Book Four

4.1.1-4.3.6

Teaching (at Carthage)

4.4.7-4.12.19

The death of a friend (at Thagaste)

4.13.20-4.16.31

Intellectual development (at Carthage)

4.1.1-4.3.6 portray A. at work as teacher, the prey of the variae cupiditates (4.1.1, sc. carnis et oculorum) to which he had yielded in the course of the narratives of Bks. 2 and 3. As G-M remark, `artem rhetoricam' at 4.2.2 means that these paragraphs deal with his career at Carthage; the rhetorical contests, with awards by the proconsul, such as those described in 4.3.5--and implied by 4.1.1, `agonem coronarum' --were surely held there. We must therefore conclude that this book is made up of reminiscences of Carthage (376/83) framing the Thagaste episode (375/6) in the mid-section of the book (see on 3.11.19, `vivere mecum').

4.4.7-4.12.19 recount the death of his friend and present an extended meditation on its meaning. The rustication to Thagaste may have been half-involuntary: cod. theod. 14.9.1 of 12 March 370 (dealing, to be sure, only with Constantinople and Rome¹ ) decreed that students could remain at Rome only `usque ad vicesimum aetatis suae annum'. It may also have been correspondingly brief (our only other information on that period is that Alypius first studied with A. then: 6.7.11), thereby partly accounting for the limited place this phase of A.'s career plays in his narrative. (R. A. Kaster, Guardians of Language [Berkeley, 1988], 129, emphasizes the infrequency with which grammatici ascended the ladder to become rhetoricians, as A. did on moving from Thagaste to Carthage. A.'s employment as a grammaticus was thus perhaps a temporary expedient.)

4.13.20-4.16.31 depict his intellectual life at Carthage, describing the circumstances surrounding the writing of the de pulchro et apto (4.13.20-4.15.27), then recalling an earlier, undated intellectual feat, his reading of the categoriae of Aristotle (4.16.28-31). The book is thus loosely bracketed by two acts of successful interpretation of difficult texts: at 4.3.5, when Vindicianus is presented as a student of astrology, and at 4.16.28, when A. reads Aristotle on the categories.

The structure of Bk. 4 offers L. Verheijen in Collectanea Augustiniana (New York, 1990), 175-201 (the same article in French in V.'s Nouvelle approche de la Règle de saint Augustin [Louvain, 1988], 2.92-114) opportunity for stimulating speculations about A.'s habitual patterns of constructing literary works in tripartite ways.

4.1.1

The construction of the paragraph is unusually artful. The first sentence is remarkably ornate, while the ring composition from `inrideant et confitear' to `sed inrideant . . . autem confiteamur' sets off a passage reminiscent of 1.1.1.

annorum novem: In all, from 372/3 to 381/2. For the first date, see on 3.4.7; for the chronological problems, see on 5.6.10.

seducebamur et seducebamus: The antitheses here are repeated in deliberate coordination through `per quos liberaremur,' juxtaposing A.'s public life as teacher of traditional rhetoric and his private activities on behalf of Manicheism. n.b., `palam . . . occulte', `hic . . . illic', `hac . . . illac'. Cf. Io. ev. tr. 8.11, `verumtamen seducti [mathematici] seducunt'.

falsi atque fallentes: Cf. 7.2.3, `deceptos deceptores' (to the analogous texts there add Macrob. somn. Scip. 1.6.64, `Hippocrates . . . tam fallere quam falli nescit'). quant. an. 33.76, `vanitas vanitantium. vanitas enim est fallacia, vanitantes autem vel falsi vel fallentes vel utrique intelleguntur'; c. acad. 2.5.12, `homini enim homo falsus docendus, fallax cavendus debet videri, quorum prius magistrum bonum, posterius discipulum cautum desiderat'; c. litt. Pet. 3.17.20, `vel falsus vel fallens', sim. (usu. fallax instead of the participle) at duab. an. 1.1, trin. 3.7.12 (`ad fallendos fallaces'), civ. 9.18, 11.13, retr. 1.7.1 (`manichaeorum iactantiam de falsa et fallaci continentia vel abstinentia').

doctrinas . . . liberales: See excursus on 4.16.30; the book is bracketed by the `liberal arts.' The form of expression, which recurs there, is somewhat pejorative, as Pellegrino saw (reviewing Courcelle, Les Confessions in Studi Medievali ser. 3, 4[1963], 651): sim. at s. Guelf. 24.4, s. 133.4, and ep. 101.1 (`litteris illis, quas variarum servi libidinum liberales vocant'), and cf. 4.3.4, `quos mathematicos vocant'.

occulte: Imperial legislation, tempered by neglect, commended prudence; compare the case of Faustus (c. Faust. 5.8): accused by Christians, brought before the proconsul, condemned; at the request of his accusers, his punishment was mitigated to relegation to an island, from which not long after he was released by a general amnesty of the emperor. (See on 5.3.3.)

superstitiosi: See on 3.6.10, `incidi'.

vani: Vanus/vanitas occurs 8x in Bk. 1, particularly of A.'s educational hopes and ambitions; recurs in the same context twice in Bk. 3, and now occurs repeatedly in Bk. 4: note especially that the address to his soul at 4.11.16 begins thematically, `noli esse vana, anima mea'. It is then a constant refrain until 8.11.26, just before the garden scene: `retinebant nugae nugarum et vanitates vanitantium'. Thereafter it appears mainly in the reading of Psalm 4 (9.4.9, where Ps. 4.3 reads `ut quid diligitis vanitatem') and in the examination of conscience under the headings both of curiositas (10.35.54-57) and ambitio saeculi (10.38.63, `de ipso vanae gloriae contemptu'). Thereafter it is rare, but note that at 11.11.13 (`cor eorum volitat et adhuc vanum est'), the reference is to Manichean opponents. See L. Chevallier and H. Rondet, REAug 3(1957), 221-234, esp. 222.

hac hac Maur. Knöll Skut. Ver. Pell.:   hac Z   (`h' sub linea):   ac C D G O S

gloriae . . . inanitatem [1] . . . spectaculorum nugas [2] . . . intemperantiam libidinum [3]: His public life alone offers a comprehensive display of the vices of 1 Jn. 2.16.

useque ad theatricos plausus: See below, 4.2.3, `theatrici carminis certamen'.

agonem coronarum: 4.3.5, `coronam illam agonisticam'.

faenearum: G-M, `perhaps “quickly fading”; . . . cf. Sueton. Nero xii “orationis carminisque latini coronam.”' An image of frailty, as at 9.7.16 (where see notes), `quantum patet aura in domo faenea' (cf. Is. 40.6, `omnis caro faenum'); sim. at 13.15.18.

nugas: See on 3.10.18.

sordibus: See on 3.1.1, `sordibus concupiscentiae'.

electi et sancti: en. Ps. 140.10, `qui sunt isti electi? quibus si dixeris: peccasti; statim illam defensionem impiam et peiorem ceteris magisque sacrilegam proferunt: “non ego peccavi, sed gens tenebrarum.” quae est ista gens tenebrarum? quae bellum gessit cum deo.'

angelos: See on 3.10.18, `anhelaret de illa angelos'.

inrideant: Again below here; see on 1.6.7.

elisi: 11.31.41, `tu enim erigis elisos'.

dedecora: See, e.g., 2.4.9, `defectum meum ipsum amavi, . . . non dedecore aliquid sed dedecus appetens [sc. in concupiscentia carnis].'

laude: Ps. 105.47, `salvos nos fac, domine deus noster . . ., ut confiteamur nomini sancto tuo et gloriemur in laude tua.' Knauer 80n2 sees in laus dei a theme uniting the first half of Bk. 4.

circuire . . . immolare: Ps. 26.6, `circuivi et immolavi in tabernaculo eius hostiam iubilationis'; en. Ps. 26. en. 2.12, `quid ergo immolamus? abundantissimum et inenarrabile gaudium, nullis verbis, voce ineffabili. haec est hostia iubilationis.' At sol. 2.20.34 and quant. an. 4.6, 7.12, circuitus is used of the roundabout ways of argument that leads to vision of higher things, hence with almost a positive sense; here, however, cf. 8.2.3, `narravi ei circuitus erroris mei.' On imperatives at the beginning or end of sections of the text, see Knauer 64 and 64n1 and (esp. on this passage) 72.

circuire: circuire O2 S Knöll Skut.:   circumire C D O1 Maur. Ver.

praesenti memoria: = contuitus [2] (see on 12.15.18, `expectatio . . . contuitus . . . memoria'), and see excursus on 10.8.12, `Memory in Augustine'.

praeteritos circuitus: = memoria [1] (or rather its contents: the past).

immolare tibi hostiam iubilationis: = expectatio [3] (the future life of ceaseless praise).

quid . . . sum: The question is asked twice: once of what A. is like without God, once of what he is like when he is with God.

dux in praeceps: 2.3.7, `et praeceps ibam tanta caecitate'. For a better dux, see 2.7.15 and 7.10.16 (`te duce').

sugens lac tuum: Cf. Deut. 33.19, `populos ad montem vocabunt; ibi immolabunt victimas iustitiae qui inundationem maris quasi lac sugent et thesauros absconditos harenarum.' Is. 60.16, `et suges lac gentium et mamilla regum lactaberis.' These references echo the vocabulary here, but are not particularly apt; better, 1 Cor. 3.2-3, `lac vobis potum dedi, non escam, nondum enim poteratis. sed ne nunc quidem potestis, (3) adhuc enim estis carnales.' Recall the suckling child at 1.6.7, and cf. 7.18.24, `ut infantiae nostrae lactesceret sapientia tua,' and 13.18.23, `tamquam parvulus in Christo lactisque potator'. R. J. O'Connell, Thought 48(1983), 188-206, emphasizes the maternal qualities imputed God by this image, and refers to numerous passages in Isaiah, but his treatment of evidence is entirely unsatisfactory. The image gives rise to the observation of R. Brändle and W. Neidhart, Theol. Zschr. 40(1984), 166n67: `Augustin findet in seinem Suchen nach Gott nur selten Begriffe der Väterlichkeit. Viel häufiger sind Attribute oder Bezeichnungen für Gott, die in die Sphäre des Weiblichen gehören.' But A. was far less insistent on drawing gender lines than are moderns: for example, at 13.22.32, A. applies the same image to Paul: `ad hoc enim dispensator ille tuus generans per evangelium filios, ne semper parvulos haberet, quos lacte nutriret et tamquam nutrix foveret.' (And see on 9.10.23, where Monnica has a masculine role to play.)

cibo qui non corrumpitur: Cf. Jn. 6.27, `operamini non cibum qui perit, sed qui permanet in vitam aeternam.' Cf. 3.1.1, `fames mihi erat intus ab interiore cibo'.

et quis homo est: G-M: `and what kind of a man is any man when he is merely a man? i.e., who is truly a man when he is only a man?'

fortes et potentes: Cf. 1 Cor. 4.10, `nos stulti propter Christum, vos autem prudentes in Christo; nos infirmi, vos autem fortes, vos nobiles, nos autem ignobiles.'

infirmi et inopes: Cf. Ps. 73.21, `egenus et inops laudabunt nomen tuum'; other echoes at 10.38.63 and 11.2.3.

