Latin Accentual Clausula as Exemplified in 14th-Century Prose Texts by Dante and Boccaccio Latin Accentual Clausula as Exemplified in 14th-Century Prose Texts by Dante and Boccaccio Annika Mikkel* Abstract. This paper studies 14th-century Latin prose rhythm as exemplified by Dante and Boccaccio. The texts observed in this analysis are samples from De Monarchia, De vulgari eloquentia, Quaestio de aqua et terra and Epistole by Dante and De mulieribus claris and De casibus virorum illustrium by Boccaccio. In ancient rhetoric, rhythmical units were used at the ends of sentences and clauses in prose texts. These units were called clausulae, and the rhythm of classical prose was based on the quantity of sylla- bles. Medieval Latin prose rhythm, however, was based on word stress and was called cursus. The aim of this paper is to study what kinds of cursus occur in the given text samples and their frequency. The research method used in this paper is comparative- statistical analysis. The distribution of cursus in these samples is also analysed by chapters and different types of cursus are distinguished. Keywords: prose rhythm, cursus, clausula, Dante, Boccaccio 1. Introduction In ancient rhetoric, rhythmical units were used at the ends of sentences and clauses in prose texts. In classical prose, these units were called clausulae, and the rhythm of classical prose was based on the quantity of syllables.1 Medieval Latin prose rhythm, however, was based on word stress and was called cursus. This paper focuses on 14th-century Latin prose as exemplified by Dante and Boccaccio. The texts observed in this analysis are samples from De Monarchia, De vulgari eloquentia, Quaestio de aqua et terra and Epistole by Dante and De mulieribus claris and De casibus virorum illustrium by Boccaccio. The use of cursus in the prose of Dante and Boccaccio has been studied before, and it has been said that the presence of cursus is obvious in both authors’ texts. * Author’s address: Annika Mikkel, Lecturer in Romance Philology, College of Foreign Languages and Cultures, University of Tartu, Lossi 3, 51003, Tartu, Estonia. E-mail: annika.mikkel@ut.ee. 1 The use of clausulae is seen, for example, in the works of Livy, Sallust and Cicero. See, for example, Aili (1979), Albrecht (2003). https://doi.org/10.12697/smp.2022.9.1.04 Studia Metrica et Poetica 9.1, 2022, 66–88 https://doi.org/10.12697/smp.2022.9.1.04 67Latin Accentual Clausula as Exemplified in 14th-Century Prose Texts by Dante and Boccaccio However, according to Paget Toynbee, the main researcher of Dante’s prose, Dante did not use cursus regularly throughout the books. Instead, he employed them more in rhetorical and personal sections appearing mostly at the begin- nings and at the ends of the books (1920: 231). Toynbee’s studies, which date back almost a century (1918, 1920, 1923), do not include systematic statistical analysis. As regards Boccaccio, Parodi (1913) notes that his letters in Latin also follow the rules of cursus.2 Previous studies have not examined the different types of cursus according to the chapters in which they occur but have instead compared the use of cursus with other structures used to end sentences. The aim of this paper is to study the kinds of cursus and their frequency in the given text samples. The distribution of cursus in these samples is also analysed according to the chapters in which they occur, and different types of cursus are distinguished. Based on this, patterns are investigated, includ- ing rhythms that the authors seem to have preferred and those they avoided, whether certain types of cursus accumulate in particular sections of the texts, and what differences appear when comparing texts. 2. Prose rhythm: theoretical basis and methodology Wilhelm Meyer (1905: 236–286) appears to have been the first to formulate a method for describing the accentual cursus of late antiquity and the Middle Ages, which has become fundamental in more recent scholarship on the sub- ject. According to Meyer, the crucial factor is the number of syllables between the last two accents and after the last one, and the length of the last word is unimportant (Tunberg 1992: 126). A number of modern studies and reference works employ this kind of method of describing cursus (Tunberg 1996: 115). Medieval prose rhythm also allows consillabicatio, in which the last word is replaced by two or three short words that include the same number of syllables (Janson 1975, Lausberg 1960). In the Middle Ages, there were four basic rhythmical units: cursus planus, cursus velox, cursus tardus and cursus trispondaicus. Cursus planus consists of a trisyllable word with stress on the penultimate syllable, which is pre- ceded by a word with a similar stress (Xx#xXx – fide servantur). In addition to the regular cursus planus, the forms X#xxXx (pars imperabant) and Xxx#Xx (tempore messis) are also used. Cursus velox comprises a tetrasyllabic word with stress on the penultimate syllable preceded by a word with stress on 2 See also the works of Vittore Branca (1967, 1975, 1983). 