Amos Philip McMahon. On The Second Book of Aristotle's Poetics and The Source of Theophrastus' Defnition of Tragedy

Harvard Studies in Classical Philology 28 (1917)
ON T H E SECOND BOOK OF ARISTOTLE'S POETICS
AND T H E SOURCE OF THEOPHRASTUS'
DEFINITION OF TRAGEDY
INCE the Renaissance any treatment of Aristotle's Poetics has

discussed and lamented the loss of a second book. Because this
book, as we shall see, is supposed to have contained a theory of comedy,
-its loss, measured by the value of the Aristotelian theory of tragedy,
is incalculable. An attempt to investigate the facts on which the
belief in this loss is based, and to determine its reliability is, therefore,
of fundamental importance.
The belief rests on the observation that the Poetics, as it is now
constituted, is incomplete, or rather fails to fulfill its apparent programme, being especially deficient in a symmetrical elaboration of
its initial divisions. Such incompleteness is usually defined as the
lack of an entire second book.
A direct statement that there were two books is to be found in the
list of Aristotle's works given by Diogenes Laertius in his life of the
phiios~pher.~By this evidence, Bywater holds? " the fact is s&iciently assured," although " we have no further direct testimony to
the existence of a Second Book." There is, however, the statement
of the so-called Anonymus Menagii, to be identified probably with
The materials in this article formed part of a dissertation, The Mediaeval Conception of Comedy and Tragedy, submitted in partial fuliilment of the requirements
for the degree of Ph.D. at Harvard University in 1916. I am under obligations for
suggestions and help, to Professors C. N. Jackson, W. H. Schofield, and especially
to Professor E. K. Rand.
a V. Rose, Aristotelis qui ferebantur librmum fragments (Teubner), Leipzig, 1886,
p. 6, 1. 83.
a Ingram Bywater, Arislotle m the Art of Poetry, Oxford, 1909,p. xx.
A . Philip McMahon
Hesychius,' where two books are m e n t i ~ n e d . ~This evidence is also

accepted by von Christ as establishing the existence of the second
book.3 I n the third place, there is the additional testimony of the only
other early index of Aristotle's works, in the "fragments of a philosopher of Ptolemy's reign," where there is mention of two books
" de arte poetica secundum disciplinam Pythagorae eiusque sectatorurn
placita."
This is explained by Wenrich: and accepted by Rose: as
being a confusion between Aristotle's two books of the Poetics and a
work on Pythagoras, but nevertheless corroborating the evidence of
the other two lists.
Zeller cites the evidence of all three lists to prove that the Poetics
as we have it is only a fragment? The relation of the three is declared
by Rose to be as follows: Andronicus made a list of the works of
Aristotle in his books on the Aristotelian philosophy; an unknown
philosopher of the time of Ptolemy did the same, with certain changes,
but using the same basis; Favorinus then derived from Andronicus in
his commentaries, upon which Diogenes and also Hesychius relied.
Although the direct assertions of the existence of two books are

comparatively few, many statements indicate that there were more
than one. Most important in this class of evidence is, of course, that
to be found in the other works of the philosopher, especially in the
Rhetoric. I n all of these the definite article is used in the plural when
reference is made to the Poetics.
I n the first book of the Rhetoric Aristotle refers to the books of the
Poetics for a discussion of the ridiculo~s.~In the third book of the
Rhetoric, again with apparent reference to the ridiculous, a treatment
E. Zeller, Aristotle atzd tke Earlier Peripatetics, trans. f r o m Philosophy of the
Greeks by B. F. C. Costelloe and J . H . Muirhead, 2 vals., London, 1897, i, p. 48, n.3.
2 Rose, Fragm., pp. 13, 75.
3 W. v o n Christ, Geschichte der griechiscken Litteratur, 5 t h ed., W . Schmid,
Munich, 1908, i , p. 702, n . 4.
V . Rose, Aristoteles Pseudepigraphus, Leipzig, 1863, pp. 79, 80.
J. G. Wenrich, De Azrctorum Graecorum zersionibus . . commentariis commentatio, Leipzig, 1842, p. 143.
8 Rose, op. cit., pp. 8 ff.
Rose, Arist. Pseud., p. 194.
9 Rhetoric, I , 1 1 , 1371 B 33.
Zeller, op. sit., i, p. 1 0 2 , n. 2.
A Lost Book o j the Poetics
of that subject is omitted on the ground that it already exists in the

books of the P0etics.l In the same part of the Rhetoric a cross-reference
is given to the books of the Poetics -to chapter 2 2 , SO Jebb believe^.^
A little later in this section another reference is made to the books of
the Poetics - chapter 21 according to Jebb.3 A few paragraphs further
on, another reference is made, for a treatment of diction, to the books
of the Poetics, corresponding, like the two immediately preceding, to
chapters 21 and 22.4 In still another place we have a reference, like
that in the first, to the books of the Poetics for a treatment of the
ridiculo~s.~
I n the Politics, also, Aristotle, speaking of the term ~hoapuis
promises to discuss it more fully in the books on Poetics! The evidence
presented by all this testimony is certainly of great weight.
Besides the indications of Aristotle's own works we have the
implications of some of the early commentators on the philosopher.
Ammonius, probably of the fifth century, in his work on the De
Interpretatione, refers to the Poetics in the plural? This comment is
taken by Vahlen to refer to chapter 20: and is cited by Zeller to
support the tradition of a lost book.
Boethius also treated the De Interpretatione, translating it once with
a commentary in two books, and again, about 507, with a fuller commentary in six books.1 I n the second work he mentions "Aristoteles
in libris quos de poetica scripsit," l1 which is taken by Zeller to mean
that Boethius knew a Poetics in two booksJ2
Rhetoric, 3 , I , 1404 A 39.
"Rhetoric, 3 , 2 , 1404 B 5 ; R. C. Jebb, T h e Rhetoric of Aristotle, ed. J. E. Sandys,
Cambridge, 1909.
Rhetoric, 3 , 18, 1419 B 2.
Rhetoric, 3 , 2 , 1404 B 26.
6 Politics, 8, 1341 B 39.
Rhetoric, 3 , 2 , 1404 B 37.
-4. Brandis, Scholia i n Aristotelem (Aristotelis Opera, i v ) , Berlin, 1836, p, g g A 12;
A. Busse, A m m o n i i i Aristotelis de Interpretatione Commentarium, I , Berlin, 1897,
P. 13, 1.
J. Vahlen, Aristotelis de Arte Poetica Liber, gd ed., Leipzig, 1885, note o n
9 Zeller, op. cit., i, p. 1 0 2 , n . 2 .
Poetics, 2 0 , 1456 B 2 j ( p . 45).
l o J . E. Sandys, A History of Classical Scholarship from the Sixth Century B. C.
to the E n d of the Middle Ages, Cambridge, 1906, i , p. 253.
l1 Boethius, Commentarii i n librum Aristotelis IIEPI EPMHNEIAZ (Pars posterior
secundam editionem continens), ed. C . Meiser, pp. 6 , 11 ff.: i n libris quos d e poetica
scripsit locutionis partes esse syllabas vel e t i a m coniunctiones tradidit, etc."
l2 Zeller, op., p. 1 0 2 , n . 2 .
'
A . Philip McMahon
The commentator Eustratius is appealed to by Bywater as showing " that there was even in the latest Aristotelian schools some faint
tradition of another Book," for, in his work on the Ethics, he speaks
of a first book of the Poetics, and this may be taken to mean that there
was also another, a second book.
Finally, the fragments of an early anonymous commentator on the
Rhetoric employ the plural when asserting that Aristotle discussed the
ridiculous in the poetic^.^ The evidence on this score is presented by
Vahlen in a series of quotations that refer to Aristotle's treatment of
the ridiculous. The words of the anonymous writer thus prove, according to Ritter, that Aristotle's work On Poets, which was in three books,
was different from the Poetics in two.'
There are in Aristotle and elsewhere grounds for believing that

certain matters spoken of as treated in the Poetics, but not found
there now, were once to be read in a second book. These references
also indicate an order of the works, according to Ritter and others,
who think that the composition of the Poetics is spoken of as a future
work not only in Politics 8, 1341 B 39, but also in the De Interpretatiowe
41 I 7 A 5 .
At any rate, in the passages of the Rhetoric which we have already
cited, the philosopher apparently referred to the work as one already
done. The close relation in general between the Rhetoric and the
Poetics is further established by the large number of correspondences
of different sorts noted by Vah1en.B
Nevertheless, the explanation of katharsis to which the eighth book
of the Politics looks forward does not appear to be sutiticiently given
in Poetics, 6, 2, where the word occurs only in the definition?
There is no treatment of the ridiculous such as we should expect
from various statements in the Rhetoric (I, 11, 1372 A I ; 3, I, 1404 A
39; 3, 18, 1419 B 2). Thus, of the references to the Poetics in the
1
6
6
Bywater, op. cit., p. xxi.
Spengel, Aristotelis Ars Rhetorics, Leipzig, 1867,i, 159,15.
Vahlen, 09. cit., p. 77.
F . Ritter, Aristotelis Poetica, Cologne, 1839,p. xi.

Op. cit., p. vi.
Poetics, 6, 1449B 24.
Vahlen, op. cit., pp. 49 ff., pp. 53 ff.
A Lost Book of the Poetics

Rhetoric,one-half find some counterpart,while there is nothing a t allcorresponding to the other.' Ritter decides such inconsistencies compel us
to conclude that either Aristotle does not mean our Poetics in these passages of the Rhetoric, or else the Poetics as we have it (Poeticam nostram)
" mancam ad nos temporum hominumque iniuria pervenisse."
In the Poetics itself there is a pledge that is not redeemed to the
satisfaction of readers, where Aristotle says that '( we shall speak
later about Comedy." It is natural to suppose that the treatment
of the ridiculous as the basis of comedy would have been found in
the part of the Poetics which discussed comedy, and that if neither
the promise of the Rhetoric nor that of the Poetics was kept, the reason
would be the same: namely, the loss of the second book in which both
were contained. The deficiency is explained on that basis by Gercke?
following Rose ;6 and Bywater, in his footnote on Poetics 6, 1449 B 21,
supplements Aristotle's words with the phrase " in the lost Second
Book of the Poetics." Thus Bywater is enabled to outline the contents of the second book as containing a further treatment of katharsis,
and a discussion of comedy, in which the laughable would have been
analysed as corresponding to the pitiful and the terrible in tragedy?
The exposition of some other matters would probably have found

a place in such a second book, according to most critics: in particular,
certain aspects of the subject of comic diction to which there exist
apparently two references; and, much more important, a defence of the
drama against the censures of Plato.
In his commentary on the Categories, Simplicius quotes Aristotle on
the subject of synonym^.^ That this topic also stood in the second book
E . M . Cope, The Rhetoric of Aristotle, ed. J. E. Sandys, Cambridge, 1877, i ,
p. 224; J. E. C . Welldon, The Rhetoric of Aristotle, London, 1886, p. 85.
* Ritter, op. cit., p. vii.
Poetics, 6 , 1449 B 21.
Gercke, "Aristoteles," Pauly-Wissowa, Real-Encyclopiidie, ii, I , col. 1053 27.
v . Rose, De Aristotelis Librwum Ordine et Auctoritate Commentatio, Berlin,
1854, P. 133.
Bywater, Art of Poetry, p. 149.
Bywater, op. cit., p. xxiii.
A. Brandis, Scholia, p. 43 A 13; Kalbfleisch, Simplicii i n Aristotelis Categorias

Commentam'um, Berlin, 1907, p. 36, 13.
A . Philip McMahon
is a supposition to which both Bywater1 and Vahlen2 incline by placing it among the fragments on comedy immediately succeeding their
texts of the extant Poetics.
The other reference is a puzzle presented by a statement in the
lexicon of the anonymous compiler, called the Antiatticist, published
by Bekker, which may derive in part from O r ~ swhom
, ~ Ritschl placed
as early as the second century of our era, while Reitzenstein assigns
him to the fifth century. The Antiatticist, in defending the use of the
word K V V T ~ T ~cites
TOV
,
Aristotle
in the Poetics.' The portion of the
Poetics proper to this definition was, according to ~ ~ i a t e rthe
, 5 lost
second book.
A defence of the drama against Plato would have been a feature of
absorbing interest in this lost second book. In Aristotle's Poetics, as
Spingarn observes,' scholars of the Renaissance discovered a satisfactory vindication of the claims of poetry against the Platonic and
mediaeval objections. In Plato the objections were grounded in a
metaphysical theory of imitation, interpreted, some would hold, in a
narrow spirit. Since he conceived imitation as mere copying, and
since he held to the theory of the objectivity of ideas, he allowed little
scope for the representative arts.8 If an object in nature is only a
comparatively unreal copy of an eternal objective reality in God, the
reproduction of that object in art is twice removed from reality, and
if certainly false probably dangerous also.
After all, the matter is not entirely settled by the answer of Aristotle
with respect to tragedy. Plato's objection to the drama as exciting
the passions without providing a means of governing them is not fully
1 Bywater, op. cit., p. 93.
2 Vahlen, op. cit., p. 81.
Sandys, 09. cit., i, p. 325.
Cf. Vahlen, op. cit., p. 81; Antiatticista i n Bekkeri anecdotis, 101, 3 2 rtvvr6rarov: 'Aprarori;X~snepi nocl)rw?s. ''76 6k n h v r ~ vK V Y ~ ~ T ~ T O Y . "

The relations of Aristotle in the Poetics to Plato have been well treated, with
results clearly demonstrating the dependence of Aristotle on his master, in: Ch.
Belger, De Aristotele etiam i n Arte Poetica componenda Platonis disci@lo, Diss.,
Berlin, 1872; Georg Finsler, Plalon und die Aristotelische Poetik, Leipzig, 1900.
7 J. E. Spingarn, A History of Literary Criticism i n the Renaissance, nd ed.,
New York, 1908,pp. 18, 19.
B. Bosanquet, A History of Aesthetic, London, 1904, pp. 47-55.
6
refuted by Aristotle's theory of katharsis, as it is usually interpreted.

I t meets the Platonic objection only so far as it is probable that the
possible vicarious aesthetic satisfaction without an inevitable impulse
to moral action will find compensation in the generalizing character
of genuine tragedy, which is one of its inherently moral functions.
That is, tragedy gives occasion for a proper conception of life, and
may thus issue forth in proper actions. But it is still possible for the
passions to be aroused without finding a direct and immediate reaction
appropriate to the character of the passions aroused and at the same
time morally profitable. Clearer intellectual perception is not inevitably succeeded by improved ethical practice, and so James urges
us never to allow ourselves an aesthetic excitation of emotion without
a deliberately beneficial consequence in our actions.'
Whatever may be the correct theory of the effect of comedy on the
spectators, the Platonic objections apply, it would appear, much more
to comedy than to tragedy, and were in fact urged against it in the
first place rather than against tragedy.
That a discussion of all the objections raised by Plato, with definite
refutation of them, clearing away the difKculties just mentioned, was
included in the Poetics is the judgment of many scholars. Although
Vahlen opposed the theory, urged by Heitz, that this treatment was
to be found in the lost ending of the Politics, he supposed that it was
contained in a lost final chapter of the Poetics. Bernays, however,
held that it was in the lamented second book, -a reasonable conclusion in view of the other arguments urged for the existence of that
book.2
5
Victorius,3 the first great editor of the Poetics, was, as Bywater
notes: the first to say that our Poetics is only part of a larger work.
This opinion grew steadily; Zeller states it specifically as the loss of
a second book, supporting his opinion with much of the evidence cited
above.6 Rose, in his work to determine the canon of Aristotle, finally
W. James, Princi9les of Psychology, New York, 1890, ii, chapters 24 and 2 5.
Finsler, Platon, p. 3.
Vettori (P.)(Victorius), Commentationes in primurn librum Aristotelis de Arle
Poetarum, Florence, I 560.
Bywater, op. cit., p. xx.
Zeller, op, cit., i, p. 102, n. 2 .
A. Philip McMahotz
listed the Poetics as having consisted of two books,l and Spengel

also thought that there must have been two books.2 I n this supposition he was in accord with his usual opponent Bernays, who even
asserted that in another treatise we still possess certain fragments
of the second book's treatment of ~ o m e d y . ~Most recently of all,
Bywater has felt justified in drawing up a sketch of the contents of
the second book.4
The loss may be partly explained by the orthodox tradition of the
vicissitudesof Aristotle's manuscripts. Sulla, according to this account,
after quelling the revolt of Apellicon, carried off his library, which
contained Aristotle's autographs that had already languished in the
cave a t Skepsis for a considerable period. Sulla entrusted these
parchments to Tyrannion, and thus the edition of Andronicus was
prepared.6 If these incidents are accurately reported, especially the
story of the unique manuscripts in the cave at Skepsis, these books
ran great risks of destruction or neglect. As it is, there exists a
total gap in the history of all the Aristotelian writings for a full century after Cicero. This is certainly due to the entire loss of all the
commentaries of that period! and a like fate may easily be assigned
to the second book in case it survived previous perils. Rose argues
that it was lost at a very early date since it was not known to the
Arabs, Syrians, or other commentators, and must have perished before
Andronicus, from whom he would date the present state of the text.'
Bernays, however, believes that the second book survived until the
fifth century, chiefly because he finds at that time in Proclus a conception of katharsis, corresponding to his own deductions from Plato
and Aristotle on the question, with which he overthrew the neoclassical doctrine in favor of a more psychological s ~ l u t i o n . ~HatzRose, De Arist. Libr. Ord., p. 241.
H. Diintzer, " D i e Aristotelische Poetik und ihr Verhlltniss zu den Biichern
I I e p i ~ O L ~ T L K ~ S Zeitschrijt
,''
fur die Alterthumswissenschaft (1842),pp. 280,281.
J . Bernays, Dwei Abhandlungen uber die Aristotelische Theorie des Drama, Berlin,
1880. ( I . Grundziige der verlorenen Abhandlung des Aristoteles iiber Wirkung der
Tragodie; 11. Erganzung zu Aristoteles' Poetik.)
6 R . Shute, On the History of the Process by which the Aristotelian Writings Arrived
at their Present Form, Oxford, 1888, pp. 47,48.
0 9 . cit., p. 66.
8 Bywater, op. cit., p. xxi.
7 Rose, De Arist. Libr. Ord., p. 133.

1
feld and Dufour go so far as to make the loss relatively recent,' because
" c'est dans le second livre de cet ouvrage que les cornmentateurs
alexandriis ont puisC la substance de leurs gloses sur les po8tes comiques grecs et de leurs trait& ' de la combdie.' "
The existence of a second book, assumed to be lost, cannot, in the

nature of the case be absolutely disproved. By the logic of such a
situation a universal negative cannot be absolutely proved even of
contemporary facts. With the increase in distance from the time of
the facts considered the difficulty is immensely increased. No amount
of rational consideration applied to a complete collection of the available facts surrounding the point a t issue - facts relatively few after
so long a time - could guarantee the truth of its conclusions. Palimpsests from some remote eastern monastery, or the accidental find
of some archaeologist in Egypt could easily overthrow the perfect
theory of an investigator. It is the obligation of scholars, however,
to erect such frail structures, based on a conscientious survey of all
the evidence, with the humble reservation in every case, that the very
paucity of the evidence must leave the structure frail. Thus, in a
question of the kind we are considering, it is possible to weigh the
value of the evidence and to judge the tradition which asserts that
there was a second book of the Poetics. Then, while we are, by the
conditions of the problem prevented from making a categorical denial,
we can, I feel sure, assert that sufficient reason cannot be shown to
warrant the belief that such a book ever existed. All the conditions
of the problem are more completely satisfied, on the basis of existing
evidence, by the hypothesis that there was no second book of the
Poetics.
To begin with, the whole tradition depends too largely on the evidence of the indices, the value of which, under the scrutiny of close
criticism, can be shown to be only limited.
1
A. Hatzfeld and M . Dufour, La PoLtipue d'rlristote, Lille, 1899, p. vii.
10
A. Philip McMahon
That Andronicus did make a list may be accepted without hesitation on the evidence of Porphyry in his life of P l ~ t i n u s . That
~
these
a i v a ~ ~were
s
copied in turn by Favorinus, from whom Diogenes
Laertius obtained his list, is the theory of scholars as different as Rose
and Bernays, whereas Shute holds that these inferences can be disproved.3 There is a great gulf between admitting that Andronicus did
make a list, and that the lists we have reproduce him, especially when
there is grave doubt about their intrinsic value, and whether their
authorship cannot with safety be assigned to another ancient scholar.
It is hard on general principles to believe that the lack of order and
arrangement in Diogenes's list could have been the result of the otherwise admirable scholarship of Androni~us.~The weight of evidence
shows rather that there is probably no relation between the index
compiled by Andronicus and the one furnished by Diogene~.~
I n the first place, although Diogenes cites Aristotle frequently,
he does not follow his own list, but by implication appears to have
used the same canon as his predecessors and c~ntemporaries.~How
could he refer to the third book of the Poetics, when he had already
said in his list that there were two?7 Indeed, Diogenes's list contains
comparatively few works, among his lengthy enumerations of titles,
that we can now accept as genuinely Arist~telian.~On poetry alone
and he seems
Diogenes ascribes five separate treatises to Ari~totle,~
elsewhere to have confused the Poetics and the dialogue On Poets.lo
In him we see clearly the beginnings of the process by which, through
including forgeries, variant editions of the same work, editions of
separate portions of whole treatises bearing another title, pupils' notes,
enlargements of later commentators, and other accretions, David the
Armenian found a thousand different works ascribed to Aristotle in
the libraries of the Ptolemies."
S h u t e , op. cit., p. 89.
A. Kirchhoff, Plotini Opera, Leipzig, 18j6, p. xxxix.
Shute, op. cit., p. 90.
S h u t e , op. cit., p. 86.

S h u t e , op. cit., p. 80.
Ritter, op. cit., p. x, n.
Zeller, op. cit., i, p. 49.

S h u t e , op. cit., p. 8.
Rose, Fragm. pp. 3 ff: 2 , 83, 118, 119, 136.
10 Rose, Fragm., p. 76; Diog. Laert., 8, 57. Cf. Diog. Laert., 3, 48 ( p p . 77, 78),
and 2146 (P. 79).
S h u t e , op. cit., p. 93.
II
The compiler of the list given by Diogenes, because of its inclusiveness, was probably an Alexandrine scholar.' In this conclusion most
critics, except Rose; agree, especially since Hermippus may be designated its author.8 A clue is afforded by Diogenes himself who,
immediately before giving a list of the works of Theophrastus, cites
Favorinus and distinctly states that the source of Favorinus was
Hermipp~s.~
Since the origin of his list for Aristotle may have been
similar, it would be easy to explain its character, whatever the additions by Favorinus or Diogenes, as merely a librarian's list of the titles
borne by books in a library.
The works mentioned by ancient authors other than the compilers
of these lists, however, generally correspond to, what we now possess,
and Cicero's statement of the range of Aristotle's works squares with
our canon.5 Dionysius of Halicarnassus uses virtually the same text
that we now have," and while Galen's canon is identical with ours
except for a few lost works, the roll of the missing does not include a
second book of the Poetics? Thus, while from the time of Cicero on,
the successors of the editors Tyrannion and Andronicus refer to a
uniform body of works nearly equivalent to our canon, of the works
which Diogenes mentions, hardly any, except the dialogues, can be
identified in the works we possess?
The index of Hesychius is plainly, as Rose points outlgonly a copy
of Diogenes, with the suppression of certain repeated titles, and the
addition of some more names, as incapable of identification as of belief.
Altogether Hesychius managed to accumulate thirteen different titles
which might have had to do with poetry.1
The index of the unknown philosopher in the time of Ptolemy is a t
once dismissed by Bywater and most other recent critics.ll Its devious
Rose tries to maintain the patently inconsistent position that the Aristotelian
works and canon were always just as we have them, and that we have them all.
a Sandys, op. cit., i, pp. 1 2 2 ff.
Shute, op. cit., p. 51. Cf. Cicero, De Fin., 5, 4,9 ff.
6 Sandys, op. cit., i, pp. 279 ff.; Shute, op. cit., p. 67.
Op. cit., p. 77.
8 Op. cit., p. 86.
Rose, Fragm., p. 11,n. I.
In
O p . cit., pp. 11ff.
Bywater, op. cit., p, xx.
2
'
12
A . Philip McMahon
history and the evident consequences of frequent mistranslation make

it practically negligible when the significance of the other lists has
entirely disappeared.
We may now approach the question from another point of view. If
various passages can be adduced to prove that there were two books
of the Poetics, a number can also be brought forward in which one is
assigned as the number of books in that work. Of course, it must be
admitted in fairness, that the singular of the definite article has not
the conclusive force possessed by the use of the plural. It is, indeed,
possible to refer to the book of the Poetics, meaning the one which I
have in mind, without asserting that there is only one book ; but the
use of the plural carries with it the inevitable consequence that there
was more than one book, and not less than two. Yet, if the occurrences
of the singular alone are sdiciently numerous and of value in themselves, their significance cannot be ignored.
Zeller cites Alexander Aphrodisiensis as using the expression Cv TQ
r c p i T O L ~ T L K ~which
S
he takes to indicate that Alexander knew only
one b0ok.l Zeller, however, apparently did not examine the passage
to note the confused reading. In the Berlin edition it was noticed that
Alexander's reference in this same passage to the Rhetoric is to be
found in the Poetics i n ~ t e a d . Following
~
an emendation proposed by
Vahlen, Wallies solved the difficulty by bracketing the reference to the
Poetics, and allowing the mistaken reference to the Rhetoric to stand?
We have here probably only a case where the original mistake of the
author or an early copyist was corrected by a succeeding scribe, to
creep in later along with the mistake. This passage, then, does not
prove that Alexander knew only one book of the Poetics; rather, that
some scribe of the third century or later, knew only one.
David the Armenian? probably in the fifth century, uses the singular in a passage where he speaks of other works by titles in the plural.
H e r m i a ~another
,~
pupil of Syrianus, was the father of Ammonius, the
Zeller, op. cit., i, p. 102, n . z.
I V , (Scholia i n Arist.), zgg B 44.
Wallies, Alexandri Aphrodisiensis in Aristotelis Sophistichos Elenchos Commentarium, Berlin, 1898, pp. 33, 26.
Vahlen, op. cit., p. 3.
4 Sandys, op. cit., i, p. 76.
2

pupil of Proclus, and he uses the expression Qv 74 r e p i ? T O L ~ T L K ~ ~ S >
This is one of the pieces of evidence presented by Zeller to show a
divergence of t r a d i t i ~ n . H
~ ere we have an apparent dserence of
opinion even within a family of scholars, since Ammonius, the son,
used the plural in spite of his father's singular. Hence Zeller's observation that the more ancient authorities were acquainted with two
books and the modern with only one loses some of its force.
When Simplicius, in his reference to synonyms,3 speaks of the book,
it may be that he vaguely recollected that some matters of diction were
taken up in the Poetics. Whatever the value of his citation of Aristotle
in this connection, it is clear that he thought of the Poetics as one book.
Disobedience to his father secured for Ammonius disloyalty in his
own pupil Philoponus, for the latter agreed with his master's progenitor and used the ~ingular.~
An Arabic commentator, Alfarabi: in the tenth century, also used
the singular regarding the Poetics, but his allusion was mistaken.$
Eustratius, it will be remembered, was cited by Bywater to prove
that when that author speaks of a first book of the Poetics, he implies
there was a second. Now, as I shall show later, neither the titles nor
the numbers of books have any definite meaning for us because of the
absolutely conflicting statements with regard to them, so that no
inference can be made from the title to the number or vice-versa, and
thus the mention of a first book in connection with a work called the
Poetics by no means proves that there was more than one book in the
work we have agreed to call the Poetics. The only safe method is to
see whether the content of a given quotation corresponds to the nature
of the works we have agreed to call the Poetics - a technical treatise,or to the work Om Poets -a dialogue. Such examination of the com1
p.
F. Ast, Platonis Phaedrus (contains also the scholia of Hermias), Leipzig, 1810,
111.
* Zeller, op. cit., i, p. 102, n. 2.
Op. cit., i, p. 102, n. 2.

Hayduck, Ioannis Philoponi in Aristotelis De Anima Libros Commentaria,
Berlin, 1897, p. 269,l. 28.
6 Sandys, op. cit., i, p. 395.
6 Vahlen, op. cit., p. 3.
"Alfarabius interprete Schmoeldersio docum. philos.
Arabum p, 21, de demonstratione omnino fallaci disseritur in ipsizcs libro de arte poetua."
4
14
A. Philip McMahon
plete passage to which Bywater refers shows clearly that it points to

what we choose to call Olz P0ets.l If the term first proves anything,
it only shows that there was more than one book in either the Poetics
or Olz Poets, and the balance of evidence inclines towards On Poets.
One of the citations, from an anonymous commentator on the
Rhetoric, is used to prove the existence of a second book, but another
commentator, also anonymous, in paraphrasing his pasiage of the
Rhetoric, uses the singular of the poetic^.^ Both had the text of the
Rhetoric before them, but where our extant version has the plural,
commentators could still vary when their works were composed.
In spite of the reasons brought forward to explain the loss of a

second book of the Poetics, the diiliculties in accounting for this disaster lead more easily to the conclusion that it never existed. The
awkwardness of attempts to make the loss plausible become more
apparent when it is recollected that the tradition assumes definite
shape only since the time of Vettori. Thus Bywater saysJ3" As for
Book 11, one thing is quite clear, that it was wanting in the common
archetype of 2, the MS. before the eighth century Syriac translator,
and Ao, our oldest Greek MS. We cannot fix the date of its disappearance; it is practically certain, however, that the loss must have
occurred during the papyrus period of the text, when Book I1 was still
on a separate roll, so as to be easily detached from Book I, which
was on another roll."
Bywater, however, also acknowledges that there is no evidence to
show that later grammarians had any information about the second
book or the theory of comedy supposed to be contained in it, while
there is, on the other hand, evidence to show that it was unknown to
them. The history of the existing book in classical times is indeed
obscure to the point of ignorance, and although Bywater thinks that
much of the teaching of the Poetics and its terminology were repro1 Heylbut, Eustratii et Michaelis et Anonyma in Ethica Nicomachea Commentaria,
Berlin, 1892, p. 320~1.36. Cf. Bywater, op. cit., p. xxi.
2 Rabe, Anonymi et Stephani in Artem rhetoricam commentaria, Berlin, 1896, p.
259. Cf. p. ix.
3 Bywater, op, cit., p. mi.

duced in the later literature of compilation,l the amount is a t best
small and comparatively late, so that the work must have been either
ignored or little studied.
Neither Dionysius of Halicarnassus nor Quintilian knew the Poetics
at a1112and Bywater confesses that in Byzantine times the second book
was completely f ~ r g o t t e n . ~
Strabo, who flourished a t about the same
time as Dionysius, has nothing to say about Aristotle which can be
verified, not excepting his story d the Aristotelian library; and he
does not mention the poetic^.^
There is, indeed, a passage in Themistius that Vahlen quotes in
his footnote to Poetics 3, 1448 A 33,6 which seems to parallel Aristotle,
at least as far as the coupling of the names of Epicharmus and Phormis
is concerned. In this passage of the Poetics, however, as in some other
ancient writers: the Sicilian origin of comedy is asserted. In 5 , 1449
B 6, the names of Epicharmus and Phormis are joined in this same
connection. The names, however, were bracketed in the text by
Susemihl and succeeding scholars, and the reconstruction of the passage, clearly ungrammatical, was effected with the aid of Themistius?
While this process may be interesting in the annotation of Aristotle,
it does not go far in proving that Themistius derived his knowledge
from the Poetics, or that it was known to him.
Vahlen also quotes a passage from one of the scholia on Dionysius
Thrax to parallel Poetics I , 1447 B 1 8 . ~The point in question is indeed
mentioned by Aristotle, but it is also mentioned by Plato and by other
later authors who do not show any knowledge of the Poetics. Indeed,
by reason of the language used, it is more reasonable to suppose that
if the scholiast derived his idea directly from Aristotle it came from
a passage treating the same theme in his dialogue On poet^.^ I n any
event neither in Themistius nor the scholiast is there any trace of a
second book of the Poetics.
Rose acknowledges that he is entirely unable to explain the loss,
and falls back on the explanation offered by Alexander Neckham for
2
Op. cit., pp. xxiii, xxiv.

Vahlen, op. cit., p. 8, n.
Ritter, op. cit., p. viii.
Bywater, op. cit., p. 123,
Bywater, op. cit., p. xx.
7 Op. cit., pp. 143, 144.
Vahlen, op. cit., p. 6 , n.
Rose, Fragm., p. 76. (No. 7 0 ; Diog. Laert., 8, 57,)
A. Philip McMahon
the disappearance of another work:' "Aristotelis viam universae
carnis ingressurus subtilissima scripta sua iussit secum in sepulcro
recondi, ne utilitati posteritatis suae deservirent."
The whole story of the cave at Skepsis is dubious, and the same
considerations that make us doubt it also render unlikely the disappearance of so notable a work as a second book of the Poetics, containing a theory of comedy. It is, however, not so much the story itself
as the value attached to it that is unwarranted. While the ordinary
causes for the loss of classical works are sufticient, when definite traces
of them at some previous time can be discovered, it has been found
necessary, forsooth, to find a specific disaster for a unique manuscript
to explain the loss of a work of which there is no definite witness elsewhere. That there are no surviving traces of a theory of comedy
derived from a second book of the Poetics will appear in the course of
this investigation, and we have reason to believe that, whatever the
merits of the narrative of Skepsis, none of the scientific treatises of
Aristotle were lost to the Peripatetic school.
Critics rely on Strabo and Athenaeus, with some aid from Plutarch,
to compose the explanation that rests upon the tale of the cave.2
Strabo veers slightly in the course of his narrative, for he begins by
speaking of the libraries of Aristotle and Theophrastus, and then later
talks of the sale of the works of Aristotle and Theophrastus by the
heirs of Neleus to Apellicon of Teos. These books might, indeed, have
been only the collections of these philosophers, but the remainder of
his tale treats them as the original manuscripts of these authors. I n
the face of evidence elsewhere that Aristotelian works were in existence
and that the Aristotelian school enjoyed a continuous career, Strabo's
remarks that the Peripatetics lacked the genuine works of Aristotle
is manifestly incredible.
In one place Athenaeus speaks of a certain Roman Laurentius who
collected the works of Greek authors including " those of Aristotle
and of Neleus, who preserved Aristotle's books, from whom our king
Ptolemy Philadelphus, having bought them all, put them together
with those which he had bought from Athens and Rhodes and brought
them to fair Alexandria." The interest of this city in Aristotle is
a Op. cit., p. 30.
I
O p. cit., p. 134.
Shute, op. cif., pp. 29 ff.
I7
certain enoughll and the passage probably refers to the works of

Aristotle. Athenaeus in another place, however, does not agree with
his own statement, for he says that in the Athenian insurrection
Apellicon took a leading part, a man who was originally a Peripatetic
philosopher and had bought the library of A r i s t ~ t l e . ~
The ready explanation of this state of affairs suggested by Shute
is that " no really published works of Aristotle were lost to the school
meanwhile," and after saying of Aristotle that " Cicero mentions him
over and over again as an author well known to all, and repeatedly
attacked by the Stoics and Epicureans," he rightly exclaims, if this
is oblivion, what is knowledge ? "
Andronicus and Tyrannion, the editors to whom the works of
Aristotle were entrusted by Sulla, do not appear to have thought they
had the autographs of the philosopher. Nor does Cicero? a close
friend of Tyrannion, mention what would have been a great discovery
if the missing manuscripts of the works on which the Peripatetics
depended were suddenly recovered and placed in the hands of immediate friends for editing.6 Indeed, Rome was the centre of Aristotelianism from the time of Cicero forward: and the character of that
philosophy seems always to have been more congenial to the Latin
than to the Greek mind. Not only were the editors Andronicus and
Tyrannion residents of Rome or Romans, but Galen and Boethius,
among the most important names in the study of Peripateticism, also
dwelt in that city. Thus, with the great probability that, if there is
any truth at all in the story of Skepsis, copies of all Aristotle's works
were still in the hands of his students, the loss of a second book of the
Poetics cannot be attributed to the damp and neglect of a cellar.
((
The evidence for a second book afforded by the cross-references in

the works of Aristotle, not only for the number of books, but for the
contents of the second, if such there was, is greatly impaired by a
critical examination of such references in general. As a first step in
this direction it may be observed that, if it is admitted that all the
Op. cit., p. 30.
Op. cU., p 31.
09. cit., pp. 33 ff.
Op. cit., pp. 35 ff.; Cicero, Fin., 4, 28, 79.

Op. cit., p. 52.
A . Philip McMaholz
18
numerous cross-references are genuine, it must be supposed that the

philosopher had a prearranged scheme of his whole system and its
execution from the very beginning, that he carried this plan in his
head both for completed works and for those not written, and that
the titles preconceived by Aristotle were neither changed then, nor
have varied since.l If, however, some of them may be genuine and
some not, then their genuineness or spuriousness will have to be
decided on grounds quite distinct from their mere presence in the text,
and it is hard to see how they can well be used as immediate proof
of the facts to which they refer.2 This argument holds apart from my
observation in a later connection, that neither the titles of the works
nor the indications of the numbers of books have any constant relation
to one another in our special question. Thus a reference to matters as
being in the Poetics is far from proving, if we fail to find them there,
that they were originally in the second book.
I t may be urged against the cross-references in general that they
are over-elaborate; that whereas there are comparatively few to works
we do not possess, these are of a vague or doubtful character; that
some of them between different works could not have been inserted
a t the time of whichever was the earlier; that in them the same work
is referred to by different names; that in some cases references in the
same work to other portions of that work contradict one another
because they cite it as both preceding and following a given portion;
that in some cases the references which act as connecting links between
two adjoining books occur a t the end of one of them and a t the beginning of the other; that the references imply an arrangement of works
in an artificial order which could not have existed until long after the
time of Aristotle; and that they contain serious errors as to the real
meaning of the doctrine which they cite or its relation to the matter
under discussion.
The attempt to defend the references on the score that they were
inserted in a second edition by Aristotle, or that they were repeated
and enlarged with such additions, is more or less futile: for the supposition that Aristotle ever prepared a formal edition of his scientific
works is unfounded, and it is d%cult to imagine a lecturer remembering so huge a course and a t the same time making minute cross1
O p . cit., p. 96.
O p . cit., p.
10.
0 p . cit., p.
26.
I9
references to questions far removed from his discussion both in thought

and order.'
For a full exposition of these charges against the cross-references I
can do no better than refer to Shute, whose fifth chapter entitled
" Of Titles and References " treats the matter thoroughly, substantiating by numerous instances all the difficulties raised above. He
concludes :
" To sum up then, we find the titles of the Aristotelian books did
not arrive a t a fixed condition till some hundred years after the death
of the master; that on the other hand the references assume all the
titles as already fixed during his lifetime; and that even so they are
not explicable, unless we grant further that he deliberately called
several books each by two or three names; that he had planned out
all his books before he began any, and carried all the details of books
both written and unwritten in his head. Even these liberal assumptions will not get rid of all the dii%culties, and the only satisfactory
way of explaining the matter as a whole is to believe that all or the
great majority of the references are post-Aristotelian, and that they
proceed from editors neither of the same date nor altogether in agreement as to the nomenclature and order of precedence of the books."
From what has just preceded we have seen that the personal authority of Aristotle is not to be invoked for any of the references, and
the motive of the editors who inserted most of them was at best to
assure an acceptance of what they believed, rather than to state a
universally acknowledged fact. The mode of composition and publication of Aristotle's works was, however, most favorable for the interpolation of such references. Indeed for their proper understanding his
works required some such aid, so that in spite of natural reverence for
the philosopher's text, scholars early inserted what they saw was
necessary for the intelligent reading of the works, and the practice once
begun was continued without system and with diminishing success.
The statement of the method of composition by Case is one of the
most a d e q ~ a t e . After
~
mentioning several hypotheses which he finds
Op. cit., p. 98.
O p cit., p. 96-116.
Encyclopedia Britannica, 11th ed., Cambridge, 1911,ii, pp. 506 ff.
A . Philip McMahon
20
himself obliged to reject, he continues: " Turning to Aristotle's own

works, we immediately light upon a surprise: Aristotle began his
However early
extant scientific works during Plato's lifetime.
Aristotle began a book, so long as he kept the manuscript, he could
always change it. Finally he died without completing some of his
works, such as the Politics, and notably that work of his whole philosophical career and foundation of his whole philosophy, - the Metaphysics - which, projected in his early criticism of Plato's philosophy
of universal forms, gradually developed into his positive philosophy of
individual substances, but remained unfinished after all. If then
Aristotle was some thirty-five years gradually and simultaneously
composing manuscript discourses into treatises and treatises into
systems, he was pursuing a process which solves beforehand the very
difficulties which have since been found in his writings."
On the question of the publication of Aristotle's works, Case shows
that printing has given us a wrong conception of publication, and that
Greek authors thought of works rather than of books issued in succession on definite dates. Philosophers especially had for their public the
immediate circle of their students, so that "it does not follow that his
own works went beyond his own library and his school. . . . There
is . . no contemporary proof that Aristotle published any part of
his mature philosophical system in his lifetime."
Any chronological ordering of the works is therefore uncertain, and
the hints for such an arrangement given by the cross-references are
useless. The double versions are results of the school, the heads of
which were probably less daring in their emendations of the actual
text than the immediate successors of Aristotle, but found it better for
the prestige of the school that the scientifk works should not be pub1ished.l Thus the distinction between the esoteric and exoteric works
grew up, the latter, which were probably the dialogues in a literary
form prepared for publication by Aristotle, being cast into disrepute
compared with the esoteric works in the possession of the school.
That most of these scientific works did reach a limited publication
later through the eager purchases of Ptolemy and the seizures of Sulla
did not prevent the growth of this tradition, especially in view of the
superior elaboration and originality of these treatises2
1 Shute, op. cit., pp. 8 ff.
09. cit., p. 37.
...
21
While the professors of the Peripatetic school could not compare in

vigorous originality with their master, natural stagnation and systematization made them rely more and more on lectures from the
Aristotelian text aided by increasing familiarity with the whole of his
works. In this way there would be evolved a set of references to facilitate their interpretation.
Again, although ~ r i s t o t l eabandoned the use of the dialogue as a
means of composition, it is probable, even certain, that question and
answer constituted a large part of the means of instruction, and probably the occasion of a more minute development of many points about
which questions were raised in the course of time.' This explains the
frequent anticipatory use of technical terms. If, then, he did not insert
the questions to which a good many passages are plainly answers, it
becomes still less probable that he would have inserted cross-references.
As Shute remarks:
There would be moreover . . . a natural
tendency among editors, who were themselves usually Peripatetics, to
exalt the esoteric and unpublished works above the exoteric and
published ones. We find, as we should expect, that references to
esoteric works are much more common in treatises which were in
vogue all through the period of darkness than in those which may be
supposed to have remained unpublished during that time. . . Thus
they (the Peripatetics) talk of the dialogues under the general and
somewhat contemptuous name of the external doctrine, without taking
the trouble to specify what special dialogue the doctrine is to be
found in."
But one of the chief reasons why any plural reference, apart from
that of the indices, is unreliable is that nearly all the titles of works
in the Aristotelian canon, both as drawn up in the indices and in other
sources, are contained in more than one book. With such a practice
in vogue it is not d a c u l t to see how any vague or general reference
to the Poetics could easily employ the plural article, especially when
to this invitation to err there was added the existence of a dialogue
dealing with related matters, plurality in the number of books in the
((
I t is interesting to note that the dialogue form was employed by Minturno in

his Italian exposition of the Poetics. I n the four books of his work he represents
himself as conversing in succession with Vespasiano Gonzaga, Angelo Constanzo,
Bernardino Rota, and Ferrante Carafa.
'
A . Philip McMahon
22
dialogue, and a confusion of names between the tratise and the dialogue. We are reduced thus for any information about the second
book and its contents to a search for any possible traces of it elsewhere.
The investigation, however, can best be made in connection with a
detailed examination of each of the references to the supposed second
book or its contents.
6
Although the cross-references have in general been discredited, it
is well, if we can, to account for them, and to examine each separately,
as far as it concerns the object of our investigation. Among the
possible sources, the distortion of genuine references is an unreliable
hypothesis for any definite results. Nor are we on solid ground in a
hypothetical work nepi XkEews,' portions of which appear to have been
absorbed into the Rhetoric, and from which some of the passages in the
Poetics, such as that on the language of poetry, may have been taken,
with a general confusion of references after this absorption, and after
this work had itself disappeared. Again, some of the genuine references may have originally alluded to the dialogue, and other references
to them may have been inserted afterwards; these last are more
practical possibilities.
The citation from Boethius De Interpretatione 290, is called a false
translation by Gercke in its use of libris, and in the case of Boethius
it is clear that he used the conception of tragedy which, I hope elsewhere to show, was contained in the dialogue On Poets. The reference
of Boethius, however, corresponds to our treatise of the Poetics, and
yet, although he was acquainted with it, his definition of tragedy, I
believe, is not derived thence. " I n libris " may be a general expression meaning " in one of the books which he wrote about poetry,"
and so may refer to the work On Poets. This possibility is interesting
in view of the fact that while Theophrastus is not mentioned in the
Consolatio Philosophiae, the definition of the dramatic species which
we afterwards find ascribed to that philosopher agrees with the idea
of Boethius. I t might follow that the Theophrastian definition which
Boethius would have found " in libris " of Aristotle was derived by
Theophrastus from Aristotle directly.
1
H. Diels, Ueber das d7itte Buch der Aristotelischen Rhetorik, Berlin, 1886.
Gercke, Pauly-Wissowa, ii,
I.
col. 1053.
23
One of the problems to solve is the reference of the Politics to

katharsis. Though Aristotle had promised to discuss this topic more
carefully in the Poetus, the treatment accorded it in the latter work
is still inadequate. Of course, on account of the manner of composition, a pledge on Aristotle's part, even if the reference is genuine,
guarantees not fultilment but only intention; in fact Finsler find only
one case, and that doubtful, in which a general promise was fulfilled.
Next, a highly technical and obscure theory such as that of katharsis
would be just the one to attract the attention of an anxious editor,
who could easily overstate the case, when all he really had to rely on
was a repetition of the word in Poetics 6. Either his memory was
poor or his intention unscrupulous, for it would appear much more
learned to say " for a fuller treatment of this subject, see the Poetics,"
than, " there is another place where this idea is involved but left
obscure." An editor is susceptible to just such temptations, and to
allow an acknowledged obscurity to remain unilluminated by his
learning may be felt a reflection on his editorial capacity.
That such a theory of katharsis was to be found in the second book
and survived until the fifth century is the hypothesis of Bernays,'
who attributed to Proclus an immediate knowledge of Aristotle's
treatmenL2 The passage of Proclus he interpreted in the light of the
Politics,3 but as Bywater shows: * '' It will be observed that Proclus
refers not only to Aristotle but also to other apologists for the Drama;
it is quite possible, therefore, that it was from one of the latter rather
than Aristotle himself that he derived his knowledge of the Aristotelian
idea of katharsis. And in the context in place of the Aristotelian term
he substitutes as synonyms dr+aiwais and hnbpaars -neither of these
words being found either in this or in any other sense in the extant
writings of Aristotle."
Vahlen goes so far as to affirm that the discussion of katharsis came
after the discussion of comedy in the second book: but Finsler shows
that the source of Proclus's idea was probably Plato, and this probability is increased by the absence of a full treatment which the refer-
Bywater, op. cit., pp. 94, 95. Cf. p. xxi.

