Linguistic evidence and Tupi-Guarani/Cariban contacts in the Eastern Guianas: Wajãpi kasi ‘be strong’, pipi ‘father’s sister’ and kasuru ‘pearl’

Evidências linguísticas e contato Tupi-Guarani/Carib nas Guianas orientais: Wajãpi kasi ‘ser forte’, pipi ‘irmã do pai’ e kasuru ‘pérola’

Abstract

This brief paper proposes a Cariban etymology for the Wajãpi (Tupi-Guarani) verb -kasi ‘be strong’, which has been incorrectly suggested as being of Tupi-Guarani provenance. It is proposed that the source language for this diffusion event was Apalai, and not Wayana, as is usually claimed when the issue of the Cariban influx into Wajãpi is addressed. Two other Cariban loans are discussed, pipi ‘father’s sister’ and kasuru ‘pearl’, and in these cases as well, it is shown that diffusion from Apalai, not Wayana, is the more likely scenario. Implications of the findings are discussed.

Keywords
Etymology; Language contact; Tupi-Guarani languages; Cariban languages

Resumo

Este breve trabalho propõe uma etimologia Carib para o verbo -kasi ‘ser/estar forte’ da língua Wajãpi (Tupi-Guarani), que já foi incorretamente tratado como tendo uma origem Tupi-Guarani. A língua identificada como fonte do empréstimo é o Apalai, e não o Wayana, como é usualmente proposto quando o influxo de elementos Carib em Wajãpi é considerado. Dois outros empréstimos de línguas Carib são considerados: pipi ‘irmã do pai’ e kasuru ‘pérola’, e também nestes casos argumentamos que difusão a partir da língua Apalai é o cenário mais provável. Algumas implicações destas propostas são discutidas.

Palavras-chave
Etimologia; Contato linguístico; Línguas Tupi-Guarani; Línguas Carib

INTRODUCTION

Although a number of lexical and grammatical parallels between Tupian (almost exclusively, Tupi-Guarani) and Cariban languages have been pointed out in the literature (most notably in Rodrigues, 1985), there has been very little in the way of careful historical linguistic investigation of the nature and significance of these parallels (see Meira & Muysken, 2017; Rose & Renault-Lescure, 2008; Rose, 2012 for exceptions). This brief paper offers a modest contribution in this direction, by advancing a Cariban etymology for the Wajãpi verb -kasi ‘be strong’, and by illustrating the methodological standards that should be applied for the rigorous investigation of language contact involving Tupi-Guarani and Cariban languages, moving beyond the programmatic and sketchy suggestions advanced in the pioneering work of Rodrigues (1985).¹

WAJÃPI -KASI ‘BE STRONG’ << APALAÍ KASE ‘FAT, GREASE’

Based on Wajãpi kasi ‘be strong’, Jensen (1998, p. 501) postulated a Proto-Tupi-Guarani (PTG) etymon *kasí, with the same meaning. The etymon in question is proposed nowhere else in the comparative literature on Tupi-Guarani (TG) languages (e.g., Lemle, 1971; Schleicher, 1998; Mello, 2000), and, as I argue below, for good reason, as it is not of PTG provenance. Nonetheless, we can go beyond merely stating a foreign origin for this term. In fact, more specific claims about the nature of Tupi-Guarani/Cariban contacts in the eastern Guianas can be motivated on the basis of a carefully proposed etymology for this term.

Setting aside the problematic postulation of PTG *s (see e.g., Carvalho, 2022), the simple absence of cognates of this form elsewhere in the family precludes its reconstruction at the PTG level, a fact that, however, has not been a deterrent to the proposal of internal etymologies. Grenand’s (1989) Wajãpi-French dictionary gives the meaning ‘force, fort’ [strength, strong] for Wajãpi kasi, and exemplifies its use with the very frequent expression nɛkasi ‘Je n’ai pas de force, je suis faible’ [I don’t have strength, I’m weak].² She proposes an internal etymology for the form, one having ka ‘graisse’ [fat] as its base, further adding that the Wajãpi associate having body fat with physical strength (Grenand, 1989, p. 221). In Grenand (1989, p. 204), kasi is given under ka ‘graisse, être gras’ [fat, be fat], as a derivative. Needless to say, the proposed etymology is problematic, as Wajãpi ka is a reflex of PTG *-kap ‘fat’ (Jensen, 1984, p. 33), and -si, in -kasi, remains, in any case, unaccounted for. Finally, the presence of the coronal fricative s in any Wajãpi wordform constitutes in itself an indication of its foreign status (e.g., kɨsɛ ‘knife’, likely from Nheengatu; Jensen, 1984, p. 121),³ as the PTG affricates have either h or ∅ in the Wajãpi varieties (Carvalho, 2022).

