Primary oral features in romance chapbooks of Northeastern Brazil.

da Silva Amorim, Marcelo

IN Stories on a String, Candace slater says that the so-called Literatura de Cordel--which can be roughly translated as "string literature"--is believed to be an heir to European popular literature, such as the Portuguese Literatura de Cego, the British Chapbooks, the French Litterature de Colportage, and the Spanish Pliegos Sueltos (56). Mark J. Curran reminds us that the word cordel had not been used to refer to these small books until the 1960s, when a new generation of scholars borrowed that name from the European tradition. It derived from the fact that sometimes these booklets were displayed, in small-town markets, hanging on horizontal strings. However, the term largely employed by early twentieth-century Brazilian folklorists, poets and their audience was folheto, which could mean either "booklet" or "pamphlet." Actually, the folhetos are brochures that typically contain eight, sixteen or thirty-two pages measuring 16x10 centimeters. Every set of eight pages in a booklet results from folding a regular sheet of paper (folha) in half twice, in a way that the total number of pages is generally a multiple of eight.

The folhetos are always written in verse. They comprise mostly the sextilha, which is a stanza composed of six lines, with XBXBXB rhymes, in which X stands for blank verses; and the decima, stanza that consists of ten lines, with several rhyme-patterns, the most common of which is ABBAACCDDC. Both types of stanza are believed to be of Iberian origin, and the sextilha is the most frequent one. Each page of a folheto contains either five sextilhas, or three decimas. As for the metrical pattern of the verses, the most common types are those containing either seven or ten syllables.

There have been several attempts to classify the folhetos by grouping them according to their themes, but there are folhetos about almost everything. As a result of such a great variety of themes, some scholars--such as Ana Maria de O. Galvao, in Brazil, and Candace Slater, in the United States --have come up with different approaches to classification. In Cordel, Galvao seeks to identify a relation between the number of pages in a chapbook and its subject matter. Therefore, she adopts the term folheto to describe either 8-or 16-page booklets, which narrate faits-divers, stories or cases in the news that are of human interest; and the word romance to define 24- to 64-page booklets that present versions of stories previously existing in the oral tradition or stories of enchantment. Braulio Tavares asserts that fantasy certainly is a trait that plays an important role in this genre, since it draws the romance closer to the universe of traditional European folktales. It is no surprise to recognize in their plots medieval settings with castles, kings, princesses, witches, dragons, talking animals, or the quest for something invaluable. On the other hand, such an environment also contains a number of elements that are typical of the Brazilian reality such as food, plants, animals, toponyms, clothes, and especially dialects.

I will also adopt, as did Galvao, the word romance to refer to this genre of booklet that is one of the few means through which the nordestino man from the Sertao--or wilderness--of Brazil could express himself in a relatively recent past. More than a century ago, poor and illiterate people, most of whom were tenants living in sublet small stretches of land on farms or in modest hamlets, used to purchase those booklets and take them home. In the evenings, neighbors and friends got together to listen to the only person who could suitably read the story to a whole crowd. Sometimes, sponsored by landowners, the poets themselves would perform the readings or, out of so much repetition and technique, would recite their own verses by memory (Tavares).

There was nothing special about the bard. He looked like anyone else in the crowd, except for the fact that he distinguished himself in the way he delivered his own compositions. He usually did not simply utter the lines: he improvised and seemed to chant them. On other occasions, two or more minstrels were invited to engage in singing duels. According to Elizabeth Travassos, "A significant part of the repertoire of published poems belonged to a genre called 'peleja' (battle or duel)" (64). Such was the scenario early in the twentieth century, when men and women from one of the neediest regions of Brazil, impelled by droughts and neglected by successive administrations, fled from their homes to more industrialized regions in the South(east) of the country. Together with the remembrance of their desolate and devastated lands, they also brought pamphlets in their baggage.

For decades, chapbooks continued to be produced and were popular among Northeastern migrants and a few dwellers in the big cities. Although the booklets have become widely spread throughout the country, according to Curran, large sectors of the Brazilian population know very little about this tradition. Many seem surprised when they encounter cordel poets performing on the streets; others cannot hide the extent to which chapbooks diverge from their urban literary tastes, associating them with the archaic and remote world of underprivileged rural classes from one of the most poverty-stricken regions of the country. At best, they are looked upon and appreciated as "folklore", i.e., a tradition that seems picturesque but, beyond any doubt, idiosyncratically bizarre, and far-fetched in style as well. As Travassos reminds us, "[...] attitudes toward cantoria vary according to the level of familiarity a person has with the tradition, which can be quite limited, even amongst the middle and upper classes of the north-east" (62), a home to the genre.

