Alliteration and Rhyme in the Traditional Kakataibo Chants of Emilio Estrella Alliteration and Rhyme in the Traditional Kakataibo Chants of Emilio Estrella Alejandro Augusto Prieto Mendoza*1 Abstract: This paper studies alliteration and rhyme in the traditional Kakataibo chants of Emilio Estrella Logía, one of the most important Kakataibo sabios of the present era. For alliteration and rhyme, consonants which are able to be in coda position accord- ing to Kakataibo syllable structure play a central role. Alliteration is sporadic and based on the repetition of fricative consonants in passages of indeterminate length. It occurs in syllable onset and freely within a line and across lines, the latter by adja- cency. Kakataibo rhyme is also sporadic, its domain is the final syllable of the line and the nucleus of it; only the nasal consonant /n/ can occupy coda in end-line syllables. Kakataibo true rhyme, as opposed to rhyme in lists created by repetition or semantic parallelism, is by adjacency and within vowel passages of indeterminate length. Keywords: alliteration, rhyme, Kakataibo, Amazonia, verbal art We are at an early stage regarding the understanding of the verbal art of Amazonian indigenous chants and music, despite its crucial importance for current anthropological theory (Brabec de Mori, Seeger 2013; Déléage 2020; Severi 2008), contemporary basic linguistic theory and language documenta- tion (Epps, Michael 2017). Nonetheless, we have a relatively incipient image of what is happening: the Amazon is a discursive area in which forms and processes of great diffusion converge in particular discursive genres (Beier et al. 2002). Thus, we know that poetic forms such as semantic parallelism, repetition, nonsense syllables, and others, are used among many languages of different linguistic families and in, for example, shamanic discourse, ritual salutation or chants. Little is known about many other poetic forms, whether meter, rhyme, alliteration, etc., which are discussed in the present document. Perhaps the most studied and confirmed poetic form in the Amazon is paral- lelism (Fox 2014; Franchetto, Stenzel 2017; Prieto Mendoza 2019). Likewise, as I have emphasized in previous works (Prieto Mendoza 2019, 2021), it has been claimed that Amerindian verbal art lacks rhyme or meter (Hymes 1977; * Author’s address: Alejandro Augusto Prieto Mendoza, Universidad Nacional de Colombia, sede Amazonia, Kilometro 2 Via Tarapacá, Leticia, Amazonas, Colombia; email: alprietom@unal.edu.co. Studia Metrica et Poetica 9.2, 2022, 92–116 https://doi.org/10.12697/smp.2022.9.2.03 mailto:alprietom@unal.edu.co https://doi.org/10.12697/smp.2022.9.2.03 93Alliteration and Rhyme in the Traditional Kakataibo Chants of Emilio Estrella Edmonson 1971; Tedlock 1977, 1983, 1987, 2010). However, this is too pre- liminary a claim, considering the under-researched situation of verbal art in the Americas, and recent works have argued against this (Michael 2019; Prieto Mendoza, 2021; Skilton 2017). With all this in mind, I present in this article perhaps the first case of alliteration and rhyme for a traditional chant of an indigenous or native Amazonian group. In addition, apart from the novelty of a new case with all its particularities, I discuss in the present paper the singing style and compositional individuality of a specific person: Emilio Estrella Logía (1925/1935? – 2020), sabio of the Yamino Native Community, who, judging by the years I have dedicated to traditional Kakataibo chant, was the most dedicated person to the Kakataibo word and its aesthetics. I hope this article serves as a tribute to his life and work, as well as the recognition of his lyrical capacity within Peruvian verbal tradition. This paper is organized as follows: I present in § 1 some important aspects of Kakataibo phonology relative to my analysis. In § 2, I detail the origin of my data and its limitations. I offer in § 3 a brief account of Kakataibo traditional chant and I present a biographical resumé of Emilio Estrella emphasizing his work in the linguistic documentation of the Kakataibo language. § 4 is a theoretical section on alliteration and rhyme based on typological-comparative studies. I analyze in § 5 alliteration and rhyme in the traditional Kakataibo chant of Emilio Estrella based on cases from no bana ‘iti and ño xakwati chants, and I discuss his use of both poetic forms in relation to the typologi- cal parameters discussed in § 4. I conclude this paper in § 6 by summarizing the main aspects of the employing and linguistic mastery of alliteration and rhyme in Emilio Estrella’s composition, and offering some reflections on the presence of both poetic forms in Amazonian verbal art. 1. Some notes on Kakataibo phonology In order to understand Kakataibo alliteration and rhyme, it is necessary to know three aspects of the phonology of the language: syllable structure and consonants in coda positions, fricative and affricate consonants, and mid vow- els. I show in Tables 1 and 2 the inventory of Kakataibo consonants and vowels, respectively – < > angle brackets indicate orthographical conventions. 94 Alejandro Augusto Prieto Mendoza Table 1. Inventory of Kakataibo consonants Labial Dento- alveolar Palato- alveolar Palatal- retroflex Velar Glottal Stop p

t k ʔ <’> kʷ Nasal m n ɲ <ñ> Flap ɾ Affricate t͡ s t͡ ʃ Fricative s ʃ ʂ Approximant β̞ Table 2. Inventory of Kakataibo vowels Front Central Back High i ɨ <ë> ɯ Mid e ɤ Low a Syllables in Kakataibo are based on the shape (C)V(C), resulting in four possibilities: V, CV, VC, and CVC. All these syllables can appear in any posi- tion. Only /n/, /s/, /ʃ/, /ʂ/, and /ʔ/ in some restricted cases, can fill the coda (Zariquiey 2011, 2018). As I show in §5, /n/ and these sibilant fricatives play a central role in Kakataibo rhyme and alliteration. As regards fricatives, these appear in syllable-final, word-final, and initial positions. Of these three, /ʃ/ is the most restricted in the lexicon and usually appears next to the vowel /i/ (Zariquiey 2018: 72–77). Nonetheless, /ʃ/ does appear next to other vowels, in a few cases. The affricates /t͡ s/ and /t͡ ʃ/ can only appear in syllable-initial posi- tion, even after a closed syllable. The phoneme /t͡ s/ might be realized as [t͡ ʃ]; however, this latter form is unpredictable. Finally, vowels and are of very low productivity in the language and are recent innovations in Kakataibo. According to Zariquiey (2018), Shell’s (1987) account gives only nine words with ; while words with are relatively more numerous. Shell (1975; in Zariquiey 2018: 84) reconstructs and as *aya and *awa, respectively; however, there are also words that cannot be explained by such origin and have no cognates in other Panoan languages. 