68 Annika Mikkel the third syllable from the end of the word (Xxx#xxXx – animus gratuletur). Another form of cursus velox is Xx#xxxXx (where the last word contains not four but five syllables: for example, cedit utilitatem). Cursus tardus includes a tetrasyllabic word with stress on the third syllable from the end of the word preceded by a word with stress on the penultimate syllable (Xx#xXxx – divina custodiet). An alternative form involves a proparoxytone followed by a trisyl- labic proparoxytone, in which the caesura comes after the third syllable of the clausula (Xxx#Xxx – omnia impleant). Cursus tardus may also occur in the form X#xxXxx (est inpossibile). Cursus trispondaicus comprises a tetrasyllabic word with stress on the penultimate syllable preceded by a word with similar stress (Xx#xxXx – declinare arbitratur). Cursus trispondaicus can also occur as Xxx#xXx (opere patrate) or X#xxxXx (pax universalis).3 Each of the cursus forms can have its longer word units resolved by combi- nations of monosyllabic and disyllabic words as long as the rhythm is retained. For example, the final tetrasyllable of the cursus velox can be represented by a trisyllabic paroxytone preceded by a monosyllable (rescribere non cessato) or by two disyllables (feliciter dici solet). Alternatively, the second word component of the tardus can be resolved by a disyllabic paroxytone and a monosyllable (oneri mihi sum) or by a trisyllabic paroxytone and a monosyllable (stel- larumque contentus sit) (Oberhelman, Hall 1984: 115). The research method of this paper is a comparative statistical analysis car- ried out in stages. In the first stage, all clauses in the samples were analysed. Cursus may appear both at the ends of sentences and clauses and in this study, all endings before punctuation marks were analysed. In the second stage, all the analysed samples were collected into different databases for each author and each text. In each database, the endings corresponding to the cursus are treated as separate categories (cursus planus, cursus trispondaicus, cursus tardus, cursus velox); endings that are different from cursus are all grouped together under the category “other endings”. The aim of this paper is to study the occurrence of cursus; therefore, the different rhythms under the category “other endings” are not distinguished. In the final stage, the data were compared (comparisons were made between the two authors, different texts by the same author, and different chapters of the same work) and statistically analysed. 3 See more about medieval cursus in Clark 1910, Janson 1975, Lindholm 1963, Norberg 1968, Oberhelman, Hall 1984. 69Latin Accentual Clausula as Exemplified in 14th-Century Prose Texts by Dante and Boccaccio 3. Dante Alighieri Malato (1999) notes that Dante adapted with the stylistic structures relying on classical rhetoric tradition and that he used cursus in his Latin prose, there- fore the regular use of cursus is obvious in De Vulgare Eloquentia, Monarchia, Quaestio de Aqua et Terra and especially in Epistole. According to Malato (1999: 1035), in De vulgari eloquentia, cursus planus is dominant, cursus tardus is frequent, and cursus velox is preferred at the end of a sentence. Epistole aside, Dante’s work in Latin prose consists of scholarly texts in which the use of cursus was not prescribed. However, even within these works, Dante does not disdain to apply the rules of rhythmic prose. For example, Chiesa and Tabarroni (2013: LXXXII–LXXXIV) note that rhythmic clauses also appear in the proemial parts and rhetorical sections of the Monarchia, and in the prologues, the cursus is used almost constantly. According to Paget Toynbee (1918, 1920, 1923), who studied cursus in Dante’s Latin prose, Dante did not use cursus regularly throughout a text but employed it in rhetorical and personal passages occurring mostly at the begin- ning and end of each book (Toynbee 1920: 231). Other scholars have also noted that the use of rhythmic clauses is not distributed