A Sociophilological Account of the Formation and Evolution of the Term Língua Geral, with Emphasis on Amazonia

Lay Summary

We examine the hypothesis that the change from calling the main language of Portugal’s South American colonies Lingua Brasílica in the 17^th century to calling it Língua Geral in the 18^th century occurred because people noticed changes in language structure caused by bi- or multilingualism amongst the descendants of Europeans and Amerindians. The term Língua Geral then passed from being seen initially as a set of closely related Tupi-Guarani languages and dialects spoken natively by many different peoples over a massive area to referring specifically to the mestizos’ language.

However, we show that there is no evidence that the mestizo population spoke in a distinct manner in the 17^th and 18^th centuries. We also demonstrate that changes in language structure were not as extreme as has been previously claimed. The contemporary sources we investigated indicate that the name Língua Geral changed its meaning in the 18^th century because of a perceived change in the way it was used. Rather than a geographically widespread complex of closely related languages, Língua Geral became thought of as a lingua franca spoken both natively and non-natively.

Introduction

This article investigates the development of naming patterns for the so-called Língua Geral (general language, i.e., lingua franca), a Tupi-Guarani language spoken very extensively in Portugal’s South American colonies. An Amazonian variety of the Língua Geral – nowadays mostly called Nheengatu good language/speech – is still spoken natively today along the Rio Negro in Brazil and Venezuela, and residually in Colombia. Nheengatu is also being revitalised as a heritage language in several Amazonian communities.

We focus on the most recent and most widely known periodisations that linguists and historians have proposed for the transition between the use of the terms Língua Brasílica and Língua Geral. In particular, we focus on Rodrigues (1986, 1996), Lee (2005), and Argolo (2011, 2012, 2016), all of which heavily emphasises ethno-racial contact and hybridization as the main catalyst for structural linguistic changes that are claimed to have, in turn, stimulated the invention of new names.

We argue that, although much structural diversity existed and was occasionally commented on, such Abstand ([structural] distance) Kloss, 1978, 1967, 1976[1], see also Goebl, 1989; Bossong, 2008, p. 25-28) was not the basis for the changes in nomenclature that modern researchers have identified, especially the shift from Língua Brasílica (Brazilian language), in the 17^th century to Língua Geral in the 18^th century in Jesuit publications (Edelweiss,1969; Lee, 2005, 2014; Rodrigues, 1984/5, 1986, 1996, 2010; Dietrich, 2010; Argolo, 2011a/b, 2012a/b, 2016; Vieira; Zanoli & Módolo, 2019).

Instead, developing Finbow (2022), we sustain that the evidence indicates that (at least) non-indigenous people between the 16^th and 18^th centuries saw the language called both ‘the’ Língua Brasílica and Língua Geral as essentially elements in a single language complex, in spite of extensive diatopic variation. Rather than the conscious perception of structural differences, the name shifted primarily from perceived changes in the language’s function between the 17^th and 18^th centuries. Thus, the idea that the terms Língua Brasílica and Língua Geral refer to “different languages” or to different diachronic phases of the same language is the result of modern Ausbau (elaboration) (Kloss, 1967, 1976, 1978)[2] and therefore, anachronistic. We propose that the best modern equivalent for the expression Língua Geral as it was initially employed by the missionaries in the 17^th century is the equally problematic term macrolanguage, used by Ethnologue. Subsequently, in the 18^th century, lingua franca is a more appropriate translation, for reasons we shall develop below.

To analyse the linguistic consciousness of Portuguese America’s colonies regarding Língua Brasílica and Língua Geral, we draw on sociophilological frameworks, developed to understand how the conceptual distinction we make today between Latin and Romance arose in the early Middle Ages (section 2). In section 3, we justify our preference for ‘Old Tupi’ rather than ‘Tupinambá’ as the generic name for the largest indigenous linguistic bloc on the eastern coast of what is today Brazil. In section 4, we present the different names used between the 16^th and the 18^th centuries by contemporaries in their linguistic publications and the hypotheses drawn previously from these data. Section 5 deals with the kinds of structural differences between diatopic varieties of Old Tupi that were recorded in the 16^th and 17^th centuries and in section 6, we present the contemporary evaluations of such differences. Section 7 presents the evidence that contemporary commentators perceived only one ‘general language’, spoken along most of the coast and far up the Amazon River. I mentioned the concept of "macrolanguage" in the previous paragraph. I don't think it's necessary to repeat it here. In section 8, we show that, in the case of the Amazonian missions, there is no evidence that the bi- or multilingual Mamelucos (Euro-Amerindian mestizos) were responsible for fostering structural changes that resulted in people perceiving what they spoke as belonging to a “different language” to the older phases they encountered and choosing a new name for the innovative variety. This leads onto a discussion in section 9 of the likelihood of pidginization and/or creolization having occurred in Amazonian Old Tupi and the problems with earlier interpretations of structural changes that have been used as evidence for this hypothesis. Section 10 contains our general conclusions.

1. Sociophilology

Sociophilology is essentially the diachronic investigation of linguistic metalanguage, especially nomenclatures, (Herman, 1996, p. 30; Wright, 1996, p. 31-44, 277-287, 2002a, p. vii). By bringing the insights of sociolinguistics into Romance historical linguistics and philology, Wright transformed the understanding of emergence of the Romance languages. He convincingly shows the insufficiencies in the classic models’ explanation for the emergence of ‘Romance’ as a metalinguistic category distinct from the traditional denominations ‘Roman’ and ‘Latin’ as the result of natural, gradual, structural evolution that produced ever greater Abstand between vernacular Latin and an artificially maintained, hyper-archaic variety (Wright, 1982, p. i–ii, 1994: 27-28, 2003, p. 676-677, see also Finbow, 2011, 2012). Instead, Wright demonstrates the impact of conscious interventions in formal written style in the form of a novel spelling pronunciation and insistence on profoundly archaic lexis and grammar. These stylistic innovations occurred over during the 8^th and 9^th century Carolingian reforms in the Frankish Empire, gradually replacing the local norms. This novel Carolingian or Mediaeval Latin was introduced into the Christian Iberian kingdoms over the 12^th century during the Gregorian reforms (Wright, 1976, 1982, 1993, 1994, 1996, 2000a/b, 2002), substituting the older, autochthonous, Visigothic or Toledan norm. Thus, the distinction we make between Latin and the (Romance) vernaculars as distinct languages is a case of Ausbau, which evolved fairly gradually from the Carolingian period and not consolidated until almost a century after the respective ecclesiastical reforms in each region, as the artificially induced Abstand with the regional vernaculars impeded vertical communication, i.e., the act of reading aloud to an audience (Banniard, 1992, 2013). The evidence from traditional Roman vertical communication practices, e.g., no instructions given to readers to change lexis or syntax as they read but simply to enunciate clearly or to writers other than to avoid grand literary flourishes, implies that the illiterate could understand reading aloud in traditional written styles before the reforms, despite numerous archaisms, but struggled to understand the reformed variety subsequently (Wright, 1994, p. 3, 126-27, 1999, p. 506-07, see also Banniard, 1992). Consequently, sermons in the vernacular were permitted. However, as the vernacular no longer possessed any written form, since the traditional spellings now had reformed pronunciations, clerics used the novel, direct grapho-phonemic correspondences of reformed spelling as models for spellings that represented vernacular phonology. The term Romance < romanice (Romanly, i.e., vernacularly) referred to this new modality of vernacular writing while Latin (latine Latinly > formally, properly) became restricted to the non-natively acquired, ecclesiastical norm (Wright, 1982, 1996, 2002; see also Müller, 1963). In time, the existence of these two written norms for increasingly mutually incomprehensible varieties in each Late Latin/Early Romance-speaking community gradually altered peoples’ linguistic consciousness (Wright, 1991, 1993, 2003, 2008; Woolard; Geneovese, 2007; Tejedo-Herrero, 2009). As the literate came to see the written modalities as two separate languages rather than the formal and the vernacular written varieties of the same language, this vision was transmitted downwards to the rest of society (Wright, 2002, p. 262-273ç Lloyd, 1996c Jensen, 1996).

By applying sociophilology to the emergence and development of the concept of LG and other associated terms, e.g., Língua Brasílica, Tupi, Tupinambá, Old Tupi, etc., we can demonstrate that, just as Wright showed that Latin and Romance were not distinguished before the Carolingian and Gregorian Reforms, in the 16^th and 17^th centuries, those who wrote about language did not see the Língua Brasílica as a different language from the Língua Geral. Instead, they appear to have considered that the former encompassed the latter, along with many other Tupi-Guarani varieties that today are thought of as separate language. In addition, we show that what has been referred to as Tupi and Lingua Geral Paulista (LGP) were not conceived of as different languages to Tupinambá and Língua Geral Amazônica (LGA), as proposed by Rodrigues (1986, 1996, 2010). We also are able to show that a semantic shift did occur in the concept of Língua Geral in the 18^th century, as the nature of its use and the predominant groups who spoke it altered, but not for the reasons that Rodrigues identifies.