4.2.2

artem rhetoricam: Possidius v. Aug. 1.2, `nam et grammaticam prius in sua civitate et rhetoricam in Africae capite Carthagine postea docuit.' On the implications for dating, see on 4.1.1.

loquacitatem: See on 1.4.4; loquacitas is apt against himself Manichee and merchant of verbal facility.

cupiditate: If in 4.1.1 the cupiditatibus denote the temptations of the flesh and of the eyes, here only ambitio saeculi is meant.

tu scis: See on 1.5.6, 3.3.6, etc.; Knauer 76-77.

caput innocentis: Cic. off. 2.14.51 (cf. Testard 1.44ff, 2.25), `atque etiam hoc praeceptum officii diligenter tenendum est, ne quem umquam innocentem iudicio capitis arcessas; id enim sine scelere fieri nullo pacto potest. nam quid est tam inhumanum quam eloquentiam a natura ad salutem hominum et ad conservationem datam ad bonorum pestem perniciemque convertere?'

de longinquo: On the remoteness of God, see on 1.18.28, esp. echoing here again Lk. 15.13, `in regionem longinquam'.

fidem meam: Here and just below, the only occurrences of fides in conf. without some Christian overtone; here, to show the lingering traces in his life of something absent in every essential way.

diligentibus vanitatem et quaerentibus mendacium: Ps. 4.3, `filii hominum, usquequo graves corde? utquid diligitis vanitatem et quaeritis mendacium?'; see on 9.4.9. A. twice wrote on lying (mend. of c. 394/95 and c. mend. of c. 420), taking an absolutist position (n.b. c. mend. 20.40: the catechumen prisoner of infidel guards, about to be put to death, is not to lie in order to gain a chance at baptism), which has discomfited later moralists (recently, S. Bok, Lying [New York, 1978]); the harsh judgment of his own professorial past is intimately linked with this attitude.

unam habebam: For the legalities and sentimentalities of A.'s `marriage', see on 6.15.25.

sane: virtually quidem (restrictive), `to be sure': Hensellek, Anzeiger Akad. Wien 120(1983), 102-103; 11x in all in conf.

proles: Adeodatus was `annorum . . . ferme quindecim' at the time of his baptism in the spring of 387 (9.6.14) and `in annis sedecim' at the dramatic date of mag. not long after; on this calculation, he was born 371/72, when A. was perhaps 17. The words `in illis annis' thus reflect another of the chronological inconsequences of conf., inasmuch as they suggest a liaison that must have begun in 371 or perhaps even 370, thus apparently probably in the first year of studies at Carthage but conceivably during the year of indolence recorded at 2.3.5-6 (the philoprogenitive optimism of Patricius did not have so long to wait). Adeodatus's mother was dismissed from Milan and returned to Africa in 385/6 (see on 6.15.25), and thus shared A.'s outward peregrinations through his entire adulescentia.

contra votum nascitur: A. accused the Manichees of resisting marriage and children, hypocritically: `filios autem inviti suscipiunt' (c. Faust. 15.7; almost the same words at c. Faust. 20.23). Brown 39 offers a delicate suggestion: `Next, his son, Adeodatus, was born: this event, unwelcome at the time, may well have had the “sobering” effect which Augustine would later recommend to young husbands [b. coniug. 3.3].' But A. had something specific in mind in that passage: `deinde quia reprimitur et quodam modo verecundius aestuat concupiscentia carnis, quam temperat parentalis affectus.'

4.2.3

A.'s reluctance to sponsor sacrifice is not a sign of personal delicacy, or anything better (`non ex tua castitate repudiavi'): mor. 2.17.54, `ab animalium nece . . . vos temperatis', reminds us that it was against Manichean doctrine to kill any living thing (see Alfaric 220-221 on this passage). As Knauer 72-73 observes, this kind of sacrifice, at one time an option A. considered for propitiating his god, exemplifies the context in which we are to take, e.g., 5.1.1, `accipe sacrificium confessionum mearum'.

certamen . . . ut vincerem: The temptation is that of ambitio saeculi [1]: vera rel. 53.103, `quare qui fines ipsos desiderant, prius curiositate [2] carent. . . . deinde accipiunt actionis facilitatem [1] pervicacia posita scientes maiorem esse facilioremque victoriam non resistere animositati cuiusquam'.

mandasse mihi: G-M and Hrdlicka 78 render `mandare' = `to send word' and cite Eutrop. 5.5, `[Mithridates] senatui mandavit bellum se ei . . . inlaturum'; but the usage is classical (Tac. ann. 14.38, `mandabat in urbem, nullum proeliorum finem exspectarent'). Souter more usefully suggests merely `send (a messenger, etc.)' and adds `(Tert. on.)': on that interpretation, here render `sent (sc. to ask) . . .'.

haruspicem: Strictly the haruspices were interpreters of prodigies (as at civ. 2.24, 3.11), but A. knew them as sources of magic power as well (en. Ps. 73.18, 134.20, doctr. chr. 2.20.30, s. 9.11.17 [linked to criticism of spectacula]).

respondisse: The claim is corroborated at 10.35.56, `nec anima mea umquam responsa quaesivit umbrarum'.

ex tua castitate: God's, not man's, as Madec rightly emphasizes at REAug 7(1961), 246, where he castigates the French translators for their unwilingness to predicate chastity of God; the same is true in English: cf. even Ryan, `it was not out of a chaste devotion to you'; better `not by means of the chastity that comes from you'. Io. ep. tr. 4.9, `ergo castificat nos sicut et ipse castus est; sed ille castus aeternitate, nos casti fide.' The metaphor is counterposed to fornicari abs te (see on 1.13.21; the phrase recurs here and at 5.12.22); but perhaps it is stronger than metaphor: beata v. 3.18, `quomodo, inquit, castus potest esse qui ab inlicito tantum concubitu sese abstinens ceteris peccatis non desinit inquinari? ille est vere castus, qui deum attendit et ad ipsum solum se tenet' (words of Adeodatus, but strongly approved by A. ad loc.). 4.3.4 uses salubritas to denote roughly the same purity.

deus cordis mei: Ps. 72.26-27, `defecit cor meum et caro mea, deus cordis mei, et pars mea deus meus. (27) ecce qui longe se faciunt a te, peribunt. perdidisti omnem qui fornicatur abs te.' See also on 6.1.1, `deum cordis mei'; cf. 9.13.35.

non . . . noveram . . . non noveram: recalls the repeated use of this phrase at 3.7.12-13, and reiterates one of the ignorances confessed there: 3.7.12, `non noveram deum esse spiritum'; Bk. 4 is haunted by the false god: 4.7.12, `vanum phantasma et error meus erat deus meus.'

fulgores corporeos: c. Faust. 22.8, `non distinguunt inter lucem quod est ipse deus et lucem quam fecit deus'.

figmentis: `products of artifice': poetic (1.13.22, 1.17.27), Manichean doctrine (3.6.10, 4.15.27, 4.16.29), his own misconceptions of catholic teaching (6.3.4, 10.34.53), and, once only in a strong metaphor, man as God's creation (13.23.33, `quoniam tuum sumus figmentum creati in operibus bonis'). trin. 4 pr. 1, `ego certe sentio quam multa figmenta pariat cor humanum. et quid est cor meum, nisi cor humanum? sed hoc oro deum cordis mei, ut nihil ex eis figmentis pro solido vero eructuem in has litteras'. Frequent in civ. in negative sense, esp. `figmenta poetarum' (e.g., civ. 2.8).

fornicatur abs te: Cf. Ps. 72.26, quoted above; see on 1.13.21; en. Ps. 72.33, `huic fornicationi contrarius est amor castus. quis est amor castus? amat iam anima sponsum suum.'

fidit in falsis et pascit ventos: Prov. 10.4, `qui fidit in falsis, hic pascit ventos'; Os. 12.1, `Ephraim pascit ventum et sequitur aestum; tota die mendacium et vastitatem multiplicat.' The Prov. text is not in the Hebrew, and is cited only in the app. crit. of the Stuttgart Vulgate as an addition to Prov. 10.4, with thin support (though it is in the Clementine Vulgate, hence its familiarity to modern readers and the verse number). It occurs in LXX as Prov. 9.12a where it immediately precedes the aenigma Salomonis (see on 3.6.11) and is catalogued by La Bonnardière Biblia Augustiniana: Proverbes 212 as Prov. 9.12b (but the passage from bapt. 6.12.18 should be struck: it arises from an Erasmian interpolation that the Maurists failed to eject). Used in various contexts, the verse consistently suggests to A. that those who employ falsehood become themselves food for demons: cf. c. Faust. 20.22, `illi quippe superbi et impii spiritus non nidore ac fumo, sicut nonnulli vani opinantur, sed hominum pascuntur erroribus'; sim. at c. Sec. 26, c. Cresc. 3.9.9, and ep. 105.2.6. (Identification by A. Vaccari, Scritti di erudizione e di filologia 2 [Rome, 1958], 12-13.)

daemonibus, quibus . . . sacrificabam: See on 1.17.27, `non enim uno modo sacrificatur transgressoribus angelis.'

superstitione: See on 3.6.10.

