Dante apparently believed that the rays of the moon, in Dante's time considered a 'cold planet' (e.g., by Jacopo della Lana [comm. to these verses]), enhanced the natural nocturnal cooling of the earth, its temperature further decreased whenever another 'cold' planet, Saturn, was visible above the horizon. The hour is just before dawn on Tuesday, the beginning of Dante's third day at the antipodes. For an earlier Dantean reference to the coldness of Saturn see Convivio II.xiii.25.

The early commentators who deal with the problems encountered here are in fairly close accord. Beginning with Benvenuto da Imola they indicate the following: geomancers are diviners who create charts based on random points on the earth's surface and drawn in the sand (and later copied onto paper or in sand on a tabletop) that can be linked in such a way, by joining various of these points with lines, to create a number of figures (Daniello [comm. to these verses], following Landino, names sixteen of these). The facts behind the passage seem to be pretty much as Grandgent, glossing it, said: “'Geomancers' foretold the future by means of figures constructed on points that were distributed by chance. Their specialty was the selection of favorable spots for burial. They were the first in Europe to use the compass. One of their figures, called fortuna major, or 'greater fortune,' resembled a combination of the last stars of Aquarius and the first of Pisces. As these constellations immediately precede Aries, in which the sun is from March 21 to April 21, the figure in question can be seen in the east shortly before sunrise at that season.” Their name, geomancer, reflects the fact that such an adept draws his figures in the sand (or earth – Greek ge) and that he is a diviner (Greek mantis). The configuration known as Fortuna maior is illustrated by Benvenuto and others as shown here:

 * *
 * *
 * *

These stars are found in the east of the pre-dawn sky, shortly before the sun rises in Aquarius in the spring, since the last stars in Aquarius and the first in Pisces are visible in the hour before dawn. Some commentators insist that all six stars are found either in Aquarius or in Pisces alone, but others object (e.g., Chiavacci Leonardi [Purgatorio, con il commento di A. M. C. L. (Milan: Mondadori, 1994)], p. 556, countering the first of these proposals). For the Christological resonance of Pisces (another reason to accept the early commentators' view that the stars are not found in Pisces alone?) see the note to Purgatorio I.19-21. This passage may lie behind Chaucer's setting for the heroine's aubade (dawn song) in Troilus and Criseyde III.1422-1442, spoken while Fortuna Major was rising in the east (III.1420).

Whatever the precise nature of the practices of geomancers, it seems clear that Dante has not taken six lines to indicate that the time was shortly before dawn without purpose. Surely the unpleasant and unsavory connotations of coldness and of divination (we remember the treatment of diviners in Inferno XX) color our reception of the dream that shortly follows. We should also remember that this dream occurs on the Terrace of Sloth, thus suggesting that it may reflect Dante's own former tardiness in seeking the good. Insofar as that affliction also encouraged his involvement in cupidity, the dream may also look back to some sort of misdirected love.

This second Purgatorial dream is at least as difficult to interpret as the first (see the note to Purg. IX.19). For a brief and cogent review of Classical, scriptural, patristic, and Scholastic views of the nature of dreams see Peter Armour (“Divining the Figures: Dante's Three Dreams in the Purgatorio,” in Melbourne Essays in Italian Language and Literature in Memory of Colin McCormick, ed. Tom O'Neill [Dublin: Irish Academic Press, 1990]), pp. 13-16.

Small seas of ink have been poured out in the quest for the source and meaning of this unpleasant woman. The far from convincing results previously obtained probably should warn anyone against advancing an opinion. On the other hand, it seems to some that the problem is easier to understand than are the attempts to solve it. The poem itself, in the words of Virgil, tells us precisely who the stammering woman is: she represents the conjoined sins of excessive love, avarice, gluttony, lust – the sins of the flesh or, in the language of Dante's first cantica, the sins of incontinence. (See Virgil's words at vv. 58-59: 'saw... that ancient witch / who alone is purged with tears above us here.') Dante's dream, nonetheless, must surely also have specific meaning for him. If the woman is the object of his affection, she must have particular reference to lust, since the poem nowhere offers any indication that Dante considered himself ever to have been avaricious (or prodigal, for that matter) or gluttonous. The 'good that fails to make men happy' (Purg. XVII.133), in Dante's case, must then nearly certainly be understood as involving wrongful sexual desire.