2. A note on nomenclature

We do not employ ‘Tupinambá’ as a generic term for the language of the largest coastal polities, as has been typical of anthropologists since Alfred Métraux and Florestan Fernandes, and linguists, following Aryon Rodrigues[1]. Instead, following Navarro (2008), we call the Tupi-Guarani linguistic and cultural complex of the Atlantic coast ‘Old Tupi’ (PT Tupi Antigo). Our preference for Old Tupi as a generic is four-fold and builds from clues in contemporary European texts that describe their perceptions and also reports of indigenous sociocultural and linguistic practices. Additionally, we take modern indigenous perspectives into account.

First, contemporary Europeans, e.g., Anchieta (1596), Cardim (1584), and Soares de Souza (1587), state that only one main language existed along the coast. It is not clear if they were influenced in this view by indigenous attitudes or not, because no records exist that can confirm if the Potiguara, for instance, conceived of what they spoke as a different language what the other indigenous polities spoke or merely a variety of the same language. Only the contrast with the Tapuia (non-Old Tupi speaker) seems to be a clear a case of Europeans borrowing an indigenous category. However, it is not evident what the term Tapuia is opposed to. Perhaps it was Tupi, used generically (see below), but given widespread Amerindian attitudes the equate ‘otherness’ with ‘non-humanness’ (Santos-Granero, 2009), another possibility is abá (person, human), which is shared by very many Tupi-Guarani languages. Such a broad tapuia/abá contrast could conceivably favour the idea that a single language was perceived as being spoken across political divisions.

Secondly, Anchieta’s use of Tupi in the poem Na aldeia de Guaraparim (ll. 183-189) shows that term could be used generically in the Jesuit variety, despite also referring to the dialect spoken by the inhabitants of the captaincy of São Vicente (Navarro, 2008, p. 11-12, see also Edelweiss, 1969, p. 69-108; Rodrigues, 1984/5c Rodrigues; Cabral, 2002).

Thirdly, only a subset of historical speakers unambiguously identified themselves as Tupinambá (Navarro, 2008, p. 11, see also Edelweiss, 1969, p. 69-111).

Finally, the modern descendants of non-Tupinambá OT-speaking peoples such as the Potiguara and Tupiniquim dislike Tupinambá being employed as an umbrella term for the linguistic and cultural tradition they identify with (José Romildo Araújo Guyraakanga Potiguara, p.c., Tiago Matheus Kaûẽ Tupinakyîa, p.c.), preferring Tupi with a qualifying gentilic, e.g., Tupi potiguara nhe’enga kuapa (Araújo et al., 2024).

On the other hand, in the case of the Estado do Maranhão[2], Portugal’s Amazonian colony, founded in the 17^th century, in our opinion, it is appropriate to employ ‘Tupinambá’ because that was the endonym of the largest polity the Portuguese initially encountered around the island of São Luís and along the coast westwards. These Tupinambá had migrated from the northeastern coast, as the Jesuit Manoel related in 1616, soon after the Portuguese conquered the region (Gomes, 1904, p. 329; see also Hemming, 1995, p. 213). This migration probably occurred sometime after 1535, as the Portuguese occupation intensified in the hereditary captaincy of Pernambuco.

Thus, within Old Tupi, we subdivide Tupinamba geographically into ‘Maranhão’ Tupinambá to distinguish this group from the ‘Bahian’ Tupinambá, spoken in the middle of the eastern seaboard, and from the Tamoio Tupinambá, spoken between the Guanabara Bay (Rio de Janeiro State) and Ubatuba (São Paulo State). However, we should not assume that the Maranhão Tupinambá were the sole contributors of Old Tupi/Tupi-Guarani material to the colony’s variety of what became known as the Língua Geral. Very many Tupi-Guarani-speaking peoples were incorporated into Amazonian colonial society. As early as 1616, Gomes records that there were Língua Geral-speaking Tapuias in Maranhão. He is probably describing non-Old Tupi Tupi-Guarani-speakers, who were very numerous on the southern shore of the lower Amazon. Bettendorf (2010 [1698]) lists, amongst others, Guajajara, Juruna, Curuba, Tocantim, Naimiguara, Usaguara, Pacajá, Nambiquara, Coatinga, Guauara, Poquiguara, Guaiapi, Taconhapé, and Aruaqui, whose names have clear Old Tupi etymologies.

3. The textual record

In Brazil, the largest indigenous language was initially called Lingua Brasílica in the Jesuits’ early publications. The Jesuit norm was mostly based on the varieties spoken between Rio de Janeiro and Bahia, which Rodrigues calls ‘Tupinambá’, although Anchieta first learned OT in the south, in São Vicente and São Paulo, which Rodrigues calls ‘Tupi’. Anchieta later moved to Bahia and finally to Espírito Santo, where several other Jesuit linguists were already working (Edelweiss, 1969). However, the nomenclature changes in texts written between the 16^th and the 18^th century (Ávila, 2021, Ayrosa, 1950, p. 9-16; Barros; Monserrat, 2015, pp. 239-40; Dietrich, 2014, p. 596-8; Edelweiss, 1969, p. 138-165; Rodrigues, 1985, p. 96).

Figure 1.

Rodrigues (1986, p. 96, 1996, p. 5, 8) claims that these data indicate that the term Língua Geral emerged in the 16^th century and that initially the adjective geral referred to the geographical extent of a language. Specifically in Amazonia, Língua Geral was also used, according to Rodrigues (1986, 1996), to refer to other Tupi-Guarani languages that were structurally similar to what the Jesuits initially referred to as Língua Brasílica, the Tupi-Guarani language of the coast. However, after new varieties arose from contact with Portuguese amongst the emergent Mameluco class of European-Amerindian mestizos, the expression Língua Geral came to refer to those novel varieties in the 17^th and 18^th centuries (Rodrigues, 1996, p. 6, see also Argolo, 2016, p. 90-93), substituting the term Língua Brasílica and no longer emphasizing the language’s geographical extension.

Additionally, Rodrigues (1986, p. 102, 1996) divides Língua Geral into ‘São Paulo’ or ‘Southern’ (paulista, meridional), descending from the ‘Tupi’ spoken on the coastal plain of São Vicente and the interior plateau of São Paulo, and ‘Amazonian’ or ‘Northern’ (amazônica, septentrional). In Rodrigues’s view, the latter variety, which became better known as Nheengatu from the end of the 19^th century, evolved from ‘Tupinambá’, i.e., the varieties spoken along the Atlantic coast from the present-day state of Rio de Janeiro to around the state-line between modern Ceará and Paraiba and in Maranhão and Grão-Pará. Rodrigues claims no Língua Geral developed from Bahian Tupinambá in the central coastal region between Rio de Janeiro in the south and Ceará in the north because the Amerindian population was rapidly either killed, driven away, or died in epidemics. As a result, few Mamelucos were born and enslaved Africans and their descendants with Europeans soon became the dominant demographic segment[1].

Rodrigues (1986) and Rodrigues (1996) model essentially reworks the proposal in Edelweiss (1969, p. 44-45, 111, 123-158, 158-204). Edelweiss affirms that what he calls (pre-contact) ‘Tupi’ was codified by the Jesuits to create the Língua Brasílica (1500-1700). This variety then suffered structural changes because of linguistic and racial mixture to become what Edelweiss calls ‘Brazilian’ (Brasiliano), ‘Middle Tupi’ (Tupi médio) or Língua Geral (1700-1800), before turning into Nheengatu (1800­–present). In the pre-contact phase, Rodrigues separates ‘Tupi’ (southern) from ‘Tupinambá’ (northern), as we have seen, and attributes the crucial structural changes to Mameluco multilingualism. In both Edelweiss’s and Rodrigues’s models, the emergence of new names follows speakers perceiving structural change.

Argolo (2011b, 2012a, 2016) essentially follows Rodrigues (1986, 1996) but claims that a Tupinambá-based Língua Geral did exist in the south of Bahia, because 18^th century documents mention ‘Indians’ who speak ‘the Língua Geral’. Unlike Rodrigues, Argolo also adds language shift and creolization to the mix in the case of Amazonian Língua Geral. Lee (2005) also mentions pidginization and creolization, claiming that a new, ‘vulgar’ language emerged from the Língua Brasílica because non-Tupi-Guarani speaking indigenous peoples were forced to learn it in the missions. Oliveira, Zilles and Modolo (2019) claim ‘Tupinambá’ began as a pidgin, which evolved into a creole they call the ‘Língua Geral do Brasil’. As we shall see, these classifications are all founded on modern perceptions of Abstand that were not stressed in contemporary accounts before the mid-18^th century.

