Like Inferno IX, this ninth canto is both liminal, marking the boundary between two large areas, and filled with classical reference. And it is the first entire canto devoted to the transition from one poetic zone to another since Inferno XXXI. This sort of self-conscious poetic behavior puts us on notice, from the very outset, that we need to pay particular attention here.

Its reference to Aurora, surprisingly enough, has made this passage among the most hotly debated of the poem. In the 'orthodox' version of the classical myth, Aurora, goddess of the dawn, arose from her couch, where she slept with her aged husband, Tithonus, to rise in the sky on her chariot, announcing the coming of day. A brief and incomplete summary of the debate yields the following (for a summary of the essential arguments over the passage and an attempt to restore Benvenuto da Imola's central and daring reading of it see Hollander [“'La concubina di Titone antico': Purgatorio IX.1,” Electronic Bulletin of the Dante Society of America (July 2001)]: Moore (Studies in Dante, Third Series: Miscellaneous Essays [Oxford: Clarendon, 1968 (1903)]), pp. 74-85, essentially solved this problem almost a century ago, but in fact the early commentators (to whom Moore pays little attention) had already done so. (Paul Spillenger [“Canto IX,” in Dante's “Divine Comedy”: Introductory Readings II: “Purgatorio”, ed. Tibor Wlassics (Charlottesville: Lectura Dantis, University of Virginia, 1993), pp. 128-41], who devotes all but two pages of his 'lectura' of this canto to the opening tercet, is a rare discussant to revisit the glosses of Jacopo della Lana and Benvenuto, only then to deny their applicability.) Nearly all of them are quite sure that Dante has invented a second myth, one in which Tithonus is married to Aurora 1 (of the sun) but has a 'relationship' with Aurora 2 (of the moon). The poetic facts are simple, according to Moore. The time is between 8:30 and 9pm, the cold animal is the constellation Scorpio (and certainly not that belated other candidate, Pisces, arguments for which identification Moore competently dismantles), and thus the aurora we deal with is that of the moon.

For a review of these tormented verses and their tormentors (up to 1975) see Steno Vazzana (“Il Canto IX del Purgatorio,” in Purgatorio: letture degli anni 1976-'79, ed. Silvio Zennaro, Casa di Dante in Roma [Roma: Bonacci, 1981]), pp. 180-85. And for one of the most interesting discussions of their meaning see Ezio Raimondi (“Semantica del canto IX del Purgatorio,” in his Metafora e storia: Studi su Dante e Petrarca [Turin: Einaudi, 1970 (1968)]), pp. 95-98. See also Gino Casagrande (“Il 'freddo animale' e la 'concubina' [Purgatorio IX, 1-6],” in Filologia e critica dantesca: Studi offerti a Aldo Vallone [Florence: Olschki, 1989], pp. 141-59) and Alison Cornish (Reading Dante's Stars [New Haven: Yale University Press, 2000], pp. 68-77). A late 14th-c. Italian MS of Virgil, written by Astolfino Marinoni at Pavia in 1393-94, states that Dante had cited Aeneid IV.585 here: 'And now early Dawn, leaving her saffron bed of Tithonus....' For this and other citations of Dante in early Renaissance commentaries to the Latin classics see Gian Carlo Alessio (“La Comedìa nel margine dei classici,” Studi di filologia medievale offerti a D'Arco Silvio Avalle [Milan: Ricciardi, 1996], pp. 3-25).

Much has been made of the phrase 'where we were' by the 'solar aurorans' in the hopes of counterposing the northern hemisphere (site of the solar Aurora at this hour in Italy) and the southern (where Dante and his companions are becoming sleepy). However, Dante is probably not contrasting the two hemispheres but the glow in the night sky of purgatory that spreads above them and the darkness of their surroundings as night advances. For a similar situation, consider Purgatorio II.8, the phrase 'là dov' i' era' (there where I was) by which Dante refers to his situation in the southern hemisphere looking at the stars from there.

His figure of speech involves mixing metaphors, as the night is given feet, by which she measures her hours, and wings that do the same thing. The meaning is that the time is between 8:30 and 9:00 pm.

In a gesture of some charm but little possible historical foundation Benvenuto (comm. to vv. 7-12) reads the passage as revelatory of Dante's studious burning, not only of the midnight oil, but even that of the pre-dawn hours: 'vir studiosus erat solitus vigilare per tres horas noctis et evigilare in aurora sicut nunc fecit.'

Dante's Adamic sleepiness, that is, the heaviness brought on by his physical being, is adumbrated by a later passage (Purg. XI.43-44), in which Virgil comments upon the difficulties experienced by this living soul as he climbs the mountain in his flesh ('la carne d'Adamo'). But the theme is introduced in the first canto of Inferno (Inf. I.10-12) where Dante's 'sleepiness' is associated with Adam's, according to Hollander (Allegory in Dante's “Commedia” [Princeton: Princeton University Press, 1969]), p. 81n., resuscitating Filippo Villani's gloss on this passage (as well as many another in his commentary), which suggests a figural relationship between the fallen Adam, sent forth into his exile from the garden, and the sinful Dante.

The reader has possibly not remembered that Dante and Virgil are still in company: Sordello, Nino Visconti, and Currado Malaspina have not yet been left behind; thus the group still numbers five.