Consideration of the femmina balba (stammering woman) has caused readers to seek out some fairly recondite sources. Giuseppe Toffanin (“La 'foetida Aethiopissa' e la 'femmina balba,'” Giornale storico della letteratura italiana 77 [1921], pp. 147-49) argued for a foul female creature described in the Vitae patrum; Francesco Mazzoni (ed., La Divina Commedia: Purgatorio. Con I commenti di T. Casini/S.A. Barbi & A. Momigliano [Florence: Sansoni, 1977]), p. 436, presented still another potential source, an apparently fair but actually foul woman described in the Tabula exemplorum secundum ordinem alphabeti. The common problem for these 'sources' is that we have no evidence indicating that they were known to Dante. In recent years still other recondite candidacies have been put forward: Paolo Cherchi (“Per la 'femmina balba,'” Quaderni d'italianistica 6 [1985], pp. 228-32) urges that of Bernard of Gordon, author of the Lilium medicinae, but admits that proving Dante's knowledge of this text or even acquaintance with the interesting passage that he cites from it (p. 230) in some other source is problematic; Michelangelo Picone (“L'Ovidio di Dante,” in A. Iannucci, ed., Dante e la “bella scola” della poesia [Ravenna: Longo, 1993]), p. 134, reaches out for the nasty widow of Ovid's De vetula; Giovanni Parenti (“Ercole al bivio e il sogno della femmina balba,” in Operosa parva per Gianni Antonini, ed. D. De Robertis & F. Gavazzeni [Verona: Valdonega, 1996], pp. 57-66) investigates possible resonances of various potential sources, including (pp. 55-59) the theme of Hercules at the crossroads found in Cicero (De officiis I.118) and (pp. 61-62) the 'idols of the gentiles' found in a Psalm certainly known well by Dante, the In exitu (Psalm 113B:4-7). For a more general analysis of the passage containing the femmina balba see Dino Cervigni (Dante's Poetry of Dreams [Florence: Olschki, 1986]), pp. 123-35.

What has rarely been noted in modern commentaries (but see Mattalia [comm. to verse 7] and, citing him, Giacalone [comm. to vv. 25-27]) is the fact that balbus is the contrary of planus, the word that describes Beatrice's speech in Inferno II.56 (see the note to Inf. II.56-57). Cf. the entry balbus in Lewis and Short's Latin Dictionary, where this view is confirmed. And see Cicero, De oratore (I.lxi.260), where stammering (balbus) Demosthenes labored so assiduously to correct himself in the pronunciation of initial 'r' that no one could eventually be considered to pronounce it more clearly (planius) than he. For the 'higher' meaning of planus see the Codice Cassinese (at verse 56): 'sermo divinus suavis et planus esse debet' (divine speech should be smooth and clear).

The woman, we learn from the preceding tercet, stammers, is crooked in her glance as well as in her extremities, and sickly in her complexion. That is her natural condition. Dante, in the logic of the dream, redresses each of these sets of flaws, making her speech fluent, straightening her limbs, and making her facial complexion the color that love desires to find in a woman (commentators debate whether this is red or white, but only since the time of Tommaseo [the early commentators do not treat the question]; Tommaseo (comm. to vv. 13-15) opts for the darker hue (purple, red); Bianchi, citing Vita nuova XXXVI.1, where the 'color of love' is the pallor Dante finds in the donna gentile, the woman who replaces dead Beatrice in his affections, chooses the lighter: white. Since then, there has been continuing dispute, which is easily enough understood, since the ugly nature of the woman has been characterized by her 'sickly pale complexion,' given the fact that this has obviously been changed under Dante's sun-like gaze. As a result, many commentators think her color must have been the rosy red of sexual excitement. But Dante's own earlier insistence on the white complexion that is the 'color of love' surely supports Bianchi's view: From an unpleasant sickly pallor (i.e., a loss of whiteness, as the hapax scialba in verse 9 suggests, from the Latin exalbare, here meaning 'unwhitened' [see Poletto, comm. to vv. 1-9]), the skin of the woman's face turns to the elegant paleness of a lady who draws the affection of her beholder. This is not the color of passion in her, but the color which causes passion in a male beholder (a male might indeed not think her a 'lady' were her face aglow with passion; one may make love to 'women,' one falls in love with 'ladies'). While the commentators remain divided, opting for a shade of red, a whiteness, or a combination of the two (all of which may be found in the lengthy tradition of the 'colors of love,' at least from Ovid onward), the context of Vita nuova, which sponsors pallor as the 'color of love,' supports only the second possibility.

That Dante's glance has transformed her may further suggest that the song which she sings is, in some sense, of his composition also, as was the tempting song sung by Casella (Purg. II.112), the second canzone included in Dante's Convivio, addressed to the donna gentile (and not to Beatrice).

For the view that the femmina balba reflects not only the potentially various flesh-and-blood ladies of Dante's sexual transgressions but also the donna gentile of Convivio, see Hollander (Allegory in Dante's “Commedia” [Princeton: Princeton University Press, 1969], pp. 136-44, 162-63). It seems more likely that Dante here means to refer to the first and carnal lady for whom he betrayed Beatrice, the lady of the last section of Vita nuova, the lady who, he later claimed, was only an allegory for the unchallengeably virtuous Lady Philosphy. Thus the main thrust of his self-correction is aimed at the straying recorded in the earlier work; but, naturally enough, it would also hold in contempt that later allegorized lady as well, also presented as an 'enemy' of Beatrice in the first three treatises of Convivio. Both move the lover from affection for his true beloved in service of one far less worthy.