uniformly and can oscillate considerably from chapter to chapter even within the same work. In Epistole, the use of cursus is more intense and regular (Mengaldo 1970: 292). For this study, I analysed the occurrence of cursus in Dante’s Latin books De vulgari eloquentia, De Monarchia, Epistole and Quaestio de Aqua et Terra.4 The sample was formed from the first eleven chapters of De Monarchia (474 sentence and clause endings) and the first ten chapters of De vulgari eloquen- tia (474 endings). From Epistole, the sample was formed from the first six letters, which contributed 478 endings altogether. In addition, 500 sentence and clause endings from Quaestio de Aqua et Terra were analysed. Unlike Paget Toynbee’s analysis, this study shows the proportions of the frequencies of different cursus types in each sample and, where possible, also analyses the distribution of cursus by chapters (including the distribution of different cursus types by chapters). The results of the analysis of De Monarchia are shown in the following chart:5 4 All samples of Dante’s texts in this paper were taken from Fallani et al. 1993. 5 Figures 1–25 in this paper also appear in Mikkel 2020: 41–126. 70 Annika Mikkel Figure 1 Figure 2 Figure 3 0% 5% 10% 15% 20% 25% 30% 35% 40% planus trispondaicus velox tardus other 0% 5% 10% 15% 20% 25% 30% 35% 40% planus trispondaicus velox tardus other 0% 5% 10% 15% 20% 25% 30% 35% 40% 45% planus trispondaicus velox tardus other Figure 1. Cursus in De Monarchia (%%) The results demonstrate the presence of cursus in this sample (63.3%) and the prevalence of cursus planus (22.8%) (e.g., primo agentem, esse credendum). Cursus velox (15.6%), (e.g., opera ordinatur, hominum salutabat) and cursus tardus (15.4%) (e.g., intellectum possibilem, habet imperium) are used nearly equally. The least frequently used form in this sample is cursus trispondaicus (9.5%) (e.g., bone voluntatis, totum universum). The analysis of the occurrence of cursus by chapters in this sample shows that there are chapters in which the frequency of a particular cursus form is far higher or lower than the average. Closer study shows a higher presence of cursus velox in Chapters 1.I (27.6%) and 1.VIII (24.0%), while the average percentage of cursus velox in this sample is 15.6%. The first chapter also has the highest occurrence of cursus tardus (34.5%) in this sample. In Chapter 1.IX, cursus tardus does not occur at all. The percentage of cursus planus is highest in Chapter 1.IX (34.8%), and cursus trispondaicus occurs more in Chapters 1.VI (22.7%), 1.VIII (20.0%) and 1.IX (17.4%). 71Latin Accentual Clausula as Exemplified in 14th-Century Prose Texts by Dante and Boccaccio Figure 1 Figure 2 Figure 3 0% 5% 10% 15% 20% 25% 30% 35% 40% planus trispondaicus velox tardus other 0% 5% 10% 15% 20% 25% 30% 35% 40% planus trispondaicus velox tardus other 0% 5% 10% 15% 20% 25% 30% 35% 40% 45% planus trispondaicus velox tardus other Figure 2. Cursus in De Monarchia Chapter 1.I (%%) Figure 1 Figure 2 Figure 3 0% 5% 10% 15% 20% 25% 30% 35% 40% planus trispondaicus velox tardus other 0% 5% 10% 15% 20% 25% 30% 35% 40% planus trispondaicus velox tardus other 0% 5% 10% 15% 20% 25% 30% 35% 40% 45% planus trispondaicus velox tardus other Figure 3. Cursus in De Monarchia Chapter 1.VI (%%) 72 Annika Mikkel Figure 4 Figure 5 0% 5% 10% 15% 20% 25% 30% planus trispondaicus velox tardus other 0% 5% 10% 15% 20% 25% 30% 35% planus trispondaicus velox tardus other 0% 5% 10% 15% 20% 25% 30% 35% planus trispondaicus velox tardus other Figure 4. Cursus in De Monarchia Chapter 1.VIII (%%) In De Monarchia, the rhythmic aspect of the initial chapter stands out. Compared to the average throughout the analysed sample, the occurrence of cursus in the first chapter is 89.7%. The results of the analysis of De vulgari eloquentia can be seen in the fol- lowing chart:Figure 4 Figure 5 0% 5% 10% 15% 20% 25% 30% planus trispondaicus velox tardus other 0% 5% 10% 15% 20% 25% 30% 35% planus trispondaicus velox tardus other 0% 5% 10% 15% 20% 25% 30% 35% planus trispondaicus velox tardus other Figure 5. Cursus in De vulgari eloquentia (%%) 73Latin Accentual Clausula as Exemplified in 14th-Century Prose Texts by Dante and Boccaccio The results show that the percentage of cursus in this sample is 73.2% and that the most frequently occurring type is cursus planus at 29.7% (e.g., aves locun- tur, rationem portare). Cursus velox is present in this sample with 21.3% (e.g., medium sensuale, operantium deridebant), and cursus tardus with 14.3% (e.g., esse