III. Accents

This web page concludes our summary of the linguistic foundations of the Divina Commedia. Accents define the pulse, the rhythm of verse, and are among the central, and perhaps most elusive, aspects of the Italian language. Their linguistic structure has plunged scholars into fierce controversies that are still not fully resolved or understood.

As a reward for the challenging reading, I can promise you that you will find something here that is not available elsewhere: a new arsenal of possibilities of accentuation that can perhaps better, more usefully, more musically absorb and express the stream of sound of the great poem.

For in order to do justice to the impressive rhythmic complexity in Dante's creation, it was necessary to expand the no longer sufficient classical accent scheme of canonical stresses. My edition introduces - this would be its scientific contribution - a three-valued accent logic, consisting of strong and weak accents as well as of alternative, optional and inclusive stress sequences. The combination of these components allows you to model real-world speech possibilities and enables you to be creative in your own reading.

With this third web page we conclude our summary of the linguistic basics of the Divina Commedia. You now have the tools of a competent recitation of this work, which on the one hand respects the strict specifications of the endecasillabo, but on the other hand also retains its particular liberties. And is it not precisely in this change, which can no longer be narrowed down, that the special charm of our poetry lies?

Should you wish to educate yourself further, have a look at the webpage Reading or pick up one or another of the books in our final selection of literature.

1. Basic definition of the Endecasillabo and project of its canonization

In order to understand the special fascination of accents and their importance for Dante's poetry, it is necessary to look a little deeper into the verse meter of the Divina Commedia, the endecasillabo (hendecasyllable), and its particular metre. As is well known, it forms every line of our poetry, with the additional presence of the terza rima, an end rhyme concatenation, which, however, need not concern us further here.

The endecasillabo has as a minimum requirement a single structural condition that applies without exception: it is stressed on the tenth syllable. In its basic form, which is given when the last word is stressed on the penultimate syllable, it has eleven metrical syllables.[30] In principle, then, the other syllables could be accented ad libitum. Soon, however, it became apparent that certain accent combinations occur more frequently than others, and therefore could be considered as an additional defining feature. Thus was born the project of canonization, the description of those very basic patterns of the endecasillabo, and finally the broader postulate that all verses of genuine poetry had to conform to this arrangement.[31]

Even if the attempt at canonization must ultimately be considered a failure, i.e., it is not possible to incorporate all verses into the system, we will nevertheless briefly introduce it, on the one hand to show its value, and on the other hand to demonstrate its limitations. We are then ready to proceed, overcoming the rigid dictate, to the actual and, as we will see, original rhythmic variety of the endecasillabo.

2. The canonical Endecasillabo

One may assume a canonical endcasillabo if, in addition to the ictus[32] already known to us, on the tenth syllable the fourth and/or the sixth syllable are stressed, which is equivalent to the fact that both must not be unstressed. In this case, one would have a non-canonical endecasillabo.

The canonical system now distinguishes two basic types, the endecasillabo a maiore and a minore, the latter also branching again. The endecasillabo a maiore is stressed on the sixth syllable, the endecasillabo a minore on the fourth syllable. In the more common variant, it has a further stress on the eighth syllable (4-8). In the rarer "dactylic" variant, however, the seventh syllable is stressed (4-7). Apart from these basic conditions, the rest of the syllable distribution is essentially free (Beltrami 2002: 182). The accentuation of both the fourth and the sixth syllable obviously serves both basic types of the endecasillabo, the a maiore as well as the a minore, but this did not worry linguists much, since it could be left to the reciter how he wanted to understand or read such a verse.

A milestone in the further development of the canonical scheme is certainly the work of Bertinetto (1973), which still points the way forward today. The author also includes marginal positions beyond the central syllables 4 to 8 and presents a more comprehensive typology of the endecasillabo. It has proven itself, with minor modifications, in the analysis of other classical works and represents the system valid today.[33] This taxonomy should be supplemented by the phenomenon of the counter accent (also Scontro di Ictus or Accento Ribattuto), already known before.[34] Here two adjacent syllables are stressed, often the syllables 6 and 7, but other combinations are conceivable and realized in the Divina Commedia.

With this last refinement, the model of permissible canonical variants has reached its limit, if it does not want to render itself completely useless by further ramifications. It consists of a list of the essential verse forms of the endecasillabo and the confidence to have almost completely represented the poetic reality. In fact, however, reality is more complex. For the system, regardless of its quality, has two problem areas, at least one of which can no longer be eliminated by even the most refined adaptation.

3. Problems of the canonical system

A. Lack of codification and irreducible existence of non-canonizable verses

Let us begin with the minor problem. We noted above that the vast majority of Dante's verses are reliably mapped in the canonical system, and we have also noted the modern improvements to this project. Nevertheless, a drop of value remains here as well. For while in the basic definition of the endecasillabo a maiore or a minore, there was basically a free choice of accents in addition to 4, 6, 7, or 8, the modern model concretized these variants and thus abolished the complete freedom again. Thus, for example, the endecasillabo a maiore with main accent on the 6th syllable has, depending on the secondary accent at the beginning of the verse, the subforms 1.6.10, 2.6.10 and 3.6.10. The same applies to the alternative version with additional accent on the 8th syllable. Here we have the variants 1.6.8, 2.6.8, 3.6.8 as well as 1.3.6.8. In the classical endecasillabo a minore we note with the main form 4.8.10 the sub-variants 1.4.8.10 as well as 2.4.8.10. And for the "dactylic" endecasillabo a minore with second stress on syllable 7, we get the main variant 4.7.10 as well as the sub-variants 1.4.7.10 and 2.4.7.10. We find similar restrictions for those variants of the endecasillabo which are stressed on syllables 4 and 6.