Joseph Mazzeo (Medieval Cultural Tradition in Dante's “Comedy” [Ithaca: Cornell University Press, 1960]), pp. 205-12, had already suggested that the Sirens are to be considered as being related to philosophical misadventure on Dante's part. His argument is not accepted by Mazzoni (“Purgatorio Canto XXXI,” in Lectura Dantis Scaligera [Florence: Le Monnier, 1965]), p. 66n., who believes that the context allows us to understand that only a carnal concupiscence is the object of scrutiny here. However, the way the mind loves its 'lady' may have seemed to Dante an even more important form of affective behavior than sexuality.

For the inverse relation between the Siren's way of naming herself 'io son, io son dolce serena' (I am, I am the sweet siren) and Beatrice's: 'ben son, ben son Beatrice' (I am, indeed I am Beatrice [Purg. XXX.73]) and their possible connection to the tetragrammaton, the two sets of two Hebrew letters (ia ia) that make up the ninth of the names of God assembled by Isidore of Seville (Etymol. VII.i.16), see Hollander (Allegory in Dante's “Commedia” [Princeton: Princeton University Press, 1969]), p. 144n.

For the biblical references to sirens (Isaiah 13:21; Job 30:29-30; Isaiah 34:13; Jeremiah 50:39, Micah 1:8), see Lino Pertile (“Ulisse, Guido e le sirene,” Studi Danteschi 65 [2000]), p. 107n. And, for patristic references to them, see pp. 107-14. And, for the similarity (and possible relationship) between Dante's portrayal of dreams and Leonardo's, see Max Marmor, “'...par che sia mio destino...': The Prophetic Dream in Leonardo and in Dante,” Raccolta Vinciana 31 (2005): 146-80.

The phrasing that expresses the Siren's power over men may put us in mind of the condition of Dante in the opening verses of the poem, when he, nel mezzo del cammin, was off his course and resembled a sailor who had nearly drowned. Does he now see himself as having been seduced by a 'siren'? Insofar as the she-wolf represents the sins of Incontinence, and thus, for Dante, lust (see the note to Inf. I.32-54), the essential reason for his having lost the true way would now seem to be predominantly related to his sexual affections.

