Virgil's brief admonition will be his only utterance in this canto, which, like the twenty-fourth, is heavily involved in questions regarding vernacular poetry, thus necessarily marginalizing Virgil's presence.

Dante, according to the calculations of Trucchi (comm. to vv. 1-6), having begun his climb of the mountain at due east at the antipodes, has now reached a point some 25^o short of due west. As for the time, it has apparently taken a bit more than two hours to climb between the two terraces (see the note to Purg. XXV.1-3).

A shadow cast upon flames does indeed make them glow a darker color. (See Richard Abrams [“Illicit Pleasures: Dante among the Sensualists [Purgatorio XXVI],” Modern Language Notes 100 [1985], pp. 1-41] for a consideration of this image as emblematic of the themes of this canto.) Once again Dante's presence in the body serves as a provocation to a group of penitents. Our translation takes into account and follows the argument of Picone (“Canto XXVI,” in Lectura Dantis Turicensis: Purgatorio, ed. Georges Güntert and Michelangelo Picone [Florence: Cesati, 2001]), pp. 407-8, based in the earlier observation of Mario Marti (“Il XXVI del Purgatorio come omaggio d'arte: Guinizzelli e Daniello nel cammino poetico di Dante,” in his Studi su Dante [Galatina: Congedo, 1984 (1978)]), p. 156, that the 'sign' created by Dante's shadow is not presented as a minor one, but as something quite extraordinary, as the notice it receives from many of the penitents, even while they are intent on their purgation, would indicate. And thus the adverb pur (which can have the very different meanings of 'only' or of 'indeed') here does not take away from the importance of the sign but stresses it, i.e., does not mean 'only,' as many commentators believe (but see, among others, Lombardi [comm. to the verses], who reads pur as having that second and intensifying sense, as do Marti and Picone). Among modern commentators who deal closely with these lines, only Chimenz, Giacalone, and Bosco/Reggio support this reading. Chimenz, for example, argues that the excited reaction of the penitents (comm. to vv. 16-24) does not seem a response to a 'small' stimulus.

At Inferno VI.36 the poet speaks of the shades of the gluttons as 'lor vanità che par persona' (their emptiness, which seems real bodies). Here the penitents remark at Dante's condition; his body does not seem to be, like theirs, 'fictitious,' airy (about which state we have just heard Statius's lengthy disquisition).

Their impulse, reversing Dante's, is not to leave the searing flames. Again we sense the eagerness of penitents to undergo their purgation. See the remark of Forese Donati, 'I speak of pain but should say solace' (Purg. XXIII.72).

The speaker is Guido Guinizzelli (see the note to Purg. XXVI.92), the previous Italian poet to whom, we will learn, Dante now feels the greatest allegiance. While Guido may or may not be correct in thinking Dante is following the two other souls out of reverence rather than from lack of zeal (or fear of the flames that he must enter), what he cannot know (and never discovers) is that these are the great shades of Virgil and Statius. We, however, do know that and realize that this scene is the last in a series that began in the cantos devoted to Statius (Purg. XX-XXII), a program devoted to an exploration of the nature of Dante's poetics in relation to those of other poets: Statius, Bonagiunta (who throws in Giacomo da Lentini and Guittone d'Arezzo for good measure), and, in this canto, Guinizzelli (adding Giraut de Bornelh and poor Guittone again) and Arnaut Daniel. No other part of the poem is as extensively or as richly concerned with the purposes of poetry.

There is some debate among the commentators about the metaphoric or literal nature of the thirst to which Guido refers: 'answer me, since I burn with thirst and fire' (verse 18). While hunger and thirst were the natural penalties undergone by the penitent gluttons, there is no such contrapasso here. Further, Guido's description of his co-sufferers' 'thirst' (Purg. XXVI.20-21) is surely metaphoric, thus suggesting that this first mention of thirst is also. For this view, among others, see Lino Pertile (“Purgatorio XXVI,” in Dante's “Divine Comedy,” Introductory Readings II: “Purgatorio,” ed. Tibor Wlassics [Lectura Dantis (virginiana), 12, supplement, Charlottesville: University of Virginia, 1993]), p. 381. On the other hand, the fire of which he speaks is literal enough, as Dante will find out in the next canto (Purg. XXVII.49-51), and his aerial body surely allows him to feel such sensations.

Benvenuto's gloss (comm. to vv. 16-18) to this verse is amusing. After identifying 'the others' as Virgil and Statius, he comments on the phrase non per esser più tardo (which he apparently took to mean, 'not for having come later than they') as follows: 'as though he were saying, “not because you are a modern poet and they ancient,” for the passage of time does not make poems improve as it indeed does improve wines.'

This remarkable simile, a rare medieval manifestation of a moment of fraternal affection between heterosexuals and homosexuals, is striking. The passage probably reflects Paul's admonition in Romans 16:16: 'Greet one another with a holy kiss,' as was suggested by Scartazzini (comm. to verse 32). For the ants, see Virgil's memorable simile in Aeneid IV.402-407. Aeneas's men, preparing their ships for departure from Carthage, are described as follows: 'Just so do ants, when winter's on their mind, pillage great stores of grain and fill their houses to the beams. Over the fields moves a black column, carrying their spoil through the grass along their narrow path; some heave the huge seeds upon their shoulders, some shape up the columns, rebuking their delay. All the path fairly shines with labor.'

