The story of an occupying army in a foreign land fighting against a well-organized native insurgency ought to claim the interest of students and other readers these days. The volume under review caters to that potential market by reprinting the Oxford Classical Text of Gal. 7 and equipping it with an introduction, vocabulary lists, notes, and an appendix glossing proper names. It is intended “for anyone who desires to read Latin prose and has had at least basic Latin grammar/morphology” (p.vii). My remarks will, of course, have to be selective.
In the Introduction M(annetter) provides basic information about Caesar and his work, but the reader who knows about only one side is ill-equipped to evaluate a war-narrative. A section would have been welcome devoted to Gaul and the Gauls as historically known and as present in the Roman imagination, perhaps including a profile of Caesar’s worthy antagonist Vercingetorix. The section of the Introduction on the Commentarii as literature cites some studies of Caesar’s style and use of rhetorical devices, but students would have benefited from examples of the different devices, perhaps with a glossary of rhetorical terms for reference at the back. A section of the Introduction devoted to the date of composition and publication of the work would have been helpful, especially since on 6.1-2 M. cites Hammond’s claim that Caesar wrote this Book in late 50, which seems to be at odds with Caesar’s statement at 90.8 ( his litteris cognitis Romae dierum XX supplicatio redditur).
M. has chosen to break up the text into small units — sometimes less than a full sentence in length — which he deems “more approachable than large chunks of Latin” (p.vii). Thus the first page presents a four-line summary (in capital letters and in a larger font-size) of the content of chapters 1-13, then one line (actually more like 2/3 of a line) of Caesar’s text, 6 lines of vocabulary (there is no general vocabulary at the back of the book), then 49 lines of commentary, then another chunk of Caesar (a little over a line) followed by vocabulary etc. The result can be awkward, as when on 20.1-2 M. has to inform the reader that Vercingetorix is the “nominative subject of respondit,” which only appears as part of the next passage forty lines further down the page. This typography also gives too little prominence to Caesar’s text, which is in the same (tiny) font-size as the vocabulary and commentary and is distinguished only by being in boldface (as are the vocabulary items and lemmata).
In reprinting the OCT M. reproduces the italicization by which Du Pontet signalled conjectural supplements and the cruces indicating corruption. But M. does not explain these conventions, nor, given his intended readership, does he reprint the critical apparatus; so students will be puzzled as to why de is italicized in 1.1 or why words are enclosed in crosses at 15.3, 19.2-3 etc.; M.’s commentary usually ignores the brackets and cruces and proceeds as if they were not there. Best either to discuss textual problems systematically or eliminate these typographical pointers to their existence.
M. has taken the definitions in his vocabulary lists from Lewis supplemented by Kelsey; only “some specialized uses” were taken from the OLD (p.vii); he does not explain why. In fact, thoroughgoing use of the OLD would have yielded better results. Thus, to take just a couple examples, OLD s.v. anteverto 3 (“give priority to, deal with before”) fits 7.3-4 better than the definition given by M. (“prefer, take precedence”); and at 16.3 occurro means “check, counter” ( OLD s.v. 6), not “resist.” Closer attention to the OLD would also have helped the commentary at a number of points.
Peskett’s text and commentary on Civ. 1, a text roughly the same length as Gal. 7, was accommodated within 200 octavo pages;1 M.’s pages are much larger (9-5/8″ x 7-3/8″), but his book totals nearly 400 pages. How so? M. does not cross-refer to earlier discussions of the same grammatical phenomena or equip his volume with an index. Rather, compliments of the “copy” and “paste” commands of a word-processing program, M. repeats verbatim the same notes on the same grammatical phenomena from passage to passage. This is a procedure typical of online commentaries (and M.’s commentary is also available as an e-book) and more palatable there if the reader can avoid the repetitiveness by simply not clicking on an item already mastered. The danger in this approach is that general principles may misapplied.
There is no alphabetized list of authors and works cited. Thus each reference sends the reader on a tedious search through the footnotes to the Introduction, oddly placed not after the Introduction but at the end of the volume. Some fairly major works of secondary literature are surprisingly not cited at all,2 and some of M.’s other choices are questionable. M. frequently relies on J.F.C. Fuller’s biography of Caesar; though the major-general can be good on military matters (e.g., his remarks quoted on 86.1-3), for politics other literature is preferable.3 For realia M. relies heavily on the old commentaries of Kelsey and Walker (p.vii).4 But archaeology has not stood still since 1900; one wishes that M. had enriched his commentary by citing and reproducing more recent archaeological finds, such as the (named) portrait of Vercingetorix on an Arvernian stater.5 M.’s only illustrations are a drawing of the climactic moment of the siege of Avaricum and a map of Gaul (with place names so blurred as to be scarcely legible), both reproduced from Kelsey.
