Princeton Dante Project Allegory
Lectures

Allegory in Dante
(Robert Hollander)

The first time I read Dante "seriously," I was twenty-five years old and, as a brand-new instructor, had to lead students through Inferno for a week in the "Great Books" course that is still required of all students at Columbia College. The one thing that I knew I did not understand (there were of course many, many other things) was the way in which the poem signified. I was able to put a word to my puzzlement: "allegory." Almost anyone coming to the Comedy for the first time has probably heard two things about its larger strategies -- it is the most Christian of poems and it is "an allegory." But an allegory of what? And what is allegory in the first place? 1

The most simple medieval definition of allegory is found in the seventh-century Spanish encyclopedist, Isidore of Seville (Lind.1911.1 -- Etymologiae I, xxxvii, 22): "Allegoria est alieniloquium, aliud enim sonat, aliud intelligitur" (Allegory is "otherspeech," for it occurs when one thing is said and another is understood). This definition, which may be taken as being either global or narrowly particular (it is in fact offered as a definition of one kind of irony by Isidore), is frequently referred to in discussions of allegory and is thus included here. (For a recent discussion of allegory as it is defined by grammarians and rhetoricians see (Ales.1987.1.) In fact, however, it does not resolve the problem in a helpful way, especially for students of Dante, who himself referred to two kinds of allegory, which he called the "allegory of the poets" and the "allegory of the theologians" (Conv.II.i.3-4). If we can understand what Dante meant in his discussion, we may be able to understand better what he did in his poem.

Allegory, as practised by poets, may be generally described as possessing the following characteristics. The work involving it (1) is to be understood as being fictive and not in any way recording events that have occurred (e.g., the Romance of the Rose [13th century]), while also (2) being developed as an extended metaphor (e.g., the Christian life portrayed as a continual "war," an inner struggle against inner temptations and external forces ranged against the would-be Christian -- see the Psychomachia of Prudentius, ca. 405). In concert with these two characteristics, allegory of the poets (3) presents its action as being internal, as taking place in the mind or soul of a single figure (e.g., Prince Arthur in Spenser's Faerie Queene, 1596) or of an anonymous "everyman" (e.g., Pilgrim in John Bunyan's Pilgrim's Progress, 1678). Further, allegorical fictions (4) tend to rely heavily on the use of personifications, generally of vices (e.g., Incontinence, Despair) and virtues (e.g., Continence, Hope), "ladies" (abstractions in Latin generally take the feminine gender, e.g., continentia, spes) who perform physical actions in battle with other "ladies," as in the Psychomachia. Not every allegorical work has all these characteristics; all, however, possess some of them.

Now let us examine Dante's definition of allegory in the opening pages of the second treatise of Convivio: "The first [sense of a text] is called the literal, and this is the sense that does not go beyond the surface of the letter, as in the fables of the poets. The next is called the allegorical, and this is the one that is hidden beneath the cloak of these fables, and it is a truth hidden beneath a beautiful [lie].... Indeed the theologians take this sense otherwise than do the poets; but since it is my intention here to follow the method of the poets, I shall take the allegorical sense according to the usage of the poets" (Conv.II.i.3-4 -- trans. R. Lansing [Lans.1990.1]). For Dante the distinguishing element of the allegory of the poets is that it is literally untrue. But are not all poems literally fictive? Before attempting to respond to that question, let us consider what Dante believes to be the distinguishing mark of theological allegory, the way in which "the theologians take this sense otherwise than do the poets." It is clear that he is now speaking of a privileged and limited class of texts, the historical passages in the Bible that medieval exegetes believed to possess four senses. 2

For now we observe only a single and crucial particular. As opposed to the literal sense of poets' allegory, the literal sense of theological allegory is historically true, found only in events narrated in the Bible (e.g., the fall of Adam and Eve, Moses leading the Israelites in the Exodus, the birth of Jesus, the Crucifixion). In his discussion in Convivio II Dante, unsurprisingly, goes on to say that he will employ the allegory of the poets to elucidate the meaning of the "allegorical" Lady Philosophy (clearly, by the way, a personification) found in his odes. What has not received sufficient attention is the astounding fact that he claims that he could have employed theological allegory in his analysis of his poems. In the Middle Ages the line separating the two kinds of allegorical exegesis was clearly drawn. All secular literature of an imaginative kind was dealt with as being fictive and not historical. 3

Further, theological allegory was limited to a single use, interpreting the several meanings found in certain (far from all) historical passages in Scripture. Dante, in one rather alarming step, had crossed that line -- in theory if not in practice. Nonetheless, the claim he had staked when he wrote Convivio (ca. 1304-6) lay ready to be put to use when he moved on to the Comedy (ca. 1307).

