From the Tree to the Labyrinth

7

Dante between Modistae and Kabbalah

7.1. The

De vulgari eloquentia

In his De vulgari eloquentia (hereinafter DVE), to explain the existence of a plurality of languages, Dante sticks to the letter of the biblical account in Genesis, which he knew in the Latin text of the Vulgate. So we must stick to the Vulgate too, setting aside any philological concerns regarding its fidelity to the original Hebrew. In any case, as we shall see, Dante occasionally strays, with the highhandedness we have come to expect of him, even from the text of the Vulgate.

If the DVE is a treatise on language and speech acts, Genesis offered Dante many examples of primal “speech acts.” The first thing we must agree on, however, is what “speaking” means. Certainly, every sign—as Augustine had already remarked—is something perceptible to the senses that serves to bring to mind something different from itself, but this definition (which could also refer to the knot I tie in my handkerchief to remind me of a task I must do) does not yet imply a communicative relationship articulated between two subjects. Rosier-Catach (2006) sees this communicative aspect underscored instead by definitions like the one in Calcidius’s Latin translation of the Timaeus, later echoed by Thomas Aquinas (“ut Plato dicit, sermo ad hoc datus est nobis ut cognoscamus voluntatis indicia” [“As Plato says, speech was given to us so we could know signs of others’ wills”], De veritate 9, 4, 7).¹ If Dante understood a speech act in this way, then we ought not to say that God “speaks” when he pronounces the fiat lux, as a result of which “there was light” (Gen. 1:3–4);² and the same could be said for other similar expressions used in the course of creation. Here God seems instead to know “how to do things with words,” bringing into play a magic, operative, performative quality of the word—thus setting a dangerous precedent for all future followers of the occult sciences, convinced they can change the course of events simply by uttering a few para-Hebraic sounds. In the same way it is not clear what God was up to when, for example, He called (“appellavit”) light “day” and darkness “night,” seeing that He had no need to communicate those names to anyone, least of all Himself.

Dante is nonetheless aware of the fact that the Bible often speaks in a figurative way, and he does not make these divine “words” the object of his reflection, considering that, as far as he is concerned, it is evident that the gift of speech has been conferred on man alone (“patet soli homini datum fuisse loqui,” DVE I, ii, 8). As he will repeat on a number of occasions, the ability to speak belongs only to mankind: the angels don’t have it (they are gifted with an “ineffable intellectual ability,” which allows each of them to understand the thought of all the others, or, alternatively, all of them read the thoughts of all the others in the mind of God) and the demons (who are already reciprocally aware of the depths of their own perfidy) don’t have it either. And—we may add—if the angels have no use for speech, the same is even truer of God when He was creating the universe.

It is the intention of the DVE, therefore, to deal solely with human speech, inasmuch as man is guided by his reason, which in single individuals assumes different forms of judgment and discernment, and requires a faculty that will permit the speaker to transmit an intellectual content through signs perceptible to the senses, in a relationship between sound and sense that he can recognize (in accordance with tradition) ad placitum, in other words, as conventionally agreed upon.³

Nevertheless, Dante still has to explain the episode recounted in Genesis 2:16–17, when the Lord speaks to man for the first time, placing at Adam’s disposal all the resources of the Earthly Paradise, and commanding him not to eat of the fruit of the Tree of Good and Evil. Clearly, what we have here is an initial act of communication, which would contradict the idea that man was the first to use language. Dante gets out of this by affirming that the fact that God communicated something to Adam does not mean that He did so verbally, but (and this traditional idea comes from Psalm 148: “fire and hail, snow and frost, stormy wind fulfilling his command” [“ignis grando nix glacies spiritus procellarum quae faciunt verbum eius”—the verb “faciunt” in the Vulgate is ambiguous and could mean “that do his word” or “that make up his word”]) He could have expressed himself through atmospheric phenomena, such as thunderclaps and lightning.

Having clarified these issues, Dante might at this point have discussed how Adam spoke when the Lord (Gen. 2:19) formed out of the ground every beast of the field and every bird of the air and brought them to the man to see what he would call them, and whatever the man called each living creature, that was its name (“omne enim, quod vocavit Adam animae viventis, ipsum est nomen eius”). Curiously enough, this role of Adam as nomothete (with the tremendous problem, touched on in Plato’s Cratylus, and which will obsess the coming centuries, that is, on what basis did Adam name the animals—with the names due to them because of their natures or with those that he himself arbitrarily chose to assign, ad placitum) is ignored by Dante. Nor is that all. Disregarding the fact that, in order to name the animals, Adam must have spoken in some way, Dante confesses to being perplexed by the fact that “according to what it says at the beginning of Genesis” the first to speak was the “most presumptuous” Eve (“mulierem invenitur ante omnes fuisse locutam” [“we find that a woman spoke before anyone else”] DVE I, iv, 3), when she engaged in dialogue with the serpent, and he finds it unbecoming that such a noble act of the human race should have emerged, not from the lips of a man but from those of a woman (“inconvenienter putatur tam egregium humani generis actum non prius a viro quam a femina profluxisse” [“it may be thought unseemly that so distinguished an action of the human race should first have been performed by a woman rather than a man”] DVE I, iv, 3, p. 9).

In fact, this observation (apart from the puzzling display of antifeminism on the part of a poet who sang the praises of a donna angelicata or a mortal woman glorified as an angel)—if we exclude the doubtful “words” attributed to God, Adam is the first to speak—first of all when he names the animals, and then when he expresses his satisfaction with the appearance of Eve. Indeed, in the latter case, an entire utterance of his is cited for the first time: “dixitque Adam hoc nunc os ex ossibus meis et caro de carne mea haec vocabitur virago quoniam de viro sumpta est” (“And Adam said, This is now bone of my bones, and flesh of my flesh: she shall be called Woman, because she was taken out of Man”). Mengaldo (1979: 42) suggests that, since for Dante people speak to externalize the thoughts in their minds, and speech is therefore a dialogical phenomenon, what Dante meant to say was that what we have between Eve and the serpent is the first dialogue, and hence the first linguistic act expressed through the physical production of meaningful sounds. Which would lead us to believe that when Adam is pleased with the appearance of Eve and “says” what he says, maybe he says it to himself, and that (but perhaps we are overmodernizing Dante) the naming of the animals ought not to be considered a linguistic act but a mere metalinguistic foundation.