A tormented tercet: what does vago mean? to whom or what does it refer? who is the serena who claims so to have held Ulysses' attention? As Barbi (Problemi di critica dantesca [Florence: Sansoni, 1934]), p. 228, maintained, in this poem the adjective vago always (it is used 13 other times) means bramoso (desirous of) and is, as here, used with the genitive (cf. Purg. XXVIII.1). (See Inf. VIII.52, Inf. XXIX.3, Purg. III.13, Purg. X.104, Purg. XV.84, Purg. XXIV.40, Purg. XXVII.106, Purg. XXVIII.1, Purg. XXXII.135, Par. III.34, Par. XII.14, Par. XXIII.13, Par. XXXI.33.) Thus, while the commentators are divided roughly evenly, with more early ones opting for vago as modifying cammin (and meaning 'wandering, indirect'), and more modern ones, beginning with Torraca (comm. to these lines), believing that it modifies Ulysses (and means 'eager'), one is more likely to be convinced, as was Giuseppina Mezzadroli (“Dante, Boezio e le sirene,” Lingua e stile 25 [1990]), p. 29, that the context and Dante's general practice allow us to resolve the first two questions as did Barbi (this woman drew Ulysses from the journey he was so eager to pursue). (For a lengthy treatment of the word, Karlheinz Stierle [“Paradiso: Canto III,” L'Alighieri 19 (2002)], p 75n., refers the reader to A. Noyer-Weidner, “Standortbestimmung zum Gebrauch eines 'echt italienischen Wortes' (vago) bei Dante und Petrarca,” in his Umgang mit Texten, I [Stuttgart: Franz Steiner, 1986], pp. 169-92.) But what of this 'siren' who so beguiled Ulysses? Commentators have at times forgotten that Dante did not know Homer's account (Odyssey XII.39-200) of Ulysses' escape from the Sirens' seductive wiles. (One of them, Benvenuto [comm. to vv. 16-24] , perhaps flaunting his newly gained acquaintance with the newly translated Homeric epics, suggests that Calypso is the siren referred to here; Padoan [“La 'serena' dell'Ulisse dantesco (Purg. XIX 19-24),” in his Il pio Enea e l'empio Ulisse: Tradizione classica e intendimento medievale in Dante (Ravenna: Longo, 1977 [1960]), pp. 200-4], encouraged by Benvenuto, also hears this ill-heard echo.) We should probably understand, following Edward Moore (Studies in Dante, First Series: Scripture and Classical Authors in Dante [Oxford: Clarendon, 1969 (1896)]) that, from Cicero's De finibus V.xviii.48-49, Dante decided that Ulysses had indeed been tempted by the Sirens. (In this vein see Hollander [“Purgatorio XIX: Dante's Siren/Harpy,” in Dante, Petrarch, Boccaccio: Studies in Honor of Charles S. Singleton, ed. Aldo S. Bernardo and Anthony L. Pellegrini (Binghamton, NY: Medieval & Renaissance Texts and Studies, 1983), pp. 80-83]; Giuseppina Mezzadroli [“Dante, Boezio e le sirene,” Lingua e stile 25 (1990), pp. 30-32]; and Lloyd Howard [Formulas of Repetition in Dante's “Commedia” (Montreal: McGill-Queen's University Press, 2001), pp. 139-41]). In any case, that is how he has the Siren portray Ulysses, and he offers no textual support for any other view. He also probably had in mind, as Moore indeed indicated, the passage near the opening of Boethius's Consolation of Philosophy (I.i[pr]) that has Lady Philosophy ordering the 'sweet Sirens' to be gone from the presence of Boethius and leave him to the care of her better 'Muses' (perhaps first noted by Tommaseo [comm. to vv. 19-21]). Hollander (“Purgatorio XIX: Dante's Siren/Harpy,” in Dante, Petrarch, Boccaccio: Studies in Honor of Charles S. Singleton, ed. Aldo S. Bernardo and Anthony L. Pellegrini [Binghamton, NY: Medieval & Renaissance Texts and Studies, 1983]), p. 82, also notes the closeness of Dante's description of the Sirens to that found in Isidore of Seville (Etymol. XI.iii.30-31). Pertile (“Ulisse, Guido e le sirene,” Studi Danteschi 65 [2000], pp. 101-18) attempting to demonstrate that the Christian tradition of Ulysses (which sees him positively, as the man who could withstand the Sirens, i.e., sensual and intellectual temptation) stands behind Dante's treatment of the hero, somehow fails to deal with the fact that in Dante Ulysses is portrayed precisely as failing to resist such temptation. For that matter, in Beatrice's later opinion, Dante himself is seen in exactly the same light, namely as yielding to the temptation of the Sirens when he withdrew his attention from her in order to fall under the spell of another lady or ladies (Purg. XXXI.43-48). Picone (“Canto XIX,” in Lectura Dantis Turicensis: Purgatorio, ed. Georges Güntert and Michelangelo Picone [Florence: Cesati, 2001], pp. 287-306) has done a lectura of the canto that is almost entirely dedicated to the Ulyssean motifs found in it. For an attempt to reassess in a balanced way Dante's treatment of Ulysses, see Patrick Boyde (Human Vices and Human Worth in Dante's “Comedy” [Cambridge: Cambridge University Press, 2000], Chapter 10, “The worth and vices of Ulysses: a case-study,” pp. 231-72). It is worth noting that at least one early commentator, the anonymous author of the chiose sincrone in a manuscript from Montecassino (the Codice Cassinese, general ote to this canto), also believes, as Dante appears to, that the Sirens were found at sea. He claims as much on the authority of the figmenta poetarum, the fables of the poets. But see Hollander (“Purgatorio XIX: Dante's Siren/Harpy,” in Dante, Petrarch, Boccaccio: Studies in Honor of Charles S. Singleton, ed. Aldo S. Bernardo and Anthony L. Pellegrini [Binghamton, NY: Medieval & Renaissance Texts and Studies, 1983]), pp. 82-83, for a possible source in a poet quite familiar to Dante, Ovid (Metam. V.552-563), who gives the Sirens wings so that they may fly over the sea in their search for Proserpina.

A number of commentators have believed, from the earliest days onward, that Dante confusedly considered Circe (to whom he has the Greek adventurer advert as his seductress in Inf. XXVI.91-92) one of the Sirens. He clearly knows better than that. On the other hand, he also may be building his own version of Ulysses' biography, as he has already done in inventing the last voyage in the earlier passage, and now includes an interlude that adds another to the single 'canonical' and Circean divagation in yet another example of Ulyssean turning.

Attempting to identify this lady, Fedele Romani (Il canto XIX del “Purgatorio” [(“Lectura Dantis Orsanmichele”) Florence: Sansoni, 1902]), pp. 15-18, one hundred years ago opted for Beatrice, but has had few followers. Gioacchino Paparelli (“Purgatorio Canto XIX,” in Lectura Dantis Scaligera [Florence: Le Monnier, 1967]), p. 717, reviewing the question, adds a piece of evidence for those who think the lady is Beatrice: as Lucy, introduced along with Beatrice in Inferno II, is central to the first dream in Purgatorio IX, so Beatrice, introduced to the poem in that same passage, operates in the second dream as its morally central character. However, Paparelli continues, the fact that Dante does not recognize her simply voids this interpretation. Perhaps Paparelli (and many another who follows this line of argument) does not take into sufficient account the fact that this scene is presented as a dream, i.e., a 'text' that requires interpretation in order to be understood. In this formulation it is essential that Beatrice not be presented as herself, but as being discoverable from her qualities. If one wanted to cite a single passage that reveals the convoluted and implausible arguments of those who seek another solution for the problem, Paparelli's discussion (pp. 714-29), which ends up opting for 'Justice' as this lady's identity, would serve admirably.