flexibile, pauci perveniunt). The least used type in this sample is cursus trispondaicus (7.8%) (e.g., sermo variatur, versus orientem). Examining the occurrence of cursus in the given sample by chapters reveals a higher percentage of cursus velox in Chapters 1.I (33.3%), 1.VII (27.3%), 1.VIII (32.6%). The average percentage of cursus velox in this sample is 21.3%. In Chapter 1.VIII, the percentage of cursus trispondaicus is the highest (10.9%), while the average in this sample is 7.8%. Figure 4 Figure 5 0% 5% 10% 15% 20% 25% 30% planus trispondaicus velox tardus other 0% 5% 10% 15% 20% 25% 30% 35% planus trispondaicus velox tardus other 0% 5% 10% 15% 20% 25% 30% 35% planus trispondaicus velox tardus other Figure 6. Cursus in De vulgari eloquentia Chapter 1.I (%%) 74 Annika Mikkel Figure 7 Figure 8 0% 5% 10% 15% 20% 25% 30% 35% planus trispondaicus velox tardus other 0% 5% 10% 15% 20% 25% 30% 35% planus trispondaicus velox tardus other 0% 5% 10% 15% 20% 25% 30% 35% 40% 45% 50% planus trispondaicus velox tardus other Figure 7. Cursus in De vulgari eloquentia Chapter 1.VII (%%)Figure 7 Figure 8 0% 5% 10% 15% 20% 25% 30% 35% planus trispondaicus velox tardus other 0% 5% 10% 15% 20% 25% 30% 35% planus trispondaicus velox tardus other 0% 5% 10% 15% 20% 25% 30% 35% 40% 45% 50% planus trispondaicus velox tardus other Figure 8. Cursus in De vulgari eloquentia Chapter 1.VIII (%%) In Chapter 1.I, it is possible to observe the accumulation of cursus tardus (23.3%, while the average percentage is 14.3%), and Chapter 1.VIII con- tains the lowest percentage of cursus tardus (4.3%) in this sample. Like in De Monarchia, in De vulgari eloquentia, we can also see a very high occurrence of cursus in the first chapter (93.3%). The results of the analysis of Quaestio de aqua et terra are summarized in the following chart: 75Latin Accentual Clausula as Exemplified in 14th-Century Prose Texts by Dante and Boccaccio Figure 7 Figure 8 0% 5% 10% 15% 20% 25% 30% 35% planus trispondaicus velox tardus other 0% 5% 10% 15% 20% 25% 30% 35% planus trispondaicus velox tardus other 0% 5% 10% 15% 20% 25% 30% 35% 40% 45% 50% planus trispondaicus velox tardus other Figure 9. Cursus in Quaestio de aqua et terra (%%) The results show that the percentage of the cursus in this sample is 55.6%, and cursus planus (21.0%) is predominant (e.g., veritatis amore, esse videntur). Next in frequency is cursus trispondaicus (15.0%) (e.g., centrum universi, membrum declarandum), followed by cursus velox (10.6%) (e.g., omnibus confirmatur, ordine discendorum), and finally cursus tardus (9.0%) (e.g., verum ostendere, esset concentrica). It is interesting to note the quite high frequency of cursus trispondaicus in Quaestio in comparison with Dante’s other Latin texts and, at the same time, the less frequent use of cursus velox and tardus. The results of the analysis of Epistole are summarized in the following chart:Figure 9 Figure 10 Figure 11 Figure 12 0% 5% 10% 15% 20% 25% 30% planus trispondaicus velox tardus other 0% 5% 10% 15% 20% 25% 30% 35% 40% planus trispondaicus velox tardus other 0% 5% 10% 15% 20% 25% 30% 35% planus trispondaicus velox tardus other Figure 10. Cursus in Epistole (%%) 76 Annika Mikkel The results show that the percentage of cursus in this sample is 79.5%. Cursus planus (28.0%) is predominant (e.g., forma conformis, signetur intentum). Next in frequency is cursus tardus (25.5%) (e.g., fecit exilium, nobis consilia) followed by cursus velox (22.0%) (e.g., persolvere attentabit, vitia repellentem), and finally, cursus trispondaicus (4.0%), (e.g., arbitrato carceratum, declinare arbitrator). Examining the occurrence of cursus in the given sample by letters (see, for example, Figures 11 and 12) reveals that, in the first letter, we can see a higher frequency of cursus, 91.8%. In Epistole, there is a very low frequency of cursus trispondaicus and, on the other hand, equal presence of all three other cursus forms. In the third letter, the percentage of planus is 10.7%, tardus is present with 30,4%, and the occurrence of all types of cursus is 71.4%. Figure 9 Figure 10 Figure 11 Figure 12 0% 5% 10% 15% 20% 25% 30% planus trispondaicus velox tardus other 0% 5% 10% 15% 20% 25% 30% 35% 40% planus trispondaicus velox tardus other 0% 5% 10% 15% 20% 25% 30% 35% planus trispondaicus velox tardus other Figure 11. Cursus in Epistole I (%%) Figure 9 Figure 10 Figure 11 Figure 12 0% 5% 10% 15% 20% 25% 30% planus trispondaicus velox tardus other 0% 5% 10% 15% 20% 25% 30% 35% 40% planus trispondaicus velox tardus other 0% 