4. Abstand in Old Tupi, Língua Brasílica and Língua Geral

Rodrigues’ identification of Tupi and Tupinambá as distinct dialects or languages (see also Dietrich, 2010) arises from Anchieta’s statement that lexical roots in the region south of Rio de Janeiro did not exhibit final consonants that existed in northern varieties (Anchieta, 2014 [1596], p.1, see also Edelweiss, 1969, p. 76 ff.; Rodrigues, 1958/59, 1985, 1986, 1996, 2010), e.g.,

(1a) apâb ~ apâ /a-′pa(β)/ (first-person singular subject, active class + terminate)

(1b) acêm ~ acẽ /a-′sẽ(m)/ (first-person singular subject, active class + leave)

(1c) apên ~ apẽ /a-′pẽ(n)/ (first-person singular subject, active class + be crooked)

(1d) aiûr ~ aiú /a-′ju(r)/ (first-person singular subject, active class + come)

Such consonantal apocope was almost certainly a wider southern areal phenomenon for Montoya (2011 [1639], p. 163; see also Navarro, 2013, p. 102, 160) records synchronic alternation of apocopated and unapocopated allomorphs in old Guarani, e.g.,

(2a) (h)endu(v)- /(h)-e′nu(β)/ (hear sth.) NB OT (s)endub /(s-)e′nuβ/ 3.p-hear

(2b) a(r) /a(r)/ (take, seize sth.) NB OT (t)ar /′(t-)ar/).

In São Paulo Old Tupi, as in Old Guarani, suffixed lexical roots retained the final consonants, e.g., Ubatuba /uɁuʋa-′tɨʋ-a/ arrow-abundant-ref (abundant arrow [cane]), Itatiba /iˌta-′tɨʋ-a/ stone-abundant-ref (abundant stones) (toponyms, São Paulo State). However, all modern Guarani varieties have reanalysed their lexical roots as possessing only open syllables, e.g., (Tekoa) Itaty /iˌta-′tɨ/ stone-abound (abundant stones, Morro dos Cavalos, Palhoça, Santa Catarina State), which would be (Tekosaba) Itatyba in Old Tupi, i.e., /t-eko-′saʋ-a ita-′tɨʋ-a/ r3-be-circ.nmzr.ref stone-abound-ref. On the other hand, Nheengatu (NHG), the descendant of Maranhão Tupinambá, generally retains final consonants through fossilising the Old Tupi (OT) referential suffix /-a/ or vocalic paragoge, e.g.,

(3a) OT taba /′taʋ-a/ (village-ref) > NHG tawa /′ta.wa/,

(3b) OT ygara /ɨ′ar-a/ (canoe-ref) > NHG igara /i′ga.ra/‘canoe’,

(3c) OT pa’i /pa′Ɂi/ (Oh my father! voc.) ~ PT pai /′pai/ (father) > NHG paia /′pa.ja/ (ibid.), cf., PT mãe /′mãi/ (mother) > NHG, manha /′mã.ja/ (ibid.)

(3d) OT sem /′sem/ (leave) > NHG sému /′se.mu/

(3e) OT syk /′sɨk/ (arrive, approach) > NHG sika /′si.ka/.

Northern Old Tupi used <-i> (/-j/) on consonant-final roots and <-û> (/-w/) followed vowel-final roots on both stative and active verbs to express the ‘circumstantial indicative’ or ‘indicative II’, which marks subordination with a third-person subject when non-arguments undergo topic-focus (Vieira, 2014; Navarro, 2008, p. 191; Rodrigues, 2010b, p. 25, 38, 41). Southern varieties marked this construction with /-j/ and /-w/ on active verbs but stative verbs employed /-(r)amo/[1].

A further difference that has been suggested in the Língua Geral phases is in the articulation of the high, central, unrounded vowel /ɨ/. In the Rio Negro Língua Geral, the Arawak substrate caused this vowel to merge with /i/ (or occasionally /u/), e.g., NHG pisirũ /pisi′rũ/ (help) < OT pysyrõ /pɨsɨ′rõ/. In the south, it has been suggested that /ɨ/ merged with /u/. For example, Martius (1867, II, p. 190-122) records putúnami (become night/dark PT enoitecer; GER Nacht werden) < OT /pɨ′tun-eme, pɨ′tũ-reme/ night-at_temp (at night[time]), in the night[time]”. Similar examples for ‘at night’ occur in Nheengatu (pituna ramé), Nhandewa Guarani (pyntũ ramõ), Mbyá Guarani (pytũ ramõ), and Paraguayan Guarani (pyhare ramo).

Another set of examples from Martius is oçuca and açuc < OT /o-, a-′sɨk(a)/ 3.a-, 1sg.a-reach (suffice ‘PT bastar; GER genügen’ (also come close to, arrive (at)).

However, clear evidence of /ɨ/ > /u/ as a systematic process is scanty, as northern varieties also exhibited /pu′tũ(n)/ for “night”. In toponyms, the Old Tupi morpheme tyb /tɨʋ/ (abound), appears written -tiba, -tiva, -ndiba, -ndiva and -tuba, -tuva, -nduba, -nduva in the north and the south[2].

Thus, structural differences certainly did exist between two broad diatopic blocs of Old Tupi. However, it is unclear how such (fairly minor) structural divergence was conceptualised by contemporary speakers, both native and non-native, which will be investigated in the following section.

5. Contemporary perceptions of linguistic diversity in Old Tupi

Sadly, no contemporary testimony exists of any OT-speaking people’s opinions about any linguistic similarities or differences they detected. However, the fundamental distinction that was recorded as having been made between themselves and the Tapuia, i.e., peoples considered non-Old Tupi by culture and/or language (Cardim, 2009 [1584], p. 205-206), Tupi-Guarani-speaking or otherwise, would suggest that Old Tupi-speakers perhaps perceived themselves rather like the Germanic and Hellenic peoples in ancient Europe did, i.e., as a broad cultural and linguistic bloc whose members opposed themselves to other cultures on the basis of certain shared characteristics, despite sociocultural and linguistic diversity, and even internecine warfare, between in-group polities.

Contemporary reports by Europeans demonstrate that the diatopic variation they detected was not interpreted by them in terms of different languages amongst the main peoples on the coast. Anchieta (1989 [1584], p. 59) separates the Carijó (a Guaranian people) from the speakers of the ‘language most spoken on the coast of Brazil’ but mentions no regional subdivisions beyond the reference to consonantal apocope discussed above. This is despite his having lived with the Tupi in São Paulo and São Vicente for many years, amongst the Tamoio and Bahian Tupinambá, as well as the Tupiniquim in the captaincy of Espírito Santo. His writings exhibit both southern and northern features (Navarro, 2008, p. 13). Cardim (2009 [1584], p. 101) states flatly that the ten coastal OT peoples, i.e., Tupi, Tupiniquim, Tamoio, Temiminó, Marakajá, Tupinambá, Tupinaé, Caeté, Tobajara, and Potiguara, all speak the same language. For Soares de Souza (2010 [1587], p. 406), the linguistic differences between the Tupinaé and the Bahian Tupinambá were comparable to the differences between the Portuguese of Coimbra and of the Beira region, i.e., mutually comprehensible diatopic varieties.

We agree with Rodrigues (1986, 1996) that Língua Brasílica is the term first used in the titles of grammars and catechisms of OT. However, Língua Brasílica was not limited to Old Tupi or even Tupi-Guarani; it referred to any Brazilian indigenous language. For example, in the interior of the Northeast, where there were few (if any) Old Tupi speakers historically, catechesis was carried out in two closely related Kariri or Kiriri languages (< OT kyrirĩ be silent) known today as Kipeá (Mamiani, 1698, 1699) and Dzubucuá (Nantes, 1709, 1986; Queiroz, 2008, 2012)[1]. Mamiani’s catechism and grammar refer to the language as ‘the Língua Brasílica of the Kiriri nation’[2], demonstrating the original use of the term Lingua Brasílica.

In 1605, Pero Rodrigues, Jesuit Provincial of Brazil from 1594 to 1603, recorded that ‘This language is the general (one), beginning above the Maranhão [i.e., Amazon] River... as far as Paraguay’[3] (Edelweiss, 1947, p. 29, Rodrigues, 1996, p. 7). Once again, Guarani seems to have been excluded from the Lingua Geral of the Portuguese conquests, perhaps because at the time it was located mainly beyond the Tordesillas Meridian in what was nominally Castilian territory. Shortly afterwards, the Jesuit Manoel Gomes wrote from Maranhão in 1616 that, ‘There are many Tapuias [i.e., non-Old Tupi peoples] of many nations, of which fourteen speak the Tupinambá lingua franca, which is almost universal [comum] in Brazil’ (Gomes, 1904 [1616], p. 334; Santos, 2011, p. 10)[4]. Gomes, and later Antônio Vieira, compare the Guajajara people with the Guaranian Carijó, showing that structural parallels were noted between the speech of non-Old Tupi Tupi-Guarani-speaking peoples in the Estado do Maranhão and the main language of the Estado do Brasil.