We have not allowed ourselves this elaborateness in order to scare off the reader, but to show that here there is no complete additional freedom of accents, but a limited one, which means that variants in which other accents or accent sequences occur are missing: for example, the stressing of the fifth syllable, the stressing of the 9th syllable, or the counter accents.

If we now take a look at the concrete distribution of accents in the Divina Commedia, we will notice many of these special variants, indeed a fascinating play of accents and accent sequences, which simply does not fit into such a comparatively clear scheme even in its final elaboration. The reason for this may be that at the time of Dante there was no explicit codification of the endecasillabo or its basic types, which gave the poet more freedom than later authors.[35] Perhaps it is simply the case that Dante consciously evaded this structural-normative narrowness, that he wanted to exhaust the linguistic possibilities, more precisely: the permutations of the endecasillabo, and that he took pleasure in the change of rhythm.

Of course, one may use as an excuse purely metrical reasons here as well, i.e., the necessity of accommodating a verse within the strict framework of the endecasillabo or terza rima, but these are our reasons. We have no cause to doubt that Dante shaped the setting of the accents and thus the rhythm of each line exactly as he wanted. We would ascribe the special charm of this variation, indeed the force of the unusual, which follows from these verse sequences, to a mere technical necessity, and thus belittle the poet's linguistic ability.

Nevertheless, one might object that an improved, more flexible canonical system, for example one that introduces even finer exception clauses, would have to be able, at least in principle, to accomplish the task sought.[36] The second problem, however, is of a fundamental nature and can no longer be cured.

B. The Problem of Performance (of the Lecture)

It consists in the fact that the two-valued accent scheme, which distinguishes stressed from unstressed syllables and does not allow any further differentiation, does not properly represent the real flow of speech. Thus, listening to professional recitations of the Divina Commedia reveals that accentuation nuances, i.e., subtle volume differences between accents, escape the bipolar system. One was now accustomed to place these subtleties in the freedom of the performer, i.e., to regard them not as a genuine primary feature of metre, as inherent in the verse itself, but as a subordinate, accidental moment of the esecuzione, the performance, or the elocuzione, the manner of speaking. This led to the result that in borderline cases of weaker accentuation, an accent was once placed and once not. In the first case, the accent appeared overprominent; in the second case, it was absent. In addition, there were constellations in which a strong accent could be set but did not have to be set, i.e., both options were permissible. What was observed for single accents also applies - as we will see shortly - to accent combinations. Here, too, inclusive, exclusive as well as optional variants are conceivable.

A true representation of the concrete performance should capture this reality and thus express that the variance is already inherent in the verse, in its semantic and poetic structure, and is not added secondarily by the performer.

4. The new edition - our contribution to an understanding of the accents.

We have undertaken - this would be the scientific contribution of our text - such a mapping. It is the first attempt at a more realistic documentation of the accent structures expressed in poetic recitation, i.e. in spoken Italian, and may perhaps serve as a basis for further research.

Let us begin with the strength of the accents and then come to the problem of canonization. Presumably, it would be possible, through a precise study of the concrete pronunciation, to establish several levels of accentuation depending perhaps also on their prolongation, and thus, similarly to music, to fix the dynamics and duration of the phoneme.

In (classical) music - we remember - the composer generally determined the tone duration precisely, but either left the volume undetermined or marked it in gradations from pianissimo to fortissimo. This usually involves longer tone phrases. A sole note may be singled out from the tonal context with a sforzato or subito piano.

In language, the length of a word is initially determined lexically. That means we know that the a in casa for example is spoken long, the a in cassa short. A intensity specification, however, is missing. We do know the accent, as in càccia, but this does not imply any additional intensity information. At best, as in città, it primarily marks the position, i.e. here the emphasis on the last and not the usual penultimate syllable.

We talked about the possibility of developing a system that would enter the volume of an accent into the text in a more differentiated way and - for purists - the length of the syllable, but here we are confronted with practical difficulties that can hardly be solved. How to design such a refinement graphically and how to realize it without restricting too much the performer, who still needs freedom also with regard to dynamics and temporal extension?

The solution we propose should be practical, realistic, flexible and instructive. It must not overwhelm the learner, i.e. it must be easily to recognize and implement. It should also better reflect the reality of the language, i.e. provide an additional, intermediate nuance to the classical system of accentuation. It is to leave to him still sufficient scope of interpretation. And it wants to inform him last. It should show him the basic structure of accents in a verse and within which limits (constraints) the neutral, i.e. in its style not consciously exalted performance may unfold.