Bernays, Z w e i Abhandllmgen, p. 47.
Finsler, op. cit., p. I .
Bywater, op. cit., p. xxi.
O p . cit., p. xxiii.
ence in the Politics promises. The scholarship of Proclus, moreover,

is generally uncritical.'
Farther, the explanation of katharsis which Bernays thought he
discovered repeated in Proclus is doubtful, although it set men t o
thinking on the problem again. In objection to Bernays, Susemihl
and Hicks argue: ('If we say that the ' painful emotion ' of fear
and pity is removed, we are reminded that the definition in the Rhetoric
(2, C 5 , C 8) makes each of these - fear itself and pity itself - a
sort of pain (XQnqT L S ) , although the emotions are generally defined a s
02s gne~a~
XQnq Ka' 7j6ovfi." 3 The proper explanation is certainly
qualitative rather than quantitative, and the discussion by Susemihl
and Hicks reviews the main positions on the question.4
There is the view, which is not very trustworthy, that the Poetics
was intended only for the use of the school in lectures, so that the
explanation of katharsis was oraL6 I t is likely, indeed, that if Aristotle explained katharsis, he did it orally, but we need not therefore
rashly jump a t the conclusion that the Poetics was only a lectureoutline."
Margoliouth, indeed, thinks that the whole question is adequately
expounded in the Problems,7 and even if this work is not authentic, i t
represents an ancient view, and one probably nearer its supposed
source than that of Proclus and Iamblichus. As Finsler points out,
such promises frequently refer to later passages in the same work.&
The Politics to a much greater degree than the Poetics demands a
more complete development of its topics, and if we admit the hypothesis of lost parts of works, or unfinished works, the Politics would be
one of them. If the latter part of the Politics were lost or left unfinished, a later editor might have changed the reference of an earlier
Finsler, op. cit., p. 3 ; Sandys, op. cit., i, p. 373.
F. Susemihl and R. D. Hicks, The Politics of Aristotle, Books I - I Y , London,
1894, p. 652, n. 2.
a Rhet., 2 , I , 8 , 1378 A 2 1 ; Nic. Eth., 2 , 5, 2 , 1105 B 23.
Susemihl and Hicks, Politics, pp. 641ff.; pp. 650 ff.; " Katharsis as an aesthetic
term."
Gercke, Pauly-Wissowa, ii, I , col. 1037.35.
Op.cit., ii, I . col. 1053. 17.
D. S. Margoliouth, The Poetics of Aristotle, London, 1911, p. 60.
Finsler, op. cit., p. 8 and n. 3.

one or of the author himself to another book where the word a t least
0ccurred.l Finsler, indeed, believes that the expression 8v rois n ~ p l
T O L ~ T LasK it
~ ~stands,
S,
refers to another part of the Politics in which
the vexed question of the relation of aesthetic interests, especially of
poetry, to social interests and the state, was disc~ssed.~In any case,
it is clear that a deliberate attack on Plato, such as some critics expect,
would not have been in harmony with the general character of the
Poetics in its close following of that philosopher, and would have
better suited the P02itics.~
The significance of the theory of katharsis was small in Aristotle's
view, and the whole modern assumption of a complete theory of art
in Aristotle is misleading. As Bywater points out: the idea of a theory
of art in general is recent, and goes back to a date no more remote than
Winckelmann and Goethe. Aristotle's ideas on aesthetics were mostly
those current in his own time. The modern preconceptions, the recent
extreme interest in psychology of the physiological kind a t which
katharsis hints, and above all the splendid opportunity for endless
scholarly disputes offered by so prominent an obscurity against
this background, explain the exaggerated present impression of its
importance.
The majority of the references to the Poetics in Aristotle's works are
to be found in the Rlzetoric, and half of these are capable of verscation.
I n addition to this, it is to be noted that none of the references in the
Aristotelian works ever specify a secolzd book. But of those which are
capable of verification, all correspond to passages in chapters 21 and 22.
Chapter 20, immediately preceding, which begins the discussion of diction with a passage on the parts of speech and other grammatical details, Butcher considers " probably interpolated," and he accordingly
brackets it.% Chapters 21 and 2 2 continue the subject with special
reference to poetic diction, and in chapter 21 there is a passage on the
gender of nouns that Butcher also rejects. Although, as we have
seen from the theory on Aristotle's method of composition, the inconsistencies that are here evident do not disprove the genuineness
Op. cit., p. 8 and n.
0 p . cit., p. 8.
2.
0 p . cit., p. 6.
Op. cit., p. vii.
S . H. Butcher, Aristotle's Theory of Poetry and Fine Art, gd ed., I,ondon, 1902,
P. 3.
O p cit., pp. 7 1 ff.
A . Philip iMcMahon
26
of the chapters in which they occur, it must be admitted that such

looseness of construction made very easy the introduction of somewhat irrelevant materials, and the task of an early editor like Andronicus must often have been complicated by short, incomplete passages
which were not closely connected nor capable of close connection with
any of the set treatises. Such a conjecture is aided by the possibility
of the inclusion in these chapters of passages from an earlier rhetorical
work aapi X8tews. Ritter accordingly doubts the validity of the
references in the Rhetoric to such matters in the Poetics, and attacks
the authenticity of chapters 21 and 22.l Gercke doubts the authenticity of the end of chapters 12, 20, and the end of 21. He thinks them
inserted under the influence of the later Stoic grammar in the third
century, and holds the position of chapter 15 d o ~ b t f u l . ~Ritter,
indeed, devotes considerable attention to developing a theory of an
interpolator and abridger in the poetic^.^
Cicero, a t any rate, in his comment on this passage in the Rhetoric
where one of the references to the Poetics occurs, says nothing about
a second book of the Poetics, nothing at all in fact about a parallel
treatment of the matter elsewhere.4
Diels elaborated the theory of the work aepi XCtews spoken of
above.5 He shows that Aristotle must have been the author, and
that it was later combined with the two books on Rhetoric into the
work as we have it. This theory provides some interesting possible
consequences; namely, either that the same person who combined the
aepi X ~ ~ E W
with
S the Rhetoric inserted some of the materials in the
Poetics, and that this enlarged edition of the Poeiics was not, however, the one which gained widest acceptance; or that early criticism
rejected part of the insertions. If the insertions had been made before
the work reached Andronicus, perhaps their partial elimination is due
to him. These conjectures, however, only add to the reasons for holding the evidence of the references from the Rhetoric to the Poetics in
slight esteem.
2
Ritter, o p . cit., pp. 230-243.

Gercke, Pauly-Wissowa, ii, I . col. 1oj3. 48.
Ritter, op. cit., pp. xx ff.
Cicero, De Or.,2 , 58.
Diels, Ueber das dritte Buch, p. 34.
27
The references to a treatment of the ridiculous, moreover, may have

been based on the Ethics, as was largely the case with the discussion
in the Tractatus Cois2iuia.n~~.Indeed, this treatise, manifestly of Peripatetic origin, and showing several different strata in its development
to its present state, may have been thought to be a part of the original
Poetics by some editors, and these references may actually correspond
to it.
It is, however, the pledge, found in Poetics 6, 1449B 2 ; , to deal
with comedy later that, next to the authority of the indices seems to
impress most critics with the necessity for a second book. But this
reference comes in at the head of a new division, in which the philosopher begins the discussion of tragedy alone, so that it was probably
used as a connecting link, perhaps originally inserted in the margin,
but afterwards creeping into the text.
Part of the promise contained in this passage of the Poetics is fulfilled to the satisfaction of critics, - that regarding " the poetry which
imitates in hexameter verse," but the remarks on comedy in chapter 5
do not appear to these persons s a c i e n t for the purpose of the initial
announcement. Aristotle says merely: " I propose to treat of Poetry
in itself and of its various kinds. . . . Epic poetry, and Tragedy,
Comedy also and Dithyrambic poetry, and the music of the flute and
of the lyre in most of their forms, are all in their general conception
modes of imitation."
There is a further possibility that, although the part of the reference
which touches epic was correct, the insertion of the words " and of
comedy" was a further addition, probably by a still later scribe after
the reference to epic had fixed a place for itself in the text. This
sentence comes immediately 'after some remarks on the relations of
tragedy and epic, and following it is the series of chapters on tragedy.
It is not strictly necessary, and without it the treatise as it stands
would be self-suilicient and consistent.
Diintzer points out that the beginning of the fifth chapter indicates
that the philosopher does not intend to discuss comedy f ~ r t h e r .This
~
reference is not a mere transition, but contains an essential step of
progress in the thought of Aristotle, and does not bear the mark of
1
Poetics, I. 1447A 1,2.

Dihtzer, Zeit. f.d. Alt. (1844,pp. 278 ff. Cf.p. 283.
A . Philip McMahon
the scribe. Here the writer implies that he closes the subject of the
nature of comedy by connecting his observations with some preceding
remarks on the object of imitation in comedy. In the treatment of
tragedy, which begins in the next section of this same chapter, he does
not refer back to the deterqination of tragedy's object, as he does for
comedy, for the very reason that he has not closed the subject, but
develops it more fully in the famous definition a t the beginning of
chapter 6. I t is therefore only reasonable to suppose that Aristotle
would not have connected his points a t the beginning of chapter 5 with
his philosophy of imitation, and then have stated his conception of the
ridiculous in this place if he had ever intended to treat it later. This
is a consideration which does not depend on the use of mere references
of transition, but indicates a greater degree of coherence than is usual
in most of the Aristotelian works, a merit which it could the more
easily have possessed had it been originally short and self-contained,
without the addition of a second book.
A further evidence of Aristotle's intention to finish up the treatment
of comedy a t this point is his balancing of various aspects of the subject, and thus a brief statement of why there is no history of comedy
is set against an outline of the evolution of tragedy. Since, however,
to his mind enough had been said about the aesthetics of a dramatic
species in which he was not much interested, he parallels the later
elaboration of his definition of tragedy with the cursory remarks on
comedy a t the beginning of chapter 5.
Still further, a t the end of chapter 2 2 before the consideration of
epic, which is promised in the first words of chapter 6, we read, "concerning tragedy and imitation by means of action this may suflice."
Now, this sentence, by whomever inserted, whether by Aristotle or by
the customary editor with his method of joining together separate
portions of the treatise a t this point, establishes several facts. One
is that the writer judged that the discussion of all drama - " imitation
by means of action "-had been iinished when the treatise reached this
point. Next, that more about comedy did not follow this observation.
Then, except for the words " and of comedy," this reference in chapter 6, whatever its origin, was correct, for the discussion of epic is
delayed by it until after tragedy; when tragedy is done, there is
another transition, summing up the drama as a whole, and the epic
is taken up according to the announcement. Yet just as the latter

reference a t the end of chapter 22 states that here ends the discussion
S TG a p h ~ ~ e p
i vi p + u e ~ sSO
,
of drama, - aepi pkv otv ~payq6ias~ a T1 ~ Cv
the promise about epic, aepi plv otv ~ i j sCv 2[apC~poispipq~i~ijs
is fulfilled, and a t the very end of the treatise we find i t said in conclusion:' "Thus much may suiKce concerning Tragic and Epic
poetry in general."2
As epic represented for Aristotle all the 6iq-yqpa~i~+,
SO did tragedy represent all the forms of poetry which imitate Cv TG a p h ~ ~ e i v .
There is also the obviously awkward connection of comedy with the
designation of epic as the poetry which " imitates in hexameter verse "
-one a conception according to medium employed, and the other
according to the aesthetic principle, if " comedy " means anything in
this context. I t is possible, indeed, reasoning on the basis of 23, I,
that, instead of the word K U ~ ~ ~ the
~ C word
I S , 8iqyqpa~i~ijs
originally
stood in chapter 6,3 while another scribe, misunderstanding or disliking the word, inserted in the margin the word Kwpq6ias, and that
this was substituted for the originaL4
As already noted, there are no traces of the supposed treatment of
comedy from the second book to be found in Varro or any of the
grammarians. There is, indeed, the Tractatus Coislinianus, which
Bernays held contained some reflections of that lost book. He himself, however, pointed out the secondary and derivative character of
this reflection, and from that beginning, scholarly opinion has gone so
far as to say, with Bywater: " There is no evidence to show that the
Op. cit., p. 283.
Poetics, 26. 1462 B 8.
a For in 23. 1459 A I 7 iv[l] phpy corresponds to b ifaphpocs but 6bt)-ft)fiarrtiis
while more suitable than rpayy6ias, is not its equivalent, although found in the cor-
responding position. There has been a good deal of doubt, indeed, as to the proper
reading in 23. 1459 A 17. Butcher, on the basis of 1449B 11 and 1459 B 32, writes
ivl fihpy, although the codices have 4v phpy, and he notes the reading bv CEafihpy
of Heinsius. This last Vahlen also notes, but adopts the reading of the codices.
Bywater agrees with Vahlen in his text.
As Aristotle seems to include all non-dramatic poetry under the term, 6r~yt)parcmj
alone is not the equivalent of that division, while, if 6~vt)parc~ijs
were there instead
of ~w&ias, it would then present a logical indication of the whole field of nondramatic poetry and of that part of it which is actually treated in chapters 23 ff.
DHntzer, op, cit., p. 282.
"water,
op. cit., pp. xxi, xxii.
A . Philip McMahon
later grammarians knew of Book 11, or of the theory of Comedy which
must have formed part of it. But there is evidence showing it to have
been unknown to them. The so-called Tractatus Coislinianus preserves
a definition of comedy, which has no doubt a certain Aristotelian look;
any one can see, however, by simple inspection that it is nothing more
than an adaptation, or rather, as Bernays calls it, a travesty, of the
well-known definition of Tragedy in the existing Poetics."
The enumeration of the parts of comedy is the same as that in
Tzetzes, who appears to have got it from Euclides. But then, if, as
some scholars think, Euclides was a grammarian of the classical period,'
it only proves that neither the compiler of the Tractatus, nor the early
grammarian had seen any second book. Nobody expects Tzetzes to
furnish any reliable evidence on such points.
Nevertheless, while Aristotelian scholars have tested the Tractatus
and found it wanting, the specialists in Aristophanes have in recent
times become aware of it, and accorded it a more hospitable reception
than it ever before received. Starkie relies implicitly on the findings
of Bernays, and asserts: " The value of this fragment was not fully
realized till Bernays demonstrated that it represented a summary,
mutilated and misunderstood in parts, of Aristotle's analysis of the
laughter in comedy. . . Rutherford alone has shown a due appreciation of its value." Later he reproduces with approval3 the definition
of comedy found in the Tractatus, and indulges in a laborious classification of " the various methods of exciting laughter employed by the
writers of old comedy, especially Aristophanes," " according to the
division of Aristotle," i. e., the Tra~tatus.~
What Starkie terms a " due appreciation " of the value of the Tractatus by Rutherford is rather immoderate zeal. He says: " I t is not
that the laughter of comedy had not been properly analyzed. Even
the scrimp and grudging abstract, now sole relic of the section in the
Poetics concerned with Comedy, will convince anybody who keeps it
in his head as he listens to Greek comic ~ p b u w a a t, hat a Greek had
0 9 . cit., p. xxii.
W . J . M . Starkie, The Acharnians of Arislophanes, London, 1909, p. xxxviii.
0 p . cit., p. xl.
Op. cit., p. xxxviii.
6 W . G. Rutherford, A Chapter i n the History of Annotation, being Scholia Aristophanica Vol. 111, London, 1905, p. 435,l. 19.
2
31
indeed read for Greeks the most secret heart of the ' mother of comedy,'
and probe in hand, had made clear wherefore it beat and what it was
made of. . . . But Aristotle thought too much and was too great an
observer to be loved by commentator and rhetor. Living at ease within
their pale of words, it was not likely they would venture outside to be
exposed to the perils and pains of thinking."
Rutherford dilates on the Tractatws and its application to the
methods and interpretation of Aristophanes.'
Except for Kayser's recent treatment, it is Bernays who has offered
the chief detailed discussion of the Tractatus; a discussion which is the
fountain-head of the whole modem belief in the second book as a fact
established to the satisfaction of scholars. He " demonstrated," as
Starkie puts it, " that it represented a summary " only by employing
in an extremely bold and often unwarranted fashion most of the evidence we have seen reason to reject.
Bernays first premises that, in order to establish the fact that
Aristotle did treat comedy fully in the Poetics, the announcement at
the beginning of the Poetics to treat all poetry should be taken in conjunction with the reference at the beginning of chapter 6 to speak
about comedy later, together with the unsatisfied references in the
Rhet~ric.~Since these premises do not necessarily lead to such a conclusion, his case is thereby almost lost.
Bernays next gives the text of the Tractatus and makes the following observations upon it. First, there is no unity apparent in the
treatment except that of subjecL4 Next, the beginning of the Tractatus
is unpromising because its division of poetry into imitative and nonimitative flies in the face of the plain declaration of the Poetics where
Empedocles is even refused the title of poet because his verse does not
imitate. In addition to this he discovers several other notable errors,
from the Aristotelian point of view, in the Tractatws. First there is
the employment of 6 i OZKTOU ~ a 6Covs
l
for Aristotle's ZXEOS ~ a 46pos.
l
Second, there is a balancing between fear and sympathy, whereas
the Rhetoric expounds a theory according to which sympathy
ought not to be & K K ~ O L J U T L K ~ VTO^) &XS.ov. Third, it is stated that
V
This assertion is based on the Rhettragedy Fxri p q ~ C p aT ~ Xirxqv.
OP.tit., PP. 435-455.
Bernays, Zwei Abhandlungen, p. 135.
09.tit., pp. 137-139.

09.cit., p. 140.
32
A . Philip McMahon
oric, but in the Poetics Aristotle plainly speaks of a proper 48ovtj in

tragedy.'
When he comes to the definition of comedy contained in the Tractaks,2 Bernays exclaims: " Diese seinsollende Definition der Komodie
ist nichts als eine jammerlich ungeschickte Travestie der aristotelischen
von der Tragodie." The sentimentality of the statement that comedy
has 76v y 6 X o ~ aas pq~kpais glaringly repugnant to all that we know
of A r i ~ t o t l e . ~
There are, indeed, as Bernays points out, certain opinions on
comedy expressed by Aristotle, especially in the Poetics, the Ethics,
and the Politics, where the laughter of comedy is touched
But,
as he also shows, later grammarians mistook Aristotle's views, confined
to Middle Comedy, for the whole of the ~ u b j e c t .Thus
~
the distinction
between Xo~Gopiaand ~ w p y 8 i amight have been derived in the first
place from the Poetics, as the wording is not un-Aristotelian! except the
expression Zp+aa~s,for which Aristotle's equivalent is b ~ b v o ~ a .
Section 5 of the Tractatus, Bernays proves, derives from the ethic^,^
and the balance in section 6 between ykXws and T ~ P + L Sis modelled
after the +bBos and ZXros of tragedy in the poetic^.^
To strengthen his case, Bernays at this point digresses on the probable influence of the Peripatetics on New Comedy, and of the relation
of Theophrastus to Menander, saying that the Poetics agree in the
main with the practice of this poet.n
Returning to the Tractatus, Bernays notes that the six elements
ascribed to comedy are modelled on those attributed to tragedy by
Aristotle.l0 The elaboration of this number of elements for comedy,
once they were obtained, shows a surprisingly ingenious though superficial manipulation of other Aristotelian texts." The pedantry of the
achievement puzzled Cramer and other editors of the work, and the
results move Bernays to term it " eine Verkehrtheit " . . . ('je
O p , cit., pp. 141ff.
Kaibel, Comicorurn Graecwum Fragmenta, i, I, Berlin, 1899,p. 50,3.
Bernays, Zwei Ablzandlungen, p. 145.
0 p . cit., p. 147.
Op. cit., p. 148.
Op. cit., p. 152.
Op. cit.. p. 150.
lo Op. cit., p. 153.
Op. cit., p. 151.
" Op. cit., p. 154.
8 Op. cit., p. 151.
33
deutlicher der mechanische Weg zu Tage liegt."
The expansion of
pBhos is believed by him, however, to point to a more complete work.2
At this point, however, Bernays feels justdied in asserting that

Was uber sie (die Komodie) daher der Excerptor in der jetzigen
Poetik nicht Nachweisbares beibringt, darf fiiglich aus dem vollstandigeren Exemplar hergeleitet werden, wofern innere Griinde nicht
dawider sind." But since, as our case now stands, genuine and independent traces of a theory of comedy must appear in the Tractatzls,
the contrary conclusion to what Bernays states naturally flows from
his previous examination.
The division and discussion of the comic characters is, indeed,
genuinely Aristotelian, but not independent, for it could all be found
elsewhere than in the Poetics, and easily accessible to the ingenuity
of the scholar who composed the Tractatu~.~T hus Bernays is compelled to admit that: b " Vielleicht hatte ein gliicklich spurender
Scharfsinn, ohne weitere Hilfe, aber dann auch wohl ohne allgemeinere
Zustimmung, blos aus diesen Stellen der Ethik und Rhetorik die nach
Aristoteles an sich komischen Characktere auf die. drei zuriickfiihren
konnen, welche der Excerptor nennt. Diesem wird nach dem Ungeschick das er schon zweimal bei Benutzung der Rhetorik gezeigt, Niemand gerade hier eine so glanzende Combinationskraft beimessen
wollen; und wenn in Ethik und Rhetorik jene Dreizahl angedeutet
scheint, so ist das nur ein Beweis mehr, dass Aristoteles sie auch in der
Poetik aufgestellt und der Excerptor sie von dort abgeschrieben hat."
Bernays then outlines Aristotle's probable treatment of the whole
question of comedy, but acknowledges that " der Excerptor hat nur
die Rubrik desselben ausgezogen."
Now when a critic makes capital of his deficits in this fashion he can
prove almost anything. Dzerences in degree of pedantic acumen and
in accuracy are easily accounted for by the assumption that the TracSatus represents not the work of one scholar, but an accumulation of
the work of a number of rhetorical investigators. Of course the excerptor made gross errors in combining his scattered hints, but it is not
necessary to suppose that he was the same who formulated the defini"
Op. cit., p. I 56.
Op.cit., p. I57.
Op.cil., p. 158.
' Op.cil., p. 159.

Op.cit., pp. 163 ff.
35
After thus disposing of the main points on which critics have depended for a theory of the second book, there remain only a few of the
less important references. Among these is the reference of Simplicius
to Aristotle for a treatment of synonyms. Rose, however, shows that
this is derived from Porphyry, and in turn depends upon the reference
in the Rhetoric.'
The allusion of the Antiatticist is very slight evidence in any case,
and it is not a t all certain that this anonymous controversialist referred
to the Poetics in citing Aristotle. Even allowing this unknown writer
the merit of honesty, we are not obliged to conclude that his memory
was sound or his source genuine. The matter to which he alludes may
possibly have been contained in the work m p l Xkt;~ws, afterwards
absorbed into the Rhetoric, or even in chapters of it inserted in the
Poetics that later editors rejected.
Bywater notes various anomalies of thought or language in the
poetic^.^ Among these he mentions: the anticipatory use of technical
terms afterwards defined; variations of terminology; inconsistencies in
the use of terms; inconsistency of thought; and lapses of memory. Yet
he also defends the philosopher on the ground of his natural limitations,
showing how the Greek play limited Aristotle's views by its conventions
with regard to stage presentation, form and structure, motives and subj e c t ~ .H
~ is ideal playwas a compromise between thedramaof the great
period and that of his own generation, seventy years after the death of
Euripides. Among the evidences of this assertion that Bywater brings
forward are: Aristotle's theory of tragic diction, and the silence about
the chorus; his concessions in plot to the more sensitive feelings of his
audience instead of the harsh situations of the older tragedy; and the
fact that his theory of comedy would have been more applicable to the
New Comedy than to Aristophanes. For the state of the text he has
the usual arguments.
Bywater also records his opinion, however, that " the book as it is
with occasional sidelights from other works is intelligible enough."
Now it is much more intelligible if we do not look for something in the
Poetics which there is no reason to suppose ever was there, or to feel
disappointed when we fail to find it.
Rose, Arist. Libr. Ord., p. 133.
Bywater, op. cit., pp. xiv ff.
0 p . cit., pp. viii, ix.
' Bywater, o p . cit., p. viii.
36
A . Philip McMahort
In the course of this investigation I have tried to deal impartially

with all the evidence urged for the existence of a second book, and
with the scholars who held such an opinion based on that evidence.
The only direct statements that there were two books, those of the
indices, have been considered; and those references have been given
which indicate that there was more than one book, -from the Aristotelian text itself, from Ammonius, Boethius, Eustratius, and an
anonymous commentator on the Rhetoric. I n addition there were
the matters supposed to have been in the Poetics, but not now found
there, indicated by the time-references to the Poetics: the promise
about comedy; the cross-references of the Rhetoric to the Poetics
about the ridiculous; the incomplete discussion of katharsis; and the
theory on comedy, reflections of which Bernays thought he found in
the Tractatus Coislinianus. Further matters which might have been
found there, according to scholars, were: a discussion of synonyms;
a treatment of the drama in defence against Plato; and something on
comic diction. This outline was followed by a brief statement of how
the tradition of a lost second book began and was developed.
Against this case it was argued that all the three indices are unreliable; that there are also references elsewhere in which the Poetics
is spoken of as contained in one book; that the inconsistent tales,
especially that of the cave at Skepsis, and other expedients demonstrate the d%culty in accounting for the loss of the second book, which
is easily avoided by a rejection of the supposition that there ever was
one. Then came a detailed examination of the cross-references in
general, where it was shown that they cannot be used as direct proof
of any of their implications; and Aristotle's manner of composition
and publication was discussed. The investigation was concluded with
an analysis of the various references individually, with negative results
as to their validity, including a refutation of the claim of Bernays to
have found traces of the complete theory of comedy. The final result
is a conclusion that there was no second book. While it is logically
impossible to prove a universal negative, there is no reason for us to
believe, in this case, that there ever was a second book; and the facts
of the case are all harmonized and accommodated to one another without such an hypothesis.
37
The dialogues were, as Shute observes, the compositions of Aristotle with which antiquity was best acquainted, and for which, next
to the IIoAi~~iai,
Since there is abunwe have the best authority."
dant evidence from many sources that Aristotle used dialogues in
the first period of his work? Rose attributes this disproportionate
influence of the Politics and the dialogues to the fact that, while the
Politics, by reason of the range of subjects treated, were especially
interesting to grammarians and historians, the dialogues, because they
treated the questions common to philosophy after Plato in an easy,
lucid and popular style, attracted not only the philosophers of the
Roman Empire, but also the later rhetoricians. Among these Rose
mentions Panaetius, Posidonius, Andronicus, Didymus, Varro, Cicero,
Dio Chrysostom, Julian, Themistius, Basil, and Plutarch.4 Indeed,
the researches of Bernays, proving that by the exoteric works reference is had to the dialogues, show that, in view of the unbroken testimony of antiquity, the dialogues were Aristotle's in a sense that can
be applied to none of the other accepted works in the Aristotelian
((
The references we have are mostly historical notes, and arouse no

suspicion against their genuineness. It is true that Fragment 76
contains a statement about Homer, apparently based on a tradition
prevalent in 10s. The questionableness of this statement cannot
prove the dialogue spurious, for it is clear that an author does not
necessarily believe all the statements put into the mouth of the
speakers in a dialogue!
The example followed by the master in dealing with poetry in dialogues seems not to have been an isolated performance, and we find
1 J. Bernays, Die Dialoge des Aristoteles im Verhaltniss zu seien &rigen Werken,
Berlin, 1863; E . Heitz, Die verlwenen Schriften des Aristoteles, Leipzig, 1865;Schlott-
mann, Ars dialogorurn quas vicissitudines apud Graecos et Romanos subierit, Rostock,
1889,pp. 19-25; R. Hirzel, Der Dialog, 2 vols., Leipzig, 1895.
2 Shute, op. cit., p. 7.
a Zeller, op. cit., i, p. 55,n. z.
Rose, Arist. Pseud., p. 23.
Zeller, op, cJt., i, p. 58, n. I.
A . Philip McMahon
Heraclides, among the Peripatetics, writing a dialogue ~ a p Ti O L ~ T L K ~ ~ S
TBV T O L ~ T Ba'.'
V
Indeed, Heraclides appears to have written a
number of works on the subjeck2 He also probably showed the
influence of Plato here; and he is coupled together with Aristotle by
Dio Chrysostom, as being a writer in that form in which Plato was
the firsL3
One of the characteristics of Aristotle's extant dialogues in which
they differed widely from his other works was their style. That Aristotle himself estimated very exactly the literary quality of Plato's
dialogues we see from what appears to be a fragment of his dialogue
On poet^,^ where he says that they are midway between verse and
prose. A distinct and deliberately planned literary excellence was one
of the qualities that antiquity specially noted also in Aristotle's dialogues. We may refer, for instance, to the passage in Ammonius
where the beauty and appropriateness of the language is mentioned:
and to another passage in E l i a ~ . ~
The style, indeed, was, as Themistius observes, the principal attraction of the dialogues for the readers of Aristotle. Cicero, except for
the Rhetoric, does not, on the evidence found in his works, seem to
have read much else of Aristotle but the dialogue^.^ His remarks
concerning their style suggest the same conclusion.8 Cicero's allusions
to Theophrastus establish similar facts for the style of Aristotle's most
famed disciple, and, as Zeller says, " In his case, as in Aristotle's, this
merit belongs chiefly to his popular writings, and especially to the
dialogues, which, like Aristotle's, are described as exoteric." It was
even said, though on insd5cient grounds, by some of the ancients, that
Theophrastus received his name on account of his graceful style.1
~ a i
1 Rose, Arist. Pseud., p. 77; Diogenes Laertius, De vitis, dogmatibus clarorum

philosophorum libri x, 2 vols., Amsterdam, 1692, 5, 86; 5, 88.
2 0. VOSS,
De Heraclidis Pontici Vita et Scriptis. Diss., Rostock, 1896, p. 31.
3 Rose, Fragm., p. 24; Dio Chrys., Or., p. 634, Emp.
* Rose, Fragm., p. 78; Diog. Laert., 3, 37.
6 Rose, Fragm., p. 23; Arnmonius, Proleg. in Arist. caleg. (p. 36 B 28 Br.).
4 Rose, Fragm., p. 23; Elias in Arist. cat., p. 2 6 B 35.
7 Shute, op. cit., p. 64.
8 Rose, Fragm., p. 23; Themist., Or., 26 p. 385,l. 28, Dind; Hirzel, op. cit., i, p.
280; Cicero Acad. #r.119.

Op, cit., ii, p. 348, n. 3.
9 Zeller, op. cit., ii, p. 352, and n. I.
39
Another known feature of the Aristotelian dialogues was their

method. Cicero testifies that the speeches of other persons were so
introduced that Aristotle himself was always the principal speaker.'
This characteristic is otherwise described by Basil as a practice of
direct statement in distinction from the contrast of opinions employed
by plat^.^ From Cicero we also learn that in dialogues of more than
one book, each part had its own prefa~e.~On this question Proclus
gives similar informati~n.~From such evidence has also been derived
the accepted belief that the dialogues were essentially distinct in form
and method from the extant treatise^.^
In support of Bernays's opinion that exoteric and published works
are identical and refer almost all to the dialogues, may be adduced
the practice of referring to the other works in ways which indicate
that only the dialogues were considered finished literary productions
and so worthy of publication.' Thus the reference in the Poetics to one
of the published works is most naturally assigned to the dialogue On
Poets, as Zeller remarks: rather than to the Rhetoric, as Rose suggest^,^
since there is no corresponding passage there. Vahlen agrees with
Zeller on this point,1 and while exoteric may include some of the more
popular of the really esoteric class, such as the Rhetoric and the
Politics, we can reasonably infer, because of the absence in these of
passages corresponding to references which require an exoteric or published work, that the dialogue 0%Poets is meant." Thus, while the
works still entire in the accepted Aristotelian canon afford little d e h i t e
proof of the genuineness of the dialogues, the unanimous agreement of
antiquity shows that the editors who inserted such references to
exoteric or published works had authentic Aristotelian dialogues which
they could have cited, and from the character of the fragments preRose, Fragm., p. 23; Cicero, E p . ad Att., 13, 19.
* Rose, Fragm., p. 23; Basil, Epist., 135.
Hirzel, op. cit., i, p. 275.

Shute, History, p. 64. Cf. Cicero, Ad Att., 4, 16, 2; 13, 19, 4.
Rose, Fragm., p. 23; Proclus, In Parmen., t. iv, p. 54 Cous.
ZeUer, 09,cit., i, p. 58, n. I.
Rose, De Arist. Libr. Ord., p. 79.
lo Vahlen, op. cit., p. 36, n. on Poetics, I;. 1454 B 18.
l1 Shute, op. cit., p. 21.
served, we believe these fragments to be Aristotelian and to show that

the dialogues contained passages corresponding to such references.
Furthermore, it is an equally reasonable hypothesis that references
to the Poetics which are not otherwise definitely accounted for, may,
on account of the confusion of On Poets and Poetics as titles of both
the dialogue and the treatise, correspond to passages in the dialogue.
Since antiquity great confusion has been caused by the similarity in
names between the Poetics and On Poets, one of them a formal treatise
and the other a dialogue. Allusions to these works are inextricably
contradictory, for with two works, each called by either of two titles,
and with references to both of the titles and both works as in from
one to three books, there are twelve possible combinations.
Beside the easy confusion occasioned by the close similarity of
names, the Poetics and On Poets, and a resemblance in subject-matter,
we must also reckon with the possibility that the dialogue, as the only
formally published work, received a title earlier, and that general agreement on the title of the treatise was not reached in antiquity.
That the work On Poets was a dialogue cannot, as we have shown,
be fully proved, but if the treatise is by agreement entitled the Poetics,
then we must call the dialogue On Poets. The real question is then:
how did the dialogue differ in method and treatment from the treatise,
and what other facts are known about it ?
The orthodox view, with its inherent d%culties more or less glossed
over, is set forth by Rose,' and may be summarized in the following
manner. That the three books On Poets, which are mentioned near
the beginning of the index of Diogenes Laertius, were really written in
the manner of a dialogue is expressly stated in the Vita Marciana of
Aristotle.3 In spite of the confusion of titles, the dialogue is distinguished by having one book more than the t r e a t i ~ e . ~
In the dialogue
were also discussed the art of poetry; its nature as a kind of imitation,
its function in the state, and so forth, - all this incapable of demonstration by Rose's method of treating the evidence. If, however, one
Op. cil., p. 25.
Rose, Arist. Pseud., pp. 77-86.
Rose, Fragm., p. 7 6 ; Vita Aristolelis Marciana (cod. 257) f. 276 A; Rose, Arist.
Pseud., p. 7 7 .
* O p cit., pp. 7 7 , 7 8 ; Zeller, op. cit., i, p. 58, n. I .
Rose, Arist. Pseud., p. 7 8 .
4I
of the works was in a t least one book and the other in three, and if
there is no reason to suppose that the treatise, called the Poetics, was
in more than one book, then the dialogue On Poets was in three.
The character of the dialogue is ascertainable with tolerable certainty. Hirzel shows that the dialogue discussed the relation of philosophy to poetry, and contends that, while critics usually consider it
a purely historical work, excluding the theoretical discussion of poetry,
they do not make it clear how the history of poetry could suitably be
made the matter of a dialogue, even of an Aristotelian one.' The
fragments themselves lead to an opposite conclusion. The examples
were probably cited to sustain points of the argument as in the
Socratic dialogues, and the fragment which maintains that when the
philosopher rhymes or the poet philosophizes, either the poetry or the
philo'sophy is inferior, indicates a philosophical discussion of a quite
Platonic ~ h a r a c t e r . ~Indeed, it is a purely philosophical distinction
which opposes historical to poetic truth as the diflerence between particular and general truth, so that a t the end of the Aristotelian dialogue Socrates, who first investigated general conceptions with worthy
results, and Homer? who above all others deserved the name of poet,
could be brought together. This conjunction, as we learn from the
fragments, probably occurred in the third and last book; and the
critical theory involved agrees with that in the Poetics and the Metaphysics.6
I t has been supposed that the dialogue also treated the art of poetry,
a supposition which would explain how the dialogue could sometimes
be entitled the P o e t ~ s . Bywater
~
thinks that, a t any rate, the dialogue did not devote special attention to the question of stage-effects?
The relation of the dialogue On Poets to the treatise the Poetics has
been several times discussed: and it has been thought possible that
many of the references to matters poetic which are not found in the
'
'
8
Hirzel, op. cit., i, p. 288 and n.

Op. cit., i, p. 288.
Bywater, op. cit., p. 19.
Note on Poetics I, 1447 B 18.
Gercke, Pauly-Wissowa, ii, I,col. 1052. 63.
Hirzel, 09. cit., i, p. 289.
Zeller, op. cit., i, p. 58, n. I.
Bywater, op. cit., p. 233.
Diintzer, op. cit., p. 278.
A . Philip McMahon
Poetics might have been in the work On Poets. This hypothesis is
surely as valid and reasonable on its face as the theory that such
matters were to be found in a second book of the Poetics. Two
of the parallels in Athenaeus to the Poetics noted by Vahlen might
conceivably have been drawn rather from the dialogue, as well as
another parallel, also noted by Vahlen, in the Anonymus de comoedi~.~
Moreover, the passage in Themistius, possibly, but by no means certainly, an expansion of information derived from the Poetics, might,
in view of his acquaintance with the dialogue, have more probably
been derived from the latter. Von Christ judges, finally, that the
dialogue was a preliminary discussion, followed later by a more profound and technical work?
The matter of the dialogue is broadly indicated in the extant fragments. Much of the surviving material is not assigned to any particular book, but there are a number of cases in which the book is indicated. Thus, in the first book there was a discussion of the dialogue
form and a reference to Plate: a fact which is further established by
the evidence of Athenaeus6
The second book, according to Macrobius, introduced the evidence
of Euripides on a question of Aetolian custom^.^
The third book, according to Diogenes Laertius, gave an anecdote
about Socrates? In this book, also, there was the discredited story
about Homer's origin? and ,from these indications the matter of the
other fragments can to some extent be grouped in the different books,
so that we may now agree with Hirzel that the place of the passage
mentioning both Socrates and Homer was a t the conclusion of the
whole work.9
Athenaeus, 8,367 B ; ii, p. 302, ed. Kaibel; cf. Vahlen, op. cit., p. 53,n : and
Athenaeus, g, 433 C ; ii, p. 442,ed. Kaibel; cf. Vahlen, op. cit., p. 51,n.
* Vahlen, op. cit., p. 13,n. on Poetics, 4. 1449B 7.
V o n Christ, Geschichte, i, p. 674.
Rose, Fragm., p. 77; Diog. Laert., 3,48.
Rose, Fragm., p. 78; Athen., xi, p. 505 C.
6 Rose, Fragm., p. 78; Macrob., Saturn., 5 , 18,19.
Rose, Fragm., p. 79; Diog. Laert. 2, 46.
Rose, Fragm., p. 79; fragment 76.
Hirzel, op. cit., p. 289.
'
43
IV. TEIEOPHRASTUS
Various sources reveal Theophrastus as the chief disciple of
Aristotle. His interest was mainly scientific, but even in science
he strove to complete and substantiate the principles of his master,
and introduced no radical diflerence~.~
As Boethius bears witness, he
advanced further than Aristotle in fields which the master had but
slightly touched; otherwise he accepted his teachings." In the same
place Boethius tells us that Theophrastus sometimes used the very
words of Aristotle without ~ h a n g e . ~Cicero, indeed, points out that
Theophrastus was more accurate in his observations, and especially
developed research in natural sciences.6 Cicero also thought Theophrastus a closer follower of the master than other Peripatetics, while
Galen seems almost never to h d any difference between them.-'
Zeller points out that Theophrastus investigated the psychological
effect of music and held that certain diseases could be healed by it?
The few fragments we possess of this discussion lead us to believe that
neither did he difler widely from Aristotle in his theory of art.
There is in Athenaeus an extract from Theophrastus " On Comedy." a Zeller, however, holds that his citations from it are " quite
incredible." Whether we have here a mistake of Athenaeus or not,
does not seriously affect the authenticity of the aesthetic theory of
Theophrastus as a practical reproduction of Aristotle's. Athenaeus
may have been familiar with the Poetics, as he evidently was with
On Poets,lo yet both the reference of 13, 608 E and that of 6, 261 D
may have been derived from some intermediary historical discussion.
The canon of the works of Theophrastus is extremely uncertain
since almost all of them, except some in natural science, are lost but
for a few fragments. The list of Diogenes gives him a work on comedy,
Zeller, op. cit., ii, p. 348.
Diels, Ueber das drittc Bwh, p. 26.
0 p . cit., ii, p. 355.
Zeller, op. cit., ii, p. 356.
Cicero, Fin., 5 , 4 , 10.
Zeller, op. cit., ii, pp. 4 15, 416.
G. Kaibel, ed. Athenaei Naucratitae Dipnosophistarum, libri X V , vols. i, ii,
Leipzig, 1887, 6 , 261 D ( p . 8 1 ) , and 8 , 348 A ( i , p. 263).
$ Zeller, op. cit., ii, p. 414, and n. 4 .
'"den,
op. cit., p. 6 , n. on Poetics, I . 1447 B 21.
'
'
44
A . Philip McMahon
and one on the ridiculous, but two separate works on p0etics.I This
last statement may have been a mere repetition on the part of a scribe
or librarian, but it is especially noted in the list that the second is
another work. Andronicus and Hermippus both drew up lists. One,
probably by Hermippus, is preserved by Diogenes, but it follows a
curious order, having first two alphabetical lists, of which the second
probably supplements the first. These perhaps show the contents, at
different times, of some great library such as the Alexandrine. They
are in turn foLlowed by a list without order, and a fourth division in
the main alphabetical. The genuineness of most of the works is beyond
our means of knowledge, but Usener thinks that some were rather the
writings of E u d e m u ~ . ~
Although Cicero says that Theophrastus passes over slightingly what
Aristotle had treated already! yet he did treat the same topics as
Aristotle; and probably reviewed the whole of the Aristotelian philosophy as head of the Peripatetic school. It is, therefore, almost certain that he would exactly reproduce the master's doctrine in a subject
in which he was apparently less original than he was in natural sciences.
Rose shows that the evidence of Cicero and Proclus proves that the
dialogues of Theophrastus were written in the same manner as those
of A r i s t ~ t l e . Plutarch,
~
whose Coasolatio, it will be remembered, contains the passage without hiatus, perhaps taken word for word from a
dialogue of Aristotle, quotes Theophrastus on Fate, and while this
quotation is short, it is in precisely the same style as the Aristotelian
quotation! Thus it is probable that Theophrastus imitated his master
as far as writing a dialogue on poetics or poets, although we are unable
to tell if one of the works mentioned in the list of his writings corresponds to such a work. Fragments of Theophrastus on the question
of comedy and tragedy, therefore, not found in our Poetics, might well
have been in Aristotle's dialogue On Poets.
Diogenes Laertius, De vitis, 5 , 36, p. 294, ed. 1692.
Zeller, op. cit., ii, p. 352, n. 4.
8 Cicero, Fin., I, 2, 6.
Cicero, Diu., 2, I, 3.
"ose, Fragm., p. 23.
6 F. Wimmer, Theophrasti Eresii opera quae supersunt omnia, vol. iii, Leipzig,
1862,p. 181 (fragment 73).
1
2
45
There is, indeed, a definition of tragedy ascribed by Diomedes to

Theophrastus.' Zeller objects to it on the ground that it is not elaborate enough; an objection which Reich meets by characterizing it as a
popularization of Aristotle's t e a ~ h i n g . This
~
is possible, but the popularization does not need to have been original with Theophrastus. I t
was much more probably the definition in the dialogue On Poets that
Theophrastus reproduced. It does not conflict with the conception of
the nature of tragedy which underlies that of the dramatic species in
the Poetics, and, as Margoliouth points out, Aristotle's own equivalent
for the word cmov8a'ios found in Poetics 6 is the word heroic of Theophrastus' definitiom4 Von Christ, indeed, rated this latter definition
higher than that of the Poetics, and declared it more correct on the
ground that, in the definition of Poetics 6, the religious factor was
omitted, and the discussion limited to an ethical and political treatment.
There are three other dehitions of dramatic species including
comedy, parallel to that of tragedy in Diomedes: and critics have
ascribed them all to Theophrastu~.~If this supposition is correct, it
is probable, as Reich asserts, that all four were derived from Aristotle.7
How they came to Diomedes is uncertain, but Koett conjectures that
they survived in some compendium.*
Although no certain proof can be offered, it is an interesting possibility that the source of Theophrastus was the dialogue O n Poets, and
it may be possible even more definitely to locate the definition of
tragedy there. It has been shown by the work of Finsler that there
is a close relation between the dialogue and Plato, and there is abundant evidence of his contention. Plato asserted that Empedocles
was not, properly speaking, a poet, and with this Aristotle agrees
both in the fragment of his dialogue O n Poets and in the Poetics.
1 Keil, Grammatici Latini, i, p. 487 of Diomedis de oratione, Liber iii; von Christ,
Geschichbe, i, pp. 248, 249.
2 Zeller, op. cit., ii, p. 414, n. 4.
Reich, op. cit., i, p. 267.
4 Margoliouth, Poetics, p. 44.
6 M. Schanz, Geschichte der ramischen Litteratur, Munich, 1904, IV. I , p. 153.
6 Reich, op. cit., i, p. 266.
7 O p . cit., i, p. 270.
E. Koett, op. cit., p. 47.
46
A . Philip McMahon
Empedocles is furthermore contrasted with Homer by Plato, and by