It has been assumed that Wayana is the main source for Cariban loanwords attested in Wajãpi, one of the few Tupi-Guarani languages (alongside Zo’é and Teko/Emerillon) spoken in the Guianas (see e.g., Grenand, 1980; Jensen, 1984; Rodrigues, 1985). Grenand (1980, pp. 27-28) considers Galibi to be the more important source of Cariban loanwords into Wajãpi, followed, in order of importance, by Wayana and then by Apalaí, the later being explicitly described as having only a minor influence (Grenand, 1980, p. 27). Later, in an often-cited appraisal of lexical and grammatical similarities noted among Tupi-Guarani and Cariban languages, Rodrigues (1985, pp. 392-393) claims that Wayana is the source of the most obvious Cariban loanwords found in Wajãpi.

Although no published study claims a Cariban provenance for Wajãpi kasi ‘be strong’, this is certainly the most plausible hypothesis, though, as far as a solid contact etymology is concerned, it constitutes merely a starting point. I propose that the Apalai noun kase ‘grease/fat (n.)’ is the ultimate source of the Wajãpi form under scrutiny. The semantic latitude between ‘grease/fat’, in the source form, and ‘be strong’, although not trivial, agrees with what Grenand (1989, p. 221) reports on the associations between body fat and strength for the Wajãpi. In any case, this disparity in meaning will be discussed in greater detail, along with other formal details on the Cariban side of the proposed etymology.

It is plausible to assume that Apalai kase ‘grease/fat (n.)’ is somehow related to the series of Panare kaʔ, kat; Bakairi gadɨ; Kuikúro katɨ; Hixkaryána katɨ, all given under the meaning ‘grease/fat’ in Meira & Franchetto (2005, p. 185). These suggest a Proto-Cariban (PC) etymon *katɨ (see Gildea & Payne, 2007, p. 46), formally and semantically identical to Proto-Taranoan (PT) *katɨ ‘fat’ (Meira, 2000, p. 156).⁴ Wayana, which the published literature considers the most likely source for Cariban loans in Wajãpi, has kat(ɨ), with the final vowel being often dropped (Camargo et al., 2009, pp. 71-72).⁵ Thus, on the one hand, if the Wayana form in question is assumed as a source for Wajãpi kasi ‘be strong’, one would be left with the unexplained shift -tɨ >> -si in the adapted loanword; if, on the other hand, Apalai kase ‘grease/fat (n.)’ is advanced as the source, a full contact etymology, taking care, as it should, of the origins of the source form as well, would have to explain how Apalai kase ‘grease/fat (n.)’ could fit in the series formed by reflexes of PC *katɨ ‘fat’. As we show below, this is indeed possible.

PC *-tɨ is regularly reflected as -tɨ in Apalai, as shown by the language’s reflexes for a sample of PC etyma including this particular syllable: PC *-petɨ ‘thigh’, *-wetɨ ‘feces (possessed allomorph)’, *-potɨ ‘beak’ (Gildea & Payne, 2007, pp. 56-61) > Apalai -petɨ, -etɨ, -potɨ. On more general phonological grounds, PC *ɨ is not among the vowels yielding spirantized or affricate reflexes of PC *t in daughter languages (see Meira & Franchetto, 2005). In sum, Apalai -kase cannot be a direct reflex of PC *katɨ ‘fat’. Note that Apalai is not alone in this regard, as other languages of the family, in particular those of the Circum-Roraima region, present comparable forms: Akawaio kasi’ [kasiʔ] ‘someone or something fat or thick’ (Stegeman & Hunter, 2014, p. 166) and Pemon kasek ‘grasa’ [grease/fat] (Armellada, 1943, p. 210). Formal developments other than regular (phonetically conditioned) sound change are needed for relating Apalai kase ‘grease/fat’ and PC *katɨ ‘grease/fat’.