Thus, it does not deprive the speaker of interpretative freedom; rather, it gives him the knowledge of the tonal centers, the accent structure of the verse, and thus the linguistic competence to deviate ad libitum from our system as well.[37]

A. The trivalent accent logic: strongly and weakly accented accents

To achieve this goal, it was necessary to abandon the two-valued accent logic and replace it with a three-valued system. I distinguish strongly emphasized accents from weakly emphasized accents and unstressed syllables. This three-valued accent logic is easy to grasp and yet amazingly effective in breaking down the real dynamics of our endecasillabo. For the reader thus has before him quite clearly those strong main accents which are usually clearly stressed. He knows where weaker secondary accents are, which are still audible but subordinate to the main accents, and he knows those syllables - the old system had already achieved this - which are not to be stressed. In this way, our system intervenes in the text certainly in a determining way, but it still leaves enough shades of accentuation for the reader, for example in volume, speed or pitch.

The problem of over- or under-emphasizing accents in the middle volume range is thus eliminated insofar as these accents now receive their own marking and do not fall victim to the bipolar decision constraint, i.e., are either marked too strongly or not at all.

B. Weakly emphasized accents versus secondary accents

We have to get one possible misunderstanding out of the way. Weak accents are not the so-called secondary accents that have occupied linguistics intensively. Secondary accents are weaker markings within a word that already has a primary accent. We follow Bertinetto's assessment (1981), who rejects secondary accents or does not recognize them as a linguistic structural element essential to Italian.[38] Dolceménte, for example, is therefore only emphasized on the middle e and not additionally weaker on the o.

Consequently, each word has only one accent. Our weak accents are therefore the only accents of a word, which means that a word can have either a strong or a weak accent or no accent at all, i.e. is unaccented. This expresses the observation that from a purely grammatical point of view each word has an accent, i.e. a stress center - in uscìto, for example, this would be the i - but that this accent can also appear unstressed in the concretely pronounced context of the sentence, for example in the case of "uscito fuòr", where the uscito is read quickly and without emphasis, while the fuor is strongly stressed on o.

We furthermore spoke above of optional accents and of accent combinations of an alternative, exclusive or inclusive nature. Here, too, we want to offer a solution.

C. Alternative accents

Alternative accents are those in which the reader can place a particular accent or the one that follows it. Thus, of two successive accents, one is stressed but the other is unstressed. Most often, these are neighboring syllables. The alternate accents can occur in stressed or weakly stressed form, and very rarely there is a dynamic mixture of the two. It is important that only one of the two accents is emphasized. Logically, this is the exclusionary or (A or B). In our edition, this variant is indicated by a slash between the accents.

D. Inclusive accents

A bit more complex are the inclusive accents. Here, too, we are dealing with two successive centers of stress, mostly adjacent syllables. In contrast to the alternative accents, however, the accentuation of both syllables is as well possible. So the reader can stress the first and not the second, the second and not the first, and finally both. And again, it is no trouble to recognize immediately that we have logically instantiated the inclusive or (A and or B). A dot between the accents marks this variant.

E. Optional (optional) accents

But what do we do in cases where the speaker wants to either emphasize or not emphasize a single accent or a well-defined accent combination? Here we are dealing with an optional accent or an optional resp. facultative accent group. So the reciter can emphasize this element or dispense with it. A bracket marks this variant. We add it either around a single accent or an accent group, which is then completely read or completely not read.

One question suggests itself to us. What happens in the case of a strong accent? Does the freedom of choice lead to its complete omission or does it lead to the accent turning into a weak accent? We have left this decision open. It should be enough to know that the strong accent in question can be presented in a reduced dynamic or not emphasized. Its dominance as a strong accent is therefore not an unconditional one. The extent of the restriction establishes itself in the concrete execution. Here it may well be conceivable that between different speakers three forms of emphasis and two of reduction are possible.

F. Advantages of the new system

a. Accent dynamics and flexibility

What have we achieved through our scheme? We have mapped the real flow of words, the concrete recitation, insofar as for the first time we have captured the different dynamics of accents more systematically and highlighted their importance. We have also described variations in the recitation. They take into account that individual accents or groups of accents are pronounced with different distinctiveness by different people and can also be omitted. These is not mere arbitrariness, i.e., accidental experiments by the individual, but possibilities inherent in the language itself which, to that extent, reflect its structure and freedom. A verse thus possesses the potential of different modes of articulation: this plurality is part of its substance, its linguistic determination. It is inherent in it and not a secondary feature.

Nevertheless, we are speaking only of meaningful main forms of performance unique to Italian, of basic structures rooted in practice. Language and interpretation can and may express themselves beyond these possibilities. For alienation and complex reshaping is part of interpretive freedom, part of what is permitted in art. It can, if it is done knowingly, have a forceful effect. Our system is intended to convey this knowledge of the rhythmic center, or more precisely: of plausible rhythmic centers, and thus to invite to experiment in one direction or another. It does not patronize, it informs. The rest is art.

b. Impact on canonization

However, we also make significant progress in terms of canonization. With the addition of weak accents, more accent combinations are conceivable and also those that are usually readily omitted. It surely is possible to accent all syllables of an endcasillabo from 1 to 9 even with the conventional system. Often, however, with secondary syllables of the positions 1 to 3 or especially 9, which is close to the always stressed 10, this designation is omitted, since one does not trust these accents the full stressing power. Our system includes them individually or in combination with other accents. Thus, non-canonical accent formations appear more frequently, i.e. the verse structure of the Divina Commedia shows its real diversity.