Aristotle in the Poetics and On Poets. Aristotle in the Poetics connects Homer closely with the subject of tragedy, and most of his illustrations are chosen from the epic. Plato, indeed, made Homer the
greatest of poets in tragedy. Now it appears probable that Aristotle
brought Homer and Socrates together in the third book, perhaps in
showing, against Plato, that the two highest species of poetry, tragedy
and epic, possessed in a high degree the generalizing or genuinely
philosophic merit of poetry. Thence it becomes very probable that
the fundamental idea of the nature of tragedy set forth in the Poetics
and concisely expressed in the Theophrastian definition, which yet
agrees with Plato and all ancient critical conceptions, would have
entered into the discussion a t this point, and may have been expressed
in the very words of Theophrastus which are reproduced by Diomedes.
The Theophrastian definition would then have been found in the third
book of the dialogue On Poets.
Harvard Studies in Classical Philology 40 (1929)
SEVEN QUESTIONS ON ARISTOTELIAN DEFINITIONS
OF TRAGEDY AND COMEDY
I. What definitions did the Greeks find most satisfactory to express their ideas of tragedy and comedy?
11. What definitions did the Romans find sufficient to express
their ideas of comedy and tragedy?
111. What definitions did the Middle Ages, in general, accept as
expressing proper ideas of tragedy and comedy?
IV. What definitions did Dante, in particular, accept as conveying proper ideas of comedy and tragedy?
V. What definitions did Chaucer, in particular, accept as conveying proper ideas of tragedy and comedy?
VI. What definitions of tragedy and comedy dominated the ideas
of Continental Europe during and after the Renaisance?
VII. What definitions of tragedy and comedy dominated the ideas
of England during and after the Elizabethan Age?
HE very words tragedy and comedy indicate an origin in the classical drama. Examples of drama, as well as other literary forms,
related to that precedent have also been called tragedy or comedy, so
that as dramatic and literary species with innumerable historical instances, the words have a definite meaning. I n addition, ever since
Aristotle, there has been an inquiry into the norm or type abstracted
from such instances, in an effort to achieve through definitions a
standard for creation and criticism; a struggle to reach an articulate
conception, with which dramatic and litzrary productions could be
compared. Finally, there has been an investigation into the problem of
what makes tragedy tragic and comedy comic, a search for the values
or essences which constitute the tragic and the comic whether in
drama, literature, and the other arts, or in the course of experience itself as directly observed. Thus it has been common practice to refer
metaphorically to disastrous events as tragedies and to ridiculous ones
as comedies.
98
A . Philip McMahon
A good deal of confusion and trouble has been due to condemning

the ideas of the Middle Ages with regard to tragedy and comedy, because the formulas of that period were not derived from a first-hand
study of the classical Greek drama. As a matter of fact, the Middle
Ages relied on broad definitions which were authentically and legitimately Aristotelian, although they knew neither Greek drama nor
Aristotle's Poetics. But, on the other hand, especially since the beginning of the nineteenth century, ideas of what is tragic and comic alien
to Aristotle's own philosophy have been used both to determine the
dramatic species and to interpret the text of the Poetics.
Bearing in mind that the meanings of tragedy or comedy are related and often interchangeable ideas, we may attempt to trace the
persistence of Aristotelian definitions of tragedy and comedy. T o
cover thoroughly the fields suggested here would require volumes;
especially in the treatment of the later questions, what is presented
must be thought of as indicating a continuity rather than a complete
survey. I t need hardly be said that I do not insist on the chronological,
linguistic, and national classifications here adopted? If any reader
wishes to employ the material in some difierent scheme, let him suit
his convenience as I have mine.2
W h a t dejitzitions did the Creeks find m,ost satisfactory to express their

ideas of tragedy and comedy?
An attempt will here be made to indicate generally the most probable
answers to the seven questions regarding Aristotelian definitions of
tragedy and comedy which are asked a t the beginning of this article.
I t is obviously impossible to list all the occurrences of the words or passages in which the idea is referred to, and inquire into their meaning
1 I n what follows I have not attempted to reform the spelling and punctuation
of the sources, but have left themas I found them. This note must be considered
a substitute for the insertion of the word "sic" several hundred times in the quotations.
2 The conclusions of this article are substantially those of my Harvard dissertation in 1916, but it incorporates new material and is condensed in form. For suggestions and advice my indebtedness to Professors E. K. Rand and C. N. Jackson
continues and increases.
Aristotelian Definitions of Tragedy and Comedy
99
based on context and historical position. But enough important citations can be produced to suggest that the ideas and definitions of
tragedy and comedy, ultimately derived from Aristotle's dialogue
On Poets, have ever since dominated European thought and language.
They have indeed become commonplaces. Accordingly, the broad
answer to the seven questions asked a t the beginning of this article is:
Definitiom ultimately derivedfrom Aristotle's dialogue On Poets.' These
questions will be treated separately in the order given and the relation
of the answers to Aristotle indicated.
We might, I believe, go even further and ask two additional questions in the light of the evidence here put forward in satisfying those
seven questions. We might ask, first: Are the prevailing modern definitions of tragedy and comedy Aristotelian? And second: I s the modern
interpretation of Aristotle's definition of tragedy in the Poetics Aristotelian? The weight of the testimony would compel us to reply to
both questions in the negative. But that is a separate matter and must
await a later opportunity for discussion.
In an article published in Volume XXVIII of the present Studies
(1917)'I undertook to examine the conclusion, accepted by scholarly
opinion since the Renaissance, that Aristotle's treatise known as the
Poetics originally consisted of two books, of which one containing a
discussion of comedy is now lost.2 I t seems to me that sufficient proof
The effort to discover traces of the Poetics in subsequent references of classicaI
literature to tragedy and comedy has been seriously pursued in but a few instances.
Numerous modern scholars have shown the general Peripatetic origin of both the
Poetics and the standard classical definitions of the dramatic species, whose historical origin I have attributed in my previous article in the IZarvard Studies (vol.
X X V I I I ) , to the dialogue On Poets. But to show the direct influence of the Poetics
generally throughout antiquity requires a theory constructed in advance to include
the statements whose relation should be the result rather than the postulate of the
discussion. Cf. A. Rostagni, "Aristotele e I'Aristotelismo nella Storia dell' Estetica
Antica," Studi Ilaliani d i FiEoEogia CEassica, N . s., 11 (1922), pp. 1-147, passim, but
particularly pp. 1 0 4 - 1 0 5 The "patrimonio quasi anonimo dell' universale cultura"
which, he claims, derived from the Poetics, cannot be proved on the basis of evidence
independent of such a n a priori postulate, but its paternity, as indicated in my previous article, may be found in Aristotle's dialogue, and its progeny is but briefly
indicated in the present contribution.
The thought that the present text of the Poetics is only the first part of the
original is k s t strongly emphasized through the title of Vettori's edition: Commen-
100
A . Philip McMahon
or reasonable grounds for such a theory can not be found; but for the
details of that investigation I refer the reader to the article itself.
Since that time no writer has demonstrated that there ever was a
second book of the Poetics.
The result of my earlier investigation, which I must re-affirm because
of misrepresentation,' was a negative one with regard to the previous
existence of a second book of the Poetics, containing a theory of comedy, and a positive one with regard to the dialogue Orz Poets as the
most probable eventual source of subsequent definitions of tragedy and
comedy.
The explanations of the grammarians Euanthius, Diomedes, and
Donatus, give clear statements of what tragedy and comedy are.
Their source is stated to be Theophrastus, and in all probability they
were ultimately derived by Aristotle's successor from the dialogue On
Poets. They are objective, almost empirical summaries of the charactarii in Printurn Librum Aristotelis de Arte Poetarum. Posiiis ante Singulas Declarationes Graecis Vocibz~sAz~ctoris:Iisdem ad V e r b t ~ mLatine Expressis. Florence, 1560.
(His translation, however, appeared before this, perhaps in 1552.) Vettori himself
supposed that the Poetics had consisted of three books originally, on the basis of
Plutarch, Diogenes Laertius, and Aristotle's promise about ~ a 3 a p n r r . I t is hard to
understand why critics should not claim three books instead of two, if they are bent
on enlarging the sphere of speculation as to the contents of the Poetics. Cf. Petrus
Victoriz~sLcctori (in edition of 1573, signature b, lolio i. recto). "Sed v t a d opus
Aristoteleum, artemque hanc redeam, iacuit diu ipsa; & paene vndique; tenebris
oppressa atque extincta fuit. quippe cum e tribus libris, primus tantum in vita relictus sit: reliqui autem duo qui consequebantur, interierint: nec vestigium eorum
vllum resliterit." [Cites I'lutarch,life of Homer; Laertius, life ol Socrates; Aristotle,
promise about katharsis.] "quarum rerum vix umbra quaedam nunc reperitur in hoe
libro, quem in manibus habemus. praeterquam quod nullo modo verisimile est ipsum, accuraLum in primis doctorem, multss partes huius operis neglecturum luisse.
& eas quidem non paruas: imperlectumque demum id, atque inchoatum reliquisse."
h Rostagni, op. ril., p. I , n. I , says, "AfTatto insignificante i. il recente studio
di A. I'h. Mc?vlahon . . . il quale, sostenendo la tesi di un unico libro, attribuisce
a1 dialog0 n e p i T O L ~ T G Vt utto cib che si cita come appartenente al perduto libro."
Four years later, however, Kostagni published his article, "Dialogo ~\ristotelico
m p i T O L ~ T G U "(Kiivsta d i l~ilologiaClus.sica, LIV, N.s., IV (1926j, pp. 433-470, and
N. s., v (1927), pp. 145 ff.) Since my article did not make the comprehensive claim
which he asserts it did, as he might easily have seen by reading it, and since he himself years later decided to write about the dialogue, i t is difficult to see why that
article was affatlo insignijcante.
IOI
teristics of tragedizs and comedies known to the ancients, and they do

not offerevidence of any knowledge of or immediate dependence on the
Poetics. The definitions and distinctions of the latter treatise differ
from those of the grammarians, chiefly in two respects: first, in the
technical points introduced into what is primarily a philosophic and
scientific analysis intended mainly for playwrights; second, in the
famous katharsis clause. On the other hand, the grammarians' distinctions do not differ fundamentally from the basic conceptions of the
Poetics, since they are also products of the Peripatetic position, unless
the katharsis clause is interpreted from a nineteenth-century point of
view.' As the grammarians' presentation is more inclusive historically,
it can thus cover developments of comedy which were not familiar to
Aristotle, and it is suited to the purpose of a public wider than students
of philosophy and p l a y ~ r i g h t s . ~
Although these passages will be cited again in their chronological
sequence, it is advisable to quote a t this point the words of Euanthius,
Diomedes, and Donatus. These quotations should be very carefully
considered a t this point, for the content of their statements. The words
ascribed to Theophrastus and the expressions which parallel them in
form are not only authentic, but, as I indicated in my previous article,
probably derived by Theophrastus from Aristotle's dialogue On Poets.
The grammarians had no intention of aiming a t originality, and they
are not the sources of the formulas which they had from Theophrastus
by means of intervening scholars. The grammarians' observations embody formulas which, as the following pages will show, were almost
Wilamowitz, while admitting Bernays's interpretation, declared i t useless for
a critical understanding of tragedy. Cf. Ezlripides Neral~les,I (Berlin, 1889), p. 109:
"dieses Kleinod der aristotelischen lehre kiinnen wir nicht brauchen, mag es auch
das unschatzbarste sein. man kann doch darhber keine worte verlieren, dass eine
kathartische wirkung weder Aischylos erstreht noch die Athener envartet haben.
mag der Philosoph auch noch so scharf und fein die wirkung beobachtet haben,
welche eine tragudie auf das publikum oder auch auf ihn bei einsamem lesen
ausiibte: diese wirkung war den dichtern und ihrem volke unbewusst."
* Cf. A. E. Taylor, Aristotle (rev. ed., London and Edinburgh, 1919), pp. 1231 2 4 : "NO book has had a more curious fate than the little manual for intending
composers of tragedies which is all that remains of Aristotle's lectures on poetry.
. . But it may be worth while to remark that the worth of Aristotle's account of
tragedy as art-criticism has probably been overrated."
A. Philip McMahon
universally accepted until the last century. They are important a t this
stage because they are the clearest and most extensive expression of
those formulas. Viewed in the perspective provided b y other Greek and
Roman writers, what they report takes a logical place a s testimony to a
continuous tradition extending from Aristotle down t o recent times.
Diomedes states:
Tragocdia cst hcroicae fortunac in adversis conprehensio. a Theophrasto
v
~ 6 x g 7s r ~ p i a ~ a a c.. .s . Comocdia cst
ita dcfinitn cst, ~ p a y q 6 i ai a ~ i 4pwi'~Gs
privatae civilisquc fortunae sine periculo vitae conprehcnsio, apud Graecos
ita definita, Kwpq6ia i a ~ i vi 6 i w ~ i ~ i javp a y p h ~ w vh~iv6vvosacpiox$ . . .
in ea viculorum, id cst humilium domuum, fortunae conprehendantur, non
ut in tragoedia publicarum rcgiarumque. . . . comocdia a tragoedia differt,
quod in tragoedia introducuntur heroes duces regcs, ih comoedia humiles
atque privatae personae; in illa luctus exilia caedes, in hac amores, virginum
raptus: deindc quod in illa frequcnter et paene semper lactis rebus exitus
tristes et liberorum fortunarumque in peius adgnitio. quare varia definitione
,
r i x v s a c p i a ~ a a i sdicta
discretae sunt. altera enim h~iv6vvosa c p ~ o x ialtera
est. (Cl. below, p. 129, note 3, and page 130, note I.)
On the differences between the two dramatic species, Euanthius tells
us, that:
inter tragoediam autem et comoediam cum multa tum imprimis hoc distat,
quod in comoedia mediocres iortunae hominum, parui impetus periculorum
laetique sunt exitus actionum, a t in tragoedia omnia contra, ingentes personae, magni timores, exitus funesti habentur; et illic prima turbulenta, tranquilla ultima, in tragoedia contrario ordine res aguntur; turn quod in tragoedia fugienda uita, in comoedia capessenda exprimitur; postremo quod
omnis comoedin de fictis est argumentis, tragoedia saepe de historia fide
petitur. (Ci. below, p. 129, note 6.)
The definitions of Donatus are strikingly similar and respecting
comedy, he informs us:
Comoedia est fabula diuersa instjtuta continens affectuum ciuilium ac priuatorum, quibus discitur quid sit in uita utile, quid contra euitandum. hanc
Graeci sic definiuerunt: Kwpq6ia i a ~ i v( ~ ~ L W T LaKp a~y~p dVl )~ w v~ c p i o ~ l j
h ~ i v 6 v v o s . comoediam esse Cicero ait imitationem uitae, speculum consuetudinis, imaginem ueritatis. comoediae autem a more antiquo dictae . . .
had ~ i j ~s h p v shoc
, est ab actu uitae hominum quia in uicis habitant ob mediocritatem fortunarum, non in aulis regiis, ut sunt personae tragicae. comoedia autem, quia poema sub imitatione uitae atque morum similitudine
103
compositum est, in gestu et pronuntiatione consistit. . . . aitque esse

comoediam cotidianae uitae speculum, nec inuria; nam ut intenti speculo
ueritatis liniamenta facile pet imaginem colligimus; ita lectione comoediae
imitationem uitae consuetudi~lisquenon agerrime animaduertimus. (Cf. below, p. I 26, note 5.)
These are, to be sure, late and derivative statements, but they
testify definitely to the uniform conception from which there are no
marked divergences in antiquity.
Spingarn refers to this conception as belonging to l L the post-classic
period" and also condemns it as "un-Aristotelian."
But it goes
back to the classical period itself. I t may appear to be un-Aristotelian
to those whose minds are saturated with the transcendental German
interpretation of the Poetics. But antiquity seems to have been largely
indifferent to the Poetics and certainly it did not participate in the
point of view which lies behind the modern interpretation of the K & ~ U ~ U L S
clause. The conception reported by the grammarians can be considered un-Aristotelian if, in contrast to the Poetics, i t is not primarily
philosophical and technical, does not reproduce the very language of
the Poetics, and is easily understood. But such grounds are insufficient
to brand their conception as un-Aristotelian.
The clauses of the standard definitions presented in the statements
of Diomedes, Euanthius, and Donatus are paralleled or implied in the
Poetics, which may be thought of as the later and professional analysis
of the drama known to Aristotle, but founded on the same point of view
and accepting the same popular grounds of distinction. It was, however, the clarity and simplicity of the definitions from the dialogue Oa
Poets rather than the obscure and complex reasoning of the treatise
which we call the Poetics, that enjoyed great diilusion and popularity.
As Spingam observed, tragedy and comedy were distinguished by
the three grammarians quoted above, and by antiquity in general on
one or all of six grounds: (I) characters represented; (2) type of action
imitated; (3) contrast between the beginning and ending of the plot;
(4) style and diction; ( 5 ) source of themes; (6) favorite topics?
J. E. Spingarn, History of Literary Criticism in the Renaissance (3d ed., New
York, I~IZ), p. 66.
Ibid., p. 67.
a Spingarn, op. cit., pp. 66-67.
A . Philip McMahon
Of these, the first three are the principal ones and the rest are their
necessary consequences. I n addition to the grounds mentioned by
Spingarn, there are two others which are mentioned by the authors to
he quoted later: (5) the emotional effect; and (8) the moral purpose.
Of these two, the former is mentioned only occasionally, but the latter
is frequent from the start. The phrases of Theophrastus cover the first
three or principal distinctions. The tears or laughter as showing the
emotional results are not often stressed, hut the moral aspect is so repeatedly emphasized that this more popular parallel to the technical
phraseology of the Poetics offered in the grammarians' formulas, derived from On Poets, would tend to show that the katharsis clause was
also moral in purpose, and not restricted to emotional experience as the
principal object of drama.
I n the passages from the three grammarians cited above, apart
from the direct quotation of Theophrastus, elements (I), (z), ( 3 ) , (s),
(6), and (S), are to be found. But (I), (2), and (3), the essentials, are
contained in the words of Theophrastus; the additional tlistinctions are
either natural implications of the three fundamental points, or are due
to the contrasts brought out between the two species.
The elements noted by Spingarn and other scholars as making up
definitions or conceptions of tragedy and comedy in antiquity and the
Middle Ages recur so constantly that we can establish a common consent regarding the topic down to the last century. The explanation for
this almost monotonous similarity is a common eventual origin in the
dialogue ON Poets, clearly not in the Poetics. So varied are the contexts
and influences afiecting the writers to be quoted that their agreement
is most significant. To show this substantial agreement from antiquity
down to the nineteenth century is the purpose of examining the words
of so many different authors and reflecting on what they have written.
I n Plato there were numerous references to tragedy and comedy

which illustrate his understanding of their nature. For example, in the
Philebus,l speaking of mental states in which pain and pleasure are
mixed, he contrasts tragedy and comedy as the two chief species of
Philebus, 48 A.
10j
drama. I n the Republic he discusses the harm done to the good by

sympathizing with the sorrowful in tragedy and tolerating the ridiculous and unseemly in comedy. Even the best of us delight in those
passages of Homer or in those tragedies that stir our feelings most. I n
the Laws2older children are to be advocates of comedy, but educated
people in general will favor tragedy. I n another passage of the Laws
principles for the regulation and censorship of tragedy and comedy are
given. These passages, as far as they indicate Plato's conception of
the characteristics of tragedy and comedy, do not conflict with Aristotle and the other writers of Greek antiquity, although they differ
with regard to value and ~hilosophicaliustification.
Without discussing Plato's other references to the subject, it may
be said that, as was the case with the use of the term pipqa~s,so the
chief characteristics of tragedy and comedy as formulated by Aristotle
and his successors were based on current, popular phrases, and it was
Aristotle7sconcise summary of those characteristics in his dialogue On
Poets which, as we shall see, determined Greek theory, rather than his
technical discussion in the Poetics.
One of the personalities whose influence was most important in disseminating the formulas of On Poets was Aristophanes of Byzantium,
the successor of Eratosthenes as head of the Library in Alexandria.
According to Sandys? "in scholarship he holds, with Aristarchus, one
of the foremost places in the ancient world." A direct line leads from
Aristotle, through Theophrastus and the Alexandrines, including Eratosthenes and Aristophanes, to the Romans, including Cicero, Varro,
and Suetonius, and thus to the Middle Ages.5 The Arguments preceding the plays of the three tragic poets and of Aristophanes represent
an abridged form of his introductions, which were founded on the reRepublic, X, 605 B-608 A.
Laws, 11, 658 B.
Lazvs, VII, 816 B-817.
J. E. Sandys, History of Classical Scholarship, I (Cambridge, 1906), p. 126.

U. von Wilamowitz-Moellendorff, Euripides Herakles, I (Berlin, 1889),
P. 55.
A . Philip McMahon
106
search of Aristotle and other Peripatetics.' These and the scholia

agree in accepting the traditional definitions, since many of the critical
judgments expressed clearly rest on them, and it is worth noting again
that in such an obvious opportunity for the citation of the Poetics, no
indisputable use is made of that s ~ u r c e . ~
For the conception of comedy entertained by Aristophanes of Byzantium, we have the celebrated phrase on Menander ascribed to him
by Syrianus: "0 Menander and Life, which indeed of you is imitation?" The poet's re-creation of life is deemed so excellent that i t
receives the highest praise known to ancient critics, the imitation and
the thing imitated can hardly be distinguished."
Various remarks in the Argument to the Orestes of Euripides show
a disapproval of the ending, based on the traditional definition. Such
criticism is found in the original Argument by Aristophanes of Byzantium himself as well as in the versions composed by later f o l l o ~ e r s . ~
The observations on Orestes 1691 expand the point, with some interestParticularly the didascaliae of Aristotle. Cf. Wilamowitz, op. cit., pp. 144 ff.;
also A. Nnuck, Aristophanis Byzantii Granttnatici Fragme?tta (Halle, 1 8 4 8 ) .
W. Dindorf, Scholia Graeca in Ezcripidis Tragoedias ex Codicibzis Aucta et
Eme?tdala ( 4 vols., Oxford, 1 8 6 3 ) ; E. Schwartz, Scholia i n Euripidem (2 vols.,
Berlin, 1 8 8 7 ) . I n the case of Aristophanes of Byzantium, as in that of others to follow, the statements of the commentators are considered together with the original
work.
Ch. U'alz, Rhetores Graeci (Stuttgart, 1 g 3 6 ) , IV, 1 0 1 : ' 0 RZCvav6pe ~ a pie,
i
rbrrpos 8p' bp&v ?r6~cpovb x t p r p $ u a ~ o ;
'
I t was natural and usual for the rhetoricians to admire hfenander for the value
t o public speakers of a study of his style. Cf. James W. Cohoon, Rlietorical Studies
i n the Arbitration Scene of Menunder's Epitrepontes (Boston, I ~ I S ) ,who cites (p. 1 4 1 )
Quintilian, X, i, 69-71. Cf. also G. Icaibel, Comicorurn Graecorum Fragmenta
(Berlin, 1 8 9 9 ) , [CGF), p. 82: Hermogenes, n t p i ptOb60u b t ~ v b ~ q ~C.o s3, 4 , n t p i r o c
K O ~ L K ~XCyerv.
~ S
1 4 . i d e m ibid. C . 36. G q p w o p i a 6rhXoyos Kopor6ia r p a y w r 6 i a uvp?rbura
ZwKpaTrKb 6rb rrvor 6rrX+js pt066ou ? r j l v r a r X i K f r a 1
. . . ~ o p w r 6 i a s62 rX0Ktj
Kai
~ ~
ytXoia, ~ i j p
v i v x r ~ p i j vu w + p o v r r b v ~ w v ,~ i j 61:
v ytXoiwv ?rapapuOoupivwv KTX.
Dindorf, op. cit., 11, p. 5: (Argurnentum Orestis.

0xbtlturc).
Ibid. p.
'Apruro+hvous ypappa.rc~oO
r b 66 6 p 6 p a K w p r K o ~ i p a vZxcr T j v ~ a ~ a u r p o + $ v .
7
(Argumentum Orestis)
~b 6k 6 p 6 p a r l j s ' E ~ b l j q s~ a 700
i r a p 6 v ~ ~6phs
TCXOS b x b ?rivBour h p x o v ~ a r~ a X$youurv

i
t i r xaphv.
Ibid. p. 9 : (Ar,qmentum Orestis) i-K Xir?rqr y i p h p x t r a r ~ a t i is Xir?rvv
~ t X t u r & ri,
r a p b v 61: 6 p 6 p h i u r r v i~ r p a y r ~ o cK O ~ L K ~ V .X$ytr y i p t i s T ~ rSa p ' 'AxbXXovos 6raX-
paror r b
Aristotelian Dejinitions of Tragedy and Comedy
107
ing comparisons, thoroughly in the spirit of this schoo1.l The same

point is not overlooked in the Poetics, but it is more concisely and emphatically expressed in the dialogue from which the grammarians seem
to have derived the idea, since there are no other indications of parallels
to the technical treatise in the productions of this scho01.~
The Alcestis of Euripides also aroused adverse comment by this
group of grammarians, led by the head of the scho01.~The discussion,
thus begun, has been continued by Tzetzes and others down to the
present day." An observation on Andronzache 32, however, takes pains
to point out the presence of the proper elements for genuine tragedy
in the play.5
This point of view is to be found also in the comments on the Ajax
of Sophocles (lines I 123 and I 127) as well as on the same poet's
Electra? But in this case, as in others, space does not permit the citaIbid., p. 347:
TOGTO76 6 ~ B p ai K T P ~ ~ L K OK FW ~ L K ~ Yi K. y d p uup4op&v is t b f l ~ p i a v
'H ~ a 7 h X q t ~r isj s r p a y q 6 i a s $ eis Bpijvov, $ t i s ?rhBos ~ a r a X b e r ,6
...
~arjvrqutv
6 i r i j s ~ w p q 6 i a st i s U T O Y ~ &~S a 6raXXayis.
i
89tv 6pBrar 7b6e 76 6pBpa
K W ~ L K
~a
~ r~ a
Xjttr
xpqubpevov. GraXXayai y d p r P b s MtviXaov ~ a 'Opiurqv.

i
dXX& ~ a i vi r i j ' A X K ~ U T LEK~ L
uup40pijv eis eb+poubvqv ~ a bvaprorjv.
i
~ a 6poiws
i
~ a kvi T u p o i Zo+o~Xious dvayvw-
ci?reiv ?roXXd r o r a 0 r a i v r i j r p a y q 6 i p e i r p i u ~ t r a c .
prupds K a r d 76 rkXos y i v e ~ a r ~, a ci?rXijs
i
Cf. A. Trendelenburg, Grantmaticorurn Graecorzrm de Arle Tragica IwEiciorum

Reliquiae. Diss. (Bonn, 1867.)
Dindorf, op. cit., IV, p. 87: (Argumenturn Alcestidis. 'Ap~uro+bvous r p a w a r r ~ o i b?r%ears).
j
...
76 61 6pBpa ~ w p r ~ w r i p a2vx t t r?jv ~ a ~ a u ~ p o + j v .
T 6 6 i 6pBpb
r&v r p a y r ~ i j vi~0bX~ a r a a ~ p k q k?rap&

~.
l u r r u a r u p r ~ i r t p o v871
, cis x a p d v ~ a j6ov?jv
i
Xerar Qs h v o i ~ e r a74s ~ p a y r ~ i ~j so l j u t w sa r e ' O p t u ~ q ~
r a ?j i " A X ~ q u r r sb, s i-K uup+opBs
p i v hpxbptva, eis tb6arpoviav 6 i ~ a x a
i pdv Xljtavra,
6 Eurr pBXXov ~ w ~ 6 l i-xbpeva.

a s
Cf. H. W. Hayley, T h e Alcestis of Ez~ripides(Boston, 1898), pp. xxiii-xxiv,

who cites seven different theories, with the relevant literature.
Dindorf, op. cit., IV, p. 129: Xiyer y d p Ljs vrv 4 a p p i r ~ o r s~ e ~ p u p p i v o r soi. +abXws
u@
r Ebpr?riGn + ~ U K O Y T ~ Sh i r p a y r ~ o i s? T ~ O U ~ T O L Kwpq6iav
S
h ? r o p v q p a 7 ~ u a p t v o~ ~y K a X o ~ T
abrbv GrartBciuBar. y u v a r ~ & vr e y d p ir?rovoias ~ a r bXXjXov
'
~ a <jXovs
i
~ a XorGopias
i
Kai EiXXa bua eis Kwpq6iav uvvreXei, r a 0 7 a dsa&inavra
~ 0 0 7 076 6pBpa h?rerXq+kvar,
i y v o o l v r e s . 6 a a y d p ttis r p a y y s i a v uuvreXei, r a l r a ?reprkxer, 76v 86varov 700 N C O ~ T O hkpou ~ aBpevov

i
IIqXkws i-v rkXer, ii?rep i u 7 i r p a y r ~ b .
P. N. Papageorgius, Scholia in Sophoclis Tragoedias Vetera (Leipzig, 1888),

p. 85: 1123. r d r o ~ a 0 r au o + i a p a r a o b ~o i a t i a r p a y y 6 i a s per& y & p .p+v bvaipeurv
i - ~ t ~ r e i v a7.6
r Gpirpa BeXjuas i-11.uwtbuaro ~ a 2Xuut
76 r p a y r ~ b v~ 6 0 0 s . 1127. r d 6&
i
70~0070~ w p y 6 i a spi?XXov oh r p a y q 6 i a s .
Iln'd., p. 103: 62.
~ a p i i r i j v Oaupaurljv e b a r 66tn ( t i ) i v r p a y y 6 i p ~ a pbBy

i
?ra-
ha&+ r e r 6 X p q ~ B71 KaK6qfles t i z t i v Kai ? ~ ~ O U K ~ O L I cis

U T robs
L K ~ ~KaB' iaur6v 8 ~ C X X O Y
ijppole ~ w p y s i p .
A. Philip McMahon
tion and close analysis of all the passages which exhibit a reliance on
the traditional formulas. Those given should, however, serve to illustrate the persistence of definitions which we find most clearly recorded,
among surviving texts, by the Roman grammarians.
The familiar conceptions of tragedy and comedy are indicated in

Polybius, who contrasts drama and history in favor of his own subject.' In another place he declares that dwelling upon misfortunes is
more appropriate to tragedy than to h i ~ t o r y . ~
Dionysius Thrax was a pupil of Aristarchus, and wrote the earliest

Greek grammar, a brief work: which continued popular for many
centuries. I t was commented upon by scholars down to the time of
the Italian Renaissance. The expression of Dionysius with regard to
tragedy was: "Iva r1jv pkv rpayqGiav 1jpwi'~Gsdlvayvij~cv,and for
S . ~ commentators, in keeping
comedy, T$VB i ~wpqBIav~ L W T L K G The
with the tradition of On Poets, stressed the words 1jpwi'~Gsand PLWTLKGS
as most significant.
I
T. Buttner-'CVobst, Polybii Historiee 11,56 (Vol. I , Leipzig, 1905, pp. 192-193):
r l ! y ~ i r v o v r a sob6i roi-s iv6e~op6vovsXbyovs r q r t i v nai s h ?rapc?rirytva roir ii?rohtr&ioors

&$aprOpeiu6'a~,~ a t l h n t pol rpayy6royph@oc, 7Gv be ?rpaxOkv~wv~ a / jl ~ O i v ~ w
K ~v T ' kX$Otrav
abrGv ~ V ~ ~ O P P n~ aEpLr aY v , ( ~ ) i i~v h v up i ~ p r aT U ~ X C L Y W U L Y B v ~ a . r h y d p ~ i X c insopias
s
nai ~payc$iar 06 r a b ~ b v ,dXXd robvavriov. &KC?
p2v y d p 6t2 6rd TGV nr0avwrcirwv
Xirywv i~?rX?Ear~ a +i u x a y w ? l j u a ~K Q T ~s;t raphv TOAS khoirov~as,&vOb6e 62 6rk T G ~
dXqOrvGv Zpywv ~ aX6-,
i wv, KTX.
I'olybius, XV, 36. (Buttner-'CVobst, 111,p. 315).

Cf. G. IIoerschelmann, De Dionysii Tlzracis Inlerpretibus Veleribzrs, Leipzig,
1874 A. I-Iilgard, De .lrtic Gram7izeticae ab Dio+zysio T h r a c ~Conzpo~ilaeZntrrpre~
Dirtribucndis. (Progr ) Leipzig,
fationibus Veteribzas i p z S i n g z ~ l oComme~zlarios
1880. A. Hilgard, Scholio i n Dionysii Thrncis Artem Grammatiram ( G r a n ~ f ~ ~ o t i c i
Gracci. Pars tertla Leipzig, 1901 ) T. Davidson, The Grammar of Vionysizts TItrax
Tra?zslated jrom the Grerk. St. Louis, Missouri, 1874.
Immanuel Bekker, A n e ~ d o f aGraeca. I1 (Berlin, 1816), p. 629: 'Avh?vwuis kurr
K ~ T ?rpouy6iav,
&
nard 6rau~ohrjv.
&K
pkv y d p r?js bnonpiuews ~ r j vbperilv,
&K
6 Tljs
Aristotelian DeJinitions of Tragedy and Comedy
109
The commentaries of Diomedes and Melampus on Dionysius Thrax

and they can for our purposes be conveniare inextricably c~rnbined,~
ently considered t ~ g e t h e r .The
~ influence of etymology in accounting
for the origin of the dramatic species and giving a suggestion of their
nature, an element present also in the Poetics, receives ingenious development in these commentarie~.~
The purposes of tragedy and comedy are stated in a short notice
prefixed to Arist~phanes,~
emanating from the school of Dionysius
Thrax.G I n this and other scholia of the same group are to be found the
eventual sources from which many of the apparently strange mediaeval
glosses were d e r i ~ e d .The
~ moral effect of tragedy, according to these
1 Floruit 7th century. Cf. F. Susemihl, Geschichte der Griechischen Litteratur i n
der Alexandrinerseit (Leipzig, 1891), 11, p 173, note.
2 Raibel, CGF, p. 10, note: Scholia haec omnia Melampodis, si codici Elamburgensi, Diomedis, si Bz~rbonicocredas.
3 Cf. Hoerschelmann, op. cit., pp. 27 ff., Kaibel, CGF, p. 11. There is apparent
reference to the censorious character of the Old Comedy in the following, but the
is interesting. uwpwr6ia eurlv r j Iv p k u w ~Xa00 ~ a r ~ y o p$i a7 0 ~ ~
explanation of PLOTLKDS
G~pouievurs. e Z p q ~ a r6; n-apd 76 K & ~ ~W ar bi wiBrj, I u ~ 61i ei6os ?ror$paros Iv u&pars uard
ywr6ia iuropiav Bxer uai dn-ayycXiav ?rphfewv yevopkvwv, r j 62 ~ w p w r 6 i an A b u p a r a

Cxcr flrwr~uijva p a y p h r w v .
p.
rep^-
Cf. Hilgard, Scholia, p. 18, lines 3 and 13; p. 20, line 7. Cf. Kaibel, CGF,
$ 62 IrvpoXuyia rjjs r p a y w ~ G i a sIuriv a t r ~ fj. BTL r p h 7 0 v kXdpPavov En-aBXov o i
11.
x e i a &~6rj. rpaxDrepov y d p ~ a +ie v ~ r f o vrtai
60ufiarov r b rGv b'pljvwv ~ 1 6 0 s700 yeXwro?roreiv, i j oiovti rerpaywvwr6ia. oi y d p xopntrai airrijv Iv rerpay&vwr u x r j p a r ~iurhpevoc
r d T ~ TU ~ C L ~ L K~ ~? ~~ CU ~ E ~ R Y U YKTX.
TO,
5 Raibel, CGF, p. 1 2 , note: "a dextra adscripsi dissertatiunculam IIepi rjjs
~ w p w r 6 i a sin libris Aristophaneis servatam Laurentiano 0 Ambrosiano A, praeterea
in editione Aldina."
6 Raibel, CGF, p. 14, line 45: uai rjjs piv r p a r w ~ 6 i a snuon-br r b cis Bpijvov urvijoa~
Oiov, $ 62 uwpwr6ia uuviu77)urv.
Cf. IIilgard, Scholia, p. 306, line 15,for a somewhat different discussion from
the Scholia Marciana: (Heliodori) ~ ~ p v 6 Iuriv

ia
Iv p f u y XaoD u a r ~ y o p i a~ a 677po-
s Iv 70:s brjpo~s$6eu@a~.*HpwCuijs
uieuurs. eipqrar 61- uwpy6ia d n b TOG kv r a i s u h p a ~ uai
A . Philip McMahon
commentators, had nothing to do with the modern interpretation of
the ~ h o a p u ~mentioned
s
in the Poetics, a work which may have been
known to them, but was by them apparently judged to be unimportant. Codex C of the commentaries of hfelampus or Diomedesll states
the case for tragedy, together with some observations on the measures
taken by the actors to secure the proper effect of grandeur. The justification of tragedy, as will be seen, depends upon the moral edification
to be derived by citizens from seeing the evil which fell to the lot of
even great heroes because of their errors. Tragedy thus furnished a
warning against wrong-doing, to the spectators in the theatre.
Some of the remarks apply rather to the duties of the actors than to
the nature of the dramatic species, but as can be seen from the context
and from other passages already quoted, they did not preclude the
traditional definitions of the dramatic specie^.^ Possible reference to
o;v perd aoXXijs utpvbrgros ~ a iaqppivqs
i
@wvijs ~ L ~ O C ~ C Y O U706s
S
+was, ~ ~ L W T L K6;~ S
i
p t r d ~ ~ X W T ~O aS aoXXijs
i
~ i u r e ~ b r g r o~s a iXapbrqros
6 a o ~ ~ r v o p i v o vr sd ~ L W T C ~K ~a r ih
n a p t r u a y b p ~ v anphuwna. Aei 6i- y r v & u ~ t r v ,61.1 aoXX$ Gra+opd rijs r p a y y 6 i a s tiai rijs
~ w p y 6 i a sBTL
, $ pi-v r p a y y 6 i a nepi tjpwi'~Dvn p a y p b r w v ~ a n pi o u I n w v XCyer, ?j 6i- tiwpy6ia
i $ pi-v r p a y y 6 i a r d r&X1a t p i u @ a y b v tiai +bvwv i x e r , tj 6;
b a t j X X a ~ r a 1r o b r w v ~ a 671
~ w p y 6 i anepi hvayvwpropo6, ~ a 671
i r j pi-v r p a y y 6 i a iuropiav ~ a bi a a y y e X i a v E X E L
npbEtwv yevo,uCvwv, t j 6i- ~ w p y 6 i a6rhaXaopa ~ i w ~ i t i i nj vp a y p d r w v . ~ a 671
i abXrv $ pi-v
r p a y y 6 i a FlaX6ec r6v plov, $ 61 ~ w p y 6 i auuviurqor.
Hilgard, Scholia, p. 17, lines 16 ff. Kaibel, CGF, p. 11: rpaywr6ia XCyerar r d
~ a AiuxbXou
i
r b v r p a y i ~ 6 vaoltjpara, I s r d 706 ELprni60u ~ a ;SO@OKX~OUS
i
~ a r ib v
yeybvaur
6 i 06701 Cai ~ b xpbvwv
v
rDv 'AOgvaiwv. r p a y r ~ o i6 i ovres ~ a i
TOLO~~W
V.
BiXovrts O@eXeiv K O L V ~ L706s rijs nbXews, napaXapl3buovris r i v a s cipxains iuropias r b v
~ aBptjvous,
i
tjphwv ixobuas n6Bg rrvb, EuO' are ~ aOavbrous
i
kv Othrpwi raiira &7re6ei~vuv~0
TOTS bpbut ~ a ~ Ki O ~ O U U L ~V ,Y ~ ~ L K V ~ aapa@vXhrrtuBar
~ E Y O L
r b dpaprbverv. ci y d p oi rgX1KOFTOL qpwes roiaGra E T ~ U X O Y , 6qXovbri d p a P r g p d r w v a b o i s npovagpypkvwv, nbuwi
pBXXov $peis ~ a oii K ~ B $pBs
'
dpaprtjuavrcs atru6ptOa; 6ei o t v , I s npoeipgrar, J s oTbv
r e Plov hvapdprgrov ~ a @rXovo@Irarov
i
p e r a 6 r I ~ t r v . ( a ' I 4 ~ X e i a ro6u r b v ~ O X L T DIU
j
i ~ 1 6 t ~ ~ v 6 p t 6v i0 T1 ~ $pi)wv
dJuavei 7 d aiirbv n p b u w n a
V
r b v 7paylKDv ~ O ~ ~ eiU7j'yc~o.
U L S
apDrov pi-v kaeXiyovro av6pas 706s pei{ova @wv$v Bxovras, 6ebr~pov6i- Pouhbptvo~~ ( a i
7 6 u h p a r a 6er~vberv $ ~ w i ' ~ Cpl366as
d
k+bpovv ~ a iip a r i a ao6ljpg. r a b r g v ofiv r $ v
r p a y w i 6 i a v 4 g u i 6eiv b rexvr~irs$ p w r ~ D sd v a y r v I u ~ e r v ,rouriurr p t y h X g i rijr @wvijr.
2 Kaibel, CGF, p. 10: kti rijs pipipews y d p Cvbpera 6eiwvvvrar r d h ~ a y r v w u e b p t v a .
6ei y d p r d phv rjpwir:d u u v d v w r rjji +wviji h v a y ~ v ~ u ~~t ravptji k ~ X e h l ~ p i v rq d~ ,6 ; BLWT L K ~T O U T ~ U T Lr d K W ~ L K
Os&&v TDLP ~ w Lrou:Curr
,
p~poupkvous y u v a i ~ a svCas t j y p a i 6 a s
r j 6c60r~brasr j h p y r ~ o p i v o u siivspas rj h a npCntr r o i s ciuayopivors a p o u I n o r s r a p &
M ~ v h v 6 ~ wij r 'Apruro@bvtr t j r o i s hXhois tiwpraois.
Aristotelian Dejnitions of Tragedy and Comedy
I Ir
the Poetics is seen in the Scholia Londinensia, but the use of the word
i i p p ~ ~ p odoes
s
not alter the fact that this discussion as a whole is
unmistakably Peripatetic and exoteric in origin.' A Byzantine commen~ S mean
tary refers to Homer for tragedy, and asserts that P L W T L K may
either censure or approbation, citing the life of St. Peter in contrast
with that of J ~ l i a n . ~
The gloss on ~ ~ a y ~ in
d ithe
a Etymologicum Magnum, as Kaibel
noted, is derived from the school of Dionysius Thrax or some writer
following that group.3
Plutarch discusses tragedy in several places. I n one passage he
cites Gorgias for a paradoxical remark on the element of illusion in
tragedy, which to many minds seems surprisingly m ~ d e r n .He
~ disapproves of the performance of tragedy in a way which seems to have
been standard doctrine among the Stoics16and he alludes to the pathos
of tragedy several times?
Hilgard, Scholia, p. 475, lines I ff.: [Scholia Londinensia in artis Dionysianae.]
r p a y y 6 i a kuri j3iwv r e r a l X l y o v $pwl'~GvZpperpos pipqurs i x o v o a uepvlrqra per6
(Heliodori) -- Z q ~ q ~ h o6kv ei ~ a ri d r p a y r ~ bCUTL B L W T L K ~ .
x h o ~ i j s rrvos.
+alverar y&p ra1 a i r d TG B i y yrvwo~6peva. K a i h e ~ ~ i o671
v ?rohh$ ?j Bra+op& sijs
s p a y y 6 i a s r a i 7ijs ~ w p q 6 i a s671
, $ pkv ~ p a y y E i axepi $pw'iKijv ?rpaypbrwv r a i ? r p o u b m v
Xkyer, $ 6& Kwpy6ia &?r?jhha~sacT O ~ ~ T W V.
K a i 6 7 ~$ pkv ~ p a y y 6 i aT & rCX1 xepl
a 4 a y G v ~ a +6vwv
i
Z X ~ L $, 6 2 ~ w p y 6 i arepi bvayvwpropoD. ~ a 671
i ?rbhrv ?j piv r p a y y 6 i a
hirer 7i)v Biov, $ 82. ~ w p y 6 i au u v i u ~ q u t v . . . Ara+&per 6e r p a y y 6 i a rwpqdias, 6 7 ~$
piv r p a y y 6 i a i u ~ o p i a v~ a di? r a y y d i a v Zxtr ?rpbEewvyevophvwv, $ 6k ~ w p y 6 i a?rXbopara
? r e p ~ k x ePLWTLKDV
~
~paypb~wv.
Hilgard, op. cit., p. 569, line 24: ~ w p y 6 i a v6; oi Lpxaibr $6yov r a l ~ e p i u u p p a
v
((PLOTLKDS))
Zheyov eivar. Eyb B i Xhyw, 6rr ~ a hi i $byow ~ a I xi l & ~ a i v oXappbverar.
O U ZUTL y d p Pies & ~ L K O S otov
,
6 j3ios
y d p eixev 6 rsxvrrls, T O U T ~ U T L V~ a s d76w ? K ~ U T PLOY.
TOO dyiov n67p0lJ Kahbs, 6 62 70; ' I o v ~ L ~K v~ oK ~~ ~S .
Kaibel, CCF, p. 16: spaywr6ia kuri Piwv ral h l y w v $pwrrGv pipqoir.
...
..
' On Plutarch's merits as a critic, cf. Sandys, op. cit., I, p. 304.
6 Quomodo adolescens poebas audire debeat, 15 D., in G.N. Bernardakis, Plzckrchi