A hypothesis that goes a long way towards accounting for the formal problem just mentioned, as well as for the semantic disparity between ‘grease/fat (n.)’ and ‘be fat/be strong’, appeals to analogical reshaping of the Apalai reflex of PC *katɨ ‘grease/fat’ on the basis of the adverbial⁶ construction derived by the *tɨ- and the suffix *-se. For Proto-Taranoan, for instance, Meira (2000, p. 156) reconstructs both *katɨ as a noun meaning ‘fat’, and the adverbial *tɨkatɨne, noting that the latter is “obviously derived” from the former. Meira (2000, pp. 115-116) reconstructs three proprietive (‘having’) adverbializers for PT, all with the same circumfix structure but differing in the form of the suffixal element: *tɨ- -ke, *tɨ- -ne and *tɨ- -ce. In the case of the Taranoan languages, the first is the most productive strategy, while the other two seem a bit more lexicalized. Such constructions are found throughout the family (see e.g., Derbyshire, 1999, p. 50), usually involving the affixation of these markers to a base noun, and resulting in a wordform whose meaning is that of ‘having’ or ‘being characterized by’ the referent of the relevant base noun.⁷ Some examples from Apalaí are given in (1), based on E. Koehn and S. Koehn (1986):⁸

(1)   Proprietive modifying constructions in Apalai (a)   t- ɨʔnapa -se   (adapted from E. Koehn & S. Koehn, 1986, p. 72)     nf poison -cmpl       ‘poisoned’     (b)   t -apuʔ -se   (adapted from E. Koehn & S. Koehn, 1986, p. 88)     nf -mold -cmpl       ‘molded’     (c)   t- onu -re   (E. Koehn & S. Koehn, 1986, p. 87)     nf- eye adjvzr       ‘seeing’     (d)   tɨ- pɨtɨ -e   (E. Koehn & S. Koehn, 1986, p. 87)     nf- wife -adjvzr       ‘with wife’  

(1)		Proprietive modifying constructions in Apalai
(a)		t-	ɨʔnapa	-se		(adapted from E. Koehn & S. Koehn, 1986, p. 72)
		nf	poison	-cmpl
		‘poisoned’
(b)		t	-apuʔ	-se		(adapted from E. Koehn & S. Koehn, 1986, p. 88)
		nf	-mold	-cmpl
		‘molded’
(c)		t-	onu	-re		(E. Koehn & S. Koehn, 1986, p. 87)
		nf-	eye	adjvzr
		‘seeing’
(d)		tɨ-	pɨtɨ	-e		(E. Koehn & S. Koehn, 1986, p. 87)
		nf-	wife	-adjvzr
		‘with wife’

The examples in (c) and (d) above illustrate the use of the ‘adverbializer/adjectivizer’ markers for deriving modifiers of nominal expressions (E. Koehn & S. Koehn, 1986, p. 87), while those in (a-b) exemplify the tɨ- ... -se construction, which, in Apalai, has a slightly wider distribution than is the case with cognate constructions elsewhere in the family (E. Koehn & S. Koehn, 1986, pp. 48-49).

For the base noun ‘grease/fat’ specifically and derived adjectival formations, Apalai uses the form tɨ-kase-re ‘fat’ for qualifying non-human referents (as in maʃipuri tɨkasere ‘fat tapir’), and a derivative of punu ‘body, flesh’, tu-punu-ne, for the modification of nouns having human referents (as in orutua tupunune ‘fat man’). The relationship between Apalai kase ‘grease/fat’ and PC *katɨ ‘grease/fat’ is here conceived according to Figure 1.

Figure 1
Proposed development for Apalai kase ‘grease/fat’.

Stage (a) above, likely characteristic of PC, has both the noun *katɨ ‘grease/fat’ and the adverbial formation *tɨ-katɨ-se, the latter with the meaning ‘be fat’ or ‘having fat’. In a second stage, (b), *tɨ-katɨ-se undergoes syllable reduction (Gildea, 1995), yielding *tɨ-kaʔ-se. Moment (c) has a reanalysis of *tɨ-kaʔ-se as *tɨ-kaʔse, with the loss of the morphological boundary between the base noun and the suffix *-se. Also necessary here in order to derive modern Apalai kase ‘grease/fat’ is the loss of the medial glottal stop, indicated by parentheses. The transition from stages (c) and (d) is crucially characterized by the analogical reshaping of *katɨ on the basis of *tɨ-ka(ʔ)se, possibly fostered by the possibly analysis of *tɨ-ka(ʔ)se as part of the nominal possessive paradigm, given the polysemy of *tɨ- as both the prefix found in a number of adverbial/adjective formations and the prefix marking a third person correferential possessor (tɨ-kase-rɨ ‘his/her own fat’). The modern Apalai modifying construction tɨ-kase-re ‘be fat, having fat’, is certainly innovative, with -re being employed after *-se was re-analyzed as part of the root.