95Alliteration and Rhyme in the Traditional Kakataibo Chants of Emilio Estrella 2. My data and its limitations My data comes from two sources: the PUCP Archive of Indigenous Peruvian Languages (Zariquiey 2014) and my own fieldwork during 2013–2017. All the songs of Emilio Estrella that I analyzed were recorded during 2010 to 2017 in the Native Community of Yamino (Aguaytía district, Padre Abad prov- ince, Ucayali region). I have analysed approximately two hours of Traditional Kakataibo chants (2500 lines approximately) from four men and four women, and of the two most divergent varieties of the Kakataibo language. I highlight this relatively small corpus in order to emphasize that rhyme and alliteration, specifically alliteration, are sporadically used in Emilio Estrella’s composition and, although other Kakataibos might use them, I have not documented any examples like the ones shown in this article on the composition of a different Kakataibo. In fact, my analysis of Emilio Estrella’s alliteration is based on four passages of a total of 14 lines. I accept that such a situation might create a problem of limitation in my analysis; however, with the years I have dedicated to Kakataibo verbal art, I can affirm that alliteration and rhyme are sporadic for Kakataibo singing composition in general. Alliteration was part of the sing- ing style of Emilio Estrella and he was one of Kakataibo sabios most dedicated to traditional singing. Finally, my examples come from no bana ‘iti and no xakwati chants, and in order to fill gaps in my analysis of Emilio’s alliteration and rhyme, I also present examples from two other Kakataibos: Irma Odicio and Roberto Angulo. 3. Traditional Kakataibo chants and Emilio Estrella In previous works (2019, 2021), I have discussed different topics on traditional Kakataibo chants, such as their learning, the available recordings and data- bases, the number of types of chants, the principal differences across them, among other themes. Therefore, I choose only to present relevant topics to understand alliteration and rhyme in Emilio Estrella’s composition. Traditional Kakataibo chants are always improvised in each different act of enunciation (Déléage 2020), which positions them along with areal trends of Amazonian singing composition (Beier et al. 2002). As Erwin Frank said, Kakataibos never “sing a song in the same manner as on other occasions [...] However, singers believe that their different versions of a song are “absolutely always the same”” (Frank 1994: 227). Thus, although the content is improvised and varies in each performance, the discursive technique remains stable across Kakataibos. 96 Alejandro Augusto Prieto Mendoza In relation to the linguistic character of the discursive technique, traditional Kakataibo chants carry a high metaphorical content, manifest reduced syn- tax, restricted/specialized vocabulary and formulaic expressions. They are composed using the poetic forms of semantic parallelism, enjambment, and repetition. It is of particular interest that a group of chants, in which no bana ‘iti and ño xakwati are found, must follow a meter specific to each of them (Prieto 2021). I present in (1) the no bana ‘iti and ño xakwati meters. (1) a. ∪ ∪ | ∪ ∪ | (∪) –– no bana ‘iti b. ∪ ∪ | ∪ ∪ | ∪ ∪ | – ño xakwati Following Prieto (2021), these meters indicate that a line or verse in a no bana ‘iti chant, for example, is made up of three metrical subgroups and a total of eight moras. The first two subgroups can be realized in two ways: with two short vowel syllables (∪ ∪) or one long vowel syllable (–), while the third subgroup is realized with one short vowel syllable plus one extra-long (three mora) vowel syllable (∪ ––) or only with this last extra-long vowel syllable and omitting the short one (––). In this way, the number of syllables can vary from line to line and, in principle, a line can be composed of at least three syllables and maximum of six, although, in general, five or six syllable lines are usual. Lastly, if the number of eight moras is not completed, vowels are lengthened following the prosody of the language (Prieto 2021). As regards the song no bana ‘iti, this male song is characterized by an exaltation of the singer through the figure of an ‘inu ‘jaguar’ that cannot be deceived, fooled or advised.1 In addition, the singer highlights his physical abilities, his mastery with shotguns and his hunting expertise. There are auto- biographical episodes that have marked the life of the singer, such as military service, travels, etc., which resembles no bana ‘iti to the caqui caqui Yaminawa or yama yama Sharanahua, both Panoan (Déléage 2007, 2008). The mythical figure of the ‘inka plays an important role in the imagery of a no bana ‘iti. The ‘inka is associated with metal tools, boats, planes; precious objects considered foreign and brought by mestizos (Brabec de Mori 2019; Frank 1990). This chant is usually sung at sunset or very early in the morning. It begins with 1 The ‘ësëti ‘to advise’ is one the most important cultural practices of Kakataibos, although its highly endangered. Traditionally, parents used to wake up their children at the middle of the night to teach them for hours the Kakataibo way of life (Frank 1994). 97Alliteration and Rhyme in the Traditional Kakataibo Chants of Emilio Estrella the formulaic phrase chira bakë xanu ‘sister’. I illustrate these points in the following excerpt (2). (2) chira bakë xanu sister chira bakë xanu sister ‘ësëtima ‘inu I am a jaguar that cannot be advised ‘ësëtima ‘inu I am a jaguar that cannot be advised parantima ‘inu I am a jaguar that cannot be fooled [...] ëxira kana I... nirakëakën I have gotten up bashi ‘inu bakë son of the jaguar of the mount chira ‘inu bakë son of the jaguar of the mount ëxira kana I... nirakëakën I have gotten up nirakëakën I have gotten up As regards the song ño xakwati, this male song is sung before hunting in order to attract the ño ‘peccary’ and to guarantee a good hunt. Kakataibos, as other Panoan groups, raise peccaries and when these peccaries are sufficiently grown to eat, families or neighboring groups gather to kill it. During this festivity, no longer practiced, men also sing ño xakwati (Montalvo Vidal, 2010). In addi- tion, the singer refers to himself through the metaphor of the runu ‘snake’, which is chasing the peccary and its brood across the rivers while sharpening the tips of his spears and arrows in (11), I offer a full but short ño xakwati. Finally, I introduce Emilio Estrella Logía (1925/1935? – 2020), a Kakataibo sabio of the Native Community of Yamino, who actively worked in the docu- mentation, maintenance, and resistance of the Kakataibo language and culture. With the advice of Emilio Estrella, the first grammar of the Kakataibo language was written, as well as dictionaries, academic articles, theses, among other works. A generation of students, linguists, and anthropologists learned from him several aspects of Kakataibo culture. In 2019, the Ministry of Culture of