Basically a post-beginners’ commentary should mediate between the grammar that students have just learned and the particular text being read. The commentator needs to show students that their study of grammar reaps dividends in helping them comprehend a real Latin text. Most of M.’s mediation involves fairly low-level operations, identifying forms and functions and the like; many of these notes will be found helpful, and some show insight, such as the remark on 88.2 that “iconically, the noise of the battle ( clamore … clamor) encloses the battle scene.” There are problems, however, that potential users of this book should be aware of.
For his grammatical notes M. leans heavily and exclusively on Allen and Greenough (hereafter AG) notorious for obfuscation and fuzzy distinctions. Thus one of AG’s disservices is to blur the distinction between adverbial and substantival ut -clauses. M. carries this still further, giving “so that” as the standard translation for ut even when it introduces an object or result clause (on 4.5, 42.2-3 etc.); and on 42.4-5 M. wrongly refers impellit ut … pudeat to AG 531.1 (“pure clauses of purpose”) rather than the substantive clauses discussed at AG 563; and purpose and result clauses are confused at 27.3.
To the quirks of AG M. adds others of his own. For instance, when there is a third person verb with no expressed subject in the Latin M. repeatedly claims that a form of is is to be understood; for this he cites “AG 271a,” who, however, raise no such claim; what they say in that passage is rather that “in Latin the subject is often implied in the termination of the verb: — sede-mus, we sit” etc. ; M. makes a similar claim for impersonal verbs on 2.3 (and elsewhere: “supply id as the subject”) and a similar misrepresentation of AG (318c, who, rather, say “A passive verb is often used impersonally without a subject expressed or understood” etc.).
Particularly troubling is M.’s repeated misapplication of grammatical and rhetorical terminology. In the Introduction he speaks of hyperbaton as “the unusual position of words” (p.v), a bad sign. Elsewhere he calls it any “reversal of the normal word order” (15.4-6 etc.); perhaps he was misled by the vague definition at AG 641 (“violation of the usual order of words”). For his treatment of hyperbaton he cites Gotoff,6 but if he had read that passage carefully and studied the examples, he would have realized that, in fact, hyperbaton refers specifically to separation of grammatically linked phenomena; there is, in fact, an instance at 15.6, but M. does not signal it: defensores oppido idonei deliguntur, where defensores and idonei are separated; the upshot is that the term hyperbaton is misapplied in every single occurrence in this commentary. M. also pervasively misapplies the term tmesis (cf. on 14.2, 17.3-4, 25.4, 47.2-3, etc.), which refers to the separation of the elements of a compound (cf. AG 640). Another misunderstanding of M.’s has to do with “the predicate position” that he claims to find in some sentences; in fact, Latin has no particular position for the predicate (perhaps M. is thinking of the Greek phenomenon), nor does AG 272 103a Note 1, to which he refers, use the term; cf. on 38.8-9, 40.7, 41.2 etc. For M.’s difficulties with the objective genitive and with explaining the double dative see below.
M.’s grip on noun morphology is sometimes shaky: he claims that in 20.12 quem … ne qua civitas suis finibus recipiat qua is an abl. but yet modifies civitas; he should have looked at AG 149 for qua as nominative (contrast the n. on 69.7, where he gets it right); on 50.1-2 M. oddly calls virtute a dative, rather than an abl.; a look at OLD s.v. confido would have convinced him that this verb is used with the abl. as well as dat.
Problems of agreement also crop up. At 3.1-2 M. cites AG 282 to undergird the claim that C. Fufium Citam stands “in apposition to his” but ignores AG 281, where it is (correctly) laid down that an appositive must be in the same case as the referent. He oddly expects collocuti to be accusative even though it modifies the subject of a quod -clause (38.5-6). There is some confusion in the n. on 63.1-4: legationes is fem., so ei is not to be supplied with it, and it cannot be the antecedent of nacti; rather the members of the embassies must be supplied, a common constructio ad sensum.
