1 Status quaestionis
In his rebuttal of Aeschines’ personal attacks at the trial against Ctesiphon, Demosthenes contrasts his own background and career with those of his rival through a sweeping series of antitheses, which culminates with the request for the testimonies of his liturgies to be read by the clerk (18.265-267). Before the reading begins, he claims that Aeschines would have nothing better to respond with than his own atrocious tragic performances. He then quotes some of the lines that Aeschines used to butcher, and wraps it up with a curse:
φέρε δὴ καὶ τὰς τῶν λῃτουργιῶν μαρτυρίας ὧν λελῃτούργηκα ὑμῖν ἀναγνῶ. παρ’ ἃς παρανάγνωθι καὶ σύ μοι τὰς ῥήσεις ἃς ἐλυμαίνου,
ἥκω νεκρῶν κευθμῶνα καὶ σκότου πύλας”
κακαγγελεῖν μὲν ἴσθι μὴ θέλοντά με”,
καὶ “κακὸν κακῶς σε” μάλιστα μὲν οἱ θεοί, ἔπειθα οὗτοι πάντες ἀπολέσειαν, πονηρὸν ὄντα καὶ πολίτην καὶ τριταγωνιστήν.1
Well then, I wish to read you also the testimonies of the liturgies that I have performed. As a comparison, you, too, read me the tragic speeches that you used to defile,
“I come [having left] the den of the dead and the gates of gloom”
“Know that I bring bad news unwillingly”,
and, bad man, may badly first of all the gods, then all these men destroy you, worthless that you are both as a citizen and as a third actor.
The lines that Demosthenes puts in Aeschines’ mouth are metrically complete tragic trimeters. The first one is the opening of Euripides’ Hecuba.2 The second quotation is an unidentified fragment from a messenger speech (adesp. fr. 122 TrGF = Nauck2). The curse that follows starts with the words
The metrical character of this sequence was already noticed by Wolf in the sixteenth century.4 According to Wolf, the sequence is part of a quotation ending after
The special status of this ‘quasi-quotation’ started being indicated in print in the early nineteenth century. Wunderlich prints a dash after
A few decades later, Meineke triangulated the curse with a sentence from a prose fragment of some Aegyptiaca of Lynceus12 (
Forms of typographical highlighting of
Meineke’s line does not survive into Kannicht and Snell’s edition of the tragic Adespota.23 Nevertheless, the beginning of the curse at De corona 267 is still commonly printed as the opening of a quotation24 and continues to be analysed as a snippet of verse by commentators. Yunis, for instance, explains that
The iambic rhythm, the structure of the sentence, and the assonances with lines of comedy (especially with Menander) are certainly intriguing, but one is left wondering whether this is enough for us to infer that Demosthenes delivered this sequence as the beginning of a tragic or comic trimeter—and print it as a broken quotation of sorts—and not simply as an imprecation. In order to address this question, we shall (a) examine more closely the uses of the polyptoton in curses in order to assess if the overlap with lines of comedy is a matter of intertextuality or merely of phraseology, (b) try and pinpoint the function of
2 The ‘Curse Formula’
Combinations of forms of the adjective
The rhetorical effects of polyptoton, along with the semantics of
The earliest extant examples of the use of
Let us now focus on some formal features in order to refine our definition of the curse formula as such. In these early epigraphic attestations,
These data are too scanty for statistical analysis, but we may still note that
(1) the most common forms of the curse are those with a compound of
ὄλλυμιas the verb and with the adjective directly preceding the adverb (10 out of 22 occurrences);
(2) outside tragedy, virtually only compounds of
ὄλλυμιare attested (with only one exception);
(3) Demosthenes only uses one pattern (adjective immediately preceding the adverb, with
ἀπόλλυμιas the verb), which matches both the fragment of Lynceus and a number of examples from Hellenistic and Imperial prose.40
Even though such a low number of occurrences does not allow us to draw solid quantitative conclusions, these observations at least raise the suspicion that tragedy allows more variety in the configuration of these curses than comedy and direct speech in prose. If comedy and prose are to be taken as more representative of ordinary usage than tragedy,41 one gets the impression that the most basic forms of the curse are those that feature a compound of
This impression is reinforced by the observation that when the curse is alluded to (but not uttered as such) by both Demosthenes44 and Aristophanes,45 the adjective is immediately followed by the adverb, as if this very sequence were the hallmark of the curse formula itself. In this connection, one may also mention a pun at Ar. Ach. 253, where the imprecation is turned into its opposite (
To sum up, Renehan is right in describing
3 Coordination and Oral Delivery
One of the elements that may encourage us to interpret
On the other hand, if we consider that the curse is an independent discourse act, we may interpret the occurrence of
Between the series of antitheses and the curse, the speech is characterized by the build-up of an explicit contrast between the first person and the second person.62 The first person refers to Demosthenes and the second to Aeschines, except in the quotations, where the first person may be constructed as referring, through the persona of a stereotypically cautious tragic messenger breaking bad news,63 to Aeschines himself. In particular, Demosthenes may have chosen to put the latter quotation (
Under this interpretation, the curse would form an entirely independent intonation unit from the quotations and, as far as punctuation is concerned, would simply need to be separated from them by a full stop.