Note that Pero Rodrigues and Manoel Gomes use Língua Geral even before the Jesuits had published anything with Língua Brasílica in the title. This proves that at that time (17^th c.) the Jesuits already thought of the largest Língua Brasílica (indigenous language), as a Língua Geral, i.e., a geographically extensive language or dialect continuum, as Rodrigues (1986, 1996) proposes was the term’s original meaning, despite the Jesuits using Língua Brasílica in the titles of their 17^th century publications. Geographical or demographic size was the primary criterion for selecting a language to be a diocese’s administrative lingua franca according to the practices developed over the mid-16^th century in Spain’s American conquests (Madureira, 1977; Zavala, 1977; Ramos Pérez, 1986; Altman, 2003; Alfaro Lagorio, 2003; Pérez Puente, 2009; Dietrich, 2014).

It is possible that increased institutional contacts during the Iberian Union (1580-1640) may have spread the Spanish model of colonial linguistics amongst the Jesuits sent to Brazil with the University of Salamanca as a major centre of diffusion (Finbow, 2022; Barros, 2023, p.c.), for the term appears later in the Portuguese sphere. Like the indigenous lenguas generales of the Spanish Empire, the norm developed by the Jesuits was from the outset a hybrid, being a formalised description of several Old Tupi diatopic varieties (Edelweiss, 1969, p. 73-79, Altman, 2003; Alfaro Lagorio, 2003). The greater usage of Bahian Tupinambá features from the central region of the coast was probably because these were thought to be intelligible on the broadest scale, again, as Spanish missionaries did (Alfaro Lagorio, 2003; Edelweiss, 1969, p. 72-79). This is the systematization published by Figueira in 1621, which became the main teaching grammar, in conjunction with Araújo’s catechism of 1618, and the Vocabulário na Língua Brasílica, also from 1621.

Despite there being a very large region between central Maranhão and the easternmost part of the northeast coast that was inhabited by many linguistically non-Tupi-Guarani peoples, and where no Old Tupi speakers are recorded, as well as several regions on the Atlantic coast where Old Tupi peoples were not predominant, the missionaries appear to have envisaged the linguistic situation in the lower Amazon as essentially a continuation of the situation they knew in the Estado do Brasil (Freire, 2011, p. 43). Their perception was justified, for the Maranhão Tupinambá and the Tobajara of the Ibiapaba hills in modern Ceará State were recent incomers who had fled from the eastern coast a few generations previously.

A modern definition that might be applied interestingly to Língua Geral as employed initially by the missionaries is macrolanguage, which is defined by ISO 636-3[5] as ‘... closely related individual languages that are deemed in some usage contexts to be a single language’ (Ethnologue, 2009). The basic criterion for a macrolanguage is a close phylogenetic relationship, which is true of the habit of applying Língua Geral as a synonym for what would nowadays be called ‘Tupi-Guarani’. Additionally, there is frequently a classical standard which speakers of several closely related individual languages understand, or which is at least known to be the source of those individual languages, e.g., Arabic, which is divided into many modern spoken varieties which exist alongside a common standard (Modern Standard Arabic) and also the classical Quranic norm. In the case of Língua Geral, this would be the Jesuit norm, although it was not the source of the other varieties but a later addition, rather as Standard German or Italian emerged from regional diversity. Secondly, a macrolanguage can have a long-lasting and deep-rooted linguistic identity amongst the spoken varieties that have undergone separate developments because of sociopolitical factors, e.g., Serbo-Croat as a macrolanguage comprising the individual Croatian, Serbian and Bosnian languages. This aspect can be applied readily to the different coastal Old Tupi varieties with their opposition to the peoples they called Tapuia, Kariri or Nheengaíba. Thirdly, a macrolanguage may comprise several closely related languages which the specialist literature often treats as either a genealogical unit of individual languages or which may be subsumed into one “language” for other reasons, e.g., political or ethnic criteria, e.g., Rajasthani being used as the macrolanguage category of several Indo-Iranian languages such as Bagri, Gade, Lohar, Hadothi, Malvi and Wagdi (https://www.ethnologue.com/language/raj/). This is similar to the missionaries’ practice of calling all the Tupi-Guarani-speaking peoples in Maranhão, Grão-Pará and the Estado do Brasil ‘Língua Geral [-speaking] Indians’ (índios de Língua Geral), despite each having its own individual identity and apparently not being seen as abá/tupi by OT-speakers.

Even the criticisms of the category macrolanguage make it appropriate to apply to Língua Geral. For example, the inconsistency of the application by SIL on Ethnologue and the lack of a unifying definition to distinguish dialects, dialect clusters, and languages. No agreement or uniformity exists in the overarching linguistic standards in terms of linguistic criteria for identifying when one should retain non-linguistically determined divisions and when one should ignore them. Thus, any classification could very easily lead to problematic results and provoke controversy. This is precisely the same kind of difficulties as the concept of Língua Geral has posed to linguists and historians, which is why ‘macrolanguage’ is such an appropriate modern correlate to this colonial exonym.

6. Conceptually, only one Língua Geral

Aryon Rodrigues (1986, 1996, 2010) argues that there were two Old Tupi-based Línguas Gerais, southern/Paulista and northern/Amazonian, which arose from two kinds of Old Tupi that he calls respectively ‘Tupi’ and ‘Tupinambá’. However, this proposal does not take into account Antônio Vieira’s attempts in the 1650s and 1660s to recruit fluent or native Lingua Geral-speaking Paulistas to work as línguas (interpreters) in Amazonia because these Paulistas were much more successful at converting indigenous peoples, presumably because of their superior linguistic abilities in contrast with the European missionaries (Barros, 2003, 2010, see also Barros, 1986, 1994, 1995, 1996). This strongly suggests that contemporaries did not see southern varieties of Língua Geral as a distinct language from what was spoken in the early Amazonian missions, despite the existence of some structural differences, as we have seen.

Gomes’ reference to Lingua Geral-speaking Tapuias in Maranhão in conjunction with a reference to the Guajajara probably means that he was referring to the non-Old Tupi Tupi-Guarani-speakers, who were very numerous on the southern shore of the lower Amazon. The Jesuits classified these peoples as índios de língua geral (Tupi-Guarani-speaking Indians), despite their diverse ethnic affiliations[1], as illustrated by the list of ethnonyms in Johann Philipp Bettendorf’s Crônica da missão dos padres da Companhia de Jesus na Província de Maranhão (Bettendorf, 2010 [1698]).

Because the groups mentioned by Gomes and Bettendorf were still unmissionized and had had little, if any, contact with the colonists, it is highly unlikely that they already knew Mameluco Maranhão Tupinambá or the Jesuit norm. These two contexts – Mamelucos and missions – are the two key sources of the changes responsible for creating the semantic shift in concept of Língua Geral according to the hypotheses advanced by Rodrigues and Argolo, respectively. Thus, in the 17^th c. Amazon, the expression Língua Geral was definitely not identified with the speech of the Mamelucos and other racial, ethnic, or structural factors did not define it. Instead, as Finbow argues (2022, p. 82-85), Lingua Geral still seems to encompass any Tupi-Guarani language that permitted communication with the missionaries and the East-coast Old Tupi-speaking settlers and their slaves and allies, as Rodrigues claimed in 1986, i.e., the primary sense of Lingua Geral continues to be ‘macrolanguage’.

Indeed, as we shall discuss below, given the diversity of Tupi-Guarani-speaking peoples that were brought as slaves and allies to Belém and São Luís from 1616 and the very large groups of mainly Tupi-Guarani-speaking peoples ‘descended’ into the missions as catechumen labourers from 1650 onwards, a Tupi-Guarani-koine, with Maranhão Tupinambá as the major contributing variety would have been the most likely outcome (Finbow, 2022; Noll, 1999, 2008; see also Barros, 2003; Mufwene, 2003, 2008).

7. Tapuias, not Mamelucos

Rodrigues uses Mameluco as a synonym of ‘mestizo’ in the modern Brazilian manner. However, this is an anachronism, for Mameluco in 16^th and 17^th-century Brazil meant someone born outside official matrimony whose mother was classed as an ‘Indian’ and whose father was legally ‘white’ and publicly recognised his offspring. This meaning dropped out of circulation in the 18^th century (Monteiro, 1994, p. 166-167). Such paternal recognition guaranteed certain freedoms that were unavailable to unrecognised illegitimate offspring, who were simply called bastardos (bastards). Thus, on the one hand, the Mamelucos’ (limited) social privileges probably did allow greater access to Portuguese and therefore did favour bilingualism, as Rodrigues (1996) claims, which, as many have claimed, can be an important factor in stimulating language change (Aboh, 2015; Mufwene, 2003, 2008). On the other hand, Rodrigues’s indiscriminate use of Mameluco hides the fact that it was not employed in the same way in the past. Moreover, no contemporary sources identify the Mamelucos as speaking a characteristic variety.