Case usage is misclassified at a number of points. M. seems not to be clear that the objective gen. must depend on a noun with strong underlying transitive verbal idea in such a way that when the expression is reformulated with the corresponding verb, the original gen. becomes its object; thus at 17.3-4 rei frumentariae can hardly be called an objective gen.; it is rather the expected defining gen. with difficultas (cf. OLD s.v. 1b); nor can ignominiae be an objective gen. depending on loco at 17.5-8 or consistendi with locum at 37.3-4 or consili with principes at 37.6-7 or magistratuum with controversia at 39.2-3 or Romanorum with amicitiam at 63.7; and the nominal correlate of an intransitive verb cannot have an objective gen. at all ( pace M.’s claim apropos proventus at 29.3-4). The fact that AG 385c2 explicitly deny that the gen. with similis is an objective gen. does not hinder M. from making the claim (and citing that very passage in support of it!). He might have invoked the obj. gen. in other passages, e.g. apropos castrorum munitiones at 52.3, which he calls instead a possessive gen. The subjective gen., on the other hand, is not in M.’s vocabulary (perhaps because the explanation at AG 343 Note1 is so vague), though he might have found the term useful in explaining eius discessu at 20.1, multitudinis studio at 20.5, consilium totius Galliae (29.6-7; 63.5), Caesaris beneficio at 37.4, and remorum sonitu at 60.4 (similarly 61.3); M. mistakenly classes the last three as objective genitives; cf. also defectione Aeduorum (61.4 and 63.1), which M. calls a possessive gen.; the n. on 62.2 ( cuius ductu) oddly calls cuius“a limiting adjective” rather than the subjective gen. of the relative pronoun (and ductu is abl. of means, not manner). While referring to AG 346a1, which should have put him on guard, M. frequently confuses partitive and possessive genitive (on 11.5-7, 12.4-5, 14.3-4, 15.4-6, 59.6, etc.).
The explanation of the double dative is confusing (4.10; similarly elsewhere): ” reliquis documento: Dative of the purpose or end (double dative)”; M. ought to have said “dative of the person affected and of the purpose” or the like; he could have compared AG 382. M. overuses that vague and unhelpful category “the ablative of specification” when “ablative of means” or “cause” or “manner” would have more explanatory power for students (cf. on 1.5-6, 3.1-2, 21.1, etc.). Some of his other classifications seem to be dubious or wrong. M. sometimes invokes the abl. of manner when the abl. of means would be more plausible: e.g., diversisque itineribus at 16.3; in that same passage incommodo is the expected abl. with afficio (cf. OLD s.v. 4a), not an abl. of manner; and multo at 14.10 should be classed as an abl. of degree of difference (cf. AG 414), not an adverb.
Conjunctions and correlative adverbs provide important clues about the way the Latin is structured, but these are sometimes missed or misinterpreted. Thus at 2.2-3 there is no need to assume that et has “a slight adversative force”: Caesar presents three parallel actions taken by the Carnutes: profitentur, pollicentur, and petunt, the first two connected with -que, the third with et; and M.’s odd notion that et … et are disjunctives seems to have resulted from a misreading of AG 223a; in fact, M. should have cited AG 317c here, not 317b. At 17.5-8 sic anticipates, not the following acc.-inf., but rather the following consecutive clause; cf. 19.2-3 and 5-6 for additional examples of the structure, though M. fails to notice that these are consecutive clauses. At 20.4-5 neque correlates with the following et, thus creating a parallel structure; hence equitum … operam should be the subject of fuisse as well as desiderari; it is awkward to supply equitem as subject of fuisse and then equites as the subject of the following clause, as M. proposes. At 20.10-11 aut is used as equivalent of vel not in the stronger sense in which the alternatives exclude each other (cf. AG 324e, which M. cites for the interpretation as mutually exclusive but which also acknowledges the other usage). M. botches the explanation of the simple correlation ut … sic (“as … so”) on 30.3-4.
Inevitably, verbal syntax takes up a sizeable portion of the commentary. Transitive and intransitive deserve special attention; M. is confused when, on 67.3 ( consistit agmen), he claims that Caesar is the subject; in fact, this is the common (intransitive) military usage of consisto“to take up a position (for fighting)”: OLD s.v. 6b.
Tense usage receives considerable attention. Caesar’s work is, of course, very largely a narrative of past events, so one regularly encounters the various past tenses, the imperfect, perfect, and pluperfect, as well as the historical present. For the pluperfect M. most frequently notes “the pluperfect denotes an action completed in past time” (e.g. on 48.1-2); but this formulation does not differentiate pluperfect from perfect or account for its use in the particular context; usually the pluperfect supplies background information in contrast to the main narrative, as M. brings out on occasion (e.g., on 48.2) but should have done much more frequently. Likewise when Caesar uses the imperfect tense there is usually a point; the commentator should try to help students understand its force (e.g. a note on Caesar … recipi prohibebat at 78.5 showing that resistance to pressure is involved).