I firmly believe that the canonical system cannot capture the metrical plurality of this work. Certainly it is a first, important fixing, and certainly the scheme retains its value in due relativization in the final expansion of a Bertinetto or his successors. However, it does not exhaust the concrete possibilities. Ultimately, the endecasillabo of Dante is free. Ultimately, it possesses an almost unlimited permutation variance. And precisely in this freedom lies its beauty and greatness.

For Dante, unlike later authors such as Petrarch in particular, was able to shape and form his language himself. He was not subjected to those standardizations that later on were offered to the verse. It is not, as some write, that Dante goes to the limit of the rule or breaks it with every verse. Rather, within the minimum requirement of the stressed tenth syllable and the central possibilities of emphasizing the fourth, sixth, seventh, and eighth syllables, he allows himself a grandiose play of dynamics that creates pleasure in alternation, in the unpredictable, and in the particular.

The power of this poetry does not derive purely from the reproduction of the system, but from the almost unlimited freedom, the discovery of ever new possibilities, then again in the return to the same canonical principle, only to leave it again in the next verse. This freedom is the very essence of the endecasillabo. Only this particular verse creates this freedom, and perhaps this is also a reason why Dante chose it. Because in this freedom is expressed what the poet wanted to realize in the work itself: the freedom and greatness of poetry, of the human spirit, and finally of human existence. And perhaps it is the vehemence and unconditionality with which Dante realizes this aspiration that have captivated us to the work for centuries and never let go.

G. Procedure

Our edition thus concludes with a marking of the accents. To do this, it was necessary to decipher the text read by six native speakers.[39] Also, two databases, that of Robey (2003) and that of the Padova Group (AMI) came to my aid.[40] First, I scanned the verses of each canto individually, i.e., I made an unbiased initial division of the accents. Then I compared these in fixed order with six audio recordings and extracted those variants that made sense to me. In parallel, I checked my result against the two databases mentioned above. Such work took months of time, but the result was deeply satisfying. For not only did the project lead me deeper into the secret of our great poet's accents, rhythm, and speech patterns. It also turned out that the components I introduced were able to effectively represent the real possibilities of recitation. This is because the combination of the parameters results in an enormous plurality of designation figures, through which even more complex structures can be successfully represented. However, since this plurality is based on a few basic symbols, it still remains comprehensible.

H. Result

The reader finally has before him those basic forms of plausible recitation extracted from and describing cultivated speech. It is a musical designation, since nuances of volume, of pointing are also involved, and thus the flow of the recitation becomes natural, more melodic. They are, invoking a famous metaphor of Otto Neurath, a ladder that leads to knowledge and then, when knowledge is achieved and artistic intuition takes its place, can be thrown aside.

To the right of each verse is its accent structure. The main accents on syllables 4 through 8 are marked in the text by underlining, both in blue font. Together with the signs for pronunciation, our system opens to the interested reader a winning aid to interpretation, with the sole aim of accompanying him more happily on the journey through the poetry of the Divina Commedia, its beguiling sound, its visionary courage, its overwhelming beauty.

I. Excursus: The Caesura

With the caesura we encounter a linguistic phenomenon which has been discussed in the literature in an exceedingly controversial way without reaching a unified solution. We want to take a clear position here and explain to the reader why we have completely dispensed with the designation of caesura in our edition.

In principle, it seems possible to divide an endecasillabo into two (unequal) half-verses and to assume a rhythmic speech pause[41] between them. This pause should immediately follow the end of the word that carries the stressed main syllable, i.e., that syllable which, apart from the always stressed tenth, defines the type of the verse. In the case of an endecasillabo a maiore, this would be the word with the sixth syllable, and in the case of an endecasillabo a minore, the word with the fourth syllable. However, the exact implementation of this separation instruction soon leads to further constellations, if not complications. For, depending on the length and ending type of the critical word, different separation points result. In the case of the endecasillabo a maiore (6-10), the caesura can occur after the sixth, seventh, or eighth syllable[42], while in the case of the endecasillabo a minore (of both subtypes, i.e., 4-7 and 4-8), it can occur after the fourth, fifth, and sometimes even sixth syllable.[43]

But what happens with the extended canonical variants we have met in Bertinetto's scheme as well as with non-canonical verses? Here, we would first proceed according to the principle established above and also locate the caesura after the word end of the stressed main syllable. If, for example, the inner main tone falls on the second, third or fifth syllable, then the caesura is also located accordingly after the word end of this syllable. In the case of a piano ending, this would be after the third, fourth or sixth syllable.

However, this is not the end of the matter. For up to now we have tacitly assumed that the syllables mentioned are the most strongly accented within a verse. In fact, however, it is up to the performer to decide which of the syllables he emphasizes more strongly and which he emphasizes less strongly. If, in addition to the stressed main syllable, another syllable is stressed just as strongly, then another caesura results, the so-called double caesura, which splits the verse into three parts (cf. Elwert 1968: 54ff.). In the canonical case 4-7, with syllables of equal stress and piano ending, a caesura would be assumed at position 5 and 8. Analogously, this applies to all extended or non-canonical variants with equal stress on the syllables.