In Maranhão, as we have seen, Gomes writes of Tapuias, not Mamelucos, speaking the ‘Tupinambá Língua Geral’ in the year of the Portuguese conquest, far too early for there to have been significant mixing of Europeans and Amerindians. Thus, with the exception of Rodrigues (1996, p. 5) and Cardeira (2006), most treatments of LG, especially in Amazonia, identify the incorporation of non-Tupi-Guarani-speaking Tapuias into the mission villages on a massive scale as the main catalyst for structural change[1]. These hypotheses see the Tapuias’ acquisition of the Jesuit norm and/or vernacular Maranhão Tupinambá as an auxiliary language in an unstructured manner as what “converts” Língua Brasílica into Língua Geral and then causes the Língua Geral to undergo further change and “evolve into” Nheengatu. Thus, the idea that the term Língua Geral underwent a semantic shift in the 18^th century as a means to refer to the speech of the Mameluco class which was subsequently generalised is not borne out.

What is certain is that Língua Geral could be acquired via two routes in the Amazonian missions from the 1650s: formal rote catechesis by the missionaries in the codified variety and immersion in the vernacular through cohabitation with speakers (Rosa, 1990; Barros, 2015, §37-40).

-

[...] catecismo acabado se sentavão todos a ouvir uma pregação ou exhortação, a qual se fazia na lingoa geral dos índios como também o catecismo, e orações eram compostas na mesma língua, e a dita exortação se acomodava sempre ao Evangelho [1] .

Resposta aos capítulos que deu contra os religiosos da Companhia em 1662 o procurador do Maranhão, Jorge de São Paio (cit. Barros; Borges; Meira, 1996, p. 195)

-

The Resposta aos capítulos shows that Língua Geral refers to not only the language of catechesis ­– which we can be confident was the Jesuit written norm, given it would have been Araújo’s Catecismo brasílico of 1618 – but also to the language of the sermon.

Barros, Borges & Meira (1996) cite this as evidence for the use of ‘LB’, i.e., the Jesuit norm, regardless of the linguistic affiliation of a mission’s inhabitants. However, the Resposta aos capítulos is from 1662, when the missions were still predominantly inhabited by OT and other Tupi-Guarani speakers, for the Portuguese were essentially only active below Gurupá at the mouth of the Xingu in the first decades of the colony (Saragoça, 2000), where Tupi-Guarani-speaking peoples were in the majority[1]. Given the structural proximity of Tupi-Guarani languages, catechesis in the codified variety is unlikely to have been problematic. Indeed, we see considerable continuity in writing between the 16^th and 17^th centuries. For example, Bettendorf’s Compêndio (1687) diverges very little from Araújo’s catechism (1618) (Edelweiss, 1969; Monserrat, 2003; Monserrat; Barros; Motta, 2010; Monserrat; Barros, 2015; Ávila, 2021; see also Rodrigues; Cabral, 2010).

Around 1660-70, however, epidemics began and intensified over the rest of the century and into the following one, decimating the original inhabitants of the missions and households (Hemming, 1987; Finbow, 2022, p. 96-67). From the final quarter of the 17^th century, the missionaries and slavers pushed beyond the Tapajós river into the middle Amazon, where Tupi-Guarani languages were largely absent. These factors combined to cause major demographic restructuring in the Amazonian colony and this would have had an impact on language.

Leite (1943, t. IV, p. 139) calculates that in 1696 the mission population was around 11,000 and in 1730 it had risen to 21,000. Raiol (1900, p. 192) estimated that the mission population in 1720 wase just under 55,000, with a further 20,000 ‘Mamelucos and slaves’ (see also Freire, 2011, p. 68; Hemming, 1995, p. 421; Finbow, 2022, p. 97, 2023a).

Another major territorial expansion occurred in the 1730s, after the Manao War (1728-1730), which saw missionaries and slavers bring very large numbers of non-Tupi-Guarani-speaking indigenous peoples, especially speakers of Arawak languages from the middle and upper Rio Negro, into the LG-speaking colonial centres on the lower Amazon. These Arawak-speaking second language learners might possibly have initiated certain structural changes still detectable in differences that exist between early- and late-19^th century Nheengatu, such as post-verbal object pronouns, although later influence from vernacular Brazilian Portuguese was probably much stronger (Finbow, 2022, p. 98, 2023a). The result of these events led to next phase in the meaning of Língua Geral, documented by Bluteau in his Vocabulário portuguêz e latino (1712-28) and in Daniel’s Tesouro descoberto no máximo Rio Amazonas (1757-76).

Bluteau built on Jesuit descriptions of LG. He contrasts Línguas Gerais, ‘spread by conquest, religion and commerce’, and Línguas Particulares (specific languages), spoken by ‘isolated, barbarous nations’ (Barros, 2015). Moreover, the speakers of Língua Geral in the Amazon, according to Bluteau, are the Tapuia. This shows that in the 18^th century, Língua Geral is not thought of as a geographically extensive macrolanguage but rather as a supra-ethnic lingua franca used as an auxiliary language by many peoples.

Evidently, this change from communities of predominantly native speakers using either their own Tupi-Guarani vernacular and/or an emerging Old Tupi/Tupi-Guarani koine, to a society containing very large numbers of non-native speakers is very likely to have had structural repercussions. Given the parallels between Portuguese Amazonia and societies in which pidgin and creole languages arose as adults acquired the primary vehicle of communication of their environment (many of which were also part of the Portugal’s colonial empire), the idea has spread that that Língua Geral should be understood as a creole language, which is what the following sections analyse.

8. Was Língua Geral a creole?

Clear proof that vernacular Língua Geral had evolved structurally comes indirectly from Daniel (2004 [1757-76] and directly from examples in the manuals written by the tapuiatinga (white Tapuia) missionaries, i.e., the northern and central European Jesuits, in the mid-18^th century (Dietrich, 2014; Monserrat, 2003; Monserrat; Barros, 2015). Note that none of these sources mention Mamelucos. Instead, Daniel explicitly identifies the Tapuia as the group that ‘corrupted’ the ‘true’ ‘Tupinambá’ LG, which he equates with the ‘Art’, i.e, Figueira’s 1621/1687 grammar. Nevertheless, Daniel also says that the ‘corrupt’ LG is spoken ‘in all the Portuguese missions of the Amazon’ and that few speak the ‘true’ ‘LG of the Tupinambá’ ‘in its native purity and vigour’ because the ‘first and true Tupinambá are already almost entirely extinguished’ (Daniel, 2004 [1757-76], v. 2, p. 365, see also Finbow, 2022; Monserrat, 2003; Dietrich, 2014).

Daniel’s statement that the Tapuias’ speech ‘seems another, different language’ to that the missionaries studied (2004, p. 365) and that the missionaries’ codified variety was incomprehensible to them, has been used to suggest that pidginization and/or creolization has occurred in the missions, giving rise to a “new” language, e.g, Lee (2005, 2010, 2014), Argolo (2011a, 2012b, 2016), Oliveira; Zanoli; Modolo (2019)[1], Castro (1991), Dietrich (2014). Dietrich, in particular develops Freire & Rosa (2003), Argolo (2011b) and Leite (2013), but he retains the emphasis on the mestizo class from Rodrigues (1996). He also discusses some structural changes (2014, p. 613-617).

Yet, Daniel is not contrasting the speech of the missionized Tapuia with the vernacular of other segments of the colony. His testimony reveals that even in old missions established in the mid-1600s with Maranhão Tupinambá and other native Tupi-Guarani-speakers the vernacular is not like the missionaries’ ‘Art’. This suggests that 18^th century Portuguese Amazonia should be seen as a diglossic society (Lee, 2005; Finbow, 2022, see also Ferguson, 1959; Fishman, 1967, 2002; Haas, 1982; Kaye, 1970, 1972,) because the tiny contingent of missionaries sought to uphold and preserve their traditional codified variety, which had to be acquired through formal instruction, for it was not acquired as a first language by any indigenous group (if it ever had been), while in day-to-day life everyone used the contemporary vernacular, which had undergone structural changes.

Daniel never exemplifies the kind of structural changes he is mentions. Thus, his testimony cannot be understood uncritically as evidence for creolization arising from a pidgin. We concur with the proponents of creolization that a great deal of structural change certainly would be expected to arise in the context of 150 to 200 years of widespread unstructured, adult second-language learning accompanied by language shift. Nevertheless, there is no evidence for a widespread stable pidgin variety of Lingua Geral in the Amazon at any period.