Those who have argued that the sin punished in Inferno XV and XVI is not homosexuality (see Inf. XV.13-21) are hard pressed to account for the obvious reference of the word 'Sodom' repeated here, used first in Inferno XI.50 to refer to the sinners on the barren sands of homosexuality. The early commentators have no doubt whatsoever about all this. See, for example, John of Serravalle (comm. to vv. 37-42): 'And from this city, Sodom, the sin against nature took its name, just as simony did from Simon [Magus].' Each group cries out the appropriate exemplar(s) of its sin, the first the homosexuals, calling out the names of these two 'cities of the plain' (see Genesis 19:1-28), the locus classicus for homosexual lust, where the men of Sodom ask Lot to give them for their sexual pleasure the two angels who have come to Lot and whom they take for men (Genesis 19:4-11).

What has long been problematic is the fact that, in Inferno, we find the heterosexual lustful punished in the realm of Incontinence, while those guilty of homoerotic behavior are in that of Violence (against nature, in their case). That these two groups are now purging themselves on the same terrace may be the result of the changed ground rules for the sins of the two cantiche more than of any supposed change of heart on Dante's part, that is, Dante no longer has the option of fitting these two different bands onto an Aristotelian/Ciceronian grid, but must associate them with one of the seven capital vices, which leaves him little choice. Nonetheless, no matter what his intentions, the effect is to make the reader feel that the poet has now softened his views. The notion of Pertile (“Purgatorio XXVI,” in Dante's “Divine Comedy,” Introductory Readings II: “Purgatorio,” ed. Tibor Wlassics [Lectura Dantis (virginiana), 12, supplement, Charlottesville: University of Virginia, 1993]), p. 388, that the impulse to love is the same in hetero- and homosexuals and that, since in purgatory only the impulse (or predisposition) toward sin is purged, there is no longer any need to distinguish between them is interesting but difficult to accept. In Inferno homosexuality is treated as a sin of hardened will, and one would be hard pressed to show that this does not make the 'impulse' that drives it different from that behind the sins of Incontinence.

The heterosexuals call out the name of Pasiphaë (see Inf. XII.12-13 and note), the wife of Minos, king of Crete, who had Deadalus build for her a wooden frame covered with a cowhide so that she could be mounted by a bull. The child of this union was the Minotaur. Pasiphaë here is clearly meant to represent the animalistic nature of unrestrained lust, not some sort of sodomy.

The hypothetical nature of this simile is underlined by Dante's use of the subjunctive mood for its main verb (volasser). Cranes do not and would not migrate simultaneously in two different directions, north to the Riphaean mountains and also south to the sands of the (Libyan?) desert. Dante has developed the passage on the model, perhaps, of some of Lucan's similes concerning cranes (see the note to Purg. XXIV.64-74), but the resemblance does not seem more than casual, if the closest would seem to be that found at Pharsalia VII.832-834. For a meditation upon Dante's cranes here and in Inferno V.46-49 and Purgatorio XXIV.64-69 see Guglielmo Gorni (“'Gru' di Dante: Lettura di Purgatorio XXVI,” Rassegna europea di letteratura italiana 3 [1994]), pp. 23-30.

The 'former song' of the penitents is the hymn Summae Deus clementiae punctuated by the words of Mary at the Annunciation and those regarding Diana's chaste anger at Callisto (Purg. XXV.121-132). Each sub-group also then sends up 'the cry that most befits' it, that is, either 'Sodom and Gomorrah' or 'Pasiphaë.'

Dante at last responds to the request of the penitents, if only to some degree, since he does not fully identify himself, in keeping with his avoidance of doing so on other terraces. He tells them what they suspected (he is here in the flesh, not in the aerial body), and that Beatrice (not named, and decidedly not the sort of woman penitents here 'do time' for) draws him heavenward. Thus he admits to his miraculous presence among them, but gives no information that might genuinely satisfy their curiosity. In return for relatively little he asks to know the identity of those with whom he speaks and the condition of the group that has moved away from them.

His evasive behavior here allows him to avoid naming himself uselessly to those who do not know him (see his similar avoidance with Sapia [Purg. XIII.133-138], Guido del Duca [Purg. XIV.20-21], and Marco the Lombard [Purg. XVI.37-42]). Those who recognize Dante in purgatory are as follows: Casella (Purg. II.76-78), Belacqua (recognized by Dante at Purg. IV.109-123; it would seem that he knows Dante but does not, in his laconic, sardonic way, reveal that he does); Nino Visconti (Purg. VIII.46-57); Oderisi (Purg. XI.76); Forese (Purg. XXIII.40-42); and Bonagiunta (Purg. XXIV.35-36).

