Simon Hornblower, A Commentary on Thucydides. Oxford: Oxford University Press, 1991. $120.00. ISBN 0-19-814880-1.
Reviewed by David Potter, University of Michigan.
A Historical Commentary on Thucydides, one of the great achievements of modern scholarship, was composed over a period of more than forty years (vol. 1, the fruit of many years' thought appeared in 1945, volume 5, equally the product of long reflection, was published in 1981) by three scholars whose sound judgment, profound knowledge of Greek, and sympathetic understanding of Thucydides raised the study of the historian and his world to a new level. Conversely, A Commentary on Thucydides, which appears to have been written very quickly indeed, does not pretend to such importance, and it often leaves its reader wondering why it was published. For, although there are some good features of this work, they are outweighed by very significant weaknesses.
First for the good part. Hornblower (H.) has read widely, and his commentary succeeds in wedding literary studies with historical. Anyone who needs to brush up on the bibliography connected with any of the standard Thucydidean problems will want to have a look at this book. H. is also quite good at presenting clear summaries of the issues that have occasioned much of the bibliography. In these discussions, while he does express a clear preference, he is usually more than fair to both sides. A remark such as his unfortunate jibe at survey archaeology ("Th. knew nothing of this sort of evidence, which essentially consists of picking up pot-sherds and counting them" p. 11) or outbursts like the gratuitous attack on Edith Hall (in his note on Thuc. 1.129.1) are fortunately infrequent. Less rare, and usually quite uninformative, is H.'s reportage of private conversation and personal correspondence. The simple statement that Professor x or Dr. y concludes this or that about something is, at times, substituted for serious analysis.
H.'s attention to the epigraphic evidence that supplements the narrative of Thucydides is also extremely good. Since much of this has either come to light or been re-examined since Gomme wrote, this is perhaps the most valuable feature of H.'s work. Thus it is all the more interesting that he leaves IG i3 174-5 out of his discussion of de Ste. Croix's explanation of the Megarian decrees (p. 111) since they show, contra de Ste. Croix, that the exclusion of trade from the empire was conceivable in the 420s, and thus, very probably, in the 430s. A more serious problem is that although H. also has a good command of the physical geography of the Aegean, the reader's ability to follow these is severely hampered by the absence of maps.
H.'s bibliographic discourse often substitutes for original contribution to our understanding of the text. This is not necessarily a vice in a book that is intended for reference, but there are cases where the citation of articles on specific points obscures their importance for more general issues. Thus H. is very good on the conventional arguments about Thucydides' omission of the peace of Callias (p. 179-81), and here notes A. Andrewes' article "Thucydides and the Persians," Historia 10 (1961), arguing that Thucydides consistently misrepresented the importance of Persia. This article is of significance not only here, of course, but for our understanding of the pentekontaetia as a whole. If Thucydides has underestimated the threat that Persia posed, then his analysis of Athens' treatment of the allies may be severely flawed. If Athenian aggression against Greek states in the 470s and 460s was the result of a justified (or even exaggerated but still real) fear of Persia, this alters the perspective of Thucydides' depiction of the growth of Athenian power.1
Similarly, H. has a tendency to cite works that are sometimes very hard to find, without giving the reader much help as to what the author is really talking about. In the note on the peace of Callias (p. 179), for example, he alludes to P.N. Stylianou's article, "The Untenability of Peace with Persia in the 460s B.C.," in Meletai kai Upomnhmata 2, 339-71 published by the Archbishop Makarios III Foundation. The periodical is extremely hard to find outside of Cyprus, and it would be a good idea if he let readers know its contents: it is a thorough demolition of Badian's notion (argued in JHS 107  1-39) that there were two peaces of Callias. In his note on 2.41.1 (TH=S E(LLA/DOS PAI/DEUSIN) H. appeals to Chr. Habicht's privately printed 1988 Magie lecture to support his own contention that this does not mean that Pericles is describing Athens as a cultural center, and that Athens' reputation as such was a development of the fourth century. In fact, Habicht (the title of whose lecture is actually Hellenistic Athens and her Philosophers, not simply Hellenistic Athens as given by H.) is specifically interested in the philosophic schools. He argues that, "while most Greeks rejected what Athens stood for in the time of Pericles, her ambition and drive for political power, we come to see that the offering of Hellenistic Athens, her contribution to higher education, was universally accepted" (p. 22), which is not quite what H. is talking about.
The sorts of problems noted above are not all that serious. The problems that arise with regard to three other areas are. These include the intended audience, the explanation of Thucydides' Greek, and the treatment of textual problems.