Dante's dream, occurring some nine hours after he fell asleep, takes place in the pre-dawn moments in which the swallow sings sadly perhaps (if we credit the mythographers, most certainly Ovid (Metam. VI.412-674) in memory of the rape of Philomel by her sister Procne's husband Tereus and her subsequent metamorphosis into a swallow (in Dante's version, where most others prefer the nightingale – see the note to Purg. XVII.19-20, a passage that makes the swallow here necessarily Philomel). Tereus, like Tithonus, has had sexual concourse with each of two sisters – if we accept the notion that the opening passage of the poem posits a lunar aurora (see the note to Purg. IX.1-9).

For the many (and mainly Ovidian) myths referred to in this canto see Picone (“Canto IX,” in Lectura Dantis Turicensis: Purgatorio, ed. Georges Güntert and Michelangelo Picone [Florence: Cesati, 2001 (1999)]), pp. 121-22. For a related topic, Dante's debt to Ovid and his Ovidian tactics in Inferno, see Leonard Barkan (“Taccia Ovidio: Metamorphosis, Poetics, and Meaning in Dante's Inferno,” in his The Gods Made Flesh: Metamorphosis and the Pursuit of Paganism [(New Haven: Yale University Press, 1986], pp. 137-70, 317-27).

Dante's insistence on identifying Procne with the nightingale and not Philomel seems surprising to a modern reader, who has learned the tale the way that he or she believes it is found in Ovid (Metam. VI.424-676). However, consideration of Ovid's verses when he indicates the avian identities of the two metamorphosed sisters yields results that are less than clear (Metam. VI.668-669): 'petit altera silvas, / altera tecta subit' (one heads for the woods, the other for the eaves). Most of us are certain that the woody one is the nightingale, alias Philomela, and that the one who lives amidst our houses is the swallow, alias Procne. Dante, along with some ancient writers, preferred to have Procne as the nightingale – and Ovid's text hardly removes this possibility. A later text in the poem (Purg. XVII.19) makes it clear that Dante refers here to Philomel as the swallow, since there Procne is clearly referred to as the nightingale. The commentary tradition is interesting, in that almost every commentator before Lombardi (1791) simply assumed that the swallow in this passage referred to Procne. Lombardi (comm. to verse 15), paying attention to the passage in the seventeenth canto, is the first to see the nature of the problem: Dante's adhesion to the aberrant classical and post-classical tradition that reverses the more usual labels. Bosco and Reggio (comm. to this passage) argue that, in addition to the possible confusions derivable from Ovid's passage, texts in Virgil (Ecl. VI.79; Georg. IV.15; IV.511-515) also might have made a clear understanding on Dante's part problematic. Still valuable is Moore's brief discussion of the problem (Studies in Dante, First Series: Scripture and Classical Authors in Dante [Oxford: Clarendon, 1969 (1896)]), pp. 209-10; and see Georg Rabuse (“Schwalbe und Nachtigall,” Deutsches Dante-Jahrbuch 38 [1960], pp. 168-92).

The greater truth, indeed the prophetic power, of dreams that came near morning was something of a commonplace in Dante's time. See the note to Inferno XXVI.7. Beginning with Torraca (comm. to this passage) twentieth-century commentators have reminded readers that in Convivio (II.viii.13) Dante had adduced our awareness of our own immortality from the fact that our dreams foretold the future for us.

The formulaic expression (mi parea + vedere) is an earmark of Dante's description of seeing in dream (see also Inf. XXXIII.36; Purg. XV.85-87; Purg. XXVII.97-98). For the consistency in Dante's oneiric vocabulary, dating back to the Vita nuova, see Hollander (“Vita Nuova: Dante's Perceptions of Beatrice,” Dante Studies 92 [1974]), pp. 3-4. For studies in English of the three purgatorial dreams see Hollander (Allegory in Dante's “Commedia” [Princeton: Princeton University Press, 1969]), pp. 136-58; Cervigni (Dante's Poetry of Dreams [Florence: Olschki, 1986], pp. 95-180); Baranski (“Dante's Three Reflective Dreams,” Quaderni d'italianistica 10 [1989], pp. 213-36). See also Stefanini (“I tre sogni del Purgatorio: Struttura e allegoria,” in Studies in the Italian Renaissance: Essays in Memory of Arnolfo B. Ferruolo, ed. G. P. Biasin, A. N. Mancini, and N. J. Perella [Naples: Società Editrice Napoletana, 1985], pp. 43-66). For a recent Italian discussion of the oneiric elements of this canto see Calenda (“Purgatorio, IX: le forme del sogno, i miti, il rito,” Rivista di Studi Danteschi 1 [2001], pp. 284-301).

The reality corresponding to the eagle outside the dream is, naturally, Saint Lucy (identified by Virgil in verse 55), who is bearing Dante higher up the mountain while he sleeps in her arms. But does this eagle have a symbolic valence? Some early commentators (the Ottimo the earliest) read the text strictly literally: the eagle is the bird of Jove (or, perhaps, Jove in the shape of an eagle). However, beginning with Pietro di Dante the eagle is allegorized as divine grace, and then, by various commentators up to and including Giacalone (comm. to vv. 19-21), as one form of grace or another (e.g., prevenient, illuminating, etc.). In the twentieth century there was a vogue for a quite different allegorical reading, the eagle as symbol of empire. (To be sure, this is often, even usually, true in this highly political poem; in this context, however, it seems a forced reading.) It would seem most likely that a literal reading is the best procedure here, following the Ottimo (comm. to vv. 19-24), and simply noting that this eagle is the one who flew off with Ganymede, as the context allows and encourages, i.e., Dante dreams that he was carried off by Jove.