As Finbow (2022, p. 87-90, 2023b) has shown regarding Argolo’s (2011, 2012, 2016) and Oliveira, Zanoli & Modolo’s (2019), attempts to identify structural features to prove the creolisation of Lingua Brasílica “into” Língua Geral Amazônica suffer from a lack of knowledge of Tupi-Guarani languages and of Old Tupi in particular. For example, Lee (2005, p. 217-220) compares hypothetical sentences in Língua Brasílica, i.e., the Jesuit norm, with their equivalents in the ‘Vulgar’ language that she envisages emerging from it. However, her first example sentence – Na eresendúipe? – actually means ‘Don’t/Can’t you hear it?’, not ‘Don’t/Can’t you hear me?’. The correct sentence in OT is Nda xe rendu(b)ipe îepé? Her second sentence, Ixé nde nheenga, literally means ‘I am your speech/words’ in OT and modern Nheengatu, not ‘I am speaking to you’[2]. This misinterpretation shows that Lee’s analysis of the structural changes between different diachronic phases of Old Tupi/Língua Geral cannot be relied on and create a more extreme picture of structural change than actually happened because she does not use any material from the 18^th century or early 19^th century.

A further issue is that, like Rodrigues (1986, p. 104-109; 1996, p. 4-5) and Argolo (2016), Lee compares later 19^th c. and modern Nheengatu to ‘classic’ Jesuit Old Tupi. The Nheengatu spoken in the later 18^th and early 19^th century did not exhibit the same object-marking strategies (Finbow, 2023a). Thus, the correct comparison between her two sentences across three diachronic phases of LG (16^th-17^th centuries, 18^th century-1850, 1850 -20^th century) is

Figure 2.

In (4a), despite the substitution of the special form for ‘1^st person patient and 2^nd person agent’, îepé /je′pe/ for singular and peîepé /peje′pe/ for plural, by the generic second-person free personal pronouns ine ~ indé /i′ne/ and penhé /pe′je/ respectively, LG retains the classic Tupi-Guarani person hierarchy with prefixes for the patient argument, as in Modern Paraguayan Guarani: Nde cherendu /ne se-r-e′nu/ ‘you.sg 1sg.inact-possm-listen’ (you listen to me) (Estigarribia, 2020, p. 139), i.e., analogical levelling has occurred across the paradigms.

In (4b), even though four physical orientation auxiliary ‘gerunds’ (/-jup-/ ‘horizontal extension’, /-Ɂam-/ ‘vertical extension’, /-en-/ ‘without extension’, /-kup-/ ‘in motion’) have not survived, the ‘be/exist’ root continues to express continuous aspect, i.e.,OT /-e′ko-/ + /-aβo/ à /-e′ko-βo/ > Mod. NHG -iku /-i′ku/. The auxiliary verb has lost the characteristicot same-subject subordinate-clause suffixes on intransitive active roots, i.e., /-a/ on consonant-final roots and /-aβo/ on vowel-final roots (Navarro, 2008, p.159-161, 172-173). The special intransitive active class number and person same-subject prefixes, i.e., /wi(t)-/ (1^st singular, /e-/ 2^nd singular) (2008, p. 161, 176-179) have suffered analogical levelling by the standard 1^st and 2^nd person active class subject prefixes, i.e., /a-/ and /(e)re-/. This has caused them to fall in line with the rest of the active class gerund personal prefixes, which are identical to those of the indicative mood, i.e., /ja-/ (1^st person plural inclusive), /oro-/ (1^st person plural exclusive), /pe-/ (2^nd person plural), and /o-/ (3^rd person subject-focus) (Finbow; O’Neill, 2022). This is also attested in Paraguayan Guarani, although the root that survived was ‘sit’, ‘unextended’ (OT v.rt /-in/ (sit) à ger /-en-a/), rather than ‘be/in motion’. It became the progressive aspect particle hína, e.g., rehai hína /re-h-ai hina/ 2sg.a-3p-write prog (you are writing) (Estigarribia, 2020, p. 164).

Thus, later 18^th century or early 19^th century examples from ‘Old Nheengatu’ phase of Língua Geral smooth the transition between Old Tupi and modern Nheengatu. The changes and the structural parallels between the three diachronic Língua Geral varieties are typical examples of language change in general and in line with tendencies observable in Paraguayan Guarani, the other Tupi-Guarani language in long-term close contact with Ibero-Romance, which is not claimed to have undergone creolisation. Therefore, it is is far less probable that unstructured, adult acquisition of koineised Amazonian Old Tupi followed by language shift created a pidgin that was subsequently creolised.

9. From macrolanguage to lingua franca

Lingua Geral’s lingua franca status should not be understood to mean that it was always or predominantly spoken non-natively, as is implied by accounts such as Lee (2005), Argolo (2011, 2012, 2016), Oliveira, Zanoli and Modolo (2019), which emphasize the demographic imbalance between native and non-native speakers as the catalyst for structural pidginization and subsequent creolisation when children begin to learn the pidgin as a mother tongue. Vernacular Lingua Geral would have certainly exhibited a very wide array of synchronic individual and collective linguistic competences. There would have been the “pidginised” speech of recent arrivals and infrequent users, both enslaved and free, Amerindian and European, at one extreme but, simultaneously, at the other, fluent non-native speakers and native speakers of the contemporary vernacular varieties whose ancestors had been Maranhão Tupinambá or speakers of other Tupi-Guarani languages, even if they no longer identified with those names. However, increased integration of indigenous communities into the colonial system would always favour ever-greater second-language fluency, followed by bilingualism that would mostly end in language shift and native-speaker competence in the contemporary vernacular Língua Geral. This is because, despite the impact of epidemic disease and chronic mistreatment, there were always fluent or native speakers of Língua Geral known as Tapejara (guides, lit., path owner, path master) available for Tapuias or Barés (newbies, PT novatos) recently brought down out the interior to model their speech (Daniel, 1757-76, vol. II, p. 258, see also Barros, Monserrat & Prudente, 2013, Barros, 2015, §39-40)[1]. This, rather than catechesis, was the primary way in which Lingua Geral was transmitted in the missions in the 18^th century, and this is what caused the functional change in the use of Língua Geral.

In Portugal’s South American colonies in the 17^th century, structural differences within the Língua Geral were primarily diatopic, within Tupi-Guarani languages, which were spoken natively by the indigenous populations and their descendants born from unions with Europeans. This changed with the largely forcible incorporation of innumerable non-Old Tupi and non-Tupi-Guarani-speaking peoples into the colonial system in Maranhão and Grão-Pará from the final quarter of the 17^th century. The missionaries’ decision not to catechise each indigenous people in their native tongue according to the Jesuit ideal, but use the missionaries’ codified variety for catechesis, except for preparing adult Tapuias for baptism or to administer extreme unction, when abbreviated versions of the catechism in the local language could be used[1] (Barros, 2015, §28-30), such as the ‘Questions on Christian Doctrine in the Manaus Language put into or taken from the Língua Geral’[2] (Joyce, 1951).

Bluteau and Daniel’s mid-18^th century descriptions emphasize diamesic and diaphasic differences that are primarily the result of the diastratic limitation of the Jesuit norm to the missionary class at that time who were the only group writing and reading frequently in the formal norm. Thus, another of Kloss’s technical terms could be applied to the Jesuit variety of LG, namely, Dachsprache (lit., roof-language), i.e., a norm that overarches varieties in a continuum. Such norms are typically deliberately elaborated, i.e., the result of Ausbau, e.g. standard ‘High’ German or standard Italian (Kloss, 1967; Muljačić, 1989, p. 256 ff.; Krefeld, 2020). In the 18^th century, Daniel shows us that the Jesuit norm is treated as a prescriptive standard by the missionaries, who regarded the Tapuias’ vernacular usage as improper but necessary for everyday communication.

Bluteau’s usage also shows that a semantic shift had occurred in the term Língua Geral between the 17^th and the 18^th century. Daniel shows that the term Língua Geral is still used to refer to the language of catechesis, but it no longer identifies a geographically extensive indigenous linguistic bloc of native speakers. Vernacular Lingua Geral is thought of as a supra-ethnic lingua franca employed in ‘war, religion and trade’ alongside the missionized indigenous communities’ native languages, which Bluteau classifies as ‘specific languages’ (línguas particulares) spoken by ‘barbarians’ and ‘savages’ who live ‘in the Interior’, ‘without hospitality or commerce’ or ‘in obstinate war’ (1721, t. VIII, p. 139).

