Bryn Mawr Classical Review

Bryn Mawr Classical Review 2002.10.05

Shackleton Bailey (trans.), Cicero, Letters to Friends. 3 vols (Loeb Classical Library 205, 216, 230).   Cambridge, MA, and London:  Harvard University Press, 2001.  Pp. viii + 497, vi + 477, vi + 475 + 3 of maps.  ISBN 0-674-99588-0, 0-674-99589-9, 0-674-99590-2.  $21.50 each.  

Reviewed by Andrew R. Dyck, University of California, Los Angeles (
Word count: 2968 words

D.R. S(hackleton) B(ailey) needs no introduction to readers of this journal. Since the first volume of Ad Atticum rolled off the press in 1965, he has produced an astonishing number of high-quality editions of classical texts, sometimes equipped with commentary. Though his work as an editor of poetic texts has suffered some cavils in recent years,1 his output is unabated since his retirement as Pope Professor of Latin at Harvard in 1988, as the Letters to Quintus etc., this set, and the two-volume Loeb Valerius Maximus (2000) most recently attest.2

The first volume of the set opens with an Introduction including a section on historical background reprinted from the 1999 Loeb edition of the Letters to Atticus, a section on Cicero's family, one on "manuscripts, text, and translation," a list of the abbreviations used in the critical notes (i.e., the names of the authors of conjectures) and a bibliographical note. Unfortunately, there is no list of sigla, though such is possible in a Loeb ed., as S.B.'s own VMax. shows; thus one must go back to S.B.'s Teubner ed. to unravel such items as vol. 2, p.454, n. 1: "l Quintio." Next follows (pp.21-34) a register of the friends (oddly placed; why not after the section on the family?)3 and finally the text and translation. The third volume concludes with an appendix on Roman dates, money, and names with a list of consuls for 68-43 B.C., a glossary of terms drawn mostly from Roman politics and public administration, a concordance to the standard numbering of the letters, an index (oddly combining entries in both the Greek and Latin alphabets), and maps of Asia Minor, Italy and Sicily, and Greece.

S.B. is no stranger to this corpus, having edited it twice previously, viz. the 1977 commented edition from Cambridge and the 1988 Teubner edition, the former sadly out of print. He explains (p.18) that this text is "almost the same" as the latter, the variations marked by asterisk in the critical notes; the Teubner in turn contains (pp.VI-VIII) a list of variants from the Cambridge. He also explains (p.19) that the translation "has been revised from" the one published in 1988 by the American Philological Association, which in turn reproduces the 1978 Penguin translation.

There is perhaps no perfect solution to the problem of the order of the letters, that of the MSS being arbitrary and unchronological. S.B.'s solution is about as good as is likely to be found. He sets up a roughly chronological framework in which the series of more or less contemporary correspondence with a single addressee is uninterrupted;4 the result is a continuously readable collection in which, with the help of the concordance, it is not difficult to find references based on the traditional numbering.

The critical apparatus is a model of concision. It is a negative one, with the manuscript reading cited, more Loebiano, in an attached footnote; the proponent of the adopted text is then indicated in parentheses. In most cases this works well; but if more than one word is affected, I would recommend adoption of positive apparatus for clarity; cf. 149.2, where going back to previous editions clarifies that tum quam is transmitted instead of tamen qu[odn]am.