Finally, the dynamics of the accents allows a further distinction. If the accents described in the last section are weakly marked inside the sentence, one speaks of a weak double caesura; if no other main tone is prominent besides the ictus on the tenth syllable, the caesura is omitted (ibid.).

All of this is intended to show us that even with the best of intentions, a simple systematization of the caesura does not result. In essence, one may assume that it has to follow a main accent, although with several of these accents, it is up to the reader to decide where to place the caesura or caesuras and whether to place them at all. For the caesura is - this would be another important limitation - a genuinely metrical feature that structures the verse in appropriate cases. The performance (esecuzione) is in principle independent of it, i.e. the temporal division of a verse allows all stylistic liberties, which means that the reciter can do without a pause in speech altogether even after a main accent.

If one now takes the real complexity of accents in the Divina Commedia, one understands why the caesura in Dante's endecasillabo is increasingly fading and why the authors dealing with it have relativized, if not canceled, its significance.[44]

It is thus an optional marking, a virtual, secondary feature of the endcasillabo that works for certain verses but not for others, and that has no binding effect on the performance. Since in our edition we have written out the accents completely, and since the caesura essentially follows the same ones, i.e., is defined by them, there is no need to mark them additionally. The reader recognizes the rhythmic structure of the verse and decides for himself whether and where to place pauses in speech. The omission offers another advantage. It maintains the neutrality of possible recitation inherent purely in the accents, does not artificially split the endecasillabo into two half-verses, thus preserving its unity and not suggesting a certain way of reading mechanically resulting from the visually perceptible separation.

Even if fixed accent structures are taken into account, the rhythmic arrangement of the same is free, i.e. the reader determines the temporal dimension of the sentence. This principle applies a fortiori to the variable accent solutions of my edition as well as to the fundamental, almost unlimited accent variance of a Dante's endecasillabo. It is my firm conviction that one can understand and read this verse only as a whole and only in its dazzling linguistic-musical permutation. The caesura usurps this freedom. It destroys the unity of the verse and imposes a restriction on the viewer that he can either make himself or does not need. We have therefore not included the caesura in our edition.

IV. Outlook

Dear Reader,

I would like to congratulate you. You now have the tools you need to immerse yourself fully in the Divina Commedia. You are able to read the text authentically and creatively at the same time. You understand the play of the basic metrical figures and you know what to do with the accents.

Nevertheless, my system is only the invitation to go on a voyage of discovery yourself. Because with every canto you absorb, new thoughts, new destinies, new worlds will open up to you. You will learn something about Dante, about our humanity and finally about yourself. You will immerse yourself in a work that has not revealed its secret even after centuries. Alongside an unmatched power of language and an admirable clarity of expression and description are visions that transcend the text itself. Among the images you see are sounds that accompany you and feelings that open up to you: harbingers of the intuitive, witnesses to a deeper truth and freedom.

Take your time with this unique work. Enjoy the process of unlocking the unknown. Embark on a journey through space, time and your own soul.

V. Literature

Battaglia, Salvatore (1961-2002). Grande Dizionario della Lingua Italiana. [Giorgio Bárberi Squarotti: UTET] Torino: Unione Tipografico-Editrice Torinese.

Bausi, Francesco & Martelli, Mario (1993). La metrica italiana. Teoria e storia. Firenze: Le Lettere.

Bayerische Akademie der Wissenschaften (2021). Dictionnaire de l’occitan médiéval (DOM). [http://www.dom-en-ligne.de/].

Beccaria, Gian Luigi (1984). Cesura. In U. Bosco, Umberto. Enciclopedia Dantesca (1970). Seconda Edizione Riveduta (1984). Edizione Speciale per la Collana «Orsa Maggiore» in Tiratura Limitata (1996). Roma: Istituto della Enciclopedia Italiana.

Bellomo, Leonardo (2016). Ritmo, metro e sintassi nella lirica di Lorenzo de’ Medici. Padova: Edizioni libriuniversitaria.it.

Beltrami, Pietro G. (2015). L’esperienza del verso. Scritti di metrica italiana. Bologna: Società editrice di Mulino.

Beltrami, Pietro G. (2011). La metrica italiana. Quarta Edizione (2002). Bologna: Il Mulino.

Bosco, Umberto (1984). Enciclopedia Dantesca (1970). Seconda Edizione Riveduta (1984). Edizione Speciale per la Collana «Orsa Maggiore» in Tiratura Limitata (1996). Roma: Istituto della Enciclopedia Italiana.

Canepari, Luciano (2018). Italian Pronunciation & Accents. Geo-social Applications of the Natural Phonetics & Tonetics Method. München: Lincom Gmbh.

Canepàri, Luciano & Giovannelli, Barbara (2012). La buona pronuncia italiana del terzo millennio. Manueletto d’italiano neutro con sonori, esercizi e test. Quarta edizione. Roma: Aracne editrice S.r.l.

Canepàri, Luciano (2009). Il DiPI. Dizionario di Pronuncia Italiana. Bologna: Zanichelli editore.

Carboni, Giancarlo & Sorianello, Patrizia (2011). Manuale professionale di dizione e pronuncia. Come educare la voce parlata. Milano: Editore Ulrico Hoepli.