Ultimately, the semantic shift detected in Língua Geral from ‘macrolanguage’ to ‘lingua franca’ also passed to the term Tapuia. From ‘non-Old Tupi Amerindian’ it came to mean the ‘detribalised’ and ‘re-cultured’ Língua Geral/Nheengatu-speaking indigenous or mestizo Amazonians of the 19^th century (Freire, 2011, Barros, 2015). In this way, Hartt (1938 [1875], ex. 684, 686, 687) registered Tapuia nheenga (Tapuia language/speech) as a synonym for Nheenga katu (good language) in the 1870s. Evidence that this semantic shift was already occurring during the 18^th century comes from Bluteau. In the entry for ‘Tapuya’ (1716, t. V, p. 140), he repeats the classic definitions from the 17^th century such as Simão de Vasconcelos (1595-1671) (Barros, 2015, §15), which characterise the Tapuia as a ‘generic nation’[1] of warlike, non-Christian savages that speak a plethora of languages. On the other hand, in the entry for ‘language’, where he develops fourteen categories of ‘parent and general languages’ (línguas matrizes e gerais) and ‘specific languages’[2] (línguas particulares) (1716, t. V, p. 138), Bluteau states that the Tapuia are the speakers of the Língua Geral ‘which occupies most of Brazil’, (1716, t. V, p. 139).

However, it is important to stress that starting with speakers of the Old Tupi varieties in the Estado do Brasil and the Maranhão Tupinambá in the North, throughout the koineization of Old Tupi that arose from cohabitation, slaving, and missionization, there ran an unbroken chain of intergenerational transmission within the oldest Amazonian colonial communities until their shift into Portuguese. The proportion of Tapuia or Baré learners to native speakers and fluent speakers (Tapeiara) certainly rose very markedly over time (Leite, 1943, t. IV, p. 139, Argolo, 2016; Finbow, 2022, 2023a; Freire, 2011; Hemming, 1987, 1995; Lee, 2005) and such a major demographic imbalance would have favoured restructuring in the direction of non-native variants (Mufwene, 2003). This is evident in 19^th century attestations of Língua Geral/Nheengatu, e.g., Hartt (1875), Couto de Magalhães (1876).

Interestingly, the kinds of structural changes that can be adduced, p. ex., loss of flexional paradigms[1], are more akin to what is seen in vernacular Brazilian Portuguese, which is the closest corollary in the Estado do Brasil to the ‘Tapuia-filtered’ Língua Geral in Maranhão and Grão-Pará (Finbow, 2022, p. 102).

10. Conclusion

Approaching Língua Geral sociophilologically reveals the problems of defining historical periodization and categorizations between languages/varieties in terms of structural differences. As Wright argues, sociolinguistics has shown that structural changes are usually gradual, and competing variants can coexist for very long periods. Moreover, there is no guarantee that modern perceptions of diversity correspond to those of the past (Finbow, 2010, 2012, 2022). The modern proposals regarding the emergence of the concept of Língua Geral that we have examined here are not backed up by solid evidence about contemporary usage and therefore have unfortunately fallen into anachronism or other kinds of misrepresentation.

In the case of Língua Brasílica and Língua Geral, modern researchers have attempted to apply these labels to divisions identified on the basis of the different kinds of Abstand detected in texts. However, in making such proposals, the categories they have used have become reified, such that many people today believe that ‘Língua Brasílica’ was an actual linguistic entity with a certain set of structural characteristics that endured for a certain amount of time before being replaced by other features which constituted another entity, namely, ‘Língua Geral’, which was divided into two varieties, the Paulista and the Amazonian. Those who think that way then ask themselves what it was that ‘changed LB into LG’, rather than looking at ‘Lingua Brasílica’ and ‘Língua Geral’ as labels which were applied in different ways in different times. However, we have shown in this paper that structural distance was not important to the naming practices employed amongst at least the non-indigenous inhabitants of Portugal’s South American domains between the 16^th, and 18^th centuries. Any indigenous language qualified for the name Língua Brasílica and diverse Tupi-Guarani languages fell under the umbrella-term Língua Geral, including the Jesuit norm which was called ‘the’ Língua Brasílica because it was the largest of the indgenous languages they had encountered. Consequently, the role of sociocultural mixing and, especially in the Brazilian context, racial mixing, which has been emphasized repeatedly as the driving force behind the structural changes that catalysed the use of new names has been vastly overstated.

Rodrigues (1996, p. 6) claims that earlier 20^th century writers had wrongly classified Língua Geral as either the same language spoken by the pre-colonial indigenous Tupi/ Tupinambá peoples or a new language created, moulded or ‘tamed’ by the Jesuits out of the pre-colonial language(s), or a pidgin/creole that arose from contact between diverse indigenous peoples and Europeans (see also Argolo, 2011a, 2012b, 2016; Lee, 2005, 2014). However, the technical definition Rodrigues seeks to reserve for the term Língua Geral as the language of the Mameluco class is every bit as ‘unfounded linguistically and historically’ (Rodrigues, 1996, p. 6) as the definitions that he criticises. In fact, it seems likely that his proposals regarding the Mamelucos’ role as multilingual innovators in the ‘emergence’ of the two varieties of Língua Geral that he identified was precisely what stimulated others to investigate the pidgin and creole hypothesis!

In the case of ‘the language most spoken on the coast of Brazil’ and its relatives and descendants, it is evident that contemporary commentators on matters of language do not regard the plurilingual Mameluco class as responsible for generating structural changes that caused the literate classes to consider that the name Língua Brasílica was inappropriate and therefore start to apply the term Língua Geral in a novel way to refer to the Mameluco’s speech. Thus, Rodrigues’s idea of employing Língua Geral to refer specifically to the Mameluco variety of Old Tupi is deeply unconvincing. An additional complication is there are no known historical records of Mameluco speech, so even if one were to accept Rodrigues’ proposals, we would know nothing about the kinds of structural changes they introduced because those that are discussed in the literature come from varieties that are too far apart chronologically for it to be safe to attribute them to the Mamelucos.

Where Rodrigues does find considerable documentary support is in Língua Geral being used to refer to other Tupi-Guarani languages besides Old Tupi, i.e., as a macro-linguistic category. We see this clearly in the writings of Pero Rodrigues, Manuel Gomes, Antônio Vieira and João Felipe Bettendorf in the 17^th century. In this regard, Wright’s concept of ‘complex monolingualism’ (1982, p. xi, 1993d: 207-8) is potentially useful. In the Late Latin/Early Romance context, this meant that speech and writing constituted a single conceptual unit for Early Medieval Romanophone peoples, despite wide structural divergences (Abstand) between diatopic, diachronic, diaphasic and diamesic modalities. Late Latin complex monolingualism mirrors how the concept of ‘Língua Geral Indians’ operated for the missionaries in Amazonia the second half of the 17^th century. Socioculturally diverse peoples were grouped together based on the proximity of their speech to that of the Jesuits’ norm.

A further point that Rodrigues identified was the semantic shift in the term Língua Geral in the 18^th century. However, he proposed that the shift was from the initial kind of usage that we have classified as ‘Tupi-Guarani macrolanguage’ to a more specific reference to the Mameluco variety. As we have shown above, it is highly unlikely that the Mamelucos were responsible for this semantic shift. The change in meaning that can be identified in the Amazon was recorded by Bluteau and by Daniel and points to the novel meaning for Língua Geral being ‘supra-ethnic lingua franca’, which many spoke natively, often alongside other indigenous languages. While structural change did occur, especially paradigm levelling, this was not beyond what would be expected from a situation in which large numbers of non-native speakers were acquiring and using the language in an unstructured manner. The way that Portuguese colonial society was structured gave considerable autonomy to unfree individuals to circulate, such that there were always ample opportunities to gain ever greater linguistic competence through regular contact with native and fluent speakers (Finbow, 2022, p. 86, 97-100). This is why no evidence can be found for the existence of a stable pidgin in the missions or colonists’ households. Thus, it is hard to see 18^th century Língua Geral as a ‘new’ language produced by classical creolization processes as Argolo, Lee, and Oliveira, Zanoli and Modolo have claimed. Despite considerable paradigm levelling, a growth in analytic structures, and some innovations, there remains a very clear structural continuity between the older and the more recent phases of what was called Língua Geral in Portuguese between the 17^th and the 19^th centuries.

Additional Information

Conflict of Interest

The author declares no competing interests.

Statement of Data Availability

The author confirms that the data, codes and materials supporting the findings of this study are available within the article [and/or] its supplementary materials.

References

ABOH, E. O. The Emergence of Hybrid Grammars: Language Contact and Change. Cambridge: Cambridge University Press, 2015.

ALFARO LAGORIO, M. A. C. “Usos e funções das línguas na área andina: o Terceiro Concílio Limense (1582-1583)”. In: ALFARO LAGORIO; FREIRE; ROSA, 2012, p. 41-54.