4 Word Order and Rhythm
Now that we have established that neither phraseology nor initial
The essential building blocks of the curse are the polyptoton, the verb, and an expression referring to the target of the curse. If we look at variation in the relative position of these elements in literary occurrences of this expression, we may note that the verb precedes the polyptoton only three times, in passages where the curse is ancillary to speech acts of a different type: at E. Med. 1386 (
When it comes to the position of the target, a number of arrangements are attested. As we have just seen, when the verb precedes the polyptoton, the target precedes the verb.67 If the polyptoton precedes the verb, the target may follow68 or precede the whole group, as is the case with the other Demosthenic occurrence.69 In drama, the target of the curse may intervene between the elements in the polyptoton.70
The pattern that we observe in De corona does not correspond to that of any of the other literary attestations. The polyptoton is immediately followed by the clitic pronoun referring to the target of the curse and, as noted above, they are both extraposed from the clause and form a separate colon. Interestingly, this arrangement has a striking parallel in Aeschines’ Embassy speech:
παρακαλῶ δὲ καὶ ἱκετεύω σῶσαί με πρῶτον μὲν τοὺς θεούς, δεύτερον δ’ ὑμᾶς τοὺς τῆς ψήφου κυρίους, κτλ.71
Ι beseech and entreat that I may be saved first of all by the gods, secondly by you who have the power of voting, etc.
Aeschines’ appeal is the polar opposite to Demosthenes’ curse:
If we discount the words
Intertextual exchanges between Demosthenes and Aeschines across different trials are nothing surprising,73 and it is plausible that Demosthenes’ wording is a subtle caricature of Aeschines’ pathetic oratorical style, while being linguistically justifiable in its own right. In particular, both
The most evident formal difference between Aeschines’ appeal and Demosthenes’ curse lies in the fact that in Aeschines, the verb
At the same time, other explanations for the position of the verb are possible. To begin with, the sequence
The results of this study are (a) that the sequence
The connection between Demosthenes’ curse and Aeschines’ plea in the peroration in his speech On the Embassy is extremely tantalizing. Rather than switching abruptly from quoting to cursing (and deploying a
The quotations and the curse may thus be described as distinct but closely connected verbal devices by which Demosthenes brings about a masterful and sophisticated parody of a number of aspects of Aeschines’ figure all at once. By quoting tragedy, Demosthenes mocks both the professional background and the oratorical style of his opponent; at the same time, the stereotypical character evoked through the second quotation and the wording of the curse are a subtle parody of Aeschines’ defence strategy as displayed in the Embassy trial. The striking rhetorical effect of interrupting a quotation would probably obfuscate the intertextual echo of the curse and, arguably, detract from the oratorical force of the curse itself. On these grounds, printing the words
Adrados, F.R. (1999). History of the Graeco-Latin Fable. Introduction and from the Origins to the Hellenistic Age. Translated by L.A. Ray. Leiden/Boston.