The asterisks denoting textual changes since the last of S.B.'s editions yield insight into the development of his ars critica.5 In general it can be said of him, as Housman said of Bentley, "he grew bolder as he grew older," most of the changes being acceptance of conjectures previously relegated to the apparatus.6 Usually the result clearly improves the sense: thus at 6.3 he accepts Tyrrell's tuo for meo, the exiled orator now more logically wishing that Cicero jr. be "in sinu semper et complexu tuo" (sc. Terentiae); frequentem senatum is an improvement on the cumbersome frequentiam senatus, as is Manutius' patuit for paruit and munitus for inimicus in 282; and the second Hammonium is likely to be a gloss in 282.2 (as Manutius saw). Another attractive conjecture of Manutius' is cantorem, now adopted for unctorem at 261.2 (bellum tibicinem . . . et sat bonum cantorem being the more likely pair of attributes of Tigellius' nephew). Logic is improved by adopting Wesenberg's atqui for atque at 6.6 (a common corruption); at 201.3 the inferential itaque is not wanted; S.B. now adopts Ferrarius' atque; but atqui might be even better; the sequel requires S.B.'s suus for suo (typical of scribes to fall into the routine pattern!) in the salutation of 264. In 219.1 Cicero's contempt for envy seems more likely to be expressed with semper (D) than with the otherwise transmitted saepe. Perspicuity is enhanced by S.B.'s et at 107.2 (already mooted in the Teubner apparatus) and Manutius' et at 425.3 (tua virtute et magnitudine animi) as well as by Ernesti's in at 355.3, giving a different function to the following ablatives; at 206.3 Wesenberg's tu now effects a sharper contrast with the following mihi quidem . . .

Other clarifying additions are welcome: [aliquid] seems to be needed as the object of moliri at 387.1, better added before quod (S.B.), whence it can easily have dropped out by saltation, than after nos (Lambinus); at 36.1 (to Trebatius) S.B. rightly prints his own "qui omnia tua ferre possum," previously relegated to the apparatus; and at 227.2 S.B. now prints his own diutius in the text (he had previously mooted diu in the Teubner apparatus) to match the apparent cross-reference at Section 3 fin.

In some cases the new text better accounts for the state of the transmission: tamen qu[odn]am of the recentiores is closer to the ductus litterarum of the transmitted tum quam than Wesenberg's previously adopted tamen quod (149.2); in 279.1 Wesenberg's ut ad illam (sc. commendationem) hanc addam enables the transmitted ad to be retained; Q. Cicero in the heading of 351 will account for the transmission, divided between Q. (MDH) and Cicero (V).

Syntactic niceties often defeated the scribes' efforts: thus S.B. is probably right to strike, with Watt, in before meis litteris (86.1) and adopt Gronovius' ne for non in an asseveration before a pronoun (195.1), Manutius' posse m[e] at 197.1, and quin of the recentiores for quid at 273.2.

Analogy recommends Lambinus' q. in the salutation of 406, as well as Wesenberg's ingeni (restoring the same set of qualities as in 74.2) at 75; similarly the restoration of diuturnae (recentiores; the form elsewhere attested for Cicero), rather than diutinae, at 360.2.

On occasion one would gladly have seen suggestions of the notes incorporated in the text, e.g. the deletion of ut at 370.3 (first suggested in this edition); the restoration of the gentilicium Latinio in 417.7 to match the other two names in this list; or the change of in re publica to in rem publicam and its transposition to follow tuorum in 419.2 (first proposed here).

Proper names pose a special problem in these letters; given the limits of our knowledge, we must sometimes content ourselves with merely probable results. Rutilius' Pomponius for Pompeius better fits the behavior described at 43.2 (S.B. had previously called the conjecture "highly probable"7); and S.B. now rejects Böhm's Terentia for the transmitted Tertia at 185.1 on grounds that Cicero was not then on good terms with his ex-wife; there is likely to have been no question of their dining together. At 185.2 S.B. now sets Bellienus (V) in the text; the transmitted text is Billienus (so MD, corrupted to Bibli- in H). As a little joke between Cicero and Tiro this word is hard to interpret. Cicero is deputizing Tiro to deal with a certain disagreeable fellow named Demetrius Bellienus; I suspect (with Ernesti) that Cicero has deliberately misspelled the name as Billienus in order to suggest bilis ("black bile, ill temper"), which, to judge from the objections attributed to him ('etsi --', 'verum tamen --', 'de illis--'), seems to have been his attribute; the contrast with Demetrius of Phalerum is also apt: "he was never polished and elegant (like Demetrius) but now is a positive curmudgeon." Such jokes usually bear their meaning on their face; no need to assume that Tiro would have instantly thought of Caelius' letter to Cicero of three years previous and its mention of a murderer identified as Bellieni verna Demetrius (if that is the correct reading at 149.2) and that the current Bellienus' "murder" consists of boring people to death (a common metaphor in English, but in the Latin of Cicero's time?).8