Carini, Claudio (2019). Dante Alighieri. Divina Commedia. Edizione integrale. Audiolibro. Terza edizione. Recitar Leggendo Audiolibri.

Castiglioni, Luigi & Mariotti, Scevola (2007). Il vocabolario della lingua latina. Quarta Edizione. Torino: La Lœscher editore (Bologna: Zanichelli editore S.p.a.).

Cecchini, Silvia (2010). La Divina Commedia. Audiolibro. Gli Ascoltalibri / Collina d'oro.

Dal Bianco, Stefano (2007). L’endecasillabo del Furioso. Roma: Pacini Editore SpA.

Dal Piai, Giorgio (2013). Dizióne e Fonètica. Un’esperienza didattica per il teatro. Prima edizione digitale. Milano: Adriano Salani Editore.

Dizionario ItalianoOccitano-OccitanoItaliano. Norme ortografiche, scelte morfologiche e vocabolario dell’Occitano Alpino orientale (1998-2001). Cuneo: Più Eventi Edizioni. [https://www.piueventi.it/dizionario-occitano.php].

Elwert,W. Theodor (1968). Italienische Metrik. München: Max Hueber.

Fasani, Remo (1992). La metrica della «Divina Commedia» e altri saggi di metrica italiana. Ravenna: Longo Editore.

Formentin, Vittorio (2010). Fonetica storica. In Raffaele Simone, Enciclopedia dell'Italiano. Roma: Istituto dell'Enciclopedia Italiana (Treccani). [https://www.treccani.it/enciclopedia/fonetica-storica_(Enciclopedia-dell%27Italiano)/].

Gabrielli, Aldo (2020). Grande Dizionario Italiano. Quarta Edizione con Versione Digitale Scaricabile Online. Milano: HOEPLI S.p.A.

Gaffiot, Félix (2016). Dictionnaire Latin Français. Nouvelle edition revue et augmentée, dite Gaffiot 2016 (version V. M. Komarov). Direction: Gérard Gréco.

Garzanti Linguistica (2017). Il Grande Dizionario Italiano. Con licenza online. Milano: Garzanti Linguistica / Novara: De Agostini Publishing S.p.A.

Gruppo Padovano di Stilistica (2000-Oggi). L’Archivio Metrico Italiano (AMI). [https://linguaetesto.wordpress.com/2013/03/12/larchivio-metrico-italiano-ami/].

Inglese, Giorgio (2016). Dante Alighieri. Commedia. Revisione del testo e commento di Giorgio Inglese. Inferno e Purgatorio: Nuova Edizione. Roma: Carocci editore.

Jeanneau, Gérard et al. (2017). Dictionnaire Latin-Français. [https://www.prima-elementa.fr/Dico.htm].

Lansing, Richard (2010). The Dante Encyclopedia. London & New York: Routledge /New York: Garland Publishing.

Lanza, Antonio (1996): Dante Alighieri. La Commedìa. Testo critico secondo i più antichi manoscritti fiorentini. Nuova edizione a cura di Antonio Lanza. Anzio: De Rubeis Editore.

Manni, Paolo (2013). La Lingua di Dante. Bologna: Società editrice il Mulino.

Marescotti, Ivano (2011). Dante Alighieri. La Commedia di Dante interpretata da Ivano Marescotti. Con introduzione ai canti di Riccardo Bruscagli. Audiolibro. Cofanetto 12 CD Audio. Zanichelli Editore.

Menge, Helmut (1996). Menge-Güthling (1911): Langenscheidts Großwörterbuch Lateinisch-Deutsch. 25. Auflage (1996). Berlin, München, Wien, Zürich, New York: Langenscheidt.

Menichetti, Aldo (2013). Prima Lezione di Metrica. Edizione digitale. Roma-Bari: Gius. Laterza & Figli Spa.

Menichetti, Aldo (1993). Metrica italiana. Fondamenti metrici, prosodia, rima. Roma: Salerno Editrice / Padova: Editrice Antenore.

Migliorini, Bruno et al. (2010). DOP – Dizionario Italiano Multimediale e Multilingue d’Ortografia e di Pronunzia. Roma: Rai Radiotelevisione Italiana.

Ouvrard, Yves et al. Collatinus web. Online lemmatiser and morphological analyser for Latin texts: [https://outils.biblissima.fr/en/collatinus-web/].

Paden, William D. (1998). An Introduction to Old Occitan. Second printing (2001). New York: The Modern Language Association of America.

Pertsch, Erich (1983). Langenscheidts Handwörterbuch Lateinisch-Deutsch (1971/1983). 6. Auflage (2000). Berlin, München, Wien, Zürich, New York: Langenscheidt.

Petrocchi, Giorgio (2003). Dante Alighieri, La Commedia secondo l’antica volgata a cura di Giorgio Petrocchi. Firenze: Casa Editrice Le Lettere. Terza ristampa. (Prima Edizione, Milano: Mondadori, 1966/67).

Praloran, Marco (2011). Metro e ritmo nella poesia italiana. Guida anomala ai fondamenti della versificazione. Firenze: Edizioni del Galluzzo per la Fondazione Ezio Franceschini.

Praloran, Marco (2007). Alcune osservazioni sul ritmo nella «Commedia». In Paolo Trovato, Nuove prospettive sulla tradizione della «Commedia». Una guida filologico-linguistica al poema dantesco. Firenze: Franco Cesati Editore.