Later in his life Dante wrote a letter to one of his most important supporters, Cangrande della Scala, a sort of preamble to his Paradiso, in which he explains many of the essential strategies of the Comedy, and most particularly its use of allegory. It must immediately be said that, for nearly two hundred years, the authenticity of this document has been hotly debated, with those finding in the negative almost always doing so because they do not find what is said in the Epistle concordant with their views of Dante's opinions or practice. This writer is among those convinced that Dante did in fact write it (see Holl.1993.2). Certainly its most astounding and controversial assertion is that the fourfold interpretation of texts used to elucidate the historical meanings of the Bible was the very method to be used in order to understand the Comedy. This is surely the stuff of heresy. For the position at the very least and unmistakably implies that the literal sense of the poem is historical, i.e., that Dante's seven-day visit to the afterworld is to be treated as historical fact. Whether or not Dante wrote this document, some contemporary students of the issue make the point that his practice in the poem is such as to indicate that the Epistle, whoever wrote it, only makes explicit what had already been accomplished in the poem (e.g., Barolini [Baro.1990.1], p. 142, and Hollander [Holl.1990.1], pp. 33, 43).

Dante, faced with the strong opposition of theologians to the idea that secular literature had any meaningful claim to purvey truth, made a bold decision. Rather than employ the allegory of the poets, which admitted, even insisted, that the literal sense of a work was untrue, he chose to employ the allegory of the theologians, with the consequence that everything recounted in the poem as having actually occurred is to be treated as "historical," since the poet insistently claims that what he relates is nothing less than literally true. We do not have to agree that such was in reality the case, only that the poet makes precisely this claim -- and no less than it. If we can acknowledge that much, we have come a long way toward demystifying this subject. However the four senses of theological allegory may function in the Comedy, we can come to the understanding that its pretext is that it is to be read historically. 4

Charles Singleton, one of the leading exponents of the "theological school," put the matter succinctly: "The fiction of the Divine Comedy is that it is not fiction" Sing.1957.1, p. 129. The crucial and noteworthy result of framing the question in this way is to be freed of the interpretive shackles imposed by forcing the "allegory of the poets" onto the poem. And that is exactly what has been its fate from the time of the earliest commentators (for a brief survey of the commentary tradition see Holl.1993.3). An example may help. When Virgil enters the poem in its first canto, the vast majority of early commentators (and the phenomenon, surprisingly, persists) treat him as an abstraction, an allegory (often of "Human Reason" or something similar). But even a cursory reading of the text that presents him ( [Inf I 67-75]) reveals that he stands before us as the son of Mantuan parents who lived at Rome at the time of the Caesars and who wrote the Aeneid.

If the student who is wrestling with this difficult matter for the first time takes only this much away from this discussion, it should be of considerable aid. You are not asked by the poem to see Virgil as Reason, Beatrice as Faith (or Revelation), Francesca as Lust, Farinata as Heresy. You may banish such abstractions from your mind unless Dante himself insists on them (as on occasion he does -- e.g., the noble castle in Limbo [see C.Inf.IV.106-111] or the Lady Poverty, beloved of St. Francis [Par XI 74], who is not to be confused with any historical earthly woman but is to be regarded as the ideal of Christ's and the Apostles' renunciation of the things of this world). It is a useful and pleasing freedom that you enjoy: "The allegory of the Comedy is not allegory as the commentators urge me to apply it. I may read this poem as history, and understand it better." When I first taught this poem, in 1958, I wish someone had given me that gift.

(February 1998)


FOOTNOTE 1: For those who are interested, introductory studies of Dante's allegory accompanied by extensive bibliographies of the subject do exist. For readers of Italian there is the overview of the subject found in the article "allegoria" by Jean Pépin (ED.1970.1), pp. 151-165; for those who read French, Pépin's somewhat fuller treatment (Pepi.1970.1); for the most recent brief treatment in English one may consult Ronald Martinez's entry "allegory" in the Dante Encyclopedia (Lans.1999.1). There is a recent volume of essays devoted this subject: (Pico.1987.1). Those interested in my view of the subject may consult a book and an article written some time ago (Holl.1969.1; Holl.1976.1). (Return to text)

FOOTNOTE 2: The four senses (and not, as all too often in modern discussions, "levels") of the Bible are generally put forth, and especially in the wake of Thomas Aquinas (Summa theologiae I.i.10), as follows: (1) historical/literal, (2) allegorical, (3) moral or tropological, (4) anagogical. This is difficult material for most modern readers. (The classic work on the subject is Lubac's [Luba.1964.1]; a brief review in English is found in Grant [Gran.1963.1].) The single most important thing to understand is that these senses occur in a historical continuum. For instance, the historical Moses, leading the Israelites out of captivity, gains his allegorical meaning in Christ, leading humankind out of bondage to the freedom of salvation. His moral (or tropological) sense is present now -- whenever "now" occurs, in the soul of the believer who chooses to make his or her "Exodus," while the anagogical sense is found only after the end of time, when those who are saved are understood as having arrived in the final "Promised Land," the blessed life in heaven. To offer a second example, one favored by Dante's early commentators: Jerusalem was the historical city of Old Testament time; it points to the allegorical Jerusalem in which Jesus was crucified; it is the moral or tropological "city" (whether within a single believer or as the entity formed by the Church Militant now) at any present moment; it is, anagogically, the New Jerusalem, which will exist only at the end of time. How the formative principles of this historical method of biblical interpretation inform the Comedy will be addressed below. (Return to text)