See Benvenuto's (comm. to vv. 61-66) paraphrase: 'as though to say, so that I may write of you souls and give your true reputation, in case there should remain in the world infamous word of your lust.' Dante's way of promising to record the penitents' presence on the road to salvation (we must remember that his charitable offer is made without knowing the identity of any of the souls whom he here addresses) so that it may draw prayers from the living and thus hasten their passage to bliss will be enlarged in Guido's final request of Dante (Purg. XXVI.130-132), where he hopes for prayers of intercession on his behalf in Heaven itself.

The language here reflects the mode of preparing a manuscript for inscription, the pencilling in of guidelines that can be erased once ink is set to vellum or paper. Dante imagines, from the vantage point of his progress through the second kingdom of the afterworld, the preparation, by his own hand, of the manuscript of the Comedy once he is back in the world. Here the preparation of the page equates with its completion, the inscription of Guido Guinizzelli in the Book of Life. See Inferno XXIX.54-57 and note; see also the less generous, but similar, offer made to Bocca degli Abati at Inferno XXXII.93. Bocca is also recorded, but in the Book of the Dead.

For a later resonance of the image of the rustic mountainman struck dumb by the sights of the city, see Boccaccio's 'meta-novella' (Decameron IV.Intro.19-20), when the normally appetitive son of Filippo Balducci, Boccaccio's version of Dante's 'montanaro... rozzo e selvatico' (the mountaineer... rough and rustic), descends with his father from their monastic mountain and sees Florence for the first time, immediately falling under the spell of the lovely women in the beautiful city. (For Boccaccio's redoing of Dante's text see Hollander [Boccaccio's Dante and the Shaping Force of Satire (Ann Arbor: University of Michigan Press, 1997)], p. 75.) However, and as William Stephany pointed out in a lecture at Dartmouth College on 31 August 1983, Dante's mountainman is a revised version of his own earlier portrait of such rude folk. In De vulgari eloquentia (II.i.6) he declaims that the illustrious vernacular is not fitting to 'mountaineers who treat of rustic matters' (montaninis rusticana tractantibus). Here, in a sort of palinode, he chides his younger self for his stylistic and intellectual snobbery, making these saved souls amazed at the presence of this writer of low vernacular texts and thus canceling the prideful assertiveness of his younger self.

Guido uses a nautical metaphor to praise Dante's on-loading of this precious cargo of knowledge, paraphrased by Benvenuto (comm. to vv. 73-75) as follows: Dante is one who 'gathers and assembles in the bark of his wit' the mountain's source of knowledge that will allow him a better chance for salvation. Dante's substitution of the verb 'to die,' where a more conventional thought would have posited 'to live,' brought the following admiring thought from Benvenuto: 'And I do not doubt but that this poet lived better and died better because he compiled this work.' Thus Benvenuto suggests that the poem, not only, as others have thought (and as Dante claimed – see Par. I.34-36), would bring at least some of its readers to salvation, but had had this result for its author himself.

Guido now addresses Dante's question (Purg. XXVI.65-66), brought on by the sight of the second group of penitents that had caught his attention (Purg. XXVI.25-27).

As part of his rather cruel treatment of Julius Caesar, who has already been put forward as a positive exemplary figure of zeal (see Purg. XVIII.101-102 and note), Dante now makes him an 'informal exemplar' (this practice is observable elsewhere only once on the mountain: see the note to Purg. XXIII.25-30) of homosexual lust (that is, he is not a part of the 'official' program of those who are presented to all penitents, but is mentioned only to Dante by Guido). This makes him perhaps the only exemplary figure in Purgatorio to have both positive and negative valences. Again we can see how complex, troubled, and unremitting Dante's response to Julius was.

Benvenuto (comm. to vv. 76-81) dutifully and ashamedly reports Caesar's one known homosexual experience but surrounds it, in his perplexity and discomfort, with a list of Julius's (at times outrageous) sexual encounters with women. In a similar mode, John of Serravalle (comm. to vv. 76-81) insists that Julius was a mere fourteen years old at the time of this misadventure. Dante's source for poor Julius's escapade in Bithynia may eventually be found in Suetonius's Life of the Caesars (ch. 49), as Daniello reports (comm. to vv. 76-78), citing the lines that were supposedly cried out against him when he returned from Gaul in triumph: 'See how Caesar triumphs, having conquered Gaul; / Nicomedes triumphs not, but he made Caesar fall.' What had happened? Apparently, when Julius was young and serving in Bithynia, the king, who admired him, got him drunk and had sexual relations with him. Other tales make Caesar a more willing accomplice to the king's desire, e.g., Suetonius, attributing the story to Cicero, cites the report that he was dressed in purple as the queen of the realm at a wild party that ended in his 'deflowering,' etc. Later Dante commentators (e.g., Oelsner [comm. to these verses]) suggest that Dante's source was a much condensed version of the events in Bithynia found in Uguccione of Pisa's entry for 'triumph' in his Magnae derivationes, which adverts to Bithynia and to Caesar's being called both 'king' and 'queen' during one of his triumphs.