Peru awarded him with the recognition of Meritorious Person of Culture, due to his important contribution to the linguistic documentation of the Kakataibo language. Emilio Estrella was indeed a Kakataibo sabio, he knew his culture like no other Kakataibo and, as far as I am concerned, his chants exhibit the greatest complexity and imagery if we judge them in relation to the aesthetics of Kakataibo verbal art. Given this, I hope the present article serves as a tribute to his work and helps to recognize his voice within Peruvian verbal and literary 98 Alejandro Augusto Prieto Mendoza tradition. Emilio Estrella passed away on July 27, 2020 during the first year of the COVID-19 pandemic, which tragically hit the Peruvian Amazon, the Ucayali region and the indigenous communities who live there.2 4. Typology of rhyme and alliteration Following Fabb (1999, 2003), alliteration and rhyme are forms of sound paral- lelism where certain parts of the syllable are repeated according to what count as “same” in a given poetic tradition: “alliteration is repetition of a coherent sequence of segments which begins with an onset; rhyme is repetition of a coherent sequence of segments which begins with a nucleus (Fabb 1999: 227). Peust (2014) described the identity or sameness as ‘rhyme phonology’, con- sidering that “the partitioning of the sound space implied by rhyme identity may differ from the partitioning of the sound space as normally practiced by phonologist (“ordinary phonology”) [...] in other words, [rhyme phonology] is an underdifferentiated version of ordinary phonology”. For example, in Swiss German, plosives and affricates sharing the same manner of articulation are treated as equal (Peust 2014: 366–367). Then, the identity of the segments seems to be language specific, even to the extreme of being dictated individu- ally. In addition, Fabb (2003) summarizes various proposal about what counts as “the same”: (i) sameness can only be captured at the level of underlying phonological representations; (ii) sameness as the sharing of features, maybe involving underspecified underlying phonemes; (iii) and sameness as a way to confirm or reveal an underlying form. Another important parameter is the size of the “identical” section, which is language specific too. In any case, it is relative to the phonology of the language in question that the identity and the size of the rhymed or alliterated segments is established or broken. Besides that, rhyme can be classified in three types according to the location of the “identical segments”: initial rhyme, internal rhyme and end rhyme. All these three types are attested across world languages. Prototypical rhyme involves the nucleus plus the coda, while variants concern only the nucleus and very rarely just the coda (Fabb 1999). Peust (2014) treats alliteration as a subtype of initial rhyme; however, as Fabb (Fabb 1999, 2003, 2022) proposes, rhyme and alliteration have important differences and can be defined as two separate poetic forms. Alliteration is attested in Germanic (Goering 2016, 2 The Ministry of Culture of Peru officially notified the death of Emilio Estrella on July 27, 2020. Sebastián Castañeda (2021) documented the burial of Emilio for Mongabay Latam. 99Alliteration and Rhyme in the Traditional Kakataibo Chants of Emilio Estrella 2020; Jakobson 1963) Finnic (Frog, Stepanova 2011), and Somali verbal arts (Andrzejewski 2011; Banti, Giannattasio 1996; Musse Jama 2021; Orwin 2011), as in other traditions (Fabb 1999; Roper 2011). In some Somali song meters, the initial sound of at least one word is alliterated in each line and the alliter- ated sound must be the same throughout the entire poem. However, there are some rules such as a poet should not use grammatical words like pronouns or particles and must avoid repeating the same word in nearby lines (Banti, Giannattasio 1996: 84). Rhyme and alliteration can be sporadic or systematic. Sporadic rhyme can be found in the situation in which “rhymed verses occur more frequently than could be expected by chance but still belong to the inventory of optional sty- listics effects, whereas “systematic rhyme” means that the rhyme has become a mandatory feature of poetry” (Peust 2014: 348). Alliteration, systematic or sporadic, is much rarer than rhyme cross-linguistically (Fabb 1999: 227), maybe because rhyme is easily borrowed across verbal traditions and end- rhyme is the most common type of rhyme, conditioned by the frequency of inflectional suffixes over prefixes (Dryer 2005; Peust 2014: 358). Finally, alliteration should not be understood as a subtype of internal rhyme, or rhyme in itself, because both poetic forms differ in three impor- tant aspects, following Fabb (2022): (i) rhyme location tends to be fixed, but alliteration is almost always free; (ii) the rhyme pattern can intersect, but an alliteration pattern requires adjacency; (iii) and rhyme is more capable of per- mitting loose similarity between sounds than is alliteration. To conclude, I use the following typological parameters to study alliteration and rhyme in Emilio Estrella’s chants: productivity (sporadic vs. systematic), location and identity of the parallel segments, and adjacency vs. intersection between lines. 5. Alliteration and rhyme in Emilio Estrella’s traditional singing 5.1. Alliteration Following what was proposed in the previous section, alliteration is a poetic form that Emilio Estrella used as a stylistic resource, sporadically; a resource that, despite not being systematic, was available to him. It is important to emphasize that other Kakataibo sabios I have worked with also use alliteration in the way I show in this part; however, their use is limited to one or two lines in a whole composition, unlike Emilio Estrella’s alliteration. 100 Alejandro Augusto Prieto Mendoza 5.1.1. Identity of the segments Following Peust (2014) and Fabb (1999, 2022), the “identity” or “sameness” of the segments may differ from ordinary phonology (Peust 2014); and same- ness can be sharing of features, maybe involving underspecified underlying phonemes (Fabb 1999). I consider this is the situation in Emilio Estrella’s alliteration. To support this assertion, let us look at the following passage (3). I do not gloss each line due to edition criteria.3 (3) [...] 