John of Serravalle (comm. to vv. 19-33), following Benvenuto, allegorizes the eagle as divine grace and then equates Dante and Ganymede, thus making Dante 'one who lived with the gods.' Casini/Barbi (comm. to verse 24) suggest that Dante had in mind Virgil's phrase in Georgics I.24-25, 'deorum / concilia' (company of the gods) when he wrote 'al sommo consistoro'; whether he did or not, his meaning seems clear. Within the dream there is a certain aura of violence and fear (implicit reference to the forcible rape of Ganymede by Jove as eagle – see Aen. V.252-257) masking the happier nature of the event: Dante being carried aloft gently by Lucy, and indeed, in a still happier understanding, on the way to the Empyrean, where he will, for a while, share the company of the immortal blessed. In Virgil's ekphrastic moment (this scene appears woven on a cloak as a prize in the funeral games in honor of Anchises) the youthful prince (son of Tros, king of Troy) was hunting on Mount Ida on the Troad (not to be confused with the mountain of the same name on Crete referred to in Inf. XIV.98) and was snatched away from his companions while his dogs barked after him; Dante's version has a more heavenly perspective, and we see the scene from Dante/Ganymede's eyes, as though he were rising in a balloon and watching the shapes of Sordello, Nino, and Currado fall back beneath him (the later poet has had to omit those wonderful Virgilian dogs).

The protagonist's thought within his dream is striking. Since, within the dream, Dante is 'thinking like Ganymede,' his thought refers to a place elevated from the normal, e.g., on this mountain near Troy. (Some commentators want to keep the usual imperial valence of the eagle by associating this Mount Ida with Troy and thus empire; however, the point would rather seem to be that the place is elevated, not that it is Trojan.) And thus Dante would be thinking that only such extraordinary, i.e., 'higher,' mortals like Ganymede and Dante Alighieri are chosen by the gods for their delight. And this thought, perfectly in accord with what we will find out on the first terrace of purgatory proper, associates Dante with the sin of Pride. Once again, however, the 'reality' tells a different story: the true God is not interested in Dante's curly locks, but in his Christian soul; and He will pluck Dante up from the mount of purgatory for reasons better than those that motivated Jove.

The issue of Jove's homosexual desire for Ganymede is mainly avoided in the commentaries. It is, nonetheless, noteworthy that, of the many myths available to Dante that might have expressed the love of the gods for a particular mortal, he has chosen this one. For the question of Dante's attitudes toward homosexuality see Hollander (“Dante's Harmonious Homosexuals [Inferno 16.7-90],” Electronic Bulletin of the Dante Society of America [June 1996]). Durling (The “Divine Comedy” of Dante Alighieri, Notes by R. M. Durling and Ronald Martinez [New York and Oxford: Oxford University Press, 1996]), pp. 559-60, on the basis of no known evidence, is of the opinion that Dante was of homosexual predisposition but had never acted on his desires. While that is probably more than can be shown to be true, the question of Dante's rather 'unmedieval' view of homosexuality (see discussion in the note to Inf. XV.13-21) has generally not been dealt with as openly as it ought to be.

The eagle's descent may have still another Virgilian provenance: Aeneid XII.247-250, as Tommaseo (comm. to these verses) was perhaps the first to suggest; he has been followed by a number of others. In that scene an eagle, described as 'Jove's golden bird,' offers an omen (arranged by Turnus's sister, the nymph Juturna) when it dives from the sky to snatch a swan out of the water and carries it off as its prey. (This much of the drama bodes ill for the Trojans, but they are heartened, unfortunately for them, in the final result, when the rest of the water fowl attack the eagle and it drops the swan.) The language is pertinent: Jove's golden bird is attacking the 'litoreas... avis turbamque sonantem / agminis aligeri' (the fowl along the shore, the clamorous crowd in their wingèd band). From among this agmen aliger the eagle picks one. For the pun available to Dante on his family name (Alighieri/aliger) see the note to Inferno XXVI.1-3. It seems possible that here Dante is conflating the two Virgilian passages in which a Jovian eagle seizes its prey and enjoying the coincidence that, in the last of them, that prey is associated with his own name, since he, too, while dreaming, is being lifted skyward in the talons of Jove.

The fire alluded to here is the ring of fire that was believed to surround the closer-to-earth sphere of air, just before one might reach the moon. That this is the 'tanto... del cielo acceso' (so much of the sky set afire) of Paradiso I.79 was possibly first suggested by Lombardi (comm. to Par. I.79-81). Thereafter it became a commonplace in the commentaries.

Once again the negative version of events put forth in the dream has a better meaning. It seemed that the eagle and Dante were consumed in the ring of fire high above the earth, while actually Dante and Lucia have risen to the gate of purgatory, as we shall shortly find out, and Dante is being awakened, not by the pain of death, but by the late-morning sun on his eyelids (Purg. IX.44). If there is a further significance to this detail, it would seem to refer to Dante's eventual arrival at the true 'sphere of fire,' the Empyrean.