In the preface, H. says that he intends this book to be "helpful to those students who are interested in the detail of Thucydides' thought and subject matter, but have little or no Greek" (v). To this end H. adds translations of the Greek lemmata that are drawn from "the corresponding passages of my revision of Jowett's Thucydides, in the World's Classics series" (also published by Oxford). As of August 1992, this revision of Jowett had not yet been delivered to the publisher, and even when it has been delivered and published, H.'s use of Greek for the lemmata, and of a great deal of untranslated Greek in the text, should be enough to deter most Greekless readers from using this commentary at all. What is such a reader supposed to make of something like the opening of this note on 1.98.4: "for the word E)DOULW=QH, 'deprived of its freedom' see above 2n.2 For the words PARA\ TO\ KAQESTHKO/S here rendered 'contrary to custom' [the word 'Greek' has been supplied to clarify the sense] literally means 'contrary to what was established'"?
The reader who does know Greek will be even more disturbed by H.'s tendency to let comparisons between translations stand in place of grammatical, rhetorical, and stylistic commentary, and by the general inferiority of the notes that he does provide to those offered by Gomme and other commentators. It is unfortunately quite common to find observations like that on 2.43.1: TH\N TH=S PO/LEWS DU/NAMIN KAQ' H(ME/RAN E)/RGW| QEWME/NOUS KAI\ E)RASTA\S GIGNOME/NOUS AU)TH=S which runs as follows, "what does AU)TH=S in Th. refer to? There are two feminine genitive nouns here [sic]. If the Athenians were being urged to become lovers of the power of Athens that would be an even more striking and aggressive idea. But Professor Dover kindly tells me that he thinks that in its context, after a series of statements about the city, the word AU)TH=S would naturally have been taken with Athens." As if there could be any doubt. This is preceded by a discussion of E)RASTH/S that draws attention to the use of the word in Aristophanes in such a way as to suggest that H. thinks that the implication of the word here is negative (citing Dover, Greek Homosexuality, 146) and points to the note that he will write on 6.54.2 where Thucydides describes Hipparchus' designs on Harmodius. Rusten more appositely points to the use of E)RASTH/S in philosophic contexts, where the meaning is "not a desire for bodily contact but love of moral and intellectual evidence (Dover, Greek Homosexuality, 157)."
To take only one more example, there is a similar linguistic problem in the note on 1.77.1. H. here gives the translation of KAI\ E)LASSOU/MENOI GA\R E)N TAI=S CUMBOLAI/AIS PRO\S TOU\S CUMMA/XOUS DI/KAIS KAI\ PAR' H(MI=N AU)TOI=S E)N TOI=S O(MOI/OIS NO/MOIS POIH/SANTES TA\S KR/ISEIS FILODIKEI=N DOKOU=MEN as "for because we found [my italics] ourselves at a disadvantage in law-suits against our allies, in cases controlled by inter-state agreements, and so transferred such cases to Athens where the laws are equal for all, we are supposed to be too fond of dragging people into court." He goes on to say, "I have followed the view which sees one set of cases here. The alternative rendering is to take the first clause as co-ordinate with the second 'although we have allowed ourselves to be at a disadvantage in cases controlled by inter-state agreements ... and have provided equal laws for all in those [different] cases which we have transferred to Athens...' This, however, ignores the different tenses -- present in the first clause, aorist in the second -- which make it natural to dissociate the function of the two halves of the sentence." H. means to say "for because we find ourselves..." or "because we are at disadvantage...," and that either the particles are coordinate (since we always lose these cases in their courts, we have made this other arrangement) or that E)LASSOU/MENOI is subordinate, explaining POIH/SANTES (since we always lost these cases, we did this). H. compounds his problem in this note by saying that [Xen] Ath. Pol. 16 shows how deeply aggrieved people felt about being dragged to court in Athens. The most important statement of this grievance is, however, not in [Xen.] Ath. Pol., but precisely in the next three sentences, as Gomme shows (HCT 1, 244) the passage in [Xen.] Ath. Pol. refers specifically to the rich and can be taken to imply that, "the repressed in the subject cities had some redress against the oppressor, whether fellow-countryman or Athenian." Furthermore, Thucydides' statement is specifically limited to cases involved with CU/MBOLA while [Xen.] Ath. Pol. 16 can be read to include other sorts of cases (such as those involving the death sentence) that are not at issue here.