The decline of the booklet tradition, despite of all the researches that have been done during the last decades, can be said to have occurred for a couple of reasons. But behind all of them lies the fact that the pamphlet literature has never been regularly or seriously taught in Brazilian schools, despite the explicit official recommendations, (2) as this genre still faces the prejudices that ensue from its association with an oral background. Repetition of events, for instance, is among the various alleged drawbacks that some from a blase intellectual elite would criticize in chapbooks as oral/written compositions. Some people would argue that the poet is not accurate with the construction of his text, since episodes can be reiterated up to three times in the course of a romance, and, according to some literate readers' judgment, that clogs the plot, making the narrative sound monotonous. What I expect to be indisputable by the end of this article is that this type of reasoning is thoroughly unacceptable, especially because it does not consider that reiteration is an inherent property of the configuration of cordel as a genre, which should be regarded as an attribute rather than as a defect.

What I would like to claim here is that some of the presumed flaws in chapbooks make them more similar to texts highly valued throughout human history, such as Homer's The Iliad and The Odyssey, than most of the books ever written all along the existence of canonic literature. It seems that only a total lack of understanding of the primary oral nature of chapbooks could explain the prejudice against them. As I have already said, this genre contains some features that integrate it into a primary oral tradition, in which the essential record of social data is made through contrived language. As happens with narratives in most illiterate societies, these romance pamphlets contain features--such as reiteration--that, in addition to being an independent trait in itself, also play an important role in the constitution of other characteristics. Reiteration rises in the so-called verbal inner compositional repetition, which encompasses the recurrence of lines, formulae and epithets; incidental inner compositional repetition, which involves the recounting of episodes; and the syntactic reiteration, which is centered on the types of enjambment that take place between lines and pairs of lines of the stanzas.

What I want to convey is that Homeric formulaic phraseology and the formulaic phraseology of romance chapbooks are somewhat similar because both employ in their construction the same "oral technology," a term that Carl J. Couch appropriates from Havelock to "designate standardized procedures for rendering discourse into poetry" (589). Differently from literate cultures, in which information is stored more permanently and reliably and with far less mental effort, illiterate societies had to rely on a technology that shaped language itself into a highly memorable and memorizable form. Erick A. Havelock points to suitable linguistic form as one of the conditions that would make language acquire that enhanced power to retain information and transmit it to the following generations. For him, "suitable linguistic form" meant "poetized language." Therefore, whatever vital information that these oral cultures needed to convey or stow away for future use, it should take the form of metrified lines and contain rhythm.

Havelock also adds that the vernacular, the everyday language spoken by a people, is not used for any statements that require preservation. Besides taking a poetized form, such statements would also have to be "narrativized," namely, they would have to be put into a kind of story format (186). Not surprisingly, critics have disagreed with Havelock on such specifications, and their arguments are so many that it would be impossible to summarize them here. But in none of them have I found any reflection on how Havelock could have meant something more than simply imposing the necessity for poetic narrative on material which could equally be expressed in another form. He certainly knew that the Greeks first used writing to compose (or transliterate) poetry, not to write prose. As Havelock says, and Couch quotes, "prose did not become common until three hundred years after the invention of the alphabet had rendered the monopoly exercised by poetry over the contrived word as theoretically obsolete" (590).

Havelock may also have meant that poetry was a highly productive medium for certain phenomena to take place. He probably had in mind the recurrence of noun-formulae and epithets in Homeric poems. In prose, it would be very hard to detect formulae or epithets, since accurate metrical locations (slots) are a requisite for the identification of such phenomena. Likewise, it is only through strict poetic metrical constructions that some characteristics become clearly visible. Such is the case with the syntactic relations between couplets and lines in a romance.

Clearly enough, all of these traits can also be present in our contemporary prose as well. The crucial difference is that, in primary oral poetry, those traits will go beyond mere stylistic usage of language. They will function in complete cooperation with a powerful economy of orality that dictates most of the rules and controls the phenomena and their recurrences with an undisguised purpose of organically constructing the text or composition. Oral economy is an intellectually-devised force that takes into account the necessities of conservation and survival of knowledge while it attempts to prevent the dispersion and loss of cultural experience. The devices employed in oral poetic compositions are intended either to resume or to restore something that is not recorded anywhere but in volatile human memory/culture. Therefore, one can assume that contrived language is a set of ingenious apparatuses that will make human beings capable of bringing back their memories at a minimum loss rate.