5% 10% 15% 20% 25% 30% 35% planus trispondaicus velox tardus other Figure 12. Cursus in Epistole III (%%) 77Latin Accentual Clausula as Exemplified in 14th-Century Prose Texts by Dante and Boccaccio 4. Giovanni Boccaccio Of Boccaccio’s Latin writings, the first that can be dated with certainty are some letters written from Naples around 1339. All these letters follow the rules of cursus. According to Ernesto Parodi, one of Boccaccio’s main models was Dante, and only from him could he learn the respect for the rhythm (Parodi 1957: 480). I have analysed the occurrence of cursus in Boccaccio’s Latin texts De muli- eribus claris and De casibus virorum illustrium.6 The sample was formed from proemio and the first five chapters of De mulieribus claris (472 sentence and clause endings) and from proemio and the first six chapters of De casibus virorum illustrium (478 endings). The results of the analysis of De mulieribus claris are summarized in the following chart: Figure 13 Figure 14 0% 5% 10% 15% 20% 25% 30% 35% 40% 45% 50% planus trispondaicus velox tardus other 0% 5% 10% 15% 20% 25% 30% 35% 40% 45% 50% planus trispondaicus velox tardus other 0% 5% 10% 15% 20% 25% 30% 35% 40% 45% planus trispondaicus velox tardus other Figure 13. Cursus in De mulieribus claris (%%) The percentage of cursus in the analysed sample of De mulieribus claris is about 56%. The most frequent form in this sample is cursus tardus (23.1%) (e.g., agitata laboribus, insignis splendoribus) followed by cursus planus with 18.4% (e.g., viros illustres, legem agentes), then cursus velox (7.8%) (e.g., publice salutare, memoria celebranda), and finally, cursus trispondaicus with 7.2% (e.g., votum advocasset, silvas incolentem). I also analysed this sample by chapters (see, for example, Figures 14 and 15). There is a high percentage of cursus tardus and, at the same time, a lesser 6 The samples of Boccaccio’s texts in this paper are taken from Branca 1967 and Branca 1983. 78 Annika Mikkel presence of cursus velox, which was more common in Latin medieval prose rhythm. We can see a higher percentage of velox in the proemio.Figure 13 Figure 14 0% 5% 10% 15% 20% 25% 30% 35% 40% 45% 50% planus trispondaicus velox tardus other 0% 5% 10% 15% 20% 25% 30% 35% 40% 45% 50% planus trispondaicus velox tardus other 0% 5% 10% 15% 20% 25% 30% 35% 40% 45% planus trispondaicus velox tardus other Figure 14. Cursus in De mulieribus claris, Proemio (%%) Figure 13 Figure 14 0% 5% 10% 15% 20% 25% 30% 35% 40% 45% 50% planus trispondaicus velox tardus other 0% 5% 10% 15% 20% 25% 30% 35% 40% 45% 50% planus trispondaicus velox tardus other 0% 5% 10% 15% 20% 25% 30% 35% 40% 45% planus trispondaicus velox tardus other Figure 15. Cursus in De mulieribus claris Chapter V (%%) The results of the analysis of De casibus virorum illustrium are summarized in the following chart: 79Latin Accentual Clausula as Exemplified in 14th-Century Prose Texts by Dante and Boccaccio Figure 16 Figure 17 0% 5% 10% 15% 20% 25% 30% 35% 40% planus trispondaicus velox tardus other 0% 5% 10% 15% 20% 25% 30% 35% 40% planus trispondaicus velox tardus other 0% 5% 10% 15% 20% 25% 30% 35% planus trispondaicus velox tardus other Figure 16. Cursus in De casibus virorum illustrium (%%) Here we can see that the percentage of cursus is 64% and the favoured form is cursus planus with 27.8% (e.g., rapte sororis, genus humanum). Next in fre- quency is cursus tardus (23.2%) (e.g., hostium viribus, volatus longissimus), then cursus velox (6.7%) (e.g., veteri civitate, incola paradisi), and finally, cursus trispondaicus (6.3%) (e.g., eventus novitate, ipsos ignoratis). In the analysis of this sample by chapters (see, for example, Figures 17 and 18), we can again note a relatively low presence of cursus velox. Only in Chapter II is the percentage somewhat higher at 12.7% (while the average use of velox in this sample is 6.7%). Cursus planus and tardus are represented more or less equally. 