However we may wish to interpret this liberty that Dante takes as a reader of the Bible, Mengaldo’s suggestion nonetheless prompts us to clarify what a linguistic act, as distinct from a language, meant for Dante, in other words to ask ourselves whether or not there is in Dante a critical awareness of the difference (to use the Saussurean terminology) between langue and parole.

Tendentious though he may be in the episode involving Eve, Dante is keen to defend his conviction that Adam ought to have been the first one to speak. And, despite the fact that the first sound uttered by human beings is usually a cry of pain, Adam’s first utterance could only be a cry of joy and at the same time an act of homage to his creator. Therefore, Adam’s first utterance must have been the name of God, El (DVE I, iv, 4, p. 9).

Confronted with this first linguistic act in human history, Dante must now come to grips with the issue he had proposed to deal with at the very beginning of the DVE, precisely because the plurality of languages confirmed by his experience finds its foundation and explanation in Genesis 11:1 and following. The story is a familiar one: after the Flood “the whole earth was of one language, and of one speech,” but pride led mankind to vie with God and to construct a tower whose top might reach unto heaven, and the Lord, to punish their pride and prevent the construction of the tower, decides to confuse their languages.

It is true that in Genesis 10, speaking of the spreading abroad of the sons of Noah after the Flood, it is said: “By these [the sons of Japhet] were the isles of the Gentiles divided in their lands; every one after his tongue, after their families, in their nations” (10:5), and in almost the same words the concept was repeated for the sons of Ham (10:20) and the sons of Shem (10:31). This hint of a plurality of languages existing before Babel will prove a sticking point, not only for many interpreters of the Bible but also for the Utopians of the Perfect Language (see Eco 1993, English trans. 1995b). But Dante does not consider these passages.

He is clearly convinced that before the building of the Tower of Babel there existed a perfect language, which Adam had used when talking to God, and with which he had spoken to his descendants, and that the plurality of languages had come about only after the confusio linguarum or confusion of tongues. Demonstrating a knowledge of comparative linguistics exceptional in his day, Dante shows how the various languages that sprang from the confusion multiplied in a ternary fashion, first according to a division among the various parts of the world, then, within the area that today we would define as Romance, they split up into langue d’oc, langue d’oil, and lingua del sì. The last-named, the language spoken in Italy, has become further fragmented into a plurality of dialects that sometimes, in Bologna for example, vary from one quarter of the city to another. This is because man is a mutable animal in customs, habits, and languages, over the course of both time and space.

Dante’s project for devising a more decorous and illustrious language (what he calls the volgare illustre) for the whole of Italy is to proceed to a critique of the various regional vernaculars, given that poets have always tended to keep a certain distance from the local dialect. His aim is to identify a vernacular that is illustrious (a bearer of light), cardinal (that functions as a cornerstone [cardine] for all the others), aulic or regal (worthy of its place in the palace of a national kingdom), and curial (the language of government, of the law courts, of instruction). This vernacular represents a kind of ideal rule that the best poets have come more or less close to, and by whose standards the existing vernaculars are to be judged.

The second, incomplete, portion of the DVE outlines the rules of composition for the one and only truly illustrious vernacular, the poetic language of which Dante considers himself to be the founder. But it is the first part of the treatise that interests us here.

The DVE defines the vernacular as the language children learn to use when they begin to articulate sounds, which they acquire by imitating their wet nurse, and he opposes it to a locutio secundaria, called grammar (grammatica) by the Romans. Grammar meant a language governed by rules that require extended study and of which one must acquire the habitus. This locutio secundaria is the scholastic Latin whose rules were taught in the schools of the day, an artificial idiom, “perpetual and incorruptible,” the international language of the Church and the university, frozen in time into a system of rules and regulations by the grammarians who had laid down the law when Latin had ceased to be the living language of Rome.

Faced with this distinction, Dante states unequivocally that the vernacular is the nobler language because it was the first one used by the human race; because the whole world uses it “though with different pronunciations and using different words” (DVE I, i, 4); and lastly because it is natural whereas the other is artificial.

On the one hand, then, he affirms that the nobler language must fulfill the requirements of naturalness, while the recognized diversity of the vernaculars confirms their conventionality (and Dante admits that the relationship between signifier and signified, a consequence of the faculty of speech, is the product of convention, in other words, ad placitum). On the other hand, he speaks of the vernacular as a language everyone shares, even though vocabulary and pronunciation may vary. Since the whole of the DVE insists on the variety of languages, how are we to reconcile the idea that languages are many with the fact that the vernacular (natural language) is common to the whole human race? The answer is that it is “natural” and common to all to learn first of all a natural language without being aware of its rules, but that this occurs because all mankind has in common a natural predisposition for language, a natural linguistic faculty, which is embodied, in Scholastic terms, in different linguistic substances and forms (see also Marigo 1938: ch. 9, n. 23; Dragonetti 1961: 32).

Dante affirms in fact (DVE I, i, 2) that the ability to acquire one’s mother tongue is natural, and this ability is common to all peoples despite the differences in vocabulary and pronunciation. He is not speaking then of a specific language, but of a general ability shared by all members of the species.