There are other instances, too, where the changes of the new edition seem to be not for the better. In 74.3 the transmitted text is tam orba civitas; in the Teubner edition S.B. had set iam orba civitas (his own conjecture) in the text; now, perhaps because of the preceding nunc vero he prefers tam[quam] orba civitas. But Cicero uses orba res publica without the apologetic tamquam already at Red. pop. 11, so that it seems unnecessary in a letter written in 50. In fact, the transmitted text can stand; for tam orba see Flac. 54: usque adeo orba fuit ab optimatibus illa contio ut . . . or letter 355.3 [in] maxima orbitate rei publicae virium talium. At 81.1 the insertion nostri [oculi Curioni]sque (cj. S.B.) rests on pure conjecture (see S.B.'s note in the Camb. ed.); more judicious to dagger the passage, as previously. Goodyear's fotus for toto is now adopted at 110.6 but is not elsewhere used by Cicero in the required sense (cf. OLD s.v. 6); I would dagger, as in S.B.'s previous editions; and I am not convinced that [facere] cannot be understood in 411.3.

A few additional notes: at 16.1 no need to delete ea with Wesenberg; 20.18 the first posse surely can stand (S.B. follows Sternkopf's deletion) in spite of its recurrence (cf. no. 144.1). At 131.2 S.B. credits himself with writing hypothecas in Latin, rather than the transmitted Greek, letters; presumably this is done for consistency with hypothecis written in Latin letters (and case-form) in the following sentence. But was Cicero consistent in such matters? Evidently not since Greek words written in Greek letters are found aplenty in this corpus. More likely that Cicero wrote the word in its first occurrence in Greek (and that the scribes simply copied what they saw) and then changed to Latin alphabet when he had to use the word in a Latin case-form.9

Other suggestions: 73.11: surely we need perfecta, not profecta. At 88.2 we read: . . . exspectes quem ad modum exeat ex hac causa denique. invidiosum tibi sit si emanarit. In translating "It would not be good for your reputation if the thing leaks out" S.B. evidently takes sit as a subjunctive of cautious assertion, but this seems unlike the author, M. Caelius Rufus, who is nothing if not brash (cf., e.g., no. 153.2). I suspect that after denique we need something like vides enim quam invidiosum tibi sit si emanarit, then punctuate with comma rather than full stop so that the following subjunctives continue the thought. At 183.4 surely we should follow Frederking in deleting tamquam, not wanted before in patria and surely anticipated from the following tamquam in exsilio. Likewise at 190.7 the train of thought is helped if we adopt C.F. Hermann's etenim for the transmitted etiam.

At 95.3 S.B. prints satis gloriose triumpharem, non essem quidem tam diu in desiderio rerum mihi carissimarum and translates "My triumph would have been glorious enough; at any rate I should not have been so long cut off from all that is dearest to me," whereby he seems to take quidem as limiting ("at any rate"), but then one expects a pronoun or word of quantity for it to limit.10 Perhaps we need to insert si from the recentiores and transpose quidem to go with it: satis gloriose triumpharem, si quidem non essem tam diu in desiderio rerum mihi carissimarum ("I would have triumphed gloriously enough if I had not so long been cut off from all that is dearest to me").11 The si will have dropped out, as small words often do; the quidem was then removed and reinserted after essem by a reader who knew it could not stand first in its clause and took it as contrasting with the following sed.