4.3.4

Astrology recurs at 7.6.8, esp. with the mention there of Vindicianus (mentioned but not named at 4.3.5). This paragraph continues the motif of curiositas from Bk. 3; cf. 10.35.56, `nec curo nosse transitus siderum', from the discussion there of curiositas; and ord. 2.15.42, `quae similiter definiendo ac secernendo in ordinem nectens astrologiam genuit, magnum religiosis argumentum tormentumque curiosis.'

Astrology recurs pastorally throughout Augustine's career. The same vocabulary, and the same scriptural quotations, appear over and over. The place of astrology in African life is vividly depicted at cons. ev. 1.23.36: If the Romans were consistent, `et mathematicos vel genethliacos maxime delerent, qui Saturnum, quem sapientium effectorem isti dicerent, maleficum deum inter alia sidera constituerent.' At Carthage, the fear of him was so profound that the vicus Saturni was often referred to euphemistically as the vicus Senis. The presence of soothsayers was also a persistent pastoral concern. Consider, e.g., Io. ev. tr. 7.7, `non quando nobis dolet caput, curramus ad praecantatores, ad sortilegos et remedia vanitatis. fratres mei, non vos plangam? cotidie invenio ista; et quid faciam? nondum persuadeo christianis in Christo spem esse ponendam?' Cf. Io. ev. tr. 8.8f and 10.5.

How early in life A. read Cicero's div. is uncertain; before civ. the only probable sighting is at conf. 10.16.25 (see notes there for the uncertainty). It is not impossible, of course, that A. knew this work; to accept that proposition we must also accept that A. could be immune to the arguments even of figures he respected highly, and this from early in life. See on 8.7.17, `petieram a te castitatem', for the history of his resistance to those (including Cicero) who preached continence. Many of his moments of `conversion' took the form of re-reading, and now accepting, books and authors he had long known.

planos: used in this sense (= pla/nos) by Cic., Hor., Petron., Plin. iun., and Aul. Gell.; OLD: `one who practices deceit or imposture, esp. as a means of making a living.' Amerbach's Basle edition of 1506 wrongly conjectured planetarios (surviving as a ghost word in Lewis and Short).

mathematicos: Cf. c. acad. 1.7.21, for the learned derision of Flaccianus (procos. Afr. 393?, a pupil of A.? [c. acad. 1.6.18]) directed against the Carthaginian soothsayer Albicerius; Courcelle, Recherches 76n3, suggests Flaccianus probably was behind the events described here in conf. (Mandouze, Pros. chr. s.v. Flaccianus, tends to agree); cf. 4.3.5, `sine verborum cultu', for a hint that Albicerius is the divinator meant here (for at c. acad. 1.7.21, F.'s derision arises from his obvious cultural superiority to the less educated abilities of Albicerius); Flaccianus was at some other date the source for A.'s acquisition of the Christian Sibylline texts (civ. 18.23). Possidius indiculum (ed. Wilmart, MA 2.185, line 49) notes a letter to Flaccianus, which was apparently extant c. 825: Courcelle, Recherches ed. 2, 262.

Two of A.'s most detailed discussions of the subject of astrology precede conf. by a relatively short period, and another probably followed not long after:

div. qu. 45.2, `adversus eos autem qui nunc appellantur mathematici, volentes actus nostros corporibus caelestibus subdere, et nos vendere stellis, ipsumque pretium quo vendimur a nobis accipere'; he instances the case of twins with different fates, though sharing the same constellatio (see on 7.6.8), to confute their pretensions.

doctr. chr. 2.21.32, `neque illi ab hoc genere perniciosae superstitionis segregandi sunt qui genethliaci propter natalium dierum considerationes, nunc autem vulgo mathematici vocantur. nam et ipsi, quamvis veram stellarum positionem cum quisque nascitur consectentur et aliquando etiam pervestigent tamen quod inde conantur vel actiones nostras vel actionum eventa praedicere, nimis errant et vendunt imperitis hominibus miserabilem servitutem. nam quisque liber ad huiusmodi mathematicum cum ingressus fuerit, dat pecuniam ut servus inde exeat aut Martis aut Veneris, vel potius omnium siderum; quibus illi qui primi erraverunt erroremque posteris propinaverunt, vel bestiarum propter similitudinem vel hominum ad ipsos homines honorandos imposuerunt vocabula.' The example of twins is used in refutation at doctr. chr. 2.22.33: Esau and Jacob are named (as at div. qu. Simp. 1.2.3, where the astrological interpretation is trumped by Pauline grace); for conventional antecedents of the argument, cf. Amb. exam. 4.4.14 and Persius 6.18-19. Cf. doctr. chr. 2.23.35, `hinc enim fit ut, occulto quodam iudicio divino cupidi malarum rerum homines tradantur inludendi et decipiendi pro meritis voluntatum suarum, inludentibus eos atque decipientibus praevaricatoribus angelis. . . . quibus inlusionibus et deceptionibus evenit, ut istis superstitiosis et perniciosis divinationum generibus multa praeterita et futura dicantur nec aliter accidant quam dicuntur multaque observantibus secundum observationes suas eveniant, quibus implicati curiosiores fiant et sese magis magisque inserant multiplicibus laqueis perniciosissimi erroris. hoc genus fornicationis animae salubriter divina scriptura non tacuit'. (He quotes Deut. 13.1-3, 1 Kgs. 28.14-20, Sirach 46.23.)

Gn. litt. 2.17.35, `de fatis autem qualeslibet eorum argutias et quasi de mathesi documentorum experimenta, quae illi apotelesmata vocant, omnino a nostrae fidei sanitate respuamus; talibus enim disputationibus etiam orandi causas nobis auferre conantur et impia perversitate in malis factis, quae rectissime reprehenduntur, ingerunt accusandum potius deum auctorem siderum quam hominem scelerum.' Jacob and Esau appear at 2.17.36.

The other most substantive discussion is at at civ. 5.1; cf. epp. 246.2, 55.4.6ff, Gal. exp. 34. For a particularly vivid document giving life to the polemics, see en. Ps. 61.23 quoted below. The attack is eventually modified to take in the Pelagians: c. ep. pel. 2.6.12, `fatum quippe qui adfirmant, de siderum positione ad tempus quo concipitur quisque vel nascitur, quas constellationes vocant, non solum actus et eventa verum etiam ipsas nostras voluntates pendere contendunt; dei vero gratia non solum omnia sidera et omnes caelos, verum etiam omnes angelos supergreditur.'

plane: here, `utterly, absolutely, quite' (OLD). Translators have strayed: Ryan: `openly to consult'; but Valentinian II had made consultation of mathematici a capital offense (cod. theod. 9.16.8). Carena's translation assumes a connection in sense to `planos': `Percio quegli altri vagabondi, che chiamano matematici, non desistevo che vagamente dal consultarli.'

quod quasi: `for the reason that it was as if' --with subjunctive in the protasis of a contrary-to-fact condition. In `quod tamen' below, quod is merely a connectiving relative.

bonum est enim confiteri tibi: Cf. Ps. 91.2, `bonum est confiteri domino'; en. Ps. 91.3, `deinde multi non accusant satanam, sed accusant fatum. fatum meum me duxit, dicit. . . . quaeris ab illo quid sit fatum, et dicit: stellae malae. quaeris ab illo quis fecit stellas, quis ordinavit stellas; non habet quid tibi respondeat, nisi deus. restat ergo ut . . . deum accuset.' This verse and the following quotation of Ps. 40.5 recur in similar contexts because astrological doctrine implies that human responsibility for sin is non-existent. For A., the opposite is the case, and Ps. 40.5, confessionally, is the thematic statement of this throughout Augustine's life.

`miserere . . . tibi': Ps. 40.5, `ego dixi: “domine, miserere mei, sana animam meam, quoniam peccavi tibi”'; en. Ps. 40.6, `in factis meis, in peccatis meis non accuso fortunam, non dico: “hoc mihi fecit fatum”; non dico: “adulterum me fecit Venus, et latronem me fecit Mars, et avarum me fecit Saturnus.”'

Ps. 40.5 appears 3x in conf., in each case introduced by confiteri (cf. 4.12.19 and 5.10.18). The latter passage is the only one in conf. where he connects the appeal of Manicheism to the freedom from guilt for sin it could offer; and it is certainly significant (Knauer 166) that astrology could offer a similar freedom, hence probable that the appeal of astrology both arose for him at this time alongside his Manicheism and is reported here in detail for its analogous function. He does once link the Manichees to astrological speculation: cont. 5.14, `et alii quidem qui sua consueverunt excusare peccata, fato se ad peccandum queruntur impelli, tamquam hoc decreverint sidera et caelum prius talia decernendo peccaverit. . . . haec manichaeorum est immundissima insania'. For the verse elsewhere, see lib. arb. 3.2.5 (implicit link to curiositas), s. 20.2 (391? c. 395?).

licentiam peccandi: Cf. Sirach 15.21, in a discussion of purgatory and penance at ench. 19.70, `“nemini” enim “dedit laxamentum peccandi”, quamvis miserando deleat iam facta peccata si non satisfactio congrua negligatur.'

`ecce sanus . . . contingat': Jn. 5.14, `ecce iam sanus factus es: noli peccare, ne quid tibi deterius contingat' --to the lame man made to walk on the sabbath.

salubritatem: Picks up the medical metaphor implied by `cura animam meam' and the John quotation, and continued and enlarged (with ironic twists) by the treatment of the medical proconsul in the next paragraph--`medicinae artis peritissimus'.