It is clear to him, then, that, while the language faculty is permanent and unchanging for all members of the species, natural languages on the other hand are capable of developing and becoming enriched over time, either independently of the wills of individual speakers or, on the contrary, as a result of their creativity—and the illustrious vernacular he is proposing to forge is meant to be a product of individual creativity. But it seems that between linguistic faculty and natural language he wishes to distinguish an intermediate moment.

In the opening chapter of the first part of the DVE, Dante, referring to his notion of the vernacular, uses terms such as vulgaris eloquentia, locutio vulgarium gentium, and vulgaris locutio, while he uses locutio secundaria for grammar. We could translate eloquentia in the generic sense either as “eloquence” or as “speech” or “manner of speaking.” But the text contains a distinction among various lexical choices that is probably not casual. In certain cases Dante speaks of locutio, in others of ydioma, of lingua, or of loquela. He uses ydioma, for example, whenever he is referring to the Hebrew language (DVE I, iv, 1; vi, 1; and vi, 7), as well as in reference to the branching off of the world’s languages, and the Romance languages in particular.

In I, vi, 6–7, in speaking of the confusio linguarum of Babel, Dante uses the term loquela. In the same context, however, he also uses ydioma, both for the languages of the confusion and the Hebrew language that remained intact. Similarly, he speaks of the loquela of the Genoese and of the Tuscans, but he also uses lingua for Hebrew and the dialects of the Italian vernacular. Writing again about the confusion of Babel. when he wants to say that, after its destruction, the builders of the Tower began to speak imperfect languages, he says that “tanto rudius nunc barbariusque locuntur,” (“the more rudimentary and barbaric the language they now spoke”) (DVE I, vii, 7, p. 14), while, a few lines down, referring to the original Hebrew language, the term used is “antiquissima locutione” (“the most ancient language”) (DVE I, vii, 8, p. 14).

It might be thought that he uses all these terms as synonyms, if it were not for the fact that ydioma, lingua, and loquela are used only when what he is talking about is a Saussurean langue, while it seems that locutio is used in a more generic sense and shows up whenever the context appears to be suggesting the activity of parole. Apropos of certain animal cries, for instance, he says that such an act cannot be called a locutio because it is not a true linguistic activity (DVE I, ii, 6–7). What’s more, Dante uses locutio every time Adam addresses God.

It would appear, then, that ydioma, lingua, and loquela are to be understood in the modern sense of “language,” while locutio seems instead to stand for discursive acts.

In DVE I, iv, 1, Dante wonders who was the first human being to be given the faculty of speech (locutio) and what was the first thing said (“quod primitus locutus fuerit”) and where, when, and to whom, and in what language (“sub quo ydiomate”) was the first linguistic act (“primiloquium”) emitted. I believe, incidentally, that we are entitled to translate “primiloquium” in this way, by analogy with “tristiloquium” and “turpiloquium” (DVE I, xi, 2; xiii, 4), used to describe the ugly manner of speaking of the Romans and the Florentines of his day.

Perhaps Dante wanted to stress the fact that Adam speaks to God before giving things their names, and that God had therefore given him the faculty of speech before he constructed a language.

But what language did Adam speak? Dante criticizes those who, like the Florentines, believe their own native language superior, whereas there exist many languages, and many of them are superior to the Italian vernacular. Next (DVE I, vi, 4), he concludes that, along with the first soul, God created at the same time a “certam formam locutionis” (“a certain form of language”). If we translate this expression as “a well-defined form of language” (as Mengaldo [1979: 55] does, how do we explain the fact that in DVE I, vi, 7 Dante states: “Fuit ergo hebraicum ydioma illud quod primi loquentis labia fabricarunt” [“So the Hebrew language was that which the lips of the first speaker moulded”]?

Dante explains that he speaks of forma “with reference both to the words used for things, and to the construction of words, and to the arrangement of the construction (“et quantum ad rerum vocabula et quantum ad vocabularum constructionem et quantum ad constructionis prolationem” [DVE, I, vi, 4]), allowing the inference that, by “forma locutionis” he is referring to a lexicon and a morphology, and hence to a language. But if we translate forma as “language,” the following passage would be hard to fathom:

And this form (forma) of language would have continued to be used by all speakers, had it not been shattered through the fault of human presumption, as will be shown below.

In this form of language (forma locutionis) Adam spoke; in this form of language spoke all his descendants until the building of the Tower of Babel (which is interpreted as “tower of confusion”); this is the form of language inherited by the sons of Heber, who are called Hebrews because of it. To these alone it remained after the confusion, so that our redeemer, who was to descend from them (in so far as He was human), should not speak the language of confusion but that of grace.

So the Hebrew language was that which the lips of the first speaker moulded. (DVE I, vi, 4–7)⁴

If we were to interpret “forma locutionis” as meaning a fully formed language, why then, in saying that Jesus Christ spoke Hebrew, does Dante use at one time lingua and at another ydioma (while, right afterward, in DVE I, vii, 7, recounting the episode of the confusion of tongues, loquela is the term chosen), whereas the expression “forma locutionis” is used only for the original divine gift? Furthermore, if we were to grant that “forma locutionis” signified only the faculty of speech, it is not clear why the sinners of Babel would have lost it (while the Hebrews kept it), seeing that the whole of the DVE recognizes the existence of a plurality of languages produced (on the basis of some natural faculty) after Babel.

Let us, then, attempt an alternative translation:

And it is precisely this form that all speakers would use in their language, if it had not been dismembered through the fault of human presumption, as we shall demonstrate below. This is the linguistic form in which Adam spoke: all of his descendants spoke thanks to this form until the building of the Tower of Babel—which is interpreted as the tower of confusion: this was the linguistic form that the sons of Eber, who were called Hebrews after their father, inherited. To them alone it remained after the confusion, so that our Redeemer, who was to be born of them through the human side of his nature, should enjoy, not a tongue of confusion, but a tongue of grace. It was, then, the Hebrew language that the lips of the first speaker framed.