According to the transmitted text at 222.2, Cicero assures Trebianus cottidie magis mihi delabi ad aequitatem et ad rerum naturam videter (sc. Caesar). Since the Camb. ed. S.B. has obelized rerum and proposed veram suam in the apparatus (this is also the text he translates). In the note ad loc. in the Camb. ed. he points to 227.2 as support: nam et res eum cottidie et dies et opinio hominum et, ut mihi videtur, etiam sua natura mitiorem facit, as well as other passages connecting Caesar's clementia with sua natura. But in none of these passages is there mention of vera sua natura. Now S.B. admits (loc. cit.) that "delabi ad suam naturam would . . . be an unusual expression"; hence the insertion of veram. But the iunctura vera natura occurs nowhere in Cicero;12 it would imply a falsa natura, an impossibility.13 In fact, the transmitted text can stand; cf. 226.4 (a similar letter to another exile): scis me antea sic solitum esses scribere ad te magis ut consolarer . . . quam ut exploratam spem salutis ostenderem, nisi eam quam ab ipsa re publica, cum hic ardor restinctus esset, sperari oportere censerem. Cicero, then, evidently regarded the res publica as the rerum natura, to which the polity would revert cum hic ardor (of the recent civil war etc.) restinctus esset.

In the first sentence of no. 317 one might have expected the praenomen of the father to have been indicated (just prior to tui necessari) given his string of descriptive epithets; perhaps an L. has fallen out following the E of unice (the name is attested at CIL I.2.2.835, where Lommatzsch mooted identification with either the father or son mentioned in our letter).14 In the next sentence of the same letter read perhaps em hic est de illis maxime qui irridere atque obiurgare me soliti sunt, whereby I follow S.B.'s deletion of ille following hic and change the verb of the relative clause to plural.

S.B. first deleted and now (since the Teubner ed.) daggers vel probabilem in the sentence qua re etsi minus veram causam habebis, tamen vel probabilem aliquam poteris inducere (427.2). Might one consider si vel[is]?

It is hard to know what sort of Latinity to attribute to some of Cicero's correspondents, given the smallness of the sample. Thus at 256 one might have expected quasi vero non iustissimi triumphi dignas in Dalmatia res gesserim; this would put the author, Vatinius, in the same category as Balbus (Att. 8.15a.1) as a user of dignus + genitive; cf. OLD s.v. 1a.

Misprints are few: nunc miser run together in 6.5; 32.1 φιλοθέωρον, not ψιλοθέωρον; 71: in the date add "Id."; 73.1 (1, 318, l. 1) read maximam; in the heading to no. 107 XII Kal. Ian. seems to have fallen out after paulo post; 188.3 should read civis e re publica (as in previous edns.), not a re publica; 294.2 read iis, not its in Latin text; 366 n. 1 read Murcus, not Marcus; a stray bracket seems to have crept in at 405 n. 2; at vol. 3, p. 386 (= 425.2) after the last line the following matter has fallen out: tantum conatus audiuntur, optimi illi quidem et praeclaris(simi).

The annotation is, as it must be in this series, lapidary; but S.B. has the advantage of being able to refer the reader, where necessary, to his detailed Cambridge commentary.15 Rarely can one take issue with him: in view of the fact that res sacrae could not be alienated, the Minerva that Cicero hopes to obtain from App. Claudius can hardly be the one he dedicated on the eve of his exile (a possibility raised by S.B. at 64.1, n. 4); he is evidently seeking a replacement. At 72 n. 3 S.B. cites Sext. Rosc. 111 as evidence that Cicero took an instrumental view of friendship "in his younger days"; perhaps he did, but it is worth remembering that Cicero cautioned that his speeches in advocacy need not show his true views (Clu. 139). At 205.1 a brief note identifying Caecilius Bassus (cf. S.B. on Att. 363.3 of the Camb. ed.) would have been helpful; similarly on 346 an indication of the province Cornificius is governing (or a cross-reference to no. 373). Occasionally a reference to an earlier edition has been retained; thus nowadays at n. 2 on 385 one would have expected to see Var. Men. fr. 36 Astbury cited rather than p. 103 Riese.