Praloran, Marco (Ed.). (2003). La metrica dei Fragmenta. Roma, Padova: Editrice Antenore.

Robey, David (2003). Alighieri, Dante (1265-1321); Petrarca, Francesco (1304-1374); Boccaccio, Giovanni (1313-1375). Italian narrative poetry of the Middle Ages and Renaissance [Electronic Resource] / Oxford Text Archive, http://hdl.handle.net/20.500.12024/2455.

Robey, David (2000). Sound and Structure in the Divine Commedy. Oxford: University Press.

Rourett, Robert (2006). La Lenga dels Trobadors (la langue des troubadours). Nîmes: L’Oiseau bleu.

Sangirardi, Giuseppe & De Rosa, Francesco (2002). Breve guida alla metrica italiana. Milano: Sansoni.

Sanguineti, Federico (2001). Dantis Alagherii Comedia. Edizione critica per cura di Federico Sanguineti. Firenze: SISMEL – Edizioni del Galluzzo e Fondazione Ezio Franceschini. Fondo Gianfranco Contini. Archivio Romanzo. Collana a cura di Lino Leonardi.

Serianni, Luca et al. (2011). Lingua Commune. La grammatica e il testo. Milano-Torino: Bruno Mondadori Scuola.

Sermonti, Vittorio (2018). La Commedia di Dante. Raccontata e letta da Vittorio Sermonti. Audiolibro. Regia: Ludovica Ripa di Meana. Emons: libri & audiolibri (1988).

Smith, Nathaniel B. & Bergin, Thomas G. (1984). An Old Provençal Primer. New York & London: Garland Publishing Inc.

Soresina, Maria (2020). Dirò de l'altre cose ch'i' v'ho scorte nella Commedia di Dante Alighieri. Milano, Colibrì.

Soresina, Maria (2009). Libertà va cercando. Il catarismo nella Commedia di Dante. Bergamo, Moretti & Vitali.

Soresina, Maria (2002). Le segrete cose. Dante tra induismo ed eresie medievali. Bergamo, Moretti Honegger.

Treccani, Giovanni (1997). Il Vocabolario Treccani. Seconda edizione. Roma: Istituto della Enciclopedia Italiana.

Vettori, Iacopo (2006-2010). Lettura integrale della Divina Commedia. [http://www.iacopovettori.it/recitazione/commedia/Default.aspx].

Warner Music Italia Srl (1961-62). Dante Alighieri. La Divina Commedia. Lettura di Giorgio Albertazzi, Tino Carraro, Antonio Crast, Carlo D’Angelo, Arnoldo Foà, Achille Millo & Romolo Valli. Warner Music (2006): 12 CD Box.

Zingarelli, Nicola (2019). Lo Zingarelli 2020. Vocabolario della lingua italiana. A cura di Mario Cannella, Beata Lazzarini. Bologna: Zanichelli editore S.p.A.

[30]It is then the "piano" ending. In the so-called "tronco" version with the last stressed syllable, the endecasillabo comprises 10 metrical syllables, in the rare "sdrucciolo" form, whose third to last syllable is stressed, 12 metrical syllables. For words, verse endings as well as rhyme forms in piano, tronco and sdrucciolo form, the terms parossitono, ossitono and proparossitono - coming from Greek - are also common in metrics.
[31]Only those who strictly adhered to this standard had done what was necessary, and it was with pride to announce that one of the greats of the classical period, Petrarch, had achieved the goal.
[32]Synonym for stressed main accent.
[33]The works of a Praloran (2011, 2007, 2003) and the volume presented by Stefano dal Bianco (2007) on the endecasillabo of Orlando furioso should be mentioned here. The treatise by Leonardo Bellomo (2016) also describes in detail the schemes of the endecasillabo.
[34]Cf. Sangirardi & De Rosa (2002): 73.
[35]Cf. Sangirardi & De Rosa (2002): 73/74: „Il Duecento, fino a Dante incluso, è caratterizzato da una varietà di schemi molto ampia, che si spiega con l’assenza di una codificazione esplicita. Dante nella Commedia adotta la maggior quantità di schemi ritmici riscontrabile nei classici italiani.” Also one finds “una percentuale molto bassa di endecasillabi che presentano problemi di scansione perché I loro schemi non si lasciano ricondurre ai modelli canonici.” (Iibid). As we will see, this number increases as the conventional bipolar accent scheme is extended.
[36]Or one goes the opposite way and adapts not the system to the poem, but the poem to the system, which in the end amounts to a denial of linguistic reality.

And indeed, there has been no lack of attempts to brand the disliked verses as irregular or to canonize them more or less forcibly by adding accents on the desired syllables. For example, Remo Fasani (1992) writes: 77: „Tirando le somme, si vede dunque che bisogna essere più che cauti quando si parla, a proposito dell’endecasillabo della Commedia, di schemi non canonici, in quanto forse nessuno di loro resiste a un esame approfondito.“ (Emphasis mine).