Among more recent proponents of Beatrice's candidacy see Poletto (comm. to vv. 25-27), who clearly prefers her as best fitting what happens in the poem, while ultimately not being quite certain; Carroll (comm. to vv. 25-33), although he is still more tentative than Poletto and, indeed, eventually prefers an unlikely choice, Matelda (whom we will meet in Purg. XXVIII); Vandelli (comm. to vv. 26-33); and Giacalone (comm. to vv. 25-27), who offers the fullest and best defense of Beatrice as being the lady in question. And see Hollander (Allegory in Dante's “Commedia” [Princeton: Princeton University Press, 1969]), pp. 141-44, who strongly supports this view. Giuseppe Mazzotta (Dante's Vision and the Circle of Knowledge [Princeton: Princeton University Press, 1993], p. 145) and (“Il sogno della Sirena [Purgatorio XIX],” in Il sogno raccontato: Atti del convegno internazionale di Rende (12-14 novembre 1992), a cura di N. Merola and C. Verbaro [Vibo Valentia: Monteleone, 1995], p. 127) also believes that this lady in Dante's dream is Beatrice. And see Picone (“Canto XIX,” in Lectura Dantis Turicensis: Purgatorio, ed. Georges Güntert and Michelangelo Picone [Florence: Cesati, 2001], pp. 287-306) for a similar view.

If this is Beatrice, then it is hardly surprising that she would recognize Virgil, in the dreamer's estimation, since he knows from what he was told in Inferno II.53 that Beatrice came to Virgil in Limbo. And as for the identity of the lady here, characterized as being 'onesta' (virtuous), it is probably worth remembering that Beatrice is later compared to a 'donna onesta' (chaste lady) in Paradiso XXVII.31.

What is the subject of the verb prendea (seized)? Some have argued that it is the holy lady. A sense of grammatical structure indicates, instead, that it is Virgil, subject of the previous verb (venìa [came forward]) that is in parallel with it. Further, if the lady indeed represents Beatrice, it would be highly unlikely that she would do the dirty work herself. Just as she, in Inferno II, called on Virgil to make Dante aware of the foulness of the sins punished in hell, so now she stands to one side while Virgil reveals the noxious nature of her rival, the femmina balba.

The stench that arises from the naked belly of the femmina has, according to Hollander (“Purgatorio XIX: Dante's Siren/Harpy,” in Dante, Petrarch, Boccaccio: Studies in Honor of Charles S. Singleton, ed. Aldo S. Bernardo and Anthony L. Pellegrini [Binghamton, NY: Medieval & Renaissance Texts and Studies, 1983]), pp. 84-86, a familiar source, not one that must be sought in out-of-the-way medieval treatises (see the note to Purg. XIX.7-9), but in Virgil's description of the Harpies in Aeneid III.216-218, as was suggested by a student at Princeton, Carl Frankel '70, in 1968: 'virginei volucrum vultus, foedissima ventris / proluvies, uncaeque manus, et pallida semper / ora fame' (maidenly of countenance, yet winged; most foul the discharge of their bellies; their hands taloned; their faces always pale with hunger). The particular similarity of the stinking bellies of Dante's Siren and Virgil's Harpies is surely striking. A further similarity lies in the purpose both creatures have in the works that contain them, which is to draw the hero away from his task, whether from proceeding to Italy or from pursuing Beatrice to a destination in Christ. In this sense both are counselors of despair. In Dante's case, it is his duty to confess that he himself had created, out of what should have been repulsive, what he came to worship; out of a Harpy he had formed a Siren. Unlike Ulysses' Siren, Dante's femme fatale is not even beautiful to begin with. It is no wonder that she will be brought back into play in his worst moment of guilt in the entire poem when he is censured by Beatrice in Purgatorio XXXI.43-48, warning him not to be lured by the 'Sirens' ever again.

In Dante's 'Surgi e vieni' (Arise and come), Mattalia (comm. to verse 35) seems to have been the first commentator (and few have subsequently joined him) to hear what is clearly a biblical echo, even if his hearing is a little dull. He cites Matthew 9:5-6, where Jesus urges the paralyzed man to walk; perhaps more applicable is Matthew 26:36-46, where Jesus three times leaves his disciples in Gethsemane in order to pray in a place apart and three times comes back to find them sleeping, finally arousing them with 'Surgite, eamus' (Rise, let us be going), for His betrayal (by Judas) is at hand. The rhythm of those three disheartening visits to those who should have been awake is preserved in Dante's 'Three times at least I've called you,' as was suggested in 1969 by two undergraduate students at Princeton, John Adams and Christopher McElroy. Lost in his dream, Dante is like the disciples who sleep while their Lord suffers alone.