The poet describes the narrator's awakening in terms that recall Statius's text (Achilleid I.247-250), describing the stratagem employed by Thetis, Achilles' mother, in order to keep him from being 'drafted' into the Trojan war. Taking him from the care of his tutor, the centaur Chiron (see Inf. XII.71), Thetis carries him in her arms, sleeping, to the island of Scyros. Again Dante adverts to a mythic narrative that has a tragic result; Thetis's benevolent caution will not prevent the coming of Ulysses and Diomedes to Scyros and the eventual death of Achilles in the war. Dante's 'comic' reality counters the Statian tragedy: Achilles is carried down from his mountain homeland to an island from which he will go off to his death; Dante is carried up a mountain situated on an island toward his eventual homeland and eternal life. Rarely in the Commedia is the contrast between classical and Christian views, between tragedy and comedy, more present than in these classicizing passages that open this canto. It is also true that the protagonist, as he experiences these new things, behaves very much as the 'old' man that he still is, and assumes that terror is a valid response to these miraculous events that, the reader can see, speak only of God's love and protection for even such a sinner as Dante.

Dante and Virgil have left their companions behind, down the mountain's slope, and are facing the east, the sun in their faces as the morning advances.

As though to remind the reader that all the material relating to Dante's dream did have a counterpart in reality, Virgil's explanation 'glosses' the dream as it explains the coming of Lucy, while Dante slept, at the solar aurora, nine hours after he had seen the lunar aurora. Sometime after dawn she began her ascent with Dante in her arms, leaving their companions in ante-purgatory.

St. Lucy remains one of the more problematic presences in Dante's poem. Exactly who is she and why is she so important to him? Most commentators take her to be the early-fourth-century martyr from Syracuse, killed while Diocletian was emperor ca. A.D. 304. She is usually associated with the well-being of the eyes, and this may have had some resonance for Dante who, in Convivio III.ix.15, reports a severe bout of eye trouble in the same year that he was composing his ode 'Amor che ne la mente mi ragiona.' For whatever reason (and we shall probably never know it), Dante was obviously particularly devoted to the cult of this saint. She has a presence in three major scenes in the work, the 'prologue in heaven' (Inf. II.97-108), the transport of Dante while he sleeps in this canto, and the prospect of the inhabitants of the 'stadium-rose' (Par. XXXII.137).

Moore (Studies in Dante, Fourth Series: Textual Criticism of the “Convivio” and Miscellaneous Essays [Oxford: Clarendon, 1968 (1917)]), pp. 235-55, addresses the problem of an allegorical understanding of Lucy. Does she represent 'illuminating grace'? 'cooperating grace'? Gradually Moore undermines both these allegorical formulations and moves toward the notion of Lucy as Dante's 'patron saint' (p. 241), offering in evidence the phrase 'il tuo fedele' (Inf. II.98) that reappears variously (Dante as Beatrice's 'fedele' in Purg. XXXI.134; Bernard as Mary's 'faithful one' in Par. XXXI.102) to suggest that the expression 'implies the relation of one person to another person as such, and not as a symbol or type' (p. 243). It was important that a reader like Moore, who still believed in the allegorical formulations that equated Virgil with Human Reason and Beatrice with Revelation or the Church (or, as Moore adds [p. 243], 'or whatever it be'), finds himself wanting to be rid of the rather fanciful associations that had previously been forced upon St. Lucy. More recently Jacoff and Stephany (Lectura Dantis Americana: “Inferno” II [Philadelphia: University of Pennsylvania Press, 1989]), pp. 29-38, crediting Pasquazi (“Il canto II dell'Inferno,” in Inferno: letture degli anni 1973-'76, ed. Silvio Zennaro, Casa di Dante in Roma [Roma: Bonacci, 1977 (1974)]), pp. 35-65) for his important contribution before them, try to restore the importance of the cardinal virtue fortitude in Dante's fashioning of the meaning of Saint Lucy. See also Anthony Cassell (“Santa Lucia as Patroness of Sight: Hagiography, Iconography, and Dante,” Dante Studies 109 [1991], pp. 71-88).

Little sense has been made out of the fact that in Convivio III.v.10-18 Dante invents two imaginary cities, which he designates Maria and Lucia, situated at the north and south poles respectively. Is it significant that Lucy appears here, at the south pole, where, in the previous work, there stands an imaginary city bearing her name? For recognition of the problem represented by those two cities, without attempts at a solution, see Poletto (comm. to Inf. II.97-99 – and Poletto seems to have been the only commentator, at least among the seventy currently gathered in the DDP, to mention this intriguing passage), Jacoff and Stephany (Lectura Dantis Americana: “Inferno” II [Philadelphia: University of Pennsylvania Press, 1989]), pp. 32-33. And now see Cornish (Reading Dante's Stars [New Haven: Yale University Press, 2000]), pp. 73-75.

