As Hidber (H.) entertainingly reminds us at the beginning of his book, Wilamowitz called Dionysius of Halicarnassus a “pitiful fellow” (“arme Geselle”), and the judgements of other scholars of the same era were hardly more friendly. The last few decades, however, have seen a resurged interest in Dionysius, and one of his most significant texts is the Preface to On the Ancient Orators, which describes the recent (near-)victory of the “Attic Muse” over the mindless bombast of Asianism, which, he alleges, reigned supreme in the Hellenistic period. Despite the shortness of this text (some four pages), the complexity of the issues it raises fully justifies the separate treatment now provided by H. Those interested in these issues will want to consult his book, but must do so with caution. It is, to this reviewer’s mind, something of a mixed bag. Though both strong and weak points are found in almost all areas, the main weaknesses concern the reconstruction of the rhetorical and cultural background, while the main strength of the book is the presence of a number of sensible observations on Dionysius’ meaning and tendencies.
H. offers (apart from bibliography and indexes) an 87 page introduction, a Greek text with facing German translation, and a commentary of 40 pages. The introduction treats all or most of the issues relevant to the Preface, starting with sections on Dionysius’ life and Roman environment and on his work, with emphasis on On the Ancient Orators. There follows a discussion of the “classicist” three-period model that we find in Dionysius (“der klassizistische Dreischritt”), a concept especially highlighted by Gelzer: the division of literary history into a classical, exemplary period (in Dionysius, that of the Attic orators), an intermediate period of deterioration (Hellenistic “Asianism”), and the present which is reviving the great past. A long section then treats the opposition between Atticism and Asianism, and the vexed questions of the rise of Atticism and the relationship between the Roman variant combated by Cicero on the one hand and the Greek variant encountered in Dionysius on the other. The next sections deal with Dionysius’ own position: his educational ideal (“Bildungsideal”) of “philosophical rhetoric”, his views on imitation ( μίμησις), and his notion of Rome as the new Athens. A relatively brief section on the form of the Preface, connecting it with historiographical prefaces, closes the introduction.
Before discussing some of the issues raised in these sections in sequence, let me mention some of H.’s good observations, which occur throughout the book. Some of them concern details, but many details in the Preface have wider implications. A good example is H.’s note on the phrase used to describe the “Asian” rhetoric in 1.7, “arrived only yesterday or the day before” ( ἐχθὲς καὶ πρῴην). He rightly remarks that the syntax of the passage ( μέν… δέ) confirms that this phrase does not indicate a concrete period, but is contrasted to “the ancient Attic Muse” (1.6); this helps to disprove the notion that Dionysius is following a second century source, and that Atticism must therefore be considerably older than Dionysius (Heldmann, Antike Theorien… [1982], 130). Equally welcome is H.’s sensible rejection (78-9, 122, 122-3) of the view of scholars like Hurst ( ANRW II.30.1 [1982], 848ff.), who refuse to take Dionysius’ positive references to things Roman seriously, holding that the mention of Roman literature in Praef. 3.2 is insincere, and sometimes even that Dionysius is implicitly critical of Augustus. A long footnote on Caecilius of Caleacte (41) is a salutary warning for those who, like myself, may have relied too much on Longinus 32.8 for their picture of Caecilius as an exclusive champion of Lysias: matters are more complicated than that. A pleasant surprise is the well-argued rejection of the too common notion that Cicero’s educational ideal in De oratore is simply Isocratean and is therefore basically the same as Dionysius’ (51-4).
H.’s first section, which discusses Dionysius’ life and (on 5-8) his contemporary milieu, is a characteristic mix of strong and weak points. H. rightly rejects the view that Dionysius’ acquaintances in Rome were all insignificant, offering the now traditional list of contemporaries who are mentioned, as dedicatees or otherwise, in Dionysius’ extant writings. His stress on the uncertainty whether Pompeius Geminus and Ammaeus were Greeks or Romans (7) is sensible, and in his commentary he quotes inscriptional evidence, not noted earlier in this connection, that the latter’s name at least may well have been Roman. But the question of the nature of Dionysius’ contacts is given short shrift. Apparently, H. writes, Dionysius and all those he mentions formed one of the numerous “literary circles” in Augustan Rome (7; the English term is H.’s). This is far too simple. Not every group of people linked in this way formed a “circle” in the stricter sense, i.e., in the sense in which we talk of the circle of Maecenas. In this case, there is actually no evidence for such a construction. And Pompeius Geminus at least was probably “not one of Dionysius’ closer associates” (Usher, Loeb II, note ad Ep. Pomp. 1). H.’s reference to other scholars on this matter (7 n. 51) is no help here, as their approaches differed widely. E.g., Roberts ( CR 1900, 439-42) did mention a “circle” in his title, but did not argue for the existence of a tightly knit group; Goold, on the other hand, did ( TAPhA 1961, 168-92), but his fanciful reconstructions and identifications have not found much credence, and run counter to H.’s own rightly cautious approach.