What, however, is this linguistic form that is not the Hebrew language nor the general faculty of language and which was given to Adam as a divine gift but lost after Babel—and which Dante, as we shall see, is endeavoring to rediscover with his theory of the illustrious vernacular?

Corti (1981: 46 et seq.) has suggested a solution to the problem, based on the principle that Dante cannot be understood if he is seen simply as an orthodox follower of the thought of Thomas Aquinas. Dante appeals, depending on the circumstances, to various philosophical and theological sources, and there can be no doubt that he was influenced by various strands of that so-called radical Aristotelianism whose major representative was Siger of Brabant (whom Dante places in the Heaven of the Sun). But Boethius of Dacia too, one of the major representatives of the Modistae grammarians (and also in the Heaven of the Sun), was associated with the circles of radical Aristotelianism (and like Siger incurred the condemnation of the bishop of Paris in 1277). Dante is alleged to have been influenced by his De modis significandi. Corti sees the Bologna of his time as the seedbed from which these influences were passed on to Dante, either as a result of a personal stay there or through contacts between Bolognese and Florentine intellectual circles.

If such were the case, it would become clearer what Dante meant by “forma locutionis.” It was the Modistae who argued for the existence of linguistic universals, that is, for a set of rules underlying the formation of any natural language. In the De modis, Boethius observes that it is possible to extract, from all existing languages, the rules of a universal grammar, distinct from either Latin or Greek grammar (Quaestio VI).

What God gave to Adam, then, was not the mere faculty for language, and not even a natural language, but the principles of a universal grammar, the formal cause, “the general structuring principle of language both as regards lexicon and as regards the morphological and syntactic characters of language, which Adam will frame little by little, as he goes on living and giving names to things” (Corti 1981: 47).⁵ The forma locutionis given to him by God could be understood as a sort of innate mechanism reminiscent of the same universal principles studied in Chomsky’s generative grammar.

It seems likely, then, that Dante believed that, with Babel, what had disappeared was the perfect forma locutionis—the only form that would permit the creation of languages capable of reflecting the very essence of things (the identity between modi essendi and modi significandi), of which the Hebrew spoken by Adam was the incomparable and perfect result—and that the surviving formae locutionis were incoherent and imperfect—just like the Italian vernaculars whose inability to express lofty and profound thought is pilloried by the poet.

If this is how the DVE is to be read, we can finally understand the nature of that illustrious vernacular that Dante claims to be tracking down like a perfumed panther, “whose scent is left everywhere but which is nowhere to be seen” (DVE I, xvi, 1).⁶ It shows up here and there in the texts of the poets whom Dante considers major, but it still appears to be unformed, unregulated, unarticulated in its grammatical principles. Confronted with the existing vernaculars, natural but not universal, and with a universal but artificial grammar, Dante pursues the dream of a restoration of the Edenic forma locutionis, which is both natural and universal. Unlike many men of the Renaissance, however, who will go in search of a Hebrew language restored to its revelatory and magical powers, Dante’s goal is to recreate the original conditions with an act of modern invention. The illustrious vernacular is to be a poetic language, his language, and it will be the means by which a modern poet is able to heal the post-Babelic wound. The whole of the second book of the DVE is not to be read as a mere treatise on style, but as an effort to create the conditions, the rules, the forma locutionis of the only conceivable perfect language, the Italian of Dante’s poetry (Corti 1981: 70). This illustrious vernacular will possess the necessity (as opposed to the conventionality) of the original perfect language, because, just as the forma locutionis allowed Adam to speak with God, the illustrious vernacular will allow the poet to make his words equal to the task of expressing what they have to express, which would otherwise be inexpressible.

This is why, instead of condemning the multiplicity of languages, Dante stresses their ability to renew themselves over time. It is on the basis of this faith in the creativity of language that he can aspire to invent a modern perfect language, without going hunting for lost models. If Dante had really thought that the Hebrew invented by Adam was the only perfect language, he would have done all he could to write his poem in Hebrew. The only reason he did not do so is because he thought that the vernacular he was called upon to invent would correspond to the God-given principles of universal form better than Adam’s Hebrew had. Dante, with characteristic chutzpah, steps up to the plate as the new Adam.

7.2.

Paradiso

XXVI

If we turn now from the DVE to Canto XXVI of Paradiso (several years have gone by between the two), it looks as if Dante changed his mind. In the DVE it was unambiguously affirmed that Hebrew sprang as a perfect language from the God-given forma locutionis, and that was the language in which Adam addressed God, calling him El. In Paradiso XXVI, 124–138, however, Adam says:

La lingua ch’io parlai fu tutta spenta

innanzi che a l’ovra inconsummabile

fosse la gente di Nembròt attenta:

ché nullo effetto mai razïonabile,

per lo piacere uman che rinovella

seguendo il cielo, sempre fu durabile.

Opera naturale è ch’uom favella;

ma così o così, natura lascia

poi fare a voi secondo che v’abbella.⁷

Pria ch’i’ scendessi a l’infernale ambascia,

I s’appellava in terra il sommo bene

onde vien la letizia che mi fascia;

e EL si chiamò poi: e ciò convene,

ché l’uso d’i mortali è come fronda

in ramo, che sen va e altra vene.⁸

Adam affirms, not only that, born out of a natural disposition for speech, languages subsequently become distinguished from each other and grow and change thanks to human initiative, but also that the Hebrew spoken before the building of the Tower of Babel was no longer the same language that he had spoken in the Earthly Paradise. In Eden Adam had called God I, whereas later he was called El.

Saying that, by the time of the tower, Adam’s Hebrew was a lost language might simply be a way to justify Genesis 10. But what is most striking is the odd notion that God might once have been called I, a choice that none of Dante’s commentators has ever succeeded in explaining satisfactorily.

