With the reader's indulgence, I would like to take a leaf from O'Donnell's book and begin this review with an extensive quote from the Author of this commentary about his methodology and purpose, about the intended audience of his work, and his role in it:
The method has in the main been to allow Augustine to be his own commentator. . . . to perform this function in some obvious and straightforward ways for Augustine is an opportunity too long neglected. In performing it a commentator may hope to make some facts where none existed before, but he must respect the text and those who have worked on the text before him; and in this case, he must respect Augustine as well.
It is meant to be a working tool, contributing to dialogue, and has no aspiration to utter the final word. It is not a commentary for the general reader, and neither is it a commentary for the passive reader. My practice has been to refrain from commentary in my own voice wherever possible, and to allow the texts to speak for themselves. Wherever possible, quotation has been preferred to paraphrase, evidence to interpretation. The aim has been to give the reader the material with which to interpret, rather than to obtrude my own views. True enough, selection and arrangement have a way of directing exegesis, but the active reader will find ample resources for independent judgement. (1:lxii-lxiii, lxiv-lxv)
These are welcome words indeed, for they promise correctives to a series of abuses. For too long the Confessions have been studied in isolation from the rest of Augustine's works, and both in studies of the Confessions and across the board, Augustine has been treated as the least reliable or relevant witness to his own meaning. At another extreme, his works have been quarried for for decontextualized one-line zingers to serve as epigraphic points of departure for whole characterizations of the man and his era. The Confessions themselves have been dissected and subjected in whole and part to so many interpretative scrutinies that they have come to be more a vehicle for the exercise of modern ingenuity than a monument of ancient religious genius. As O'Donnell more soberly observes:
Philological scholarship takes its departure from one text and generates another, and the movement is all to often away from the object of the researches to the investigating subject . . . . That is the reason to write commentary rather than interpretative essay: to facilitate the movement past the commentator's words once again to Augustine's words -- to Augustine's life. (1:xxx)
But of course, neither with Augustine's words nor with his life is our contact as immediate as we might like. Whether in clarifying Homer from Homer or Augustine from Augustine, allowing texts to speak for themselves is neither so simple, antiseptic, nor passive a procedure as once was thought, and, as O'Donnell acknowledges (1:lxv), selection, arrangement, and interpretation go cheek by jowl. Even so, the commentary itself bears, I think, clear witness to the author's sincere intent and to the success of his method in achieving his goals. What improvement in our understanding of the Confessions requires is not so much more interpretation as more information, and this commentary surely supplies us with what we need. Its publication ought to be a watershed in the study of Augustine and his Confessions.
Despite O'Donnell's assertion that This is not the full philological, source-critical, historical, and philosophical commentary that has been a declared desideratum of scholars for more than a generation, each of these areas receives, in fact, generous attention. But the present observations will attempt to address some of the commentary's more narrowly philological aspects, more narrowly, because the term philological in its broadest, best, and most correct sense could be applied to the commentary as a whole. I will take briefly the topics of textual criticism, lexicography, and literary criticism.
The Text
The text is provided handsomely printed in Vol 1:3-205. Its only subdivision (with rare exception, 1:lxi) is according to the traditional paragraphing (with associated numeration) of the Confessions. Since there are no line numbers, cross reference between text volume and the volumes of commentary can be made only by lemmata, a procedure which works well for the close, active reading which the author demands, but is less precise and convenient than one might like for reference or casual consultation.
The history of the text is presented in a brief (1:lvi-lx) but prudent discussion in the introduction. Principal MSS are identified with remarks about date, provenance, etc.; the text tradition is indicated through M. Gorman's stemma; earlier printed editions receive a nod, with more attention devoted to those in current use (CSEL, BT, CCSL). There is no critical apparatus accompanying the text, but variant readings and conjectures are provided in the commentary in formats varying from simple report, to remarks adversarial and explanatory and extended discussion. The benefits of this are several: a clean text handsomely displayed, an informed and selective report of major witnesses to the text and of attempts at its improvement, a focus on selected text critical materials without the distraction and ennui that an exhaustive apparatus of variants might occasion, and an integration of text critical issues into the commentary with room for discussion that treats them as an integral part of achieving an understanding of the text. The obvious drawback of this procedure is the loss of the thorough but economically schematized conspectus of text critical materials that a good apparatus criticus provides, but again, this is more than compensated for by the selectivity and abundant contextualized discussion of variants and conjectures which integration of text critical material into the commentary provides.