H. promises a discussion of the text in volume 2. This is not very helpful since the reader needs to know what he thinks about the tradition at the start. Moreover, when H. does turn to textual issues (many of them minor and of no interest to the Greekless) his discussions do not bear close comparison with those in Gomme or other commentators. Here are a few examples:
At 1.2.2 (OU)/TE TH=| A/)LLH| PARASKEUH=|) H. offers "For the possibility that what Th. wrote was DIANOI/A|, "intention," see E.G. Turner, 'Two Unrecognized Ptolemaic Papyri', JHS 76 (1956), 98, citing K.J. Dover... However, it is not easy to see what this could mean: 'they had no great cities, nor did they think big in other ways'?" The problem here is that P. Hamb. 163 (third century BC) reads OU)/TE TH=I [A)/LLHI] DIANOI/AI, which "cannot be explained as due to a mechanical failure of the copyist" (Turner, p. 97). Moreover, the same reading appears in H (Paris gr. 1734), a manuscript which elsewhere appears to descend from an independent tradition that was not collated in the OCT (a serious lapse). Since P. Hamb. shows that this reading was in this tradition at a very early date, "this agreement ... makes it possible to argue that DIANOI/A|, as lectio difficilior, is the right reading" (Turner, p. 98). As Thucydides is arguing that the comparative weakness of early Greece made it impossible for the Greeks of that time to undertake operations comparable to the Peloponnesian war, the sense is less strained than H. thinks.
When H. comes to the crux at 1.2.6, DIA\ TA\S METOIKI/AS E)S TA\ A)/LLA MH\ O(MOI/WS AU)CHQH=NAI, he says only that "I would accept, with Gomme but against Stubbs and Marshall, that E)S TA\ A)/LLA ('in other respects') makes no sense, and that TA\ A)/LLA (meaning 'Greek states other than Attica') is needed." Gomme in fact thought that E)S TA\ A)/LLA could mean what the scholiast who wrote TO\ TH\N *E(LLA/DA KATA\ TA\ A)/LLA AU(TH=S ME/RH MH\ O(MOI/WS TN=| *A)TTIKH=| AU)CHQH=NAI TW=| PLH/QEI TW=N A)NDRW=N thought that it meant, though he also thought that it might be easier to change this to TO\ A)/LLO (cp. Thuc. 1.15.2). Otherwise, as Stubbs points out ["Thucydides 1.2.6," CQ 22 (1972), 76], Thucydides does use E)S TA\ A)/LLA to mean "in other respects" in five other places, two nearby (1.1.3, 1.6.4).
At 2.65.12, on TRI/A ME\N E)/TH, H. simply states that "the manuscripts have TRI/A, 'three', but this is obviously wrong," while duly noting Connor's suggestion that Thucydides had not included a number at all. Gomme, however, did not think that the universal agreement of the manuscripts here was "obviously wrong," and noted, with Bury, that " -- ME\N E)/TH suggests that another period is to be added, and that this is represented by KU/RW| TE U(/STERON..." This makes better sense of the Greek, and the view has recently been restated (with new arguments) by Rusten in his note on this passage. H. does not let his reader know about any of this.
At 3.17.1, after a good note justifying the retention of this passage in its present position, H. comments as follows on E)NERGOI\ KA/LLEI "'effective and in good trim'. The expression is harsh in Greek (though no worse than some expressions found in Th.'s speeches) and some word or words may have dropped out." It would have been preferable if H. had made some mention of alternative readings (e.g. Stahl's attractive KAI\ A)/LLH|) since the passage has some significance for our understanding of Athenian naval power.
At 3.30.4 H. misrepresents the tradition behind NOMI/SANTES OU)K A)/LLO TE EI)=NAI TO\ KENO\N TOU= POLE/MOU when he argues for Steup's KOINO/N. As Gomme shows, the phrase is proverbial, and the variant reading KAINO/N can be explained as a result of confusion in pronunciation. H. might also have benefited from the useful discussion of POLLA\ KENA\ TOU= POLE/MOU in Walbank's note on Polybius 24.16.3.
Whatever one thinks of the answers to these textual problems, it is important that they be discussed completely and competently if they are going to be discussed at all. Although it would be absurd to expect that it is possible to write a commentary that can satisfy everyone on every point, it is reasonable to expect that the author will try to represent the evidence accurately.
 See, for example, M.I. Finley, "The Fifth Century Athenian Empire: A Balance Sheet," in P.D.A. Garnsey and C.R. Whittaker, eds., Imperialism in the Ancient World (Cambridge, 1978), 103-26. H.'s discussion of this at 1.99 misrepresents the main point, which is precisely that Thucydides is not a completely reliable guide to the rise of the empire.  H. never explains his cross referencing system, so the reader will have to deduce that this means the note on 1.98.2.