It has been suggested that I stands for the Roman numeral corresponding to the Arab numeral 1, and that it symbolizes therefore the perfect unity of God, but elsewhere in Paradiso (XIX, 128), the Roman numeral I stands for the smallest of quantities and is opposed to M, which stands instead for 1,000; it does not seem likely, then, that the poet would decide to designate the divinity with a numeral that indicates a minimal value.

A second interpretation appears to be inspired by a curious case of linguistic ethnocentrism—the conviction, that is, that there exists only one language and it is the most perfect one. The last thirteen cantos of Dorothy Sayers’s English translation of the Comedy were completed after her death by Barbara Reynolds. Lines 133–136 of Canto XXVI in Reynolds’s version read as follows:

Ere I descended to the pains of Hell

Jah was the name men called the highest Good

Which swathes me in this joy. Thereafter El

His title was on earth.…

Clearly, if Dante’s I had been preserved in the English, it might have been mistaken for the first person singular pronoun. It is understandable, then, that the translator should have changed it to Jah. We might be tempted to believe that Jah is simply the first syllable of Jahveh, if it were not for Reynolds’s footnote, which suggests that Dante must have been thinking of Psalm 68:4, which she naturally cites in the King James Version: “Sing unto God, sing praises to his name; extol him that rideth upon the heavens by his name Jah, and rejoice before him.”⁹

What makes this explanation “suspect”? The twin facts that unfortunately Dante did not know Hebrew and that neither was he especially conversant with the King James Version of the Bible.¹⁰ The Bible Dante knew was the Vulgate, in which the verse in question is translated as follows: “Cantate Deo psalmum, dicite nomen ejus, iter facite ei qui ascendit super occasum Deus est nomen illi. Exultate in conspectus ejus.” So the name of God Dante knew was Deus (for what it’s worth, Luther’s German translation also has, not Jah, but Herr).

For the same reasons we must exclude the hypothesis that Dante was influenced by Exodus 3:15, because in that case the Vulgate speaks of “Dominus Deus.” As for the theory that Dante may have taken his I from the frequently used abbreviated Florentine form i’ of the pronoun io—it is true that in Exodus 3:14 God says to Moses “Ego sum qui sum,” but what he is saying is that his name is “Qui sum,” in Hebrew Ehyieh.

There is yet another hypothesis. In the seventh book of his Etymologies, Isidore of Seville lists the traditional names of God in the Hebrew tradition and, along with El, Eloi, Eloe, Sabaoth, Elion, Eie, Adonai, Tetragrammaton, and Saddai, he also mentions Ia (“which is only applied to God, and which sounds as the last syllable of ‘alleluia,’ ” Etymologies, p. 153). But if Dante had followed Isidore, whom he certainly knew, why did he use I and not Ia? Certainly not for metrical reasons (the only consideration that could justify the abbreviation), since his hendecasyllabic line would have scanned correctly in either case.

The mysterious appearance of this I can only be explained by concluding that Dante had changed his mind about Adam’s original Hebrew, and that he had done so on the basis of information directly or indirectly acquired, just as we hypothesized that he had taken his idea of the forma locutionis from Modistae sources. We must therefore take one step, if not backward, at least to one side, and see what was happening at more or less the same time in Hebrew circles.

Let us take a look, then, at the principles of the Kabbalah of names, or ecstatic Kabbalah, theorized and practiced in the thirteenth century by Abraham Abulafia.¹¹

The Kabbalah of names is practiced by reciting the names of God hidden in the Torah, playing on the various combinations of the letters of the Hebrew alphabet. The so-called theosophical Kabbalah, while making occasional recourse to practices of numerological reading through acrostics or anagrams, remained basically respectful of the sacred text. The Kabbalah of names, on the other hand, alters, rearranges, dismantles, and recombines the surface of the text and its syntagmatic structures, all the way down to the linguistic atoms constituted by the individual letters, in a process of continuous linguistic re-creation. If, in the theosophical Kabbalah, the text still stands between God and the interpreter, in the ecstatic Kabbalah, the interpreter stands between the text and God.

The practice of reading by permutation tends to provoke ecstatic effects. As Abulafia himself says:

And begin by combining this name, namely, YHWH, at the beginning alone, and examine all its combinations and move it and turn it about like a wheel returning around, front and back, like a scroll, and do not let it rest, but when you see its matter strengthened because of the great motion, because of the fear of confusion of your imagination, and the rolling about of your thoughts, and when you let it rest, return to it and ask [it] until there shall come to your hand a word of wisdom from it, do not abandon it. Afterwards go on to the second one from it, Adonay, and ask of it its foundation [yesodo] and it will reveal to you its secret [sodo]. And then you will apprehend its matter in the truth of its language. Then join and combine the two of them [YHWH and Adonay], and study them and ask them, and they will reveal to you the secrets of wisdom.… Afterwards combine Elohim, and it will also grant you wisdom. (Hayyê ha-Nefes, as cited in Idel 1988b: 21)

If in addition to this we consider the breathing techniques that are meant to accompany the recitation of the names, we can see how the adept may progress from syllabification to ecstasy and thence to the acquisition of magical powers, because the letters the mystic combines are the same sounds with which God created the world. This aspect will become still more evident in the fifteenth century. Apropos of Yohanan Alemanno, the friend and inspirer of Pico della Mirandola, Idel (1988a: 205) remarks: “the symbolic cargo of language was transformed into a kind of quasi-mathematical command. Kabbalistic symbolism thus turned into—or perhaps returned to—a magical language of incantation.”

All of this was possible because for Abulafia the atomic elements of the text, its letters, had meaning in and of themselves, quite apart from the syntagmata in which they occur. Every letter is already a divine name: “since for the letters of the Name each letter is a Name in and of itself, be aware that the Yodh is a name and YH is a name” (Perush Havdalah de-Rabbi ‘Akiva).