73. ‘ën xabun ‘ë /x/ ‘my sister’ 74. ushin chaxu xaka /sh/ /ch/ /x/ /x/ ‘red deer skin’ 75. uxu chaxu xaka /x/ /ch/ /x/ ‘white deer skin’ 76. ‘ë xuka xuan /x/ /x/ ‘peel it for me’ 77. ‘ë minan chaxu /x/ ‘to my purple deer’ 78. ‘ë ushin chaxu /sh/ /ch/ /x/ ‘to my red deer’ 79. ushin chaxu xaka /sh/ /ch/ /x/ /x/ ‘red deer skin’ 80. ‘ë xuka xuan /x/ /x/ ‘peel it for me’ [...] First of all, each line within this passage has at least one fricative consonant, the retroflex palatal , /ʂ/, which is the only fricative segment appearing in every line at least once. Taking this into account, alliteration in this passage occurs interlineally, since all the lines in the example are related by the presence of the retroflex palatal fricative , /ʂ/, and intralinearly in lines 74–76, and 78–80. Secondly, alliteration in this passage (3) also is connected to semantic par- allelism and enjambment through lexical-phonological selection by similar identity. Let us now turn our attention to lines 74–75, which are composed by 3 I use the orthographic conventions of Table 1, section 1, to identify the alliterated conso- nants, instead of IPA conventions. [...] implies that there is a previous or a next line. I only mark lengthened vowels in examples (9) and (10) to illustrate the free order of alliteration. 101Alliteration and Rhyme in the Traditional Kakataibo Chants of Emilio Estrella semantic parallelism. This poetic form implies that two elements, prototypical names in Kakataibo verbal art have a semantic relationship, in this case, type of colors (Prieto Mendoza 2019). What is interesting here is that the element followed in line 75, and semantically related to ushin ‘red’ in line 74, also shows a fricative consonant, as is the case for uxu ‘white’. We can interpret this relationship not only semantically, but establishing the identity of what an alliterated consonant might be in Emilio Estrella’s singing. Thus, identity or sameness in Emilio Estrella’s alliteration is established by sharing of fricative features. This same situation happens between lines 78–79, which are com- posed by enjambment, having the added element xaka ‘skin’ in line 79 which also includes the fricative , /ʂ/. Thirdly, alliteration not only contributes to lexical selection, but is also a product of poetic forms. I say this due to the repeated lines 74–79, 76–80 and the repeated word xaka ‘skin’ in 74–75, 79; repetitions that create a greater alliterative sonority in (3). Note that this pair of repeated lines (74–79, 76–80) are also the ones that show a higher number of alliterative consonants in the passage. Finally, other segments show a lower degree of alliteration, as /k/ in lines 74–76, 79–80. I will explain this control of saturation in brief. See the next example (4). (4) [...] 27. rëchitë bëxin /ch/ /x/ ‘peeling reed’ 28. bëxin bëxin okin ka /x/ /x/ /k/ /k/ ‘peeling, peeling’ 29. paka xëta bëxin /k/ /x/ /x/ ‘peeling the tip of the bamboo’ 30. bëxin bëxin okin ka /x/ /x/ /k/ /k/ ‘peeling, peeling’ [...] Similar to the previous example, (4) begins with a line with the retroflex , which begins the chain of fricative repetitions. However, one might wonder if the affricate really participates in the alliteration as it is not properly a fricative. In this case, lines 27 and 29 of [rëchite] and [paka xëta] respectively are composed by semantic parallelism. In theory, any other word related to rëchitë ‘reed’ may have been used. My proposal is that alliteration conditions the selection of [paka xëta] as it has to continue with the “fricative identity” already established by in the first and second lines, but also it must create 102 Alejandro Augusto Prieto Mendoza a fricative sonority between the slots of the semantic parallel lines – as in lines 74–75 of example (3) –, thus treating the affricate as a “fricative” in an expanded sense. At the same time, the selection of [paka xëta] also contributes to the control of alliteration by including the consonant , thus avoiding a saturated sonority of fricatives, and creating what I propose to call “minor alliteration” as opposed to the central alliteration of “fricatives”. As in the pre- vious example (3), other segments, apart from the obvious repetitions, might be alliterated in a passage, but with a lower degree in relation to “fricatives”. Then, the final line repeats line 28 closing the passage in a constant of sound parallelism. Now, see the next example (5). (5) [...] 36. xón pacha ‘ëo /x/ /ch/ ‘big red pomfret’ 37. xón ruti ‘ëo /x/ ‘big red pacu’ 38. uxu ruti ‘ëo /x/ ‘big white pacu’ 39. chirimipabian /ch/ ‘making them jump’ 40. xón ruti ‘ëo /x/ ‘big red pacu’ 41. chirimipabian /ch/ ‘making them jump’ 42. uxu sanin chiri /x/ /s/ /ch/ ‘white anchovy...’ 43. chirimipabian /ch/ ‘making them jump’ [...] This third example (5) exemplifies what has been proposed up to this point. The fricatives repeated across lines and within a line create a passage of allit- eration. The first line in the passage has the palatal retroflex /x/. And semantic parallelism participates in alliteration as a result of the lexical selection so producing “fricative” identity, as in lines 36–37–38. What is important in this new case is the employing of the third Kakataibo fricative , which closes the set of central “fricative” consonants used by Emilio Estrella. Let us look the following fourth example in (6): 103Alliteration and Rhyme in the Traditional Kakataibo Chants of Emilio Estrella (6) [...] 45. parantima baëxun /x/ ‘you cannot fool me, I am brave’ 46. ’ësëtima baëxun /x/ ‘you cannot advise me, I am brave’ 47. ‘axunima ‘ikinma /x/ ‘I will not do it to you, you cannot be with me’ 48. ‘axunira ‘axunwë ‘doing it right, do it’ /x/ /x/ Apart from what has already discussed so far, I include this last case in the present article because with it my examples are ended. As I said in previous sections, alliteration was of very low productivity in Emilio Estrella’s com- positions; a few lines if we compare it with the three main poetic forms of Kakataibo verbal art, and unique to him in that only he used it as in the four cases discussed so far. To summarize, Emilio Estrella’s alliteration follows these characteristics: (i) Alliteration occurs in passages of indeterminate length. (ii) Every line must have at least one retroflex fricative , /ʂ/. (iii) Alliteration treats fricative consonants and the affricate as the same. (iv) Alliteration can occur within a line and interlineally; the latter by adjacency. (v) Alliteration participates with semantic parallelism and enjambment by phonological-lex- ical selection. (vi) In order to avoid a saturated sonority of fricatives, other segments might be used in the alliterated periphery as opposed of central allit- eration of “fricatives”. Lastly, (vii) it is striking that the dento-alveolar affricate consonant did not appear in any of the documented cases; nonetheless, this situation can be due to database limitations. 