In this article, I will concentrate on reiteration as a means that contributes to evoke and restore information in the romance chapbooks. The formulae, epithets and syntactic relations between verses are features governed by reiteration, which show a high incidence in the romance. The density of these phenomena will be demonstrated by means of four pamphlets from which I will take examples:

1. Casamento e Mortalha no Ceu se Talha (CMCT), by Leandro Gomes de Barros.

2. Historia de Juvenal e o Dragao (HJD), by Joao Martins de Athayde.

3. Historia da Princeza da Pedra Fina (HPPF), by Leandro Gomes de Barros.

4. O Pavao misterioso (PM), by Manoel Camilo dos Santos.

Notice that the samples to be shown here are drawn from a corpus that contains around two thousand four hundred verses. Therefore, it would not be possible to analyze all of the instances of such phenomena in this article. Hence, I opted to simply make a concise introduction of formulae and epithets, saving more room to look a little deeper into the presentation of the syntactic mechanisms and some of their statistical figures, followed by a brief interpretation. I also left out the incidental inner compositional repetition (reiteration of scenes), which is a subject worth an article in itself.

Firstly, I will address the question of commonplace reiteration of formulae in the cordel tradition. For milman Parry, a formula is a word or "a group of words which is regularly employed under the same metrical conditions to express a given essential idea" (The making of Homeric Verse 272). Referring to the ancient Greek tradition, he asserts that a formula has to fill a particular metrical pattern and that, except for slight changes of inflection, it must be repeated verbatim, that is, exactly as in previous occurrences.

In CMCT, one can see words and expressions that are repeated within the romance, in metrical slots 1/2/3/4 (Table 1); 1/2/3 (Table 2); and 5/6/7/[8] (Table 3), with unstressed syllable -NA in slot 8, not computed for metrical purposes: Table 1 Recurrence of formulae in s ats 1/2/3/4 in CMCT FORMULA 5 6 7 Stanzas (N)O NO ME DE 22, 23, 28, 44, 93, 94, 107 Table 2 Recurrence o formulae in slots 1/2/3 in CMC FORMULA 4 5 6 7 Stanzas CHA MA DO[A] 1, 6, 11, 17, 41, 67, 82 Table 3 Recurrence of formulae in slots 5/6/7/[8] in CMCT 1 2 3 4 FORMULA Stanzas SAL VE LI NA 28, 35, 36, 46, 49, 52, 58, 61, 62, 69, 100, 107

And here there is another romance, HPPF, in which the formula PRINC EZA(S) is concentrated in slots 6/7/[8] and A PRINCEZA in slots 1/2/3/4 (Table 4). Notice that the latter occurs exclusively in blank rhyme type lines (X), whereas the former appears evenly in b-rhyme type lines and blank rhyme type lines (X). That means the poet employs the strategy to fill out the two last slots indistinctively, whether or not he needs to find a rhyme to close the line. One can also note that the formulae show either at the onset or at the end of the lines, inasmuch as, in these ambiences, it becomes much easier to have a better control of improvisation for the remaining slots. Therefore, when at the end of a B-type line, the formula meets two requirements at a time: besides providing the verses with a rhymed closure, it also grants them the perfect preservation of their hepta-syllabic metrical patterning. At an onset position, formulae basically function as a theme (topic) and enable the poet to advance adjustments necessary to accommodate rhematic content (comment) in the remaining slots. Refer to the examples in Table 4 below.

The high frequency of BARBEIRO at the closure of the lines is also conspicuous, as can be easily seen in Table 5 below. The somewhat significant recurrence of the word in intermediary slots 2/3/4 and 3/4/5 is due to determiners (o: the, and este: this) placed before the noun and which occupy slot 1 and slots 1/2 respectively.

The function of all this verbal repetition is precisely to serve the meter. In "Studies in the Epic Technique," Parry reasons that the poet who composes orally, "unlike the poet who writes out his lines [...], cannot think without hurry about his next word, nor change what he has made, nor, before going on, read over what he has just written. [...]. He must have for his use wordgroups all made to fit his verse" (77).

Secondly, we have the repetition of epithets. In "The Text Tale of Frodo," Maria Prozesky, arguing about heavy characters in the Homeric poems, says that mnemonic impulses contribute to "another characteristic of oral literature, namely a tendency to epithetic identification" (26). Elements of thought and d scourse are remembered more easily if they occur in clusters, and particula ly in parallel clusters, such as noun-epithet aggregations: a brave soldier, a beautiful princess, a sturdy oak.