Although the meaning ‘be strong’ of Wajãpi kasi matches most closely that of the adverbial tɨ-kase-re ‘be fat’, its form fits instead that of the base noun kase ‘grease/fat’. Borrowing kasi from -kase- in tɨ-kase-re would have required Wajãpi speakers enough command of Apalai to be able to parse this adverbial form, extracting -kase- and adopting with it the derived meaning ‘be fat’, later extended to ‘be strong’. At the present state of our knowledge about the nature of contacts between Wajãpi and Apalai speakers it is unclear whether this was characterized solely by the fleeting diffusion of a few open class items or unanalyzed wordforms, or whether, instead, a more or less stable bilingualism can be assumed. In any case, the participation of -kase ‘fat’ in the above-mentioned possessive paradigms in modern Apalai (ɨ-kase-rɨ ‘my fat’, a-kase-rɨ ‘your(sg) fat’, i-kase-rɨ ‘his/her fat’, and so on) certainly offers a morphological model from which the root -kase ‘fat’ could be abstracted. Gaps in any proposed account will remain until more is known about this particular contact situation, but the contact etymology advanced here is certainly preferable to any purely internal etymology, and it should be provisionally accepted, unless an improved account is offered for the etymology of Wajãpi kasi ‘be strong’.

MEANING AND FORM, AND FURTHER APALAI LOANS INTO WAJÃPI

There are other cases for which Apalai, and not Wayana, offers the best candidate for the source of Cariban loanwords in Wajãpi. The form pipi is attested in Wajãpi as a term of address for the father’s sister (Grenand, 1989, p. 360), and the term lacks cognates elsewhere among Tupi-Guarani languages (see Carvalho & Birchall, 2022). Although a form pipi is attested in both Wayana and Apalai as a kinship term, they have utterly distinct meanings in the two languages, and these differences show that Apalai, not Wayana, is the immediate source for Wajãpi pipi. Wayana pipi has the meaning “vocatif pour époux, époux de la sœur, frère aîné, fils aîné du frère (femme parlant) [vocative for husband, sister’s husband, older brother, brother’s eldest son (woman speaking)]” (Camargo et al., 2009, p. 109). Apalai, on the other hand, has pipi as a vocative for father’s sister (as well as to ‘mother brother’s wife’), used by both male and female ego (S. Koehn, 1975, p. 85), exactly as Wajãpi pipi. Despite being formally similar, Apalai pipi and Wayana pipi are likely not cognate, as evidenced by their meanings, and Wayana pipi is more probably a cognate of Apalai piʔpi instead, a vocative form used by women when addressing their elder brothers (E. Koehn & S. Koehn, 1995, p. 41). In fact, Wayana pipi itself has a ‘phonetic variant’ showing a glottal element: pipih (Camargo et al., 2009, p. 109). This glottal element is seemingly etymological, as it could be related to long vowels found in cognates of this form elsewhere, as in Tiriyó piipi ‘older brother’ (male and female ego) (Carlin, 2004, p. 140). In any case this makes Wayana pipi/pipih both formally and semantically idependent from Apalai pipi ‘father’s sister, mother brother’s wife’, the latter in turn being the best candidate for the immediate source for Wajãpi pipi ‘father’s sister’.

Many presumed Wayana loans in Wajãpi have the peculiarity of featuring a coronal fricative s (see e.g., Grenand, 1980, pp. 27-28; Rodrigues, 1985, p. 392). The problem is that outside of the context of a following or preceding i the fricative s was debuccalized in Wayana, but not in Apalai (Camargo, 1996; Tavares, 2006, pp. 19-20). Indeed, this has been explicitly pointed out in the literature as an important isogloss between the two languages (Camargo, 1996, pp. 20-21), and speakers seem to be well aware of this fact.⁹ For these reasons, whenever a presumed Cariban loanword is found in Wajãpi, Apalai is a more plausible candidate source for the loanword than Wayana. A good example is found in Wajãpi kasuru, or, in the Upper Oyapock variety, kasulu, a form whose meaning is ‘pearl’. Grenand (1989, p. 221) presents it as a Wayana loan, citing a Wayana source form kahulu, also with the meaning ‘pearl’. Modern sources on the language confirm the form kahulu, but give a more encompassing meaning: kahulu ‘rocaille, verroterie’ [rockery, glassware] (Camargo et al., 2009, p. 67).¹⁰ Here again, the proposed contact etymology faces a formal hurdle, as it is unclear why h would be adapted in Wajãpi as s, and once more the problem can be avoided by assuming a loan from Apalai instead: Apalai kasuru ‘miçangas’ [beads] (E. Koehn & S. Koehn, 1995, p. 25). Wajãpi preserves, for the meaning ‘beads’, the TG native form moʔɨrɨ, Apalai kasuru being thus adopted with the meaning ‘pearl’ only.