80 Annika MikkelFigure 16 Figure 17 0% 5% 10% 15% 20% 25% 30% 35% 40% planus trispondaicus velox tardus other 0% 5% 10% 15% 20% 25% 30% 35% 40% planus trispondaicus velox tardus other 0% 5% 10% 15% 20% 25% 30% 35% planus trispondaicus velox tardus other Figure 17. Cursus in De casibus virorum illustrium Chapter II (%%) Figure 16 Figure 17 0% 5% 10% 15% 20% 25% 30% 35% 40% planus trispondaicus velox tardus other 0% 5% 10% 15% 20% 25% 30% 35% 40% planus trispondaicus velox tardus other 0% 5% 10% 15% 20% 25% 30% 35% planus trispondaicus velox tardus other Figure 18. Cursus in De casibus virorum illustrium Chapter VI (%%) 81Latin Accentual Clausula as Exemplified in 14th-Century Prose Texts by Dante and Boccaccio 5. Comparison and analysis The results of the analysis of all samples from Dante’s texts are compared in the following chart: Figure 19 Figure 20 0% 5% 10% 15% 20% 25% 30% 35% 40% 45% 50% Cursus pla nus Cursus velox Cursus tardus Cursus trispondaicus other DVE De Mona rchia Quaestio Epistol e 0 5 10 15 20 25 30 35 40 1 2 3 4 5 6 7 8 9 10 11 obser ved occurrence averag e 0 5 10 15 20 25 1 2 3 4 5 6 7 8 9 10 11 obser ved occurrence averag e Figure 19. Cursus in all samples of Dante (%%) As shown in the chart, cursus planus appears more often than other cursus types in all samples. In Epistole, the presence of cursus was usual and expected, including a nearly equal occurrence of planus, velox and tardus and a lower frequency of trispondaicus. The percentage of other endings in this sample is 20.5%. Compared to the Epistole, in De vulgari eloquentia, the frequency of planus and velox is similar, but the frequency of tardus is slightly decreased. In the sample from De Monarchia, planus still the most preferred ending, but velox has been used less and trispondaicus more. Quaestio de aqua et terra is different in that the frequency of planus is comparable, for example, to the sample from De Monarchia, but there are fewer occurrences of velox and tardus in Quaestio. A higher rate of trispondaicus can be seen instead. The percentage of other endings in this sample from Quaestio is also higher at 44.4%. Previous Dante researchers have noted that the use of rhythmic clauses is not distributed uniformly and can oscillate considerably within the same work. In earlier studies, different cursus types have not been distinguished in favour of a more general observation on the concentration of cursus. The analysis of samples of Epistole, De vulgari eloquentia and De Monarchia by chapter, which separately observes the frequencies of every cursus type, also demonstrates a clearly uneven distribution of different rhythmic clauses. 82 Annika Mikkel In general, cursus planus is distributed through the samples more or less equally, but at the same time, the percentages of velox, tardus and trispondaicus are very different in different chapters. In the Epistole, both velox and tardus occur throughout the sample. Analysing by chapters, cursus occur more in the first chapters of the books: in the first chapter of De vulgari eloquentia (93.3%), in the first chapter of De Monarchia (89.7%) and even in the first letter of Epistole (91.8%). In the sample from De Monarchia, there are also chapters in which certain cursus types do not appear at all. For example, in Chapter 1.I, trispondaicus does not occur, and in Chapter 1.IX, tardus is not present. The presence of cursus tardus in this sample is higher in the first chapter (34.5%), but at the same time, there is no trispondaicus. However, in the samples from De vulgari eloquentia and Epistole, all cursus types are present in every chapter. As an example, we can see here in the sample from De Monarchia, the occurrence of cursus by chapters compared to the average percentage of cursus in the whole sample.Figure 19 Figure 20 0% 5% 10% 15% 20% 25% 30% 35% 40% 45% 50% Cursus pla nus Cursus velox Cursus tardus Cursus trispondaicus other DVE De Mona rchia Quaestio Epistol e 0 5 10 15 20 25 30 35 40 1 2 3 4 5 6 7 8 9 10 11 obser ved occurrence averag e 0 5 10 15 20 25 1 2 3 4 5 6 7 8 9 10 11 obser ved occurrence averag e Figure 20. Cursus planus in De Monarchia (%%) The frequency of cursus planus remains nearly the same in all chapters. The presence of planus is a little higher at the beginning of the book and in Chapters 7–9. 83Latin Accentual Clausula as Exemplified in 14th-Century Prose Texts by Dante and Boccaccio Figure 19 Figure 20 0% 5% 10% 15% 20% 25% 30% 35% 40% 45% 50% Cursus pla nus Cursus velox Cursus tardus Cursus trispondaicus other DVE De Mona rchia Quaestio Epistol e 0 5 10 15 20 25 30 35 40 1 2 3 4 5 6 7 8 9 10 11 obser ved occurrence averag e 0 5 10 15 20 25 1 2 3 4 5 6 7 8 9 10 11 obser ved occurrence averag e Figure 21. Cursus trispondaicus in De Monarchia (%%) Cursus trispondaicus shows rather big differences from the average: in the first chapter, the percentage of trispondaicus is 0%, but in Chapter 6, it is 22.7%. The presence of trispondaicus is also greater than the average in Chapters 7–9. Figure 22 Figure 23 0 5 10 15 20 25 30 1 2 3 4 5 6 7 8 9 10 11 obser ved occurrence averag e 0 5 10 15 20 25 30 35 40 1 2 3 4 5 6 7 8 