According to donaldo schuller, the function of epithets in heroic poems serves to delineate the characters showing whatever stable characteristics in his personality. Of course, in epic poetry, the hero's character is not susceptible to changes: he belongs in noble and divine lineage; his often supernatural deeds are praised; his acts of living up to his full potential, his virtues or excellence--in short, his arte--will be acclaimed. This becomes very apparent in the famous homeric epithets: odysseus is always wily; nestor is wise; achilles is swift-footed, and so on. However, like noun-formulae, the celebrated epithets must also occur with regularity and fill a certain metrical pattern.

In the romances the poet will employ these attributes according to the way that characters behave in the narrative. In chapbooks, the protagonist is not necessarily a hero, since his personality presents unsteady traces. His willpower and actions are not immutable. That is the case with Afonso Durval, in CMCT. Durval is a rich farmer who has everything he wants except a child. He and his wife are depicted as a distressed couple all throughout the first pages of the booklet, in a household devoid of happiness. One night, an angel appears to him in a dream and predicts that his wife would become pregnant. The baby, a girl named Salvelina, is destined to marry a poor boy in the neighborhood called Guilhermino, who actually is one of Durval's godsons. In order to change the inescapable fate, on the following day, Durval sends one of his slaves, Manuel, to kidnap and kill the toy. Manuel, filled with compassion is alt g er incapable of ca rryi g out the t sk assigned to him. He puts the boy in a basket and sets him adrift on a river. Guilhermino is found by a rich Engli shman an 1 is token t 3 Euro e, but when his stepfather dies, he retu ns to his pla of origin, meets Salvelina an i marries her. During the wedding party, the whole story is unveiled, and the truth about Durval's crimes is reveled.

Although Durval's description ranging from good to evil can be clearly tracked down and pinpointed in the booklet, it is interesting to remark that only his and his wife's qualities--not anyone else's--as a righteous couple are portrayed in the form of epithets in most of the first half of the narrative, as shown in Table 6: Table 6 Repetition of epithets in slots [1]/2/3 in CMCT EPITHETS 4 5 6 7 [8] Stanzas [O] GRAN DE A fon so Dur val 2, 9, 24 [A] GRAN DE A Ma ri a Ro sa 7, 18 [A] NO BRE Ma ria a Ro sa 8, 10, 15 [DO] GRAN DE A fon so Dur val 57, 72

The typical stability of epithets on the second and third slots will persist until approximately half of the romance. From Guilhermino's account of past events in stanza 78 onwards, adjectives like "mad," "delinquent," and "mean" begin to be erratically employed: the epithet system becomes unstable and does not follow any metrically strict or regular pattern.

Lastly, there is the repetition of syntactical mechanisms. To use one of Walter Ong's preferred expressions, the "additive" nature of the lines in the romances (which is clearly well marked by syllabic definition and a typical closing rhythm) draws them much closer to The Iliad and The Odyssey, when it comes to the configuration of enjambment found in them, than to slightly later written poetry, as will be demonstrated. Just as Homer tends to avoid periodic (necessary) enjambment (NE)--wherein the sentence is semantically and syntactically incomplete at the end of the hexameters and, therefore, must be finished in the following line--so the Northeastern minstrel of romance chapbooks shows a marked preference for either unperiodic enjambment (UE) or no enjambment (OE) at all. (3)

In Homer's poetry, one verse line is often connected to the next, but generally loosely linked, in the famous adding or strung style always associated with him. It is this peculiarity that first led Parry to his investigation of syntax. Valerie Krishna affirms that, "to get at the basis of Homer's distinctive style, Parry studied the interplay between the sentence and the hexameter verse line in his poems, and then compared his verse with that of Virgil and Apollonius, distinguishing three types of line endings" (64). In order to illustrate the recurrence of such phenomena in the romances, I have here selected a number of stanzas from the booklets previously listed. I will not show examples pulled out from The Iliad (refer to Krishna 65 to see examples in ancient texts).

