
Lindsay.
[18] In chap. I.