A cautionary note on any use of the h : s isogloss must be presented, however, and it relates to the chronology (and spatial diffusion) of the *s > h development in Wayana. One cannot exclude, in principle, the possibility that Cariban loans in Wajãpi showing a coronal fricative were borrowed from Wayana before the operation of said change - or, at least, from a dialect of Wayana still not subject to the change. On both the spatial (dialectal) and chronological sides of the question there are, still, a number of open issues, a few of which will be commented upon here. In the end, however, none of the gaps in our knowledge seem to decisively affect the conclusions of the present paper.

Jackson (1972), who described Wayana as spoken in the Tapanahoni river, Suriname, notes that s and h have merged in the speech of some of the Wayana of this region, but that these remain nonetheless distinct for the majority of the individuals. This might suggest that some varieties of Wayana were not affected by the debuccalization of the coronal fricative s. However, as noted by Tavares (2006, p. 23, fn. 17), whose grammar is based essentially on ‘Brazilian’ Wayana of the Paru river, the varieties of the Maroni (described by Camargo, 1996; Camargo et al., 2009) and of the Tapanahoni (described by Jackson, 1972) seem to retain s only when adjacent to i, that is, in the same context where this coronal fricative is found in the Paru variety, that dialect where the *s > h debuccalization seem to have come to completion. In any case, some of the earliest available materials on Wayana suggest that the varieties spoken in the Jari river - that is, those closer to the Wajãpi - were affected by the debuccalization of s quite early, which agrees with our assumption that any Wayana dialect working as a source for loans into Wajãpi would have lacked s in contexts other than of an adjacent i. The Wayana materials of Henri Coudreau were recorded on two occasions, first among the northernmost Wayana of the Maroni river (1887-1889), and then among the southern groups of the Jari river (1889-1890). A brief comparison of the materials, organized and published by Lucien Adam in Coudreau (1892), reveals that <s> and <ch> (the latter only before <i> and likely standing for IPA [ʃ]) in the northern dialect correspond to <h> in the Wayana of the Jari: ‘Gras’ [fat] <capsac> : <i-capehac>; ‘Malade’ [sick, ill] <yétoumsac> : <étoumhac>; ‘Cheveux’ [hair] <yomset> : <yomhet>; ‘Chemin’ [path] <chima> : <hema> (see Coudreau, 1892, pp. 16, 18, 43). Though Wayana dialectology still calls for a comprehensive and adequate treatment, including both the early written documents and the extant spoken varieties of the language, the facts as they stand seem at least compatible with the hypothesis that bona fide Wayana loans into Wajãpi should not be expected to have a coronal fricative, except when adjacent to i.

CONCLUSIONS AND THE BROADER CONTEXT

The present work sought to contribute to a rather important topic in the regional history of the indigenous languages and peoples of South America, that of the contacts between speakers of Tupi-Guarani and Cariban languages. It has argued for a specific set of etymological associations at the phonological, morphological, and semantic levels. One also hopes that it has illustrated the potential held by careful and detailed etymological investigations of contact phenomena in general, and of the diffusion of loanwords in particular, which are often approached without the necessary rigour, if not in a dismissive manner, by researchers (see Boček, 2013; Mailhammer, 2013; Carvalho, 2017 for comments and examples).

The specific conclusion of interest to the indigenous history of the Guianas - that is, that Apalai, and not Wayana, constitute the main source of the Cariban influx in Wajãpi - is consistent with what is known about the history of the relevant populations. In their migration coming from the lower Xingu river (see Gallois, 1980, pp. 55, 58-59), the Wajãpi would have met the Apalai before meeting the Wayana, as the former once dwelled in a more southern location, closer to the main course of the Amazon river (e.g., Farabee, 1924, p. 221; Frikel, 1957, p. 512).¹¹ Although the statement of the close cultural proximity between Apalai and Wayana - a virtual cliché in the Cariban literature - does reflect a fact (Frikel, 1957, p. 527; Basso, 1977, p. 11), linguistic evidence and, in particular, careful historical linguistic work, can help cast further light on the different historical trajectories of these otherwise very similar Cariban-speaking populations. It remains to be seen whether more general claims about Wayana influence on Wajãpi (see e.g., Jensen, 1984, p. 132) will be vindicated or not, but one hopes that the present contribution has underscored the potential offered by etymological research in this particular domain.