On a related note, the contemporary setting of some of Dionysius’ ideas is sometimes obscured. E.g., H. is no doubt right that Dionysius went back to Isocrates for his “Bildungsideal”, and (as noted above) to reject Cicero as an intermediary on this point (44-56). But when H. adduces only Isocratean parallels for the concept of Athenian autochthony (100, ad Praef. 1.6), this wrongly suggests that this is also taken from him. The concept was not only widespread in classical times (see, e.g., Thuc. 1.2.5 with Hornblower’s note, and in general N. Loraux, Les enfants d’Athènes [1981], 35-73), but appears as a matter of course in Cicero, Pro Flacco 62. Likewise, the relationship between the alleged Peripatetic origin of the “term” “political philosophy” ( πολιτικὴ φιλοσοφία) and the contemporary use of the phrase by Strabo remains unilluminated (131, ad 4.2).
The sections on Dionysius’ Atticism, however, firmly put this central aspect in the contemporary context (esp. 33-44). If I find H.’s discussion disappointing, it should be said that I am not completely impartial, having gone on record as holding a different view (see note 2). It took me some effort to arrive at a clear picture of H.’s opinions, not least because he separates two questions that are not really independent: one section, entitled “Zur Entstehung des Attizismus”, only discusses the origins of the Roman variant (30-37), while the problem of the connection between this and other variants is postponed to the next section (37-44). Also, H. has hidden in a footnote (35 n. 158) his crucial (and correct) observation that there are no indications that Atticism is earlier than the first century BC. Finally, the notion that it was “perhaps” (“vielleicht”) Greek rhetoricians in Rome who were responsible for the movement is introduced in a parenthesis, embedded in the most complicated sentence of H.’s whole text, which is otherwise written in very clear German (35). The clear summary at the end of the sections in question (42-44) is therefore welcome, but does not address all the issues.
All in all, the picture emerging from H.’s account relies heavily on Dihle’s influential articles, and is more or less as follows. Calvus and the other representatives of Roman Atticism advocated a simple style on the model of Lysias and Hyperides (this rhetorical aspect rightly receives more stress than in Dihle). Moreover, they attached particular importance to achieving linguistic purity, to which end they invoked the doctrine of grammatical analogy (30-2, 38-9). The Greek version of the movement as seen in Dionysius was much broader in its choice of models, and did not need a specific grammatical theory, since enough recognized models of pure Greek were available (39). To these differences between the Roman and the Dionysian variant H. adds another: the former was restricted to forensic oratory, the latter concerned itself with all literature (39). H. holds (39), as does Dihle (esp. 1977: 176), that these differences show that Dionysius’ Atticism cannot have been derived from the Roman variant, as has been argued by Kennedy, Bowersock and myself. Both must stem from an earlier, Greek version. As compared to Dihle, H. adds (in the wake of Gelzer [see note 1], 15-6), that the earlier existence of a broader Atticism from which the Dionysian variant could be derived is indeed reflected by a passage from Cicero’s Brutus, viz. 285 (32-3, 40; cf. 71).
There are grave problems to this account. To start with, the close link between Roman Atticism and analogy rests on tenuous evidence. One of its main supports is the connection of Caesar with the Atticists, a connection always open to doubt (cf. Douglas, CQ 1955, 245), and now apparently even doubted by Dihle himself, as H. notes (32 n. 145). The difference between the two variants of Atticism is therefore probably smaller than H. allows. Moreover, H.’s interpretation of Brutus 285 will not stand scrutiny. The allegedly crucial phrase Atticos, inquit, volo imitari (“My aim is, he says, to imitate Attic orators”) does not reflect a broad Atticism different from that of Calvus; as the context shows, it is actually presented by Cicero, whether fairly or unfairly, as Calvus’ own view (cf. also J.M. May, Athenaeum 1990, 177-80). Finally, I fail to see why the differences between the Roman and the Dionysian variants should preclude a direct connection; why, in other words, the development of a broad Greek version into a narrower Roman one is inherently more likely than the other way round. The model of (gradual) development is actually far from unproblematic for what was basically a literary fashion, which may partly have served to distinguish its proponents from their contemporaries.