Dante's line, 'poi ella e 'l sonno ad una se n'andaro' (then she and sleep, as one, departed) reminded Vellutello (comm. to vv. 64-69) of the end of the dream in Cicero's Somnium Scipionis (VI.29): 'Ille discessit, Ego somnio solutus sum' (He departed, and I was released from sleep). Gabriele (comm. to this verse) had earlier signaled this resonance, and added also that of Aeneid VIII.67, 'nox Aenean somnusque reliquit' (night and sleep both left Aeneas), a passage also cited by his student, Bernardino Daniello, discussing this verse.

A final simile prepares us for the entrance to purgatory proper, comparing Dante to one who moves from dubiety to confidence, a movement that required that he reinterpret the dream and his associations in a positive light.

At almost the precise midpoint of the canto Dante situates his second address to the reader of this cantica (see Purg. VIII.19-21 for the first). It has caused two major interpretive problems, twice dividing its readers into two basic groups. First, there are those who believe that it speaks of an increase of quality in the artistry employed by the poet, while others contend that it speaks rather of an increase of quantity. This dispute is most readily understood by referring to translations of the poem; those in the first group have Dante say that he will employ better art; those in the second, more of it. The early commentators, here represented by Benvenuto, think of the more elevated matter of purgatory needing to be supported by 'fictionibus magis artificiosis et sententiosis' (more elaborately worked and more highly signifying fictions). Or, in another formulation, the art to describe the higher mystery represented by the portal of purgatory proper needs to be more artful (in this case allegorical) than at other times, but not necessarily more beautiful, the eventual Romantic view, yet one that is found even in John of Serravalle (comm. to these verses), who argues that the poem must from here on be more beautiful. Costa (comm. to verse 71) continues Benvenuto's spirit: the reader should not be surprised, in his opinion, that, as Dante says, 'io cerco di sostenere con più artificiose parole la materia sublime di che favello' (I seek to reinforce the sublime matter of which I speak with more elaborately worked words).

A second question remains: does this heightening reflect just the scene that follows (as in the opinion of a minority, all responding to these verses, e.g., Tozer; Momigliano; Porena; Sapegno, who has a foot in both camps; Mattalia, or so it would seem; Giacalone; while Bosco insists that the phrase refers to both 'arty' halves of the canto), or all of the Purgatorio that is to come? For the latter opinion see Francesco da Buti (comm. to vv. 64-72), who argues that, since the subject is now penitence, that is the higher matter that requires more art; in one form or another, this is the position taken by most commentators. In the opinion of this writer, the address to the reader refers to the description of the gate, its warder, and the three steps in the rite of confession, all of which need to be understood in the tradition of the allegory of the poets, as we shall see. It regards, in other words, matter both local and temporary. In a similar vein see Vazzana (“Il Canto IX del Purgatorio,” in Purgatorio: letture degli anni 1976-'79, ed. Silvio Zennaro, Casa di Dante in Roma [Roma: Bonacci, 1981]), pp. 177-80, arguing that what is at stake is the allegorical nature of what immediately follows.

The angel is seated not upon (as some translate) the highest step but above it, as will become clear from verse 104, where we learn that he is seated on the threshold of the gate with his feet upon the third step.

Beginning with Scartazzini/Vandelli (comm. to vv. 79-84), commentators resort to the Bible (Matthew 28:3) for the stern brightness of the angelic countenance: 'Erat autem aspectus ejus sicut fulgur' (His countenance was as the lightning).

The warder's sword is 'the sign of one who has authority to pronounce sentence,' in the words of Singleton (comm. to this verse), citing Benvenuto da Imola. Singleton goes on to add that it is also 'reminiscent of the Cherubim with the flaming sword that were placed to guard Eden after Adam and Eve were driven forth.'

The warder's words of challenge, 'Dite costinci' (Say it from there) repeat, as many note, the challenge of one of the centaurs to the approach of Virgil and Dante (Inf. XII.63) and, like that one, probably reflect Charon's similar words (fare age... iam istinc [tell me... right where you are]) to the invading Aeneas in Aeneid VI.389, as Daniello (comm. to vv. 85-87) was perhaps the first to note.

This scene represents an epitomizing replay of Virgil's and Dante's encounter with the guardian of all purgatory, Cato (Purg. I.40-93): challenge by the guarding presence, who wants to know if some higher authority permits this visit; Virgil's response indicating a female who had sent these 'pilgrims' on their journey (this time with no attempt to flatter the warder); the warder's courteous acceptance of the aspirants' desire to enter a sacred precinct.