Venus . . . Mars: en. Ps. 140.9, `totum hoc doctum et magnum, defensio peccati est. eris adulter, quia sic habes Venerem; eris homicida, quia sic habes Martem. Mars ergo homicida, non tu; et Venus adultera, non tu'; cf. s. Mai 17.1, civ. 5.1 (Mars), en. Ps. 31. en. 2.16 (Mars and Venus), en. Ps. 128.9 (Saturn, Mars, and Venus), en. Ps. 61.23 (where even a converted mathematicus is represented using the standard exemples of Mars and Venus: quoted below). For the conventionality of the forms of expression, cf. Plotinus 2.3.6.1-4, *)/Area de\ to/nde h)\ *)Afrodi/th qeme/nous moixei/as poiei=n, ei) w(di\ ei)=en, w(/sper e)k th=s tw=n a)nqrw/pwn a)kolasi/as au)tou\s e)mpipla/ntas w(=n pro\s a)llh/lous de/ontai, pw=s ou) pollh\n a)logi/an e)/xei; Immediate influence is unlikely, for Plotinus is rather subtler than A., who has to find a way for Mars to stand for violence, rather than for collaboration in adultery.

caro et sanguis: Mt. 16.17, `beatus es, Simon Bariona, quia caro et sanguis non revelavit tibi, sed pater meus, qui in caelis est'; 1 Cor. 15.50, `quia caro et sanguis regnum dei possidere non possunt'. Caro et sanguis is not so common in scripture as its equivalent in English. One could read the Matthew text to say that the revelation that comes from caro et sanguis would be owed to a purely human means of divination or prognostication.

superba putredo: Cf. 2.6.14, `o putredo, o monstrum vitae et mortis profunditas!' The exclamation there arises from a diagnosis of the sin of pride. Cf. also 2.1.1 (`computrui') and 4.11.16 (`reflorescent putria tua').

suavitas: Also of God at 2.6.13, 9.1.1, 10.17.26. Cf. Ps. 144.7, `memoriam abundantiae suavitatis tuae eructabunt, et iustitia tua exsultabunt.'

qui reddes unicuique: Ps. 61.13, `quia tu reddes unicuique secundum opera eius'; Mt. 16.27, `venturus est in gloria patris sui cum angelis suis, et tunc reddet unicuique secundum opus eius'; Rom. 2.6 (see Knauer 61n1), `in die irae et revelationis iusti iudicii dei, qui reddet unicuique secundum opera eius'; Job 34.11 (VL), `quia reddit homini opus suum, et iuxta viam suam unusquisque reperiet.'

Just here in Ps. 61 A.'s homily on the text records a dramatic moment: en. Ps. 61.23, in an appendix `post tractatum de psalmo, cum mathematicus in populo monstraretur' : `diu mathematicus fuit; seductus seducens, deceptus decipiens, inlexit, fefellit, multa mendacia locutus est contra deum. . . . iste dicebat quia adulterium non faciebat voluntas propria, sed Venus; et homicidium non faciebat voluntas propria, sed Mars; et iustum non faciebat deus, sed Iovis: . . . portat secum codices incendendos, per quos fuerat incendendus ut, illis in ignem missis, ipse in refrigerium transeat.' Despite the spontaneousness implied by the unusual stage-direction in the text there, it is likely that this last paragraph of en. Ps. 61 was integral to the sermon, with the converted mathematicus lurking to be brought out as a show piece. Clearly there was much suspicion of the bona fides of this conversion, but A. vouched for him.

cor contritum: Ps. 50.18-19, `quoniam si voluisses sacrificium, dedissem utique; holocaustis non delectaberis. (19) sacrificium deo spiritus contribulatus; cor contritum et humilatum deus non spernit.' This is the sacrificium (cf. above) that A. offers as an alternative to what all the unbelievers--himself when young, the astrologers who lured him, and all the others--had to offer.

humilatum: For the spelling adopted in text and commentary, see D. De Bruyne, MA 2.558-561.

4.3.5

eo tempore: If Vindicianus is the proconsul, this event dates to 379/82, and hence belongs unequivocally to the Carthage years of A.'s life (see on 4.1.1), not long before Faustus' appearance recounted in Bk. 5 (n.b. `rhetoricam' below, which is what he taught at Carthage). The episode here is carefully matched against the comparable discussion in 7.6.8; in both there is the intersection of curiositas and ambitio saeculi.

vir sagax: 7.6.8, `acutus senex', where Vindicianus is named; also ep. 138.1.3, `magnus ille nostrorum temporum medicus Vindicianus . . . acerrimus'. On Vindicianus, cf. PLRE 1.967: comes archiatrorum c. 379, serving Valentinian II; translated some Hippocrates; not otherwise attested as proconsul. See Jones, LRE 387, on appointment to administrative office as a reward for serving as physician to the court. O'Daly 81n5 suggests that V. may have been A.'s source for a fair amount of sophisticated medical information, notably his view of the mechanics of sense-perception; in this O'Daly follows Agaësse-Solignac (BA 48.711-713) tracing echoes of V.'s gynaecia at nat. et or. an. 4.2.2ff. An anecdote of Vindicianus from 411 shows him again the bastion of science against superstition for A. (but still thought to possess some `inlicita potentia'): ep. 138.1.3.

peritissimus: On peritia, see on 1.14.23; often associated by A. in this neutral way with medicine (cf. e.g., civ. 22.30 and elsewhere).

proconsul: proconsul G O2 S Knöll Skut.:   proconsule C D O1:   pro consule Ver. Pell.
At 8.4.9, the editors read proconsule, where the MSS divide essentially the same except that Sadds the final -e. On the most generous reading of the texts, A. used the connected form far more often than the etymological, and editorial practice in our editions probably masks the realities. In the editions contained in the Würzburg data base, there are eleven occurrences of the separated form, in five different works of A.; the connected form is cited 97x from 18 different works, including four of the five that also display the separated form.

qui resistis . . . gratiam: Jas. 4.6, 1 Pet. 5.5 < Prov. 3.34 (LXX): `deus superbis resistit, humilibus autem dat gratiam.' See on 1.1.1.

tamen etiam: The particles suggest Vindicianus was not Christian.

iucundi: Cf. 5.6.10, `hominem gratum et iucundum verbis,' but also 8.5.10, `deus, sola certa iucunditas'.

conloquio: Of private and intimate conversation: 4.8.13 (`conloqui et conridere et vicissim benivole obsequi'), 6.7.11, 9.9.19, 9.10.23, 12.16.23.

genethliacorum: doctr. chr. 2.21.32 (quoted above at length on 4.3.4), `genethliaci propter natalium dierum considerationes, nunc autem vulgo mathematici vocantur.'

monuit: Of benevolent advice from one in (often parental or pseudo-parental) authority, and perhaps especially advice that is not particularly welcome: 1.9.14 (`recte mihi vivere puero id proponebatur, obtemperare monentibus'), 2.3.7 (`monitus muliebres'), 2.3.8, 6.7.11, 9.5.13, 10.30.41. A. makes no direct response, only asks the (perhaps teasing, perhaps hostile) question given below.

professionem . . . deferre: G-M allege this `the technical phrase for making a return to the censor of one's occupation.' They seem to be inferring an idiom from the context of a single passage. TLL 5.1.314-318 s.v. deferre does not corroborate. Deferre is often used with (in) censum, but not with professio.

Hippocraten: Courcelle, LLW 195, quotes several texts to show that A. had access to Hippocrates in Greek in later years. Already in the late 390s, he knew of the problem of ps.-Hippocratean writings (c. Faust. 33.6).

gravis: gravis/graviter 30x in conf. 1-9, only 2x in 10-13. Cf. 9.6.14, `ingenio praeveniebat multos graves et doctos viros'.

vim sortis: See on 8.12.30 for A.'s own crucial experiment with the sortes, and note esp. ep. 55.20.37 quoted there.

instinctu: The word is infrequent in A. and almost always used in just this indefinite sense in a similar context: cf. 7.6.10, `tu enim, domine, . . . occulto instinctu agis, ut dum quisque consulit hoc audiat quod eum oportet audire'. For parallels, see civ. 2.26, 5.7, 8.26, and esp. 21.8 (differentiating monstra, ostenta, portenta, and prodigia): `sed viderint eorum coniectores, quo modo ex eis sive fallantur sive instinctu spirituum, quibus cura est tali poena dignos animos hominum noxiae curiositatis retibus implicare, etiam vera praedicant sive multa dicendo aliquando in aliquid veritatis incurrant.' For the classical origins of the usage, see Cic. div. 1.6.12, `est enim vis et natura quaedam quae tum observatis longo tempore significationibus tum aliquo instinctu inflatuque divino futura praenuntiat' (similarly at div. 1.18.34, Cic. makes instinctus an explanation of accurate prognostication that comes without ars--see next note).

non arte sed sorte: Cf. 4.3.6, `forte vel sorte, non arte,' and see on 7.6.9, `non arte dici sed sorte'; 7.6.10, `non ergo arte sed sorte vera diceret' --the latter texts from the revived discussion of astrology and divination at 7.6.8ff; div. qu. 45.2, `quod si non arte de codicibus exit saepe versus futura praenuntians, quid mirum si ex animo loquentis, non arte, sed sorte exit aliqua predictio futurorum?' The distinction goes back to Greek wordplay on te/xnh and tu/xh (Löfstedt, Symb. Osl. 56[1981] 106, citing E. R. Dodds on Plato's Gorgias [Oxford, 1959] 192: Gorgias 448c, e)mpeiri/a me\n ga\r poiei= to\n ai)w=na h(mw=n poreu/esqai kata\ te/xnhn, a)peiri/a de\ kata\ tu/xhn). BA 13.416n1, quoting passages from Plotinus 4.4.32-43: `Ces vues de Vindicianus laissent percer une mentalité néo-platonicienne; en effet, il n' attribue pas au seul hasard les prédictions heureuses, mais reconnaît une certaine consonance de l'univers aux divers états de l'âme.'

4.3.6

deliniasti: `Sketched', suggesting an imprecise figure; cf. civ. 17.8, `scriptura sancta etiam rebus gestis prophetans quodam modo in eo figuram deliniat futurorum.'