However, we do not want to deal with these maneuvers any further, since the other problem is even more fundamental.
[37]It deliberately leaves the syllable length free, and does not set any caesura or other pause markings, since these temporal parameters arise quasi-naturally from the strength and position of the accents or the rhythm of the verse and, similar to what is known from music, are the responsibility of the performer.
[38]Bertinetto (1981): 247/48: „Per ciò che riguarda il problema dell’accento secondario, ho cercato di mostrare la scarsa rilevanza di questa entità teorica nel sistema prosodico italiano. Da un lato ho osservato infatti che molti presunti accenti secondari sono in realtà dei meri fatti di esecuzione (che ho chiamato ʿ accenti ritmici ʾ), legati a certe esigenze di articolazione ritmica del discorso, e strettamente dipendenti dalla diversa rapidità di elocuzione.”
[39]These were first the authors of the historic Warner Italia recording (1961/62). They were followed by Claudio Carini (2019), Vittorio Sermonti (1988) and Ivano Marescotti (2011). Iacopo Vettori (2006-10) and Silvia Cecchini (2010) rounded out the list.

It is not the place here to point out differences between the reciters. We may freely highlight with Iacopo Vettori one of the interpreters: he has hit the right measure in most cases with a medium speaking speed, an andante, and a calm, but at the same time committed manner of presentation.
[40]The reader will find the link to Robey's database in footnote 25. The L'Archivio Metrico Italiano of the Padova Group (AMI) was introduced in 2000 by Sergio Bozzola and has been continuously developed since then (https://linguaetesto.wordpress.com/2013/03/12/larchivio-metrico-italiano-ami/).

. We do not want to hide the fact that there are substantial differences between these two databases, but above all in our assessment of the accents to them as well. This is also, but not only, due to the introduction of the trivalent accent logic. Where Robey sets too many accents, we find too few in the Padova group. This may well be due, as Praloran (2006: 50, cited in Beltrami 2015: 307), one of their mentors, points out, to the American's over-reliance on automatic scansion of the texts, and therefore loss of the contextuality of the accents, indeed destruction of the true character of the verse rhythm. The Italian database promises improvement here. (Computer science is supposed to be a help, but not a dogma, and be replaced by an individual phonological and syntactic approach.)

We feel sympathy for this approach, so we can agree in principle with this assessment and also did not make a syntactic-statistical assignment. Admittedly, it turns out that the manual processing of the Italians cannot really satisfy either. It is incomplete and - like that of its counterpart - not without errors. Both databases were therefore allowed to be consulted and overcome. We would consider ourselves fortunate if our work had succeeded, mainly also due to the new accent system, in raising the scansion of the Divina Commedia to a new, more realistic level, i.e. in having broken down and reliably expounded its rhythmic essence, which illuminates itself from the lively recital.

To our regret, we must state that the Padova Project has suspended its presence on the Internet for an indefinite period of time. Thus, its database is currently no longer available online. Nevertheless, we want to put the old link here in case of a future remedy: http://www.maldura.unipd.it/ami/php/index.php.
[41]For the moment, we equate caesura with (speech-) pause. Later, we will see that this is a metrical interruption, which may suggest a pause in the speech, but is in principle independent of it.
[42]One gets then before the separation a) a Settenario tronco: „di bella verità / m’avea scoverto”, b) a Settenario piano „Allor distese al legno / ambo le mani” or c) a Settenario sdrucciolo: „per la similitudine / che nacque“. The caesura is marked here as below with the slash.
[43]This results analogously in a) a Quinario tronco: „noi fuggirem / l’imaginata caccia”, b) a Quinario piano: „O voi che siete / in piccioletta barca“ or c) a Quinario sdrucciolo: „Tutti gridavano: / "A Filippo Argenti!“. All of the examples in this paragraph are taken from the Beccaria article cited at the beginning of the next footnote.
[44]Thus, for example, Gian Luigi Beccaria writes in the Enclicopedia Dantesca (1984: 929): „In D[ante] anzitutto non c'è alcuna obbligatorietà di c[esura]: il suo verso pare piuttosto organismo fortemente unitario, che non risulta dalla somma di due unità ritmiche appartenenti a versi differenti (Pernicone); la c[esura] non è mai così forte da spezzare la salda unità ritmica del verso (anche se pare più netta nei versi con parola tronca, e avvertibile ancora con parola piana [ricominciaron / le parole mie], lo è meno quando avviene sinalefe: Ma se le mie parole / esser dien seme // che frutti infamia / al traditor ch'i' rodo).“ Elwert (1968: 57) adds: „Die Zäsur ist somit nicht ein wesentliches Merkmal des Elfsilbers.“

And Menchietti (1993: 467) notes after a nuanced account of the controversial assessments in the literature: „A nostro modo di vedere, invece, la cesura dell'endecasillabo - anche quando si realizza (cioè, come vedremo, non sempre) - non ha in quanto tale nessun riflesso necessariamente percettibile sull'esecuzione; la sua percettibilità, di per sé, è solo virtuale, anche se il lettore è messo tanto piu chiaramente sulla sua traccia quanto piu frequente e percettibile è il suo coincidere con pause o sospensioni effettive. In altri termini, la cesura dell'endecasillabo, come quella di ogni altro verso non doppio o composto, si situa secondo noi prima di tutto al livello della scansione.” The caesura, in other words, is not a phonetic but primarily metrical phenomenon (ibid: 467).

Impressum/Privacy Policy