Though the text offers no advance in recensio (1:lx), issues of palaeography and textual transmission receive more discussion than one might have anticipated (enough to placate, at least, those given to codicolatry), and the text itself has been subjected to a very thorough scrutiny leading to extensive correction and revision (carefully described and discussed ad loc. in the commentary). The author has thoroughly revised the punctuation of the text, displaying a welcome independence and attention to this too- often-neglected indicator of meaning and determinant of reading; punctuation is discussed in the commentary passim, in its own right or in connection with a preferred reading or a suggested interpretation of a passage. This text of the Confessions should be regarded as a new edition, for it is plainly the result of more thorough care, boldness, and thought than is apparent in very many scholarly productions to which the label editio noua has been attached.
A subsidiary text critical issue in this commentary is the text of the Latin Scriptures which are the stuff of such abundant allusion and quotation in the Confessions. The Vulgate has exercised a kind of tyranny, both in the transmission of Augustine's works through the Middle Ages (see 1:lxvi) and, until fairly recently, in our conceptualizing about Augustine's Bible(s). The author states as his goal the citation/quotation of scriptural texts in versions that come as close as possible to what Augustine would have known, and offers both a brief but clear account (1:lxx-lxxi) of the difficulties involved in identifying these versions and a conspectus of the witnesses to the scriptures known by Augustine, including, especially, Augustine's own quotations, the Vulgate serving as a version of last resort. Fidelity to Augustine's versions of Scripture is manifested in attention to specific usages that cast light on the choice of words in the Confessons (e.g., 2:147 Muscipula is common in A.'s Psalter, though rare in the Roman and Gallican versions . . . Augustinian Studies 25 (1994) 101-116. .), in abundant quotations of passages from Augustine's scriptures to unpack biblical allusions, and in virtual reconstructive quotations of whole sections of Scriptures (see, e.g., 2:429- 430, 3:91-92, 141, 344-345). O'Donnell's commentary sets a new standard of sensitivity to ancient texts of the Latin Bible in a work not wholly devoted to the recovery or study of Latin biblical texts, and the Index of Scriptural Passages (3:462-477) facilitates access to much Old Latin textual material unavailable elsewhere or not so conveniently available.
Lexicography
Not every commentary makes a contribution to lexicography, but in this area too the commentary goes far beyond its modest hope to make some facts where none existed before The lexicographical contributions are of several kinds: identification of unusual meanings and usages in the text of the Confessions, statistics and other information on usage of words/phrases within the Confessions, and contributions to Augustinian lexicography and orthography in the form of information on Augustine's usage elsewhere or the usage of other Latin authors. This information is provided for its own interest, to clarify meaning, and to establish readings and punctuation. The following varied specimens will give the reader an idea of what to expect:
consolabatur . . . consolari: Deponent, then passive; both uses classical, but the mixture is unusual. [2.331]
vel vitiose atque perverse: vel here is concessive: `quamvis, licet'. Cf. Hensellek, Anzeiger Akad. Wien 120 (1983), 106, with texts. [2:138]
sane: Virtually quidem (restrictive), `to be sure': Hensellek, Anzeiger Akad. Wien 120 (1983), 102-03; 11 X in all in conf. [2:207]
misereor CDGO Maur. Ver. Pell.: miseror S Knöll Skut. Vega. The question here is morphology rather than semantics. On available evidence, the following may be said of A's practice. . . . [2:156]
sursum corde: . . . The phrase sursum cor is liturgical . . . its presence here is a sure sign that the church as liturgical community is meant . . . [3:322-323]
ecce: 115 X is conf., evenly distributed; less common in his other works of sustained exposition (over a comparable amount of text in civ., 17 X, in trin., 24 X), but common in en. Ps. (85 X in one sample of comparable length). A spoken punctuation mark, adding emphasis. [2:29]
exaperit: Christian Latin (Souter cites Ex. 34:19 in one VL manuscript, and cf. Iren. 2.19.8 and Tertull. apol. 18.8 (= `translate'): `to disclose, disentangle, explain'. [2:143]
quondam: The word is rare in A. (60 X overall, 7 X in en. Ps., 7 X in civ., 3 X at most in conf. (and once, 1.16.26, in quotation from Terence)), much more frequently with adjectives or (as here) ps.- adjectivally with nouns than with verbs; cf. 9.13.37, `Monnicae . . . cum Patricio, quondam eius coniuge'. Wherever used, the meaning is consistent with that passage: `at one time'. In view of that the comma introduced after `quondam' by Skutella should be either deleted or moved to follow `Victorinus'; but the question presents itself whether quidam might not have been the original reading here (note 4.14.21, `Hierium, Romanae urbis oratorem', where the insertion of quondam would have been just as apt as here). That would be consistent with A.'s practice elsewhere (see on 3.4.7). [3.15]