• 1
  In addition to the sources cited throughout the text, this paper relies on the author’s fieldwork on both Wajãpi (Tupi-Guarani) and Apalai (Cariban).
• 2
  The expression in question is formed by prefixing the pre-vocalic allomorph n- of the (verbal) negator prefix, plus the Set II (or stative) first person singular prefix ɛ-. The Wajãpi also employ the expression nɛkasi with the meaning ‘I am hungry’ (author’s field notes). For the notion of ‘Set II’ person marker, see the overview in Jensen (1998).
• 3
  Interestingly, kɨsɛ ‘knife’ is considered ‘an old word’ by majority of younger Wajãpi speakers, which use instead marija, itself a (likely) loan from a Cariban language.
• 4
  The Taranoan languages form a low-level branch of the Cariban language family, and include Tiriyó, Akurió and Carijona/Karihona (see Meira, 2000).
• 5
  Tavares (2006, p. 138) analyzes the final -t as one of the possessive suffixes found throughout the Cariban family.
• 6
  ‘Adjectives’ and ‘adverbs’ are not formally distinguished in most Cariban languages, with most adjectival notions being expressed within the same lexical class expressing predicate modifiers (Gildea, 2012, pp. 455-459). For this reason, I use ‘adjective’, ‘adverb’ or ‘adjective/adverb’ interchangeably in the body of the text.
• 7
  But in Wayana, with ‘fat’ specifically, see another adverbial formation: i-ka-phakë-anu ‘the fat one’ (Tavares, 2006, p. 128; Camargo, et al., 2009, p. 39).
• 8
  I have retained the glosses used in E. Koehn and S. Koehn (1986). The prefix t(ɨ)- is apparently taken by the authors to be the same that codes a reflexive or correferential third person possessor in nouns, but in these cases its function, together with a suffix of varying shape (-re, -se, -e, -ne) consists in deriving either adjective/adverbs from nouns or non-finite verb forms. This explains the use of the glosses NF = ‘non-finite’, CMPL = ‘complement’ and ADJVZR = ‘Adjectivizer’ (see E. Koehn & S. Koehn, 1986, pp. 48-49 for a brief discussion of the differences between the Apalai construction with -se and its cognate construction in Hixkaryána).
• 9
  Camargo (1996) and Tavares (2006) differ in some aspects of their description of Wayana, and while some of these differences can be explained by reference to dialectal variation, others seem to result from analytical choices preferences or decisions by the linguists themselves. Note that while Camargo (1996) proposes an underlying fricative s for Wayana, Tavares (2006) opts for h instead. In any case, the distribution of the two attested allophones, [h] and [ʃ], is essentially the same in both accounts, and do not affect the arguments presented here.
• 10
  Rodrigues (1985, p. 392) mentions a Wayana form <kašuru>. This has to be an error, as Wayana lacks a coronal fricative in this context. Since first-harnd or primary sources are not given for any of the forms cited in Rodrigues (1985), the nature of the error and its origin cannot be identified.
• 11
  The claim in Gallois (1980, p. 61, fn. 1) that the lower Jari populations identified under ethnonyms such as ‘Tuare’ or ‘Aroaqui’ were speakers of Arawakan languages will necessarily remain speculative, as no language material associated to these names exists; on the other hand, lower Paru groups such as the Aracaju were definitely Cariban, as shown by the language data available (see Loukotka, 1968, p. 202).

• Carvalho, F. O. (2024). Evidências linguísticas e contato Tupi-Guarani/Carib nas Guianas orientais: Wajãpi kasi ‘ser forte’, pipi ‘irmã do pai’ e kasuru ‘pérola’. Boletim do Museu Paraense Emílio Goeldi. Ciências Humanas, 19(3), e20240001. doi: 10.1590/2178-2547-BGOELDI-2024-0001.