The word 'hermaphrodite,' here doubtless means (and only means) 'heterosexual' (from Ovid's tale of Salmacis and the son of Hermes (Mercury) and Aphrodite (Venus), Hermaphroditus [Metam. IV.285-388], in which their two genders are eventually included in a single double-sexed human being). If 'hermaphrodite' here meant other than that, the only souls saved from the sin of Lust would have been homosexual and bisexual, that is, there would be no heterosexual penitents on the mountain. The commentary tradition yields some amusing missteps on this subject. Francesco da Buti (comm. to vv. 76-87), selecting 'hermaphrodite' (in the sense of bisexual) as the second category of the penitent lustful on the mountain, tells the tale of a person he had seen, while he was a youth, who dressed as a man but who sat at the distaff and spun wool, using the name 'Mistress Piera.' It is only with Gabriele (comm. to this verse) and Daniello (comm. to vv. 82-87) that commentators get the problem cleared up: these are the penitent heterosexual lovers. There was still so much confusion three centuries later that even Scartazzini (comm. to this verse) felt that he had to do a full review of the question. In our own day, however, the confusion has again become manifest. See Bernard Knox, reviewing W. S. Merwin's translation of Purgatorio (New York Review of Books, 21 September 2000), where the word is translated as 'the performance as both sexes,' and the corrective rejoinder by Peter D'Epiro in the same journal (15 November 2001). It should be said that it was provocative for Dante to have used the myth as he chose to. Ovid's original tale is quite startling in its sexual role reversals: Salmacis behaving like a traditional slavering male while Hermaphroditus behaves like a traditional inviting female (even performing an unintended striptease before the excited Salmacis, peeping from her hiding place in the woods). The tale seems fully intended to serve as the foundation myth of hermaphroditism. That was not enough, however, to protect it from Dante's by now unsurprisingly elastic and eclectic reading of classical material. His meaning for the word is clear: 'heterosexual'; to arrive at that unriddling, he probably foresaw, would cause his readers some exertion.

These lines themselves possess the capacity to settle a problem that some readers prefer to keep open. How can the name of Pasiphaë, who was involved not in heterosexual lust between humans, but in sodomy (sexual contact between human and beast), be used here to indicate the former? He and his companions, Guido makes plain, did not commit 'unnatural' sexual acts, but broke the laws that govern human sexual concourse, specifically those of marriage, and did so with an untamed energy that is more fit for beasts than humans, and is thus symbolized by a woman who conspired to be entered by a bull.

The large number of those who repine their former lust seems to be commensurate with those who are condemned forever to relive it; see Inferno V.67-69, where Virgil points out to Dante a vast number of identifiable sinners – and this after we have already seen huge flocks of essentially anonymous lovers (Inf. V.40-42). And these sinners here, like those among whom we find Francesca and Paolo, are also compared to cranes (Purg. XXVI.43-45; Inf. V.46-49).

The speaker finally reveals himself as Guido Guinizzelli. 'The most illustrious of the Italian poets prior to Dante, he belonged to the family of the Principi of Bologna, in which city he was born ca. 1230. In 1270 he was Podestà of Castelfranco; in 1274, when the Ghibelline Lambertazzi were expelled from Bologna, Guido with the rest of the Principi, who belonged to the same party, was forced to leave his native city; he is said to have died in exile at Verona in 1276' (Toynbee, “Guido Guinizelli” [Concise Dante Dictionary]). His most famous poem is the canzone 'Al cor gentil rempaira sempre amore' (Love always finds shelter in the noble heart). It sets out the doctrine, embraced by Dante in the fourth treatise of Convivio, that true nobility is not determined by birth but by inner virtue.

Since Guido had died ca. 1276, he has made his way from Ostia in very good time indeed, passing through ante-purgatory and the first six terraces in less than a quarter century. Such speedy passage through purgation is a feature common to all the major figures whom Dante meets on the mountain, one forced on him by his predilection for the recently dead (Statius being the only ancient of note upon the slopes of the mountain allowed a speaking part, while Hugh Capet is the oldest 'modern'). However, we have no idea how long any of the souls whom we see will be at their penance (if Statius is a model, a good long time, since he spent more than 1200 years purging himself [see the note to Purg. XXI.22-24]). We simply have no idea how long Guido must stay in the fire for his lust. Time is not over for any of these sympathetic figures, and the mysteries of penance and redemption leave such concerns unresolved. For all we know, Manfred or Belacqua may finish purgation before Guido does.