FOOTNOTE 3: There did exist a category to describe "realistic" fiction, called argumentum, as opposed to fabula (things clearly untrue) and to historia (things that have clearly occurred). Argumenta are things that, while not having actually occurred, could have done so. For general discussion see Curtius (Curt.1948.1), pp. 452-455; for discussion with particular reference to Dante see Hollander (Holl.1969.1), pp. 256-258. (Return to text)

FOOTNOTE 4: If the Comedy is to be understood as being intimately related to the allegory of the theologians, in what precise ways may we understand that Dante has adapted its methods? (An important study of this subject remains Auerbach's study "Figura" [Auer.1944.1].) The main technique he borrows from the theologians centers on the relationship between the first two senses, historical/literal and allegorical. These two senses have more than several sets of synonyms, among which the two most commonly used are "type and antitype," "figure and fulfillment." In addition, historical events are sometimes referred to as "shadows" of things to come. Theologians refer to Moses as the type of whom Jesus is the antitype (a warning: this word is often misused in discussions of Dante's allegory); of the sacrifice of Isaac intended by Abraham as the figure of which the Crucifixion is the fulfillment; of Jonah's three days in the innards of Leviathan as the shadow of Christ's three-day sojourn under the earth between the Crucifixion and the Resurrection. While all these historical senses may have moral/tropological and anagogical ones as well, most discussions are directed to their second meaning, the allegorical (or "figural" or "typological") sense. The main contention of those who believe that Dante borrowed this technique of ordering meaning by historical referentiality is that his appropriation of the first two senses is what clearly indicates this indebtedness. One may indeed argue that each sinner or saved soul is seen as the fulfillment of the earthly life that brought him or her to a place in the afterworld -- and this is the view most significantly advanced by Auerbach. To this formulation one may justly respond that, if it is true, it is also not very interesting, except insofar as it guarantees the exceptional claim the poet makes for his poem, that its mode of signifying is to be understood as being that of the allegory of the theologians. Some recent discussions of the question have centered on the far more interesting typological, or figural (these words, one should remember, are synonyms), relationships exhibited by personages in the poem to other figures or events. To take a single instance, one fifteenth-century reader of Dante (Filippo Villani) reported that the Florentines referred to Guido da Montefeltro (canto XXVII) as the "new Ulysses" (Holl.1980.1, p. 142). In such an observation, in which laymen are "reading" a historical being "figurally," we can observe the adaptation of a theological technique to a secular understanding. (For studies of such adaptations see (Mine.1977.1.) In Dante's poems one finds any number of such moments. Pope Celestine V (if it is he who is referred to in [Inf III 59-60], as the most convincing arguments make probable [C.Inf.III.58-60]) is the "antitype" of others before him who showed a similar form of cowardice -- e.g., Pontius Pilate); Francesca, reading a book that leads to her "conversion" to sin and death in the company of a man named Paul, is the "negative antitype" of St. Augustine, reading a book by Paul that leads to his conversion (Confessions VIII.xii [Pine.1961.1, p. 178] -- see Swing for what seems to be the first observation of this striking connection [Swin.1962.1], p. 299, and further discussion by Hollander [Holl.1969.1], pp. 112-113); Cato, the guardian of Purgatory, is presented (cantos I and II) in such ways as to portray him as the pagan antitype of Moses and type of Christ (for the second correspondence see E. Raimondi [Raim.1962.1], pp. 81-83; for both see Hollander [Holl.1969.1], pp. 124-129). Statius, rising from his purgation ready to ascend to heaven ([Purg XXI 7-10]), has long been understood to be in the role of a "post-figured" resurrected Christ because the language of the poem makes this seem only a necessary view of the matter.

Users of this electronic version of the poem and of its accompanying apparatus will find, in various notes, further evidence for the presence of the allegory of the theologians as the central underlying technique of signifying in Dante's poem. The main point a first reader of Dante should take from these remarks is that he or she is essentially freed from reading the poem as an allegory, at least as literary allegory is traditionally defined. Cato is not the personification of an abstract virtue, but the historical being whose looks, actions, and significance all point to other historical entities and actions. This way of reading him is not easier than traditional allegorical exegeses, but is much more satisfying. Insofar as that is a true statement, it underlines the importance of attending to Dante's choice of a theological allegory that is denied him by every competent authority, but which he simply chose to make his own.(Return to text)

Copyright © 1997-2000 The Trustees of Princeton University and Prof. Robert Hollander. All Rights Reserved. Use of this site is governed by Terms and Conditions of Use/Legal Notices in connection with the use of this web site. Thank you for visiting. And for reading the fine print. Bugs? Comments?