9 10 11 obser ved occurrence averag e 0% 5% 10% 15% 20% 25% 30% 35% 40% 45% 50% planus velox tardus trispondaicus other De casibus De mulieribus Figure 22. Cursus velox in De Monarchia (%%) 84 Annika Mikkel With cursus velox, we can see higher occurrence in the first chapter (27.6%) and in Chapter 8 (24%). At the same time, there are chapters with a lower presence of velox, for example, in Chapters 6 (4.5%) and 7 (6.7%). Figure 22 Figure 23 0 5 10 15 20 25 30 1 2 3 4 5 6 7 8 9 10 11 obser ved occurrence averag e 0 5 10 15 20 25 30 35 40 1 2 3 4 5 6 7 8 9 10 11 obser ved occurrence averag e 0% 5% 10% 15% 20% 25% 30% 35% 40% 45% 50% planus velox tardus trispondaicus other De casibus De mulieribus Figure 23. Cursus tardus in De Monarchia (%%) Also, in the case of cursus tardus, there is a higher occurrence in the first chapter (34.5%) and Chapter 10 (26.9%). There are some chapters in which the percentage of tardus is not so different from the average percentage of this sample, and in one chapter, this cursus does not occur at all (Chapter 9). The results of the analysis of Boccaccio’s texts are compared in the follow- ing chart: 85Latin Accentual Clausula as Exemplified in 14th-Century Prose Texts by Dante and Boccaccio Figure 22 Figure 23 0 5 10 15 20 25 30 1 2 3 4 5 6 7 8 9 10 11 obser ved occurrence averag e 0 5 10 15 20 25 30 35 40 1 2 3 4 5 6 7 8 9 10 11 obser ved occurrence averag e 0% 5% 10% 15% 20% 25% 30% 35% 40% 45% 50% planus velox tardus trispondaicus other De casibus De mulieribus Figure 24. Cursus in De mulieribus claris and De casibus virorum illustrium (%%) The chart shows that cursus planus and cursus tardus occur most frequently in the Boccaccio samples. In the sample from De mulieribus claris, the presence of tardus even exceeds planus, but at the same time, velox occurs less often, at a similar rate to cursus trispondaicus. In the sample from De casibus virorum illustrium, planus and tardus appear to be the preferred types, and there is a nearly equal distribution of velox and trispondaicus. The analysis of those samples by chapters does not show that Boccaccio used more cursus in some chapters and less in others. Comparing the works of Dante and Boccaccio, we can conclude the following: 1) In Dante, there is a higher occurrence of cursus in the first chapters of the books. This is not the case in the samples from Boccaccio.7 2) The two authors use cursus planus more often and cursus trispondaicus the least. The exception is the Quaestio de aqua et terra by Dante, where a high frequency of trispondaicus is apparent. 3) Comparing the use of cursus velox and tardus by both authors, there is a clear preference for tardus and lower presence of velox in the samples from Boccaccio, while in the samples from Dante, they tend to be used nearly equally. 7 See, for example, Figure 14; in the proemio of De mulieribus claris, the percentage of cursus is 56,8%. 86 Annika Mikkel 0% 5% 10% 15% 20% 25% 30% 35% 40% 45% 50% DVE De Mona rchia Quaestio Epistol e De casibus De mulieribus planus velox tardus trispondaicus other Figure 25. Cursus in all samples from Dante and Boccaccio (%%) 6. Conclusion In conclusion, cursus occur in all samples from both analysed authors, and there are more cursus planus, tardus and velox and less trispondaicus. Comparing Dante’s Latin works, planus is the most used cursus form, and tardus and velox appear less often. The exception is Dante,s Epistole, where planus, tardus and velox are nearly evenly present. The presence of all three cursus types in Epistole is predictable, as constant usage of cursus was a rule in letters. However, in De vulgari eloquentia and in De Monarchia, it also appears that the presence of planus and velox is not significantly lower than in Epistole, the proportion of tardus has decreased, and the proportion of trispondaicus has slightly increased. Both in the samples of De vulgari eloquentia and De Monarchia, the percentage of cursus is higher than average in the first chapter of both works, which may indicate that for the author, cursus is one of the stylistic elements to be presented to the reader at the beginning of a book. Quaestio de aqua et terra differs from the three works mentioned above: here we can see more modest use of cursus velox and tardus and a greater presence of cursus trispondaicus. In his Latin samples, Boccaccio clearly prefers cursus tardus and cursus planus. Cursus velox, generally considered the most elegant form of cursus in Latin prose, is less present in his samples. Previous researchers