Dante has slept late, unsurprisingly, given his late-night activities on the Terrace of Sloth (Purg. XVIII.76-78), and now finds the sun, at his back, risen above the horizon. For his posture Bosco/Reggio (comm. to vv. 40-42) suggest the similitude of a broken supporting pier of a bridge (and thus only half an arch) rather than that of one of the two halves of a pointed arch, as proposed by others, since pointed arches were rarely found as features of medieval bridges.

Whatever the architectural model, the protagonist is obviously concerned about the meaning of his dream.

The Angel of Zeal's words ('Come, here is the passage') may not be like any heard here on earth, but they do resemble those spoken by Beatrice when she was described by Virgil as being 'soave e piana' (gentle and clear) in her speech (Inf. II.56), as Poletto (comm. to vv. 40-45) suggested.

As for the word marca, commentators point out that it is of feudal (Carolingian) origin, and referred to the borderline that a lord established for his terrain; in Dante's use of the extended sense of the word (which had come simply to mean 'region') there may reside some sense of the original term: our earth is like a small fiefdom when compared with God's kingdom.

Perhaps this only reference to a particular bird as being similar to a welcoming angel includes swan-like features because of the avian characteristics of the Siren that it intrinsically counters. The soft wings of the angel also stand in sharp contrast to the rocky sides of the narrow passage upward, framing these, as it were. Some commentators argue that this angel opens its wings in order to point the way to the passage; since its voice has already done so, this seems an unnecessary precision.

The nine verses devoted to the presence of the angel here represent the briefest scene yet devoted to the interplay between angel and mortal (but see the note to Purg. XXII.1-6). The Beatitude referred to, Matthew 5:4 (5:5 in the Vulgate), 'Blessed are they that mourn, for they shall be comforted,' has caused some to wonder what specific relevance these words have to those formerly guilty of Sloth. Federigo Tollemache, “beatitudini evangeliche” (ED.1970.1), p. 540b, explains that, given Thomas Aquinas's definition (ST II.ii.35.2) of accidia as tristitia de spirituali bono (dejection over one's spiritual health), the phrase 'qui lugent' (those who mourn) is relevant. For the program of the Beatitudes in this cantica, see the note to Purgatorio XII.110.

The exchange clearly reflects that between Virgil and Dante in Purgatorio XV.120-126. Once again Virgil begins by asking Dante 'che hai?' (what's wrong?), not at first understanding his charge's removal from present reality. Once again Dante insists on his other-mindedness. In the first instance Virgil quickly understood that Dante was having a visionary experience; now he becomes aware that Dante has been having a dream of what his guide's words had prepared him for, coming to grips with the 'good that fails to make men happy' (Purg. XVII.133). In the first case, Virgil seems to have insisted that he knew what he did not know; here he seems to have intuited correctly what his charge was dreaming. Scartazzini (comm. to verse 58) points out that some texts have question marks following the words 'si piagne' (verse 59) and 'si slega' (verse 60), making what Petrocchi's text records as observations into questions. Were we to know that such was indeed the punctuation used by Dante, the hypothesis outlined above would be supported, i.e., Virgil did not 'see' Dante's dream, but divined it from the situational context.

While some of the early commentators are sure that here Virgil only refers to the sin of Avarice, repented on the next terrace, beginning with Benvenuto da Imola (comm. to vv. 58-63) others have clearly understood that Virgil is referring to all of the three sins of incontinence he had described in Purgatorio XVII.133-139. For a full review of the interpretive problem represented by the antica strega see Gabriele Muresu (“Il richiamo dell'antica strega [Purgatorio, XIX],” La Rassegna della letteratura italiana 100 [1996], pp. 5-38).

Virgil's formulation causes a problem for those who would argue that the holy lady is Beatrice, since it generalizes the nature of the lady who opposes the femmina balba and makes Dante's dream applicable to all sinners, no others of whom, we may assume, are lovers of Beatrice. For this reason Giovanni Parenti's understanding (“Ercole al bivio e il sogno della femmina balba,” in Operosa parva per Gianni Antonini, ed. D. De Robertis & F. Gavazzeni [Verona: Valdonega, 1996]), pp. 62-63 (resuscitating Torraca's opinion [comm. to vv. 25-33]), that the holy lady equates with Charity, seems the most adequate solution, good love that operates against the forces of 'the good that fails to make men happy' (Purg. XVII.133). Charity may well be the general meaning of the lady in the dream; for Dante, however, that theological virtue is the core of the meaning of Beatrice.

Some commentators take Virgil's exhortation metaphorically, i.e., 'crush the things of this world under your feet,' but it seems to be literal enough, urging Dante to cease his dreamy self-absorption and push up (they are on the slope between terraces) and forward with his feet in order to continue his journey.

Virgil's metaphoric expression and the poet's following simile return to falconry (see the note to Inf. XVII.127-136), now in as central an image of the basic movement of the entire poem as may be found. Mortals look down, consumed by their own concerns while God, the falconer, wheels his lure (the celestial heavens) around his 'head,' thus drawing us back to Him. Dante had been looking at the earth (verse 52) and Virgil urges him to push off against it in order to move on (verse 61); in the simile the falcon, too, looks down, perhaps to see if he is still bound to the falconer's wrist now that his hood has been removed (or merely in his habitual attitude, his head inclined downward, resting on his breast). Both bird and Dante, urged on, look up and travel upward, in Dante's case by climbing through the passageway in the rock so that he may resume his circling of the mountain on his approach to God.