Chaeronensis Moralia, I (Leipzig, 1888), p. 35: r o p y i a s 62 r$v s p a y y 6 i a v eixev d?rC
s p$ d ~ a ~ f i u a v r~aoi s6 b?rarq@eisuo+b~epos700
~ q v+)v
, 6 7' b ~ a r f i u a s6 r r a ~ 6 ~ e p oTOO
b?raq@ivros. Cf. F. M . Padelford, Essays on the Stz~dyand Use of Poetry by
clti
Plutarch and Basil the Great (Yale Studies in English, XV, New York, 1902).
6 De Gloria Atheniensium, 348 D. (Bernardakis, op. cit., 11, p. 463.)
Qua quis ratione se ipse sine invidia laudet (Quaestionum convivalizcm, VII,
7" E). (Bernardakis, op. cit., IV, p. 289.)
'
In the ninth chapter of the treatise On the Sublime (first century

ascribed to Dionysius Longinus in every edition from 1554 to
1808 occurs a passage which, according to Saintsbury? illustrates l ' the
Classical distrust of, and distaste for, the Romantic, and the comparatively low estimate of manners and characters." The significant
sentence is: "For such are the details which Homer gives, with an eye
to characterization, of life in the home of Odysseus; they form, as it
were, a comedy of manners." But this is merely another instance of
the familiar ancient view expressed by Aristotle and others, that
tragedy finds a parallel in the Iliad and comedy in the Odyssey?
As Roberts points out,Qhis chapter presetlts another familiar manner of contrasting the two species, that between 380s and ~hOos.7 This
distinction, touched upon by Aristotle in the Rhetoric, and continued
by other rhetoricians, is made very clear in the treatise On the Szcblime,
where it is stated that, "passion is as intimately allied with solemnity
as sketches of character with entertainment."
.I.D.)
The Stoic attitude toward tragedy necessarily implies the tradiA MS. of the treatise in Florence is more properly entitled: bvwvbpov r e p i G$ovs.
Sandys, op. cit., I, pp. 288 ff.
George Saintsbury, Loci Cridici (Boston, 1qo3), p. 44
W.R. Roberts, Lotzgifzz~so n ths Sztblinze (Cambridge, 1899), I X , 15 (pp. 68-69).
"Longinus," On the Sztblime, translated by W. Hamilton Fyfe (Loeb Classical Library, London and S e w York, 1927), p. 154: rora3ra y b p ?rev r h r e p ; rijv roi,
1
'06vuuhws <f3i~Djsa h @ ~ i o X o y o L p e v ao i ~ i a v oiovei

,
tiwpy6ia r i s lurru <tloXoyoupivq.
poetic^, 1448 b 9. Cf. Kaibcl, C G F , p. 37, on Homer, according to Tzetzes:
Roberts, op. cil., p. zoo, note on I X , Ij.
Roberts hcre cites Cicero, Orator, 37, 128. (Should be 28, 128.)
Cf. Chr. Walz, Iiiretores Graeci, IX, p. joq: [ih- ~ 0 1 i4017FIKOT

,
IIEPI ETPE-
ZEiiZ] 76 ~ 6 0 0 ns oX<, Cv ?TOL$UEL

rpayih.5.
Roberts, op. cit., pp. 116-117 (XXIX,
perhxer rouo3rov, d ~ b u o v$008 ?j6oves.
2).
Fyfe, op. cii., p. 208: rireos 61- Z+ous
I 13
tional formulas. Epictetus exclaims in words echoed by Boethius;

"For what else is tragedy, but the dramatized sufferings of men, bewildered by an admiration of externals?" Elsewhere he points out,
in showing the intrinsic worthlessness of title and rank, that tragedies
deal only with kings and rich men, but the poor do not experience such
great disaster^.^ Again, he refers to the subject of tragedies, resulting
from the pursuit of appearances instead of realities, and he employs
the word ~ h 6 w a ~with
v regard to the tragic theme.3
The attitude of Marcus Aurelius toward the drama is not different

from that of Epictetus. In one place he advises a rational acceptance
of death, as opposed to the tragic obstinacy of the Chri~tians.~
In the
same chapter he condemns comedy and refers to tragedy as an inspiration to stoic^.^ Again, it will be noted, although there is a reference to
1
H. Schenkl, Epicteti Dissertationes, I, 4 (I,eipzig, 18~8),p. 18:
60~eire
~
T 6Lp i v
T i s Xkyer 7aOra;
ii6oE6v r l v a kpG ~ a raaerv6v;

i
E p i a p o s a h ~ i oh
r X k y e ~ ; Oi6iaovs oh X2yer;
LAX' 6 ~ 6 ~ fiaurXe?s
0 1
X k y o v u ~ . T i y i p eiarv &XXo r p a y y b i a c ,
4 bvOphawv
ahOq ~ e 0 a u -
p a ~ 6 7 w ur i r C K T ~ S , 61b pk7pov 70~0ii6' ka16er~vhpeva;
Epictetus, I, 24, 15. (Ed. of Schenkl, p. 76.)
Ibid., I, 28, 31. (Ed. of Schenkl, p. go.)
H. Schenkl, Marci Antonii Imperatoris in Semet Ipsum Libri X I I (Leipzig,

U
Tva bai) i 6 ~ ~ i Kj s~ ~ U ~Z W
p xSq ~ a l ,p7) Ka78
1 ~ 1 3 P.
) ~137. (XI, 3): 76 6 i ~ T O L ~ O70670,
~ a Bi u r e ~ a i
$~X7)v a a p & ~ a E ~ vAS
, 01 X p ~ u r ~ a v o &
i
i , X i & XeXoy~upkvws ~ a uepvDs
iiXXov m i o a r , &rpay&6ws.
Schenkl, 09. cit., pp. 138-139 ( X I , 6):

u ~ q v i j fsiuxaywyeiuOe,
T O ~ T O L S~ T ' J iix0toOe
H p G ~ o va i r p a y y 6 i a ~aap$xOquav 380-
kai r i j s pei<ovos u ~ q v i j s .6p&ar y i r p , BTL OGTWS
6ei r a 6 r a aepaiveuOar, ~ a 671

i $ i p o v u ~ v a t r b ~ a o i ~ e ~ p a y 6 7 e s"
. 2 T i ~ e a ~ p h v . "KaL
X l y e r a ~62 TLva bah 7Gv r i r 6 p b p a r a aorobvrwv x p q u i p w s . 0I6v k u ~ l~~ ve i v ophX1ura.
TOTS a p c i y p a u r v y i r p o h x l BupoDuOar.
Piov Oepirerv B o r e ~ + ? r r p o v u r k x v v .
A. Philip &fciMahon
114
the moral utility of tragedy, there is no mention of the

Poetics in any sense.
K&&Z~ULS
of the
Among the precious relics of erudition preserved in Athenaeus are

passages from ilntiphanes %nd Timocles which emphasize the same
distinctions as to proper subject, style, invention, and moral purpose
which were traditional.
L
r a u r i a a p e X 6 p B a v t ~ .per& raOr7v 68 3 pkaq swpw6ia
oKouoa. rpbs oi6v rr K ~ Aroykvqs
~ aXorahv
i
I j vka apbs r i aore aapeiXqarar, q K ~ T ' hXiyov i?rl rrjv ?Kprpljacws +~XorexvLav
haeppirq, l r i a r q o o v . d ~ prhv y d p Xkyerar ~ a 6x6
i T O G ~ W V rrvd xpljurpa, obs b y v o e i r a r
6XXd $ BXq iar@oXrj r i j ~70rabrqs a o ~ l j a t w ss a i bpaparoupylas ~ p h sr i v a r o r h axo?rhv
iahBXe$ev;
1 A. hleineke, Athcsaei Deiprtosophistae (Leipzig, 1858); C . Kaibel, Athenaei
Nazirratitae Dipnosophislarum Libri X V (Leipzig, 1887).
2 hleineke, op. cil., I, p. 394: 'EatrG? b a a r r t i s auvtxijs d?rau~Wv,
eraipe T r p 6 ~ p a -
hey6peva, saruir rrva vopirwu $pas e6piuscrv, baopvljres, r d a a p d sois 6e~avouoq5~arais

b IIorrjoer r6v6c rhv rp6aov
aopkv ue r d r a p ' ' B v r r @ h u e Xtybpeua
~
a o i q p a arb a h v r ' . eZ y t rpGrou oi Xbyor
b r h rDv OtarDv eiorv i y v w p ~ u p k v o ~ ,
a p l v ~ a T Li V ' eiat?v, i)s iiaopviluar pbvov
6.2; rhv ?rorqr1jv. O i 6 i ~ o u vy a p BY p6vov
q5D, r&XXa airvr' i u a o r v 6 a a r $ p Ahros,
pljrqp ' I o s h u r q , Buyarhpts, r a i s e s rives,
r i atiuet7' osros, r i aeaoLqrcv. ilv ahXrv
eZ?rn T L S ' A X ~ p k w v a ~, a r di aar6ia
abur' tbObs t i p q x ' , 871 paveir bairrovev
r$u pqrkp', d y a v a ~ r D v6' "A6pauros eb0kws
+&r, a6Xrv T ' t i ~ e r u r .
2aer03,Brav p+hv 6bvwvr' ei?reiv Z T L ,
(Note 3 on page 1x5).
115
Tzetzes (1110-1180) is called by Sandys "dull as a writer and

His first prooemium to Aristountrustworthy as an authority."
phanes is similar to part of the scholia of Diomedes on Dionysius
Xpkpqs T L S ?j @tiSwv T L S & K U U ~ ~ T T E T ~ L ~
IIqXe? 62 nbvr' EEeurr wai Tebwpw nor&v.
Ai@rXos 6' Cv 'EXevq+opoDurv
& 7 6 ~ 6 2aoa~a6ovua
'
~ a ~it w ~ r ~ p k v q
BpavpGvos irpoD Bt+rXku~arov ~ d a o v ,
b s oi rpayq6oi +auiv, 0;s ZEouuia
Zurrv X k y t ~ vBaavra xai aoreiv p6vors.
706s y d p ~payy6oirsap&ov, el PobXer, owdaer,

6 s &@cXoGu~~ h v r a s . d p;v d v y i p aivqs
ykpwv Trs b r v x t i , wa~kpa6evrbv Oivka.

Bnavra y d p r d ~ t i r o v ' ai-aov86 T L S
b r u ~ + p a r 'BXXors ytyov6.r' Ivvoo6pevos
r d s abrbs a6~oGuvp@opds bgov +;per.
Sandys, op. cit., T, 419.

Cf. Chr. Harder, De Ioannis Tzetiae Historiarz~mFontibus Quaestiones Selectae
(Kiel, 1886). Tzetzes places Aristotle before all others who have dealt with comedy:
cf. Kaibel, CGF, pp. 20, 23: ~ U T E P O V63 ra6ras d x h u a s U K ~ V L K ~T SE wai K O L ~ T L K ~aSh ? U T O L i ~ q y + o a v r o ,. . . b s wai r p b a 6 ~ i j u
ahvrwv ' A p ~ u ~ o r k X qBut
s . he actually seems
to have read little of the Philosopher and he confuses Aristotle with Aristarchus.
Cf. Harder, p. 69: "Aristotelis pauca scripta legerat: De mundo c. 2: ad Lyc. 255.
Exeg. 11. p. 33.-De coelo I, 9: Alleg. 11. 0,180. -Hist. Anim. VIII, 6. cf. 11,
17. Inc. anim. 10: Chil. IX, 113. Exeg. 11. p. 70, 27 confundit Aristotelem cum
Aristarcho, v. Orion, p. 93, 17s." Harder, p. 49, says of him: "Quamquam neque
tales eius sunt errores neque tam gravia quae de suo addidit, ut alterum Ptolemaeum
Chennum nominare liceat, et quamquam excusari poterit respicienti similiter
permultos aequales eius suos composuisse, tamen summae levitatis malaeque fidei
convictus est."
A. P h i l i p M c M a h o n
116
Thrax.l I t is, however, the reference to ,t6Oaputs which has attracted

the attention of some writers to Tzetzes and the Tractatus CoisliaiL ~ Z L Sbut
, ~ the inferences drawn thence with regard to an Aristotelian
theory of comedy as originally found in a lost booB of the Poetics are
more than doubtful. Tzetzes repeats the adverzc judgments of the
school of Aristophanes on the Orestes ant1 .4lccstis of Euripitles, on the
Electra of Sophocles, and condemns the Hcruclcs particularly, his basis
being the formula of Aristophanes on the aims of tragedy and
comedy."
In his Prolegomcna to Lycophron: Tzetzes briefly repeats points
already made.' Again, in his "Verses on the Differences Between
Poets," he reproduces the usual matter, and his view of the moral
purpose of tragedy expressed in lines 57 ff., is strangely overlooked by
1 Kaibel, CGF, p. 17: "Proemia . . . nullo auctoris nomine insignita e codice
Parisino . . . recensuit Studemund. Ioannes Tzetza prooemii auctor videtur."
2 Kaibel, CGF, p. 17: t u r i 6 l ~ w p o r 6 i a
p i p q u ~ sr p b t w s . . . ~ a b ' a p r f p ~ ozab'qpbs
pwr6ias 6 r 1
pkv r p a y W c 6 i a i u r o p i a v
ijstl y l v o p k v a s
~ X B LI C ~ b; z a y y t A . i a v
z p h t t w v yevopkvwv,
a b r b s , $ 6 k Kwpwc6ia z X h u p a r a z t p c i x e r
3V s
P L WLKGV
~ zpaypbr~v.
r i a i 6 ~ c~ l j sp i v ~ p a y W c 6 i a suriozhs 7 6 t i s 8pijvov ~ ~ v i j ruo ia' s~& ~ p o a ~ a rsl j,s 6 i
K W ~ W L ~ ~ ~ S
t i s ykXwra.
3 Rutherford, Starkie, and Cooper. Cf. Cooper, Aristotelian Theory of Comedy
(New Yorlr, 19221, pp. 10ff., 224 ff.
l
pcylb'ous
4 Kaibel, CGF, p. 50. tiwpwc6ia & u r i p i p q u ~ si r p b t t w s yeXoiou ~ a cipoipov
r t X t i o u , x w p i s I x h u r o u r G v popiwv &v ~ o i sc Z 6 t u ~ ,6pOvrwv ~ a (06)

i
61' b r a y y t X i a s , 61'
rhv y k X ~ ~ a .
6
Kaibel, CGF, p. 21, 1. 50:
Zs~ov6 i ~ w p w c 6 i a sp l v 7 h p t p c y p i v o v Zxtcv 70:s utiOp-
p a r t y i A w r a , r p a y w 1 6 i a s 6 l xkv8q riai u v p 4 o p b s .
Cf. also 1. 64:
rkXos 62 r p a y w c 6 i a r p l v Abtcv r h v P i o v , ~ ~ p ( c ~ c f i 5i ai su u v ~ u r B vabri)v
KTX.
6 Lycophron (born c. 3goljzj B.c.) was responsible for the arrangement of the
comic poets in the Alexandrine library. His work on comedy, in eleven books, is
of such inferior quality, to judge by available fragments, that it may well have been
the pedantic source of much that is unprofitable and confusing in Tzetzes and the
Tr(zrta1us Coislitiiiinrrs. Cf. Sandys, op. cil., I, p. 122.
Kaibel, CGF, p. 34.
8 Kaibel, CGF, pp. 34 ff.: IUANNOT T O T T Z E T Z O T Z T I X O I 1IEI'I AIA@O-
P A Z 1IOIIITON.
I I7
those critics who value his use of the word K & ~ C L ~ U L Sso highly.' He
frequently cites the early grammarian Euclid12particularly in connection with his own verses on tragic poetry, where he gives a definition
of tragedy of some interest."
Together with Tzetzes, the Tractattls Coislinianas is usually cited
by those who wish to discover traces of the survival of an Aristotelian
theory of comedy deriving ultimately from the Poetics. I n my previous
article in 1917I discussed the mechanical character of the Tractatus
and showed how i t fails to supply evidence unavailable to the keen
pedants who composed it from other extant writings of Aristotle. The
Ibidem., pp. 36-37.
~ X i j u r s6& 70;s uhpaaurv $v ~ p u y w r 6 i a .
xp6vwr G~qrpiOq6 i
xij jots eis ~ p i a ,
~ w p w ~ d i ab vp a r e ~ a ~ ip a y o ~ b i a v
V ijvde
T?V p t u a ~ ~ h ~ q v .
~ a U i~ T U P L K ~ ~
duov pkv o;v EoxqKe 75v Opqvw~6lav

~ p a y o r 8 i a vE+auav oi ~ p ~ 7 7a6i 7 ~ .
duov 6& 70O ykXw70s $v ~ a u i~ w p p d r w v ,
~ o p w r 6 i a v%e
' v~o
73v K X + T L V
q5ipe~v.
Bp+w 61 ap6s u h u ~ a u r vquav TOOBiou.

d y d p ~ p a y r ~ b7 isj v &Xar
a b e q Xkywv,
'Pljuous, ' O p i u r a s , @ o l v r ~ a s IIaXap$bets,

,
TOGS{ O v ~ a s&)Xauvev ciyepwxias.
d
K W ~ L K ~
66S
aws y t k i j v
K O ~ W L ~ ~ ~ L S
i l p a a y b n v a ~ a ~ ia ~ o O p y o~ v a +86pou
i
76 Xotadv lj6paiwuev eis eh~oupiav.
0570 X h t ~pl-v
6 ~ p a y w r b i aDiov,
Daepoi 61 ~ a xljyvuurv
i
$ ~wpw~6ia
l j
b p a Kwpor6ia1,
~ a u ia r u p ~ ~uJv
dpoO u ~ u ~ p w x o7171
i s
x a p & p e p ~ y p i v q .
Raibel, CGF, p. 43: lIEPI TPArIKHZ lIOIIIZEQZ

" A K O UXo:adv
~
rtai r d ~ i j r p a y w ~ 6 i a ; ,
x6vors aepruuois a p i v &poi uuvqypkva
it
;v 6 E b ~ X t i 6 q s7~ ~ a X O
i L T O ~a 6 u o ~
E"ypa\l.av bv6pes i v X6yors 6 y p p 6 o r .
Ibid., p. 48:
i a e i 61 ~ a X & sd v ~ uoiaep
a
iypb+q,
a~ovX
t o~abvT i T ~ X O7Sp a y w ~ G i a s .
pipqurs IjtlDv, ~ p b t e w v ,a a t l p p b ~ w v ,
ljpwr~oO7p6aov
uepvoxpemjs
~t
~ i j spayw~8ias,
~t
~ a b~qppkvq.
i
pedagogic purpose of Lane Cooper's recent Aristotelian Theory of

Comedy results in much valuable material, but in the main its use of the
Tractatz6s simply continues the Byzantine methods of the work on
which it relies.
The traces in classical antiquity of direct influence from the Poetics
are slight and problematical. But there can be little doubt, judging
from the words of the ancient authors themselves, that the Greeks
found the definitions of O n Poets most satisfactory, for they used practically nothing else.'
W h a t definitions did the Romansjind suficient to express their ideas

of tragedy and comedy?
There is an even greater unanimity in Latin literature than in Greek
as to the nature of tragedy and comedy. With the exception of the
late compilations of Tzetzes and of the Tractatus Coislinianz~sthere are
no treatises in Greek devoted to the drama which even appear to have
derived the distinctive ~&Baparsclause immediately from the Poetics.
There are none a t all in Latin literature. On the other hand, the sources
which justify us in believing that the stock formulas of Greek and Latin
literature were ultimately to be found in Aristotle's dialogue O n Poets
are preserved for us by Roman scholars. Nor is there any hint that
the essence of drama lies in a conflict of wills; such a theory, so popular
in recent times; is almost as remote from ancient thought as the
s
itself.
modern interpretations of the ~ 6 8 a p a ~clause
The citation of Greek and Roman authors does not, of course, mean that the
ancient definitions of tragedy and comedy were in every instance immediately derived from a first-hand knowledge of the dialogue, On Poets; i t means only that
directly or indirectly such authors were influenced by the definitions originally
found there. The problem of how long the dialogues survived is still another question.
Cf. Brunetihre's theory, similar to Hegel's, in The Law of the Drama, With
an Introduction by Henry Arthur Jones. (Papers or2 Play-makifzg, 111, Dramatic
Aluseum of Columbia University, h'ew Uork, 1914), pp. 79-80: "The general law
of the theater is defined by the action of a will conscious of itself; and the dramatic
species are distinguished by the nature of the obstacles encountered by this will.
And the quality of will measures and determines, in its turn, the dramatic value
of each work in its species."
119
The dramatic species as well as the theory pertaining to them

were borrowed from the Greeks by the Romans. The Roman grammarians were not alone in assimilating the Greek theories; the great
writers of comedy also drew upon Greek theory to justify their own
practices.
Plautus in celebrated passages of the Amphitruo presents his apology
for his work as a tragi-comedy, because characters proper to a tragedy
were introduced in the play,l and again he defends the appearance of
Jupiter on the stage.2 Such remarks were addressed to a public which
would distinguish between the species on the familiar grounds derived
from Peripatetics and preserved to us by Donatus and Diomedes
among the Roman grammarians.
Terence not only brought over Greek themes and presented them to
Latin audiences; he brought over the idea of comedy as representing
PLWTLKGS.A phrase ascribed to Cicero by Donatus is similar to the
1
W. M. Lindsay, T. Macci Plauti Comoediae ( 2 vols., Oxford, 1903), Amphitruo
58 ff.:
teneo quid animi uostri super hac re siet:
faciam ut commixta sit; (sit) tragico [co] moedia;
nam me perpetuo facere u t sit comoedia,
reges quo ueniant et di, non par arbitror.
quid igitur? quoniam hic seruos quoque partis habet,
faciam sit, proinde ut dixi, tragico[co]moedia.
Zbid., 11. 88 ff.;
ipse hanc acturust Iuppiter comoediam.
quid? admiratin estis? quasi uero nouom
nunc proferatur Iouem facere histrioniam;
etiam, histriones anno quom in proscaenio hic
Iouem inuocarunt, uenit, auxilio is fuit.
praeterea certo prodit in tragoedia.
hanc fabulam, inquam, hic Iuppiter hodie ipse aget
et ego una cum illo. nunc (uos) animum aduortite,
durn huius argumentum eloquar comoediae.
A . Philip McMnhon
idea given in the Adelphoe,' and it is not unlikely that Menander, receiving the formula from Theophrastus, passed it on also to Terence.
The moral justification of comedy here briefly indicated became an
essential part of dramatic theory in the Renaissance, and it was vigorously used to entrench comedy against the puritan^.^
The great erudition of Varro included a work on Plautus,3 and for

his vast grammatical learning he seems to have depended on Dionysius
Thrax, who was in Rhodes, when Varro's master, Stilo Praeconinus,
was there.4 Tyrannion, a pupil of Dionysius, who taught a t Rome in
the time of Pompey the Great, and aided in the editing of manuscripts
of Aristotle and Theophra~tus,~
was also followed by Varr0.F There
is no evidence that the Poetics was among the manuscripts brought to
Rome to rejuvenate the Peripatetic school, and no trace of its direct
influence can be established in Latin literature. But Varro wrote of
poetry and poets,' and the Peripatetic foundations are apparent in
Sidney G. Ashmore, P. Terenti A j r i Comoediae. T h e Comedies of Terence
(New York, 1908), p. 264. Adelphoe, III, iii, 6 0 (414) :
denique
inspicere tamquam in speculum in uitas omnium
iubes atque ex aliis sumere exemplum sibi.
Zbid., 11. 74 (428) :
postremo tamquam in speculum patinas, Demea,
inspicere iubeo et moneo quid facto usus sit.
fishmore, op. cil., p. 6, in our day iecls c:illed upon to defend Terence and
Plautus against the charge of immorality, as did scholars in the seventeenth century.
He asserts: "To be sure, there are plays of Plautus extant which remind us that
the limits of true decency were occl~sionallytransgressed; but even these plays,
though often coarse in tone, fall short of any radical departure from the moral
tenets of the agc which gave them birth."
3 The Pnbzilae t'arronii~izne, consisting of the plays which he recognized, determined the canon of those which now survive. Cf. G. Goetz and 17. Schoell, M.
Tererzti Varronis dc Litzgui~Lntinn qztne Sz~perszlnl(Leipzig, 1910).
Sandys, op. cit., I, p. 176.
6 Otto Jahn and Keifierscheid have stressed the dependence of Varro on Tlieophra~tus. Cf. H. Reich, Der iMimlts (Eerlin, 1903), p. 272, note.
6 Sandys, op. cit., p. 140.
7 M. Schanz, Geschichtc dcr romischeri Lilt., V I I I . 1. 2 (~gog),
p . 433.
12 I
what we possess of his the0ry.l This situation becomes important when

it is remembered that Varro and Suetonius were the chief sources of the
later theorists in poetic^.^
A derivation of tragedy is ascribed to Varro by Diomedes in the
same passage in which the well-known definitions
Another
passage in Diomedes on the Roman varieties of comedy seems to be
derived from V a r r ~ . ~
CICERO
When discussing the dialogues of -3ristotle in my previous article in
the present Stztdies, I have shown how important is the evidence of
Cicero for the character of those compositions15and besides the dialogues Cicero does not seem to have read much else of Aristotle, except
the Rhetoric.O At the beginning of the Topics he complains that no
philosopher in Rome had been able to read the treatises of Aristotle,
but his dialogues were in every hand.? Cicero defined comedy, according to Donatus; s comoediam esse Cicero ait imitationem uitae, s$eculunz
consuetudi~zis,intagigzem ueritatislgnot unlike Terence, as shown above.
I-I. Funaioli, Grammaticae Romanae Fragmenta (Leipzig, 1907)~pp. 209 ff.
Cf. Kettner, Kritisclze Benzerkungen z u Varro urzd Lateinischelt Glossaren (Halle,
1868), and Varronische Stfulien (ISalle, 1865). I n the latter it is shown that Isidore's
citations of Varro were probably through intermediate sources. Cf. also, E. Samter,
Quaestiones Varronianae (Diss., Berlin, 1891); on page 87 there is a stemma to show
the relations of Varro to the great grammarians, but in the light of what is shown in
this article, several additional lines should be drawn in it.
Funaioli, op. cit., p. 320.
Ibid., p. 322.
narvard Studies i n Classical Philology, XXVIII ( I O I ~ )pp.
, 37 ff.
J. I,. Stocks, Aristotelianism (New York, 1927)~p. 122, erroneously asserts:
"There is little evidence that the Romans seriously studied any works of Aristotle,
apart from the published dialogues, except the Rhetoric and the Poetics."
7 Cf. W. W. Jaeger, Studien zur Entstehungsgeschichte der Mekzphysik des Aristoteles (Berlin, rgrz), pp. 147-148
8 Iiaibel, CGP, p. 67.
This definition has been identified as a fragment of the De R e Pz~blica. Cf. F.
Osannus, M . Tul!ii Cireronis De Republica Librorum Fragmenta (Gottingen, 1847),
and his edition of the De Repzrblica, p. 329. Cf. I<eich, op. cit., p. 265; Cicero, De
R e Pzrblica, De Legihus, with an English translation by C. W. Keyes, Loeb Classical
Library (New York, r928), pp. 238-241; (De R e p , IX, X , 11): "Numquam
comoediae, nisi consuetudo vitae patuetur, probare sua theatris flagitia potuissent." Cf. Wilamowitz, Euripides Herakles, p. 55, for the Greek equivalent:
1
His works display Peripatetic ideas but no influence from the Poetics.l
St. Xugustine cites phrases similar to those preserved by Donatus as
found in C i c e r ~ .The
~ idea is repeated in Cicero's oration Pro Sento
Roscio.The public speaker's interest in the diction of drama finds frequent excression i n C i ~ e r o . ~
Of Horace's Art of Poetry,"airltsbury "justly spoke as a "famous

document - the traditional and influential importance of which cannot possibly be exaggerated, though its intrinsic critical value may be
much more disputable." Horace accounts for the etymology of the
word tragedy and the historical origin of the species? Because of
Horace's pedagogic method and probable intention he was, however,
Cf. R. Wurzer, De Cic-erone Tragoediae Iudice (Vienna, 1385).
U. Dombart, Sa~lctiAz~reliiAugz~stiniL'piscopi De Ckitiite Dci, Libvi X X I I
(2 vols., Leipzig, 1908), 11, 9 (pp. 62-63): "Quid autem hinc senserint Romani
veteres, Cicero testatur in libris, quos de re publics scripsit, ubi Scipio disputans
ait: 'Numquam comoediae, nisi consuetudo vitae pateretur, probare sua theatris
flagitia potuissent. "'
3 C. F. R. Jliiller, Cicerol~isScripta Onzlziu, I'ars 11, I (Leipzig, 188o), p. 46:
"et certe ad rcm nihil intersit, utrum hanc ego comicum adulescentem an aliquem
ex agro X'eiente nominem. lctenim hanc conficta arbitror esse a poiitis, ut eflictos
nostros more in alienis personis expressamque imaginem (nostram) vitae cotidianae
videremus."
Cf. I<. lclotz, A l . Tzlllii Cicerollis Scripfa quoe Muliserulit Ontliia, Pars I , Val.
I1 (I,eipzig, 1852). Orator, 1, 28, 128 (p. 27), and 111,8, 30 (p. 151); Brutus, 55, 203
(pp. 243-244), and elsewhere. Also, Ue A*utura Deorzrm, XXIX, pp. 71-72.
6 Cf. Norden, "Die Composition und Litteraturgattung der Horazischen Epistula ad Pisones," IIernzes, XL (~gog),p p. 481-5223, where it is shown that the sequence of topics and their treatment are dependent on Greek precedents. Wilhelm
Kroll, "Die I-Iistorische Stellung von I-Iorazens Ars Poetica," Sofzrabcs, LXXII;
N. F. Y I ( 1 ~ 1 8 )p.
, 81.
6 Saintsbury, Loci Cilici, p. 54. A modern work, similar in scope to Batteux's
Les Quatre Po6tiqucs ( r j j ~ ) is
, Albert S. Cook, The Art oj Poetry (Boston, 1892),
containing the treatises of Horace, Vida, and Boileau.
7 .4rs Poetica, 1. 2 2 0 . This passage is cited by Uiomedes in the same passage in
which Varro is mentioned and the fornlula of Theophrastus occurs. Cf. Keil, Gram.
Lat., I, p. 487. Note the translation of this passage in Horace by Goldsmith, in
Cook, op. cit., p 238
I 23
read chiefly as a technical manual for poets,' developing more conspicuously the aim of Aristotle's Poetics in this r e ~ p e c t . ~
While Vitruvius did not have an influence on the Middle Ages,%is

description of the stage-settings proper to tragedy and comedy responds to the traditional ideas of those dramatic species, and his few
words had a considerable effect on the ideas of Renaissance scholars
and architects4
Even the astronomical poem of M a n i l i u ~ reflects
,~
the traditional
conception in his reference to the contrast between the intrigues and
humorous characters of comedy, on the one hand, and the royal subjects and great disasters, on the other.
Ovid in the Tristia refers to the style of tragedy and in the Amores
describes tragedy in a way that may have influenced Milton's I1 Penseroso?
1 Ars Poetica, 11. 89 ff., zzoff., 275 ff., are the passages which record Horace's
acceptance of the traditional characteristics of the dramatic species. There is nothing
of tragic purgation or purification; no apparent relation to the Poetics.
The utilitarian and pedagogic purpose of Ars Poefica is indicated by the fact
that it could inspire such a work as this: William King, The Art of Cookery, in
Inzitation of IIorace's Art of Poetry (London, 17097).
3 Vitruvius, The T e n Books on Architecture, translated by Morris Hicky Morgan
, I 50.
(Cambridge, ~ g r q )p.
4 M. Vitruuii Pollionis De Architectura Libri Decem (Venice, 1567)~V, viii (p.
193): "Genera autem sunt scenarum tria, unum, quod dicitur tragicum, alterum
comicum, tertium satyricum. Horum autem ornatus sunt inter se dissimiles disparique ratione; quod tragice deformantur columnis, fastigiis & signis reliquisque
regalibus rebus. Comicae autem aedificiorum privatorum et moenianorum habent
speciem, prospectusque fenestris dispositis imitatione communium aedificiorum
rationibus. Satyricae vero ornantur arboribus, speluncis, monlibus, reliquisque
agrestibus rebus in topiarij operis speciem deformatis."
T. Rreiter, ill.fifanilii Astronolizica, I . Carlnina (Leipzig, 1907), V, 11. 472 ff.
(P. 139)Tristia 11, 1. 381.
Amores, 111, 2, 13. Cf. R. C. Browne and H. Bradley, The English Poems of
John Milton (Oxford, 1906), I, p. 276. (I1 Penseroso, 11. 97 ff.)
'
124
A . Philip McMahon
The influence of Seneca on Humanist and Elizabethan drama is

well recognized. Several critics have also noted the correspondence
between the contents of his tragedies and the preferred themes mentioned by the Roman grammarians.' I t is probable, however, that the
composition of the dramas was influenced by the definitions preserved
by the grammarians rather than vice-versa. There seems to be present
in the minds of numerous writers on Greek drama the seldom formulated and obviously unfounded idea that the Greek tragedi,os were composed with the principles of the Poetics in view. As a matter of fact
late Greek and Roman comedy was itself influenced, as we have seen,
by the definitions of the dialogue On Poets, but we have no surviving
Greek plays from the later period which show traces of the doctrines or
language peculiar to the Poetics.
Quintilian has a good deal to say of the utility to the orator of a

study of the dramatic poets. Numerous passages display his acceptance of the traditional formulas. He oflers no innovations, and he does
not show any direct influence of the Poetics. He draws the distinction,
popular with theorists of rhetoric, between the fabula of tragedy and
of ~ o m e d y .Hc
~ continues, impressively, the contrast
the argun~e~~ttcnz
between 400s and ~6L0os.Verefers very approvingly to Menander,
and his reference obviously implies the traditional estimate of that
poet and of comedy in generaL6 Quintilian in discussing the laughter of
1 Cf. F I,. Lucas, Seizeca and Elizabethan Tragedy (Cambridge, 1922). Rluch of
the materlal used by St. Jerome against Jovinian is believed to derive from a treatise
De Matrinzonio by Seneca. The reference by St. Jerome to the contents of tragedy,
a5 well as other passages, is accepted by the editor of the Teubner Seneca. Cf. Fr.
IIaasc, L. A ~ z n u e iSrnelae, Opera qzbae S z i p e r ~ n a t ,Szrpplemoztzinz (Leipzig, 1902)~
p. 29, fragment 67.
2 L. Kadermacher, AT. Fabii Qzli~ttilianiInstitzrtionis Oratorine Lib. X I I (Leipzig, 1907).
Ibid., I , 8 , 7 and 8, 8.
Ibid., 11,4, 2.
Ibid., V I , 2 , 8 fi.
0 Ibid., X , i, 69: "Hunc et admiratus maxime est, ut saepe testatur, et secutus
quanquam in opere diverso Menander: qui vel unus, meo quidem iudicio, diligenter
'
I2 j
comedy does nct become strangely dogmatic and arbitrary like some
modern psychologists and critics, but acknowledges the psychological
difficulty without declaring it solved.'
PLINY
Pliny the younger has a humorous reference to the garb of the actors
in one of his letters, where he calls one of his villas comedy and the
other t r a g e d ~ . ~
SUETONIUS
Suetonius, the leading immediate source for most of the later grammarians, is important as an inheritor of the learning of Varro. Parallels
to his statements are to be found in Dionysius Thrax and he survives
in the treatises of the grammarians.3 He is one of the sources mentioned by I ~ i d o r e . Because
~
Diomedes, Euanthius, and Donatus are
our sources for his definition of comedy15of tragedy: and his statement of the differences between comedy and tragedy,' these texts will
be discussed below.
SERVIU~
Servius, discussing Book N of the Aeneid, stresses its comic character, as judged by the current formulas of what constituted comedy
lectus, ad cuncta, quae praecipimus, effingenda sufficiat . i t s omnem vitae itnaginem
expressit; tanta in eo inveniendi copia et eloquendi facultas; i t s est omnibus rebus,
personis, affectibus accomodatus."
VI, 3, p. 7: "Nequeenim ab ullo satis ex~licariputo, licet multi temptaverint,
unde risus, qui non solum facto aliquo dictove, sed interdum quodam corporis
tactu, Iacessitur." Cf. J. Y. T. Greig, The I'sychology o/ Laughter and Cof~zcdy
(London, 1923), Appendix, pp. 225 ff., "Theories of Laughter and Comedy, A Historical Summary."
R. C. Kukula, C. Plini Caecili Secundi Epishlarum Libri Nooem (Leipzig,
1912)~p. 240 (Lib. I X , Epist. vii.)
A. Keiflcrscheid, C. Sz~etuniTranquilli Praeter Caesarzim Libros Reliquiae (Leipzig, 1860), p. 5.
hf. Manitius, Geschichte der Lateinischen Literatur des Mittelallers (hrunich,
1911)~I, pp. 55-57. Cf. e.g. Isidore, Orig., \TIIl, 7, 5.
"eifierscheid, op. cit., p. 7.
IZeiKerscheid, op. cit., p. 5.
Zbid., p. 8.
A . Philip McMahon
and tragedy,' for which he has been duly condemned by modern
scholars just as they have criticized Dante and Chaucer for participating in the same tradition. Saintsbury's comment2 is intended to be
devastating; ((So the Fourth book, with its steady rise toward the
hopeless, the helpless, the inevitable end, is paene rornicus. Certainly
the criticism is, from our point of view." This failure to understand
the difierence between all ancient definitions of comedy and tragedy
and nineteenth-century formulas is repeated even by C o l l i n ~ . ~
The treatment of comedy and tragedy by Donatus was for later

ages most important and influential. His sources seem to have been
the same as those of Diomedes and Chari~ius.~
Donatus defines c ~ m e d yintroducing
,~
the same Greek equivalents
G. Thilo and H. I-Iagen, Scrvii Grammatiri i n Vergilii Car9nina Cowzmcntarii
(Leipzig, 1881), I, p. 459, on Aen. IV: "Apollonius Argonautica scripsit et in tertio
inducit amantem hfedeam: inde totus hie liber translatus est. est autem paene
totus in affectione, licet in fine pathos habeat, ubi abscessus Xeneae gignit dolorem.
sane totus in consiliis et subtilitatibus est; nam paene comicus stilus est: nee mirum,
ubi de amore tractatur."
G. Saintsbury, IZistory of Critirism and Literary Taste i n Europe (3 vols , Edinburgh and London, ~ g o w o q )I, , p. 339.
S. T . Collins, The Interpretation of Virgil with Spcrial Riferivzce lo Macrobius
(Oxford, rgog), pp. 8-9.
Keil, Grammatici Latini, I V (Leipzig, 1854), p. x1: "cum Iliomede ita per universum librum consentit Ilonatus, ut multis locis . . . non solum eadem res tradantur, sed etiam verbs verbis respondeant . . . ab antiquiore auctore, tamquam
communi fonte, uterque suam doctrinam derivaverit . . . non solum cum Iliomede,
sed etiam cum Charisio i t s consentiunt, u t haec amnia ex eodem fonte ducta esse
plane adpareat." Cf. Keifferscheid, Szcetoniz~s,p p. 379-380.
Kaibel, CGF, p. 67: "Comoedia est fabula diuersa instituta continens affectuum ciuilium ac privatorum, quibus discitur quid sit in uita utile, quid contra euitandum. hanc Graeci sic defniuerunt: ~ w p q a i a6 u ~ i v(iFiw~rti&u)~ p a y p i r r w v~ c p i o x i
b ~ i v f i v v o r . comoediam esse Cicero ait imitationem uitae, speculum consuetudinis,
imaginem ueritatis. comoediae autem a more antiquo dictae, quia in uicis huiusmodi carmina initio agebantur apud Graecos (ut in Italia compitaliciis ludicris),
admisto pronuntiationis modulo, quo dum actus commutantur populus attinebatur:
67rA T ~ tihpqs,
S
hoe est ab actu uitae hominum qui in uicis habitant ob mediocritatem
fortunarum, non in aulis regiis, u t sunt personae tragicae. comoedia autem, quia
I2 7
which Diomedes ascribes to Theophrastus, as we noted near the beginning of this article. He then quotes Cicero, and develops the latter's
phrases in a way which shows their substantial agreement with Theophrastus,' and his conception of the dramatic species is implied in his
commentary on T e r e n ~ e .The
~ vast influence of the ideas of Donatus
can hardly be over-estimated; his presence is to be seen in the Middle
Ages f r e q ~ e n t l yand
, ~ his discussion was reprinted in most editions of
Terence from the Renaissance until a comparatively recent date. His
grammar, as is well known, was so popular for ages that his name became synonymous with an elementary grammar."
St. Jerome, the pupil of Donatus and the secretary of Pope Damasus,
had been a diligent student of Terence15and, as one might expect, he
poems sub imitatione uitae atque morum similitudine compositum est, in gestu et
pronuntiatione consistit . . . aitque esse comoediam cotidianae uitae speculum,
nec inuria; nam u t intenti specula ueritatis liniamenta facile per imaginem colligimus; ita lectione comoediae imitationem uitae consuetudinisque non agerrime
animaduertimus."
If what is valuable in the commentary of Servius is really derived from Donatus, the judgment on Aen. IV, mentioned above is all the more easily understood.
Cf. E. K . Rand, "Is Donatus' Commentary on Virgil Lost?" Classical Qzcarlerly, X
(1916), p. 158.
Cf. H. T. Karsten, Commenli Donatiani ad Terenli Fabulas (Lugduni Batavorum, 1912-13); P. Wessner, Aeli Donali qz~od Ferlz~r Commentzcm Terenli, I1
(Leipzig, 1905)~pp. 90 and 92, on Adelphoe, 111, 3,11. 61 and 75; Paul Rabbow, D e
Donati Commenlo in Terenlizcm (Leipzig, 1897).
Dante's estimate of Donatus is to be seen in the fact that he placed him with
St. Bonaventura among the Doctors of the Church, and refers to him in Par. XII,
11.137-138, as:
"quel Donato
Ch'alla prim' arte degno par mano."
'
W. J. Chase, T h e A r s M i n o r of Donatzcs, Madison, Wis., 1926.