In a striking example of abbreviatio, Dante boils down a lengthy scene in Statius's Thebaid (V.499-730) to two lines. The story to which he refers runs roughly as follows. Hypsipyle was the daughter of the king of Lemnos, whom she saved from death when the women of the island determined to kill all the males on it. She was subsequently seduced and abandoned by Jason (Inf. XVIII.88-95), by whom she had twin sons, Thoas and Euneus. When the Lemnian women discover that their king is still alive, Hypsipyle flees, and after a misadventure with pirates ends up in the service of the king of Nemea, Lycurgus. While caring for Archemorus, the king's son, she was approached by the seven exiled warriors marching against Thebes in civil war. Since they were thirsty, she agreed to show them to water (the fountain of Langia – see Purg. XXII.112), but left the baby behind long enough for him to be killed by a snake. Lycurgus would have killed her but for the chance arrival of her twin sons. Once they have quelled the attempt on their mother's life, they embrace her eagerly, more eagerly than Dante will move to embrace Guido because he fears the flames.

Benvenuto (comm. to vv. 94-99) remembers the reference to Guido's having been 'driven from the nest' at Purgatorio XI.97-99, and that he had been cast forth precisely by Dante. But who are these other poets who wrote (the verb is in the past tense) better poems than Dante? While most commentators simply avoid this problem, those who deal with it tend to favor Guido Cavalcanti and Cino da Pistoia. The first precision of any value is only made by Tommaseo (comm. to these verses) pointing out that Dante here refers to style, and not language (i.e., he is not limiting the field to Italians, as some argue), as we can see by his later inclusion of Arnaut in this canto. (But Arnaut cannot have been a 'son' of Guido because he lived and wrote before the Bolognese master.) Steiner (comm. to these lines) suggests that Cino da Pistoia is one of these elect. Trucchi (comm. to vv. 94-99) insists that the distinction is purely formal, stylistic, since Dante had insisted, two cantos earlier (Purg. XXIV.52-54), that his and no other's was the true 'intelligence of love,' which alone accounts for his innovations in the content and substance of poetry; but Trucchi does not say who he thinks are indicated. Porena (comm. to vv. 61-62) considers the matter in some detail and suggests that the most probable names are those of Cavalcanti, Cino, and Lapo Gianni. Cavalcanti's name is heard in Chimenz (comm. to these verses) along with Cino's. And that pair is mentioned also by Giacalone (comm. to vv. 96-99) and by Bosco/Reggio (to this passage), who also tentatively offer the name of Lapo Gianni, as had Porena. On the whole, the commentary tradition has been extremely cautious in trying to piece together Dante's thoughts about Guinizzelli's other 'progeny.' However, Roberto Antonelli (“Cavalcanti e Dante: al di qua del Paradiso,” in Dante: da Firenze all'aldilà. Atti del terzo Seminario dantesco internazionale, ed. Michelangelo Picone [Florence: Cesati, 2001]), p. 294, insists on an implausible reading of verse 98 to make the comparative adjective miglior refer to Guinizzelli alone; but see the sensible objection raised by Del Sal in the discussion of Antonelli's paper, pp. 366-67. For the later reappearance of the adjective 'sweet,' see Purgatorio XXVI.112 and note.

It is not difficult to understand how much Dante wants his reader to understand the importance of Guido to his sense of himself as poet. No one else on the mountain is treated as respectfully as he.

His oath to Guido evidently was to swear to tell the truth (see Purg. XXVI.109).

Guido, like Bonagiunta, refers to hearing Dante's voice with the relative clause 'ch'i' odo' (that I hear). See Purgatorio XXIV.57. Both clauses draw attention to the importance that each of these predecessor poets places on listening to the living voice of this extraordinary visitor to their realm.

Guido's request allows Dante to formulate another defining moment that will use no more than the space of a difficult tercet to respond to the remarks of a poetic predecessor (see Purg. XXIV.49-51).

Beginning in these lines, words for truth and for speaking or for writing poetry wend their way through 22 lines (ver here in vv. 109, 121, 126; dir or its derivative detto [poem] in vv. 111, 112, 119, 130). Their conjunction reminds the reader of the importance of the issue of the possibility of poetic truth, given the traditional view that poets are liars. It may be helpful to know that in Dante's day a poet in the vernacular was known as a dicitore per rima, a 'speaker in rhyme.'

For the rarity of Dante's use of the honorific voi, here given to Guido, see the note to Purgatorio XIX.131.

If Guido's poems are dolci (sweet), in what way are they different from those of Dante that are written in the 'sweet new style' (Purg. XXIV.57)? This is a question that has had a variety of answers. Some make Guido the first practitioner of the 'sweet new style' (which is impossible, in Dante's view, given his precisions at Purg. XXIV.49-51 that make his canzone 'Donne ch'avete,' written at least ten years after Guido's death, the first of the 'new' poems that constitute the dolce stil novo). The better understanding is that Guido Guinizzelli, Guido Cavalcanti, others among the Tuscans, and surely Dante himself, all wrote in the 'sweet style' (the most effective demonstration of this is found in Lino Leonardi [“Cavalcanti, Dante e il nuovo stile,” in Dante: da Firenze all'aldilà. Atti del terzo Seminario dantesco internazionale, ed. Michelangelo Picone (Florence: Cesati, 2001), pp. 331-54]). But what both Guidos, the early Dante, and just about everybody else failed to do was to find the 'new style' that Dante developed from his understanding of Guinizzelli's sweet style of praise, a mode that he took one step farther when he developed his own theologized poetics of Beatrice. In that he was the first, and had perhaps no more than one companion (Cino da Pistoia, who also wrote 'theologically' of Beatrice after her death). This is what the evidence of the texts seems to suggest. In short, Dante honors Guinizzelli for being what he was, as far as the younger poet was concerned, the 'father' of the dolce stil novo, but not one of its practitioners. Thus we may understand that a number of poets wrote in the 'sweet style,' but hardly any of them achieved the new 'sweet style' that is the hallmark of Dante's praise of Beatrice. In all aspects of this debate, it is essential to remember that we are trying (or should, at any rate, be trying) to negotiate an answer from what Dante said happened, not from what actually happened (or what we imagine actually happened).