Nebridius: One of A.'s two important companions (the other is Alypius, not mentioned until 6.7.11) at Milan. A. always speaks warmly of N., though there was undoubtedly always some distance between them in their attitudes, right up to N.'s premature death. For fuller notice, see on 6.7.11. N.'s family home was near Carthage, reducing further any chance that this episode occurred at Thagaste.

castus: castus G S Knöll Skut. Pell. Vega  (wrongly reporting O in support of the reading):   cautus C D O Ver.
Cf. 9.3.6, `castitate perfecta atque continentia' --of N. G-M think cautus better here, since castitas is better predicated of him after conversion. BA defends castus by arguing that it is not continentia that is meant here, but rather just that he was untainted by this superstition: cf. civ. 8.18, `homo castus et ab artium magicarum sceleribus alienus,' and see above on 4.2.3, `ex tua castitate' (with Madec's note cited there). But the BA translation weakly offers only: `jeune homme de grande vertu et de grande religion'. Juergens-Schaub (in Skutella [1969], 389) cite 1.13.21, 2.6.14, 4.2.3, 5.12.22, passages where, as they say, `omnis a deo aversio fornicatio appellatur' (see on 1.13.21, `fornicabar abs te'). Cautus is infrequent in A. and hardly offers unqualified praise--it has an overtone of scrupulosity when in 7.6.8 the astrologers are said to count the minutes of a horoscope `cautissima observatione'. (A similar substitution of caut- for cast- appears at c. acad. 2.2.5--see prolegomena.)

nullum certum quale quaerebam documentum: The first sign of this hankering for certitude (the emphasis at 3.6.10 is rather different), which is presented as reaching its most acute stage at the end of Bk. 5 and thenceforward as defining one of the main conditions of A.'s philosophical and religious inquiries; cf. 5.14.25.

4.4.7

The death of A.'s friend, centerpiece of Bk. 4, performs several functions.

1. It marks a stage on the trajectory of temptation, a falling into sin whose punishment is such that we react by sinning further:

a) since Bk. 2, we have seen the disordering of the power of friendship in his life (concupiscentia carnis);

b) since Bk. 3, we have seen that his compassion and pity are out of joint (concupiscentia oculorum);

c) here now he is punished for his sins in an apt way, and the result (4.7.12) is that he pursues his career to Carthage (ambitio saeculi).

2. It displays him living in a world falsified with respect to both of the two great commandments (see 3.8.15):

a) belief that God is only a phantasma (4.7.12, `vanum phantasma et error meus erat deus meus' : cf. 3.6.10);

b) the inability to love his neighbor (4.7.12, `o dementiam nescientem diligere homines humaniter!'), though `amare amabam' (3.1.1) had characterized his ambition since his first coming to Carthage.

3. It displays A. in the grip of violent and genuine grief, condign punishment for one who had studied long in the school of sham pity (3.2.2).

Two deaths in conf. affect A. deeply, his friend's here and that of Monnica: for comparison of the two, see on 9.11.28. Many other persons who figure in this narrative die in its pages (especially in Bk. 9: see on 9.1.1), but no other receives comparable attention. The only other death in A.'s life to show a comparable effect in his texts is that of Marcellinus, executed in 413 in circumstances that left A. devastated (ep. 151.).

L'innominato: Courcelle, Recherches 41, `Augustin a sans doute jugé que le nom de ce jeune homme de Thagaste, mort a la fleur de l'âge, était sans intérêt pour les lecteurs et la postérité.' Would A. neglect a chance to pay edifying tribute to a deceased member of Alypius' flock (with so exemplary an attitude to baptism)? But Courcelle examines this as one of the many unnamed figures in the text; consider instead the pattern of the named ones. Leaving aside persons A. never met, there remain in conf. only: Adeodatus, Alypius, Ambrose, Elpidius, Evodius, Faustus, Firminus, Monnica, Nebridius, Patricius, Ponticianus, Romanianus, Simplicianus, Symmachus, Verecundus, Vindicianus, 16 in all. The list is sharply pared to include only individuals who were, witting or unwitting (e.g., Faustus and Symmachus), agents of conversion; the sole exception is Patricius, not named until Bk. 9, there only once incidentally in the narrative of Monnica's life (9.9.19) and once formally in a request that his readers remember Monnica (herself named only there in all A.'s writings) and Patricius at the altar. There are few names in these books of downfall (only A., in such a mood and for such a reason, could write the paean to friendship that appears at 4.8.13 and mention not a single friend by name!), more later as A.'s life takes a better turn (Vindicianus has been described at 4.3.5, but is not named until 7.6.8; Bk. 6 is notable for the way the coterie of friends comes together--Augustine, Alypius, Nebridius; in Bk. 8, Simplicianus and Ponticianus appear). Finally, both the friend here and the mother of Adeodatus are, to put it bluntly, people A. treated shabbily--A. even refuses himself the consolation of thinking that he treated this one as a true friend (`sed nondum erat sic amicus'): it is not hard to believe that he felt he had no right to draw even their names into his story. (See on this W. Steidle, Romanitas-Christianitas [Festschrift J. Straub: Berlin, 1982], 446. The pattern of biblical names cited in conf. is interesting in its own right: see on 7.21.27.)

The friend is never mentioned elsewhere. But could A. have written this passage (on Mt. 5.29, `si oculus tuus dexter scandalizat te . . .') without thinking of him? s. dom. m. 1.13.38, `cogit quaerere diligentius quid dixerit oculum. in qua quaestione nihil mihi occurrit congruentius quam dilectissimum amicum. nam hoc est utique quod membrum recte possumus appellare, quod vehementer diligimus, et hunc consiliarium, quia oculus est tamquam demonstrans iter, et in rebus divinis, quia dexter est. . . . in rebus autem divinis consiliarius scandalizans est, si in aliquam perniciosam haeresim nomine religionis atque doctrinae conatur inducere.' On that assumption, note that the younger A. becomes the scandalous consiliarius, to be tossed aside--which is just what the dying friend did when A. sought to lure him back to heresy, `nomine religionis atque doctrinae'.

illis annis . . . coeperam: This memory dates from the year teaching at Thagaste (375/6), age 21 (as was the friend: `coaevum'). On the chronological displacement, see on 4.1.1.

municipio: See on 2.3.5.

in scholam ieramus pariterque luseramus: Cf. the description of childhood activities at 1.9.15.

vera amicitia: A.'s notion of friendship had remained consistent since Cassiciacum, but the old textual underpinnings were carefully pulled out and new ones installed: c. acad. 3.6.13, `mecum enim familiarissimus amicus meus non solum de probabilitate humanae vitae verum etiam de ipsa religione concordat, quod est veri amici manifestissimum indicium, si quidem amicitia rectissime atque sanctissime definita est [cf. Cic. amic. 6.20], “rerum humanarum et divinarum cum benevolentia et caritate consensio”.' The revised version takes on wider and fuller expression at trin. 3.4.9, `illic enim dei voluntas, qui facit angelos suos spiritus et ministros suos ignem ardentem, [Ps. 103.4] in spiritibus summa pace atque amicitia copulatis, et in unam voluntatem quodam spiritali caritatis igne conflatis, tamquam in excelsa et sancta et secreta sede praesidens. . . . inde se, quibusdam ordinatissimis creaturae motibus, primo spiritalibus, deinde corporalibus, per cuncta diffundit [Rom. 5.5], et utitur omnibus ad incommutabile arbitrium sententiae suae, sive incorporeis sive corporeis rebus'. That A. never wrote in Christian terms a treatise de amicitia (as one could well imagine Ambrose, for example, writing), despite the powerful role he gives to friendship both in his description of his life and in the metaphors he uses (see esp. on 2.2.2, `luminosus limes amicitiae'), is a measure of a loss that A. never knew how to redress, perhaps because it affected him too intimately for him to articulate and thus to control. (The same may be said of s. 385.2.3, a discussion of friendship that is, however, not in its present form A. but Caesarius, though depending [see Verbraken] on a lost sermon of A.'s; quoted below on 4.8.13, `alia erant'.)

The Ciceronian definition recurs in A. at ep. 258.1, in a letter to another friend of A.'s youth, Marcianus, who had been similarly estranged from Christianity when he and A. were young, but who had by the time of the letter (Mandouze, Pros. chr. s.v. Marcianus 2, puts it close to 395) changed his sentiments, though not so far as to accept baptism (ep. 258.5). There the harsh judgment on the inauthenticity of the first friendship is the same: ep. 258.1-3, `antiquissimo amico, quem tamen non habebam, quam diu in Christo non tenebam. . . . (3) nolo autem suscenseas nec tibi videatur absurdum quod illo tempore cum in vana mundi huius aestuarem, quamvis me multum amare videreris, nondum eras amicus meus, quando nec ipse mihi amicus eram sed potius inimicus. diligebam quippe iniquitatem'.

haerentes: haerentes O S Knöll Skut. Ver.:   cohaerentes G:   inhaerentes C D Maur.
For haereo in a good sense, but not of man clinging to God, see 6.10.16 (2x) and 9.4.8.

tibi tibi C D O Ver. Pell.:   sibi GS Knöll Skut. Vega

caritate diffusa: Rom. 5.5, `spes autem non confundit, quia caritas dei diffusa est in cordibus nostris per spiritum sanctum qui datus est nobis.' The quotation is repeated to bracket Bk. 13, at 13.7.8 (see notes there) and 13.31.46. Note the linking of caritas, the sanctus spiritus, and the spirit of friendship: it was particularly friendship that A. thought perverted by the fault of concupiscentia carnis (2.2.2). A.-M. La Bonnardière, Aug. Mag. 2.657-665 reports the verse at least 200x in A., from early (mor.) to latest (c. Iul. imp.), a constant intimation of the presence of the Spirit in the church for A.

cocta: cocta G O1 S Knöll Skut. Ver.:   coacta C D O2

a fide vera: The Christian milieu of A.'s upbringing is underscored: the friend is another so brought up.

superstitiosas: See on 3.6.10.

mater: Monnica's only appearance in Bk. 4, a reminder of the ending of Bk. 3 (3.11.19 - 3.12.21).

errabat: For error as particularly a form of deviation in the order of knowledge (and hence corresponding to concupiscentia oculorum), recall 1.20.31 (`dolores, confusiones, errores'), and cf. 4.7.12 (`vanum phantasma et error meus erat deus meus'); error/errare occur by far more often in Bk. 4 than in any other (13x in 4, 33x in all others together), right from the outset (4.1.1, `praeteritos circuitus erroris mei').

fugitivorum: At 4.7.12, A. flees again (`et tamen fugi de patria'), but such flight is in vain (4.9.14, `quo it aut quo fugit nisi a te placido ad te iratum?'). Cf. 2.6.14, 3.3.5 (`fugitivam libertatem'), 6.11.20. For `tu imminens dorso fugitivorum tuorum', cf. 4.16.30, `dorsum enim habebam ad lumen'.

deus ultionum: Ps. 93.1, `deus ultionum dominus, deus ultionum fidenter egit' (A.'s text, against the usual libere egit).

ultionum . . . misericordiarum: in one act, divine punishment for Augustine (duly merited [in his eyes] by the deeds he has narrated) and divine mercy for the friend (snatched away just after baptism, before the Manichean Augustine had a chance to work on him again).

miris modis: The same combination at 5.7.13, 7.21.27, 10.40.65, 13.11.12; `miris et occultis modis' at 5.6.10 and 6.12.22.