As to the remaining sections of the introduction, I will limit myself to comments on two of them, first that on Dionysius’ view of “eclectic mimesis” (56-75): the selective imitation of the best features of the great writers. This is the best-written part of the book, and in it, H. makes a good case for the great importance of mimesis to Dionysius. A small lacuna here is a description of Dionysius’ criteria for judging the authors he discusses; this long section could have accommodated more than a mere reference (65-6) to Classen’s recent treatment in the 1994 Entretiens on La philologie grecque (331 ff.). A little more attention to the criterion of ἄλογος αἴσθησις ( Lys. 11) and to the related difference between natural and artificial mimesis (as described in the striking passage Din. 7.5-6) would also have been welcome. But although most of H.’s points will be uncontroversial, the section offers enough as it is. H.’s attention to the mimesis of the life of the worthy orators and writers, mentioned by Dionysius alongside the mimesis of their writings ( Praef. 4.2), allows him to make more sense of Isocr. 5-10, with its stress on the moral effect of Isocrates’ speeches (72-4); contrast, e.g., the surprise at these sections in Grube’s good discussion of Dionysius ( Greek and Roman Critics, 215-6). H. also observes that in Dionysius’ view, the classical orators themselves already used the method of eclectic mimesis, a point that neatly ties Dionysius’ aims even more closely to his venerated models (68-9); and the interpretation of Dionysius’ historical work (obvious once pointed out, but missed by many), that it is meant to exemplify this method, allows us to see the unity of Dionysius’ endeavours.
The last section of the introduction links the Preface with the genre of the historiographical prooemium. The idea is interesting and worth pursuing, but H.’s own discussion is extremely brief (85-87), giving the impression of an afterthought. One misses, e.g., references to and comparisons with Longinus’ first chapter and Cicero’s De oratore 1.1-5 and 2.1-11.
It is H.’s text, translation and commentary that most clearly show the unevenness of his book. The translation is the most accurate one available, though some corrections may be suggested. the commentary is generally informative and contains some good interpretations (see above), but there are some surprising omissions; e.g., I for one would have profited from a discussion of Dionysius’ reasons for insisting at such length not just on the novelty of his treatise On the Ancient Orators, but also on his uncertainty whether there were really no predecessors in existence (4.3). More generally, some of the notes show considerable overlap with the introduction. Also, the latter would have been a better place for some general points, such as the note ad 1.5 on the application of moral categories to the criticism of Asianism (something similar is sometimes true of the footnotes to the introduction).
H. is very weak on manuscripts and on grammar. He offers no apparatus criticus, a surprising omission in a book-length study of such a short text. A footnote in his preface (p. x) tells us that he generally follows the text of Usener-Radermacher, not Aujac’s recent Budé, though he does employ the latter’s paragraphing, which is fortunately becoming standard. But he seems unaware of Aujac’s extensive account of the Mss (Budé vol. 1, 29-34, 54-63) and even of that of Usener-Radermacher; in what is, as far as I have seen, the only remark on an individual Ms (ad 2.3 ἤρκαντο), he measures the importance of the Ambrosianus by quoting Sadée’s judgement from 1878 (sic), that it is an “optimus codex”.
His handling of this problem in 2.3 in the commentary (the only real textual problem, he claims) is actually confused on several other counts. He is right that the plural ἤρκαντο is problematic because of the switch to the singular παρεσκεύασε in the next clause (the former is transmitted unanimously, unless Sadée was right that the Ambrosianus has ἤρκατο; but Aujac reports no variants). But since he adduces a grammatical explanation for this switch, referring to Kühner-Gerth II 1, 79, he should not have obelized ἤρκαντο. However, the explanation is impossible: Kühner-Gerth only offer parallels for a switch from singular to plural, and that only in cases where a second subject is added (e.g., Xen. An. 2.4.16, ἔπεμψέ με Ἀριαῖος καὶ Ἀρτάοζος… καὶ κελεύουσι…). We must either read ἤρκατο (for the singular with plural subject see K.-G. o.c., 80-1), or follow Radermacher’s suggestion to emend παρεσκεύασε into – ασαν.