The notion that the name of God can be expressed by a single letter of the Tetragrammaton is also confirmed by the way in which the divine name is written in many manuscripts. I am referring to Perani and Sgradini (2004: 131–143), where we see that it was the custom in medieval Hebrew texts to represent the divine name with a calligraphic arrangement of a series of three or four yodhs. The fact that these manuscripts were produced in an Italian context encourages us to entertain the hypothesis that Dante was aware of this tradition.

If we transliterate the yodh to an I, as Dante may have done, we have a possible source for the poet’s volte-face. But this notion of the divine name is not the only idea that Dante seems to share with Abulafia.

According to the ecstatic Kabbalah, language is a universe unto itself, and the structure of language reflects the structure of reality. Already Philo of Alexandria had attempted in his writings to compare the intimate essence of the Torah with the Logos, the World of Ideas, while at the same time Platonic concepts had filtered into the Haggadic-Midrashic tradition, in which the Torah was perceived as the schema according to which God had created the world. The eternal Torah was therefore identified with Wisdom and in a number of passages with a world of forms, a universe of archetypes. In the thirteenth century, adopting an unmistakably Averroistic approach, Abulafia will postulate an equation between the Torah and the Active Intellect, “the form of all the forms of the separate intellects” (Sefer Mafteah ha-Tokhahot).

Nevertheless, for Abulafia, this matrix of all languages (which is one and the same as the eternal Torah, but not necessarily the written Torah) does not yet coincide with Hebrew. It appears that Abulafia makes a distinction between the twenty-two letters (and the eternal Torah) as matrix and Hebrew as the mother tongue of humankind. The twenty-two letters of the Hebrew alphabet represent the ideal sounds that must preside over the creation of each of the seventy other languages in existence. The fact that other languages have a larger number of vowels is a result of variations in the pronunciation of the twenty-two basic letters (the other foreign sounds would be called, in modern linguistic terms, allophones of the basic phonemes).¹²

For Abulafia the twenty-two letters represent all of the sounds naturally produced by the organs of phonation. It is the way the letters are combined that makes the creation of the different languages possible. The word zeruf (combination) and the word lashon (tongue) have the same numerical value (386): to know the laws of the combinatorial system is to possess the key to the formation of every language. Abulafia admits that the choice of representing these sounds by certain graphic signs is a matter of convention, but he speaks of a convention established between God and the prophets. He is perfectly familiar with the current theories of language according to which the sounds for certain things or concepts are conventional (because he found this Aristotelian and Stoic idea in authors like Maimonides), but he seems to overcome his embarrassment with a rather modern solution, implicitly distinguishing between conventionality and arbitrariness.

Hebrew had its origin in convention like all languages (Abulafia rejects the idea, endorsed by other scholars, some of them in the Christian camp, that a child left to itself from birth would automatically speak Hebrew), but Hebrew is still the Holy Mother Tongue because the names given by Adam were in accord with nature and not chosen arbitrarily. In this way, Hebrew was the protolanguage, and as such it was necessary if all the other languages were to be created, for “if such a language did not precede it there couldn’t have been mutual agreement to call a given object by a different name from what it was previously cald, for how would the second person understand the second name if he doesn’t known the original name, in order to be able to agree to the changes?” (Sefer ‘Or ha-Sekhel, in Idel 1989: 13–14).

Abulafia bemoans the fact that, during the course of their exile, his people have forgotten their own original language, and his project is naturally for the Kabbalist to work toward the recovery of the true matrix of all seventy languages. It is the Messiah who will finally reveal the secrets of the Kabbalah, and the differences between languages will cease at the end of time, when all the existing languages will be reabsorbed into the Sacred Language.

Once again, we find that the positions of Abulafia and Dante have something in common. For Abulafia there existed an equation between the Torah and the Active Intellect, and the schema according to which God had created the world coincided with the linguistic gift he had given to Adam, a kind of generative matrix of all languages that did not yet coincide with Hebrew. We find, then, on the one hand, Averroistic influences in Abulafia, which lead him to believe in a single Active Intellect common to all mankind, and, on the other, undeniable and proven Averroistic sympathies in Dante—the conception, for instance, that Nardi (1985: ch. V) sees as having its origins in Avicenna and Augustine, of a divine Wisdom that offers its forms to the Possible Intellect. The Modistae too (especially the group based in Bologna) and other defenders of a universal grammar were no strangers to the Averroistic tradition. So here we have a shared philosophical position that (without insisting on a direct influence) might incline both thinkers to consider the gift of tongues as the handing-down of a forma locutionis, a generative matrix comparable to the Active Intellect.

This is not all. Historically speaking, Hebrew, for Abulafia, had been the protolanguage, but the Chosen People, in the course of the Diaspora, had forgotten that original language. Therefore, as Dante will say in the Paradiso, at the time of the confusion of Babel the tongue of Adam was “all extinct.” Idel (1989) quotes an unpublished manuscript by a disciple of Abulafia which says:

Anyone who believes in the creation of the world, if he believes that languages are conventional, he must also believe that they are of two types: the first is Divine, i.e., agreement between God and Adam, and the second is natural, i.e., based on agreement between Adam, Eve and their children. The second is derived from the first, and the first was only known to Adam and was not passed on to any of his offspring except for Seth, whom he bore in his likeness and his form. And so, the tradition reached Noah. And the confusion of the tongues during the generation of the dispersion [at the tower of Babel] occurred only to the second type of language, i.e., to the natural language. (Idel 1989: 17)

If we bear in mind that the term “tradition” refers to the Kabbalah, then the passage quoted is once more alluding to a linguistic knowledge, to a forma locutionis as a collection of rules for the construction of different languages.

If the original form is not a language, but the universal matrix of all languages, this confirms the historical mutability of Hebrew itself, but also the hope that that original form can be rediscovered and made to bear fruit once again (in different ways, obviously, for Dante and Abulafia).