"pologia adversus Libros Rz~$rzi (Migne, Patrol. Let., XXIII, cols. 428-429):
"puto quad puer legeris . . . commentarios . . . in Terentii Comoedias praeceptoris mei Donati, aeque in Vergilium." Eusebii Chronicorz~mLib. 11.Interprete S.
Bieronymo. (Migne, Patrol. Lat., XXVII, cols. 501-502). A.D. 359: "Victorinus
rhetor et Donatus grammaticus prxceptor meus Romae insignes habentur."
A. Philip McMahon
expresses again the traditional definition of comedy.' And he cites
tragedy in his controversy with Jovinian to prove the wickedness of
wornanl~ind.~
For Euanthius we have the evidence of St. Jerome. I t is probable

that the first part of the treatise by Donatus belongs to E ~ a n t h i u sa, ~
passage which appears to have come frcm Suetonius. I t is interesting
to compare what he has to say about the origin of tragedy and comedy
with what is said in the Poetics.Vhe accounts are entirely different,
and if evidence were needed, this would again show that the Poetics had
little to do with establishing the tradition. The distinctions he draws
between tragedy and comedy are consistent with those of the scholiasts and other inheritors of the Peripatetic tradition.6
Sancti IIieronymi Epistula L I V , A d F u r i a m de Viduilate Seruanda. Corpus
Script. Eccles. Lat., LIV (Vienna and Leipzig, I~IO),p. 476: "cum etiam comicus
cuius h i s est humanos mores nosse atque describere dixerit, sine Cerere et Libero,
friget Venus" (Eunuchus, 111,v. 6 ) . St. Jerome's words were often cited by educational authorities as a justification for including Latin comedy in the curricula
of boys' schools.
S. Eusebii IIierony?ni Aduersus Jouinianum, Lib. I (Migne, Patrol. Lal., XXIII,
col. 292): "Quidquid tragoediae tument, et domos, urbes, regnaque subvertit,
uxorum pellicumque contentio est. Armantur parentum in liberos manus; nefandae
apponuntur epulae; et propter unius mulierculae raptum, Europa atque Asia
decennali be110 confligunt."
4. Reifferscheid, Euanthi et Donati Commenturn de Comoedia, Prog. (In Index
Scholarurn in Linivcrsitate Litterarum Uratislaviense, 1874). Ed. Scheidmantel,
Qunesliones Euanthianae (Diss., Leipzig, 1883). P. Wessner, U n l e r s u ~ h z ~ n g zur
e~t
Lmteinische~zScholien-Litteratur, (Hremerhaven, 1899).
JVessner, Aeli Donali qzlod Fertur, I (~goz),
p. 14,1. I j: "quamuis igitur retro
prisca uoluentibus reperiatur Thespis tragoediae primus inuentor e t comoediae
ueteris pater Eupolis cum Cratino Aristophanes esse credatur, IIomerus tamen,
qui fere omnis poeticae largissimus fons est, etiam his carminibus exempla praebuit
e t uelut quandam suorum operum legem praescripsit: qui Iliadem ad instar tragocdine, Odyssiam ad imaginem comoediae fecisse monstratur."
i'oetics, 1q48a 3 and 1449 a 10.
kyessner, op. cil., p. 21: "inter tragoediam autem et comoediam cum multa
tum imprimis hoc distat, quod in comoedia mediocres fortunae hominum, parui
impetus periculorum laetique sunt exitus actionum, a t in tragoedia omnia contra,
ingentes personae, magni timores, esitus funesti habentur; et illic prima turbulenta,
I2 9
The grammarian Diomedes, also of the fourth century, is one of the

most elaborate writers on the dramatic species; his influence, combined
with that of Donatus, determined what most men held to be the meaning of tragedy and comedy for fourteen centuries. There has been
some controversy over his sources,l and in particular his relation to
Varro through Suetonius has been debated; but the embodiment of
the tradition in an authoritative form is our present concern. His
quotation of the Greek formulas, and his ascription of that for tragedy
to Theophrastus,3 makes it almost certain that the definitions of the
tranquilla ultima, in tragoedia contrario ordine res aguntur; tum quod in tragoedia
fugienda uita, in comoedia capessenda exprimitur; postremo quod omnis comoedia
de fictis est argumentis, tragoedia saepe de historia iide petitur."
Cf. E. Koett, De Diomedis Artis Poeticae Fontibus (Diss., Jena, 1904); Buchholz, "Ueber die Abhandlung De Poematibus des Diomedes," N. Jahrbnch, XXIII
(18971, PP. 1 2 7 ff.
Cf. Keil, Grammatici Latini, I, p. liv: "In tertio libro Diomedis memorabilis
et plena antiquae eruditionis disputatio est de poematibus 482 ,I3-492,14. eam
totam a Suetonio, quem in fine eius capitis 4 9 1 ~ 0Diomedis appellavit, petitam
esse docuit 0.Iahnius. Mus. Rhenan. noviss. VIII, p. 629. nam quod satirici
poetae tres tantum appellati essent 485, 32, Lucilius Horatius Persius omisso
nomine Iuvenalis, id non posse intellegi, nisi ab antiquiore aliquo auctore, qui
aetate superior fuisset quam Iuvenalis, haec recepta essent. deinde quia Suetonium
in libris suis maxime Varronis auctoritate usum esse constaret, ea quoque quae
Varronis nomine Diomedes adscripsisset 486, 8. 487, 15. 488, I. 489, 18 Suetonio
deberi. ea vero quae de origine bucolici carminis 486,17 sqq. tradita essent, quoniam
cum iis fere consentirent quae de eadem re Probus ante commentarium in Vergili
bucolica scripsisset, ex eo ips0 loco a Suetonio, qui saepius Probi copiis usus esset,
recepta videri. quem autem Suetonii librum Diomedes exscripserit non magis
certo definiri potest, quam quo potissimum ex Varronis libris Suetonius usus sit
. . . ceterum ut ipse Diomedes vix quicquam antiquo commentario addidit, ita
Suetonius in plerisque satis accurate Varronis auctoritatem secutus esse videtur."
But Wessner does not agree with this conclusion (Hermes, LII, 1917, p. 2 1 1 ) :
" . . . wobei ich nicht unterlassen will darauf hinzuweisen, dass die Ansicht
Reifferscheids die ganze Poetik bei Diomedes gehore dem Sueton, nach den Darlegungen von Buchholz und besonders von Kott unhaltbar ist."
Keil, op. cit , pp. 487 ff.: "Tragoedia est heroicae fortunae in adversis conprehensio. a Theophrasto ita definita est, rpayqaia kuriv rjpwi'rijs ~ 6 x 7 s~spiuraurr.
tragoedia, ut quidam, a r p h y q et 465 dicta est, quoniam olim actoribus tragicis
rphyos, id est hircus, praemium cantus proponebatur, qui Liberalibus die festo
A. Philip McMahon
other species came ultimately from the same source. The definition of
tragedy stands between two passages which are directly attributed to
Varro, so that the genealogy of this definition is: Aristotle-Theophrastus-Varro-Suetonius - an unknown number of grammarians and compilers - Diomedes.
The definition of comedy is parallel to that of tragedy and involves
pointing out the differences between the two.
Thus in Roman, as in Greek antiquity, there was a consistent reliance on the definitions of tragedy and comedy received a t the beginning from 0 1 2 Poets. There is plenty of evidence showing the persistence of Peripatetic ideas in Roman literary theory, but this evidence
does not include any indubitable influence from the Poetics. The
formulas which were to be passed on to the Middle Ages were those
from the dialogue written by Aristotle.
Libero patri ob hoc ipsum imnlolabatur, quia, u t Varro ait, depascunt vitem; et
IIoratius in arte poetics
carmine qui tragic0 vilem certavit ob hircum,
mox etiam agrestis Satyros nudavit,
et Virgilius in georgicon secundo, cum et sacri genus monstrat et causam talis
hostiae reddit his versibus,
non aliam ob culpam Eaccho caper omnibus aris
caeditur.
alii autem putant a iaece, quam Graecorum quidam ~ p i r r aappellant, tragoediam
nominatam, per mutationem litterarum u in a versa, quoniam olim nondum personis a Thespide repertis, tales fabulas peruncti ora faecibus agitabant, ut rursum
est Horatius testis sic,
ignotum tragicae genus invenisse Camenae
dicitur et plaustris vexisse poemata Thespis,
quae canerent agerentque infecti faecibus ora.
alii a vino arbitrantur, propterea quod olim rpb: dictitabatur, a quo ~ p C r v r o s
hodieque vindemia est, quia Liberalibus apud Atticos, die festo Liberi patris, vinum
cantoribus pro corolla~iodabatur, cuius rei testis est 1,ucilius in duodecimo."
1 Ibid , p. 488: "Comoedia est privatae civilisque fortunae sine periculo vitae
conprehensio, apud Graecos ita definita, hwpw8ia i u r i v i 8 r w r r ~ G v7rpaypbTwv &~iv611vos
i r s p r o X i . comoedia dicta &7r6 TGV K W ~ G Vh. i j p a ~enim appellantur pagi, id est conventicula rusticorum itaque iuventus *\ttica, u t ait Varro, circuln vicos ire solita
fuerat et quaestus sui causa hoc genus carminis pronuntiabat. aut ccrte a ludis
vicinalibus. nam postea quam ex agris Xthenas conmigratum est et hi ludi instituti
Aristotelian Defilzitions of Tragedy and Comedy
13I
What definitions did the Middle Ages, in general, accept as expressing

proper ideas of trczgedy and comedy?
I t is not necessary for the present purpose to state a principle for

separating the late classical period from the early mediaeval. Many
of the characteristic tendencies of the Middle Ages are already well
developed before the political events occur which are usually accepted
as dividing these eras.'
Boethius, however, is in many ways particularly significant for the
Middle Ages, and began one of its principal theoretical activities, the
harmonizing of Aristotle and revealed religi~n.~Hugo of Trimberg
called Boethius dignus et laudabilis in omni n ~ t i o n eand
, ~ a more modern
editor has said of him: "No philosopher was so bone of the bone and
sunt, sicut Romae conpitalicii, ad canendum prodibant, et ab urbana ~ & p ~g a +6ij
i
comoedia dicta est: vel quod in ea viculorum, id est humilium domuum, fortunae
conprehendantur, non ut in tragoedia publicarum regiarumque: vel E H ~ 7, 06 ~Llpou,
id est comessatione, quia olim in eius modi fabulis amantium iuvenum ~ G p o ci anebantur. comoedia a tragoedia differt, quod in tragoedia introducuntur heroes duces
reges, in comoedia humiles atque privatae personae; in illa luctus exilia caedes, in
hac amores, virginum raptus: deinde quod in illa frequenter et paene semper laetis
rebus exitus tristes et liberorum fortunarumque priorum in peius adgnitio. quare
S
varia dehnitione discretae sunt. altera enim b~ivEvvosacpioxlj, altera T ~ I X ~mpC
urauis dicta est. tristitia namque tragoediae proprium; ideoque Euripides petente
Archelao rege u t de se tragoediam scriberet abnuit ac precatus est ne accideret
Archelao aliquid tragoediae proprium, ostendens nihil aliud esse tragoediam quam
miseriarum conprehensionem."
While Donatus and Diomedes embodied the Peripatetic conception as i t was
handed on to the Middle Ages, it hardly seems accurate, however, to follow the
practice of what is, on the whole, a useful book: Earrett H. Clark, European
Criticism
Theories of the Drama (New York, 1918). I n the chapter on LLDramatic
of the Middle Alges," two authorities are reproduced: Donatus, On Comedy a d
Tragedy, and Dante's Epistle to C a n Graizde (pp. 41-47).
Cf. A. Hildebrand, Boetl~itiszind seine Stellz~ngs u m Cl~ristentume(Kegensburg,
1885)~P. 5 ff.
J. Huemer, "Das Registrum multorum auctorum des IIugo von Trimberg,"
Sitzz~fzgsber.d. Wiener Akad. d. W i s s . Phi1os.-Hist. Cl., CXVI (1888), pp. 145-190.
A. Philip ilIcMahon
flesh of the flesh of Middle-Age writers as Boethius. Take up what

writer you will, and you find not only the sentiments, but the very
words of the distinguished old Roman."
The same formula that Donatus andDiomedes had inherited through
the ages is repeated by Boethius or implied in several places. I n the
Consolatio P h i l o ~ o p h i a ehe
, ~ asks: Quid tragoediarum clamor alizid deflet
nisi indiscreto ictzi fortunam felicia regna uertentem? This Chaucer
translated as "What other thing biwailen the cryinges of tragedies but
only the dedes of Fortune, that with an unu7ar stroke overtorneth
realmes of grete nobley?" Cloetta observes that; "If in these words
nothing wrong is contained, yet they helped to establish the Middle
Ages in error, as is proved to be the case with Chaucer." This remark, and the adoption of his views by most historians of the drama,5
proceeds from two prior misconceptions. One is his idea of the nature of
tragedy and comedy based on German philosophy of the early nineteenth century, and the other is his scholarly purpose, which is to determine how much the Middle Ages knew of classical drama and how
successful they were in imitating it, rather than to discover what
license the Middle Ages had for their theory of tragedy and
comedy.
Boethius, as did Cicero and many others, used tragedy in a metal Chazccer's Translation of Boethius' " D e Consolalione," edited by Richard Morris
(London, 1868), p. ii.
R. Peiper, Anieii Manlii Severini Boelii Philosophiae Consolationis Libri Quinque (Leipzig, 1871)~pp. 28, 1. 36.
3 W. W. Skeat, The Student's Chaucer (New York, n. d.), p. 143. Cf. Bernard L.
Jefferson, Chaucer and the Consolation of Philosophy of Boethius (Diss., Princeton,
1917).
* I&'. Cloetta, Beitrage zur Litteraturgeschichte des Afittelalters t ~ n dder IZenaissance
(Halle, 1890-92), I, p. 17.
6 W. Creizenach, Geschichte des Netheren Dramas (Halle, 1893), I , p. 9: "Unter
diesen Umstanden konnte es um so leichter geschehen, dass allmahlich in den
Lehrbiichern des Mittelalters in Bezug auf die Grundbegriffe des Dramas cine unglaubliche Verwirrung urn sich griff ." But cf. Lynn Thorndike, Science and Thought
i n the Fifteenth Century (New York, 1929), pp. 9-10: "The old slurs and disparaging
generalizations a t the expense of the middle ages are now repeated only by mechanical creatures of habit, by those who stopped thinking and reading twenty or thirty
years ago, and who refuse to give up any catchword or prejudice that was instilled
into their minds in childhood."
I33
phorical sense of suffering.' Their example does not prove, as Cloetta

seems to think in this and other instances which he cites, that, although
Boethius was acquainted with classical drama, when acquaintance with
i t had practically disappeared in later times, a proper theory of tragedy
and comedy had also been lost.
Among the commentaries on Boethius is one formerly ascribed to
St. Thomas A q ~ i n a s .The
~ gloss on the reference to tragedy might give
some hint of a scholastic theory of tragedy if the commentary were
really his. But it has been established that Aquinas did not compose
this work; and although it was very popular and ran through many
editions after the invention of printing, it merely repeats material derived from Isidore in the first place.
One of the most popular and useful books of the Middle Ages was the
compilation of the saintly encyclopedist Isidore of Seville, in whom the
aim of originality was completely a b ~ e n t He
. ~ carried on the traditional
definitions as best he could, although etymology a t times usurped the
r61e of more difficult historical descripti~n.~
Among the most import-
'
Peiper, op. cit., p. 205 (Contra Eytychen et Nestorium, 1. 80): "At si noua ueraque non ex homine sumpta car0 formata est, quo tanta tragoedia generationis?
ubi ambitus passionis?"
Cf. various editions in the Harvard College Library; e.g., Boetii V i r i Celeberrimi De Consolatione Philosophie Liber cum Optimo Comento Beati Thome. [rqgo?.]
3 Cf. P. Mandonnet, Des dcrits Az~thentiques de S . Thomas d'Aquin (zd ed.,
Fribourg, Switzerland, 1910).
H. Dressel, De Z~idoriOriginum F o n t i h s (Turin, 18 74). A. Schenk, De Isid.
Hisp. de hTatura Rer. Libelli Fontibus (Jena, 1909). C. H. Beeson, Zsidor-Studim
(Munich, 1913). W . M. Lindsay, Isidori Eiispalensis Episcopi Etymotogiarum sine
Originum L i l n i X X (2 vols., Oxford, 1911). Lindemann, Corp. Gramm. Vet. Lat., I11
(1833).
His ideas about the etymology of tragedy may have been disturbed by his
literary disapproval of the goat. Cf. Etym. XII, I, 14 (Lindemann, op. cit., p. 74):
"Hircus lascivum animal et petulcum et fervens semper ad coitum: cuius oculi ad
libidinem in transversum aspiciunt, unde et nomen traxit. Nam hirci sunt oculorum
anguli secundum Suetonium: cuius natura adeo calidissima est, ut adamantem
lapidem, quem nec ignis, nec ferri valet domare materia, solus huius cruor dissolvat."
I34
A . Philip McMahon
tant passages is that in his Book VIII of the Etymologies.' Many of

Isidore's unfavorable remarks about the stage of antiquity; which were
those most apt to be carried on by the glossators, were apparently derived from T e r t ~ l l i a n . ~
Kayser held that the common source of Diomedes and Isidore was
probably S ~ e t o n i u s . ~He judged that the references to Varro came
from the same s ~ u r c e But
. ~ Wessner contested this con~lusion.~
At any
rate, Creizenach, like other historians of the drama cites Isidore to
prove that the Middle Ages did not know what tragedy or comedy was,
because the mediaeval understanding of the tragic and comic did not
L'tyn~.V III, vii (Lindemann, 111, p. 26j): "Tragoedi dicti, quod initio canentibus praemium erat hircus, quem Graeci 7 p i r y o s vacant. Unde et IIoratius:
Carmine qui tragic0 vilem certavit ob hircum.
Olim enim dehinc sequentes tragici multum honorem adepti sunt, excellentes in
argumentis fabularum ad veritatis imaginem fictis. Comoedi appellati sive a loco,
quia circum pagos agebant, quos Graeci &pas vocant, sive a comessatione. Solebant
enim post cibum homines ad eos audiendos venire. Sed comoedi privatorum hominum praedicant acta, tragici vero res publicas ct regum historias. Item tragicorum
argumenta ex rebus sunt luctuosis: comicorum ex rebus laetis."
Cf. Etynz. XVITI, xlv: "Tragoedi sunt, qui antiqua gesta atque facinora
sceleratorum regum luctuoso carmine spectante populo concinebant." And Etynz.
XVIII, xlvi: "Comoedi sunt, qui privatorum hominum acta dictis, aut gestu canebant, atque stupra virginum et amores meretricum in suis fahulis exprimebant"
(Lindemann, op. cil., pp. 577-578).
Af. Klussman, Excerpla Tertullianca i n Isidori Hispalensis Etynzologiis (Hamburg, 1892)~pp. 30-31.
J. Kayser, De Vetcrum Arte Poctica Quaestiones Selcctae (Leipzig, 1906), pp. 4445.
Zbid., p. 51.
P. WJessner, "Isidor und Sueton," H e n ~ t c s L
, I I (1917)~p. 292: "Seine Quellen
hat Isidor nur verhiiltnismiissig selten genannt; gerade diejenigen, die er a n meisten ausgebeutet hat, nennt er in der Kegel nicht. Die Citate aus alteren Autoren
hat er fast durchweg aus zwcitcr und dritter Hand, und zu ihncn gehoren unter
anderen auch die Suetoncitate. Es kann zum allermindesten als sehr wahrscheinlich
angesehen werden, dass er keine Schrift von Sueton in Hiintien gehabt hat ausser
vielleicht die Kaiserbiographien. Mann lrann miiglicherweise mit EIilfe von sichergestelltem Eigentum Suetons bei anderen Schriftstellern hier und dort zu der Fcststellung gelangen, dass die von Isidor benutzten Quellen in einzelnen Fallen
irgendwie mit Sueton zusammenhingen, aher mit Hilfe Isidors suetonisches Gut
andemarts nachweisen zu wollen, ist ein eitles Beginnen und muss scheitern, wie
Schmekels Buch in abschreckender mTeisezeigt."
'
135
correspond to nineteenth-century philosophical principles, and the

Middle Ages were largely ignorant of the dramatic p ~ e t of
s antiquity.'
A good deal of value has often been attached to glosses, as indicating

the nature of the mediaeval conception of tragedy and comedy. The
strange statements presented by such glosses are supposed to indicate
that the Middle Ages were entirely mistaken and in no way possessed
an adequate theory of tragedy and comedy. Several observations are
pertinent on this score. I t is to be noted, first of all, that the grammarians Donatus and Diomedes maintained an influence continuously;
that Isidore, in particular, was persistently popular, being the chief
source of both truth and error for most of the glosses; that the misinterpretations are usually due to faulty etymology or to hints derived from
Tertullian; and that the mistakes are mistakes in understanding the
traditional formulas derived from On Poets ultimately and not misinterpretations of the Poetics. I t is also to be noted that glosses have not
yielded the gains once expected from their collection and study. The
amount of independent evidence to be had from them, covering points
not available in surviving classical literature, is indeed surprisingly
W. Creizenach, Geschichte des AAeueren Dramas (Halle, 1893), I , p. 9.
Cf. A. Reifferscheid and G. Wissowa, Quinti Septinzi Florentis Tertulliani
Opera. (Corpus Script. Eccles. Lat., vol. XX, Prague, Vienna, and Leipzig, 18901,
De Spectaculis, passim. But Tertullian introduces 3 keen appreciation of tragic
values when he anticipates the confusion of his enemies on the Day of Judgment.
Cf. op. cit., pp. 28-29: "at enim supersunt alia spectacula, ille ultimus et perpetuus
iudicii dies, ille nationibus insperatus, ille derisus, cum tanta saeculi uetustas et tot
eius natiuitates uno igni haurientur . . . etiam poetas non ad Rhadamanthi nec
ad Minonis, sed ad inopinati Christi tribunal pdpitantes? tunc magis tragoedi
audiendi, magis scilicet uocales in sua propria calamitate; tunc histriones cognoscendi, solutiores multo per ignem; tunc spectandus auriga in flammea rota totus
ruber; tunc xystici contemplandi, non in gymnasiis, sed in igne iaculati. . . ." An
even more frequent source of misleading glosses is to be found in Lactantius, Divin.
Inst., VI, xx (Corpus Script. Eccles. Let. X I X , p. 560 ff.): "In scenis quoque nescio
an sit corruptela uitiosior. nam et comicae fabulae de stupris uirginum loquuntur
aut amoribus meretricum, et quo magis sunt eloquentes qui flagitis illa finxerunt, eo
magis sententiarum elegantia persuadent et facilius inhaerent audientium memoriae
uersus numerosi et ornati. item tragicae historiae subjiciunt oculis parricidia, et incesta regum malorum, et coturnata scelera demonstrant "
136
A . Philip McMohon
small.' Too many of them are the result of courageous ignorance rather
than rare fragments of authentic classical origin.
The Glossarium Ansileubi, edited by Lindsay, illustrates several of
these point^.^ Glosses reproduced by Kaibel go back to I ~ i d o r ewhile
,~
etymology is emphasized in the Commentum Einsidlense in Donati
Artem Minorem4 The glosses developed before the general diffusion of
Isidore are the most apt to err.5 Interesting material of this sort is provided in the C o r p ~ sand
, ~ it is a comparatively easy task to trace the
sources of the mingled fact and fiction in most of the glo~ses.~
On the
whole, however, little positive evidence for the mediaeval theory is to
1 Cf. W. M. Lindsay and H. J. Thomson, Ancient Lore i n Mediaeval Latin Glossaries (St. Andrews University Publications, No. XIII, London, I ~ Z I )p., viii: "We
must banish from our minds the notion that each glossary is an isolated work, the
result of the learned labour of a life-time, the slowly amassed collectanea of some
wide reader like Bede or Lupus. . . . Glossaries are much more hasty makeshifts, the mere result of massing the word-collections that were available a t this
or that monastery and then re-arranging the mass."
2 Glossaria Latina, Iussu Academiae Britannicae Edita; vol. I, Glossarium Ansileubi sive Librum Glossarunt (edited by W. hi. Lindsay and others, Paris, 1926),
p. 129, 398, Comes(s)atio: cotzvivium meretricum; 423, Comicam: tragicam, satiricam, p. 568, 57, tragoedia: luctuosae relatiortes; 60, luctuosum carmen; 66, tragoedia: est quae res publicas amplissimas et regum historias continet; tragoediam autem
a Melpo[ejmene Muss assuerunt poefae inventam. . . .
3 Kaibel, CGF, p. 72: "Comoedia, historia tragoedia.
Comoediae, cantica
agrestia graece. Comoedia est quae res privatorum et humilium personarum comprehendit non tam alto ut tragoedia stilo sed mediocri et dulci. Comoedia est quae
privatorum hominum continet acta. . . . Comoedi sunt qui privatorum hominum
acta dictis aut gestu canebant atque stupra virginum et amores meretricum in suis
fabulis exprimebant."
4 Grammatici Latini ex recensione HenAci Keilii, Supplementum Continens
Anecdota Eleluetica, ex recensione Hermanni Hageni. (Gramm. Let., vol. VII, Leipzig,
1870)~p. 236: "Comoedia autem dicitur a Graeco, quod est 'comos' et 'ode.'
Comos enim Graece uilla, ode cantus dicitur, inde comoedia carmen uillanum de
uilibus ct inanibus rebus compositum. Orestes tragoedia. . . . 'Tragos' Graece
hircus, inde tragoedia dicta est, quia poetis talia carmina componentibus hircus
dabatur pro beneficio."
6 A. S. Napier, Old English Glosses (Anecdota Oxoniensia, vol. XI, Oxford, rgoo),
P. 93.
Cf. especially, G. Goetz, Thesaurus Glossarum
6 Corpus Glossariorum Latinorum.
Emendatarum, vols. VI-VII (Leipzig, 1899-1901).
7 J. F. Mountford, Quotations jrom Classical Authors in Medieval Latin Glossaries (New York, 1925), contains additional material
137
be had from the glosses.' Some of the inadequate ideas are to be traced
to E u ~ e b i u s .The
~ explanation in many early German and AngloSaxon glosses is also to be found in the original use of the word in its
metaphorical sense.
I t is difficult to distinguish between such collections of glosses and
the fuller dictionaries and encyclopedias into which they grew, but the
work of Papias is best treated as a lexicon. His Elementarium Docbut he presents
trinae Erudimentum drew upon the Liber Glossar~m,~
further data secured from Isidore, Diomedes, and Donatus, of whom
the last is expressly mentioned.
JOHN OF
SALISBURY
and UGUCCIONE
DA PISA
I n John of Salisbury are to be found reflections of the traditional

theory of tragedy and ~ o r n e d y .The
~ Magnae Derivationes of Uguccione
da Pisa will be discussed in relation to Dante, but it was based on
Isidore and Papias and affected, among others, Bartholomaeus Anglicus?
Vincent of Beauvais, another very popular encyclopedist, of the
thirteenth c e n t ~ r ywas
, ~ frequently printed, down to the seventeenth
c e n t ~ r y .In
~ that part called Speculum Doctrinale, chapter 109, poetry
Cf. Corpus, IV, pp. 185, 220, 221, 425,and 572; V, pp. 181, 250, 396, and 426.
C o r p ~V,
, pp. 418 and 426. Cf. Eusebius, Eccl. Hist., I, 8.
L. Diefenbach, Glossarium Latino-Germanicum Mediae et Injimae Aetatis
(Frankfurt, 1857), p. 155. Diefenbach, Novum Glossarium Latino-Germanicum
Mediae et I~zfimaeAetatis (Frankfurt, 1867), p. 102. E. Steinmeyer and E. Sievers,
Die Althochdeutschen Glossen (Berlin, 1879-98).
T. Wright, Anglo-Saxon and Old English Vocabularies (2d ed., R. P. Wiilcker,
2 vols., London, 1884), s.v. tragoedia; conzocdia.
Cloetta, op. cit., I, pp. 23-24.
Polycraticus, 111, 8 and VIII, 9. (Migne, CXCIX, col. 491 A. and col. 743 B.)
7 P. Toynbee, Dante Studies and Researches (London, ~goz),pp. 97 ff.
Cloetta, op. cit., I, p. 33. E. Boutaric, Vincent de Beauvais et la Connaissance
de I'Antiquiti Classique au Treizic'me SiLcle (Paris, 1875).
Bibliotheca Mundi. Vincerztii Bellovacensis speculum quadruplex; naturale,
doctrinale, morale, historiale . . . Omnia nuno accurate recogttita . . . opera et studio
dheologorum Benedictinorum collegi Vedastini i n alma academia Duacensi (Duaci,
1624).
138
A. Philip McMahon
is classified into seven kinds, where tragedy and comedy are defined, in
accordance with Isidore. In the Specz~lz~m
Doctrinale, a section, De
poetis,' is carried over from Isidore, and chapter IIO is simply Etymologiarz~wzVIII, 7 reproduced completely.
JOHANNES JANUENSIS
The Catl~olicon,~
another store-house of mediaeval learning, by
Johannes de Balbis or Januensis, depends on Papias, and on Diomedes
or Donatus either immediately, or through such compilations as that of
Uguccione, for its views of tragedy and ~ o m e d y . ~
I n the fifteenth-century Latin-English dictionary, the first of its

kind, the Catholicon of Johannes Januensis is cited for tragedy and
Uguccione for ~ o r n e d y . ~
The metaphorical use of tragedy, which depends on the traditional

understanding rather than upon the special sense of the Poetics, is frequent from the time of Cicero down. I n the Middle Ages it is often
found; among others, in Ekkehard IV,5 Lampert of H e r ~ f e l d ,and
~
Otto of F r e i ~ i n g . ~
One of the works most frequently referred to, by historians of the
drama and others, for data on the mediaeval conception of tragedy and
Ibid., I , lxiv, col. 76.
A copy in the Library of Harvard College: Catholicon. Edita a Fratre de J a n u a ,
Ordinis Predicatorurn. (Cologne, 1497.)
Cf. Cloetta, up. cit., I , p. 26.
' '1. L. llayhew, Tfze P r o ~ ~ p t o r i uParvzblorum,
m
The First Englisfz-Latin Dictionary, 1440 A.D. (E. E. T. S.), 1890, col. 240 and col. 351. Cf. note, p. 629.
G. Rleyer von Kronau, Casus Sancti Galli (St. Galliscke Geschichtsqzbellen, St.
Gallen, 1877-791, PP. 228, 237, 238, 346, 411, and 445.
0. IIolder-Egger, Lamperti fifonachi Hersfeldensis Opera (Hanover and Leipzig, 1894), p. 172.
A. IIofmeister, Ottonis Episcopi Frisingelzsis Chronica s h e IIistoria de Duabus
,
2-3, 7, 317.
Ci-vitatibus (Script. Rer. Germ. sep. ed., Hanover and Leipzig, I ~ I Z ) pp.
1
139
comedy, as well as of the tragic and comic, is Cloetta's book. He bases

his investigation on the classical conception of classical drama, as determined by modern criticism.' He considers only those literary productions which have some close relation to the classical drama. Within the
pre-determined limitations he produced a scholarly treatise of permanent value,but as critics pointed out a t the time of its publication,
the terms tragedy and comedy were not applied solely to survivals and
imitations of the classic drama. Cloetta's work was expanded and in
some points improved by Bahlmann.3
The compositions which Cloetta examined, although they do not
always conform to his pre-determined criteria, always find justification
for their being known as comedies or tragedies in the mediaeval theory
as it has been outlined here. The terms were extended to many kinds
of narrative, as we have already seen, and if Cloetta had attempted to
analyze every one to see how far it conformed to the mediaeval rather
than his particular and narrower basis, the task would have been enormous and of no more than corroborative value. The Middle Ages were
not willfully eccentric but no one will deny that in the course of time a
great deal of classical learning was forgotten just as certainly as much
classical literature was lost. I t is, therefore, suificient for the present
purpose to concede the value of Cloetta's work within its limits, but to
point out that the Middle Ages inherited an authentic and legitimate
conception of classical origin and that, with very elastic inclusiveness,
they applied the theory they had received from classical antiquity.
Cloetta, Beifruge, I, pp. 1-2.
Cloetta himself in vol I1 corrects his data on the knowledge of Seneca in the
Middle Ages, as Traube pointed out in the Krit Jahresber. 11. d. Fortschr d. Rom.
Phil (1890), I , p. 90. Manitius has written many articles on mediaeval knowledge
of the classics, including the dramatic poets
P. Bahlmann, Die Ernezlerer des Antiken Cramas zbnd ihre Erstet~ Dramatischen Versuche, 1314-1478 (ZLIunster, 1896); Die Laltinischen Dra?)ienvotl I.lrimphelings Stjfpho bis ZZLT -1Iitte des Sechsehnfen Jahhz~nderts,1480-1550 (Munster,
1893).
1
A. Philip M c M a h o n
W h a t dejinitions did Dante accept as conveying proper ideas of tragedy

and comedy?
There are several places in the Divine Comedy in which Dante's
theory of tragedy and comedy is indicated. For comedy:
e pcr le note
Di questa Cornmedia, lcttor, ti giuro,
S'elle non sien di lunga grazia vote. . . .
Inferno, XVI, 127-129.
Cosi di ponte in ponte, altro parlando

Che la mia commedia cantar non cura,
Venimmo. .
Ircferno, XXI, 1-3.
. .
In these passages he uses the word with reference to his own poem.
Virgil, however, refers to his own work as a tragedy:
e cosi il canta
L'dta mia tragedla in alcun loco.
Inferno, xx, 112-113.
Dante's basis was almost certainly the Magnae Derivationes of Uguccione da Pisa,' gathered in turn from the Elementarium Doctrinae Rudimentum of Papias and the Etymologies of Isidore. Uguccione died in
1210, and although manuscripts of his work are frequent, and Du
Cange drew upon him; it was never printed, probably because it was
superseded by the Catlzolicon.3 Of the derivations and definitions given
by Dante in his letter to Can Grande, Toynbee says, they are ('taken
Uguccione seems
directly from Uguccione . . . under the word oda."
also to be the source of what Greek Dante knew.
1 Toynbee, op. cit., p. 97. (Reprinted with additions, from Romania, XXVI,
P P 537-534.)
Cf. Du Canee.
(Paris, 1846),
- . Glossariunz Mediae et Infimae Latinitatis, vol. \'I
p. 633, s.v. Tragoedia, " Irouia, laus facta de ztlibus et fetidis, uttde Tragoedisare,
dictare, in Gloss. Bibl. MSS. anonymi ex Ugutione in Bibl. reg."
3 Toynbee, op. cit., pp. 98-99.
Toynbee, op. cit., pp. 103 ff. Uguccione under the word oda says: "Oda, quod
est cantus vel laus, componitur cum comos, quod est villa, et dicitur hec comedin,
- e, idest villanus cantus, vel villana laus, quia tractat de rebus villanis rusticanis,
141
The Epistle to C a n Grande della Scala is now accepted by nearly all

scholars,l and in it Dante accounts for the title of his poem, defining
tragedy and comedy, while noting the difference between them, in the
usual, authentic manner.2
et aifinis est cotidiane locutioni, vel quia circa villas fiebat et recitabatur, vel
comedia a commensatione, solebant enim post cibum homines ad audiendum eam
venire. . . . Item oda in eodem sensu componitur cum tragos quod est hircus, et
dicitur hec tragedia, - e, idest hircina laus, vel hircinus cantus, idest fetidus; est
enim de crudelissimis rebus, sicut qui patrem vel matrem interficit, et commedit
filium, vel e contrario et hujusmodi. Unde et tragedo dabatur hircus, idest animal
fetidum, non quod non haberet aliud dignum premium, sed ad fetorem materie
designandum. . . . E t differunt tragedia et comedia, quia comedia privatorum
hominum continet acta, tragedia regum e t magnatum. Item co?nedia humili stilo
scribitur, tragedia alto. item comedia a tristibus incipit, sed in letis definit, tragedia
e contrario; unde in salutacionibus solemus mittere et optare amicis tragicum
principium et comicum finem, idest principium bonum et letum, et bonum et letum
finem."
Cf. E. K. Rand, T h e L a t i n Concordance of Dante and the Genuineness of Certain
of his L a t i n W o r k s (Boston, 1912)~pp. 7-39: Paget Toynbee, Dantis Alagherii
Epistolae (Oxford, rgzo), pp. 160 ff.
Paget Toynbee, "Dante's Letter to Can Grande (Epist. X.), Emended Text,"
Modern Language Kevitxw, XIV ( ~ g r g )p.
, 286: "Libri titulus est: 'Incipit Comoedia Dantis .4lagherii, Florentini natione, non moribus.' Ad cuius notitiam sciendum est, quod comoedia dicitur a comos villa, et oda quod est cantus, unde comoedia
quasi villanus cantus. E t est comoedia genus quoddam poeticae narrationis, ab
omnibus aliis differens. Differt ergo a tragoedia in materia per hoc, quod tragoedia
in principio est admirabilis et quieta, in fine sive exitu est foetida et horribilis;
e t dicitur propter hoc a tragos quod est hircus, et oda, quasi cantus hircinus, id
est foetidus ad modum hirci, ut patet per Senecam in suis tragoediis. Comoedia
vero inchoat asperitatem alicuius rei, sed eius materia prospere terminatur, u t
patet per Terentium in suis comoediis. E t hinc consueverunt dictatores quidam
in suis salutationibus dicere loco salutis, ' tragicum principium, et comicum finem.'
Similiter differunt in mod0 loquendi: elate et sublime tragoedia; comoedia vero
remisse et humiliter; sicut vult Horatius in sua Poetria, ubi licentiat aliquando
comicos ut tragoedos loqui, et sic e converso:
Interdum tamen et vocem comoedia tollit,
Iratusque Chremes tumido delitigat ore;
E t tragicus plerumque dolet sermone pedestri
Telephus et Peleus etc.
E t per hoc patet quod Comoedia dicitur praesens opus. Nam si ad materiam
respiciamus, a principio horribilis et foetida est, quia Infernus; in fine prospera,
A. Philip McMahon
References to the style of tragedy and comedy are to be found in the
Convivio and in the De Vulgari Eloquentia? and they are in accordance
with the letter to Can Grande.
Since commenting on Dante was formerly one of the necessary occupations of most Italian scholars, and the reading of Dante is still an
essential part of Italian education, his use of the words was constantly
reconsidered and perpetuated itself despite other influences3 Scaliger
was to disturb many nineteenth-century critics and scholars by his
obstinate refusal to follow the Poetics blindly.
Not all of the commentators discuss the matter fully, but many of
them do, and a few will be cited here as illustrations. Jacopo della
Lana: for examples, states the differences in accordance with the same
authorities from whom Dante derived. Pietro, the son of Dante, explains the title of the work in accordance with the traditional theory,"
and besides citing Isidore and Horace, he quotes Boethius, who may
perhaps have also been remembered by Dante in this c~nnection.~
Francesco da Buti discusses the question with regard to Inferno XX,
113, remarking that Virgil calls his work a tragedy on account of the
style, the persons, and the progress made from happiness to disaster?
Giovanni Boccaccio treats the matter fully, including such questions as
etymology, style of language, and verisimilitude. He finally decides
that the poet spoke figuratively, but compares the progress from turbudesiderabilis et grata, quia Parndisus. Ad modum loquendi, remissus est modus et
humilis, quia locutio vulgaris, in qua et mulierculae communicant. E t sic patet
quare comoedia dicitur."
1 I.:. hfcore, Tzifte le Opere di D. A. (Oxford, rgoq), 11. C o l r i . I , cap. 5, 50
(P. 242).
De Vzclgari Eloq., 11,iv, 4 ff.
3 G. J. Ferrazzi, Elzciclopedia Gafltesrcc (5 vols. Bassano, 1865-1877), 11, pp.
431 fi.
4 L. Scarabelli, Conzedia di D. degli A. col comfncntodi Jaropo dclla Lana (3 vols.,
Bologna, 1866), I, p. 351.
5 V. Nannucci, Pelri Allcglrer.ii strpcr D a ~ t i sipsius Gelzitoris Cowzocdiam Commentarium (Florence, 1845-18~6) pp. 9 if.
6 G. A. L. Baur, Boetitts uttd Dante (Leipzig, 1873); R. R'lurari, Danle e Boaio
(Bologna, 1905).
7 C. Giannini, Commefzto di Francesco da Bzcti sopra la D. C. (Pisa, 1858), I,
pp. 7 and 531 fl.
143
lence to tranquillity in the Commedia, with the comedies of Plautus

and Terence.l He refers back to his introduction when he comes to
I?!fenzoXVI, I 28.2
The work of Benvenuto da Imola adds something from the study of
formal rhetoric and scholastic philosophy, while citing I ~ i d o r e . He
~
1 G. Milanesi, I1 Conzento di Gionanni Bocraccio sopra la C. ( 2 vols., Florence,
1863)~I , p. 84. Domenico Guerri, Giovanni Boccaccio, I1 Comefzto ella D. C (3 vols.,
Bari, 1g18), I, p. 116: "I1 tutto della commedia 5 (per quello che per I'lauto e per
Terenzio, che furono poeti comici, si pub comprendere): che la commedia abbia
turbolento principio e pieno di romori e di discordie, e poi l'ultima parte di quella
finisca in pace e in tranquillitb. A1 qua1 tutto 5 ottimamente conforme il libro
presente: percioche egli incomincia da' dolori e dalle turbazioni infernali, e finisce
nel riposo e nella pace e nella gloria, la quale hanno i beati in vita eterna."
Milanesi, op. cit., 11,p. 453. Guerri, op. cit., 111,p. 2 2 9 .
Cf. Luigi Rossi-Cad, Di Maestro Benvenuto da Imola (Pergola, 1889).
J . P. I'acaita, Benvenuti de Rambaldis de Inzola, Comenlztm szcper Dantis Aldigherij Comoediam, nunc primzrm integre i n lucem editum (Florence, 1887), I,
p. 18: "Tragoedia est stylus altus et superbus; tractat enim de memorabilibus
et,horrendis gestis, qualia sunt mutationes regnorum, eversiones urbium, conflictus
bellorum, interitus regum, strages et caedes virorum, et aliae maximae clades; et
talia describentes vocati sunt tragoedi, sive tragici, sicut Homerus, Virgilius, Euripides, Statius, Simonides, Ennius, et alii plures. . . . Comoedia est stylus bassus
et humilis, tractat enim vulgaria et vilia facta ruralium, plebeiorum, et humilium
personarum; e t talia describentes vocantur comoedi, sive comici, sicut Plautus,
Terentius, Ovidius. hlodo est hie attente notandum quod, sicut in isto libro est
omnis pars philosophiae, u t dictum est, ita est omnis pars poetriae. Unde si quis
velit subtiliter investigare, hie est tragoedia, satyra, et comoedia. Tragoedia
quidem, quia describit gesta Pontificum, Principum, Regum, Baronum, et aliorum
magnatum et nobilium, sicut patet in toto libro. Satyra, idest reprehensoria;
reprehendit enim mirabiliter et audacter omnia genera viciorum nee parcit dignitati, potestati, vel nobilitati alicujus. Ideo convenientius posset intitulare satyra,
quam tragoedia, vel comoedia. Potest etiam dici quod sit comoedia, nam secundum Isidorurn Comoedia incipit a tristibus et terminatur ad laeta. E t ita liber iste
incipit a tristi materia, scilicet ab Inferno et terminatur ad laetam, scilicet ad
Paradisum, sive ad divinam essentiam. Sed dices forsan, lector: cur vis mihi
baptizare librum de novo, cum autor nominaverit ipsum Comoediam? Dico quod
autor potius voluit vocare librum Comoediam a stylo infimo et vulgari, quia de rei
veritate est humilis respectu litteralis, quamvis in genere suo sit sublimis et excellens. Deinde tangitur in titulo causa efficiens, cum dicitur Dantis Aldigherii; et
materia, cum dicitur, in qua agitur de Inferno, sive causa materialis et subjectum
libri primi."
I44
A . Philip McMahon
distinguishes between the species on grounds of style,' repeating the

point on Inferno XX, 113,2 and Iftferno XXI,
The writer usually
known as the h o n i m o Fiorentino, deals with Iqferno XX, 113, in the
customary manner, but also employs material from the letter to Can
Grande.4 Christophoro Landino explains Virgil's references on the
basis of style and the nature of his themes5 Velutello mentions only
the style as a reason for calling the Aeneid a t r a g e d ~ .Tasso
~
later observes that the Aeneid is in his time usually considered an epic, but he
justifies Dante on the basis of the remarks in De Vulgar; Eloquentia?
It was not until the middle of the sixteenth century that any difference between Dante and the Poetics troubled the critics. Gelli, in his
lectu~eson the Cornmedia says, in defending llante, that his error was
not of much importance in the poet's time, and he quotes the letter to
Can Grande. He also declares that if Aristotle had read Dante as he
had read Homer, he would have framed his theory to include the Italian
poet. I n similar fashion he compares the work of M.ichelangelo and
that of the ancients, concluding that Michelangelo's achievement is in
no way inferior to that of antiquity. Although Dante did not construct
his poem on the basis of rules, his performance did not fall below that
of the dramatic poets of classical t i r n e ~ .Another
~
rabid defender of
Dante in the controversy was Giacopo Mazzoni, who, however, in 1572
made an attempt to reconcile Dante and the Poetics without introducIbid., I, p. I 54: "Homerus, Virgilius, et Lucanus scripserint in alto stilo,
scilicet tragedia, tamen Horatius scripsit in mediocri stilo, puta satira, et Ovidius
in basso, scilicet comedia."
2 Ibid., 11, p. 88: " e l'alta mia tragedia, quia est stylus altus, et de rebus altis
tractans, canta, idest poetice describit, i n alcun loco, scilicet tertio Eneidos."
3 Ibid., 11, p. 9 j: "che la mia comedia, idest meus liber vulgaris, cantar non cura,
idest poetice describere. . . ."
4 P. Fanfani, Conzmento d l a D. C. d'dnonimo Fiorentino, I (Bologna, 1866),
pp. 451 and 461.
Comento di Christophoro Landilto sopra la Com. di D. A . (Opere d d Divino
Poela Danthe, Bibliotheca S. Bernardini) Venice, I 512, f . 131 r.
0 La Comedia di Dante Aligieri con la Nova Espositione di A. velvtello (Venice,
I 544), in loc.
7 La D. C. di D.A . Postillata da T . Tasso, 3 vols., I (Pisa, 1 8 ~ 0 p.
) ~ 153.
8 C. Negroni, Cioaan Batista Gelli sopra la C. d i D. (Florence, 1887), Lettura
Ottava, 11,p. 296-297.
Aristotelian Definitions of Tragedy a n d Comedy
145
ing considerations of historical development.' The critic Daniello da

Lucca, on Inferno xx, 113, mentions reasons of style and the death of
T u r n u ~ .It
~ is interesting to see the comments and solutions of other
writers: but those given will suffice to show the persistence of Dante's
use and its acceptance.
W h a t definitions did Cha.ucer accept as conveying proper ideas of

tragedy and comedy?
Chaucer's conception of tragedy was the usual one among scholars of
his and preceding times14and is clearly stated in his translation of Boethius: "'What other thing biwailen the crqinges of tragedies but only
the dedes of Fortune, that with an unwar stroke overtorneth the
realmes of grete nobley?';
Mario Rossi, Discorso di Giacopo Mazzoni in Difesa della Conzmedia del Divino
Poeta Dante (Citti di Castello, 1898), p. 74.
Dante con 1'Espositione di M . Bernardino Daniello da Lucca (Venice, 1568),
P 134.
3 K. Andreoli, I,a C. d i D. A. (Naples, 1863), p. 131; L. G. Blanc, Gottliche
Komodie (Halle, 1864), on Inf., XX, 113; G. da Siena, Com. d i D. A. (Naples, 186770), on Inf , XX, 113; K. Vossler, Die Gdttliche Komodie, 4 vols. in 2 (Heidelberg,
1908), 111,306 ( S ~ I )and
, IV, 57 (963). P H. Wicksteed and E. G. Gardner, Dante
and Giouanni del Virgilio, (Westminster, 1902), p. 293.
Cf. G. L Kittredge, Chaucer and his Poetry (Cambridge, 1915), p. IIO.
Skeat, op. cit., p. 143 (Prose 11, 1. 75). Cf. B. L. Jefferson, Chaucer and the
Consolation of Philosophy of Boelhizhs (Princeton, 1917), p. 165: "How much the
Consolation determined Chaucer's own attitude toward life, it is difficult to determine with precision At the least, it may be said that Boethius and Chaucer
were compatible in point of view and that Chaucer found in Boethius, in many
ways, a congenial spirit. At the most, it may be said that Boethius was an influence
so profound that he completely determined Chaucer's view of the meaning of life
and of the way in which life should be conducted. The truth no doubt lies somewhere between the two extremes, and Boethius probably accentuated and extended
views which Chaucer already had temperamentally."
The influence of Chaucer on all students of English made even Archer, PlayMaking (New York, 1928), p. 261, seek to reconcile Chaucer and the Poetics. He
notes that Chaucer's lines show that peripeteia or reversal of fortune was the "very
essence and meaning of tragedy" in the Middle Ages.
A. Philip McMahon
To this is added: " Glose. Tragedie is to seyn, a ditee of a prosperitee
for a tyme, that endeth in wrecchednesse."
Ten Brink quoted Dante's letter to Can Grande and thought that
Chaucer had thence derived his t h e ~ r y .But
~ since Miss Petersen's investigation,%weh o w that theory is wrong, as well as others which supposed the source of the gloss was the French prose translation ascribed
to Jean de Meung or the Latin commentary on Boethius formerly attributed to St. Thomas Aquinas. illiss Petersen established the fact that
Chaucer's source was the unpublished commentary by Nicolas Trivet,
which included those of Jean de Meung and the pseudo-Aq~inas.~
Trivet's theory of tragedy in his edition, with commentary, of Seneca's tragedies,%arly in the fourteenth century, was based largely on
I s i ~ l o r e .His
~ long note cn the passage in Roethius was kindly sent me
by Miss Petersen from unpublished material?
Chaucer's agreement with that theory is indicated in various passages of his own poems. For example, in the T r o i l z ~ sand Criseyde: s
Go, litel book, go litel myn tragedie,
Thcr god thy maker yet, er that he dye,
So sende might to make in some comedie!
Rut litel book, no making thou n'cnvyc,
But subgit be to alle poesye;
And kis the steppes, wher-as thou sccst pace,
Virgile, Ovyde, Omer, L u c q and Stace.
Skeat, op. cit. (Prose 11, 1. 78).