One other example must suffice. Kennedy (see note 6) translated 3.1 γενναῖοι τὰς κρίσεις γενόμενοι as “exhibit the judgment associated with noble birth”. H. rightly rejects this translation (120), but does so only tentatively (on grounds of content?), and without mentioning its grammatical impossibility. A successful textual note, on the other hand, is that ad 2.5, where H. is right to adopt Radermacher’s insertion ( τοὺς < τοιούτους > λόγους).
The widest implications of points of language and text are those at the much-discussed passage 3.1, where Dionysius describes the salutary effect of Roman influence on the Greek cities. These implications cannot be drawn out here, but two points seem worth mentioning. (1) Rome, Dionysius writes, forces “the cities in their entirety” to look to her. Astonishingly, this phrase ( τὰς ὅλας πόλεις) is translated by “all the cities” by almost all previous scholars, and H. follows suit (“alle Städte”). (2) According to the transmitted text, Dionysius writes that being “honoured” by the Roman elite, “the sensible element of the city” ( τὸ… φρόνιμον τῆς πόλεως μέρος) has gained more strength. Gabba explicitly ( CA 1982, 48), and others implicitly, have taken the singular τῆς πόλεως as “collective”, i.e., as equivalent to “the cities”. Goudriaan rightly deemed this grammatically impossible ( Over Classicisme 568 n.: “taalkundig ontoelaatbaar”). H. (121) quotes Goudriaan here, but characteristically fails to press the point, even though he does arrive at accepting a singular meaning. “The city”, he thinks, refers to Rome, which serves as an example for the Greek cities (thus already Goudriaan l.c., unacknowledged by H.). I find this hard to accept in the context, and suggested < ἑκάστης > τῆς πόλεως, “of city” (art. cit. in note 2, 77). However that may be, H. is inconsistent: his translation has “der vernünftige Teil des Gemeinwesens”.
Despite some strong points, then, H.’s book as a whole is disappointing. Perhaps the project was too ambitious for the time he could spend on it. The significant problems offered by Dionysius’ four page Preface are both more knotty and more numerous than meet the eye.
Thomas Gelzer, “Klassizismus, Attizismus und Asianismus”, in: Le classicisme à Rome aux Iers siècles avant et après J.-C. (Entretiens 25, 1979), 1-41. “Greeks, Romans, and the Rise of Atticism”, in: J.G.J. Abbenes, S.R. Slings, I. Sluiter, eds., Greek Literary Theory after Aristotle (Festschrift Schenkeveld). Amsterdam: VU University Press, 1995, pp. 65-82 (here p. 70). H. graciously mentions this article in his bibliography, even though (as he adds) it came too late for him to use it. Albrecht Dihle, “Analogie und Attizismus”, Hermes 85 (1957), 170-205; “Der Beginn des Attizismus”, A 23 (1977), 162-77. Kennedy, The Art of Rhetoric in the Roman World (1972), 241-2, 351-3; Bowersock, “Historical Problems in Late Republican and Augustan Classicism”, in: Le classicisme (above note 1), 57-75; Wisse (above note 2), 73-7. Dihle earlier regarded Caesar as the founding father of the Roman movement (1977: 166), but recently, he wrote more cautiously: “Die Attizisten, denen wohl auch der Diktator Caesar nahestand” (entry “Attizismus”, Historisches Wörterbuch der Rhetorik I [1992], 1164). Others are those by Kennedy, Art of Persuasion in Greece (1963), 337-40 (quite inaccurate, a surprising ending to this epoch-making book); Russell in Russell-Winterbottom, Anc. Literary Criticism (1972), 305-7 (purposely loose, with some anachronisms); Usher, Loeb vol. 1 (1974; generally good but inaccurate at some points); and Aujac, Budé vol. 1 (1978; often close to a paraphrase). Esp. (correspondences with the other translations are not signalled): 1.6 ἄτιμον σχῆμα”a position without civic rank” fits the metaphors better than “eine entehrte Erscheinung”; 2.5 (and elsewhere) λόγους”oratory” (“Beredsamkeit”) rather than “Rhetorik”; 3.1 (see below); 3.2 “herausgegeben” for ἐκφέρονται invites anachronistic interpretations; 4.3 “erinnere mich” for οἶδα is obviously meant to avoid conflict with following εἰδῶς, but unduly weakens the statement. This includes myself in the uncorrected version of my article (above note 2, 77); also Russell, Usher and Aujac. Kennedy’s “whole cities” is better but ignores the definite article. κοσμούμενον; hardly “ordered”, as in H., Russell, Usher.