4.4.8

quis laudes tuas: Cf. Ps. 105.2, `quis loquetur potentias domini, auditas faciet omnes laudes eius?'

laudes tuas: Only here in conf. in this sense constrained by the relative clause, `deeds worthy of praise'; elsewhere `acts or praise directed to you': 1.17.27 (`laudes tuae, domine, . . . per scripturas tuas'), 5.1.1, 6.7.12, 9.7.16, 10.6.8.

Carena relies on De Marchi 312, and translates: `Chi puó da solo enumerare i tuoi vanti, che in sé solo ha conosciuto,' effectively reading a comma after unus and none after uno (Skut. and other editors print a comma after uno only). Unus and uno are brought together for euphonic effect, but the first belongs to the main clause, the second to the relative clause.

investigabilis abyssus: Cf. Rom. 11.33, `o altitudo divitiarum sapientiae, et scientiae dei: quam incomprehensibilia sunt iudicia eius et investigabiles viae eius!' Sirach 42.18, `abyssum et cor hominum investigavit.'

The Romans passage is frequently quoted in texts dealing with grace and free will as a way of ending inconclusive analysis with avowal of the paradox of grace and freedom (e.g., div. qu. Simp. 1.2.22, pecc. mer. 1.21.29, spir. et litt. 34.60, 36.66, c. ep. pel. 4.6.16, gr. et lib. arb. 22.44, corrept. 8.17-19, praed. sanct. 2.4, 8.16; persev. 12.30, ss. 26.12.13, 27.6.6-7.7, and 294.7.7). Here at a moment of obscure turning in his own life, the dynamics of punishment and grace are beyond reach.

investigabilis: See on 2.9.17.

abyssus iudiciorum tuorum: Cf. Ps. 35.7, `iustitia tua sicut montes dei, iudicia tua sicut abyssus multa.' Other abysses of divine judgment at 7.6.10 and 13.2.3. en. Ps. 41.13, `abyssus enim est profunditas quaedam impenetrabilis, incomprehensibilis; et maxime solet dici in aquarum multitudine. . . . denique quodam loco dictum est: “iudicia tua abyssus multa”; hoc volente scriptura commendare, quia iudicia dei non comprehenduntur.'

baptizatus est: breviarium Hipponense 32, `ut aegrotantes, si pro se respondere non possunt, cum voluntatis eorum testimonium sui periculo proprio dixerint, baptizentur'; = conc. Carth. III (397) can. 45. Cf. adult. coniug. 1.26.33, `catechumenis ergo in huius vitae ultimo constitutis, seu morbo seu casu aliquo si compressi sint, ut, quamvis adhuc vivant, petere sibi tamen baptismum vel ad interrogata respondere non possint, prosit eis quod eorum fide christiana iam nota voluntas est, ut eo modo baptizentur quo modo baptizantur infantes, quorum voluntas adhuc nulla patuit. [He allows, however, that some may be disinclined to baptize the unconscious for fear that they `contrarium gerant voluntatis arbitrium'] . . . si voluntas eius incerta est, multo satius est nolenti dare quam volenti negare'. G-M suggest this was written `doubtless with the present instance in mind'; but by 419/20 the bishop A. had seen many more sickbed baptisms.

recreatus est et salvus factus: Similar expressions at 1.11.17 (A.'s childhood illness and brush with baptism), 4.8.13 (A.'s grief after this friend's death), 5.10.18 (A.'s illness on coming to Rome).

temptavi . . . inridere: haer. 46.17, `baptismum in aqua nihil cuiquam perhibent [manichaei] salutis afferre: nec quemquam eorum quos decipiunt baptizandum putant.'

at ille: This veneration for baptism is a token of the power that ritual held over the late antique imagination. What could be smiled over by a non-initiate was a different matter altogether for the initiate, however initiation was attained.

libertate: See on 2.6.14 for the dangers of libertas to A.; here closer to `impertinence' than `freedom' or `liberty.'

dementiae: 4.7.12, `o dementiam nescientem diligere homines humaniter!'

febribus: See on 5.9.16, `ingravescentibus febribus'.

defungitur: Regular in conf. for death (3.4.7, 6.2.2, 8.2.3, 9.8.17, 9.11.27), and elsewhere in A. (e.g., civ. 5.18, c. Faust. 14.1) but the euphemism is not attested earlier than Pliny the elder, and in CL only in perfect tenses.

4.4.9

The depiction of grief is almost literally God-less: only in a vain attempt to enjoin hope does God appear here.

contenebratum est cor meum: Lam. 5.17, `propterea maestum factum est cor nostrum, ideo contenebrati sunt oculi nostri.' In the few other passages where contenebrare occurs, it is set against the illumination that comes from God, e.g., en. Ps. 103. s. 4.2, `erigunt ipsam mentem suam, non ad se sed ad artificem suum, . . . et unde recedentes contenebrantur, et quo revertentes inluminantur'; sim. at en. Ps. 57.22, 65.13, 75.8.

mors: Bk. 4, largely because of this extended meditation on his friend's passing, is more marked by death (counting mor./morior and related nouns and adjectives) than any other in conf.; Bk. 9 comes closest (and see preceding comm. on 9.1.1 for the transformation of death there. Bks. 10-13 show a marked decline (and a corresponding upswing in vita/vivo, etc.).

patria supplicium: A first hint that at the end of this episode A. will leave Thagaste for Carthage (4.7.12).

paterna domus: Here a sign that at Thagaste A. took up residence in his father's house, though Patricius was 3-4 years dead (see on 3.4.7); no mention of Monnica, or of her having any say whether A. would reside there (see on 3.11.19). The phraseology is further redolent of the story of the prodigal (see on 1.18.28).

cruciatum: Elsewhere of school-punishments (1.14.23), of Monnica's grief when A. abandons her in Carthage (5.8.15), and of A.'s stifled grief for Monnica (9.12.31).

expetebant: expetebant G S Knöll Skut. Vega Pell.:   expectebant CDO Ver.
Expetere is `to look for', expectare only `to wait for, to have the opinion that he would turn up'; the former is much better with oculi; cf. 4.7.12, `minus enim eum quaerebant oculi mei'.

oderam omnia: Things A. `hates' in conf. (see on 3.1.1): his schoolwork in general (1.12.19), Greek in particular (1.13.20, 1.14.23), `securitatem et viam sine muscipulis' (3.1.1), students who did not pay their fees (5.12.22--`quamvis non perfecto odio'! see notes there), himself and his own iniquity (8.7.16-17); then nothing at all through Bks. 9 to 13 except 12.14.17, `odi hostes eius [sc. scripturae sacrae] vehementer.'

nec mihi iam dicere poterant [sc. omnia]: This virtual personification of `all things' is not unparallelled: cf. 9.10.25 (`quoniam si quis audiat, dicunt haec omnia') and 10.6.8 (`et caelum et terra et omnia quae in eis sunt, ecce undique mihi dicunt').

quaestio: Cf. 2.10.18, `factus sum mihi regio egestatis,' and 10.33.50, `me, in cuius oculis mihi quaestio factus sum, et ipse est languor meus'. Note (Knauer 150n2) at 4.11.16 the echo of Ps. 102.3, `sanabuntur omnes languores tui'; the passage from 10.33.50 provides its own commentary on the present text.

interrogabam: This interrogatio anticipates the formal apostrophe of 4.11.16; or rather parallels it, for the present passage represents his thoughts of 375/6, while 4.11.16 presents a line of thought contemporary to conf. The Psalm-text is anachronistic. A. does not say that in 375/76 he saw in the Psalm a parallel to his situation; rather, he uses the Psalm's words to embody in his text an anxiety that he then felt and to comment upon indirectly. The Psalm-text's presence here implies, without making explicit, an extra-textual answer to a question posed within, not merely the text, but within the past life narrated in the text. If A. had known at the time to apply Ps. 41 (`et si dicebam “spera in deum”': i.e., if the word of God had intruded on the event as it here intrudes upon the narrative), his soul would have known how to respond to the interrogatio--with hope and confession, the instruments with which he wrote this book twenty years later.

quare tristis esset: Ps. 41.6, `quare tristis es, anima mea, et quare conturbas me? spera in deum, quoniam confitebor illi'; en. Ps. 41.10, `ait [homo] sibi constituto inter has tristitias, et comparans haec illis ad quae videnda ingressus est, et post quae visa egressus est: “quare”, inquit, “tristis es, anima mea, et quare conturbas me?” . . . et quasi responderet illi anima eius in silentio: “quare conturbo te, nisi quia nondum sum ibi, ubi est dulce illud quo sic rapta sum quasi per transitum? numquid iam bibo de fonte illo, nihil metuens? iam nullum scandalum pertimesco? iam de cupiditatibus omnibus tamquam edomitis victisque secura sum?” . . . sed “spera in deum” respondet conturbanti se animae suae, et quasi rationem reddenti perturbationis suae, propter mala quibus abundat hic mundus.' (A. presents Ps. 41 [en. Ps. 41.1] as the chant of `catechumeni [qui] ad gratiam sancti lavacri festinant', here expressing the desire that brings them to the church: echoing that Psalm here captures both the attitude that should have motivated A.) Ps. 41 plays an important role at 13.12.13-13.14.15, where see the Psalm text and additional notes. Ps. 42.5, `utquid tristis es, anima mea, et utquid conturbas me? spera in dominum, quoniam confitebor illi.' See the similar exposition at en. Ps. 42.5-6.

phantasma: See on 3.6.10. BA ad loc., `Le “dieu matérialisé”, auquel croit alors Augustin, lui paraît manquer de réalité en regard de l'ami qui vient de mourir: un dieu irréel ne console pas d'une perte réelle.'

fletus erat dulcis: On tears in conf., see on 3.2.4 and see next paragraph here. (With the other echoes here of Ps. 41, it is apt to think of Ps. 41.4, `fuerunt mihi lacrimae meae panes die ac nocte'; cf. then en. Ps. 41.6, `fuerunt mihi, inquit, lacrimae meae, non amaritudo [cf. 4.5.10, `suavis fructus de amaritudine vitae'], sed panis. suaves erant mihi ipsae lacrimae'); sim. at en. Ps. 127.10, `dulces sunt et ipsae lacrimae . . . quia factae sunt tibi et ipsae panis die ac nocte'.

in deliciis animi mei: Ps. 138.11, `fortasse tenebrae conculcabunt me et nox inluminatio in deliciis meis'; en. Ps. 138.14, `quia in nocte me desperaveram posse transire tantum mare, et tantam viam superare, et venire ad extremum perseverando usque in finem.'