All these remarks would make more sense if it could be demonstrated that Dante was familiar with Hebrew Kabbalistic thought and with Abulafia in particular.

Abulafia had come to Rome in 1260 and had stayed on in Italy until 1271, when he returned to Rome with the idea of converting the pope. Then he continued on to Sicily, where we lose track of him toward the end of the 1290s. His ideas, then, undoubtedly influenced Italian Jewish circles. In fact in 1290 we witness a debate between Hillel of Verona (who probably met Abulafia twenty years earlier) and Zerakhya of Barcelona, who arrived in Italy at the beginning of the 1270s (cf. Genot-Bismuth 1988: ch. II). Hillel, who had been frequenting intellectual circles in Bologna, writes to Zerakhya with a question first broached by Herodotus, that is, what language would a child brought up deprived of linguistic stimuli express itself in? For Hillel (who appears not to be aware, or chooses to ignore, that Abulafia had been of a different opinion), the child would express itself in the same Hebrew that had originally been given to man as part of his very nature. In his reply Zerakhya accuses him of having surrendered to the sirens of the “uncircumscised” of Bologna. The sounds produced by a child who had not been exposed to a linguistic education, he objects, would be similar to the barking of a dog, and it cannot be argued that the sacred language was given to man by his very nature, because the aptitude man possesses for language is merely potential, and the only way he learns to speak is through the education of his phonatory organs.¹³

This exchange is sufficient to demonstrate that Abulafia’s themes were debated on the Italian peninsula, to be precise in the same Bolognese circles that influenced Dante (and where, according to Corti (1981), he might have picked up a number of ideas concerning the forma locutionis). But the research of Genot-Bismuth supplies additional details about the period, in which we encounter a certain Yehudi Romano, who lectured to his coreligionists on Dante’s Comedy, or Lionello di Ser Daniele who will do likewise, using a copy of the Comedy transliterated into Hebrew, to say nothing of a figure like Immanuel of Rome who, in his own poetic compositions, seems almost to parody Dante’s themes, as if he were nursing the ambition of writing an anti-Comedy in Hebrew.¹⁴

We are not talking simply about the influence of Dante on Italian Jewish circles. Genot-Bismuth proves that the influences went both ways, going so far as to posit a Jewish origin for the theory of the four senses of Scripture mentioned in Dante’s Epistle XIII—a bold thesis, when we think of the abundance of Christian sources Dante had at his disposal on that subject. Far less extravagant, and in many ways more convincing, is her thesis that Dante may have caught echoes of the Hillel-Zerakhya polemic in Bologna in the years following.

We might conclude that in the DVE Dante comes close to the position espoused by Hillel (or by Hillel’s Christian inspirers, as Zerakhya suggested in his rebuttal), while in Paradiso XXVI he has become converted to Zerakhya’s thesis, which was also that of Abulafia—though it is also true that, by the time he came to write the DVE, Dante could already have been familiar with both opinions.

Though Genot-Bismuth is able to document in detail a number of Jewish contributions to historiography that would appear as echoes and suggestions in the De regimine principium of Giles of Rome, it is enough for our purposes to recognize the existence of an intellectual climate in which ideas circulated as part of a constant polemic, made up of written and oral debates, between Church and Synagogue (cf. Calimani 1987: ch. VIII). Assuredly if, before the Renaissance, a Christian thinker had come close to embracing Hebrew doctrine, he would never have admitted it publicly. Like the Christian heretics, the Jewish community belonged—as Le Goff (1988) cogently puts it—to a category of outcasts that the official Middle Ages seemed to detest and admire simultaneously, with a mixture of fascination and fear, keeping them at a distance, but making sure the distance was close enough for the outcasts to be within reach. “What it called its charity towards them was like the attitude of a cat playing with a mouse” (Le Goff 1988: 316).

Before its rehabilitation by the Humanists, Christian notions of the Kaballah were hazy, and it tended to be lumped together with the black arts. On the other hand, it has been suggested (Gorni 1990: ch. VII) that Dante refers a little too insistently to various divinatory and magical arts (astrology, chiromancy, physiognomy, geomancy, pyromancy, hydromancy and, of course, necromancy). He appears to have been somehow familiar with an underground and marginalized culture of which the Kabbalah was confusedly a part, at least in popular opinion.

Thus, the interpretation of the forma locutionis as a universal matrix of language, even without referring it directly back to the Modistae, becomes still more persuasive.

The only drawback is that, in the absence of concrete proof of these contacts, this is all merely conjecture—as Busi (2004) pointed out in his review of Debenedetti Stow’s (2004) book on Dante and Jewish mysticism, for whom the hypotheses we have just set forth are the object of passionate conviction, a conviction that results in her treating a number of hypotheses as if they were proofs.

Still, when all we have to work with are the texts, certain textual analogies, while they cannot be taken as irrefutable proofs, deserve nevertheless to be stressed, if for no other reason than to encourage further research.

We may close by imagining that, on his journey to Paradise—the one he took post mortem, not his literary journey—Dante actually met Abulafia. They may even be conversing amiably together at this very moment, smiling indulgently perhaps at the efforts we have been making to discover whether they had anything in common. And if at a certain point Adam too were to join in the conversation, it would be fascinating to discover what language the three of them were using to make themselves understood. But since the present author is somewhat skeptical concerning the existence of a Perfect Original Language, he prefers to think that the Angels will no doubt provide state-of-the-art simultaneous translation.

A reworking of “Forma locutionis” published in Vattimo (1992), which also appeared in English, with the title “The Perfect Language of Dante,” in Eco (1995) and, in a somewhat different form, as “Languages in Paradise,” in Eco 1998b.

1. Or again: “Nihil est enim aliud loqui ad alterum, quam conceptum mentis alteri manifestare” (“For to talk to someone else means precisely to make known one’s thoughts to them”) (Summa Theologiae I, 107, 1.). This idea of a relationship with another person or persons reappears in various other authors.