humilitatis D2G1O1S Knöll Skut. Ver.: humiliatis CD1G2O2 Maur. On the rare verb form humilitare, see De Bruyne MA 2.561 (else in A. most likely only at cons. ev. 4.10.20, quoting Sirach 3:20 (cf. Amm. Marc. 30.4.1, `ad humilitandam celsitudinem potestatis')). All TLL refs. for verb forms from humilis are late antique. [3:89]
O'Donnell alludes to the troubled issue of Christian Latin, . . . (the question of `Christian Latin' as Sondersprache is ripe for fresh and venturesome treatment) . . . (1:lxiii), and many of his comments either consolidate or elaborate evidence for Christian Latin usage. For some interesting examples, see notes on animositas (2:117), antistite (2:338), collyrio (2:412), communionem (3:332), in celebritate sollemnitatum tuarum (2:159-160), incrassatus corde (2:397), promereri and offendere (2:375-376), sanctuarium (3:195), satis episcopaliter (2:323- 324), typhus (2:420-421). The substantial and frequent contributions of the commentary to lexicography, orthography, and syntax make the absence of an index of words commented upon an unfortunate lack. Again, this is to demand of these volumes more than is claimed and intended for the commentary, but it goes to show that the commentary can have a utility far beyond that claimed for it.
Literary Criticism
As far as I can tell, O'Donnell has no critical axe to grind save to give Augustine and his work due prominence, attention, and respect, and there is no noise of theoretical machinery as one goes through the commentary. He makes his points of departure plain in the Introduction: his views of what the Confessions are about (A Reading of the Confessions, 1:xxxii-xli) and how he situates the Confessions in the context of A.'s life and literary production (The Confessions in Augustine's Life, xli-lvi; note also the excursus Alypius, Paulinus, and the Genesis of the Confessions, 2:360-362). In the commentary the author's own critical orientation seems in evidence only in his toleration for the Confessions' multiplicity and in his restraint. His voice is heard mainly in the occasional witticism and in quotations from T.S. Eliot and Emily Dickenson.
Appropriate attention is given to structure. Commentary on each book is prefaced with a brief structural analysis, and time is taken at crucial points, e.g., beginning of Book 7, of Book 10, of Book 11, for discussion of structure.
An amount of attention is given to Scripture that seems unprecedented in a work not focused on the study of biblical texts or of exegesis. The multiple roles of Scripture in the Confessions, as object of commentary, as quotation, as integrating allusion, are identified and abundantly illustrated. Biblical texts are not merely identified, but generously quoted, both the texts themselves and Augustine's interpretations of them. Knauer's insights on Augustine's artistry in the use of Scripture are reported and elaborated upon, and whether through brief remark or copious quotation, the reader is guided towards a realization of the fundamentally biblical character of the Confessions. Another set of observations call attention to the intensification of the use of liturgical and specifically eucharistic language as the Confessions progress, with suggestions of a greater stylistic multiplicity in the work than has usually been supposed.
Stylistics is the subject of a formal recusatio in the introduction (1:lxiii). A brief survey of works on prose rhythm and style is provided (ibid nn125, 126), and, as the author suggests, much that is provided in the commentary (notes on lexicography, collocations of words, use of quotation and allusion to pagan and Christian texts, etc.) is of immediate relevance to stylistic analysis and appreciation of the Confessions, but I should like to have seen more of the author's views on stylistic considerations. It is unfair to demand yet more of an astonishingly conscientious commentator, but style is so very inseparable from every other sort of consideration, especially in the case of a peculiar work like Confessions, and I would have like to have heard less about neoplatonists, manichees, and the other usual suspects, and more about the Confessions' style, particularly from J.J. O'Donnell.
This commentary is a wonderful and momentous contribution. In it we have an approach to Augustine and his text that is constructive, rigorous, learned, and, if one may apply these terms to a contemporary scholarly publication without prejudicing many against it, respectful and honest. St. Augustine has been canonized, lionized, demonized, summarized, and trivialized, and the Confessions have been the locus and focus for much of this activity. O'Donnell's commentary will not put a stop to the silliness, nor, for all its learning and skill, is it the last word, indeed, because of its learning and skill it is not the last word. But if received and employed as intended, it offers a real opportunity for a fresh start in the understanding of Augustine and his most popular work.