The image of the star-filled heavens as God's lures for us, his falcons, is central to the progress of the poem that concludes each of its cantiche with the word stelle (stars).

Dante's arrival on the fifth terrace, that of Avarice and Prodigality, is immediately greeted by the sight of those who are purging themselves there, prostrate on the earth. This terrace is unique in that it is a stage for three increasingly lengthy conversations, first with a pope (Adrian V) in this canto, with a kingly figure (Hugh Capet) in the next, and finally with a poet (Statius) in XXI and XXII, a sample of callings that reflects Dante's most pressing concerns: Church, empire, and letters.

The penitents' cries, muffled because they lie facedown on the floor of the terrace and are uttering them through painful sighs, are 'my soul cleaves to the earth' (Psalm 118 [119]:25). The Ottimo (comm. to vv. 70-74) connects this confessional outpouring with Virgil's earlier remark to Dante ('Press your heels / into the ground' [verse 61]), thus suggesting that the avaricious repent their longing for the things of the earth, exactly what Virgil is urging Dante to do.

The terms in which Virgil puts his request may remind us that his own condition in Limbo lacks precisely what these penitents enjoy: hope in the justice of God for eventual salvation. Virgil and the other inhabitants of Limbo long for that justice, but without any possible hope of achieving it (Inf. IV.42).

As Singleton (comm. to this verse) observes, this is the first time we learn that some penitents do not have to spend penitential time on every terrace, since the nameless speaker (we will learn that he was a pope at verse 99) assumes, from Virgil's request for help, that both of these newcomers are saved souls exempt from the sin of avarice (or of prodigality) who are ascending to a destination higher up the mountain without having to stay here.

Dante cannot make out the identity of the speaker, but is able to individuate the source of the words he has just heard; he seeks Virgil's permission to question him.

The cenno (sign) made by Virgil in assent to Dante's request is apparently either a facial gesture (a smile?) or an ocular one (a kindly glance?). He will make a responsive and welcoming gesture to the shade of Statius in Purgatorio XXI.15, the nature of which has caused more puzzlement than this one.

Dante's circumlocution, 'that without which there is no return to God,' refers to the satisfaction each penitent must offer to God, showing that he or she is finally pure of the traces of the defiling sin purged on each terrace.

Dante's three questions will be the basis for this penitent's speech, which will fill most of the rest of the canto, vv. 97-126 and 142-145. He wants to know the identity of his interlocutor, the nature of the sin reflected in his prone posture, and, as is customary, what service he (as living soul) can perform back on earth for him.

The speaker, in good papal Latin ('Know that I was a successor of Peter'), informs Dante that he was once a pope, not boastfully, but humbly and ashamedly, as though to say 'I, of all people, who should have known better.' Benvenuto (comm. to vv. 97-99) makes a similar point; it is as though he said he was a successor of Peter 'sed non pauper ut Petrus' (but not a poor man, as was Peter).

Geographical indications (two towns on the Ligurian coast and the stream taken by members of the Fieschi family for their title: they are 'counts of Lavagna') leave no doubt as to the identity of the pope who speaks: 'Adrian V (Ottobuono de' Fieschi of Genoa), elected Pope at Rome, in succession to Innocent V, July 11, 1276; died at Viterbo on Aug. 16 following, before he had been crowned' (Toynbee, “Adriano [2]” [Concise Dante Dictionary]); thus Dante's 'a month and little more' to indicate Adrian's term of office. Longfellow (comm. to verse 99) reports the following papal remark: 'When his kindred came to congratulate him on his election, he said, 'Would that ye came to a Cardinal in good health, and not to a dying Pope.''

See Purgatorio XVI.127-129 for Dante's earlier view of the papacy's descent into the 'mud' of wrongful activity.

Bosco (comm. to vv. 127-138) draws attention to the parallels between the previous nineteenth canto and this one, both deeply involved with the papacy in both similar and opposed spirits. At least here we understand what the papacy might be if the pope were an Adrian V. It is perhaps by design that the first saved pope whom we meet in the poem (there will be more [see the note to Inf. VII.38-39]) should be distinguished by having died shortly after his election and thus without having served 'officially' at all.