The allegory of the three steps should be less difficult than it has proven to be. Considering this problem, Carroll (comm. to these verses) cites Milton, describing the gate of heaven (Paradise Lost III.516): 'Each stair mysteriously was meant.' But what exactly does each of Dante's steps 'mean'? Catholic doctrine, as variously expressed, presents the path to absolution from sin as running, in the Sacrament of Penance, from contrition (the recognition and heartfelt rejection of a sin) to confession (of the sin, voiced to a priest), to satisfaction (in the promise to perform an act of penance as ordered by that priest, thus showing the genuineness of the confession). This is the psychologically correct order as well as the one given by 'Scholastic and other Church writers uniformly' (Moore [Studies in Dante, Second Series: Miscellaneous Essays (Oxford: Clarendon, 1968 [1899]), p. 47), i.e., one is contrite, confesses, and then performs acts that will lead to absolution, in the culmination of the sacrament. Dante's first commentators are, however, more or less evenly divided as to whether the three steps found here represent, in sequence, (1) contrition, (2) confession, (3) satisfaction (this group is saving Dante from himself, as it were, i.e., they record what he should have said) or (1) confession, (2) contrition, (3) satisfaction. Holding for the first order see the Ottimo (comm. to vv. 94-95), Benvenuto (comm. to vv. 94-99), Anonimo Fiorentino (comm. to vv. 94-96), Serravalle (comm. to vv. 106-108); (2): Lana (comm. to vv. 76-77), Pietro (Pietro1, comm. to verse 76), Buti (comm. to vv. 94-105), Andreoli (comm. to verse 94). However, it is fairly clear from the text that Dante has reinvented the order to suit his own purpose, beginning with confession and only then proceeding to contrition. That this is almost certainly true is confirmed by a later text, Purgatorio XXXI.31-90, which offers a carefully orchestrated presentation of Dante's own penance before Beatrice, with the steps of that paralleling the steps found here, namely, confession, contrition, satisfaction, in that order. Moore also points out that Dante's presentation of the third stage, satisfaction, is unorthodox, since he represents it as the love that came from Christ's self-sacrifice. Thus, in Dante's scheme, the sinner first confesses his sin, then feels true contrition for it, and then moves beyond it in his imitation of and love for Christ. Why he should have wanted to repackage the elements of what was a standard body of doctrine and belief is not a subject that has received adequate attention. But we should not be surprised to find that this poet remakes any text or any doctrine to his own liking and for his own reasons. The one element that does run through Dante's version of this sacrament is that its priestly element is curtailed in favor of inward recognition and performance on the part of the sinner, i.e., confession in Dante seems more a private form of self-recognition than is generally the case, contrition contains mainly internal elements (if it must eventually be given voice), and satisfaction seems more attitudinal than performative. It also seems that Dante has moved elements of satisfaction into the second stage of the process, contrition.

Pietro di Dante (Pietro1, comm. to verse 130) is alone in referring to Psalm 99:4: 'Introite portas eius in confessione' (100:4, Enter his gates confessing your trust), as reflecting the situation here in the poem.

Some years ago Gregory Curfman (Princeton '68) suggested that the 'narrative' told by the three steps offered an epitome of the history of the human race: purity, sin, redemption.

For convincing disagreement with the notion that the steps actually represent not the process of the Sacrament of Penance but three stages in the sinner's contrition see Bosco/Reggio (comm. to verse 94).

The first step, of white marble, serves as a mirror to the protagonist and thus seems associated with confession.

The second, darker than the purple-black of the color perse and broken by a cruciform crack, represents the sinner's recognition of his 'broken' state.

The third, red with the color of spurting blood reminiscent of the sacrifice of Jesus Christ, is, like the first, of a stone of lofty character, porphyry (the second seems to be humble geological matter indeed), and suggests the sacrifice the sinner must make in imitation of the great sacrifice made by Christ.

Only recently have commentators to this passage (Mattalia, Fallani, Bosco) turned, for the source of Dante's diamante, to Ezechiel's adamantem, when God gives his prophet a stony forehead to wear against his enemies (Ezek. 3:9). And since the priestly angel, seated upon the adamant threshold, is iconographically related to St. Peter, a number of commentators think of Matthew 16:18: 'You are Peter and upon this rock will I build my church.'

Either in December 1310 at Vercelli or in January 1311 at Milano, Dante was apparently presented to the new emperor, Henry VII (see Frugoni's notes in his edition of Dante's Epistles, p. 564). Dante himself reports that he embraced the emperor's feet (Epistle VII.9). It is at least conceivable that this verse remembers that experience, especially since, if it was written after the event, it was probably written soon after it. Once again we are unable to be certain, because of the uncertainties that attend dating the stages of the poem's composition.

Dante's threefold beating of his chest has been glossed, since the time of the Ottimo (comm. to vv. 106-111), as signifying 'mea culpa, mea culpa, mea maxima culpa.' Andrea Lancia (the probable author of that commentary) proceeds, followed in this, as well, by many later commentators, to say that these three in turn signify three kinds of sin: sins of thought, of the tongue, and in deed.

The seven P's, generally understood as deriving from the first letter of peccata (sins), evidently stand for the seven mortal sins (or capital vices), but here, since the protagonist has learned to hate sin, they stand for what remains of the predisposition to these seven vices that is inherited by every mortal through Adam's (and Eve's) original sin. For the P's as deriving from the letter tau see Sarolli (“Noterella biblica sui sette P.,” Studi Danteschi 34 [1957]), pp. 217-22, arguing for a source in Ezechiel 9:2-6, where God commands a scribe to write that letter – as a positive sign in that case – on the foreheads of all the inhabitants of Jerusalem who repent the abominations done in the city. In the following slaughter of the rest of the inhabitants, only Jeremiah and the other just Jerusalemites are preserved. See also Revelation 7:3, for the true believers who bear the sign of God on their foreheads, as well as Revelation 13:16 and 20:4, where those who worship the Beast have his sign on their foreheads.