REFERENCES

• Armellada, R. P. C. (1943). Diccionario de la lengua Pemon (Arekuna, Taurepan, Kamarakoto) Artes Gráficas.
• Basso, E. B. (1977). Introduction: the status of Carib Ethnography. In Autor (Ed.), Carib-speaking Indians: culture, society and language (pp. 9-22). The University of Arizona Press. http://www.etnolinguistica.org/biblio:basso-1977-carib
  » http://www.etnolinguistica.org/biblio:basso-1977-carib
• Boček, V. (2013). Etymology and language contact studies: some notes on mutual relationships. In B. Vykypel & V. Boček (Eds.), Etymology: an old discipline in new contexts (pp. 13-21). Lakladatesltví Lidové Noviny.
• Camargo, E. (1996). Aspect de la phonologie du Wayana. Amerindia, 21, 115-136.
• Camargo, E., Tapinkili, Akajuli, Atajumale, & Hernandez, L. (2009). Hakëne omijau eitop Wajana-Palasisi (Dictionnaire bilingue Wayana-Français). CELIA. http://www.etnolinguistica.org
  » http://www.etnolinguistica.org
• Carlin, E. (2004). A grammar of trio: a cariban language of suriname P. Lang.
• Carvalho, F. O. (2017). Tupi-Guarani loanwords in southern Arawak: taking contact etymologies seriously. Revista Linguística, 13(3), 41-74. https://doi.org/10.31513/linguistica.2017.v13n3a16383
  » https://doi.org/10.31513/linguistica.2017.v13n3a16383
• Carvalho, F. O. (2022). On the Guaranian evidence for two Proto-Tupi-Guarani affricates. Journal of Language Relationship, 20(1), 81-112.
• Carvalho, F. O., & Birchall, J. (2022). A comparative reconstruction of Proto-Tupi-Guarani kinship terminology. LIAMES: Línguas Indígenas Americanas, 22, e022001. https://doi.org/10.20396/liames.v22i00.8666489
  » https://doi.org/10.20396/liames.v22i00.8666489
• Coudreau, H. (1892). Vocabulaires méthodiques des langues Ouayana, Aparaï, Oyampi, Émérillon J. Maisonneuve. http://www.etnolinguistica.org/biblio:coudreau-1892-vocabulaires
  » http://www.etnolinguistica.org/biblio:coudreau-1892-vocabulaires
• Derbyshire, D. (1999). Carib. In R. M. W. Dixon & A. Aikhenvald (Eds.), The Amazonian languages (pp. 23-64). Cambridge University Press. https://ribeiro.wdfiles.com/local--files/temp%3A1/The_Amazonian_Languages__Cambridge_Language_Surveys.pdf
  » https://ribeiro.wdfiles.com/local--files/temp%3A1/The_Amazonian_Languages__Cambridge_Language_Surveys.pdf
• Farabee, W. C. (1924). The Central Caribs The University of Pennsylvania. https://etnolinguistica.wdfiles.com/local--files/biblio%3Afarabee-1924-caribs/Farabee_1924_TheCentralCaribs_BDCN.pdf
  » https://etnolinguistica.wdfiles.com/local--files/biblio%3Afarabee-1924-caribs/Farabee_1924_TheCentralCaribs_BDCN.pdf
• Frikel, P. (1957). Zur linguistisch-ethnologischen Gliederung der Indianerstämme von Nord-Pará (Brasilien) und den aliegenden Gebieten. Anthropos, 52(3/4), 509-563. http://www.etnolinguistica.org/biblio:frikel-1957
  » http://www.etnolinguistica.org/biblio:frikel-1957
• Gallois, D. (1980). Contribuição ao estudo do povoamento indígena da Guiana Brasileira, um caso específico: os Waiãpi [Masters’ Thesis, Universidade de São Paulo].
• Gildea, S. (1995). A comparative description of syllable reduction in the Cariban language family. International Journal of American Linguistics, 61(1), 62-102. https://doi.org/10.1086/466245
  » https://doi.org/10.1086/466245
• Gildea, S., & Payne, D. (2007). Is Greenberg’s “Macro-Carib” viable? Boletim do Museu Paraense Emílio Goeldi. Ciências Humanas, 2(2), 19-72. https://doi.org/10.1590/S1981-81222007000200003
  » https://doi.org/10.1590/S1981-81222007000200003
• Gildea, S. (2012). Linguistic studies in the Cariban family. In L. Campbell & V. Grondona (Eds.), The indigenous languages of South America (pp. 441-494). De Gruyter Mouton.
• Grenand, F. (1980). La Langue Wayãpi (Guyane Française): Phonologie et Grammaire SELAF. http://www.etnolinguistica.org/biblio:grenand-1980-langue
  » http://www.etnolinguistica.org/biblio:grenand-1980-langue
• Grenand, F. (1989). Dictionnaire wayapi-francais; lexique francais-wayapi SELAF.
• Jackson, W. S. (1972). A Wayana grammar. In J. E. Grimes (Ed.), Languages of the Guianas (pp. 47-77). Summer Institute of Linguistics (SIL).
• Jensen, C. J. S. (1984). O desenvolvimento histórico da língua Wayampi [Masters’ Thesis, Universidade Estadual de Campinas]. https://doi.org/10.47749/T/UNICAMP.1984.51508
  » https://doi.org/10.47749/T/UNICAMP.1984.51508