5.1.2. Domain of alliteration As for the position of the alliterated fricative consonant within the word, the segment occurs in syllable onset as can be seen in each example discussed in the present article. An important fact to take into account is that the frica- tive consonants , /s, ʃ, ʂ /, can appear in coda position according to Kakataibo phonology. However, they do not appear in this latter position in Emilio Estrella’s alliteration. I formalize this in (7): (7) C1V(C) C1= /ʂ, s, ʃ/, t ͡ ʃ/ 104 Alejandro Augusto Prieto Mendoza Emilio Estrella’s alliteration seems to be restricted only to the onset of syllables with natural or non-lengthened vowels. The only apparent “contradiction” are monosyllables with alliteration, as in , example (5). However, monosyl- lables in Kakataibo must be phonologically analyzed as disyllabic for stress assignment criteria (Zariquiey 2018: 127–128). This seems to confirm that every monosyllable should be represented as (∪ ∪) for metrical purposes (Prieto Mendoza 2021). Secondly, regarding the order of the alliteration relative to the line, we need to remember the metric pattern of no bana ‘iti and ño xakwati in (1, a–b). Considering this, alliteration in the no bana ‘iti examples (3–5) appears freely in any metric subgroup and in any syllable, as shown in (8) – (*) represents an alliterated syllable: (8) * a. chirimipabiaaan ∪ ∪ | ∪ ∪ | ∪ ––– ‘making them jump’ * * * b. uxu sanin chiriii ∪ ∪ | ∪ ∪ | ∪ ––– ‘white anchovy...’ * * * c.‘ë ushin chaxuuu ∪ ∪ | ∪ ∪ | ∪ ––– ‘to my red deer’ It is the same situation for the ño xakwati examples (4–6), where alliteration appears in every metric subgroup and almost every syllable as shown in (9). However, there is no case of alliteration in the first syllable of the first metric subgroup or the fifth syllable of the third subgroup. This apparent restriction might be caused by database limitation considering that alliterations appears freely in no bana ‘iti meter. (9) * * * * a. bëxin bëxin okin kaa ∪ ∪ | ∪ ∪ | ∪ ∪ | –– ‘peeling, peeling’ * * * b. paka xëta bëxiin ∪ ∪ | ∪ ∪ | –– | –– ‘peeling the tip of the bamboo’ 105Alliteration and Rhyme in the Traditional Kakataibo Chants of Emilio Estrella 5.2. Rhyme As we saw in the previous section, alliteration is a sporadic poetic form used almost exclusively by Emilio Estrella in his singing composition. On the con- trary, rhyme is also sporadic in his chanting but with some levels of systematicity as I show in this section. In other words, Emilio Estrella was not the only Kakataibo that used rhyme and the principles of Emilio Estrella’s rhyme can be found in every other Kakataibo performer. In this section, I explain Kakataibo rhyme with examples from Emilio Estrella and in order to fill gaps in my analysis I also discuss examples from two other Kakataibos: Irma Odicio and Roberto Angulo. Let’s see the following example (10), which is a complete ño xakwati. (10) 1. nukën papa mia my father. 2. tama ruru piminun ...is going to invite you peeled peanut 3. nukën tita mia my mother... 4. tama chankë piminun ... is going to invite crushed peanut 5. mi ini rëkwënan come with your daughter 6. mi bakë rëkwënan come with your son 7. shinkun runu pibëtsin the snake is coming to eat you 8. basi runu pibëtsin the snake is coming to eat you 9. uantia kwëchinkin come snouting along the banks of Uantia river 10. banaoka kwëchinkin come snouting along the banks of Banaoka river 11. kwëchinkin kwëchinkin snouting, snouting 12. naneoka kwëchinkin come snouting along the banks of Naneoka river 13. rëchinkin rëchinkin sniffing, sniffing 14. naneoka kwëchinkin come snouting along the banks of Naneoka river 15. kwëchinkin kwëchinkin snouting, snouting 16. basi ño rëkwënan come with your mount peccary 17. ‘inu ño rëkwënan come with your big peccary 18. shinkun runu pibëtsin the snake is coming to eat you 19. basi runu pibëtsin the snake is coming to eat you 20. mi ini rëkwënan come with your daughter 21. uxu ño rëkwënan come with your white peccary 22. bunpa ño rëkwënan come with your dark peccary 23. mi ini rëkwënan come with your daughter 24. mi bakë rëkwënan come with your son 25. ñon pirui the peccary is coming 26. ñon pirui the peccary is coming 27. ñon piruikiri the peccary is coming 106 Alejandro Augusto Prieto Mendoza The first aspect to highlight is that this ño xakwati in (10) is obviously and very productively composed according to a principle of semantic parallelism and repetition. Then, one possibility is that the end of each line might be conditioned by the use of these two poetic forms, a very common fact among Amerindian verbal art that probably has led to proposal of the absence of rhyme (Fabb 2017; Hymes 1977; Edmonson 1971). Nonetheless, if we focus our attention from line 4 to 24, we can see that the final syllable of each line has a nasal coda . What is crucial is that this nasal coda is not conditioned by any poetic form, apart from the obvious repetitions, because it belongs, for example, to -nun ‘DS/A/P: POE’ or -bëtsin ‘coming.trans’. In other words, the presence of this nasal coda is not a product of repetition, neither of morphology, but an example of rhyme itself. Taking this into account, the first principle of Kakataibo rhyme is that only the nasal segment can occupy coda at the last syllable of the line. This restriction is important consider- ing that, in principle, can occupy this position according to the Kakataibo syllable structure, but they cannot be in coda of end-line syllables for Kakataibo rhyme purposes. Furthermore, the last syllable of each line not only rhymes by that nasal coda, but also in other ways and parts of the syllable. For example, lines 4–5 share the same onset and coda, but not the nucleus, in a (NVN) rhyme. Similar situation between lines 8–9, which share the nucleus, , plus the coda in a (CIN) rhyme. I show these two cases in (11a) and (11b), respectively. (11) a. 4. tama chankë piminun ... is going to invite crushed peanut 5. mi ini rëkwënan come with your daughter b. 8. basi runu pibëtsin the snake is coming to eat you 9. uantia kwëchinkin come snouting along the banks of Uantia river Another point to note about (11a) and (11b) is that they both occur in contiguous lines in (11) and, as Fabb (1999) proposes, rhyme allows intersection or ABAB patterns, while adjacency is mandatory for alliteration. In the Kakataibo case, true rhyme, as opposed of rhyme in lists created by repetition or semantic parallelism, seems to be only by adjacency as in lines 4–5 and within vowel passages, as in lines 7–15. In other words, Kakataibo rhyme needs adjacency – perhaps because it is unpredictable in whether it will occur. Thus, two lines with different vowels in end-line syllables can rhyme as in (11a) and lines can rhyme over vowel passages of indeterminate length, as in 7–15 of example (10). Though there are apparent rhyme intersections these arise only as a side effect of semantic parallelism, repetition or enjambment. In any case, intersection in Kakataibo rhyme needs further investigation. 