The rendering of so various a collection poses special problems. The letters to powerful members of the aristocracy tend to be as elaborately polished as Cicero's published writings; here "the high sentiments, stately flatteries, and courteously veiled rebukes might have transposed naturally into eighteenth-century English, but put a modern translator at a disadvantage," as S.B. remarks (vol. 1, p.3). It is nonetheless hard to imagine how these could have been rendered in a more elegant and supple form than S.B.'s "sparkling" English.16 Connoisseurs will relish such points as the rendering of ruinae periculum as "booby trap" (237.3). Only rarely are there oddities that may give pause to some readers, e.g. the rendering of tumultus (127.3) as "alarms and excursions."

The current set not merely supersedes but completely eclipses the old W. Glynn Williams edition (3 vols., 1927-29); it is eloquent testimony to the regeneratio imperii underway at the Loeb Classical Library, which is increasingly offering cutting-edge texts partnered with elegant translations and parsimonious but helpful notes. Now that the letters have been so well catered for, dare one hope that other items of the Ciceronian corpus will receive new editions as well?


1.   Cf. the critical but by no means onesided remarks of H. Tränkle, "Von Keller-Holder zu Shackleton Bailey. Prinzipien und Probleme der Horaz-Edition," in Horace, L'oeuvre et les imitations. Un siècle d' interprétation, ed. W. Ludwig, Entretiens Hardt 39 (Geneva, 1993), 11 ff., with literature.
2.   The former reviewed by S.M. Goldberg, this journal 2002.08.10, the latter by H.-F. Mueller, ibid. 2002.06.35.
3.   It would have been helpful if this register of friends could have been repeated at the front of each of the three volumes, since readers may not wish to carry the whole set with them.
4.   As in the Cambridge edition and in the translation, but unlike the Teubner edition.
5.   However, the asterisk at 347.1 is falsely placed: the text printed here is the same as in the Teubner ed.
6.   M. Annaei Lucani Belli civilis libri decem editorum in usum ed. A.E. Housman (Oxford, 1970; orig. 1926), xxxii, citing Napoleon on Turenne. -- An exception is 16.1, where S.B. now defends the transmitted nimium as a colloquialism by reference to Leg. 1.27 (he had previously adopted Weinhold's inimicorum).
7.   Onomasticon to Cicero's Speeches, 2nd ed. (Stuttgart, 1992) s.v. Pomponius, Cn.; cf. R. Johannemann, Cicero und Pompeius in ihren wechselseitigen Beziehungen bis zum Jahre 51 v. Chr. (Emsdetten, 1935), 85, who denies that there was a close friendship ("eine wirklich nähere Freudschaft") between the two.
8.   S.B. argued his interpretation of the passage at PCPhS 5 (1958-59), 11-13; cf. now I. Oppermann, Zur Funktion historischer Beispiele in Ciceros Briefen, (Munich-Leipzig, 2000), 76-81 with further literature.
9.   On the whole question of Cicero's use of the Greek alphabet in the letters cf. G. Nieschmidt, Quatenus in scriptura Romani litteris Graecis usi sunt (Marburg, 1913), 36-39.
10.   Cf. examples at J.B. Solodow, The Latin Particle quidem (Boulder, 1987), 108-9.
11.   For emphatic quidem following si even in contrary to fact conditionals cf. ibid. 127.
12.   According to a search of the Packard Humanities Institute CD-ROM 5.3.
13.   For Ciceronian usage of natura cf. E. Zellmer, Die lateinischen Wörter auf -ura, 2nd ed. (Frankfurt a.M., 1976), 23-26. Perhaps the closest he comes to the idea is the quasi altera natura of N.D. 2.152.
14.   Cf. H.-O. Kröner, RE Suppl. 10 (1965) s.v. Precilius 1.
15.   Sometimes indeed he takes the opportunity to withdraw an idea mooted there: cf. n. 1 on no. 263.
16.   G.O. Hutchinson, Cicero's Correspondence: A Literary Study (Oxford, 1998), 1 n. 1.

Read Latest
Index for 2002
Change Greek Display
Books Available for Review

HTML generated at 13:28:11, Friday, 03 April 2009