have claimed that Dante does not use cursus evenly throughout his works: they occur more frequently at the beginnings and ends 87Latin Accentual Clausula as Exemplified in 14th-Century Prose Texts by Dante and Boccaccio of works, and their occurrence may be very different within the same book. In previous studies, analyses made by chapters did not distinguish between different types of cursus and simply examined the presence of cursus versus other sentence endings. The results of the present research confirm that Dante’s samples have a very high incidence of cursus in the first chapters (for example, 93.3% in De vulgari eloquentia). Analysis by chapters ascertained that cursus planus is distributed evenly, but the presence of velox, tardus and trispondaicus varies to a great extent in different chapters. The results of this analysis are certainly not definitive, but they offer several directions for future analyses. References Aili, Hans 1979. The Prose Rhythm of Sallust and Livy. Stockholm: Almquist & Wiksell International. Albrecht, Michael von 2003. Cicero’s style: A Synopsis, Followed by Selected Analytic Studies. Leiden: Brill. Branca, Vittore 1967. Tutte le opere di Giovanni Boccaccio, vol. X: De mulieribus claris. Milano: Mondadori. Branca, Vittore 1975. Boccaccio medievale, Firenze: G.C. Sansoni editore. Branca, Vittore 1983. Tutte le opere di Giovanni Boccaccio, vol. IX: De casibus virorum illustrium. Milano: Mondadori. Chiesa, Paolo; Tabarroni, Andrea 2013. Dante Alighieri. Le opere, vol. IV: Monarchia (Introduzione). Roma: Salerno Editore. Clark, Albert Curtius 1910. The cursus in Medieval and Vulgar Latin. Oxford: Clarendon Press. Fallani, Giovanni; Maggi, Nicola; Zennaro, Silvio (eds.) 1993. Dante Alighieri. Tutte le opere: Divina Commedia, Vita Nuova, Rime, Convivio, De vulgari eloquentia, Monarchia, Egloghe, Epistole, Quaestio de aqua et terra. Milano: Newton Comptor editori. Janson, Tore 1975. Prose Rhythm in Medieval Latin from the 9th to the 13th Century. Stockholm: Almquist & Wiksell. Lausberg, Heinrich 1960. Handbuch der literarischen Rhetorik: eine Grundlegung der Literaturwissenschaft. München: Hueber. 88 Annika Mikkel Lindholm, Gudrun 1963. Studien zum Mittellateinischen Prosarhythmus, seine Entwicklung und sein Abklingen in der Briefliteratur Italiens. Stockholm: Almquist & Wiksell. Malato, Enrico 1999. Dante. In: Malato, Enrico (ed.), Storia della letteratura italiana. Volume I. Dalle origini alle Dante. Roma: Salerno editrice, 773–1052. Mengaldo, Pier Vincenzo 1970. Cursus. In: Bosco, Umberto (ed.), Enciclopedia dantesca, vol. 2. Roma: Istituto della Enciclopedia Italia, 290–295. Meyer, Wilhelm 1905. Die rhythmische lateinische Prosa. In: Meyer, Wilhelm, Gesammelte Abhandlungen zur mittellateinischen Rhythmik. Berlin: Weidmannsche Buchhandlung, 236–286. Mikkel, Annika 2020. Rütmilised lauselõpud 14. sajandi ladinakeelsetes ning itaalia rahvakeelsetes proosateostes. Tartu: Tartu Ülikooli Kirjastus. Norberg, Dag 1968. Manuel pratique de latin médiéval. Paris: Picard. Oberhelman, Steven M.; Hall, Ralph G. 1984. A New Statistical Analysis of Accentual Prose rhythms in Imperial Latin Authors. In: Classical Philology 79(2), 114–130. https://doi.org/10.1086/366843 Parodi, Ernesto Giacomo 1913. Osservazioni sopra il cursus delle opere latine e volgari del Boccaccio. In: Miscellanea storica della Valdelsa XXI. Castelfiorentino: Società storica della Valdelsa, 231–248. Parodi, Ernesto Giacomo 1957. Intorno al testo delle epistole di Dante e al cursus. In: Folena, Gianfranco (ed.), Lingua e letteratura. Studi di teoria linguistica e di storia dell’italiano antico. Parte seconda. Venezia: Neri Pozza Editore, 399–442. Toynbee, Paget 1918. Dante and the ‘Cursus’: A New Argument in Favor of the Authenticity of the ‘Quaestio de Aqua et Terra’. In: The Modern Language Review 13(4), 420–430. https://doi.org/10.2307/3714033 Toynbee, Paget 1920. Dantis Aligherii Epistolae. The Letters of Dante. Emended Text with Introduction, Translation, Notes, and Indices and Appendix on the Cursus. Oxford: Clarendon Press. Toynbee, Paget 1923. The Bearing of the Cursus on the Text of Dante’s De Vulgari Eloquentia. Oxford: Oxford University Press. Tunberg, Terence O. 1992. A Study of ‘Clausulae’ in Selected Works by Lorenzo Valla. In: Humanistica Lovaniensia 41. Leuven: Leuven University Press, 104–133. Tunberg, Terence O. 1996. Prose styles and cursus. In: Mantello, Frank; Rigg, Arthur Georg (eds.), Medieval Latin. Washington: The Catholic University of American Press, 111–121.