107Alliteration and Rhyme in the Traditional Kakataibo Chants of Emilio Estrella In summary, Kakataibo rhyme is of the end rhyme type. The domain of rhyme is the last syllable of the line and is centred on the nucleus of it. Only the nasal consonant can occupy coda position in end-line syllables. Kakataibo rhyme needs adjacency and rhyme is possible in coda, nucleus + coda, and onset + coda. There are other rhyme possibilities, such as onset + nucleus and onset + nucleus + coda, which I present later. Now consider the next example (12) from a no bana ‘iti. (12) 1. chira bakë xanu sister 2. chira bakë xanu sister 3. ‘ësëtima ‘inu I am a jaguar that cannot be advised 4. ‘ësëtima ‘inu I am a jaguar that cannot be advised 5. parantima ‘inu I am a jaguar that cannot be fooled [...] 60. xanun tuama ka not the son of another woman 61. ëxira kana I... 62. nirakëakën ...have gotten up 63. bashi ‘inu bakë son of the jaguar of the mount 64. chira ‘inu bakë son of the jaguar of the mount 65. ëxira kana I... 66. nirakëakën ...have gotten up 67. nirakëakën ...have gotten up As in the previously discussed song ño xakwati, in this song no bana ‘iti, the coda position is reserved for the nasal consonant in end-line syllables and, apart from identical end-line syllables by repetition, rhymes occur within vowel passages as in lines 1–5, 60–61, and 62–64. For example, the u-vowel passage in lines 1–5 shows identical rhyme between xanu e ‘inu in lines 1–2 and 3–5, being of the type onset + nucleus. The same situation holds for lines 60–61 and 62–64, whose end line syllables rhyme in onset + nucleus. It is important to note these three group of lines rhyme without being conditioned by repetition, as in example (10), confirming that end line syllables rhyme independently of other kinds of parallelism. See the next example (13) from a no bana ‘iti. 108 Alejandro Augusto Prieto Mendoza (13) [...] 79. ushin chaxu xaka red deer skin 80. ‘ë xuka xuan to me red deer 81. bëtsi baka uka to another river 82. bëtsi me uka to another land 83. suñuanën ‘apa the strong wind... 84. ‘apaakexa took him 85. puna puna buan the strong wind 86. buankëxa brought him 87. bakëxunbi ka being a child 88. tuaxunbi ka being baby 89. buankëxa brought him 90. bëtsi me uka to another land 91. buankëxa brought him 92. suñuanën nua the strong wind 93. nuamiakëxa made him flight [...] Kakataibo true rhyme within vowel passages can involve more than a few lines as shown in example (13). Apart from the repetitions and semantic parallelisms in (13), end-line syllables rhyme in the nucleus. Thus, apart from the mini- mum rhyme of vowel passages and nasal codas, I have presented the following rhyme types: (i) onset + nucleus, (12); onset + coda, (11a); and nucleus + coda, (11b). As we have seen, Kakataibo rhyme demands full identity of whichever segments are involved. It is important to note that there is no case of onset + nucleus + coda rhyme type that does not involve repetition in the chants of Emilio Estrella at my disposition, but this may be due to a database limitation. In order to conclude this section, I summarize in the following Table 3 the principal characteristics of Kakataibo rhyme and alliteration according to the typological parameters discussed in section 4. Table 3. Emilio Estrella’s rhyme and alliteration according to typology Rhyme Alliteration Line domain end-line syllable free Syllable domain nucleus (+ n) onset Identity identical segments fricatives and Pattern adjacency adjacency 109Alliteration and Rhyme in the Traditional Kakataibo Chants of Emilio Estrella 5.2.1. Mid vowels and possible counterexamples As I discussed in § 1, Kakataibo mid vowels are of very low productiv- ity, perhaps only a few cases in the entire lexicon. This situation also happens in Kakataibo traditional chats. In all my data from Emilio Estrella or any other Kakataibo, there is not a single case of an end-line syllable with an nucleus – there are a few cases with - and there is no case of a vowel pas- sage of or . Nonetheless, both vowels do appear normally in other positions. Cases with end-line syllable with are ‘ó ‘tapir’, no ‘foreigner, enemy’, ‘ëo ‘big’, tsoo(t)- ‘to.live’, kon- ‘to.get.used’, pënpëro ‘butterfly’, and ‘ió ‘new’. Of these, the first two are innovations produced by the contraction of *aw(a)>o (Shell, 1975). The third one is reconstructed by Olivera (2014) as ʔɨwa. Words tsoo(t)- ‘to.live’ and kon- ‘to.get.used’ are cases of enjambment, so their use is conditioned by meter as the line needs to be broken. Lastly, there is no reconstruction for pënpëro or ‘ió and I do not know if they have cognates in another Panoan language. Considering this, the only productive vowels for Kakataibo end rhyme are , making it like the protoPanoan vowel inventory (Shell 1975). Is it possible that behaves like in end-line syllables? In any case, Kakataibo mid vowels need more research. Regarding the restriction of nasal coda, I have only identified in my entire database four lines with a different coda, , in a bana tuputi by Irma Odicio, as shown in (14). I do not interpret this case as a counterexample but as a case of stylistic deviation that endows the composition with originality. (14) 7. ura bakanuax ura baka-nu-ax far river-LOC-PA:S ‘along that distant river’ 8. ura menuax ura me-nu-ax far land-LOC-PA:S ‘along that distant land’ [...] 18. ënë bakanuax ënë baka-nu-ax this river-LOC-PA:S ‘along this river’ 110 Alejandro Augusto Prieto Mendoza 19. ënë menuax ënë me-nu-ax this land-LOC-PA:S ‘along this land’ 5.3. Mid rhyme and Emilio Estrella To end this article, I would like to discuss one last example of a ño xakwati chant by Emilio Estrella, a unique example that illustrates his linguistic mastery. (15) 47. ‘axunima ‘ikima ‘a-xun-i-n=ma ‘i-kin=ma to.do-BEN-IMPF-1/2P=NEG to.be-ASSO=NEG ‘I will not do it to you, you cannot be with me’ 48. ‘axunira ‘axunwë ‘a-xun=ira ‘a-xun-wë to.do-BEN=INTF to.do-BEN-IMP ‘doing it right, do it’ Let us recap what this example offers us. Lines 47–48 are part of the allit- erative passage analyzed in (6), in which the retroflex palatal fricative is repeated intralinearly (as in the case of line 48) and interlineally. Now, consid- ering only line 47, this line shows two similar segments [nima] – /ninma/ and [kima] – /kinma/, which are only distinguished by the initial consonant /n/ or /k/ and whose similarity is not given by repetition of suffixes (except for the suffix =ma ‘NEG’). Then, lines 47–48 repeat a similar segment now through [nima] – /ninma/ and /nira/, which in their phonetic realization end up being [nima] – [nira] and so more alike; these latter two segments do not resemble each other by repetition either. This is a fact that illustrates a game of rhymes both phonologically and phonetically. Likewise, if we look at the position of such segments in lines 47–48, the similarities continue since the segments in question are in exactly the same position. I formalize this in (16) – I have underlined the parts of the similar segments. http://to.do http://to.do http://to.do 111Alliteration and Rhyme in the Traditional Kakataibo Chants of Emilio Estrella (16) ∪ ∪ | ∪ ∪ | ∪ ∪ | – ‘axu/nima / ‘iki/maaa ∪ ∪ | ∪ ∪ | ∪ ∪ | – ‘axu/nira /‘axun/wëë Lastly, in order to avoid a saturated sonority, since all the words end in a similar way, Emilio Estrella breaks the repetition in the last part of line 48 by using the Shipibo-Konibo suffix –wë ‘IMP’.4 Thus, this final example shows that Emilio Estrella can rhyme a segment in the middle of a single line and continue to rhyme that segment in the next line and in the same mid position; this opens up the creative possibilities of Kakataibo rhyme. 