The moods are also dealt with at length. On 32.6 M. twice remarks that “the tense of the subjunctive is normally in secondary sequence after an unexpressed verb of saying,” but he really needs the qualifier “an unexpressed past verb of saying.” M.’s idea that the subjunctive is expected after quam in 1.8 is based on a misunderstanding. At 2.2-3 quoniam … possint is an example of virtual oratio obliqua, not attraction; it is no integral part of the ut … sanciatur clause. On the other hand, the quod -clauses at 20.3-4 are not virtual oratio obliqua, as M. claims; the subjunctives are rather the expected mood in dependent clauses in indirect discourse (cf. AG 580); and M.’s claim that the quod -clauses are “grammatically unrelated to the main sentence” is contradicted not only by this sentence, in which they function respectively as subjects of factum (esse) and persuasum (esse), but also by AG 572 Note. Subjunctive would have been required in quibus … cognoverint (5.4-6) even if it were not embedded in indirect discourse because it is causal (AG 535e); similarly quae … explorata at 20.7. The consecutive clause with limiting force seems not to be part of M.’s vocabulary, though he could have usefully invoked it on 37.4.
Issues of non-grammatical interpretation are broached from time to time. I wish it could be said that M., author of a dissertation on Caesar,7 is more reliable here, but unfortunately that is not the case. To cite just one instance: in a note on 78.3-5 (p.332) M. quotes (at second hand) Dio 40.40, reporting that the civilians expelled from Alesia died between the camp and the city. M. comments: “If the reader does not know this result, it would not be extrapolated from the text” (he evidently means “could not”). But this is wrong: the actions of the Gallic resistance in expelling them and of Caesar in refusing to accept them are clearly narrated; given the fortifications on both sides, there is no other possible inference than that they perished between the town and the besiegers.
Finally, far greater care was needed in editing and proofreading. I count, for instance, four typos on the very first page. The typos I have found that affect the Latin are as follows: advisarios for adversarios (4.3-5); proviciam (6.3-4), relniqueret (11.1-2), concitatem for concitatam (three times) (13.2-3); Avericum (16.2-3); prohibebat for -ant in the first note on 22.4-5; nostross on 22.4-5; curaret replaces curat in the note on 31.3-4; summum for summam in the note on 39.1-2; facles murales on 84.1-2. The notes are sprinkled with an assortment of baffling miscellaneous errors, of which I offer a brief sample: it is claimed (on 20.2-3 and 12) that sine means “with”; detrimentosum esse is called an impersonal verb (on 33.1-3); monet is declared to be a participle (on 45.8-9); the note on 46.1-2 confuses protasis and apodosis; on 54.1-2 through an editing error Viridomaro … Eporedorige were somehow substituted for ipsos as lemma and thus claimed as accusative; similarly on 71.6-9 M. meant to gloss eis but instead used qui … paverint as the lemma; on 77.8 cogentur is claimed as subjunctive, rather than future; on 80.3 expeditos is said to be nom.; on 80.4-5 superiores is said to be superlative; the Teutones become the Teutoni on 77.12 and in the Appendix; Balsdon becomes Ballston (p.327) etc. The idea of producing a commented edition of Gal. 7 usable by post-beginners is laudable, and I had hoped to be able to recommend this book. What M. has produced would, however, be a problematic choice as a classroom text. Most teachers will want to encourage students to gain fluency in reading, but that will be hindered by their having to search out embedded bits of Caesar. Teachers will also want students to be able to do more and more of the grammatical analysis themselves, rather than to be spoonfed from beginning to end. M.’s volume resembles an earlier generation of books that offered a text “completely construed” and, like them, will probably be used primarily for reference and private study, but even for such purposes caution is advised because of its errors.
1. Caesar, Bellum Civile I, ed. A.G. Peskett (Cambridge, 1934).
2. E.g., M. von Albrecht, Masters of Roman Prose from Cato to Apuleius, tr. N. Adkin (Leeds, 1989); A.K. Goldsworthy, The Roman Army at War 100 BC-AD 200 (Oxford, 1996); K. Welch and A. Powell, eds., Julius Caesar as Artful Reporter: The War Commentaries as Political Instruments (London-Swansea, 1998).
3. J.F.C. Fuller, Julius Caesar: Man, Soldier and Tyrant (New York, 1965); for Caesar’s relations with Pompey in the late 50s (on 6.1-2) M. could have referred instead to R. Seager, Pompey the Great, 2nd edn. (Oxford, 2002) 130-32.
4. F.W. Kelsey, Caesar’s Gallic War (Boston, 1898); A.T. Walker, Caesar’s Gallic War (Chicago, 1907).
5. Cf. C. Goudineau, César et la Gaule (Paris, n.d.) 212; for other possible illustrations cf. P. MacKendrick, Roman France (New York, 1972) 46-49 and 51.
6. H.C. Gotoff, “Towards a Practical Criticism of Caesar’s Prose Style,” ICS 11 (1984) 1-18 at 6-10.
7. D.A. Mannetter, Narratology in Caesar. Diss. (University of Wisconsin – Madison, 1995).