B. Ten Brink, Clzclucer (Xliinster, 1870)~1, p. 78.
X. 0. Petersen, "Chaucer and Trivet," Pzibl. Moder?t Lnng. Assoc., X V I I I
(1903), PP- 173-1934 The sources which Miss Petersen used were Addit. RlSS. 1958j and 27875,
British RIuseum.
5 Given in full detail in Peiper, De Se?zerae Tragoediarzhm Vulgnri Lertiolze
(Breslau, 1893).
Creizenach, o p . i i l . T, p 487 ff.
7 "Secundo cum dicit Quid tragediarum probat mutabilitatem fortune divulgare
cotidianis clamoribus cluia clamores poetarum cotidic in theatro recitancium tragedias nichil aliud continebant quam mutabilitatem fortune. ICt nota quod tragedi
dicuntur secundum Ysidorum, T4:thi. li. 18c. de Ludo Scenico: illi qui antique gesta
atque facinora scelerstorum regum luctuoso carmine spectante populo concinebant.
Unde tragedia est carmen de magnis iniquitatibus a prosperitate incipicns et in
adversitate terminans. E t dicitur tragedia a tragos quod est ycrus. . . ."
0 Skeat, op. i i l . , p. 324 (V, 1786-1792).
1
147
the envoy corresponds to his initial purpose:

The double sorwe of Troilus to tellen,
That was the king Priamus sone of Troye,
In lovinge, how his aventures fellen
Fro wo to wele, and after out of joye,
My purpos is, er that I parte fro ye.
Thesiphone, thou help me for t'endyte
Thise woful vcrs, that wepen as I wryte!
I n the Mon,k's Prologue again, we find:

Tragedie is to seyn a certeyn storie,
As olde bokes maken us memorie,
Of him that stood in greet prosperitee
And is y-fallen out of heigh degree
Into miserie, and endeth wrccchedly.
-4nd they ben versifyed comunly
Of six feet, which men clepe exanzelron.
In prose eek been endyted many oon,
And eek in metre, in many a sondry wyse.
Although Skeat supposed that the gloss on the passage in Chaucer7s

translation of Boethius was original with Chaucer himself: he noted
correctly that the authors mentioned a t the end of the first passage
from Troilus quoted above were models which, in accordance with
mediaeval usage, prompted him to term works in heroic hexameters
tragedies. This superficial confusion is really testimony to the persistence and weight of the definitions derived from the dialogue On Poets,
because of the similarity of characters and themes in epic and tragedy,
and the term heroic was used both with regard to the metre of epic and
also in the traditional f ~ r m u l a . ~
Ibid., p. 206 (1, L-7).
Ibid , p. j p (T. 3163-3171).
ktT. W. Skeat, Chaqlcet, The Prior~ssesTale, Sire Thopas, et cet. (4th ed., OXford, 1888), p. 173, on line 3163, ar,d p. 193.
Several important topics have necessarily been omitted in this whole discussion, because of lack of space. One of them is the history of the terms used to
designate the metres of tragedy and comedy. Another is the relation of the conceptions of fortune, chance, and luck to tragedy. The latter is treated from the
mediaeval point of view in 13.12. Patch, The Goddess Fovkna in Mediaeval Literature
(Cambridge, 1927), pp. 68 ff.
148
A. Philip McMahon
The Monkes Tale, a t the very beginning, recurs to Chaucer's conception of tragedy:
I wol biwayle in maner of Tragedie
The harm of hem that stode in heigh degree,
And fille11so that ther nas no remetlie
To bringe him out of hir adversitee.
The miseries of Sampson arouse a n echo of the definition which

Chaucer a c ~ e p t e das
, ~well as the fate of Ugolino,3 and in the story of
Croesus,4 the gloss on Boethius is versified:
Tragedie is noon other maner thing,
Ne can in singing crye ne biwaille,
But for that fortune alwcy wol assaille
\Vith unwar strook the regnes that ben proude.
The effect of Chaucer's views is t o be seen among his contemporaries

and successors, particularly in Lydgate, who repeats the idea frequently,%nd in one place also puts the gloss into verse.6 The theory is
found frequently in his Troy Book; he explains comedy as well as
tragedy:
A comedie hath in his gynnyng,

A prime face, a maner compleynyng,
And afterward endeth in gladness.
But tragidie, who so list to knowe,
I t begynneth in prosperite,
.4nd endeth euer in aduersite;
And it also doth be conquest trete
Of riche kynges and of lordys grete.
Skeat, Student's Chaurer, p. 531 (T. 3181-84).
Ibid., p. 532 (T. 3267): "Sith thou fro wele art falle in wrecchednesse."
3 Ibid., p. 537 (T. 3647-3648).
Ibid., p. 541 (T. 3951-3954)5 J. 0. Halliwell, il Sdection frollz the M i n o r Puems uf Dan J o h n Lydgnle (Percy
Society, No. z, London, 1840), pp. 25 and 128.
6 H. N. RlacCracken, T h e M i n o r P o a ~ n sof J o h n Lydgate (IC. E. T . S.), I, 1911
(for 19lo), P. 73.
7 H. Bcrgen, Lydgate's T r o y Book, A.R. 1412-30 (I?. E. T. S.), I (London, 1906),
pp. 168-169.
1
Aristotelian Defcnitions of Tragedy and Conzedy
149
The same idea is found in his Fall of Prilzces,' and also in Henryson's
Testament of C r e ~ s e i d . ~
Throughout the sixteenth century Lydgate's translation of Boccaccio7sDe Casibus Virorum et Femi~tarumIllustrium was popular, and
thus served to illustrate the persistence of the mediaeval ~ignificance.~
I n 1554 we also find Hawes speaking of Chaucer's Legend o j Good
Women as consisting of tragedies4
Thus, in both Dante and Chaucer the Peripatetic inheritance survived in extended form, although they are the mediaeval authors most
frequently patronized by recent critic^.^ Such critics fail to consider
several obvious points: (I) There is a clear line of descent for this
C. F. E. Spurgeon, Five Hundred Years of Chaucer Criticism and Allz~sion
(London, 19141, 1, P. 43.
G. Gregory Smith, The Poems of Robert Henryson (Edinburgh and London,

1908), 111, p. 3.
Cf. the title of a copy in the Library of Harvard College: "A Treatise Excellent and Compendious, Shewing and Declaring, in Manner of Tragedye, the Falles
of Sondry Most Notable Princes and Princesses with Other Nobles, Through ye
hlutabilitie and Change of Unsteadfast Fortune Together with Their Most Detestable and Wicked Vices. First Compyled in Latin by the Excellent Clerke
Bocatius, an Italian Borne. And Sence that Tyme Translated into our English
and Vulgare Tongue, by Dan John Lidgate Monk of Burye, etc. In Aedibus Richard
Tottelli, London, 1554."
Stephen Hawes, Pastime oJ I'lensure. (In C. F. E. Spurgeon, Five Hundred
Years of Chnz~cerCriticism and Allusion, E . E. T. S., I (1914), p. 67.):
And then the tragedies, so piteous
Of the ninetene ladyes, was his translation.
The extremes to which some writers go to reconcile the Middle Ages and the
Poetics are exemplified in the following. K. Vossler, Poetische Theorien in der
italienischcn Fruhrenaissnnce (Berlin, lgoo), p. 26: "Es versteht sich van selbst,
dass der Verfasser der Eccerinis die TragGdie als Dichtungsgattung und nicht
mehr als Stilart auflasste, wie das game hIittelalter und Dante noch gethan hatten.
Ja er muss sogar auf irgend welchen Wege schon die tragischen Theorien des
Aristoteles kennen gelernt haben, wie wir aus den folgenden Versen hervorzugehen
scheint :
Vox Tragici mentes ad contingentia fortes
Eficit, ignavus diluiturque metus,
Vincit in adversis semper constantia rebus,
Nan habet hanc illis qui rude pectus habet
Tunc cum victor eris, vinci potuisse putabis
Constringit mavens anxia corda timor.
150
A . Philip McMahon
theory back to the earliest Peripatetics. ( 2 ) I t was a theory which

abundantly satisfied Greek and Roman antiquity. (3) I t was held by
people who were perfectly familiar with classic drama. (4) I t was
found more satisfactory by people who knew both the classical drama
and the Poetics of Aristotle. (5) I t survived and references to it continued both while there was a wide-spread knowledge of classic drama
and afterward. (6) I t has also seemed, until the nineteenth century, to
state more concisely than the definition in the Poetics the tragic essence
of tragedy and the comic essence of comedy. (7) I t more adequately
corresponds to the metaphorical use of the words tragedy and comedy.
(8) The metaphorical use and the persistence of the formulas preserved
by the Roman grammarians are sufficient reasons for the practice of
Dante and Chaucer. The Italian and the English mediaeval poets can
hardly be held to account because they lacked a knowledge of classical
drama and modern classical philology on the one hand, or, on the
other, a prophetic vision of what would happen when transcendental
philosophy and classical philology devoted themselves to a combined
attack on the K C ~ ~ U ~clause
U L S of Artistotle's Poetics.'
VI
W h a t dejinitions of tragedy and comedy dominated the ideas oj Corttinental Europe dzcring and after the Renaissance?
I t would have been a unique event in the history of thought if the
general classical and mediaeval theory of tragedy and comedy had been
immediately abandoned on the recovery of Aristotle's Poetics, but such
an unparalleled development has been postulated by many historians
of l i t e r a t ~ r e .I~t is part of the exaggerated contrast between the Mid1 As Schelling, Hegel, Schlegel, and Schopenhauer determined the German interpretation of the Poetics throughout most of the nineteenth century, one finds
Croce dominating the Italians. Cf. liostagni, "Aristotele e I'Aristotelismo nella
Storia dell' Estetica Antica," S t z ~ d iIlaliil?zi d i 1;ilologiu Clussicu, N . s., 11 (rgzz),
pp. 83-84. I t is hardly necessary to point out the injury to historical criticism suffered from this cause.
Giuseppe Tofianin, La Fina dell' Cr~ancsinzo,T urin, 1c)zo, however, points
out that the systematic study of t l ~ ePoetics, codification of the rules of the drama,
and treatment of it as an inrpircd and uniquely authoritative document develop
conspicuously as a result of forccs active until the Council of Trent; forces wliich
after i t were directed to the study of poetry rather than theo1o:y.
151
dle Ages and the Renaissance. J. A. Symonds, to exalt the Renaissance, must correspondingly depress the significance of the Middle
Ages. Of the revival of learning, for example, he says: "For a generation nursed in decadent scholasticism and stereotyped theological
formulas it was the fountain of renascent youth, beauty and freedom,
the shape in which the Helen of art and poetry appeared to the ravished
eyes of mediaeval Faustus." But the break with the Middle Ages was
by no means complete or sudden, and the theory of tragedy and
comedy in the Renaissance and after was far from being an entirely
original or independent study of an inspired document, newly revealed.
The divergence of Renaissance theory and practice from the Poetics,
especially with regard to tragedy, has puzzled many historians, and
many explanations have been offered, including ignorance, perversity
of spirit, sheer eccentricity, and, more credibly, the aim of reconciling
Horace and Aristotle, or the slavish imitation of S e n e ~ a .If~ the mediaeval conception had been correct and authentic but too inclusive,
then the ideas of the Renaissance tended to err in being too restrictive
and legalistic.
Scholars were never entirely without sound theory of the drama,
even when they lacked the Poetics, and their renewed knowledge of
classical drama and classical grammar yielded, to their minds, confirmation of the traditional theory. The scholastic principles of analysis and classification grounded on Aristotelian logic and metaphysics
found a new outlet for scholars in the Poetics, a document obviously
exemplifying the same characteristics. Whereas theology and philosophy, had, for the time being, been thoroughly worked by that method,
the Poetics provided the generations immediately after the Council of
Trent with a new and fresh field, where there was no danger to the
faith and a splendid model was ~ f f e r e d . ~
Encyclopaedia Britannica (11th ed., I ~ I I ) ,XXIII, p. 86.
C. IF. Conrad Wright, History of French Literature, New York, 1912, p. 2 1 1 .
3 Cf. Toffanin, op. cit., pp. 1-2: "Nell' anno 1548, mentre a Koma si lavorava,
fra illusioni e dubbi, a preparare e a procrastinare quel Concilio da cui doveva
uscire cosi mutato e contrastante I'aspetto spirituale dlEuropa, un giovine udinese,
sceso a Firenze con molta dottrina umanistica, oflriva a Cosimo de nIedeci il
primo comment0 all' 'hrte poetics' di hristotele. L'animo del Robortelli non
sospettava - ne siam certi - che quella sua ambizione erudita, iniziava, per la
letteratura, l'eti del Concilio di Trento: ma il genio della storia aveva stabilito
2
The history of the editions and study of the Poetics has been outlined
by Sandys,' following Spingarn12and there is no need to repeat it here.3
The traditional theory of tragedy and comedy was, however, present in
the minds of most of the Renaissance scholars,4 and Lanson has shown
how reluctantly the traditional formulas were abandoned, if a t all.5
The classical authors read during the Middle Ages were not surrendered in exchange for the newly recovered texts, including the
Poetics. Convincing evidence of the persistence of the traditional formulas is offered by a study of the early printed editions. The pressure
of need for widespread approval is felt by the publisher of a printed
book even more keenly than by the editor and scribes of a manuscript.
When, therefore, the earlier editions of classical texts are accompanied
by the traditional apparatus, it is certain that this was no blind following of precedent but the response to an imperative demand.
cosi perch*, d a quel giorno - proprio da quel giorno - 1' ' Arte poetica' diventa il
canovaccio su cui una gente, preoccupata e offuscata da grandi pensieri e da meschini pregiudizi, tesse le tramc d'una scolastica letteraria e si prepara due secoli di
decadenza che si chiameri prima secentesimo e poi Arcadia c a v r i fine solo col
romanticismo."
1 Sandys, op. cit., 11,pp. 133 fl.
J. E. Spingarn, IIistory of Literary Criticism in the Renaissance (3d ed., New
York, 1~12). The Italian translation contains valuable additional material: J. E.
Spingarn, L a Critica IJelteruria ncl Ri~tuscinle~tto.T raduzione italiana dcl Dr.
Antonio Fusco (Ilari, 1905).
"413 indispensable aid in any study of the I'oetirs is: Lane Cooper and Alfred
the I'ortics of .:iristotle (Cornell Studies in English,
Gudeman, A Bibliogruphy
XI, New Haven, 1928).
4 Cf. Custave Lanson, "L'IdCe de la TragCdie en France avant Jodelle," Ren~e
dlIIistoire Litternire dc la France, X I (rqoq), p. 585, for Scaligcr's acceptance of
the traditional definitions: "On y remarquera surtout quc, chez Scaliger, la dkfinition de la tragCdie, la notion du sujet tragique restent traditionelles. I1 retient
toute l'idce que Cloetta a constatCe dans les glossaires et les sommes du moyen
jge," ctc.
5 Ibid., pp. 541: "Un des plus ficheus cffets de la distribution traditionelle
du travail entre les Crudits appliquCs au moyen 3ge et les critiques qui ne voient
rien au deli de la Renaissance, a CtC de faire croire 2 une coupure rCelle et prCcise
dans le dCveloppemcnt litt6raire de notre pays. Cettc croyance a entrain6 diverses
erreurs dans la representation du mouvement intellectuel du XVIC sii.cle. On a
commenc6 de nos jours i les percevoir. J'cn voudrais signaler et, si je puis, dCtruire
encore une."
153
Boethius was, for example, an author read as widely during the

Renaissance as during the Middle Ages, and constantly reprinted.
The note on the passage referring to tragedy was merely the mediaeval
gloss repeated or expanded.' The same explanation is found in editions
such as that printed a t Basel in 1570,~and the note in the Delphin edition a t the end of the seventeenth ~ e n t u r yneglects
,~
any reference to
the Poetics but clings to Horace and the grammarians. This was the
most popular and accepted text until the German editions of the nineteenth century, but the Delphin edition was reprinted in London in
1823: and it was reproduced verbatim in Migne's e d i t i ~ n . ~
Terence, in particular, was reprinted with the same apparatus that

had been used in the latter days of the Roman Empire and through the
Middle Ages. Nearly all the printed editions, until recent times, carried the treatise of Donatus as well as his life of Terence.= The great
popularity of these comedies in the Renaissance and satisfaction with
the explanation of the grammarian is one of the best attested facts in
the history of literature.' In cases where Renaissance scholars provided
1 Severini Boethii de Consolatione Philosophiae Iihri Quinque. Czcnz Praeclaris
Joannis Murmellii Comnzentwiis, cnmgz~eRodolphi Agricolae Phrisii et Augustini
Enarratione (2d ed , n. p. ISIS), fol. xxxii, v., quotes Diomedes and repeats the
anecdote about Euripides and Archelaus.
Plzilosophorum et Theologornnz Principis Opera Omnia (Joannis Ilfz~rmellii
i n V . Lib. De Consol. Phil. Conz ), (Basel, 1570).
Boetii De Consolatione, et cet. Znterpretatione et iVotis Zllzcstravit Petrzcs Callyus.
I n Usum Screw. Delphini (Paris, 1695).
Boethii De Consolatione Phil. Lih. V . E x Ed. Vdpiana cum Xotis et Znterpuetatione i n LTsum Delphini (London, 1823).
Migne, Patr. Lat., LXIII, col. 667 ff.
Cf. the following editions in the Library of Harvard College: Terentii Comoediae czcm Donati Interpretis Commentario (Treviso, 1477; Terentius cnm Donato
(Venice, 1492); Terentiz~sC U ~ ZDuobus Commentis L7idelicet Donato et Guidone
(Venice, 1494).
H . W. Lawton, Terence en France azc XVZe Sidcle (Contribution a I'Histoire de
I'Humanisnze en France), Paris, 1926, follows out the ideas of Lanson. This study
lists all the editions, so far as the author could discover, printed during the fifteenth
A. Philip McMa hon

new discussions of tragedy or comedy, they relied on the standard
sources on which the Middle Ages had also depended.' The popularity
of such equipment for the intelligent reading of Terence is shown by
many other editions during the following centuries, of which that puhlished a t Amsterdam and Leyden will serve as an e ~ a m p l e . ~
Seneca, who, with Terence, provided models most acceptable to early

Humanistic drama, appeared with commentaries relying in greater or
less degree on the traditional theory. An edition of 1498, for example,
had two scholars to explain it, to whom a third, Badius, mas added in
1514.~
B ut as Lanson points out, while Aristotle is mentioned as an
authority for dividing tragedy into six parts, the discussion is actually
based mainly on Donatus and D i ~ m e d e s . ~The same writer shows
that the tragedies of Seneca conformed to the definitions of Diomedes,
and, interpreted in this light, they provided the most influential model
for the Renaissance."
and sixteenth centuries. He mentions 446, as well as 59 editions of separate plays. I n
those editions which Lawton himself examined a full statement of the contents is
given. To judge from the information afforded by Lawton, we find most editions
reprint the traditional treatises of Donatus or material derived from him.
1 Lanson, op. cit., pp. 544: "Joducus Uadius a donc ou conservi ou retrouv6
la doctrine du moyen bge. Sujets historiques, ca1amitC.sroyales et bouleversements
des Btats, chute du bonheur au malheur et d6nouements funestes, style majestueux
et abondance de rhktorique esclamative ou plaintive; c'est exactement la th6orie
que 1%'. Cloetta a constatie au moyen age."
V ~ u b l i Terentii
i
Carthaginiensis rifri Comoediae VZ. IIis Accedunt Integrae
Notae Donati, Eugraphii, Faerni, Boecleri, Farnabii, Mer. Casauboni, Tan. Fabri.
Amstelodalmi, et Lugd. Batav., 1686. Contains: (I) Danielis Ileinsii ad IIoralii
de Plarllo et Terentio Jndicizmm Dissertatio; ( 2 ) Euanthins de Trag. et Con?., ( 3 ) Donati Fraglnenlzmz tie Conloedia et Tragocdia; ( 4 ) De Fabzrlanlm, Lzrdorr~m,Theatvorunt,
Scentlnint, ac Scenicorz~nlafcliqua consuelztdine libellzis, ex optinzis azlctoribzis collectzis,
ud Comicos facilizis inlelligendos. Praecipzie conscriptzls i n gratiam stztdiosae juventtitis; (5) T . Eahri Observatiztncz(1ae Miscellaneae; (6) Tanaqztilli Fahri Arolzllae
ud Terentizim, et cet.
Tragoedia Senecae czt~ftDzlohzls Commenlariis: ITideliret Uernardini ~lfarr?zitae
et Danielis Gaetani (Venice, 1498).
Lanson, op. cit., p. 547.
Ibid., p. 546.
I 55
Even Horace, who was cited by the grammarians for his derivations,
was not left without scholarly comment, and in the French edition of
1529, four authorities united to explain the classic author, quoting the
pertinent passage in Diomedes on the ode dedicated to Pollio, while the
Ars Poetica was made clear by reference to the same ancient source?
The appearance of a Greek text did not, moreover, do away with

all Latin texts or translations, as many enthusiastic students of the
Renaissance imagine.2 The translations of Greek plays into Latin, such
as those of E r a s m q 3 fail to show any effect of the Poetics, but rely
wholly on the classical grammarian^.^ I t is probable, indeed, that such
plays as Hecuba enjoyed the greater popularity because of fulfilling the
definition of Diomedes."
Lanson c o n c l ~ d e s that
, ~ when the Humanistic drama appeared, it
was determined in its essentials by Donatus and Diornede~,~
supported
by Horace and V i t r ~ v i u s . ~
Ibid., pp. 572-573.
Ibid., p. 545.
Cf. Necuba, & Zphigenia i n Aulide Euripidis Tragoediae i n L a t i n z m Tralatae
Erasmo Roterda~noZnterprete, Venice (Aldus), 1507. Erasmus (f. 3 I., ibid.) took
Horace's ironic reference to the metre of the tragedy seriously and apologized for
the lack of such a quality in his own translation: "Iam ueroque 1,atinae Tragoediae
grandiloquentiam, ampullas, & sesquipedalia (ut Flaccus ait) uerba hic nusquam
audient, mihi non debent imputare, si interpretis officio fungens, eiusquem uerti,
pressam sanitatem, elegantiamque; referre malui, quam alienum tumorem, qui
me nec alias magnopere delectat."
' Lanson, p. 549. For Erasmus's neglect of Aristotle's Poetics, cf. p. 545.
Ibid., pp. 549-5 jo.
6 Zhid., p. 580: "Ainsi i la veille du jour oh Jodelle Ccrivait, la notion de la
tragCdie Ctait, pour I'essentiel, composCe dans les esprits f r a n ~ a i par
s les definitions
et les rimarques de Donat et Diom*de, aux-quelles s'ajoutait Horace, et un teste
de Vitruve."
On the significance of Vitruvius for Renaissance theory and criticism, cf. Lan(March-April, ~ g o q ) pp.
, 72-84.
son, Revlie de la Re~zaissa~zce
The usual explanation of the situation suffers from a misunderstanding with
regard to the influence of the Poetics on the Renaissance drama. Cf. Arthur Tilley,
The I.iteraltrre of the French Renaissance, I1 (Cambridge, rgoq), pp. 97-98: "Ac1
A. Philip McMahon
P O L Y D ~ RVIRGIL
E
The work of Polydore Virgil, De I~zventoribusRerum,' was popular
throughout Europe, appearing in numerous editions. Book I, chapter
10,on tragedy and comedy, derives from Donatus and D i o m e d e ~ . ~
4
, systematic treatise on tragedy and comedy was produced by G. B.
C a ~ a l iwhose
, ~
discussion is preserved in the collection of G r o n o v i ~ s . ~
His explanation of ~ t L B a p a ~iss still another variation on the most
discussed passage in the Poetics."e
cites the relevant words in
the Politics, and develops them to make it seem that Aristotle's
thought is a sort of spiritual d e c o r ~ m . Both
~
forms of drama are
essentially moralizing and instructive in nature; the difference
cording to the theory which the Renaissance critics built up partly on the practice of
the ancients, and partly on a misunderstanding or a t least a careless reading of
certain precepts in Aristotle's Poetics, tragedy and comedy were two perfectly
distinct species of drama. Tragedy deals with princes, ends unhappily, and is
written in a lofty style Comedy on the other hand draws its characters from the
middle or lower classes, employs a familiar style, and ends happily."
Paris, 1499
Lanson, op. cit , p. 545.
Floruit I joo--1525.
"ronovius,
Thesazsrz~sGraecarz~ntAn.tiqzril(~fzrnz,VIII, cols. 1598 ff., J o a n n i s
Baptisloe Casalii Rontani, Dc Tragoedia el Comoedia Lzsczrbratio.
Ibid., col. 1600: " D i ~ i t u rTragoedia ejusmodi personarum illustrium actiones
idcirco imitari, et spectatoribus proponere; u t contendat animi moerorem, ac
commiserationem concitare ad illas ipsas, quas commovet, perturbationes absterquendas."
"bid., "Adde quod saepe numero perperam homines dolent, ac pertimescunt
iis de causis, ob quas minus dolere, ac extimescere par esset: & cum in Tragoediis
proponuntur res cornmiseratione, ac terrore dignissimac, apprehendunt homines
quid et quo tempore dolendum, et commiserandum sit, quae utilitas et vitae fructus
est maximus, u t docet Arist. Politic. lib. VIII. sub finem."
7 An effective (?) argument about the didactic and moral value of drama was
made several centuries later by the critic who invented the term "Beaux-~lrts":
L'Abb6 J. B. du Bos, RtJlczions Critiques szlr la Po;sic el szsr la Peintz~re (3 vols.,
Dresden, 1760), I, pp. 424-425: I ' . . . je veux dire seulement que les PoEmes dramatiques corrigent quelquefois les hommes, 8: que souvent ils leur donnent l'envie
d'etre meilleurs. C'est ainsi que le spectacle imaginC par les LacCdi.moniens, pour
I 57
consists in the kind of people they represent and to whom they

appeal.'
Trissino's work on poetry, published in 1529 and 1563,~defines

tragedy in accordance with the Poetics, but for comedy he merely
parallels the pattern of that for tragedy:? He criticizes Dante for calling his poem a comedy, whereas it is more properly an epic4 Trissino
inspirer I'aversion de I'yvrognerie ?
leur
i
jeunesse, faisoit son effet. L'horreur que
la manie et I'abrutissement des esclaves, qu'on exposoit yvres sur un thkktre,
donnoit aux spectateurs, laissoient en eux une ferme resolution de resister aux
attraits de ce vice. Cette resolution empkchoit quelques jeunes gens de prendre du
vin avec excss, quoiqu'elle ne fbt point capable d'en retenir plusiers autres. . . .
L a Tragedie purge donc les passions A peu prss comme les remedes guhrissent, et
comme les armes dCfensives garantissent des coups des armes offensives. L a chose
n'arrive pas toujours, mais elle arrive quelquefois."
Gronovius, op. cit., col. 1606: "Ex omni mortalium numero Comoedia sibi
proposuit imitandos deteriores. Quoniam duo hominum genera turbas in civitate
faciunt; unum eorum qui potentia atque opibus pollent: alterum eorum qui rerum
inopia, & desperatione laborant. Igitur ad continendos in oflicio atque instituendos
utrosque hominum ordines, duabus Poeticae dramaticae formis Tragoediae &
Comoediae datus est locus. Tragoedia quippe admonet primaries & Principes, u t
intueantur in ewitus infelicissimos virorum potentium, qui limites aequi & recti
per vim egredi voluerunt: Comoedia docet inopes & destitutos de felici rerum
deploratarum, involutarumque successu bene sperare; etenim videtur magis fuisse
consentaneum rationi successum illum exprimere in ordinibus deterioris fortunae,
qui saepissime solent a rebus adversis gradum ad secundas & prosperas facere,
quam in Principibus, qui plerumque a statu vitae secundo corruunt in adversum."
Four divisions of the subject were published first by Trissino, with a long
intermission before the last two: L a Poctica d i M . G o v a n Giorgio Trissino (Vicenza,
1529); L a Qzlinta et la Sesla Divisione Della Poctica (Venice, 1562, 1563). Both
are found in Tzsttc lc Opere (Verona, 17zy), under the title, L c S e i Divisioni Della
Poelica.
G. F. Trissino, T u t t e Ic Opere, I1 (Verona, 1729)~p. 95 and p. 120. Cf. Spingarn,
op. cit., p. 76.
'I'rissino, 11,p. I 2 0 : "Resta adunque a trattare la imitazione dele azione, e costumi de i piu bassi, e peggiori, la quale si fa col deleggiarli, e biasmarli, et'a quel mod0
insegnare a gli uomini la virth, il che comunemente si suol fare con le Cornmedie, ne
le quali il Poeta non parla mai da se, come avemo veduto, che si fa ne le Tragedie,
ma sempre induce persone, che parlino, e facciano; e cosi ancora si fa ne le Egloghe
pastorali. . . . La Commedia adunque imita le azione peggiori con sermone, ritmo,
'
158
A . Philip McMahon
was also one of the first to emphasize the unity of time,' and to make
the epic a routine topic of scholarly criticism ."
For some reason, it is to be noted, Aristotle had not treated comedy
in the Poetics with the same elaboration with which he hcd discussed
tragedy, and thus when scholars afterwards treated comedy, they were
obliged to have recourse to a dull mechanical imitation of the definition
of tragedy in the Poetics, such as the Tractatz~sCoislinianz~soffers, or to
the traditional Peripatetic formula. The various difficulties of comedy
have placed later scholars, who accepted the verbal inspiration of the
Poetics, in an embarrassing position. Most of them have placed their
trust in the principle of cataclysm, like an older generation of geologists, and supposed that Aristotle did write on comedy in a second part
of the Poetics which has disappeared. A few recent writers have even
welcomed the Tractatus. But most scholars since the Renaissance have
bridged the gap by a judicious combination ol the Poetics with the
traditional theory of comedy.
et armonia, come la Tragedia; et imita una azione sola, compiuta, e grande, la quale
abbia principio, mezo, e iine; ma in questo e differente de la Tragedia, che come
quella fa la sua dottrina con la misericordia, e con la tema, cosi questa la fa col deleggiare, e col biasmare le cose brutte, e cattive . . . e per tal causa alcuni hanno pensato, che Dante nominasse Commedia il suo Pocma, percia che termina in bene, cio*
ne I'essere stato in Cielo fra l'anime beate; et altri vogliono, che pi;l tosto cosi lo
nominasse, per lo stile mediocre, in cui volea mostrare avcrlo escritto, percia che
ancora nomina Tragedia lo Eroico di Virgilio, per esscre in lo stile alto. E poi nel suo
libro de la Volgare Eloquenza egli nomina lo stile alto Tragico, et il mediocre
Comico, ct il basso Elegiaco, ma sia per qua1 causa si voglia, quel suo poema non
si puo nominar Commetlia, per non aver nulla di quello, che a la Commedia s'appcrtiene; ma essendo Dante nato in quella eta roza, et imbarbarita, chc non
conobbe ne vaghezza di stile Latino, ni? arte retorica, n? poetics, quantunque egli
fosse di profundissima memoria, e di ir~gegnoacutissimo, et elcvato, e di natura
quasi miraculosa, c fosse di quclla Teologia, c Filosofia, ct rZstrologia gi2 imbarbarite
instruttissimo, e ne le lettere sacre molto esercitato, e de le istorie, e favole Greche,
Latine, et Ebraiche dottissimamente informato, e de le cose de i suoi tempi mirabilmente instrutto, non potca per6 fare, che per lo difctto di quei sccoli, non incorressc in alcuni piccioli errori, come fu qucsto di nominarc Commedia la opera sua,
la quale (como ho detto) non ha nulla di qucllo, che a la Commcdia si richiede,
anzi pi& tosto tien de lo Eloico.
1 Spingarn, Literary Criticism, pp. 92-93.
Zbid., p. 109.
Aristotelian Deji~zitionsof Tragedy and Comedy
159
Daniello, who was also an interpreter of Dante, published a Poetica

in 1536.l He sets forth the traditional definitions and refers apparently
to that of the Poetics, but not as if they were contradictory or essentially different.2 His discussion with regard to the tragic hero is termed
" a curious misconception of Aristotle's meaning" by Spingarn? but
it merely reproduces the traditional treatment of the topic and
Spingarn himself discusses thc sources of Daniello's views earlier in
his book.4
I n 1537 there appeared a translation into French of the Electra of

Sophocles by Lazare de Baif. The work is prefaced by a definition of
tragedy whose relation to the mediaeval expression is o b v i ~ u s . ~
I n Giraldi Cintio, who wrote in 1543, we have knowledge of both the
Poetics and the traditional t h e ~ r y .Spingarn

~
considers his statement
of the unity of time as the very first enunciation of this law of the
drama? The traditional element is conspicuous in his view that the actions of tragedy are called illustrious because of the rank of the persons
repre~ented.~
Bernardino Daniello, Della Poetica (Venice, 1536).
Zbid., PP. 34, 38, 39.
3 Spingarn, op. cit., p. 83.
Ibid., p. 65.
Lanson, op. cit., p. 574: "TragCdie est une moralit6 compos6e des grandes
calamitez, meurtres et adversitez survenus aux nobles et excellents personages
comme Aias qui se occist pour avoir CtC frustrC des arrnes d'Achille, Edipus
qui se creva les yeux apr&squ'il fut dCclarC comme il avait eu des enfants de sa
propre m&re,apri-s avoir tuC son pi-re."
G. B. Giraldi, Discorsi . . . inlorno a1 Comporrc de i Romanzi, delle Comedie,
e delle Tragedie, e d i d t r e Maniere d i Poesie (Venice, 1554).
7 Spingarn, op. cit., p. 91.
8 Ibid., p. 62.
'
There is in the collection of treatises assembled by Gronovius a work

on comedy by another Giraldi%nd in it both Aristotle and the grammarians are cited, but his interest is in the latter.
Instead of attempting, as Rostagni has recently done, to prove the

influence of the Poetics on Horace, an earlier writer sensibly explains
the differences on a sound historical basis."
From "I'ib. VI De Scena et Poetarum Scaenicorum Historia," in Ilistoria
Poelartmt t a m Graecort~mquum Latinor.zrm Diulogi Decem, Basel, 1545.
Jacobus Gronovius, Thesaurus Graecorum Antiquitatzm (Venice, 1735), VIII,,
col. 1474: "1,ilii Gregorii Gyraldi, Ferrariensis, De Comoedia ejusque Apparatu
Omni et Partibus Commentarius: Comoedia est privatae civilisque fortunae sine
vitae periculo comprehensio. Graeci itidem sic definiunt: x o p y 6 i a i u ~ i vi 6 r w T r ~ S v
~ a TiO ~ L T L K G
~V
p a ~ p d l ~~ o
K ~v V ~ U VT O
~ PSL O X + . Donatus vero ita: Comoedia csl fabula,
dioersa institzrta continens affectuum civilium ac privutorum, qua discitur quid sit in
vita utile, quid contra mita?zdum. M . Tullius Comoediam esse ait, imitationem vitae,
speczdum conszietudinis, imaginem veritatis: id quod est visus accepisse a Livio Andronico, qui Comoediam ante Ciceronem esse dixerat quotidianae vitae speculum:
nec injuria: nam ut intenti speculo veritatis lineamenta facile per imagines colligimus, ita lectione Comoediae imitationem vitae, consuetudinisque non aegerrime
animadvertimus. Unde etiam scriptum est, Comoediam esse poema sub imitatione
vitae, atque similitudine compositzrm. Aliam tamen definitionem affert Aristoteles,
quae 8E alium habet finem: ita enim ait: Comoedia est imitalio improbioris quidem,
n o n ad omnent tamen n~alitiant,sed turpitudinis quaedum est ridicztla particula.
Kidiculum e n i m peccatum qzroddam est, & burpitudo doloris expers, correptionem
mininte afferens: veluti ridictila facies, quae stotim est tzrrpe quiddanc & perversum,
sine dolure. Videtur ex his verbis Philosophus innuere, Comoediam inventam esse
a d oblectandos populos: & propterea varia quae tradunt de ejus inventione recitabo." . . . [The differences between the species are distinguished thus; cols.
1478 ff 3: "Nunc quid inter Tragoediam Sr Comoediam distet, disquiramus. I n
Comoedia quidem mediocres fortunae hominum, parvi impetus, periculaque,
laetique sunt exitus actionum: a t in Tragoedia omnia contraria, ingentes personae, magni timores, esitus funesti habentur, & illic turbulenta prima, tranquilla ultima: in Tragoedia contrario ordine res aguntur, tum in Tragoedia fugienda
vita, in Comoedia capessenda exprimitur. Postremo omnis Comoedia de fictis
argumentis. Tragoedia saepe de historica fide petitur. . . . Sed enim IIomerus,
qui omnis poeticae largissimus fons est, his carminibus exempla praebuit, & velut
quadam suorum operum lege praescripsit, qui Iliaden instar Tragoediae, Odysseam
ad imaginem Comoediae fecisse monstratur."
Francisci Philippi Pedmontii Ecpkrasis i n Iloratii Flacci Arlent Poeticam
(Venice, I j46), f. 31 v.: "Kt ideo non mirum, si cum Aristotele Flaccus quandoque
161
The first critical edition of the Poetics, together with a commentary
on Horace, was that of Robortelli in 1548,' and it appeared in the same
year as the dictionary oi Junius. He translates ~ h 6 a p c r ~by

s
purgan~,~
and explains it by reference to the politic^.^ The value of tragedy is
in the satisfaction which spectators have in seeing that there is no
mortal, however highly placed, who is not subject to the greatest calamities. But in the commentary on Horace there is clear evidence of
his study of the Roman grarnn~arians.~
Capriano seems to be based on the Poetics, in the light of the grammarians, and much of what he says shows an advance over some of his
predecessor^.^ His observations on the epic have an interesting
parallel in the work of the French critic Pelletier, published in the
same year.'
non consentit. Quippe ille ab antiqua poesi non discedens artis poeticae normas
tradidit; hic autem uates, quod quidem permultam interest, cum maiorum tum
iuniorum poemata perpendens quaedam admittit, quaedam uero non probat. Nam
& ipse peripateticorum princeps ueterum philosophorum dogmata taxans antiquam philosophiam balbutientem nuncupauit. quod si a b eorum opinione in rebus
philosophicis quasi semper dissentire sibi licuit, cur & auctori nostro in poeticis
cum ipso semel non conuenire crimini dabitur?"
1 Francisci Robortelli Utinensis i n Librum Aristotelis de Arte Poetica Explicationes, qui ab eodem Allthore ex Manuscriplis Libris, M d t i s i n Locis Emendatus Fuit,
u t j a m Dificillimus ac Obscz~rissimz~s
Liber a NuRo ante Declaratus Facile ab Omnibus
Posset Intclligi. [Together with the preceding is]: Paraphrasis i n Librum Horatii q u i
Vulgo de Arte Poetica ad Pisones Inscribitur; ejusdem Explicationes de Satyra, de
Epigrammate, de Comoedia, de Salibus, de Elegia. (Florence, 1548.) His readings
were drawn upon in the edition of the Poetics published by E. A. W. Crafenhan
a t Leipzig in 1821. References below are to the edition printed a t Florence, 1658.
I n Libr. Arist., p. 52.
Ibid., pp. 52-53.
Ibid., p. 42.
Giovanni Pietro Capriano, Della Vera Poetica, Venice, 1555.
Cf. Spingarn, op. cit., pp. 42-43> 83-84.
7 Ibid., p. 211.
A . Philip McMuhon
Jacque Yelletier's Art Po&tique(1555) shares in a moderate way the

point of view of the PlEiade,' and although his work appeared before
that of Scaliger, Spingarn surmises that he rnay have known the great
critic or have gone to the same sources. His definitions are the traditional ones; so that Spingarn's conclusions do not immediately follow
as far as Pelletier is concerned? for the words of his formula for tragedy
might easily have been taken from a mediaeval a ~ t h o r . ~
Eleven years after the publication of Robortelli's edition, there appeared the work on poetry by lintu urn^,^ which he himself afterwards
translated into Italian? He insists on the appropriate endings of tragedy and comedy, in terms reminiscent of Diomedes, but he appeals to
the Poetics to establish his point.? IIis paraphrase of the definition of
tragedy is quoted, immediately after a translation into Latin of Aristotle's text, by Vossius, who with Scaliger and Heinsius was among the
most powerful critics of the seventeenth and eighteenth centuries.8
s
to be expiutio. I n the
hlinturno's equivalent for ~ h ( 3 a p u ~appears
Ibid., p. 191.
Ibid., p. 200.
3 [bid. p. 2 0 0 : "By this time, then, Aristotle's theory of tragedy as restated
by the Italians, had hecome part of French criticism."
4 Cf. Tilley, op. cit., p. 08, n. I : 'La\u
lieu des personnes comiques qui sont de
hasse conctition en la Tragedie s'introduisent lcois, Princes et grands Seigneurs.
E t au lieu qu'en la Comedic les chases ont joyeuse issue, en la Tragedie la fin est
tousjours luctueuse et lamentable, ou horrible 2 voir. . . . La comedie parle facilement, ct comme nous avons ciit, populairement. La trag2die est sublimc, capable
de grandes matieres."
6 A~zto~zii
.Scbastia?zi fifintzcrfzi Dr:.Pocla Libri Sex (Venice, 15 jq).
Antonio Schastiano fifinturno, L'Arle Poctira (Venice, 1564).
De Yoetcz, p. 1 2 j.
8 In Vossius, Ilzst. Port, 11, xi, z (p. 94 of 1696 ed.): "Imitatio insignis, seriaque
aliqua, & absoluta, 8L magnitudine quackam comprehensam actionern oratione suavi
exprimens: ita u t ejus partes suo sinpulac quidem loco, atque seormlm, adhibeantur; nec sinlplici narrrrtionc, sed inductionc Illorum, qui i t s agunt ac ciicant, u t
miserationem terroremque concitcnt ad id genus morborum expiationem."
1
2
'
Aristotelian DeJi+zitions of Tragedy and Comedy
163
Italian translation by his own hand, he notes explicitly that the ideas
of Cicero and of Aristotle in the Poetics do not conflict.'
Bernardo Tasso, the father of Torquato, however, relies entirely

upon the witness of Cicero and the mediaeval tradition for his definition~.~
One of the most important of Renaissance scholars and critics, whose

dominance of opinion in his own and the following generation can
hardly be exaggerated, is S~aliger.~
Of him, Spingarn says: "He was
the first to regard Aristotle as the perpetual lawgiver of poetry. He
was the first to assume that the duty of the poet is to find out what Aristotle says, and then to obey these precepts without question. He distinctly calls Aristotle the perpetual dictator of all the arts: 'Aristoteles
imperator ~zoster,om~ziumboftnrum artium dictator perpetuus.' " His
effect on French classical drama was decisive. A good deal of Elizabethan criticism also reflects his dicta.5
Modern historical criticism, colored by modern philosophy, has been
'
L'Arte Poetica, p. 116.

Bernardo Tasso, Ragionanzento della poesia (Venice, 1562), (in Belle Lettere d i
M . Bernardo Tasso, 2 vols., Padua, 1733,pp. 511-s38), 11, pp. 513and 515.
Julius Caesar Scaliger, Poetices Libri Septenz. Geneva, 1561; Lyons, 1561;
Heidelberg, 1581,1586, 1607, 1617, et cet. Cf. Select Translations from Scaliger's
Poetics, translated by F. hf. Padelford (Pale Studies i n English, vol. XXVI), New
York, 1905. E. Brinschulte, J. C . Scaligers Kunsttheoretische Anschauungen. (Diss.,
Bonn, 1913);J. C. Scaligers Ktbnsttheoretische Anschazrzingen zlnd deren Hazrptquellet>
(Bonn, 1914).
4 Spingarn, up. cit., p. 141. But cf. Thomas Pope Blount, Censure Celebriorum
Azfthorum, London, 1690,p. 570: "Dii Boni! quam multa ille, quam multa vetera,
non lecta, non visa, non audita aliis depromit! Daemonium hominis. Lips. epist.
Quaest. I. 3. epist. 20." . . .. "Literarum omnium Dictator perpetuus, & per omnia
Diis magis quam IIominibus comparandus, Josephus Scaligcr, Gasp. Scicpp. praefat. de Art Crit." . . . "In Criticis omnium recte aestimantium judicio princeps sine
controversia, sine aemulo ac rivali dominatur. Baud. orat. in obit. Scalig."
G. Gregory Smith, Elizabethan Critical Essays, I (Oxford, ~goq),p. lxxvi, n. 7.
Cf. noteson I, 68.25 and 80.7.
A. Philip McMahon
prone to regard him as a false prophet or a t least a perverse pedant.'
Until Lanson pointed out the general sources of this energetic critic,
both his method and his influence were a mystery. The fact is that
Scaliger, in addition to his ideals of scholastic system and formal, legalistic classification, analyzed fully as much the traditional views and
their sources as he did Aristotle's Poetics, and the popularity of his
work on poetry was due to a favorable attitude on the part of his readers and followers toward such a method.
He considers that the Odyssey is more truly a tragedy than the I l i ~ d . ~
He criticizes the definition of comedy by the grammarians on the
ground that it would include other literature in addition to the drama:
but his objection that danger is involved even in comedy obviously has
reference to the grammarians' expression4
When contrasting tragedy and comedy, however, the points to which
he draws attention are precisely those stressed in the traditional formulas, and he utilizes the material provided by Diomedes, even to the
anecdote about Euripides and King Archelau~.~
And, in general, the
* E. Lintilhac, De J. C. Scaligeri Poetire (Thsse), Paris, 1887, is a futile and

elaborate discussion of Scaligcr's verbal and internal divergence from the Poetics
of Aristotle.
Poet., I , v (p. 23, ed. 1617).
I'oet.,I,v (p. 24,ed. 1617).
* Gronovius, op. cif., col. 1498: "De Comoedia illa quoque falsa Grammatici
docuerc: quia esset poema positum in imitatione, totum in gestu consistere atque
pronunciatione. Profecto nihilominus Comoedia est, etiam quum legitur vel
tacitis oculis. Quin gestus recitantium solus est: non omnes qui legunt, recitant.
Praeterea nimis jam saepe dictum est, imitationem univcrsae poeseos finem esse.
Comoediam igitur sic definiamus nos, poema dramaticum, negotiosum, exitu lactum,
stylo populari. Errarunt enim, qui Latinis sic definivere. privatarum personarum,
civilium negotiorum comprehensio, sine pcriculo. Principio aliis quoque fabulis
convenit, non dramaticis quac simplici narratione recitari possunt. Deinde in
Comoedia semper est periculum, alioquin esitus cssent frigidissimi. Quid enim est
aliud periculum, quam immincntis mali aditio sive tcntatio? Praetcrea non solum
pcricula, sed etiam damna lenonibus, rivalibus, & servis, 8r heris: quemadmodum in
Asinaria, & in Afost~llaria,ipsi quoque heri male mulctantur. Adhacc Praetextatas Comoediae nomine appellare nequeant ex ea definitione: non enim sunt privatae
personae. l'ostremo & Mimis est communis definitio, & Satyrae dramaticae."
"bid., col. 1498: "Tragoedia, sicut & Comoedia in exemplis humanac vitae confirmata, tribus ab illa dilfert, I'ersonarum conditione, fortunarum, negotiorumque
qualitate, esitu: quare stylo quoque diaerat necesse est. I n illa e pagis sumpti
Chremetes, Davi, Thaides loco humile: Initia turbatiuscula: fines lacti. Sermo
165
Renaissance theorists were more concerned about tragedy than about

comedy, for, as already noted, Aristotle had not discussed comedy with
the same detail that he had tragedy, and the difficulties of harmonizing
contemporary practice with Aristotle's Poetics were consequently
greater. Besides, contemporary comedy was more independent of the
scholarly critics for its support.
I t is noteworthy that Scaliger does not blindly worship the text of
Aristotle; he even rejects the philosopher's definition of tragedy in
favor of his own, ruling out those technical elements in Aristotle's formula which pertained only to stage presentation in the Athens of his
day, and objecting to ~ciflapncras irrelevant. His actual words, in his
own definition of tragedy, which he expressly prefers to that of the
Poetics, are simply a return to the formula of the Roman grammarian^.^
GF&VIN
GrCvin in 1562 thought that French drama was already able to display productions perfect according to the rules of Horace and Aristotle,
but his theory of tragedy shows faint traces of Aristotle and none of
S ~ a l i g e r .H
~ is theory of comedy goes back to tradition, and he repeats
the formula of C i ~ e r o . ~
-
de medio sumtus. I n Tragoedia Reges, Principes, ex urbibus, arcibus, castris.