6. Final remarks As far as I am concerned, this is the first documented case of alliteration and rhyme for an indigenous Amazonian group, despite its previously postulated absence (Hymes 1977; Edmonson 1971). Apart from the novelty, Amazonian verbal art is still an under-researched area in modern linguistics, so we do not know how common or uncommon are alliteration and rhyme in Amazonia, or more generally in the Americas. I hope this article opens up new horizons in verbal art typology and Amazonian linguistics. My principal findings are as follows. For both poetic forms, Kakataibo consonants that can appear in coda position according to the syllable structure of the language play a central role in Emilio Estrella’s alliteration and Kakataibo rhyme. On one hand, alliteration is sporadic and Emilio Estrella is mainly the only Kakataibo that employs it in his composition. Furthermore, it is based on the repetition of an expanded notion of “fricative” consonants in a passage; the fricative consonants and the affricate count as the same. In this passage of alliteration, every line must have at least one retroflex fricative . Alliteration occurs in syllable onset and freely within a line and across lines, the latter by adjacency. However, it seems to be restricted only to the onset of syllables with natural or non-lengthened vowels. In order to avoid a saturated sonority of “frica- tives”, other consonants or vowels are used in what I call minor alliteration 4 Kakataibo traditional chants shows an important lexical and morphological Shipibo-Konibo imprint (Prieto Mendoza, Zariquiey 2018). 112 Alejandro Augusto Prieto Mendoza as opposed to the central alliteration of “fricatives”. Finally, the dento-alveo- lar affricate consonant did not appear in any of the documented cases; nonetheless, this situation can be due to data base limitations. On the other hand, Kakataibo rhyme is also sporadic, but other Kakataibos tend to use it. Its domain is the final syllable of the line and the nucleus of it; only the nasal segment can occupy the coda in a rhyme, and there is a unique case of mid-line rhyme. Kakataibo true rhyme, as opposed to rhyme in lists created by repetition or semantic parallelism, seems to be by adjacency and within vowel passages of indeterminate length. I have identified the following rhyme types: coda, nucleus + coda, and onset + coda. Intersection and the role of mid vowels in Kakataibo rhyme needs further investigation. To conclude, I hope the present article serves as a tribute to Emilio Estrella Logía as to Roberto Angulo and Irma Odicio, and to recognize their verbal art within Peruvian lyrical tradition.5 Glosses 1 ‘first person’ 2 ‘second person’ ASSO ‘associative’ BEN ‘benefactive’ DS/A/P ‘different subjects and objects’ LOC ‘locative’ IMPF ‘imperfective’ INTS ‘intensifier’ NEG ‘negative’ PA:S ‘participant agreement oriented to S’ POE ‘posterior dependent event’ trans ‘transitive’ 5 I would like to thank Prof. Nigel Fabb for his useful comments and grammatical checking of this article, as well to the anonymous readers. 113Alliteration and Rhyme in the Traditional Kakataibo Chants of Emilio Estrella References Andrzejewski, Bogumił W. 2011. Alliteration and scansion in Somali oral poetry and their cultural correlates. In: Journal of African Cultural Studies 23(1), 37–47. https://doi.org/10.1080/13696815.2011.581456 Banti, Giorgio; Giannattasio, Francesco 1996. Music and Metre in Somali Poetry. In: Hayward, Richard J.; Lewis, Ioan M. (eds.), Voice and power: Essays in honour of B. W. Andrzejewski (African Languages and Cultures. Supplement 3). London: School of Oriental and African Studies, 83–127. Beier, Christine; Michael, Lev; Sherzer, Joel 2002. Discourse Forms and Processes in Indigenous Lowland South-America: An Areal-Typological Perspective. In: Annual Review of Anthropology 31, 121–145. https://doi.org/10.1146/annurev.anthro.31.032902.105935 Brabec de Mori, Bernd 2019. The Inkas Still Exist in the Ucayali Valley – What We Can Learn from Songs. In: Rivera Andia, Juan Javier (ed.), Non-humans in Indigenous South America – Ethnographies of Indigenous Cosmologies, Rituals and Songs. New York, Oxford: Berghahn Books, 167–196. Brabec de Mori, Bernd; Seeger, Anthony 2013. Introduction: Considering Music, Humans, and Non-humans. In: Ethnomusicology Forum 22(3), 269–286. https://doi.org/10.1080/17411912.2013.844527 Castañeda, Sebastián 2021. Dolor y rituales ancestrales: retratos de la pandemia en la Amazonía peruana. In: Mongabay Latam, August 18. https://es.mongabay. com/2021/08/retratos-pandemia-amazonia-peruana-reportaje-fotografico/ Déléage, Pierre 2007. A Yaminahua Autobiographical Song: Caqui Caqui. In: Tipití: Journal of the Society for the Anthropology of Lowland South America 5(1), 79–95. https://digitalcommons.trinity.edu/tipiti/vol5/iss1/5 Déléage, Pierre (2008). La tradition des autobiographies chantées chez les Sharanahua (Amazonie occidentale) / Traditional Autobiographical Songs among the Sharanahua (Amazonie occidentale). In: Bulletin de l’Institut Français d’Etudes Andines 37(3), 535–551. https://doi.org/10.4000/bifea.3011 Déléage, Pierre 2020. Toward an Epidemiology of Ritual Chants. In: Social Analysis 64(3), 113–144. https://doi.org/10.3167/sa.2020.640307 Dryer, Matthew S. 2005. Prefixing versus Suffixing in Inflectional Morphology. In: Haspelmath, Martin; Dryer, Matthew; Gild, David; Comrie, Bernard (eds.), The World Atlas of Language Structures. Oxford: Oxford University Press, 110–113. https://doi.org/10.1080/13696815.2011.581456 https://doi.org/10.1146/annurev.anthro.31.032902.105935 https://doi.org/10.1080/17411912.2013.844527 https://es.mongabay.com/2021/08/retratos-pandemia-amazonia-peruana-reportaje-fotografico/ https://es.mongabay.com/2021/08/retratos-pandemia-amazonia-peruana-reportaje-fotografico/ https://digitalcommons.trinity.edu/tipiti/vol5/iss1/5 https://doi.org/10.4000/bifea.3011 https://doi.org/10.3167/sa.2020.640307 114 Alejandro Augusto Prieto Mendoza Edmonson, Munro S. 1971. Lore: An introduction to the science of folklore and literature. New York: Holt. Epps, Patience; Michael, Lev 2017. The Areal Linguistics of Amazonia. In: Hickey, Raymond (ed.), The Cambridge Handbook of Areal Linguistics. Cambridge: University Press. https://doi.org/10.1017/9781107279872 Fabb, Nigel 1999. Verse constituency and the locality of alliteration. In: Lingua 108(4), 223–245. https://doi.org/10.1016/S0024-3841(98)00054-0 