(1) Those containing NO ENJAMBMENT (0E), in which the end of a line coincides with the end of a sentence:

Perguntou evangelista quanto custa o seu invento? -de-me cem contos de reis acha caro o pagamento? o rapaz lhe respondeu: acho pouco, dou duzentos. (Pm stanza 60, p. 14)

(2) Those with UNPERIODIC ENJAMBMENT (UE), in which the end of a line marks the completion of a thought, but the sentence is only resumed in the next line:

creusa sabendo da historia chorava de arrependida Por ter marcado o rapaz com banha desconhecida disse: nunca mais terei sossego na minha vida. (Pm stanza 110, p. 25)

(3) Those with NECESSARY ENJAMBMENT (NE), in which the end of a line falls in the middle of a syntactic unit or at the end of a word group that is not a complete thought:

-exijo que va comigo Para meu pai conhecer esse homem destemido Que me salvou de morrer mesmo pra recompensa-lo da forma que merecer. (HJD stanza 50, p. 11)

Let us scrutinize what Parry obtained when he researched various compositions from different types of literacy backgrounds (Table 7 was edited from Krishna 69): Table 7 Comparison--different types of compositions (Oral Tradition and Written Tradition) OE UE OE + UE NE ORAL Iliad 48.5% 24.8% 73.3% 26.6% Odyssey 44.8% 26.6% 71.4% 28.5% Serbo-Croatian poets 44.5% 40.6% 85.1% 14.9% Averages 45.9% 30.7% 76.6% 23.3% WRITTEN Argonautica 34.8% 16.0% 50.8% 49.1% Aeneid 38.3% 12.5% 50.8% 49.2% Averages 36.6% 14.3% 50.8% 49.2%

The qualities that Parry's investigation highlights in Homer--a high incidence of end-stopped and potentially end-stopped lines and a low incidence of necessary enjambment; the prominence of unperiodic enjambment; and the tendency of line endings, even those with necessary enjambment, to coincide with a pause or a natural break--prove to be, in spite of the differences of language and time, quite close to those in the songs by the oral poets of Northeastern Brazil.

Let us see what I obtained conducting the same kind of statistical approach to the romance chapbooks (see Table 8): Table 8 Comparison between romances Romances OE UE OE + UE NE CMCT 31 30 61 39 HJD 40.8 29.2 70 30 HPPF 47.6 28.9 76.5 23.5 PM 42% 31% 73% 27% Averages 40.3 29.8 70.1 29.9

And here we have Table 9 in which the average results are compared: Table 9 Comparison between averages in Oral Tradition, Romances and Written Tradition Averages 0E (%) UE (%) 0E + UE (%) NE (%) Oral Tradition 45.9 30.7 76.6 23.3 Romances 40.3 29.8 70.1 29.9 Written Tradition 36.6 14.3 50.8 49.2

Commenting on the figures that Parry obtained in his research, Lord says, in The Singer of Tales, that "this absence [sic] of necessary enjambment is a characteristic of oral composition and is one of the easiest touchstones to apply in testing the orality of a poem" (52). If Lord is really accurate as regards to what the figures represent, one can assume that the romances analyzed here are surely closer to oral compositions than to written ones considering this aspect, since 29.9% is a figure that stands closer to the 23.3% of Homer 's poems than to the 49.2% of written works.

In order to close this article, I would like to restate that reiteration is a conspicuous trait within the romance chapbooks briefly discussed here. From a perspective of the conception of those songs, what Ong used to call redundancy or copiousness (and which I have called reiteration) is a necessary consequence of the oral economy, a pattern that the singer, poet or minstrel will keep in mind when he creates his poems. He generally changes into poetry the prose existing in his routine by means of a technology that turns regular, but important, information into a composition of highly retrievable characteristics. From a point of view of the reception, during the performance, in which the minstrel meets with a public that will follow his story, such devices of repetition gain a meaningful purpose. He will use repetition as a way to keep what is being presented securely in view, since backtracking to check up on previous matters would be difficult for his audiences. As Prozesky poses, "an oral story teller has to keep the audience engaged" (31). The mechanisms of repetition of the items listed here, together with other phenomena employed by the poet, besides helping him to memorize his piece, also make the romances more rhythmic.

While talking about what rhythm represented for "primitive people," Franz Boas argues that "the liking for the frequent repetition of single motives is in part due to the pleasure given by rhythmic repetition" (330). I maintain that the same laws of oral literature still apply nowadays for oral compositions made by people who are far from being "primitive," such as the authors of the romance chapbooks. One can assume that the practice of whatever kind of reiteration that formally marks out the romance poets from others can be regarded to serve the function of advancing the stories that they intend to deliver to their public, rather than clogging them with redundancy and hence retarding them. Therefore, it would not be an overstatement to say that the authors of romance chapbook have full command of what they are doing and by no means could be judged as naive or half-witted. The way they conceive and transmit (in performances) their compositions is the very essence of what the romance chapbooks are--and that is, beyond question, vital to their audience's understanding and fruition of their own literature dynamics.