• Jensen, C. J. S. (1998). Comparative tupí-guaraní morphosyntax. In D. Derbyshire & G. K. Pullum (Eds.), Handbook of Amazonian Languages (Vol. 4, pp. 487-618). Mouton de Gruyter. https://doi.org/10.1515/9783110822120
  » https://doi.org/10.1515/9783110822120
• Koehn, S. S. (1975). Apalaí kinship and social behavior. Arquivos de Anatomia e Antropologia, 1(1), 81-108. http://etnolinguistica.wdfiles.com/local--files/biblio%3Akoehn-1975-apalai/Koehn_1975_ApalaiKinshipSystem.pdf
  » http://etnolinguistica.wdfiles.com/local--files/biblio%3Akoehn-1975-apalai/Koehn_1975_ApalaiKinshipSystem.pdf
• Koehn, E. H., & Koehn, S. S. (1986). Apalai. In D. C. Derbyshire & G. K. Pullum (Eds.), Handbook of Amazonian Languages (pp. 33-127). Mouton de Gruyter. https://doi.org/10.1515/9783110850819.33
  » https://doi.org/10.1515/9783110850819.33
• Koehn, E. H., & Koehn, S. S. (1995). Vocabulário básico Apalaí-português. Sociedade Internacional de Linguística. https://www.silbrasil.org.br/resources/archives/16984
  » https://www.silbrasil.org.br/resources/archives/16984
• Lemle, M. (1971). Internal classification of the Tupi-Guarani linguistic family. In D. Bendor-Samuel (Ed.), Tupi Studies I (Summer Institute of Linguistics Publications in Linguistics and Related Fields, Vol. 29, pp. 107-129). Summer Institute of Linguistics. http://www.etnolinguistica.org/biblio:lemle-1971-internal
  » http://www.etnolinguistica.org/biblio:lemle-1971-internal
• Loukotka, Č. (1968). Classification of South American Indian languages University of California Press. http://www.etnolinguistica.org/biblio:loukotka-1968-classification
  » http://www.etnolinguistica.org/biblio:loukotka-1968-classification
• Mailhammer, R. (2013). Towards a framework of contact etymology. In Autor (Ed.), Lexical and structural etymology (pp. 9-31). Mouton de Gruyter.
• Meira, S. (2000). Proto-Taranoan Lincom Europa.
• Meira, S., & Franchetto, B. (2005). The southern Cariban languages and the Cariban family. International Journal of American Linguistics, 71(2), 127-192. https://doi.org/10.1086/491633
  » https://doi.org/10.1086/491633
• Meira, S., & Muysken, P. (2017). Cariban in contact: new perspectives on Trio-Ndyuka pidgin. In K. Yakpo & P. Muysken (Eds.), Boundaries and bridges: language contact in multilingual ecologies (pp. 197-228). De Gruyter Mouton.
• Mello, A. A. S. (2000). Estudo histórico da família lingüística Tupí-Guaraní: aspectos fonológicos e lexicais [Tese de doutorado, Universidade Federal de Santa Catarina]. http://www.etnolinguistica.org/tese:mello-2000
  » http://www.etnolinguistica.org/tese:mello-2000
• Rodrigues, A. D. (1985). Evidence for Tupi-Carib relationship. In H. E. M. Klein & L. R. Stark (Eds.), South American languages: retrospect and prospect (pp. 371-404). University of Texas Press. https://biblio.wdfiles.com/local--files/rodrigues-1985-evidence/rodrigues_1985_evidence.pdf
  » https://biblio.wdfiles.com/local--files/rodrigues-1985-evidence/rodrigues_1985_evidence.pdf
• Rose, F., & Renault-Lescure, O. (2008). Contact-induced changes in Amerindian Languages of French Guiana. In T. Stolz, D. Bakker & R. S. Palomo (Eds.), Aspects of language contact: new theoretical, methodological and empirical findings with special focus on romancisation processes (pp. 349-376). Mouton de Gruyter. https://doi.org/10.1515/9783110206043.349
  » https://doi.org/10.1515/9783110206043.349
• Rose, F. (2012). Borrowing of a Cariban number marker into three Tupi-Guarani languages. In M. Vanhove, T. Stolz, A. Urdze & H. Otsuka (Eds.), Morphologies in contact (pp. 37-69). Akademie.
• Schleicher, C. O. (1998). Comparative and internal reconstruction of the Tupi-Guarani language family [Doctoral dissertation, University of Wisconsin]. http://www.etnolinguistica.org/tese:schleicher-1998
  » http://www.etnolinguistica.org/tese:schleicher-1998
• Stegeman, R., & Hunter, R. (2014). Akawaio-english dictionary SIL International.
• Tavares, P. S. (2006). A Grammar of Wayana [Doctoral dissertation, Rice University]. http://www.etnolinguistica.org/tese:tavares-2006
  » http://www.etnolinguistica.org/tese:tavares-2006

Edited by

Editorial responsibility: Hein van der Voort

Publication Dates

• Publication in this collection
  07 Oct 2024
• Date of issue
  2024

History

• Received
  05 Jan 2024
• Accepted
  14 Mar 2024