Fabb, Nigel 2003. Linguistics and Literature. In: Aronoff, Mark; Rees-Miller, Janie (eds.), The Handbook of Linguistics. Oxford: Blackwell, 446–465. https://doi.org/10.1002/9780470756409.ch18 Fabb, Nigel 2017. Poetic parallelism and working memory. In: Oral Tradition 31(2), 355–372. https://doi.org/10.1353/ort.2017.0014 Fabb, Nigel 2022. Rhyme and alliteration are significantly different as types of sound patterning. In: Fabb, Nigel; Sykäri, Venla (eds.), Rhyme and Rhyming in Verbal Art, Song and Language. (Studia Fennica Folkloristica 25). Helsinki: Finnish Literature Society (SKS), 155–171. https://doi.org/10.21435/sff.25 Fox, James 2014. Explorations on Semantic Parallelism. Canberra: Anu Press. https://doi.org/10.22459/ESP.07.2014 Franchetto, Bruna; Stenzel, Kristine 2017. Amazonian narrative verbal arts and typological gems. In: Stenzel, Kristine; Franchetto, Bruna (eds.), On this and other worlds. Voices from Amazonia. Berlin: Language Science Press, 1–19. Frank, Erwin H. 1990. Mitos de los uni de Santa Marta. Enex ca Santa Martanu ice unin bana icen. Cayambe, Ecuador: ABYA-YALA. Frank, Erwin H. 1994. Los Uni. In: Granero, Fernando Santos; Barclay, Frederica (eds.), Guía etnográfica de la alta Amazonía. Quito: FLACSO, 129–237. Frog; Stepanova, Eila 2011. Alliteration in (Balto-)Finnic Languages. In: Roper, Jonathan (ed.), Alliteration in Culture. Basingstoke: Palgrave Macmillan, 195–218. https://doi.org/10.1057/9780230305878_13 Goering, Nelson 2016. The linguistic elements of Old Germanic metre: phonology, metrical theory, and the development of alliterative verse. In: Transactions of the Philological Society 114(2), 171–197. https://doi.org/10.1111/1467-968X.12079 Goering, Nelson 2020. (Proto-)Germanic Alliterative Verse: Linguistic Limits on a Cultural Phenomenon. In: Friedrich, Matthias; Harland, James M. (eds.), Interrogating the “Germanic”. Berlin: De Gruyter, 241–250. https://doi.org/10.1515/9783110701623-013 https://doi.org/10.1017/9781107279872 https://doi.org/10.1016/S0024-3841(98)00054-0 https://doi.org/10.1002/9780470756409.ch18 https://doi.org/10.1353/ort.2017.0014 https://doi.org/10.21435/sff.25 https://doi.org/10.22459/ESP.07.2014 https://doi.org/10.1057/9780230305878_13 https://doi.org/10.1111/1467-968X.12079 https://doi.org/10.1515/9783110701623-013 115Alliteration and Rhyme in the Traditional Kakataibo Chants of Emilio Estrella Hymes, Dell 1977. Discovering oral performance and measured verse in American Indian Narrative. In: New Literary History 8(3), 431–457. https://doi.org/10.2307/468294 Jakobson, Roman 1963. On the so-called vowel alliteration in Germanic verse. In: STUF – Language Typology and Universals 16(1/4), 85–92. https://doi.org/10.1524/STUF.1963.16.14.85 Michael, Lev 2019. Lines in Nanti karintaa chants: An areal poetic typological perspective (An essay in honor of Joel Sherzer). In: Cadernos de Etnolinguistica 7, 56–62. Montalvo Vidal, Abner 2010. Los kakatai. Etnia amazónica del Perú. Lima: Instituto del Bien Común. Musse Jama, Jama 2021. Tidcan: Multiple Alliteration of Somali Songs – New Insights. In: International Journal of Literature and Arts 9(3), 124–142. https://doi.org/10.11648/J.IJLA.20210903.13 Oliveira, Sanderson Castro Soares 2014. Contribuições para a Reconstrução do Protopáno. PhD Thesis, Universidade de Brasília. https://repositorio.unb.br/ bitstream/10482/17129/1/2014_SandersonCastroSoaresdeOliveira.pdf Orwin, Martin 2011. Alliteration in Somali Poetry. In: Roper, Jonathan (ed.), Alliteration in Culture. Basingstoke: Palgrave Macmillan, 219–230. https://doi.org/10.1057/9780230305878_14 Peust, Carsten 2014. Towards a typology of poetic rhyme. In: Grossman, Eitan; Haspelmath, Martin; Richter, Tonio Sebastian (eds.), Egyptian-Coptic linguistics in typological perspective. Berlin, Boston: De Gruyter/Mouton, 341–385. Prieto Mendoza, Alejandro Augusto 2019. Semantic Parallelism in Traditional Kakataibo Chants. In: Open Linguistics 5(1), 383–404. https://doi.org/10.1515/opli-2019-0021 Prieto Mendoza, Alejandro Augusto 2021. Meter in Traditional Kakataibo Chants. In: Studia Metrica et Poetica 8(1), 117–138. https://doi.org/10.12697/smp.2021.8.1.04 Prieto Mendoza, Alejandro Augusto; Zariquiey, Roberto 2018. Shipibo loans in traditional kakataibo songs: lexicon, morphosyntax and compositional strategies. Paper presented at Amazonicas VII, Baños, Ecuador. https://www.academia.edu/37247913 Roper, Jonathan 2011. Introduction: Key Topics in the Study of Alliteration. In: Roper, Jonathan (ed.), Alliteration in Culture. Basingstoke: Palgrave Macmillan, 1–20. https://doi.org/10.1057/9780230305878_1 https://doi.org/10.2307/468294 https://doi.org/10.1524/STUF.1963.16.14.85 https://doi.org/10.11648/J.IJLA.20210903.13 https://repositorio.unb.br/bitstream/10482/17129/1/2014_SandersonCastroSoaresdeOliveira.pdf https://repositorio.unb.br/bitstream/10482/17129/1/2014_SandersonCastroSoaresdeOliveira.pdf https://doi.org/10.1057/9780230305878_14 https://doi.org/10.1515/opli-2019-0021 https://doi.org/10.12697/smp.2021.8.1.04 https://www.academia.edu/37247913 https://doi.org/10.1057/9780230305878_1 116 Alejandro Augusto Prieto Mendoza Severi, Carlos 2008. El Yo-memoria. Una nueva aproximación a los cantos chamánicos amerindios. In: Cuicuilco 15(42), 11–28. https://www.scielo.org.mx/pdf/cuicui/v15n42/v15n42a2.pdf Shell, Olive A. 1975. Estudios Pano III: Las Lenguas Pano y su Reconstrucción. Yarinacocha: Instituto Lingüístico de Verano. Shell, Olive A. 1987. Vocabulario cashibo-cacataibo. (Serie Lingüística Peruana 23.) Lima: Instituto Lingüístico de Verano. Skilton, Amalia 2017. Esoteric morphology: vocable affixes in Máíhɨk̃i shamanic song. In: Memorias Del Congreso de Idiomas Indígenas de Latinoamérica VII. Austin: University of Texas, 29–31. Tedlock, Dennis 1977. Towards an Oral Poetics. In: New Literary History 8(3), 507–519. https://doi.org/10.2307/468297 Tedlock, Dennis 1983. The Spoken Word and the Work of Interpretation. Philadelphia: University of Pennsylvania Press. Tedlock, Dennis 1987. Hearing a voice in an ancient text: Quiché Maya poetics in performance. In: Sherzer, Joel; Woodbury, Anthony C. (eds), Native North American Discourse. Cambridge: Cambridge University Press, 140–175. Tedlock, Dennis 2010. 2000 Years of Mayan Literature. Berkeley, Los Angeles, and London: University of California Press. Zariquiey, Roberto 2011a. A grammar of Kashibo-Kakataibo (Issue Abril). PhD Thesis. La Trobe University. Zariquiey, Roberto 2014. Canciones tradicionales (kakataibo). In: Archivo Digital de Lenguas Peruanas. http://repositorio.pucp.edu.pe/index/handle/123456789/35164 Zariquiey, Roberto 2018. A Grammar of Kakataibo. (Mouton Grammar Library 75). Berlin: De Gruyter Mouton. https://www.scielo.org.mx/pdf/cuicui/v15n42/v15n42a2.pdf https://doi.org/10.2307/468297 http://repositorio.pucp.edu.pe/index/handle/123456789/35164