Adrados, F.R. (2000). History of the Graeco-Latin Fable. The Fable During the Roman Empire and in the Middle Ages. Translated by L.A. Ray. Leiden/Boston.
Bers, V. (2009). Genos dikanikon. Amateur and Professional Speech in the Courtrooms of Classical Athens. Washington/Cambridge, MA.
Carrara, L. (2018). Edipo all’altare? Per una lettura ed interpretazione di Euripide, fr. 554a K. (Edipo). In: L. Austa, ed., The Forgotten Theatre. Mitologia, drammaturgia e tradizione del teatro frammentario greco-latino, Alessandria, pp. 111-132.
Dalby, A. (2000). Lynceus and the Anecdotists. In: D. Braund and J.S. Wilkins, eds., Athenaeus and His World. Reading Greek Culture in the Roman Empire, Exeter, pp. 372-394.
Gambetti, S. (2016). Lykeas of Naukratis (613). In: I. Worthington, ed., Brill’s New Jacoby, http://dx.doi.org/10.1163/1873-5363_bnj_a613, consulted online on 12 August 2020.
Gomperz, T. (1910). Die Apologie der Heilkunst. Eine griechische Sophistenrede des fünften vorchristlichen Jahrhunderts. Leipzig.
Jones, N.F. (2016). Amelesagoras of Athens (330). In: I. Worthington, ed., Brill’s New Jacoby, http://dx.doi.org/10.1163/1873-5363_bnj_a330, consulted online on 12 August 2020.
Lambert, F. (2017). Les emplois de καί initial en grec ancien. In: F. Logozzo and P. Poccetti, eds., Ancient Greek Linguistics. New Approaches, Insights, Perspectives, Berlin/Boston, pp. 193-209.
Natalicchio, A. (2000). Per Ctesifonte, Sulla corona. In: L. Canfora, ed., Discorsi e Lettere di Demostene, Vol. 2, Turin, pp. 19-216.
Renehan, R. (1976). Studies in Greek Texts. Critical Observations to Homer, Plato, Euripides, Aristophanes and Other Authors. Göttingen.
Vince, C.A., and Vince, J.H. (1926). Demosthenes, Orations, Vol. 2: Orations 18-19. De corona, De falsa legatione. Cambridge, MA.
D. 18.267. Translation by the author.
Interestingly, the quotation is not syntactically complete, since the accusatives are governed by the participle
Cf. Voemel 1862, 316 and Wankel 1976, 1167.
It is noteworthy that Hermogenes comments on Demosthenes’ use of quotations in this passage but makes no mention of the curse (Id. 338.6-10 Rabe).
See Marini 2007, 236. Among ancient rhetorical sources cf. in particular Hermog. Meth. 34. Both Wolf and Markland apud Reiske 1774, vol. 1, 312.
Reiske 1774, vol. 2, 534-535.
Wunderlich 1810 (non vidi), according to Schäfer 1824, 348.
Wunderlich 1820, 44.
Bekker 1823; Dissen 1837; Dindorf 1846.
Schäfer 1824, 348.
The only known Lynceus is a comic poet from Samos who was the brother of Duris and a pupil of Theophrastus (Sud. s.v.
Meineke 1846, 7.
Voemel 1862, 316. Cf. also Goodwin 1904, 165, who misconstrues Meineke’s restored trimeter as attributed to Lynceus and a potential direct source for Demosthenes. As Renehan 1976, 115 shows, this is impossible.
Blass 1890, 159.
So for instance Butcher 1903 (OCT), Fuhr 1914 (Teubner), and even Vince and Vince 1926 (Loeb), whose text is based on Dindorf’s third edition (Teubner, 1866).
The passages read:
So Mette in Snell’s Supplementum to Nauck (Snell 1964, 25).
Renehan 1976, 114-116.
After Pearson 1917, 17 and followed by Beroutsos 2005, 81.
Cf. Wankel 1976, 97.
Wankel 1976, 1167-1168.
Kannicht and Snell 1981, 51. In their view, the curse in the fragment of Lynceus would only amount to an affectation of the sound of tragedy.