2. Biblical quotes, here and elsewhere, are from the King James Version.

3. “Oportuit ergo genus humanum ad comunicandum inter se conceptiones suas aliquod rationale signum et sensuale habere; quia, cum de ratione accipere habeat et in rationem portare, rationale esse oportuit; cumque de una ratione in aliam nichil deferri possit nisi per medium sensuale, sensuale esse oportuit; quare, si tantum rationale esset, pertransire non posset; si tantum sensuale, nec a ratione accipere, nec in rationem deponere potuisset. Hoc equidem signum est ipsum subiectum nobile de quo loquimur: nam sensuale quid est, in quantum sonus est; rationale vero, in quantum aliquid significare videtur ad placitum” (DVE I, iii, 2–3). “So it was necessary that the human race, in order for its members to communicate their conceptions among themselves should have some signal based on reason and perception. Since this signal needed to receive its contents from reason and convey it back there, it had to be rational; but, since nothing can be conveyed from one reasoning mind to another except by means perceptible to the senses, it had also to be based on perception. For, if it were purely rational, it could not make its journey; if purely perceptible, it could neither derive anything from reason nor deliver anything to it. This signal, then, is the noble foundation that I am discussing, for it is perceptible, in that it is a sound, and yet also rational, in that this sound, according to convention, (ad placitum) is taken to mean something” (Dante 1996: I, iii, 2, p. 7. Subsequent citations in English are from Steven Botterill’s translation (Dante 1996). Botterill’s facing Latin text is based on the critical text established by Mengaldo (1979).

4. “Qua quidem forma omnis lingua loquentium uteretur, nisi culpa presumptionis humane dissipata fuisset, ut inferius ostendetur. Hac forma locutionis locutus est Adam; hac forma locutionis locuti sunt omnes posteri eius usque ad edificationem turris Babel, que ‘turris confusionis’ interpretatur; hanc formam locutionis hereditati sunt filii Heber, qui ab eo dicti sunt Hebrei. Hiis solis post confusionem remansit, ut Redemptor noster, qui ex illis oriturus erat secundum humanitatem, non lingua confusionis, sed gratie, frueretur. Fuit ergo hebraicum ydioma illud quod primi loquentis labia fabricarunt” (DVE I, vi, 4–7).

5. Maria Corti’s thesis has been challenged, especially by Pagani (1982) and by Maierù (1983): (i) there is no convincing proof that Dante knew Boethius of Dacia’s work, (ii) in a number of instances Corti draws untenable analogies between the two texts, and (iii) the linguistic notions we find in Dante were already circulating among other philosophers and grammarians even before the thirteenth century. Even if we grant the first two points, however, there still remains the third, that the idea, that is, of a universal grammar enjoyed wide circulation in medieval culture and, as none of Corti’s critics has placed in doubt, Dante was familiar with these discussions. To say, as Maierù says, that there was no need to be acquainted with Boethius’s writings to know that “grammar is one and the same in all languages, even though there may be surface variations,” because the same affirmation is already to be found in Roger Bacon, is if anything proof that Dante could indeed have been thinking of a universal grammar.

6. It was thought in the Middle Ages that the panther had a richly perfumed breath and left a trace of its passage wherever it had been. But, for the hunters who attempted to entrap it, it was practically impossible to locate. So they would smell its perfume but never succeed in catching it. This explains how the panther became a metaphor for poetry itself.

7. See in Marmo (1994: 124, n. 39) the interesting reference to Simon of Faversham, who claimed that there exists in language a difference between natural signification and positive (or conventional) signification.

8. “The tongue I spoke was all extinct before Nimrod’s race gave their mind to the unaccomplishable task; for no product whatever of reason—since human choice is renewed with the course of heaven—can last forever. It is a work of nature that man should speak, but whether in this way or that nature then leaves you to follow your own pleasure. Before I descended to the anguish of Hell the Supreme Good from whom comes the joy that swathes me was named I on earth, and later He was called El; and that is fitting, for the usage of mortals is like a leaf on a branch, which goes and another comes” (Dante 1961: 379).

9. Even a contemporary Hebrew scholar like André Chouraqui translates: “Poétisez pour Elohim, chantez son nom; frayez passages au chevalier des nues: Yah est son nom! Exultez en face di lui!”

10. I am reminded of that nineteenth-century congressman from Texas who opposed the introduction of foreign language teaching in the schools declaring: “If English was good enough for the Lord Jesus Christ, it’s good enough for me!”

11. All my information about Abulafia and the quotations that below come from Idel (1988a–c, 1989).

12. Other Kabbalists point out that Christians are lacking the letter heth and the Arabs do not have the pe; and in the Renaissance Yohanan Alemanno will be of the opinion that the variations in pronunciation with regard to the twenty-two Hebrew letters are comparable to the sounds made by the different animals (some are like the grunt of a pig, others like the croak of a frog, others still like the honking of a crane). So that the very fact that they produce different sounds reveals that the other languages belong to peoples who have abandoned the proper conduct of life. In this sense, the multiplication of letters is considered to be one of the results of the confusion of Babel. Alemanno is aware of the fact that other peoples have recognized their own languages as the best in the world, and he cites Galen, for whom the Greek tongue is the most pleasing and the most respondent to the laws of reason, but, not daring to contradict him, he admits that this is because there are affinities between Greek, Hebrew, Arabic, and Assyrian.

13. Zerakhya uses a proof that we shall encounter after the Renaissance in other, Christian authors—cf. Brian Walton, In Biblia polyglotta prolegomena (1657) or Francisco Vallesio, De sacra philosophia (1652)—if the divine gift of an original sacred language had ever been made, every human being, no matter what their mother tongue, would have to have an innate knowledge of the sacred language as well.

14. See Romano (2000). Cf. Battistoni (1995, 1999).