Lino Pertile, with whom this writer disagrees cordially about a number of issues, including this one, some ten years ago suggested, thoughtfully and usefully, that the interpretation advanced here boiled down to understanding the phrase dolce stil novo as though it had been ordered alla rovescia, that is as though Dante had written it as “novo stil dolce” (which he might easily have done, but chose not to). In other words, Guinizzelli had written in the “old” stil dolce, while Dante's theologized Beatrice made for a sweet “style” that is radically novo. Further, in this line of analysis it is possible that Dante thought of his own former poetry in celebration of other women as being similar to Guinizzelli's, as lacking precisely the theological “edge” that the prose of the Vita nuova (and the later poetry of the Commedia) gave to Beatrice, thus making a breakthrough that the younger Dante turned his back on. For instance, Casella's song in Purgatorio II is referred to as deeply affecting its auditor, Dante himself. The poet marks the moment in the following terms: “la dolcezza ancor dentro mi suona” (the sweetness sounds within me still – verse 114). Are we to think of the second ode of Convivio, “Amor che ne la mente mi ragiona,” as being in the intellectualizing “style” of Guinizzelli, praising his lady as being the repository of every value except the most important one, at least in a Christian poet? and thus as falling short of the dolce stil novo? This would again set the poetry in praise of the donna gentile against that in praise of Beatrice, the allegorical poetry of earthly love against a poetry based in a Psalm (113) itself based on the Exodus (see Hollander, “Purgatorio II: The New Song and the Old,” Lectura Dantis [virginiana] 6 [1990]: 28-45).

Dante's affectionate remark bestows less than it may seem to do, for all his unquestionable admiration of Guido. In Purgatorio XI.97-108 we are told both that this Guido had been eclipsed by Guido Cavalcanti in the public's esteem and that fame for artistic excellence does not last very long, in any case. In short, 'modern custom' is not enduring. As long as such poems (i.e., those in the Italian vernacular) are written, Guido's will be read; he will have an honored place amongst the vernacular poets. Dante, partly because of his own endeavors on Guido's behalf, got it right. His praise, however, is not of the order of his praise of Virgil; according to Beatrice his fame will last as long as the world lasts (Inf. II.59-50), a life that even Dante probably thought would be longer than that enjoyed by the Italian vernacular (see Par. XXVI.133-138). For Dante's relationship to Guido see discussions in Mario Marti (Con Dante fra i poeti del suo tempo [Lecce: Milella, 1966]), Gianfranco Folena (“Il canto di Guido Guinizzelli,” Giornale storico della letteratura italiana 154 [1977], pp. 481-508), Vincent Moleta (Guinizzelli in Dante [Rome: Edizioni di Storia e Letteratura, 1980]), and Teodolinda Barolini (Dante's Poets [Princeton: Princeton University Press, 1984]), esp. pp. 125-36.

Guido now, in a moment of evidently heartfelt humility, insists on Arnaut Daniel's superiority in the vernacular, but in the vernacular of Provence, an area in the southeast of what is now France; it only came under the official rule of the French kings in 1245.

'Arnaut Daniel, famous Provençal poet (fl. 1180-1200); but little is known of his life beyond that he belonged to a noble family of Ribeyrac in Périgord, that he spent much of his time at the court of Richard Coeur-de-Lion, and that he visited Paris, where he attended the coronation of Philip Augustus, as well as Spain, and perhaps Italy. His works, such as they have been preserved, consist of eighteen lyrical poems, one satirical, the rest amatory' (Toynbee, “Arnaldo Daniello” [Concise Dante Dictionary]).

Arnaut's poems are brilliantly difficult, written in the so-called trobar clus (a modern translation could be 'hermetic verse'), typified by its harsh tones and challenging phrasing; in other words, he avoids the trobar leu (a more open style, easier and more mellifluous). On Dante's knowledge and use of Arnaut's poems see Maurizio Perugi (“Arnaut Daniel in Dante,” Studi Danteschi 51 [1978], pp. 59-152.). For a brief overview in English of Dante's relations with Provençal poets and poetry see Thomas Bergin (“Dante's Provençal Gallery,” Speculum 40 [1965], pp. 15-30). And for a substantial study of Dante's response to previous modern poets, Provençal and Italian, see Teodolinda Barolini (Dante's Poets [Princeton: Princeton University Press, 1984]), pp. 85-187.