Principia sedatoria: exitus horibiles. Oratio gravis, culta, a vulgi dictione aversa,
tota facies anxia, metus, minae, exilia, mortes. Memoriae proditum est, Euripidem
ab Archelao rege Rfacedoniae, cujus in fide, ac clientela esset, rogatum, ut de se
Tragoediam scriberet. Ne, inquit ille, Jupiter, ne tantum mali."
Ibid., col. 1499: [After quoting the Poetics,] "Quam nolo hic impugnare aliter
quam nostram subnectendo. Imitatio per actiones illustris fortunae, exitu infelici,
oratione gravi metrica. Nam quod harmoniam & melos addunt, non sunt ea, ut
Philosophi loquuntur, de essentia Tragoediae. etenim Tragoedia in scena tantum
esset, extra scenam non esset. Quod autem dixit, p i y t e o s ixobulls, positum est, ad
differentiam Epopoejae, quae aliquando prolixa est. Non tamen semper. cujusmodi
vides apud Musaeum. Praeterea ~ri6'apurs vox neutiquam cuivis materiae servit:
sicut p k y t e o s mediocritatem significat hic. Paucis enim versibus nequit satisfieri
populi expectationi: qui eo convenit, ut multorum dierum fastidia, cum aliquot
horamm hilaritate commutet. Quemadmodum inepta quoque est prolixitas: adeo
ut facete dicas illud Plautinum,
L u m b i sedendo, o c d i spectando dolent."
Spingarn, op. cit., p. 2 0 1 .
Jacques Grkvin, Thddtre Complet, avec Notice et Notes par Lucien Pinvert
(Paris, 1922)~p. 7: "Or je reviens i% la ComCdie, qui est un discours fabuleux, mais
A . Philip McMaholz
Castelvetro is credited with the first affirmation of the unity of

place and with formulating the three unities in their definitive form, the
form that was accepted by French classicist critics and p l a y ~ r i g h t s . ~
His acute remarks on the function of the chorus in ancient drama, presented in his controversy with Pigna, give added reason for distinguishing between tragedy and comedy on the traditional basis? He treats
approchant dc veriti., contenant en soy diverses manihes de vivrc entre les citadins
de moyen estat, et par lequel on peult apprendre ce qui est utile pour la vie, et au
contraire cognoistre ce que lon doit fuir, enseignez par le bonheur ou malheur
d'autruy. C'est pouquoy Cictron l'appelle imitation de vie, mirouer des coustumes,
et image de viriti.."
Lodovico Castelvetro, Portico d'ilristolclc, ~ J z ~ l g u r i ~ s uetl aSpostu (Vienna,
1570); (revised and enlarged, Hasel, 1576, r 582, 1678). Opere Vurie . . . ATon
P i d Stumpale, collu V i t a dell' Aulorc Scritta dal Sig. Lodmlico Antonio Muralori
(Milan and Berne, 1727). Cf. H. B. Charlton, Castelvetro's Theory of Poetry (Manchestcr, 1 ~ 1 3 ) .Antonio Fusco, L o Poelicu di Lodovico Castelvetro (Naples, 1904).
Spingarn, op. cit., pp. 67-65, loo.
Opere Varie, pp. 81 ff.: "Primicramente adunque nel prcdetto Libro egli
[G.-B. d(l.11~I'ignu] vuole, che si crcda, che egli sia il trovatorc di quclla opinione,
che la Tragedia non possa aver per soggetto azione procedentc ddl' ingegno del
l'oeta, convenendogli di necessitd, che sia stata prima riccvuta dal popolo, come
manifesta, o per Istoria, o per fame in generale; poichit delle azioni Reali si tiene
conto dal mondo. e se ne fanno Istorie, e passano a notizia di tutti, almeno in generale; siccomc dall' altra parte la Commcdia non pu6 aver per soggetto se non
azione tutta imaginatasi dal l'oeta, non essendo verisimile, chc il grido delle azione
Cittadinesche private si rallarghi fra molte persone, e se ne conservi la memoria o
per Istoria, o per fama lungo tempo, Ia quale azione Cittadinesca privata E la
matcria della Commedia siccome l'azione Keale i- la materia della Tragetiia. Appresso in difendcndo egli Eratosthenc-, il quale avvisava, siccome mc, chc la I'oesia
dovcssc solamentc dilcttare, e non giovare, tlalle riprcnsioni tli Strahonc protlucc
artlitamcnte in mczzo qucsta risposta, come stla, clie posto cl:c la I'oesia possa
alcuna volts giovare, si pub nondimcno sicuramentc dirc, cha punto non giovi, non
giovando ma1 sc non pcr accidente, e procedentlo il giovamento piti tosto tli fuori,
e dall' acutczza tlcl lcggente, chc tlentro tlalla Pocria, e tlall' intenzione dcl Poeta.
Ultimamcntc senza punto arrossare, rcntle la veracc ragionc, come trovata da lui,
pcrch? il Coro, quantunquc sia constituito di pcrsone popolari br vili, non convenga
alla Commcdla rappresentanti popolari, & i vili, come fa alla Tragctlia rapprcs e ~ t a n t Si ignori & i Nobili, la quale E cosi fatta: Non pu6 il giudicio dcl Popolo
tutto, il quale & soggetto del ragionamento tlel Coro, aver luogo se non ncllc azioni
'
167
the definitions of tragedy as primarily a defense against Plato,l and he

says Cicero's remarks on laughter prove that he never read the poetic^.^
He also holds that the text of the Poetics does not give sufficient
grounds for believing in the previous existence of a second book, now
lost.3 He refers to Donatus for data on the costume of the comic
actor^.^
Reali, le quali di parte in parte, mentre che si fanno, non che tutte, poich? sono
fornite, si divulgano tra Sudditi riguardanti, consideranti, e giudicanti i detti, & i
fatti de' lor Signori: 12 dove le azioni Cittadinesche private pervengono a gli
orecchi di pochi, n@sono subito sapute, ni? danno da pensare, e da ragionare a tutto
il populo o facendosi, o ancora poich? son fatte. "
Poelica, 1570, f . 64 v. ff. (At the end of the book Castelvetro asks the reader
to pardon the many typographical errors in the book due to the fact that the German
printers did not know Italian.)
Ibid., f. 51 r. ff.: "Ma con tutto che la materia pertenente a riso fosse, si come
io m'imagino distesa da Aristotele ne libri poetici, non dimeno Cicerone non la
lesse mai, percioche se l'hauesse letta, non direbbe sotto persona altrui, che i libri
di questo soggetto liquali haueua veduto dessero piutosto materia da ridere che
insegnassero certa dottrina di riso, conciosia cosa che gli nsegnamenti d'Aristotele
per isciocchezza non dieno da ridere, ma per sottilita rendano altrui stupefatto."
Ibid., f . 61 v.-62 r.: "Ha detto Aristotele infino a qual termine l'epopea habbia
fatta compagnia alla tragedia, & quale forma le habbia data, hora restaua a parlare
infino a qual termine l'epopea habbia fatta compagnia alla comedia, & quale forma
le habbia data percioche haueua detto che il Margite, il quale poema senza dubbio
era epopeico haueua date le figure alla comedia. hfa egli promette di parlar poi di
questo. La qual parte manca, & perauentura in questo volume non se scrisse mai
nulla. Itla perche alcuni adducono questo luogo a prouare che Aristotele promette
di parlare della comedia, cio 2 dalla sua natura & di tutto cio che le appertiene come
fara della tragedia, la qual promessa vogliono che habbia attenuta nel second0 libro
che s'imaginano essere perduto per ingiuria di tempo, 2 da sapere che Aristotele non
promette qui di parlare della comedia se non in quanto ha riceuuta forma dell' epopea nella guisa che in questa particella ha parlato della tragedia non ragionando
se non di quello che ells ha riceuuto dall' epopea. Hora io non tralasciero di dire
che alcuna volta ho sospettato che questo testo nella voce ~ w g w 8 i a ns on sia cambiato
volendo hauere ~ p a r w o ' i a rpercioche
,
@ assai verisimile che parendo ad Aristotele per
la conclusione generale che haueua posta che le cose che ha l'epopea si truouano
nella tragedia, & che tutte le cose che ha la tragedia non si trouano nell' epopea si
douesse a raccontare particolarmente le cose che ha l'epopea di meno che ha la
tragedia, & le cose che ha la tragedia di piu che l'epopea dica che non voglia dire
a1 presente, ma che dira poi luogo piu conueneuole & perpoco necessario quando si
questioners quale tra l'epopea o la tragedia sia di antiporre."
* IIbid., p. 56 v.
A. Philip McMahon
Jean de la Taille in 1572 first distinctly formulated the doctrine of

the unities in France, deriving the principle from Castelvetro.' His
Art de TragPdi~appeared as a preface to his Saul le Furieux, and he
emphasized form as against the irregularity of the moralitie~.~
He
esteems tragedy as being different in tone from such plays, in terms
derived from Seneca? but he objects to the grammarians' fornrula, in
order to admit Biblical themes into tragedy: and he looks to the
naturalization of classic drama in F r a n ~ e . ~
Iason Denores, who had commented on Horace in 1553, was apparently among the first to repeat the methods adopted some centuries
before him, by the compilers whose work is seen in the Tractatus Coislinianus. Lacking a definition of comedy in the Poetics to parallel
that of tragedy, he constructs a mechanical definition in the same patSpingarn, op. cit , p. 2 0 6 .
Ibid., p. 202.
A. Werner, JCUVL
de la Taille u n d sein Saul le F?irieux (Leipzig, ~goX),p. 10:
"La Tragedie done est vne espece, et vn genre de Poesie non vulgaire, mais autant
elegant, beau et excellent qu'il est possible. Son vray subiect ne traicte que de
piteuscs ruines des grands Seigneurs, que des inconstances de Fortune, que bannissements, guerres, pestes, famines, captiuitez, execrables cruautez des Tyans. et bref,
que larmes et miseres extremes, et non point de chases qui arriuent tous les iours
naturellement et par raison commune et. cet."
Ibid., p. 11: "Quant 2, ceulx qui discnt qu'il fault qu'vne Tragedie soit tousiours
ioyeuse au commcnccment et triste 5 la fin, et vnc Comedic (qui luy est semblable
quant 5 l'art et disposition, et non du suhiect) soit au rcbours, ic leur aduise que
cela n'aduient pas t,ousiours, pour la diuersitk des subiccts et bastiments de chascun
de ccs dcux poemes."
"bid., p 12: " E t voudrois bicn qu'on eust barmy de France tellcs ameres
espicerics qui gastent le goust de nostre langue, et qu'au lieu on y cust adopt6 et
naturalis6 la vrayc lragcdie et Comedie, qui n'y sont point cncor 2 grand' peine
paruenues, et qui toutcfois auroient aussi bonne grace en nostrc languc Franroise
qu'cn la Grccque et Latinc "
lason Dcnores, I12 Epistola~nQ. Horatii Flacci de Avte Yoetica . . . Inlerprelatio (Venicc, 1553).
169
tern as his expansion of the definition of tragedy,l introducing elements

from various source^.^
One of Scaliger's bitterest opponents, Martin Antonio del Rio, agreed

with him nevertheless in the theory of tragedy and comedy; he published a large work on Latin tragedy, in which he cites various
standard authoritie~.~On tragedy he quotes Theophrastus, Scaliger,
Minturno, and Aristotle, translating the ~&Bapursclause of Aristotle
in a medicinal sense." I n defining comedy he reproduces Diomedes,
and accepts his distinctions between the two species. He expressly
claims that Diomedes does not conffict with Aristotle, and he also
quotes the anecdote about Euripides and Archelau~.~
Poetica d i Iason Denores. Xella qua1 per V i a d i Defitzitione, O Diuisiotze si
Tratta Secondo l'Opi?zione d'dristotele della Tregedia, del Poema Ileroico, O della
Comedia (Padua, 1588), f. 6 r.: "E dunque la Tragedia imitation per rappresentation di una attion marauigliosa, compita, & conueneuolmente grande d
i persone
Illustri, mezzane fra buone, & cattiue negli errori humani per qualche horribiliti,
che cominciando da allegrezza finisce in infeliciti nello spacio di vn giro di Sole,
composta con parole altiere, & graui, & con uersi sciolti endecasillabi; 6 per il piu
de sette, & di cinque silabe; 6 con ambidue mescolatamente, & ne' chori con canzioni, & con madrigali, per purgar gli spettatori col diletto, che nasce dalla imitatione, & dalla rappresentatione dal terrore, & dalla misericordia, & per fargli
abhorrir la uita de' tiranni & de' piu potenti."
? Ibid., F 117 r.: "E dunque la Comedia imitation per rappresentation di una
attion marauigliosa, compita, & conueneuolmente grande di persone priuate,
mezzane fra buone, & cattiue negli errori humani per qualche sempieta, cbe principiando da trauaglio fmisce in riso, & in allegrezza nello spacio di un giro di Sole,
composta con parole humili, & con uersi corti di sette, & di cinque sillabe, per
purgar gli spettatori col diletto, che nasce dalla imitatione, & dalla rappresentatione,
da quelle passioni, & discontentezze, che turban0 la lor quiete, & tranquiliti, per
gl' inamoramenti della mogli, delle figliole, de' figlioli, per gl' inganni, & tradimenti
de' seruitori, de' ruffiani, delle nutrici, & di altre persone simili, & per fargli inamorar della uita priuata a conseruation di quella tal ben regolata Republica populare, nella quale si troueranno."
M a r t i n i Antonii Delrii Syntagma Tragoediae Latinae ( 2 vols., Antwerp, 1593).
Ibid., I, pp. 1-2.
Ibid., I , p. 5.
A . Philip McMal~on
Vauquelin de la Fresnaye, the theorist of the Plkiade, relies eventually on the Poetics for his definition of tragedy,' but his statement of
comedy is illustrated with original themes, while depending on the
grammarians2
A Poetics compiled by certain professors of Giessen, which reached

its third edition in 1617, relies upon Diomedes and Cicero for its understanding of c ~ m e d yfor
, ~ tragedy upon Diomedes and Scaliger? with
interesting local references in the discussion of the persons of ~ o m e d y . ~
One of the most interesting references of the period to a theory of

comedy, to be found in the literature which is not exclusively critical,
is that in Don Quixote where the Curate digresses, while inveighing
against books of chivalry, to attack Spanish comedy on the grounds of
Georges Pellissier, L'aziqz~din de la Fresnaye, L'Art PoCtiqzte (Paris, 1885),
PP. 733-134 (1, 153 ff.):
AIais le suiet Tragic est vn fait imitC
De chose iuste et grave, en ses vers limit&,
Auqucl on y doit voir de l'affreux, du terrible,
Vn fait non attendu, qui tienne de l'horrible,
Du pitoyable aussi.
Ibid., 1. 143:
Ida comedie est donc vne Contrefaisance
D'vn fait qu'on ticnt meschant par la commune vsance,
Alais non pas si meschant qu'a sa meschancetk
I7n rcmcde nc puissc cstre bicn aportC:
Comme quand vn garcon, vne fille a rauie
On peut en l'espousant luy racheter la vie.
Poetica per Acuden~iueGessinue i\ronlzz~llos Projessores (gd ed., Giessae Hassorum, 1617), p. 331.
Ibid., p. 335.
Ibid., pp. 337-338.
I I
its lack of unity and plausibility. I n contrast, the definition of Cicero

and the correctness of foreign writers are cited.'
Lope de Vega wrote an apology for the unclassic drama of Spain, and
his Arte Nuevo de Hacer Comedias en Este Tienzpo defends his own
practice and that of other writers of the time. I n his essay he refers to
the Ciceronian definition13and advises the reader who wishes to know
the rules to study R~bortelli.~
His essay concludes with a versified
loose paraphrase of part of the traditional f ~ r r n u l a .The
~ grammarians'
definitions could thus include Spanish comedy of the Golden Age.
The brief treatise of Daniel Heinsius, the pupil of Joseph Scaliger,

De Tragoediae Constitutione, appeared in 1610;~because of its authority
and conciseness it was termed by Chapelain, "the quintessence of Arktotle's Poetics." I t was accepted as an authority by Racine and Corneille; it was really through this work that the influence of the teacher's
father was dominant in France: and even penetrated into England,
affecting Ben Jonson and Dryden. Heinsius seems to have anticipated
Bernays in considering the Pythagoreans and Neo-Platonists in relation to the controversy over ~ h 6 a p a t s and
, ~ on the tragic hero he invokes scholastic ethicsg So much of his material is reproduced by
Vossius that it need not be discussed extensively here.
hfiguel Cervantes de Saavedra, El Ingenioso H i d d g o Dorz Qztixote de la Manchu
(Madrid, 1608). Chapter xlviii (Part I\'), f. 255 r.
Translated by William T. Brewster in the Papers on Play-Making, Vol. I , with
an introduction by Iirander hrathews (Dramatic RIuseum of Columbia University, New York, 1914). Reprinted in Clark, European Theories, pp. 89 ff.
Sec. I I (Clark, op. cit., p. 90).
Sec. 13 (Clark, op. cit., p. 91).
Sec. 28 (Clark, op. cit., p. 93).
Aristotelisde Poetica Liber (Leyden, 1610 and 1611; Paris, 1625 and 1645). De
Tragoediae Constit~~tione
(Leyden, 1643). Cf. Kacine's marginal notes on the lastnamed work in Racine, CEztnres, ed. Paul RIesnard, V I (Paris, 1865), pp. 288-290
Spingarn, op. cit., p. 245.
De Tragoediae Const., p. 10 (1643 ed.).
Ibid., p. 76 (1643 ed.).
A . Philip McMahon
\'ossius was the most indefatigable in his collection of material bearing on tragedy and comedy.' EIe may properly be considered the best,
as he is the most comprehensive, single authority for study as a representative Renaissance theorist and scholar. I-Ie compiles and codilies
all the data he could find in classical literature and the critical work of
the preceding c e n t ~ r y ..At
~ the head of one section he writes dfetl?odus
Scribendae Tragoediae,%hich might well have been used as a general
title for most of the Renaissance poetic treatises we have been considering, if not for the Poetics of Aristotle himself. Before giving Aristotle's
definition of tragedy from the Poetics, he gives his own, in which he admits purgation if desired, but does not insist on it.4 After translating
the definition of tragedy from the philosopher, he cites Scaliger, but disagrees with him on the necessity for the unhappy ending, and he offers
an ingenious explanation for the spectators' happiness in witnessing
unhappine~s.~
H e quotes Theophrastus's and Diomedes's definitions
Cf. Sandys, op. cit., 11, pp. 307-309.
Gerardi Joanftis Vossii, De Artis Poeticae 11-atzrraac Co7tstilz~tioneLiber (Amsterdam, 1647). Poeticarunt Institutionunt Libri Tres (Amsterdam, 1647); Tractatus
Philologici de Rhetorics, de Poetica, de Artium et Sciefztiarum Xatura ac Constitntione (Amsterdam, 1697).
3 Poet. Znst., 11, xviii. Cf. also: Compendiunz Artis Poeticae Aristotelis ad
LTsumCmz$ciendorum Poematurn, Ab Antonio Riccobono Ordinatum, & quibusdam
Scholiis Explanaturn (Patavii, 1591).
* Zbid., 11;xi, 2 : "Tragoedia est poems dramaticum, illustrem fortunam, sed
infelicem, gravi et severa oratione imitans. Quibus et finem hunc, si voles, adde:
a d affectus ciendos, animumque ab iis purgandum."
5 Zbid., 11, xi, 2 : "I'er haec vero tragoedia aepaiverv TGV T C ~ o r o i r ~ w
.rraOqp&~wv
v
~(rOapurv,purgare, ac leoare animum ab huiusmodi perturbationibus. Quale est quod
(ut unam e multiplici purgatione afiectuum memorem) homines videntes adversa,
in quibus jam olim inciderint magnae animae, facilius discunt ferre praesentia. Qua
de re elegantes Timoclis versiculos legere ?st spud Athenaeum initio libri sexti.
"Julio Scaligero lib. I , de re I'oetica cap. vi. definitur, imitatio illustris fortunae,
exitu infelici, oratione gravi metrica. Ubi illud probare non possum, quotl requirat
exitum infelicem. Plurimum quidem id sit; sed non est de obuip tragoediae. In
multis enim id Graecorum tragoediis non videas; u t postea dicetur. Quare d8erentia ei6o.rrords, qua difiert a comoedia, in eo consistit, quod graves actiones imitatur, eoque graves etiam personas assumit. Sed, utcumque non semper exitus sit
I 73
of tragedy, as if not contented to rely on the Poetics alone.' I n another

place, however, he rates the happy ending in tragedy as a concession to
the mob.2 He presents various definitions of comedy, but defends Diomedes against Scaliger, whose formula he also reproduce^.^ Diomedes
and Donatus are both given on the differences between tragedy and
infelix; semper tamen infelix conditio, vel grave periculum, ob oculos ponitur,
Nam affectus ei movere propositum est, in primis misericordiam.
"Quaerat aliquis, quomodo delectationem, quae poetae omni est proposita,
adferre tragicus possit, cum tristia et luctuosa repraesentet: nec viri boni sit
gaudere alienis malis, praesertim virorum illustrium. Huic difficultati occurritur,
si cogitemus, spectatorem capere voluptatem, non ex eo, quia aliis fuerit male:
verum ex artificio po&tae: quomodo exhorrescimus conspectum draconum, &
monstrorum: a t jucunda est eorum pictura, quia pictoris nos ars delectat. Etiam
alterum hic concurrit. Nam suave est nosse casus tantos, quia haec res prudentiam
auget."
Ibid., 11, xiii, 10: "Trahitur argumentum tragicum ex calamitatibus atrocibus,
quae heroibus, & regibus, accidere. Ut duo hic spectare oporteat; personas, & res.
Personae sunt illustres; u t heroes, & reges.
"Unde tragoedia a Theophrasto dicebatur esse fipwi~iisr i r x ~ sacpiu7auts, u t apud
Diomedem legere est lib. III. Cui ea itidem definitur heroicae fortunae in adversis
comprehensio. Ac similiter Etymologici magni auctori describitur, Biwv r e , ~ a i
X6ywv 7 j p w t ~ S vpipqurs. Sed Calamitatum hic nulls est mentio: quod fit in illa Diomedis. Quas etiam expressit Elias Cretensis in Nazianzeni orat. XXIII: Tragoedias, inquit, veteres appellabafzt, qzcae ob gravissimas calamitates canebantur: quemadmodz~mcontra comoedias eas, quae ridicule argumenta contifzebant. Rarius est,
u t atroces illae calamitates in heroum domos incidant: quae et caussa est, cur
pauca olim tragoediarum fuerint argumenta."
Ibid., 11, xiii, 31: "Neque, si exitus sit laetus, eo nomen tragoediae amittitur: quia non est de oiruiq ejus, u t exitus sit tristis. Alioqui minor pars tragoediarum Euripidis, quae quidem hodie exstent, tragoediae esse desinerent. Quare,
. . . satis est, si facies ejus sit luctuosa & anxia, ita u t in atroci, & ancipite illustrium
personarum fortuna, natura tragoediae clare eluceat.
"Non tamen negaro, tragoedias ejusmodi aliquid ex natura comoediae trahere.
Natura enim tragoediae est luctuosa; quod negari nequit; cum terror, et misericordia, ei inprimis proponantur. Aliter cum fit, datur aliquid, u t diximus populi
voluptati. Unde Aristoteles sit, a poetis fieri id 6th rfiv T S V e~6Lrpwvhufliv~tav,hoc
est, ob injirmum, et imbecille eorum jztdicizbm, qui in theatro spectant. Sed populus
varius est, & inconstans: nunc hoc probat, nunc illud: eoque satius erit, si naturam
tragoediae retineamus penitus; nec, nisi gravis caussa justerit, ab ea recedamus."
Ibid., 11,xxii: "De Comoedia; ac primum de vocabulo, definitione, discrimine
ejus a tragoedia, ac fine. . . .
"Comoedia est poema dramaticum, ciuium, ac vulgi actiones stylo populari
A . Philip IlfcMahon
comedy, but Vossius decides that the only thing that really matters is
the prevailing mood.'
The influence of Heinsius and Renaissance scholarly criticism is to

be seen not only in Kacine, as noted above, but also in C~rneille,~
who
imitans, non sine salibus, ac jocis. Quod si rationem habeamus Comoedixe, quae
obtinet: dicere possis, imitari actiones non civiles modo, sed etiam privatas.
"Comoedia a Graecis i t s definitur, ut Diomedes, et Donatus, referunt: I i w p y 6 l a
i u r l v i 6 i w ~ i x G i ,r r p a y p & ? w v iihiv6u~osX L P L O X ~ ~ .Quorum mentem ita expressit Diomedes: Comoedia cst privatae, cbilisque fortunac, sine pcriculo vitae, conzprehenrio.
E a in definitione duo scrupulum injiciunt. Unum est, quod dicitur comoedia esse
hriv6uvos. Sed eo spectarunt, quod, utcumque periclitentur amantes, et similes;
tamen illud leve sit. Non enim de vita agitur; sed solum, an amica excidant, vel re
simili. Accedit, quod i n catastrophe omnem illam curam sequitur securitas Praeterea culpatur, quod comoedia dicatur esse r r p a y p & ~ w vi8iw71hiju: hoe est, u t Diomedes ait, privatar rivilisque jortzmae. Atque ejus sententiam est Donatus, seu
Euanthius; cui Comoedia estfabz~la,divcrsa instituta continc~zr,ajectuumquechiliurn,
ac privatarum rerum; qua dircitur, quid sit in oita utile, quid contra evitandum. Sed
Julio Scaligero, auia omnis comoediae habere rationem voluit, placuit utrumque
omittere. U t cui definiatur (Lib. I. de poet. cap. 5) pofnza dramaticum, negodiosum,
exitu laetum, stylo populari."
1 Ibid., 11, xxii, 4 : "De discrimine hoc comoediae, et tragoetliae, Diomedes
scribit his verbis: Tragoedia itztroduclrntur dzcces, heroes, reges: i n Comoedia humiles,
atqzre privatae personae. I n illa luclus, e.-csilia, caedes: i n hac amores, virginum raptus.
I n illa frequenter, el pene senzpcr, laetis rebtrs cxitzls tristes, et liberorunt,fortunarzcmqtre
priortrm, i n peius agnitio: i n lzac tristibus laetiora succedzmt. Quare oaria definitione
:
$pwir+j~7 6 ~ 7 s~ t p i ~ r i a udicta.
~s
discrclae sunt. Altera e n i m iixivduvos r t p ~ o x f i altera
Donatus, sive Eugraphius, de eodem hoc pacto: Ilz comoedia nzediocres fovlzrnae
honzinzcm, paroi impetzcs, periculaqzcc, laetiqzie snrnt exitus actionzcm, A t in lragoedia
omnia contraria; ingeiztes personae, magni timores, exihus fz~nerlih a b c n t ~ ~ rE. t i l l i ~
tt~rbzilenlnprima, Lranqtrilla zclbima. I n tragoedia contrario ordine rer nglmitrr T a m
quod itz tragoedia fugiendn vita, i n comoedin c a p e r s e ~ d amprimittcr. Postremo, quad
omnir comoedia de fictis esl argzr?izentis: trccgoedia saepe ab historica jide petitzcr
''Irerum ex mu!t;plici hoe discrimine illutl inprimis attendere oportet, quotl essential~est, ac quod idcirco in tragoediae tlefinitione Xristotcles exprimendum
putarit. Quale est, quotl tragoedia continent gravem personae gravis actionem, in
qua terror ac mi5ericortlia; Comoed~n%era personae levioris habeat actionem levem,
ac subinde ridiculam. Caetera tliscrimina perpetua non sunt; quale illud, quod ab
exitu petitur "
et
2 1) Corneille, UZzrzlicr, XI1 (i'aris, IPz~),p. 26. Jules Leniaitre (C'orf~cille
la I'oktiqz~ed'Arislote, I'aris, 1888)devotetl a small volume to a superGuou\ task.
I 75
distinguished between the two species in accordance with the theories

of the Roman grammarians.'
Rapin was an influential critic whose work was translated into English in the same year that it appeared in F r a n ~ eHe
. ~ explains purgation
as the regulation of pride and hardness by pity and fear; and he also
1 E. F. Jourdain, Afz Introduction to the French Classical Drama (Oxford, I ~ I Z ) ,
notes the discrepancy between Corneille and Aristotle, but ignores the whole
historical development of theory when she says (pp. 12-13): "The broad distinction between tragedy and comedy Fas perhaps this. I n the thought of the seventeenth century tragedy dealt with ideal conditions, which might be foreign or
ancient, and much symbolism was used in its ex~ressionon the stage. Comedy,
on the other hand, was intended to be a picture of real life; and it is interesting
to see the transition of thought from one to the other. For instance, the ideas of
personal and political liberty, of honour and duty, are seen in their ideal aspects in
the tragedies of Corneille. I n the comedies we see the same ideas struggling for
expression in faulty natures and everyday surroundings." The distortion of values
and misinterpretation of classical theory and practice in her work are clearly due
to the influence of Brunetisre, whom she cites (pp. 18-19): "The problems presented in seventeenth century French drama may all be described as problcms of
the will in relation to reason and action. . . . And i t was his observation of
seventeenth-century drama that led Brunetisre to formulate his Law of the Drama,
by which this genre can be distinguished from the epic or novel."
RenC IZapin, REjlexions sur l a Poitique d'ilristote, et sur les O ~ v r a g e sdes Podtes
Anciens et Modernes (Paris, 1674); Thomas Rymer, Rejlections o n Arislotle's Treatise
of Poesie, Containing the Necessary, Rational, and Universal R u l e s f o r Epick, Dramatick, and other Sorts of Poetry. . .. by R . R a p i n (London, 1674).
The ethical value of drama was, however, seriously questioned, as for instance,
in C. Desprez de Boissy, Leltres szrr les Spectccles; avec une Histaire des Ouorages
pour et conire lcs Tht'atres (2 vols., 6th ed., Paris, 1777), I T , p 1 2 : "Le Th6atre
comique ne devint pas moins nuisible aux moeurs que le tragique. On en fit un
recueil de stratagkmes, pour faire rCussir tous les crimes, favoriser toutcs lcs passions, mknager toutes lcs intrigues, traverser tous les peres, maris, maitres, exciter
l'amour du libertinage, & le faciliter par le jeu infame dcs valets, des soubrcttcs &
des confidens, qui furcnt toujours dans la ComCdie lcs rBles les plus intkressans.
Zbid., I, p. 94: "Je n'ai jamais entcndu, dit &I.de Fontenelie 2 ce sujet, la purgation
des passions par le moyen des passions memes." 36id., p. 11, 385 (quoting SaintEvremond): "C'est inutilement qu'on y opposeroit la Doctrine la plus sainte, les
actions les plus chrktiennes, & les vCritCs les plus utiles pour produire cette purgation qu' Arislole avoit eu la simplicit6 d'admettre comme un remede propre i
arr&terles mauvaises impressions des Poemes 1)ramatiques. Ce RhCteur Philosophe
est i cet Cgard en dCfaut; car y a-t-il rien de si ridicule que de se former une science
176
A . Philip McMalzo?z
ascribes a similar moral effect to comedy, citing Cicero.' The popularity of Rapin and Rymer continued in England for many years, so
that this work is one of the chief sources for the ethical interpretation of
purgation in tragedy and a moralistic analysis of ~ o m e d y . ~
DACIER
Another French critic, -4ndr6 Dacier,* attacks Corneille for permitting the introduction of royalty into comedy and considers tragiqui donne sfirement une maladie qui travaille incertainement 2 la gukrison d'une
autre? y a-t-il rien de si ridicule que de mettre la perturbation dans une ame pour
ticher aprss de la calmer par des r6flexions clu'on lui fait faire sur le honteus 6tat
oil on l'a mise? "
Rapin, G z ~ s r e s11, (Amsterdam, r;og), xxv, p. 173.
"Basil
Kennet, T h e 14'hole C r i t i ~ i ~Itrorks
l
of Monr. Rapz'fc (3d. ed , 2
vols ,
T,ondon, 1731). [Preface is dated 1705.1 New title, 11, p 107: H i s RcjEectiorzs o n
rlrisfotle's I'reatisc of Poesy; z~bith a large Preface by flfr. R y m e r , Chapter xvii
(pp. zo4-20;). "Tiagedy, of a11 Parts of I'oesy, is that which :\ristotle has most
discuss'd; and where he appears most exact He alledges that Tragedy is a publick
Lecture, without comparison more instructive than Philosophy; because it teaches
the Mind by the Sense, and rectifies the Passions by the Passions themselves, calming by their Emotions, the Troubles they excite in the Heart. The Philosopher had
observ'd two important Faults in >Ian to be regulated, Pride and Hardness of
Heart, and he found, for both Vices, a Cure in Tragedy. For i t makes 3Tan modest,
by representing the great Masters of the Earth humbled; and it makes him tender
and merciful, by shewing him on the Theatre the strange Accidents of Life, and the
unforeseen Disgraces to v,hich the most important Person? are subject But because Man is naturally timorous and compassionate, he may fall into another
Estreme, to be either too fearful, or too full of Pity; the too much Fear may shake
the Constancy of hfind, and the too great Compassion may enfeeble the Equity.
'Tis the Business of Tragedy to regulate these two XYeaknesse5; i t prepares and
arms him against Disgraces, by shening them so frequent in the most considerable
Persons; and he shall cease to fear ordinary Accidrnts, when he sees such estraordinary happen to the highest Part of &fankind. Hut as the 7:.nd of Tragedy is
to teach Alcn not to fear too \veakly the common Misfortunes, and manage their
Fear; it makes account also to teach them to spare their Compassion, for Objects
that deserve i t not."
Chapter xsv (I). 219) : "Comedy is an image of common Life; its End is to
shew, on the Stage, the Faults of Particulars, in order to amend the Faults of tlie
I'ublick, and to correct the I'eople thro' a Fear of being render'd ridiculous. So that
which is most proper to excite Laughter, is that which is most essential to Comedy."
XntlrC I)acier, La Po&iiq~~-le
d'zlristoie, Contenant les Rdglcs l~csPlzrs Eracles pour
Jzqer dl6 I'oe~ne II&rozpt~e,el des PiPces de Thkcitre, l a Trczgtfdie el In Combdie, Tra-
177
comedy as really comedy.' I n his exposition of the definition of tragedy

in the Poetics, he stresses the formal elements and neglects p ~ r g a t i o n . ~
He wisely remarks later that the extirpation of pity and terror, desired
by the Academics and Stoics, is beyond the capacity of tragedy, but he
agrees with the Peripatetics that it is not the presence of these emotions
in the soul that is evil, but their excess, so that the drama may seek to
reduce them to proper proportions and moderate them. The moral
benefit is derived from the sympathy of the spectator with the subjects
represented and an increased ability to face calamities with a reasonable estimate of their importance. To support this view he quotes
Marcus A ~ r e l i u s . ~
LuziC~
LuzLn, writing in Spain in the eighteenth century, after giving the
definition of tragedy from the Poetics, undertakes to give one of his
own, but he holds that its moral effect is restricted, in the main, to
royalty.4 But comedy deals with common life and is of general benefit.5
duite en Fran~oisavec des Remarques Critiques (Paris, 1692,1698; Amsterdam, 1692,
1717, 1733).
Pottique (1692), p. 60: "Quand Plaute fit son Amphitryon, oh il introduit des
Roys & des Dieux, il l'appella en plaisantant TragicocomCdie, mais c'est pourtant
une veritable Comedie, oh il tourne en ridicule un sujet tragique, & voila de quelle
manisre seulement la Comedie peut prendre ses sujets dans les actions des Roys
et des Heros. Le ridicule doit toOjours Ctre le caractere de ce Poeme, et une marque
certaine, que c'est la Nature m&mequi a fait ce partage, c'est que toutes nos pieces
qui sont conformes i cette dCfinition, rkiississent tobjours mieux que les autres que
nous ne voyons jamais sur nos thtLtres qu'avec un ennuy mortel."
Ibid., p. 72.
Ibid-, p. 81.
Ignacio de LuzAn, La Poetica 6 Reglas de la Poesia (1st ed. Saragossa, 1737)
(2 vols., Madrid, 1789), 11, p. 77: "En tanto, en gracia de 10s que no entendieren
bien la definicion de Arist6teles, que es algo obscura, seame permitido proponer
aqui otra mas clara 5 mi entender, y mas inteligible, como asimismo mas adaptada
6 10s dramas modernos. Pareceme pues que se podria decir 'que la Tragedia es
una representacion dramhtica de una grande mudanza de fortuna acaecida A Reyes,
Principes y personages de gran calidad y dignidad, cuyas caldas, muertes, desgracias y peligros exciten terror y compasion en 10s Animos del auditorio, y 10s
curen y purgen de estas y otras pasiones, sirviendo de exemplo y escarmiento 5, todos,
pero especialmente fi 10s Reyes, y A las personas de mayor autoridad y poder."'
Ibid., p. 225 "La Comedia, pues, 5 mi parecer, como quiera que otros la definan,
es una representacion dramfitica de un hecho particular, y de un enredo de poca im-
A . Philip McMalzon
Gons&lez de Salas, another Spanish author of the same period, explicitly declares that he relies mainly on Heinsius.'
I n the Jesuit manuals of studies, the traditional definitions are continued down through the nineteenth century. The definition of tragedy
from the Poetics is given ior the sake of its ethical ihterpretati~n,~
and
the distinctions between the species are established in the manner of the
Roman grarnrnarian~.~
portancia para el pfiblico, el qua1 hecho 6 enredo se finja haber sucedido entre personas particulares 6 plebeyas con fin alegre y regocijado: y que todo sea dirigido f~
utilidad y entretenimiento del auditorio, inspirando insensiblemente amor B la
virtud y aversion a1 vicio, por medio de lo amable y feliz de aquella, y de lo ridfculo
6 infeliz de esto."
1 Jusepe Antonio Gong&lezde Salas (Xuena Idea de la Trogedia Antigzla o Ilzistracion Ultiw~aa1 libro singular De Poetica de Aristoteles Stogiritn, &!adrid, 1633)~
p 9, note 4, cites Aristotle, "Caput 4. Edit. Heinsii, 'quam perpetuo sequimur."'
2 P. G. F. Le Jay, e Societate Jesu., Bibliothecam Rketorum Praeceptc~el Exempla
complectentem, quae ad oratoriam et poeticam fac~rltatem pfrtinent, et cet , ed. J. A.
Amar (Paris, 1809) 11, p 3 1 . Tragoedza deJinitur ab Aristolcle (Poet c. 6 ) . . .
"Denique, per misericordiam ac metum animum purgat ac levat a b ejusmodi per
turbationibus. Designatum hic habcs Tragoediae finem, quae res atroces exhibendas suscipit, u t pravis animorum motibus medeatur. . . ." Ibid., 11, p. 46:
" Haec pcrro durn subjicit oculis gravissimas calamitates, in quas illustres viri,
errore lapsi potius quam culpa sua: incidere, misericordiam et metum naturaliter
in nobis excitat, eosdemque interim reprimit motus et coercet. Docet enim non
mod0 ferre praesentia, quae leviora sunt gravibus malis quorum miserescimus, sed
animum praeparat ad similes casus, si forte contigerint, constanter tolerantlos."
3 Ibid., p. 51: "Uicitur primo Imitatio actionis communis, per quotl Comoetlia
a Tragoedia distinguitur: neque enim imitatur res illustres ac terribiles, ut Tragoedia, neque admirabiles ac prodigiosas; non luctus, non exilia, non caedes, sed
civiles et privatas actiones, u t Juvenum curas, avaritiam Senum, fraudes Servorum;
et similes, quae alios solicitudine, alios laetitia afficiant. Neque etiam Principes,
Keges, Heroas, sed tenuiores atque humiles personas in scenam inducit.
"Dicitur secundo repraesentare vitae privatae imaginem non sine salibus et
jocis, qua in re multum etiam differt a Tragoedia. Haec enim gravissimas animorum
perturbationes, quales sunt misericordia ac metus, movet: illa sectatur unam oblec-
Aristotelian Dejiliitions of Tragedy and Covzedy

ATTACKSON
THE
I 79
THEATRE
An attack against the theatre as a public institution with a bad

effect on public morals was made by the early Christian fathers in the
third and fourth centuries. Again in the sixteenth and the seventeenth
centuryJ1throughout Europe; theatrical productions, especially comedies, were condemned. Together with the opposition to the stage, in
the latter period, there was wide opposition to the continued use of
certain classical texts in educational circle^.^ Comedies were easily
disapproved as fit reading for the young, but even tragedies were den ~ u n c e d .This
~
feeling was not confined to the Puritans in England,
tationem, quam ridendo parit. Denique Comoediae finis est humanos mores nosse,
describere, et describendo corrigere, ne, quod in illis vitiosum est, privatorum honlinum familiis nocere possit. Tragoedia etsi hoc proponit sibi pariter, ut mores
instituat, tamen nec eadem id perficit via, nec mediae tantum conditiones homines
respicit, sed ad ipsos etiam Principes ac Keges assurgit."
Cf. J. D. Wilson, "The Puritan Attack Upon the Stage," in T h e Cambridge
,
11, pp. 421 ff.
IIistory of Englislz Literature, VI ( I ~ I O )Part
Cf. in addition to the bibliography in Wilson, op. cit.; Emilio Cotarelo y Mori,
Bibliografia de las Controversias sobre la Licitud del Teatro en Esparia (Madrid,
1904), (739 pages, 213 authorities cited; pp. 619 ff., report legislation against
theatre from year 1534 to date of publication).
Cf. Franz Freiherr von Lipperheide, Spruchworterbuch, (Berlin, 1go7), P. 461,
s. v. Komodien: "-Und Christen sollen Comodien nicht ganz und gar fliehen,
darum, dass bisweilen grobe Zoten und Buhlerei darinnen seien, da man doch um
derselben willen auch die Bibel nicht durfte lesen. (Dr. Martin Luther, Tischreden
oder Colloquia (1566). Nr. 68. Tischreden u. Studien. 2. B. Comodien.)" The question is referred to in Thomas Crenius (pseudonym of Thos. Theo. Crusius), De Philologia, Studiis Iiberalis doctrinae, informatione et educatione Litteraria generosorum
adolescentum, et cet. . . . Tractatus (Lugduni in Batavis, 1696), pp. 252-253: "Jam
etiam post Terentium perlectum, Plautus evolvi potest, ea adhibita cautione, ne
omnia ibi reperta, sibi imitanda, imo multa sibi solicite fugienda ducat adolescens.
Nam ne repetam, quod ante jam monui, & cuivis eruditio satis notum est, quod
dissolutos seculi sui, praecipue autem juventutis gentilis mores, Comici isti describunt, a quibus Christiana pubes quam maxime abhorrere debet: etiam sermo
eorum, inprimis IJlauti, nonnisi judiciosa imitatione efingendus est juventuti."
Cf. Antonii Possevini, Mantvani Societatis Jesu, Ribliotheca Selecta de
Katione Studiorz~m (Coloniae Agrippinae, 1607)~ 11, p. 430: "In Tragicis, &
Comicis Dionysius ipse Lambinus, quae moribus adversari possent, haud dissimulavit. Itaque ille sic: Veniamus ad Tragicos, q u i o m n i z ~ mmanibus teruntur, q u i
assidue leguntur, qui ediscuntur. P r i m u m i f s a tragoediarum argumenta partim sunt
Aristotelian DefLnitions of Tragedy and Comedy
181
Lexicography is not only an influential but a very conservative

science, and while dictionaries and encyclopedias were slow to admit
any significance from the Poetics for the words tragedy and comedy, to
this day they emphasize the traditional meanings accepted by classical
antiquity, the Middle Ages, and the early Renaissance.
The lexicon of Firmin Le Ver, prior of the Carthusians of Saint
Honor&-les-Abbeville,was composed between 1420 and 1440. Its mediaeval foundations are obvious in its definition of tragedy.'
The commentaries of Budaeus on the Greek language, published in
Paris in 1529, appear to have been embodied in the dictionaries of
chapter headings indicate the nature of the contents: IT. De idolatria scenae.
VIII. Abominationes. XII. V i t i a poetarum. XIII. Sententia theologorum de histrionibus. XIV. Sententia jurisconsultorum. XIX. Sententia philosophorum. XX.
Voluptates non semper damnari. That Erasmus approved of the drama is apparently
good reason for condemning it; it is also claimed that comedies lead to atheism, and
the fact that Machiavelli himself composed comedies is a significant bit of evidence.
Nevertheless, the refutation of the claim to moral value in comedy is keen (col.
1675): ''Erasmum defendunt: qui nec est nosier, nec est sanus, & est pendulus,
& incertans ubique, & ridens (ita de eo Lutherusque) religionem, vid. Flac. praef.
N. T. Sleid. IX Apage. I n his quae adjidei integritatem pertinent, ediam deterius est
dubium, quam pravum: ita contra Erasmum etiam Cardanus, 111. de S a p . Jocos
Erasmi defenditis? E t sancta per jocum doceantur? discantur? dedoceantur? dediscantur? Nosque ita monent Apostoli, & Prophetae? Acerbissima ubique, et fere
omnia sacrorum sunt librorum: Imo apagete vos lusores, risores, doctores religionis
jocosi mali histriones. Nihil e n i m est t a m naturale, q u a m eo genere quidque dissolvere, quo colligatum est, ut 1.35. de reg. jur. c. 4. 27. q. I I . E t risus res est plebeja:
ducitque ad vilitatem, Epict. ench. XLIX. Trahit rem risus in suspicionem: u t
artificium sophistae est, facetiis diluere, risu discutere serium, Arist. 111. Rhetor.
Haec via ad atheismum prona, praeceps. Haec via (audenter dico) athea magis,
quam est Macchiavelli." Another comprehensive collection of anti-dramatic authority is to be seen in Gronovius, op. cit., cols. 1713 ff.: Job. Ludovici Fabricii. . . .
De Ludis Scenicis.
Lanson, op. cit., p. 5 4 2 : "Tragedia. Oda, quod est cantus, seu laus, componitur cum tragos quod est hircus. E t dicitur hec Tragedia . . . carmen luctuosum
quod incipit a leticia et finit in tristicia. Cui contraria est comedia, quia incipit
a tristicia et finit in leticia. Unde Tragedia dicitur de crudelissimis rebus, sicut qui
patrem seu matrem occidit, seu comedit filium, et e converso s. hujus modi. Unde
et tragedo dabatur hircus animal fetidum. Ad fetorem materie designandurn. Tragoedicus . . . luctuosus, funestus."
182
A. Philip McMaholz
1554.' On the Greek words significant in the definition of tragedy in

the Poetics meny authors are cited, but there are no direct references to
Aristotle's Poetics. The only mention of tragedy is with reference to
Polybius and not to the drama.2
Another early Greek-Latin dictionary, first publishetl in 1548, makes
no reference to the Poetics, although Plato, Polybius, ant1 Erasmus are
mentioned in connection with the words of special ~ignificance.~
A dictionary compiled by Pierre Gilles, and published a t Basel in
1577, directs the student, under the word ~ p u y y d i a , to read Aristotle's Poetics, but apart from this there is no reference to that source
under the words significant in the Poetics,l although he had drawn
upcn the work of several eminent scholars.
A lexicon of great influence was that of Forcellini (1688-1768) in
which, under cowtoediu, Fest. up. Paul. Dinc. and Isid. VIII. Orig. 7, 6,
are cited, and, while classical texts are referred to, the definitions proposed are founded on the Roman grammarians.
The dictionary of the Spanish Academy retains the icleas of the Koman grammarians, without considering the Poetics of Aristotle. Under
the worc! comedia it begins with the traditional use, but asserts that in
Spain it hat1 been extended to all kinds of dramatic productions. The
formula of Theophrastus is quoted and Cicero is cited. For tragedy
there is a short reference to the definition of Dio1nedes.j
The edition of the dictionary of the Accademia della Crusca still in
Italuse, relies mainly on Dante for its definition of c o r n e d ~ .Another
~
C. Budaeus, Conz?nerztarii Ligzguae Graecae. Venundantur Iodoco Badio Ascensio (Paris, 1529). Cf. Lanson, op. cit , p. 573. The dictionaries are: "1,exicon
Craeco-Latinurn, seu Thesaurus Graecae Linguae," Jean Crispin, 1554, with a
Charles Estienne, 1554.
preface by Claude Baduel; Uiclionariu?n Grcecc-Lati+~zl?jt,
Commentarii, p. 714.
J u n i i IIornafzi Ifadri(~?zi,L c ~ i c o n sive Diciio+zariz~+n,
Graeco-Latinz~nz (Basel,
s
~pa?yFia.
1557)~s v. ~ & B a p o ~and