For some time now a debate among the commentators has involved the question of whether or not others on the mountain possess these P's (i.e., whether or not the P's on Dante's forehead are unique). Two differing reasons help us to be fairly certain that they are in fact unique to him, the first one positive: Dante is the sole visitor to purgatory who climbs there in the flesh; his uniqueness thus has this further sign. Second, and arguing from negative evidence, one may say that, since no other character is ever observed bearing these stigmata, one may reasonably conclude that none has them. One may imagine the various human scenes that the poet could have developed, as various characters on the mountain compared P's with Dante, etc. He simply did not choose to handle his material in that way and those who wonder whether others bear P's on their foreheads have simply invented a problem. See the note to Purgatorio XXI.22-24.

Sin as a 'wound' is a biblical topos (see Isaiah 1:6 [first noted by Scartazzini (comm. to Par. VII.28)], Psalm 38:11 [39:10 – as noted by Singleton (comm. to this verse)].

The color of the warder's garments is the gray of ashes and gives expression to his humility, despite his high office.

For reasons to associate the color of the angel's vestments with Franciscan garb, see A. Pegoretti, “Immaginare la veste di un angelo: il caso di Purg. IX, 115-16,” L'Alighieri 27 (2006): 141-50.

Sensible allegorical expositions of the two keys are found variously, but of particular use is Poletto's gloss (to vv. 115-117). The golden key denotes the authority to absolve granted by Christ directly; the silver, the judgment necessary in the priest to be sure the penitent is truly deserving of absolution. Poletto cites passages in St. Thomas to show Dante's closeness to them in this part of his description of the process of entering purgatory (Summa Th., III, Suppl., qq. 17-20). Once the priest has judged the penitent ready for absolution (using his silver key), he then uses the golden one to complete the opening of the door. The priest, of course, may err in wanting to allow an unfit soul to enter; in that case the golden key will not turn in the lock – but even so, God is disposed to err on the side of mercy and will overrule a prelate who is niggardly in pardoning.

The fact that a priest may err in his judgment makes it disturbing that this figure is presented as being literally an angel (who thus should be free of such weak discernment). It would seem more logical if we dealt here with an allegorical figure, Priestliness, the Petrine warder of the gate, a composite figure representing a class, not a particular historical being. And, indeed, the angel does not behave in any other way.

See Christ's words to his disciples (Luke 9:62): 'Nemo mittens manum suam ad aratrum, et respiciens retro, aptus est regno Dei' (No man, having put his hand to the plough, and looking back, is fit for the kingdom of God). This apt passage has been cited in this connection since the time of Pietro di Dante (Pietro1, comm. to these verses). While Lot's wife (Genesis 19:17; 19.26) may also be remembered here, the passage from Luke's gospel is more closely related. The same may be said for the resonance of Orpheus's backward glance at Eurydice. Luke's passage, insisting that a backward glance will cost the kingdom of Heaven, is perfectly a propos (as Timothy Hampton, then a graduate student at Princeton, suggested in 1984) because it alone has the object that may be lost ahead of (rather than behind) its seeker.

This is the first part of one of the most difficult canto endings in the Divine Comedy. These lines remember a terrible moment of Roman history, while the second part (Purg. IX.139-145) reflects the musical practice of Dante's day.

For hundreds of years Rome had kept a part of its treasure secure from any use, locked behind a portal that was never opened, until Caesar, in order to fund his pursuit of Pompey and Cato in 49 B.C., broke into the treasury and looted it, overcoming the resistance of a single brave republican, Metellus, loyal to Pompey. Dante's source here is the violently republican Lucan (Phars. III.108-168). It is worth reading the entire passage, which most commentators apparently do not do, for it drips with sarcasm about Julius Caesar, from its inception, in which Caesar, having marched on Rome and conquered by arms, has become everything and the senate has become a mouthpiece for this 'private man,' to its conclusion, in which the city is portrayed as being poorer than the one man who rules her as a result of his plundering the temple of Saturn, her treasury. There is nothing good here about Caesar, despite Dante's respect for him as the person he considered the first emperor of Rome (see Stull and Hollander, “The Lucanian Source of Dante's Ulysses,” Studi Danteschi 63 [1991 (1997)], pp. 33-43, for Dante's mainly negative, if mixed, views of Julius in the poem). And thus the sound that sounds so shrilly at Dante's entrance into purgatory is reminiscent of what, for Lucan and for Dante, is perhaps the nadir of Roman history, the accession of Caesar and the destruction of the republic. For Dante's fervent belief in the republican virtues and form of government see Hollander and Rossi (“Dante's Republican Treasury,” Dante Studies 104 [1986], pp. 59-82). Along with the passage in Lucan, undoubtedly Dante's main source here, Tommaseo (comm. to vv. 133-135) brings into play Aeneid VI.573-574: tum demum horrisono stridentes cardine sacrae / panduntur portae (then at last, grating on their hinges, the impious gates swing open). Virgil is describing the gates of Tartarus, swinging open (the Sibyl and Aeneas do not enter, but she does tell him of the horrors of the punishments therein). Here, too, we can see how Dante has juxtaposed two similar objects, the gates of Tartarus, the pagans' hell, and those of purgatory, and make the reader aware of the crucial similarity that marks their utter difference.