Adrian's remarks have caused a certain puzzlement, since historical records give no sense of his involvement in avaricious behavior (nor, consequently, of his turning from that sin only once he was elected pope); as Scartazzini (comm. to verse 99) observed, such notice derives only from this passage in Dante's poem and from his perhaps too gullible commentators. Bosco (“Particolari danteschi,” Annali della Reale Scuola Normale di Pisa, Lettere, storia e filosofia 11 [1942]), pp. 136-43, followed by Sapegno (comm. to verse 99), argues that Dante thought that what he had read in John of Salisbury's Policraticus (VIII.xxiii.814) or, more likely, since Petrarch later also made the same mistake, in some (unknown) later source that created this error, about the twelfth-century pope Adrian IV, concerned instead his thirteenth-century namesake. Renucci, cited by Giacalone (comm. to vv. 100-102), has argued that Dante was well acquainted with John of Salisbury's text and deliberately conflated the two Adrians for reasons of his own. Since Dante almost certainly had recourse to John's text, it is difficult to counter Renucci's view, even if it leaves one wondering why Dante should have wanted to rewrite the life of Adrian V in this way. For the Policraticus as source of several of Dante's exemplary figures see E.R. Curtius (European Literature and the Latin Middle Ages, tr. W.R. Trask [New York: Harper & Row, 1963 (1948)]), pp. 364-65; see also André Pézard (“Du Policraticus à la Divine Comédie,” Romania 70 [1948]), esp. pp. 163-91 (for John of Salisbury's Adrian), and Renucci (Une source de Dante, le “Policraticus” de Jean de Salisbury [Paris: n.p., 1951]). For the texts in the Policraticus that are pertinent here, see Singleton (comm. to vv. 99-114).

Conversion here signifies any turning to God. Even confirmed Christians are likely to experience a continuing need for 'conversion.' See Singleton (Journey to Beatrice [Cambridge: Harvard University Press, 1967 (1958)]), pp. 39-56.

Adrian here answers Dante's second question, why these souls are in the posture Dante sees them in, by explaining their contrapasso: since they sought the things of earth energetically, they now are facedown on that earth and restrained, immobile, upon it.

Dante's reverent kneeling before Adrian is apparent from his voice, which sounds louder because his face is now closer to the recumbent pope's body. The interruption in Adrian's answers allows this little exchange that offers a lesson in fellowship that trumps Dante's gesture of respect.

Adrian earns one of Dante's relatively few uses of the honorific 'voi' (see the note to Inf. X.49-51). In the first cantica only Farinata, Cavalcante, and Brunetto have this honor bestowed upon them. Adrian, the fourth of seven to share the honor, will be joined only by Guinizzelli (Purg. XXVI.112) and Beatrice (first at Purg. XXXI.36) in this cantica, and then by Cacciaguida (first at Par. XVI.10) in the next.

Adrian's response is so urgent that he only gets to his fraternal salute, nearly always found, elsewhere in the poem, at the beginning of direct address and never at its end, last. His 'frate,' rhyming with 'dignitate,' is the answer to the hierarchy underlined by that second term and by Dante's kneeling. In God's kingdom there is no specialness, only brotherhood of the equally special.

For a survey of the presence of the address 'frate' in the Commedia, see the note to Purgatorio IV.127. There are twenty occurrences of the word, when used to address another, in the poem, the first in Inferno XXIII.109, where Dante addresses the Jovial Friars rather nastily, the last in Paradiso XXII.61, where Benedict addresses Dante, who is saluted a total of seventeen times in this way. Only Forese Donati (twice) and Beatrice (four times) are permitted to use this form of address more than once. In the last two cantiche the only addressee other than Dante is Statius (Purg. XXI.131), in Virgil's humble and warm disclaimer of a respect similar, on Statius's part, to that displayed here by Dante for Adrian.

From the time of Benvenuto (comm. to vv. 133-138) it has been understood that this scene clearly replays a similar scene in the Bible: Revelation 19:9-10, where the angel addresses John, commanding him to write of the blessedness of those who share the marriage supper of the Lamb. When John falls before the feet of the angel to worship him, the angel says: 'You must not do that; I am your fellow-servant (conservus).' Dante's hapax, conservo, surely cements the relationship between the two texts.

The next biblical reference is to Matthew 22:23-30, 'neque nubent' (nor do they marry), a passage in which Christ deals with the sardonic and hairsplitting Sadducees, who do not believe in resurrection and who wish, cynically, to know which of six brothers, who had in turn married an eldest brother's widow, would be her resurrected husband. Jesus answers them by saying that after the Resurrection there will be no marrying in Heaven, where all will share, one might add, in the marriage supper of the Lamb as equals, where all are married to all and to none. Adrian's insistence on the lack of hierarchical distinction is his version of Jesus' saying.

'That of which you spoke' refers to Dante's previous understanding (verse 92) of Adrian's desire to complete his penance and thus achieve purification.

Answering the third element in Dante's question, regarding what Dante might do for him, Adrian can supply only the name of a niece, Alagia, who might pray for him, thus suggesting that his avaricious former life had much in the way of familial bad company among all the rest of his relatives.

Alagia was married to Moroello Malaspina, with whom Dante was on friendly terms, and thus his good words about her probably reflect a positive impression gained from personal knowledge and may also serve to express gratitude for the Malaspina family's hospitality in Lunigiana in the early years of Dante's exile.