UNIVERSIDADE FEDERAL DO RIO GRANDE DO NORTE

Works Cited

Athayde, Joao Martins de. Historia de Juvenal e o Dragao. Juazeiro do Norte: Jose Bernardo da Silva, 1974.

Barros, Leandro Gomes de. Casamento e Mortalha no Ceu se Talha. N.p.: n.p., n.d.

--. Historia da Princeza da Pedra Fina. Recife: n.p., n.d.

Boas, Franz. "Stylistic Aspects of Primitive Language." The Journal of American Folklore 149 (1925): 329-39. Web. 5 July 2014.

Couch, Carl J. "oral Technologies: a Cornerstone of Ancient Civilizations?" The Sociological Quarterly 30.4 (1989): 567-602. Web. 10 July 2014.

Curran, Mark J. "A Literatura de Cordel: Antes e Agora." Hispania 74.3 (1991): 570-76. Web. 10 July 2014.

Galvao, Ana Maria de O. Cordel: Leitores e Ouvintes. Belo Horizonte: Autentica, 2001.

Havelock, Eric A. The Literate Revolution in Greece and its Cultural Consequences. Princeton: Princeton UP, 1982.

Krishna, Valerie. "Parataxis, Formulaic Density, and Thrift in the Alliterative Morte Arthure." Speculum 57.1 (1982): 63-83. Web. 20 June 2014.

Lord, Albert B. The Singer of Tales. Cambridge: Harvard UP, 2000.

Ong, Walter. Orality and Literacy: The Technologizing of the Word. London: Methuen, 1982.

Parametros Curriculares Nacionais: Terceiro e Quarto Ciclos do Ensino Fundamental: Lingua Portuguesa. Secretaria de Educacao Fundamental. Brasilia: MEC/SEF, 1998.

Parry, Milman. "Studies in the Epic Technique of Oral Verse-Making: I. Homer and Homeric Style." Harvard Studies in Classical Philology 41 (1930): 73-147. Web. 20 September 2014.

--. The Making of Homeric Verse. The Collected Papers of Milman Parry. Oxford: Clarendon Press, 1971.

Prozesky, Maria. "The Text Tale of Frodo Nine-fingered: Residual Oral Patterning in The Lord of the Rings." Tolkien Studies 3 (2006): 21-43. Web. 22 August 2014.

Santos, Manoel Camilo dos. O Pavao Misterioso. Campina Grande: n.p., 1960.

Schuler, Donaldo. Aspectos Estruturais na Iliada. Porto Alegre: UFRGS, 1972.

Slater, Candace. Stories on a String: The Brazilian Literatura de Cordel. Berkeley: U of California P, 1982.

Tavares, Braulio. "Stories on a String. Literatura de Cordel." Classes. Web. September 2015.

Travassos, Elizabeth. "Ethics in the sung duels of North-Eastern Brazil: Collective Memory and Contemporary Practice." British Journal of Ethnomusicology 9.1 (2000): 61-94. Web. 17 September 2010.

(1) This article has been produced during the author's Postdoctoral research conducted at Universidade Federal do Rio de Janeiro, under the supervision of Professor Violeta Virginia Rodrigues.

(2) The Parametros Curriculares Nacionais include "cordel" in a list of oral literary genres to be taught in Brazilian schools (54).

(3) I borrowed the enjambment categories from Krishna (69). Table 4 Recurrence of formulae in slots 6/7/[8] and 1/2/3/4 in HPPF 1 2 3 4 5 6 7 [8] Type of Stanzas rhymes PRIN CE ZA(S) B 1, 49, 63, 87, 101, 108, 142 X 2, 10, 48, 50, 86, 124, 154 A PRIN CE ZA B -- X 55, 70, 77, 113, 122, 138, 145 Table 5 Recurrence of formulae in slots 2/3/4, 6/7/[8], and 3/4/5 in HPPF 1 2 3 4 5 6 7 [8] Type of Stanzas rhymes BAR BEI RO B -- X 32, 59, 106, 107 BAR BEI RO B 57, 121, 124 X 30, 35, 36, 65, 68, 92, 127, 128 BAR BEI RO B 58, 125 X 66, 93, 124