I'ierre Gilles. AE?II<OW. IIoc cst Dictionariir?~~
Graccolalin~~?n,
S u p r a O?ilncs
L . Tuscanz~~rz,
G. Gcsnerum,
Ediliones in hoc A n n o cz: Variis. . . . G. Bz~daczl?i~,
fI. Izmzunz, R Constantizl?n, l o . Ilartingtlm, M a r . IIoppcn~?n,Gizrl X y l a ~ i d r u n ~
(Basel, 1577).
Diccionario dc la Lerzgua Castclla?za. Con;pl~estopor la Real Acaden:ia Espaliola
(Madrid, 1729) s. v. comedia and tragedia.
6 Lrocabolario degli Accadctizici d d l a Crusca.
Quinta impressione (T;lorence,
1875), p. 204, s. v. "Commcdia. Sorta d i componinzento dranznzaiico; i n prosa o anche
183
ian dictionary still current is also content with mediaeval sources for
tragedy.' The Dictiotz~zairede l'Acad6mie Fran~aise,(Paris 1878, 7th
ed.) agrees word for word with the Nouveau Dictiolrnoire National de la
Langue Fralz~aise,par Bescherelle AinC (Paris, 1887), and the definitions common to both are the traditional o n e s . V n English, Richardson's dictionary refers the reader to V o s ~ i u s . ~
Without pursuing this line of research further, it may be said that
practically the whole weight of lexicographical authority, from the be-
i n verso, nel quale s i rappresentano per lo pizi fatti e personaggi della vita privata. H a
esito quasi se~nprelieto, e intende, mediante i l ridicolo, a correggere i vizj e i dqetti degli
uomini. . . . Segn. B. Poet. volg. 288: La commedia . . . 5 una imitazion di cose cattive; ma non gii che abbino il sommo grado della cattivitii: ma & una imitazione di
quella parte ridiculx, che contien la brutezza. . . . I. Consmedia dicevasi, second0 la
opinione espressa da Dante nel libro De vulgari eloquio, Qualunque componimento i n
lingua volgare, i n quanto che questa non si credeva atta a trattare se non soggetfliumili o
mezzani."
3 Dizionario della Lingua Italialza.
Nuovamente compilato dai signori Nicoli,
Tommaseo e Bernardo Bellini. (Turin, 1879), p. 1535. S.V."Tragedia. Poemarappresentative, che 1 imitazione di azione grande, fatta da personaggi illustri, con parlar
grave, e che ha dolorosa catastrofe. . . . But. Purg. 2 2 . 2. Tragedia i: canto in sublimo
stilo, e tratta de' principi, ed ha felice principio, ed infelice fine. Dante I n f . 20. Euripolo ebbe nome, e cosi'l canta L'alta mia Tragedia in alcun loco. .. . But. iui: Dice
Virgilio, chela sua Eneide 5 alta tragedia. Questo finge Dante per dimostrare che in
alto stilo & fatta, e che si d?e chiamare tragedia, perche tratta de' fatti dei principi, e
incomincia dalle cose liete, e finisce nelle triste e avverse. D. Volg. Eloq. 2.4. Per la
tragedia intendo lo stilo superiore; per la commedia l'inferiore. Galat. 25. Per tal
cagione egli affermava essere state da principio trovate le dolorose favole, che si chiamaron tragedie, acciocchi?raccontate, ne' teatri, come in quel tempo si costumava di
fare, tirassero le lagrime agli occhi di coloro che avevano di cii, mestiere. Com. Boea.
2 . Nota che tragedia sono quelli, li quali scivono le geste luttuose delli re; onde tragedia 5 verso di grandi iniquitadi, incominciante da prosperitate, ed in adversitate
terminante."
"ComCdie. (Euvre dramatique, pi&cede t h t i t r e dans laquelle on reprksente
une action de la vie commune, et qui peint d'une mani6re plaisante les moeurs, les
difauts ou les ridicules des hommes.
"TragCdie. P i k e de thCLtre qui offre une action importante, des personnages
illustres; qui est propre 2 exciter la terreur ou la pitie, et qui se termine ordinairement par un CvCnement funeste."
A New Dictionary of the English Language, by Charles Richardson, London.
1838 (s.v. tragedy).
184
ginning until our own times, has emphasized the traditional definitions
derived from O n Poets.
I t appears, then, that the definitions of the grammarians were not
immediately supplanted when the Poetics was recovered, but that the
traditional formulas continued to occupy men's minds. Until a recent
date, the printed editions of classical authors carried the same material
that had been popular in the Middle Ages, and even to-day dictionaries carry on the ideas derived from O n Poets.
VII
W h a t dejitzitions of tragedy and conzedy dominated the ideas of England
duritzg and after the Elizabethan Age?
In Elizabethan England also there is evidence that when a reliable
definition of tragedy or comedy was required, the same traditional formulas that prevailed elsewhere were relied upon. The evidence, to be
sure, is scanty, but no more so than critical discussion in general during
the period when drama flourished most actively, so that it is sufficiently representative. I n the earlier phases, English thought depended
on mediaeval sources, supported by Cicero and justified by its reading
of Seneca and Terence. I n later times it carried on the same principles
under the guidance of Sca1iger.l
Douglas had already called Dido's story a tragedy, in 1513; and mediaeval plays in the vernacular were often termed comedies. When
references to classical drama began, they were accompanied by the
moralistic defense, as in Sir Thomas Elyot, for in speaking of " Therence and others that were writers of comedies," he said that the mirror
of life which comedy presents does not instruct in wickedness but rather
serves as a warning to spectators."
Cf. Henry Peacham (Spingarn, Critical Essays of the Seoentecnth Century,
Oxford, 1908, I, p 128) who, in 1622 said: "Thus haue I, in briefe, comprised for
your behoofe the large censure of the best of Latine Poets, as it is copiously deliuered by the Prince of all learning and Iudge of iudgements, the diuine Iul. Cues.
Scaliger."
Douglas, Aeneis, IV, Prol. 264.
Sir Thomas Elyot, The Boke Named the Gouernour, ed. H . H . S. Croft (2 vols.,
London, 1880), I, p. 124.
185
The metaphorical use, exemplified throughout the preceding ages

and founded on the traditional formula rather than on the definition in
the Poetics, had continued in English literature also.'
Mirror for Magistrates

The meaning accepted by Lydgate was carried on in another influential series of disastrous narratives, The Mirror for Magistrates.
Tragedy is applied to the fate of the two sons of King E d ~ a r dand
, ~ in
~ influence of this compilathe Induction it is similarly e m p l ~ y e d .The
tion on the poetry and drama of the Elizabethan Age is well recognized.
Galateo
According to Symmes, the first record in English of the idea of tragic
purgation is to be found in the translation of the Galateo. In this work
it is asserted that man has better cause to weep than to laugh, and
therefore tragedies were devised, that "They might draw fourth
teares out of their eyes that had neede to spend them. And so they
were by their weeping healed of their infirmities." * But it was long before anything more clear than this obscure reference to the purgation
became general in English criticism, and the traditional formula has
1 NED S.V.tragedy. 1598-99 [ E. Forde] Parismus I (1661) 68: "I fear he is destroyed by the treachery of that wicked homocide . . . who is not contented with his
tragedy, but also seeketh my destruction." 1535, Laytan in Lett. Suppress. Monasteries (Camden), 76: "To tell yowe all this commodie, but for thabbot a tragedie, hit
were to long."
Mirror for Magistrates, edited by Joseph Haslewood (3 vols., London, 1815),
I, p. xi.
Ibid., I, pp. 16-17:
For some, perdy, were Kinges of highe estate,
And some were Dukes, and came of regal1 race:
Some Princes, Lordes, and Tudges greate that sate
I n councell still, decreeing euery case.
Some other Knightes, that vices did imbrace,
Some Gentlemen, som poore exalted hye:
Yet euery one, had playde his tragedye.
EI. S. Symmes, Les D C ~ Ude~ Sla Critique Dramatique en Angleterre (Thesis)
Paris, 1903, P. 46. (Giovanni della Casa, Galateo of Manners and Bekauiours, Englished by Robert Peterson, London, 1576, p. 31.)
156
A . Philip McMahon
not yet been supplanted in ordinary usage. For this there seem to be
several reasons, of which the following may be mentioned: (I) A reliable tradition founded ultimately on Aristotle's dialogue, known
through the Roman grammarians, had already been accepted in England, and the Poetics introduced later did not provide a sufficiently
strong motive for abandoning it. (2) Whatever the Poetics might
suggest regarding a moral justification of poetry, which could be considered an intrinsic element of the definition of tragedy, there was no
clear parallel statement for comedy, so that critics were obliged to rely
on the traditional theory of comedy in any case. (3) Where certain
erudite men of letters and critics did expound tragedy in terms of the
Poetics, there was such difference of opinion on details, and such a confusion as to its real meaning and function in a definition of the species,
that it never prevailed as against the authoritative and accepted
formula.'
1 I n the Ragionamento di M. Agnolo Segni, Gentilhz~omoFiorentino, Sopra le
Cose Pertifzenti a11a Poetica (Florence, 1581), the difficulties of the katharsis clause
as an element in the definition of tragedy had been clearly seen and stated in a way
that is worth remembering now. Cf. pp. 47 ff.: "Nora la purgazione dlAristotile
come si debbe intendere, questo hP dificult2 non piccola. E' non E dubbio, che
second0 lui la Tragedia ci empie di passioni, di misericordia, & di timore: & questo
6 il primo fine, nel q u d e con Platone conuiene: ma non si ferma, & ne troua un'
altro pi& innanzi, il quale ? la purgazione degli affanni mediante que' due, & questo
? l'ultimo fine. Tutto questo ? chiaro & indubitato della Tragedia, ma non s'intende
di q u d i affetti sia quella purgazione, ni: in chc mod0 ells si faccia. Alcuni dicono,
che la purgazione i: de' medesimi affetti misericordia, & timore, si che la Tragedia
de' medesimi empia prima l'animo nostro, Sr poi lo voti: & hanno loro ragioni.
Altri, che la purgazione sia pur de' medesimi misericordia, & timore, ma non in
tutto estirpazione, ma moderazione: che la Tragedia modera, & diminuisce in noi
questi due affetti, & mediante questi gli altri simili h loro. N6 l'una, n? l'altra interpretazione si pui, accettare per le ragioni, che vdirete: ma prima vi voglio dire
come intende breuemente tutta questa purgazione. hfediante la misericordia, e'l
timore si fa purgazione in noi, dice Aristotile, d'dtri afietti, di quali? di qucgli,
dico, che sono contrarij 2 que' due: & che 2 loro siano altri contrarij, che non possono con loro stare insieme, i. manifesto in Aristotile nella Kettorica: & che l'uno
affetto cacci l'dtro, egli meclesimo lo manifesta nel medesimo libro. h l a diccndo
lui (di tali affetti) vuol dire d'dtrui simili 2 questi, simili, perche tutti sono affetti,
et passioni dell' appetito: che se hauesse inteso i due nominari, harebbe detto di
questi, & non di tali. I-lora mostreremo, che Aristotile non poteua intendere i due
affetti misericordia, et timore, che la purgazione sia di questi ni. nel primo modo,
187
In 1580 Thomas Lodge published his Defense of Plays against the attacks of Gosson. He asks: "What made Erasmus labor in Euripides
tragedies?" l On this the editor of the important Elizabethan critical
essays has a note which suggests Lodge's source in a characteristic edition of Erasmus's tran~lation.~Lodge cites Donatus and Iodocus
Badius for the etymology and origin of tragedy and comedy, and refers
to the "sower fortune of many exiles, the miserable fall of hapless princes" in traditional v e h 3 A little later he quotes Cicero's definition of
comedy.4
Gosson challenged Lodge to locate this definition in the text of
comments: "The fact was that Lodge had found the
C i ~ e r o .Klein
~
definition in Donatus. The strangest thing about it is that such a
thorough scholar as Jonson should have attributed the phrase to Cicero
twenty years later." I t would have been stranger yet if he had failed
n& nel secondo delle due interpretazioni predette." p. 50: "10ho sempre inteso &
sperimentato, che il fare qualunque cosa pib volte, & l'auuezzarsi 2 fare i: causa,
che poi si ritorna a1 somigliante, che il fare insegna fare, & si fa venire dietro sempre
il medesimo. A questo Assioma fermissimo, & verissimo contradice quella opinione,
che vuole, che auuezzandoci noi nella Tragedia 2 piangere, poi non piangiamo, b
moderatamente piangiamo secondo l'altra, & che'l timore ci faccia sicuri & arditi.
hlale V. A. il pianto asciugherj gli occhi, il terrore assicurerj, & l'intenerire spesso ci
pot& indurare; anzi tutto il contrario interuerrk della Tragedia. Per quella opinione rouinerebbe tutta la dottrina d'Aristot. morale, che sempre dice, che gli
huomini col fare le cose giuste diuentano giusti, & poi di nuouo fanno le medesime
cose giuste meglio che prima, & cosi in tutte le cose."
G. Gregory Smith, Elizabethar~Critical Essays (z vols., Oxford, rgoq), I , p. 68.
Ibid , p. 366; " E r a s m u s 'interpreted' or translated Hecuba and Iphigenia.
Lodge's reference to these, to Buchanan, and to Donatus suggests the idea that he
was familiar with a popular edition of T r a g ~ d i a eselectae issued by Henri Estienne,
printer to Huldrich Fugger (I j67, &c.), which contains the interpretations of Hecuba
and Iplzigenia by Erasmus (pp. 115-II~),the tract by Donatus De Tragoedia et
Comoedia (pp. 118-128), the interpretation of the Medea and Alcestis (pp. 129-133)
and of the A j a x , Antigone, and Electra of Sophocles, by Georgius liotallerus."
Ibid., I , p. 80.
Ibid., I, p. 81.
I n Playes Confuted (Nov., 1579). Cf. G. Gregory Smith, op. cit., I, pp. 369-370.
David idlein, Literary Criticism from the Elizabethan Dra~?zatists(New Uork,
A . Philip McMahon
to use the phrase and to attribute it to Cicero. We have already seen
that the words are paralleled in Cicero's extant works, if not in this
identical grammatical form. Modern classical scholars accept the formula and assign it a definite position in collected fragments. The
weight of tradition for about 2 0 0 0 years was thus behind Lodge, both
as to the form and content of his quotation from Cicero.
Sir Philip Sidney appears to uphold the Ciceronian definition also,

interpreting it in the manner characteristic of criticism until a recent
date.' He also attempts to refute the example from Plutarch, oi the
tyrant Alexander Pheraeus, used by writers who attacked the stage, as
showing that a man could be emotionally affected by tragedy, and yet
remain a bloodthirsty criminaL2
A passage in Kyd's Spanish Tragedy reaffirms the traditional concept i ~ n .Webbe,

~
again, in his Discourse of English Poetry (1586) repeats
1 G. Gregory Smith, op. ril., I, p. 176-177: "Onely thus much now is to be said,
t h a t the Comedy is a n imitation of the common errors of our life, which he representeth in the most ridiculous and scornefull sort that may be; so as i t is impossible
t h a t any beholder can be content to be such a one."
"l3ut how much i t can mooue, Plzctarch yeeldeth a
2 Ibid., I , pp 177-175:
notable testimonie of the abhominable Tyrant Alexaflder Phcracus; from whose
eyes a Tragedy, we1 made and represented, drewe aboundance of teares, who, witho u t all pitty, had murthered infinite nombers, and some of his owne blood. So as
he, that was not ashamed t o make matters for Tragedies, yet coulde not resist the
sweet violence of a Tragedie. And if i t wrought no further good in him, i t was t h a t
he, in despight of himselfe, withdrewe himselfe from harkening to that which might
mollifie his hardened heart." (Plutarch, I i j c cf Pclopidns, 29.)
Klein, op. rit., p. 22:
Hallhazer. Hieronimo, methinks a comedy were better.
IIier.
A comedy! fie! comedies are fit for common wits;
I3ut to present a kingly troop withal,
Give me a stately written tragedy;
Tragoedia cothurnata, fitting kings,
Containing matter, and not common things.
189
the familiar distinctions between tragedy and comedy.' These were not
merely iischolastic," as termed by Klein; they originated in the very
source from which scholastic philosophy gained i n s p i r a t i ~ n .Webbe's
~
account of the etymology and origin of the dramatic species is directly
from D o n a t ~ s . ~
I n Shakespeare, likewise, there is no divergence from the traditional
and reliable definitions. Bosanquet notes his acceptance of the ideas:
Coming upon the arena thus prepared for him, Shakespeare adopts a distinctly traditional dramatic form. He accepts the complicated organic structure of Latin comedy, with its five acts and separate scenes. He is more
careful than his crude predecessors to motive or excuse his violation of the
unities. He observes, except in the histories, with hardly any deviation, the
sharp distinction between tragedy and comedy which Dante applied so
strangely. That is to say, in the plays of which the catastrophe is not tragic,
the happy ending or reconciliation is absolutely complete, and no irrevocable misfortune befalls any character in the play.
Bosanquet's observation is accurate, with the exception of the word
"strangely" applied to Dante's use. I n that particular he is obviously
influenced by Hegel, Bernays, and other moderns rather than by the unbroken, authoritative tradition that extends from Aristotle's dialogue
Ofz Poets down to the popular usage of to-day.
I n Love's Labour's Lost (V, ii, 515)'~Shakespeare refers to Christmas
plays as comedies; in the same play (V, ii, 950-952) he speaks of the
G. Gregory Smith, op. cit., p. 249; William Webbe, Of English Poetry (1586);
"There grewe a t last to be a greater diuersitye betweene Tragedy wryters and
Comedy wryters, the one expressing onely sorrowful1 and lamentable Hystories,
bringing in the persons of Gods and Goddesses, Kynges and Queenes, and great
states, whose partes were cheefely to expresse most miserable calamities and
dreadful1 chaunces, which increased worse and worse, tyll they came to the most
wofull plight that might be deuised. The Comedies, on the other side, were directed
to a contrary ende, which, beginning doubtfully, drewe to some trouble or turmoyle, and by some lucky chaunce alwayes ended to the ioy and appeasement of
all parties."
Klein, op. cit., p. 20.
G. Gregory Smith, op. cit., I, pp. 245-249.
Hosanquet, History of Aesthetic (London, I ~ I O ) p.

, 155.
Furness, Variorum Shakespeare, XIV, p. 272.
Ibid., p. 315.
190
A . Philip McMahom
standard ending of comedy; and in the Midsunznzer Night's Dream

there is stressed the fatal ending of tragedy:
Lis. A tedious breefc Sccnc of yong Piranl~ts,
And his louc Thisby; very tragicall mirth.
The. hlerry and tragicall? Tedious and briefe? That
is, hot ice, and wontlrous strange snow. (V, i, 63-66)
Anti tragicall my noble 1,ord it is: for Pirumus
Therein doth kill himselfc. (V, i, 73-74) '
The most important reference is, however, in fjravzlet 111, ii, 19-2,3,
where the phrase of Cicero receives perpetuation in English Literature:
. . . the purpose of playing, ~\rhoscend, both a t the first and now, was and
is, to hold, as 'twerc: thc mirror up to nature; to show virtue her own fcature, scorn hcr own image, and the very age and body of the time his form
and pressure."
The example of Shakespeare has given pause to many critics, whose
minds are naturally filled with the definition of tragedy found in the
Poetics, but interpreted by the mid-nineteenth century. Some of the
most acute have fairly acknowledged the bearing of Shakespeare's historical background on his conceptions, as for example, A. C. Bradley,
who points out that for the poet tragedy is "essentially a tale of suffering and calamity conducting to death." EIe shows further that i t
conforms to the traditional view: "The suffering and calamity are,
moreover, exceptional. They befall a conspicuous person. They are
themselves of some striking kind."
Again, he repeats: '(Shakespearean tragedy as so far considered may be called a story of exceptional calamity leading to the death of a man in high estate."
The theories of critics were not inconsistent with the reflections of

opinion in the dramatists. In the Arte of Poesie, attributed to Puttenham (1589)) appears the familiar distinction between tragedy and
Ibid., X , pp. 206-207.
Ibid., 111, pp. 227-228.
A. C. Bradley, Shukespereun Tragedy ( L o n d o n , ~ g o g ) p.

, 7.
Ibid., p. 8.
Ibid., p. 11.
ArisLoteliun Dejinitions of Tragedy and Conzedy
191
c0rnedy.l The standard moralistic interpretation or defense is to be

seen in the title of a chapter following that in which the distinction is
~tated.~
Ben Jonson, who protested against the tendency to make Aristotle
a dictator of literary t h e ~ r yaccepted
,~
the traditional definitions and
distinctions. I n the Prologue (1598?) to Ecery Man in his Humour there
is a repetition of Cicero, with moral application, in striking phrases!
I n Every Man out of his Humour, 111,i, there is a vivid paraphrase of
the same idea.5 Shortly following this passage is one in which Jonson
1 G. Gregory Smith, op. cit., 11,p. 27: "There were also Poets that wrote onely
for the stage, I meane playes and interludes, to recreate the people with matters
pageants, accompanied
of disporte, and to that intent did set forth in shewes
with speach, the common behauiours and maner of life of priuate persons, and
such as were the meaner sort of men, and they were called Comicall Poets: . .
Besides those Poets Conzick there were other who serued also the stage, but medled
not with so base matters, for they set forth the doleful1 falles of infortunate 8r
afflicted Princes, & were called Poets Tmgicnll: . . ."
J. Haslewood, Ancient Critical Essays upon English Poets and Poetics (London,
I ~ I I ) I, , p. 25: "HOWvice was afterward reproued by two other maner of poems,
better reformed then the satyre, whereof the first was comedy, the second tragedie."
Spingarn, Critical Essays of the Seventeenth Century, I (Oxford, 1go8), p. 42.
Cf. G. Gregory Smith, op. cit., 11, p. 389; from The tVrorkes of Beniamin
Ionson. Folio 1616. (Bodleian Library. Douce, I, p. 302.) The Prologue may be
dated 1598.
"But deedes, and language, such as men doe vse,
And persons, such as Contoedie would chuse,
IVhen she would shew an Image of the times,
And sport with humane follies, not with crimes,
Except we make 'hem such, by louing still
Our popular errors, when we know th' are ill.
I meane such errors as you'll all confessee,
By laughing a t them, they deseiue no lesse:
IVhich when you heartily doe, there's hope left then,
You, that haue so grac'd monsters, may like men."
[&I
The TVorlzs of Ben Jowson, edited by W . GiBord (London, 1875), 11, p. 17:
"Well, I will scourge those apes,
And to these courteous eyes oppose a mirror,
As large as is the stage whereon we act,
Where they shall see the time's deformity
Anatomized in every nerve and sinew."
challenges "these autumn-judgments" to define comedy better than
Cicero had d0ne.l The five sections of Jonson's Discoveries dealing
with tragedy, instead of being independent errors, as many critics in the
last century have judged, are a literal translation from the Dutch critic
I-Ieinsius, whom we have discussed a b o ~ e . ~
Milton did not deviate from the orthodox doctrine in his views which
are due partly to classical sources but also to the Italian theorist^.^ The
most important passage treating of tragedy is to be found in his preface
to Samso~zAgofzistes, where the famous definition of Aristotle's Poetics
is reproduced and translated: per misericordiam et metum perjiciens
talium ajectuum lustrationem." Spingarn correctly notes that Milton's
discussion points to a reading of Minturno, but he reads into Aristotle,
Minturno, and Milton something that did not occur to many minds
until several centuries later, when he adds: "both Milton and Minturno
clearly perceived that by katharsis Aristotle had reference not to a
moral, but to an emotional effect." Milton did perceive, in common
with previous criticism, that the phrase in Aristotle is best adapted to
purposes of defense against the enemies of the drama.6 He paraphrases
1 Zbid., 11, pp. 108-109: "Cor. You say well, but I would fain hear one of these
autumn-judgements define once, Quid sit conzoedia? if he cannot, let him content
himself with Cicero's definition, till he have strength to propose to himself a better,
who would have comedy to be: imitatio nitae, speczilunt consuctudinis, imago oeritatis;
a thing throughout pleasant and ridiculous, and accomodated to the correction of
manners."
Cf. Maurice Castelain. Ben Jonson, Iliscoveries, A Critical Edition. (Thesis)
Paris, 1907, p. xx.
Cf. Spingarn, Critical Essays, I, pp. 196 and 206.
E~tglishPoems of John Milton (edited by R. C. Browne, revised by 11. Bradley; Oxford, ~ g o z ) 11,
, p. 204.
6 Spingarn, Literary Criticisnz, pp. 80-81.
6 English Poems, 11, p. 205: "This is mentioned to vindicate tragedy from the
small esteem or rather infamy, which in the account of many i t undergoes a t this
day with other common interludes; happening through the poet's error of intermixing comic stuff with tragic sadness and gravity; or introducing trivial and vulgar persons, which by all judicious hath been counted absurd; and brought in
without discretion. corruptly to gratify the people."
Aristotelian Definitions of Tragedy and Conzedy
193
the idea, saying of the emotions that katharsis is "to temper and reduce
them to just measure."
Another document in the controversy over the theatre was produced
by Thomas Heywood. I n the modern edition of his work the editor left
the Greek quotation in the corrupt state due to Heywood's printer,
without suspecting that Donatus was the authority for the definition
of comedy stated, although he is expressly cited a t the beginning of the
passage ."
,4n interesting suggestion in the discussion of the moral effect of
tragedy and comedy is provided by Thomas Shadwell, incidentally,
of course, implying the traditional formula. Tragedies, according to
Shadwell, are of moral benefit particularly to royalty, whereas comedy
possesses ethical value for ordinary people. But teaching royalty is
such a hazardous business that it is wiser to be concerned with ~ o r n e d y . ~
More significant, however, is the reference of Edward Phillips because the preface of his Tlieetrum Poeticrum is so much better than his
Ibid., p. 204.
Thomas IrIeywood, A n Apology for ilctors in Three Books. From the edition
of 1612, compared with that of W. Cartwright. Reprinted for the Shakespeare
Society (London, 1841), p. 49: "Tragedies and Comedies, saith Donatus, had their
beginning a rebus diuinis, from divine sacrifices. They difier thus: in comedies
tnlrhule~zta priwza, tra~zyz~illau ltima: in tragedyes, tranquilla prima, turhzdenta
zdlima: comedies begin in trouble and end in peace; tragedies begin in calmes,
and end in tempest. . . . The deffinition of the comedy, according to the Latins:
a discourse, consisting of divers institutions, comprehending civil1 and domesticke
things, in which is taught what in our lives and manners is to be followed, v h a t to
bee avoyded. The Greekes define i t thus: K w e G i a iarrv LBWTLKDV ~ a POXLTLLOV
i
~ p a y p b r w v& X L V BOYOS X O P O L X ~ Y . Cicero saith a comedy is the imitation of life, the
glasse of custome, and the image of truth."
Thomas Shadwell, Preface to T h e tlumorists, A Comedy (1617) (Spingarn,
Critical Essays, 11,p. I 54): "And in this latter I think Comedy more useful than
Tragedy; because the Vices and Follies in Courts, as they are too tender to be
touch'd, so they concern but a few, whereas the Cheats, Villanies, and troublesome
Follies in the common conversation of the World are of concernment to all the Body
of hIanlrind."
194
A. Philip McMahon
other work that the aid of his uncle, Milton, has been suspected.l The
phrase "passionately sedate and moving" recalls Milton, but the discussion as a whole relies largely on the Roman grammarian^.^
In his Essay of Dramatick P o ~ s i eDryden

,~
presents a definition of the
drama in general which is placed in the mouth of Lisideius, thought to
be Sir Charles Sedley, ant1 this definition is hn interesting combination
of the elements common to both tragedy and comedy, some of which
come from the grammarian^.^ The same speaker suggests that; "Of
that book which Aristotle has left us irepi ~ T j s~ T O L ~ / T L Horace,
K~S
his
Art of Poetry is an excellent comment, and, I believe, restores to
us that Second Book of his concerning Comedy which is wanting
in him."
Among the neo-classic critics Rymer was powerful, a t a later time,

through his translation of Rapin and through his own work in the same
temper. Thus, through his translation, the Ciceronian definition of
Spingarn, Critic01 Essays, IT, p. 3.50.
Ibid., Edward Phillips, Prejace to Theetrum Poetarz&m (1675), 11, p. 269: "Next
to the IIcroic Poe+17 (if not, as some think, equal) is Tragedy, in conduct very different, in heighth of Argument alike, as treating only of the actions and concernments of the most Illustrious Persons, whereas Colnedy sets before us the humours,
converse, and designs of the more ordinary sort of People: the chief parts thereof
are the $BOY 8r aciOor, by which latter is meant that moving and Pathetical manner of expression, xvhich in some respect is to exceed the highest that can be delivered in Heroic Poesie, as being occasioned upon representing to the very life
the unbridled passions of Love, Rage and Ambition, the violent ends or downfalls
of great Princes, the subversion of Kingdoms and Estates, or what ever else can
(be) imagined of funest or Tragical, all which will require a style not ramping,
but passionately sedate & moving. . . ."
"ublished in 1668. Reprinted in Clark, Ez~ropcenTheories, pp. 174 ff.
4 Clark, op. cit., p. 176: " A just and lively image of human nature, representing its passions and humors, and the changes of fortune to which it is subject, for
the delight and instruction of mankind."
5 Clark, op. cit., p. 177.
1
Aristotelian Defifzitions of Tragedy and Comedy
195
comedy and the moralistic explanation of tragic purgation were

again impressed on English readers. Elsewhere he expands his moralistic interpretation of Horace and Ari~totle.~
He uses the Ciceronian
formula as another basis to condem Othello,4 and he disagrees with
Rapin only when the latter claims that the English should have a special capacity for tragedy because of their c r ~ e l t y . ~
An interesting and amusing publication is the grammar attributed to Steele. Its prose is relieved by versified passages, among
1 [Rymer] Reflections on Aristotle's Treatise of Powie Containing the $ecessary,
Rational and Universal Rules for Epick, Dra?izatick, a d the other Sorts of Poetry.
B y R . Rapin (London, 1674), p. 124. Monsieur Rapin's Rejections on Arislotle's
Treatise oj Poesie . . . Made English by M r . Rymer. [In the Whole Critical Works
o j Monr. Rapi~z. Translated by Basil Kennet, and 0thers.j (gd ed., London, 1731),
11, p. 219.
Ibid., pp. 204-205.
Thomas Rymer. Tragedies of the Last Age Consider'd and Examin'd by the
Practice of the Ancients and by the Common Sense of all Agw. (London, 1678), p. 140:
"Some would blame me for insisting and examining only what is apt to please,
without a word of what might profit.
" I. I believe the end of all Poetry is to please.
" 2 . Some sorts of Poetry please without profiting.
"3. I am confident whoever writes a Tragedy cannot please but must also profit;
'tis the Physick of the mind that he makes palatable.
"And besides the purging of the passions, something must stick by observing
that constant order, that harmony and beauty of Providence, that necessary relation and chain, whereby the causes and the effects, the vertues and rewards, the
vices and their punishments are proportion'd and link'd together; how deep and
dark soever are laid the Springs, and however intricate and involv'd are their
operations." (Spingarn, Critical Essays, 11, pp. 206-207.)
4 Rymer, A Short View of Tragedy (London, 1 6 ~ ~[in
1 ,Spingarn, Critical Essays,
11, p. 2271: "But besides the Alanners to a iMaguijro, humanity cannot bear that
an old Gentleman in his misfortune should be insulted over with such a rabble of
Skoundrel language, when no cause or provocation. Yet thus i t is on our Stage;
this is our School of good manners, and the Speculufn Vitae."
Rymer, Mofzsieur Rapi~z'sRc$ertions, IT, p. 112: " . . . he confesses, that
we have a Genius for Tragedy above all other People; one Reason he gives we
cannot allow of, uis. The Dispositiolz of our fiation, which, he saith, i s delighted zeith
cruel Things."
196
A. Philip McMahon
which the traditional def~nitionsof comedy

introduced.
and of tragedy
are
The influence of the Renaissance philologists and of Rapin continued

down to the end of the century, and in 1770 a work in which they are
much quoted was judged a suitable book to give to the Library of Harvard College by Sir Thomas H01lis.~ Blount notes the opinion of Vossius with regard to the Poetics, that its present state represents substantially the original scope of the work! He cites Rapin r e ~ e a t e d l y , ~
quotes Donatus by way of the AbbC tl'Aubignn~,~
and in tliscussing
Terence relies on Heinsius, Erasmus, Scaliger, Vossius, Lipsius, and
Rapin, among others?
1 [Sir Richard Stcele], A Grunt?nar o j the English Tongzrc . . . T h e Whole Atlafzing a Cornpleat Ssslenz oj art English Edz~catiort. For the Gsc of the Schools of Great
Britain and Irelaqzd (4th ed London, I jzr), p 150.
I n Comic Scenes the common Life we draw,
Accordinx to its Tbmours, Ictions, Law,
And Vice and Folly laughing, kecp in awe.
n u t what is yet a nobler, juster End,
To all the Charms of Virtue do's commend."
Ibid., p. 154:
"One only Action; that's entire and grave,
And of just length, the Tragic l l u s e must have
The Object of its artful Imitation,
And that without the Help of the Narration,
Dy the ztronq I'ow'r of Terrour and Compassion.
All sorts of I'assion perfectly refines,
And whrtt is in us to Passion else inclines."
3 Sir Thomas I'cpe I%lount,De Rc i'i~elicu, or, Kcnzarlzs erpo?z I'oclry.
with
Charucters und Cerrs~iresi f the ,%lost Considcrizble l'oc,ts, whetlzcr Ant-ieizt or ~ l l o d e r n .
[New pagination [or Churuclcrs iirid Censzrres, contrtined in same volume.] London,
1694.
Ibid., Chizrizcters, p. 16: " ~ l r i s l ~ l lhad
l c fnish'd; and given the last strobes to
this most Excellent Worb; Ant1 this, says Lhssizls, may casilp 1)c prov'd hy that
czrriozrs Afcthod, and ail~niiable co?zcalmirliow, which hc hath observ'd from first
to last."
Ibid.; cf. pp. 45, 5j.
Ibid., p. 54.
7 Ibid., p. 227.
Aristotelian Delinitions of Tragedy and Comedy
197
The Latin lectures on poetics of Joseph Trapp were apparently popular, for the third edition had been reached by 1736 and an English
translation came out in 1742.' He combines Cicero and Horace ingeniously in his definition of the drama? and his definition of comedy is
a logical expansion of the result? with the moral effect emphasized.
He also combines Aristotle and Vossius to secure a definition of trage d ~ . ~
In 1738 was published an interesting work on poetics by Henry Pem,~
on Aristotle, and of
berton. He sets forth a view of t r a g e d ~ relying
comedy depending on Horace and C i c e r ~ . ~
Joseph Trapp, Lectures o n Poelry, translated from the Latin (London, 1742).
Trapp, Praelecliones Poeticae i n Schola Naturalis Philosophiae Oxon. Habitue
(2 vols., London, 1736)~(gtl etl.), 11, pp. 153-154. "Drama hoc motlo definimus:
Nimirum quotl sit, Pocma cerbam quandam actioncm continens, et veram humanae
aitue imaginem exhibens, delectationis atque z~tilitatis:causa.
Ibid., 11, p. nor: "Nimirum, quod sit Poema Dramaticum, vitae communis et
privatae imagincm cxhibens, virtutem commendans, ti vitia qz~aedam,atqz~eineptias
h o m i n z ~ mperstringens, jocosa praecipue, sive lepida scribendi ratione."
Ibid., 11, p. 241: "Ex duabus igitur Definitionibus simul collatis tertia conficienda, Aristotelica quidem clarior, Vossiana vero perfectior: E s t itaqz~e. Tragoedia
Poema Dramaticz~m,illwtrcm fortz~nam,scd luctuosanz, gravi, et severa, sed vidctur i n
jzuunda oratione, imitans; ad affectus, praesertim Misericordiam, & Terrorem, ciendos,
animz~mqueab i i s purgandum."
Henry I'emberton, Obseruotions o n Poclry, Especially the Epic, Occasioned by
the Late Poenz on Lconidas (Lontlon, 1738), p. 21: "The genuine tlcsign of cometly
is to represent the true source of private enjoyment from family ariections, ant1 the
judicious choice of our acquintances and frientls; to shew the inconveniences
arising from imprudent conduct, and the irregular sallies of passion, together with
the ridicule due to capriciousness of temper, and other particularities of humorists:
tragedy on the other hand is adapted to form the mind to compassion, to give just
apprehensions of the uncertain state of human felicity, to set forth the excellence
of fortitude, public benevolence, and the other great virtues, and to inspire a detestation of the contrary vices."
Ibid., p. 23: "Whereas tragic and epic poetry relate chiefly to men in high
station, ant1 comedy or similar narrations reqard the busy part of common life as
i t is found in cities and large societies; so the true office of pastoral is to express
the cares and the amusements of the rustic condition."
19s
A. Philip McMahon
In a work dedicated to David Garrick,' William Cooke draws the

usual distinctions between tragedy and comedy in his chapter XXVI,
and he quotes Cicero by way of Rapin on c ~ m e d y . ~
The passages cited in the foregoing pages are, of course, but a fragment of the available material. They are, however, representative and
significant. Further investigation would serve to make the relation of
the standard definitions to each period and author more precise, but
not to destroy the fundamental thesis of this article. That thesis is:
T h e definitions of tragedy and conzedy, ultimately derived fronz Aristotle's
On Poets dominated European tlzeories from the time of Aristotle down to
tlze Ronzantic nzovenzent. Both Greek and Roman literature show that
the Poetics had comparatively slight influence in determining ancient
icleas of the nature of tragedy and comedy, but the presence of the
standard definitions, the essence oi which is contained in the words of
Theophrastus, is everywhere to be found. The Middle Ages accepted
the same statements, and Dante as well as Chaucer did no more than
participate in the universal inheritance. During the Renaissance and
afterwards, both on the Continent and in England, ideas of the tragic
and comic, of tragedy and comedy, were not radically or immediately
altered by the recovery of Aristotle's Poetics, but continued fundamentally to depend on the traditional conceptions. I t was not until the
Romantic movement that any other understanding of tragedy and
comedy gained ground, and not until some time afterwards that classical philology, influenced by the Romantic philosophies, reinterpreted
the Poetics in a manner unknown to all previous ages and to Aristotle
himself. Even to-day popular usage and the dictionaries ignore this
Romantic interpretation of the Poetics and carry on the definitions derived from the dialogue On Poets.
1 William Cookc, The Elcmeuts of Drnn~ntirCritirism (T,ondon, I j 7 5 ) , p. 136.
T o supplement the suggestions made above, interesting material on the
study of the Poetics in the eighteenth century in England, is to be found in
John W. Draper, "Aristotelian 'Rfimesis' in Eighteenth Century England," Publ.
o f t h e Mod. Lung. Assoc., X X X V I (I~zI),p. 372.
Amos Philip McMahon

Warren, Ohio, 14 August 1890 21 June 1947, 27 Washington Square North, New York, New York

Amos Philip McMahon. On The Second Book of Aristotle's Poetics and The Source of Theophrastus' Defnition of Tragedy

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Amos Philip McMahon. On The Second Book of Aristotle's Poetics and The Source of Theophrastus' Defnition of Tragedy

Uploaded by

Copyright:

Available Formats

Harvard Studies in Classical Philology 28 (1917)

ON T H E SECOND BOOK OF ARISTOTLE'S POETICS

AND T H E SOURCE OF THEOPHRASTUS'

INCE the Renaissance any treatment of Aristotle's Poetics has

Hesychius,' where two books are m e n t i ~ n e d . ~This evidence is also

Although the direct assertions of the existence of two books are

A Lost Book o j the Poetics

of that subject is omitted on the ground that it already exists in the

There are in Aristotle and elsewhere grounds for believing that

Bywater, op. cit., p. xxi.

Spengel, Aristotelis Ars Rhetorics, Leipzig, 1867,i, 159,15.

Vahlen, 09. cit., p. 77.

F . Ritter, Aristotelis Poetica, Cologne, 1839,p. xi.

A Lost Book of the Poetics

The exposition of some other matters would probably have found

Bywater, op. cit., p. xxiii.

A. Brandis, Scholia, p. 43 A 13; Kalbfleisch, Simplicii i n Aristotelis Categorias

Bywater, op. cit., p. xxiii.

A Lost Book of the Poetics

refuted by Aristotle's theory of katharsis, as it is usually interpreted.

listed the Poetics as having consisted of two books,l and Spengel

H. Diintzer, " D i e Aristotelische Poetik und ihr Verhlltniss zu den Biichern

8 Bywater, op. cit., p. xxi.

7 Rose, De Arist. Libr. Ord., p. 133.

A Lost Book of the Poetics

The existence of a second book, assumed to be lost, cannot, in the

A. Hatzfeld and M . Dufour, La PoLtipue d'rlristote, Lille, 1899, p. vii.

Shute, op. cit., p. 90.

S h u t e , op. cit., p. 86.

Zeller, op. cit., i, p. 49.

Rose, Fragm. pp. 3 ff: 2 , 83, 118, 119, 136.

A Lost Book of the Poetics

history and the evident consequences of frequent mistranslation make

A Lost Book of the Poetics

* Zeller, op. cit., i, p. 102, n. 2.

Op. cit., i, p. 102, n. 2.

plete passage to which Bywater refers shows clearly that it points to

In spite of the reasons brought forward to explain the loss of a

A Lost Book of the Poetics

Op. cit., pp. xxiii, xxiv.

A Lost Book of the Poetics

certain enoughll and the passage probably refers to the works of

The evidence for a second book afforded by the cross-references in

Op. cit., pp. 35 ff.; Cicero, Fin., 4, 28, 79.

numerous cross-references are genuine, it must be supposed that the

A Lost Book of the Poetics

references to questions far removed from his discussion both in thought

himself obliged to reject, he continues: " Turning to Aristotle's own

A Lost Book of the Poetics

While the professors of the Peripatetic school could not compare in

I t is interesting to note that the dialogue form was employed by Minturno in

Gercke, Pauly-Wissowa, ii,

A Lost Book of the Poetics

One of the problems to solve is the reference of the Politics to

Bywater, op. cit., pp. 94, 95. Cf. p. xxi.

ence in the Politics promises. The scholarship of Proclus, moreover,

Finsler, op. cit., p. 8 and n. 3.

A Lost Book of the Poetics

Op. cit., p. vii.

O p cit., pp. 7 1 ff.

of the chapters in which they occur, it must be admitted that such

Ritter, o p . cit., pp. 230-243.

A Lost Book of the Poetics