4.5.10

A quaestio de fletu dulci:

1. Is it sweet because we are derelict here? But then it would not be sweet, for there would be no remnant of hope.

2. Because we hope God will hear? True for prayer (see the fig tree in 8.12.29), but not true here, where he had no hope of recovering his friend.

3. Is it in fact bitter, but pleasant only by comparison with our revulsion from the thing we had possessed before with pleasure?

The first two alternatives are presented with refutations; the third, left unrefuted, expresses by implication his position about the tears that came with his friend's death. He no longer loved the friend in the inappropriate way in which he had loved him in life, and did not know how to react except in this preference for bitter tears.

qui veritas es: Jn. 14.6, `ego sum via, et veritas, et vita'; see on 1.5.6, and cf. (in identical words) 3.6.10, 5.3.5, 10.23.33.

et admovere aurem cordis mei ori tuo: The heart has ears to hear (more often than eyes): 1.5.5, 4.11.16, 4.15.27. What is here possibility becomes act at 13.6.7, `tibi admoveo cor meum, ne me vana doceat'.

ubique: See on 1.3.3, `totus ubique'.

longe: See on 1.18.28.

tu in te manes: Wisd. 7.27, `in se ipsa manens innovat omnia'. Other echoes at 7.9.13, 7.11.17, 9.10.24 (`et quid simile verbo tuo, domino nostro, in se permanenti sine vetustate atque innovanti omnia'). This is parody of Christian doctrine, a Manichean alternative--that the Logos would remain serene and undisturbed, apart from human affairs, and men would be left to their miseria. A.'s distaste for such an argument predates his conversion: cf. 6.5.7.

speramus: The present passage marks the entry of Christian hope into the text. Cf. here 4.4.9, `spera in deum' (a scriptural voice he did not hear at the time); here (hope become a question and a possibility); 4.6.11, `spes mea' (the narrator's confidence); 4.16.31, `in velamento alarum tuarum speremus' (itself the object of prayer). In later books, spes becomes a recurring motif: in Bk. 6 a possibility (6.11.18, `magna spes oborta est') and a contrast to secular hopes (6.11.19, `relicta spe saeculi'); in Bk. 10 a way of going beyond the present remnants of sin (10.30.42, `sperans perfecturum te in me'); in Bks. 11-13, a recurrent asseveration (from Rom. 8.24, `spe enim salvi facti sumus', echoed at 11.9.11, 13.13.14, 13.14.15).

istuc: istuc154n1 C1 D2 GO2 S Ver. Knauer:   istud C2 O1 Knöll Skut.:   istis D2

miser . . . eram: The thought is continued in 4.6.11.

tunc tunc O1 SV Skut. Ver.:   dum C D A H F:   cum O2:   tunc, dum G EM Knöll Maur.:   dum tunc B P:   nunc Z

4.6.11

Reflection on confession in the midst of confession again leads to the tension between quaerere and confiteri: both play a part in conf., the former subordinated to the latter.

amicitia rerum mortalium: 1.13.21, `amicitia enim mundi huius fornicatio est abs te.'

requiescebam: First occurrence of requiesco/requies since 1.1.1, `inquietum est cor nostrum'. What he regrets is not that life was bitter to him (it always is: see citations below), but that he found repose in it: only in God is true repose (6.16.26, `et tu solus requies'), but the place of repose is the place to which our love leads us (13.9.10, `requies nostra locus noster', and see notes there on love's `gravity'). The discussion of A.'s grief here is repeatedly marked by restlessness (4.7.12, `nec requies erat nec consilium,' 4.10.15, `et requiescere amat in eis, quae amat,' 4.12.18, `requiescite in eo et quieti eritis . . .; non est requies ubi quaeritis eam').

in amaritudine: Cf. Iob 3.20 (VL), `utquid enim datur eis qui in amaritudine sunt lux?' (adn. Iob on 3.20: `“eis qui in amaritudine sunt lux”: peccatorum honor'.) Is. 38.15, `recogitabo omnes annos meos in amaritudine animae meae.' The bitterness of alienation from God also at 4.9.14, 6.10.17, 7.3.5. Cf. en. Ps. 85.6, `solus enim tu es iucunditas: amaritudine plenus est mundus.'

ita miser eram: The focus of this paragraph, to specify the degree of wretchedness (--> `sic eram omnino, memini').

nescio an vellem: Takes away any possibility of reading his love for his friend as disinterested caritas.

de Oreste et Pylade: Cic. amic. 7.24, `qui clamores tota cavea nuper in hospitis et amici mei M. Pacuvii nova fabula! cum ignorante rege uter Orestes esset, Pylades Orestem se esse diceret, ut pro illo necaretur, Orestes autem, ita ut erat, Orestem se esse perseveraret. stantes plaudebant in re ficta; quid arbitramur in vera facturos esse?' The same episode at Cic. fin. 5.22.63., and, deriving from Cic. fin., at Amb. off. 1.41.216, in both cases emphasizing the theatrical response. It is no coincidence that the exemplum is one met in so theatrical a context; whether A. ever saw a representation of the story (or, improbably, knew the text of Pacuvius), it came to him in authoritative texts as just the sort of tear-jerker (cf. Nietzsche's `augenverdreherisch' quoted below) that enthralled him at Carthage (thus representing the triumph of sentimentality over any authentic love: see on 3.2.2), and the exemplum is just the sort of mythic context in which A. would have attempted to understand his own grief at the time.

vel simul vel simul C D G O Ver. Pell.:   simul S Knöll Skut. Vega Pell.
Pell.: `vel simul . . . risponde bene al senso: giacché non è possibile vivere insieme, almeno morire insieme.'

qua: qua D1 G O1 Ver.:   quia CD2 O2 S Maur. Knöll Skut. Vega Pell.
Cf. Ovid trist. 4.4.75-76(on Orestes/Pylades):

nec tamen hunc sua mors, nec mors sua terruit illum:
    alter ob alterius funera maestus erat.

taedium vivendi . . . moriendi metus: Ovid., met. 10.482, `mortisque metus et taedia vitae'; cf. J. Kern and O. Hammerstein, `I'm tired of livin' but I'm feared of dyin'' (`Ole Man River'). Cf. below `et ideo mihi horrori . . .', explaining his feeling.

ecce cor meum: 6.6.9, `vide cor meum, domine, qui voluisti ut hoc recordarer et confiterer tibi'; see on 1.5.5, `ecce'.

spes mea: Ps. 70.5, `domine, spes mea a iuventute mea'; en. Ps. 70. s. 1.7, `ante enim non in te sperabam; quamvis tu fueris protector meus, qui me salvum perduxisti ad tempus quo in te discerem sperare.' (Taken literally in the Augustinian scheme of the ages of man [see on 1.8.13], this text would remind A. that it was in early iuventus [6.1.1, citing Ps. 70.5, and 7.1.1] that he acquired this hope.)

mundas . . . immunditia: 1.5.6, `ab occultis meis munda me, domine'.

evellens de laqueo pedes meos: Ps. 24.15, `oculi mei semper ad dominum, quoniam ipse evellet de laqueo pedes meos'; en. Ps. 24.15, `nec timeam pericula terrena, dum terram non intueor; quoniam ille quem intueor, evellet de laqueo pedes meos.' Knauer 178: `Der Vers wird also in zwei parallelgeordnete Kola zerlegt, die durch das hinzugefügte zweite Verbum (“dirigo”) einander angeglichen werden. Da das Zitat an einen Gottesnamen angeschlossen wird, der bereits durch einen Relativsatz näher bestimmt ist, bilden die Partizipien einen ruhig auslaufenden Schluss der Periode.'

quidam dixit: Horace carm. 1.3.5-8:

navis, quae tibi creditum
debes Vergilium, finibus Atticis
     reddas incolumem precor,
         et serves animae dimidium meae.

suae: suae C D G O S Knöll Ver. Pell.:   meae Maur. Skut.
The idea had become conventional: Hier. ep. 3.3, `partem animae meae' (of a dead friend: see fuller quotation on 4.10.15). For other echoes of the motif in classical and patristic literature, see Pellegrino, Les Confessions 121n5, and cf. Ennodius, ep. 3.2.2, `qui maior animae fuit portio'. Both exempla of friendship given are pre-Christian.

unam fuisse animam: 9.12.30, `vita, quae una facta erat'. Ovid, trist. 4.4.72 (on Orestes/Pylades), `qui duo corporibus, mentibus unus erant.' ord. 2.18.48, `amici quid aliud quam unum esse conantur? et quanto magis unum, tanto magis amici sunt.' Cassian, conl. 1.1 (written at least twenty years after conf.), speaking of himself and Germanus: `. . . ut cuncti ad significandam sodalitatis ac propositi nostri parilitatem pronuntiarent unam mentem atque animam duobus inesse corporibus'.

et ideo forte: retr. 2.6.2 (the only qualification he offers there to conf. 1-9), `in quarto libro, cum de amici morte animi mei miseriam confiterer, dicens quod anima nostra una quodammodo facta fuerat ex duabus, “et ideo”, inquam, “forte mori metuebam, ne totus ille moreretur quem multum amaveram.” quae mihi quasi declamatio levis quam gravis confessio videtur, quamvis utcumque temperata sit haec ineptia in eo quod additum est “forte”.' Writing thirty years later, giving no sign of feeling the spell of this episode, A. found that the conf. text distanced itself too little from the atmosphere of classical sentimentalism about friendship (note that the Horace tag is quoted here not as something he thought at the time, but as a general principle true at the time he wrote). A. was thus at least as dismayed as Nietzsche with the rhetorical posturing of this passage. But a closer inspection of the passage shows that the attitude in `ne totus ille . . .' was not that of A. of 397, but of A. of 376, aged 21, a sentimental age. A. judged himself to have exercised elsewhere a retroactive censorship on what he said of his youth, either identifying attitudes as inappropriate or censuring them--and he judged himself as having done so inadequately here.