What does it mean to say that Arnaut was 'a better craftsman of the mother tongue'? Not, it would appear, that he wrote poems that were better in substance than those of others, but better in style, better made. It is possible that Guinizzelli's language reflects the claim that lay latent (but clear enough) in some poems of Cavalcanti, who liked to use the image of the file (limo), an instrument used to refine one's handiwork, to suggest the careful nature of his own poeticizing. Dante here would seem to be, through the testimony of Guinizzelli, taking some of that distinction away from the other Guido and, in a sense, replacing him with Arnaut, a craftsman not only better than he, but better than Cavalcanti, too.

The 'mother tongue' here not only refers to Arnaut's langue d'oc, his native Provençal, but to any Latin-derived vernacular (see Sordello's similar remark, addressed to Virgil, at Purg. VII.16-17, which also makes Provençal the child of Latin: 'O glory of the Latins... through whom our language showed what it could do').

Guido also seems to mean that Arnaut's vernacular verse was stylistically better than anyone else's writing in the vernacular, whether in verse (in the langue d'oc – Dante is cavalierly unconcerned, at least publicly, with French poetry) or in prose fiction (in the langue d'oïl, since according to De vulgari eloquentia I.x.2, all vernacular prose was written in French). This passage is the cause of much debate, one issue in which centers on the question of whether or not Dante means that Arnaut himself wrote prose fiction. This seems more than dubious. See Poletto's comment to this effect (gloss on vv. 118-120).

'Giraut de Borneil, one of the most famous troubadours of his century, born at Essidueil, near Limoges, ca. 1175, died ca. 1220' (Toynbee, “Geradus de Borneil” [Concise Dante Dictionary]). Dante refers to him a number of times in De vulgari eloquentia and once in Convivio, never slightingly – quite the opposite is true. Hence Dante's about-face here is dramatic. In De vulgari eloquentia II.ii.9 he had made Girault his own Provençal counterpart: Dante presented himself as the leading poetic celebrant of virtue in Italian, while Giraut was presented in exactly the same role for his language group. In the same passage Arnault Daniel was put into the identical relationship with Dante's then first friend, Cino da Pistoia; they are the twin wonders of amorous poetry. However, and as Barolini points out (Dante's Poets [Princeton: Princeton University Press, 1984]), pp. 98-99, Giraut is clearly ranked above Arnaut. For consideration of Dante's possible reasons for wanting now to distance himself from Giraut see Picone (“Giraut de Bornelh nella prospettiva di Dante,” Vox romanica 39 [1980], pp. 64-85) and Mengaldo (“Dante come critico,” La parola del testo 1 [1997], pp. 36-54). And for a set of formulations that attempts to deal with the whole range of the Comedy's revisionism regarding Dante's poetic precursors, see Barolini, pp. 175-87.

'Guittone del Viva, more commonly known as Fra Guittone d'Arezzo, one of the earliest Italian poets, was born ca. 1240 at Santa Firmina, about two miles from Arezzo. But little is known of the details of his life, a great part of which was spent in Florence, where Dante may have known him. About the year 1266 Guittone, who was married and had a family, entered the Order of the Frati Gaudenti [Jovial Friars, see Inf. XXIII.103]. In 1293 he helped to found the monastery of Sta. Maria degli Angeli at Florence, in which city he appears to have died in the following year' (Toynbee, “Guittone” [Concise Dante Dictionary]). Dante, surely unfairly, is getting even, less perhaps with Guittone than with his admirers, who were many (and continue to exist today). It was not enough for him to have Bonagiunta (Purg. XXIV.56) include him, along with himself, among those who failed to come up to Dante's measure. Now Dante uses the great Guinizzelli (who had indeed come to dislike Guittone's poetry in reality) to skewer Guittone a second time, and much more harshly. As early as De vulgari eloquentia (e.g., I.xiii.1; II.vi.8) Dante had been deprecating Guittone. One may sense a certain 'anxiety of influence' at work here, especially since Guittone did so many of the things that Dante himself took on as his own tasks: love lyrics, moral canzoni, and eventually religious poems.

On Dante's rejection of Guittone see Contini (Un' idea di Dante [Turin: Einaudi, 1976]), pp. 60-61, the contributions of Gorni, Antonelli, and Mazzoni in the volume devoted to studies of Guittone edited by Michelangelo Picone (Guittone d'Arezzo nel settimo centenario della morte: Atti del Convegno Internazionale di Arezzo [22-24 aprile 1994], ed. M. Picone [Florence: Franco Cesati, 1995], pp. 307-83), and Barolini (“Guittone's Ora parrà, Dante's Doglia mi reca, and the Commedia's Anatomy of Desire,” in Seminario Dantesco Internazionale: Atti del primo convegno tenutosi al Chauncey Conference Center, Princeton, 21-23 ottobre 1994, ed. Z. G. Baranski [Florence: Le Lettere, 1997], pp. 3-23).