E.g. by Zürcher 1983; Usher 1993; Natalicchio 2000; Yunis 2001; and Dilts 2002.
Yunis 2001, 260.
Pontani 2009, 402-404.
West 2007, 111-116.
Cf. Coray 2016, 134 and Brügger 2018, 64.
The polyptoton extends to the following sentence into a wide-ranging repetition:
Cf. also Hippon. fr. 126.3 Degani, with (restored) ⟨
The following is the list of the occurrences of the polyptoton (except in curses) up to the age of Demosthenes (by genre). Tragedy: A. fr. 349 TrGF, Pers. 253, 531, and 1041 (three forms of
Cf. LSJ s.v.
Voutiras 1998, 25.
Pace Voutiras 1998, 32.
Cf. O’Neil 2006, 197.
Cf. Dittenberger 1915, 768.
See Bubeník 1989, 89-90.
E.g. Plb. 7.3.2; Ev.Matt. 21.41; Alciphr. 2.2.1; Luc. JTr. 38 and Pseudol. 24, but not Plu. Cic. 26, where the adjective and the adverb are separated by a particle and the verb in an imprecation ascribed to Cicero.
Cf. e.g. Slings 1992.
Literary attestations would suggest that the adjective normally precedes the adverb unless some elements intervene, in which case the reverse order seems more frequent (cf. already Elmsley 1818, 202), but the epigraphic data do not confirm this.
Cf. Gygli-Wiss 1966, 82.
The passages read:
Cf. Gygli-Wiss 1966, 82.
I.e. one may curse a bad man with a bad end because bad men must suffer. The fragment itself would be uttered by a character pondering on a hypothetical situation; see Carrara 2018, esp. at 130-132.
For a gnomic use of the polyptoton in tragedy cf. for instance
See Finglass 2006, 260-263.
Finglass 2018, 253.
Renehan 1976, 115.
As described by Arist. Rh. 1403b22-35.
The only configurations in which the sequence would not scan iambic are
Cf. Wray 2002, 70.
The same applies to the universal quantifier (
Mathieu 1958 prints suspension points between
See e.g. Dik 1995, 36-37 for a brief explanation and Goldstein 2016, 44-84 for the particulars.
We can assume that the phraseological connection of the adjective and the adverb can make them function as one prosodic word and as the host of clitic
Whether or not Demosthenes recited the iambics in a theatrical manner in order to caricature Aeschines’ antics (cf. e.g. Bers 2009, 31), it is likely that his audience would be able to identify the trimeters as quotations based on their syntax, meaning, and vocabulary, not to mention the fact that the first line of Euripides’ Hecuba must have been well known to the audience (it was certainly famous at the time of Aristophanes, who quotes it almost literally in the Gerytades, fr. 156 K.-A., and parodies it in one Aeolosicon, fr. 1 K.-A.). In particular, the lexeme
See for instance Liapis 2012, 323. A number of the examples presented by Denniston 1954, 322-323 may be interpreted in this direction.
Cf. Lambert 2017, 202.
Cf. de Jong 1991, 71 n. 26.
Cf. De corona 139-140, 191, 198-199, and 272-273.
Cf. Wankel 1976, 1168.
The verb intervenes between the adverb and the adjective at S. Ph. 1369 (
At Ar. Pl. 65 the target intervenes between the preverb and the verb in tmesis.
Eub. fr. 115 K.-A. 1-2.
D. 7.45; cf. also Men. Dys. 220-221, Epit. 424, etc.
S. Ph. 1369; E. Cyc. 268-269; Ar. Eq. 2-3.
Aeschin. 2.180. Translation by the author.
The position of the clitic pronoun
See e.g. MacDowell 2000, 29.
In general, clitic personal pronouns tend to occur as the last word of a preclausal colon most frequently when they are the extraposed subject of an infinitival clause (e.g.
I have tried to reproduce the effect in my translation of
Accordingly, it is not necessary to identify a literary or rhetorical source for the curse in the fragment of Lynceus discussed above (pace Gambetti 2016).
I wish to thank the anonymous reviewers and the editorial board of Mnemosyne for their helpful comments and suggestions.