Nietzsche's criticism is in a letter to Overbeck, 31 March 1885 (writing from Nice, between Thus Spake Zarathustra and Beyond Good and Evil² ): `Ich las jetzt zur Erholung³ die Confessionen des h[eiligen] Augustin, mit großem Bedauern daß Du nicht bei mir warst. Oh dieser alte Rhetor! Wie falsch und augenverdreherisch! Wie habe ich gelacht! (zb. über den "Diebstahl" seiner Jugend, im Grunde eine Studenten-Geschichte). Welche psychologische Falschheit! (zb. als er vom Tode seines besten Freundes redet, mit dem er eine Seele gewesen sei, "er habe sich entschlossen, weiter zu leben, damit auf diese Weise sein Freund nicht ganz sterbe." So etwas ist ekelhaft verlogen.) Philosophischer Wert gleich Null. Verpöbelter Platonismus, das will sagen, eine Denkweise, welche für die höchste seelische Aristokratie erfunden wurde, zurecht gemacht für Sklaven-Naturen. Übrigens sieht man, bei diesem Buche, dem Christenthum in den Bauch: ich stehe dabei mit der Neugierde eines radikalen Arztes und Physiologen.' (Nietzsche Briefwechsel: Kritische Gesamtausgabe [Berlin/New York, 1982], 3.3.34)

4.7.12

On this paragraph, see also C. Carena, Riv. stor. e lett. rel. 3(1967), 65-70, emphasizing the link to the argument of the third book of Cicero's Tusculans (see on 6.16.26 for A.'s reading of Cicero reflected in conf.). The parallels Carena draws to Cicero and others are mainly ones of topic and treatment, or else conventional turns of phrase, but he well captures the atmosphere of the passage. It is undeniably apt that A.'s treatment of his state of mind at this time should reflect the analysis he should have given it at the time, from his classical readings.

Alfaric 50-51 uses this passage to inveigh against A. the insincere friend: hostile to his father, mistreating his mother, casting off his concubine, tearless at his mother's deathbed, locking himself up in Thagaste and refusing to visit a friend's deathbed in Carthage (Nebridius: ep. 10.1-2), confining his caritas to God and living himself shut away like a monk. There is certainly an element of truth to that: caritas never took A. out into the towns and crossroads in the manner of Francis of Assisi.

concisam: concisam G O S Knöll Skut. Ver.:   conscissam CD Maur.
The loss of his mistress evokes similar language: 6.15.25, `cor, ubi adhaerebat, concisum'.

non in amoenis nemoribus . . .: When A. speaks of the senses, he regularly invokes them in the deliberate rhetorical order here (or its reverse): sight, sound, smell, taste, touch (see further on 10.6.8; cf. his treatment of the temptations of the flesh at 10.30.41-10.34.53). That the senses are the arena of carnal love is suggested in 4.10.15, `non in eis figatur glutine amore per sensus corporis'. We have already seen A. chronicle his early life as a sequence of surrenders to concupiscentia carnis and concupiscentia oculorum. Here, on the death of his friend, he faced a crisis, in which he sought repose. After tears (4.6.11), he now sought rest (`adquiescebat') elsewhere:

1. `non in amoenis nemoribus' (sight),

2. `non in ludis atque cantibus' (sound),

3. `nec in suave olentibus locis' (smell),

4. `nec in conviviis apparatis' (taste),

5. `neque in voluptate cubilis et lecti' (touch). That pattern gives sense to the anticlimax in the final element: `non denique in libris atque carminibus'. First, carmina must mean something other than `song' (as in `wine, women, and song'), which is already represented by cantus. Sense is found if books and poems are taken as the instruments of curiositas--that pursuit of wisdom whose derangement A. has been tracking since 3.4.7. Confirmation comes from civ. 10.9 (on theurgy), `fiebant autem simplici fide atque fiducia pietatis, non incantationibus et carminibus nefariae curiositatis arte compositis, quam vel magian vel detestabiliore nomine goetian vel honorabiliore theurgian vocant.' A. thus says here that his first recourse in crisis was to give way to the temptations to which he had already submitted, and he presents those vices schematically and in the order in which they have already been presented in conf. But there is no rest for A. in these familiar faults, so he will be driven further (see on `veni Carthaginem' below).

Carena, art cit. 69, rightly notes that the structure of the sentence anticipates that of the verse of Paul (Rom. 13.13) that A. will encounter in the garden scene at 8.12.29: but see there for the reflection of the three temptations in the Pauline verse. For the phrases that make up the sentence, Carena supplies the following classical parallels:

amoenis nemoribus: Aen. 6.638-639, `amoena virecta fortunatorum nemorum'.

ludis atque cantibus: Cic. Tusc. 3.20.46., `et corporum complexum et ludos atque cantus'; Hor. carm. saec. 22, `ut cantus referatque ludos'; Val. Flacc. 5.443, `ludos . . . et cantus'.

suave olentibus locis: Hor. serm. 1.4.76, `suave locus voci resonat'; Catullus 61.7, `suave olentis amaraci'; Priap. 3.13, `suave olentia mala'.

conviviis apparatis: Cic. off. 3.14.58, `apparatum convivium'; Livy 24.16.17, `apparata convivia'.

neque: neque C D G O Maur. Ver. Pell.:   nec S Knöll Skut. Vega

aliquantula requies: See on 4.6.11, `requiescebam'.

sarcina: Worldly ways are a burden in conf. (6.6.9, 8.5.12, 8.7.18, 10.40.65), and later, the same noun was often attached to his churchly office (e.g., ep. 31.4, 69.1 [`episcopatus sarcinam']). An implicit alternative underlies his use of the term: 9.1.1, `quo subderem cervicem leni iugo tuo et umeros levi sarcinae tuae' (< Mt. 11.30, where `et sarcina mea levis est' is A.'s preferred text [Milne 36]): also echoed, emphasizing iugum and without the word sarcina at 8.4.9, 10.36.58, 13.15.17. Cf. en. Ps. 59.8, `alia sarcina premit et aggravat te; Christi autem sarcina sublevat te; alia sarcina pondus habet; Christi sarcina pennas habet.' On the word in A., see M. Jourjon, Rech. sc. rel. 43(1955), 258-262; S. Poque, Le langage symbolique (Paris, 1984), 1.64-66.

levanda: Ps. 24.1, `ad te domine levavi animam meam'; en. Ps. 24.2, `“levavi” . . . desiderio spiritali, quae carnalibus desideriis conculcabatur in terra.' Cf. en. Ps. 85.7, `quocumque ergo se converterit, in terrenis rebus amaritudinem invenit; unde dulcescat non habet, nisi levet se ad deum.' N.B. here not `levavi' but `levanda erat . . . sed nec volebam nec valebam' : as often, in retrospect the solution to a problem is implied by the scriptural echo, a solution to which A. was blind at the time.

phantasma: Cf. 4.4.9, `phantasma', and see on 3.6.10.

ibi ponere: i.e., apud phantasma quod mihi erat deus.

per inane labebatur: Carena, art. cit. 70, detects Lucretian flavor in the phrase.

ego mihi remanseram infelix locus: Cf. 2.10.18, `factus sum mihi regio egestatis'; and 13.9.10, `requies nostra locus noster'.

fugeret: Flight from self and flight from God are equally dangerous, reminiscent of the prodigal's career (Lk. 15.13, `profectus est in regionem longinquam' : echoed again at 4.16.30; see on 1.18.28; in the most literal sense, A. is fleeing his father's house [4.4.9] and land, to which he will in the most literal sense return in 388). A. in flight (all pre-conversion): 2.6.14 (`servus fugiens dominum'), 3.3.5 (`fugitivam libertatem'), 4.4.7 (`tu imminens dorso fugitivorum tuorum'), 4.9.14 (`quo it aut quo fugit'), 5.2.2 (`eant et fugiant a te inquieti iniqui . . . quo enim fugerunt, cum fugerent a facie tua?), 6.11.20 (`et ab ea [sc. beata vita] fugiens quaerebam eam'), 8.7.16 (`et videbam et horrebam [cf. here horrebant], et quo a me fugerem non erat').

quo a me ipso fugerem: Hor. carm. 2.16.18-20:

     quid terras alio calentis
sole mutamus? patriae quis exsul
     se quoque fugit?

(Cf. Hor. ep. 1.11.27, `caelum non animum mutant qui trans mare currunt.') Cf. Lucr. 3.1053-1075, on those whom the fear of death inspires to flight, esp. 1068-1072:

hoc se quisque modo fugit, at quem scilicet, ut fit,
effugere haud potis est . . .
. . . morbi quia causam non tenet aeger;
quam bene si videat, iam rebus quisque relictis
naturam primum studeat cognoscere rerum.

Lucretius's solution, like A.'s, came as culmination, after considering the delights of the flesh. L.'s lines depict the frightened man racing distractedly to his villa, seeking release in sleep, then rushing back to the city again, all without clear purpose: against that, knowledge of the natura rerum is the key to happiness. A. likely knew the passage, since it contains both the anticipatory echo of Vergil's famous homage (quoted at civ. 7.9) to L. from geo. 2.490-492 and the virtual title of L.'s poem. (Reference not in Hagendahl: suggested by Verheijen in his ed. ad loc.) Seneca (cited by G-M) ha