What is the effect of such negative reminiscences as the protagonist begins to attain the Promised Land? One must conclude that we are dealing here with antithesis: as brutally shrill as was the sound of the squealing doors of the temple of Saturn, of the gates of Tartarus, exactly so terribly loud is the rare victory of a penitent being allowed to enter the kingdom of Heaven – or its vestibule. What was tragic in its consequence for Rome is marked by a sound exactly as loud and grating as this one that announces the victory of a new (and better) Caesar who enters not against the will of the warder, but in accord with it. Only the sounds are similar; all else is changed. And, as we have seen occur several times in the first half of this canto, tragic classical myth or history gives way to comic Christian narrative. In the words of Jesus (Luke 12:34), 'There where your treasure is, there your heart is also.' Caesar's treasure is far different from the treasure in Heaven sought and found by only relatively few Christians, their low numbers suggested by the infrequent screechings of this gate.

As for the positive resolution for the unpleasant sound of the opening gate, Denise Heilbronn (“Concentus musicus: The Creaking Hinges of Dante's Gate of Purgatory,” Rivista di studi italiani 2 [1984], pp. 1-15) points out that medieval concepts of musical 'sweetness' had more to do with harmony than with the sounds themselves. She (p. 4) gives the examples of the hurdy-gurdy and drum, both of which would hardly seem to be 'sweet' to modern ears, but did to those of the time who recorded their responses. This point is a pivotal one for those who cannot bring themselves to see how the grating screech of a gate can be in harmony with another sound. Yet when we reflect, along with Heilbronn, that what the gate's sound announces is very sweet indeed, we may begin to understand Dante's strategy here. M.A. Buchanan (“At the Gate of Purgatory [Purg. IX, 130-145, X, 1-6],” Italica 25 [1948], pp. 6-7) points out that there are three musical moments in the scene: the roaring of the hinges, the harmonious singing of the angels who are apparently within the gates, and the sound of the gate closing behind Dante in the fourth verse of the next canto.

The noun regge represents a relatively rare term (one never used elsewhere by Dante) for the main portal of a church. Denise Heilbronn (“Concentus musicus: The Creaking Hinges of Dante's Gate of Purgatory,” Rivista di studi italiani 2 [1984]), p. 5, suggests that 'like a cathedral door, the gate of purgatory is the mystical image of the gate of heaven.'

This passage, too, has caused a great deal of difficulty. Denise Heilbronn (“Concentus musicus: The Creaking Hinges of Dante's Gate of Purgatory,” Rivista di studi italiani 2 [1984], pp. 1-15), a Dante scholar with a musical background, has dealt with a number of the issues that have puzzled readers and offers a helpful review of the extensive discussion. Some of the essential matters in dispute involve the words tuono, voce, suono, and organo.

According to her, tuono (understood as 'a note' and not as 'thunder') should be seen as positive, since it is the sound that accompanies a soul's entrance into purgatory; voce and suono are, respectively, technical terms for the human voice and an inanimate, instrumental sound (pp. 6-7), while organo refers either to polyphonic singing or singing accompanied by an organ. Heilbronn is illuminating about the use of impressively large organs in churches in the tenth and eleventh centuries (pp. 8-10). However, some contemporary scholars have moved back to the notion of the first commentators that the phrase 'singing with organs' (cantar con organi) referred to vocal music sung polyphonically, no modern commentator more convincingly than Fallani (comm. to verse 144), revisiting a conference paper given by Casimiri in 1925. But counter-arguments by Giacalone (comm. to vv. 142-146) and Bosco/Reggio (comm. to vv. 144-145) seem even more convincing. Giacalone points out that Dante himself had described an organist as accompanying a song (De vulgari eloquentia II.viii.5-6) and Bosco/Reggio, influenced by the paper given by Damilano in 1974, remind us that one of Dante's commentators, Cristoforo Landino (1481), grandnephew of the famous organist Francesco Landino, is of the opinion that the practice was to alternate passages of singing and organ-playing in church services and that this is referred to by Dante here.

Dante believes he hears the words of a hymn being sung (and we must imagine that, if there was actual singing to greet his coming [Dante only says that he seemed to hear voices], it was done by angels, since the penitents we eventually see in the next canto, the prideful, are bent under their weights and far from lyrical). Te Deum laudamus has an interesting history in the commentaries. (For the text of the hymn in Latin and English see Singleton, comm. to verse 140.) Benvenuto, responding to vv. 139-140, claims that St. Ambrose wrote this hymn after he had served as St. Augustine's spiritual doctor and cured him of his terrible errors (in Milan shortly before Augustine's conversion); it is thus, Benvenuto continues, a most fitting accompaniment to Dante's – another great intellect's – turning to penance. Other early commentators also associate the hymn with Augustine's conversion, whether it was sung while he was being baptized or spoken by Ambrose in his sermon on that occasion or, indeed, according to Francesco da Buti, spontaneously spoken responsively by these two great men on that day. While in our time it is not believed to have been composed by Ambrose, in Dante's it was. That Dante should have chosen to present himself, entering purgatory, as a new (and better) Julius Caesar and as the new Augustine is both altogether extraordinary and completely Dantean. For the opinion that Dante presents himself as being like Augustine in the sins he must conquer, lust and bad philosophizing, see Hollander (Allegory in Dante's “Commedia” [Princeton: Princeton University Press, 1969]), p. 165n.