Guido's request speaks now of a better kind of poetry, the Lord's Prayer, that this other poet can offer in his name in Heaven. He concludes by suggesting an 'edit' in the text, namely of that part which speaks of leading not into temptation and delivering from evil (Matthew 6:13; Luke 11:4). Similarly, in Purgatorio XI.19-24 the penitents do sing this part of the text, but only on behalf of their earthly brethren, since it no longer pertains to them.

Guido's departure reminds the reader that those who speak with Dante need to come to the edge of the flame, without leaving it, in order to be visible (see Purg. XXVI.13-15). His disappearance back into the flames like a fish from the surface of a pond recombines images of fire and water, as did his opening sally to Dante, in which he speaks of his thirst, while he burns in flame, to know of a living man's reason for being here.

Why does Dante now, in a canto composed of so much direct address, shift into indirect discourse? Steiner, glossing these lines, was perhaps the first commentator to note that the style of the remark is 'elegant and studied in form,' as befits a discussion with Arnaut or any Provençal poet, all of whose verse is similarly ornate. But Steiner does not remark upon the use of indirect discourse per se, nor does any other commentator. In fact, Dante had already employed it in describing his oath to Guido (Purg. XXVI.104-105). A desire for stylistic variation may explain his choice, or there may be a more interesting reason that has escaped us.

That Arnaut's speech forms a little Provençal poem eight lines in length may reflect, according to Nathaniel Smith (“Arnaut Daniel in the Purgatorio: Dante's Ambivalence toward Provençal,” Dante Studies 98 [1980]), pp. 101-2, the fact that this was generally a favored stanza length for Provençal poets and indeed of six of the eleven Provençal poems cited by Dante in De vulgari eloquentia. Smith also demonstrates that Dante's pastiche of Provençal lyric is wider than an imitation of Arnaut, and reflects numerous other poems and poets in the lingua d'oc. He also comments on the deliberate 'dumbing down' of Dante's version of Arnaut's language (recognized by its difficult and elliptical style) in which almost every word of this invented poem has an obvious Italian (or Latin) cognate (pp. 106-7).

It is interesting to consider that both the first and last lines of Arnaut's poem contain fairly obvious imitations of lines of two other notable vernacular poets. The first, as has been noted by commentators, first by Chimenz (comm. to these verses), is meant to recall the first line of a canzone of Folco di Marsilia, the last vernacular poet to appear in the poem and the most highly placed, the only one seen in paradise (he is named at Par. IX.94). Dante cites the opening line of one of his canzoni as being among the illustres (i.e., components of the 'illustrious vernacular' that his treatise champions) in De vulgari eloquentia (II.vi.6): 'Tan m'abellis l'amoros pensamen' (So pleases me the very thought of love). Arnaut's opening line derives from Folco's. And the same is true of his closing verse, derived, as Mario Marti indicates (in his edition, Poeti del Dolce stil nuovo [Florence: Le Monnier, 1969] – note to verse 7 of Cino's poem numbered 64), from Cino da Pistoia: 'e sovverrebbe a voi del mio dolore' of which 'sovenha vos a temps de ma dolor' (remember, when the time is fit, my pain) is a fairly close Provençal translation. If we are supposed to notice these redoings, are we also supposed to consider a new side to the daring and flashy Arnaut, now content to borrow from his fellows? And they are particularly good ones, Folco, already in Heaven, and the Cino who has had, in De vulgari, Dante's imprimatur as fellow poet of Beatrice. (For a speculative treatment of Cino's absence from the Commedia, based on the hypothesis that he was indeed originally intended to be the last modern poet included in the poem at the center of Paradiso, a plan that was discarded once Dante became aware that Cino had deserted the imperial cause and, surely worse, joined with his Florentine enemies sometime after the death of Henry VII in 1313, see Hollander [“Dante and Cino da Pistoia,” Dante Studies 110 (1992)], pp. 215-19.)

Arnaut, like Guido, is now more interested in salvation than in dazzling the world with verse. These are penitential moments, not ones that encourage thoughts of emulous poetic striving. Giacalone (comm. to vv. 139-147) says it this way: 'The poet of the trobar clus, who was ever at the ready to make his poems “concealing,” now at the end conceals himself, but in the penitential fire.' There seems no question but that Dante was staggered by Arnaut's virtuosity; and surely we were staggered, in turn, by Dante's perhaps most virtuosic versification, heavily indebted to Arnaut, found in the last cantos of Inferno. But now there is no dwelling on the excellence of poetic technique, but rather on prayer and hopes for the joys of Heaven. Guido's last words (Purg. XXVI.92-93), which seem to imply that, unlike Dante, he came to God not in his poetry but even in spite of it, and Arnaut's Provençal stanza, which also speaks to life rather than to art, are matched by Dante's own 'sweet new style' here, singing of salvation, the subject it took this poet some time to find again after his first attempt in the concluding chapter of his Vita nuova. In this sense Guido's words about prayer as poem and Arnaut's poem, which is a request for prayer, posthumously join these two poets to Dante's new style, a poetry in tune with God.