ALFARO LAGORIO, M. A. C. “Elementos da política lingüística colonial hispânica: o Terceiro Concílio Limense”. In: FREIRE; ROSA, 2003, p. 43-56.

ALFARO LAGORIO, C.; ROSA, M. C.; FREIRE, J. R. Bessa (org.). Políticas de línguas no Novo Mundo. Rio de Janeiro: EdUERJ, 2012.

ALTMAN, C. “As línguas gerais sul-americanas e a empresa missionária: linguagem e representação nos séculos XVI e XVII”. In: FREIRE; ROSA, 2003, p. 57-84.

AMMON, U. (org.). Status and Function of Languages and Language Varieties. Berlin: Walter de Gruyter, 1989.

ANCHIETA, J. de. Arte de Gramática da lingoa mais usada na costa do Brasil (1595). 5 ed. facsimile, Salvador, BA: EdUFBA, 2014.

ANCHIETA, J. de. Cartas. Correspondência ativa e passive [1553-1596]. 2 ed., Sao Paulo: Loyola, 1984.

ARAÚJO, A. de. Catecismo brasílico da doutrina crista. Edição fac-similar da 2. ed. de 1686, corrigida por Bartolomeu de Leão. Edited by Júlio Platzmann, Leipzig: B. G. Teubner, 1898.

ARGOLO, W. Línguas gerais na história social-linguística do Brasil. PAPIA, São Paulo, v. 26, n. 1, p. 7-52, 2016.

ARGOLO, W. História linguística do Sul da Bahia: levantando hipóteses e iluminando caminhos. Entrepalavras, v.1, n. 2, p. 270-292, 2012a.

ARGOLO, W. “A língua geral da Amazônia como um sistema historicamente novo: jesuítas e tapuias na origem do contexto com interrupção de transmissão linguística entre gerações”. In: MATTOS E SILVA; OLIVEIRA; AMARANTE, 2012b, p. 479-513.

ARGOLO, W. Introdução à história das línguas gerais no Brasil: processos distintos de formação no período colonial. Dissertação 2009 (mestrado em história). Universidade Federal da Bahia, 2011a.

ARGOLO, W. “Língua geral na Bahia: comarcas de Ilhéus e Porto Seguro”. In: CARVALHO; ROCHA; PARCERO, 2011b, p. 99-109.

ÁVILA, M. Twardowsky. Proposta de dicionário nheegatu-português. Tese (Doutorado em Estudos da Tradução). Faculdade de Filosofia, Letras e Ciências Humanas, Universidade de São Paulo, 2021.

AYROSA, P. O caderno da língua ou vocabulario portuguez-tupi de Frei João de Arronches, 1739: notas e commentarios á margem de um manuscripto do sec. XVIII. São Paulo: Imprensa Official do Estado, 1935.

AYROSA, P. (org.) Orações e diálogos da Doutrina Cristã na língua brasílica – mss. do século XVIII, transcritos e anotados por Plínio Ayrosa. São Paulo: Boletim CVI, Etnografia e Língua Tupi-Guarani Nº. 17, 1950.

BANNIARD, M. Viva voce: Communication écrite et communication orale du IVe au IXe siècle en occident latin. Paris: Institut des études augustiniennes, 1992.

BANNIARD, M. “The transition from Latin to the Romance languages”. In: MAIDEN; SMITH; LEDGEWAY, 2013, p. 57-106.

BARROS, M. C. Drummond de Menezes. Um caso de política lingüística: a questão do interprete e do discurso religioso no Brasil colonial. Amerindia n. 11, p. 69-77, 1986.

BARROS, M. C. Os línguas e a gramática tupi no Brasil (século XVI). Amerindia. Revue d'ethnolinguistique amérindienne. Actes du colloque international 'La "découverte" des langues et des écritures d'Amérique' (Paris, 7-11 setembre, 1993), 1994/95, p. 4-14.

BARROS, M. C. “Notas sobre a política jesuítica da língua geral na Amazônia (séculos XVII-XVIII)”. In: FREIRE; ROSA, p. 85-112, 2003.

BARROS, M. C. “O uso do tupi na Capitania de São Paulo no século XVII. Indícios na vida de um jesuíta “língua””. In: NOLL; DIETRICH, 2010, p. 141-153.

BARROS, M. C. Em razão das conquistas, religião, commercio’. Notas sobre o conceito de língua geral na colonização portuguesa da Amazônia nos séculos XVII-XVIII. Mélanges de la Casa de Valázquez v. 45, n. 1, p. 99-112, 2015.

BARROS, M. C. Drummond de Menezes; BORGES, L. C.; MEIRA, M. A língua geral como identidade construída. Revista de Antropologia, v. 39, n. 1, p. 191-219, 1996.

BARROS, C., MONSERRAT, R., PRUDENTE, G. Os “ocaporas” em listas de repartição de índios e nas fontes da língua geral na Amazônia no século xviii. Comunicação apresentada em 2013 na Jornada de Tierras Bajas. História e Antropologia, Santa Cruz de la Sierra, unpublished ms.

BEOZZO, J. O. Leis e regimentos das missões. Política indigenista no Brasil. São Paulo, 1983.

BETTENDORF, J. P. Crônica da missão dos padres da Companhia de Jesus na Província de Maranhão. Brasília: Senado Federal, Conselho Editorial, 2010.

BETTENDORF, J. P. Compêndio da doutrina crista na língua portuguesa e brasílica, etc. Lisboa: Officio de Miguel Deslandes (reimpressão por Frei José Mariano da Conceição Vellozo, Lisboa: Offic. de Simão Thaddeu Ferreira, 1800).

BLUTEAU, R. Vocabulario portuguez, e latino, ...: autorizado com exemplos dos melhores escritores portuguezes, e latinos; .... 8 v; 2 Suplementos. Coimbra: Collegio das Artes da Companhia de Jesu: Lisboa, Officina de Pascoal da Sylva, 1712-1728.

BOSSONG, G. Die romanischen Sprachen. Eine vergleichende Einführung. Buske, Hamburg 2008.

CABRAL, A.S. Arruda Camara; RODRIGUES, A. Dall’Igna (eds.), Línguas indígenas brasileiras: fonologia, gramática e história, Belém: UFPA, 2002.

CÂMARA JR., J. Mattoso. História e estrutura da língua portuguesa. Rio de Janeiro: Padrão, 1979.

CARDIM, F. Tratados da gente e terra do Brasil. São Paulo: Editora Hedra, 2009.

CARDEIRA, E. História do português. Lisboa: Caminho, 2006.

CARVALHO, C. dos Santos; ROCHA, F. Aninger de Barros; PARCERO, L. M. de Jesus (orgs.). Discurso e cultura: diálogos interdisciplinares. Salvador: EdUNEB, 2011.

CASTRO, Ivo. Curso de história da língua portuguesa. Lisboa: Universidade Aberta, 1991.

COSTA, J. M. Tamburelli; Tosco, M. Contested Languages: The Hidden Multilingualism of Europe. Histoire Épistémologie Langage, v. 44, n. 1, p. 201-205, 2022.

COUTO DE MAGALHÃES, J. Vieira Couto de. O selvagem. Rio de Janeiro: Typ. da Reforma, 1876.

DANIEL, J. Tesouro descoberto no máximo Rio Amazonas. Rio de Janeiro: Contraponto Editora, 2004.

DICCIONARIO anonymo da lingua geral do Brasil, publicado de novo com seu reverso por Julio Platzmann. Edição facsimilar. Leipzig: B. G. Teubner, 1896.

DIETRICH, W. O conceito de “Língua Geral” à luz dos dicionários de língua geral existentes. D.E.L.T.A., n. 30, p. 591-622, 2014.

DIETRICH, W. O “Vocabulario da Lingua Brazil” de Anselm Eckart (por volta de 1757-1759). Revista Brasileira De Linguística Antropológica, v. 16, n. 1, p. 258–285, 2024.

DIXON, R. M. W.; AIKHENVALD, A. (eds.). The Amazonian Languages. Cambridge: Cambridge University Press, 1999.

ECKART, A. O Exemplário da língua corrente do Brasil. Tradução do latim de Carlos Antônio Kalil Tannus e Miguel Barbosa do Rosário. Terceira Margem, UFRJ, Rio. ano 2, n. 2, p.176-180, 1994.

EDELWEISS, F. G. Tupis and Tupi-Guarani: estudos de etnonímia e linguística. Bahia: Secretaria de Eduação e Saúde, 1947.

EDELWEISS, F. G. Estudos Tupis e Tupi-Guaranis: confrontos e revisões. Rio de Janeiro: Livraria Brasiliana Editora, 1969.

ESTIGARRIBIA, B. A Grammar of Paraguayan Guarani. London: UCL Press, 2020.

ETHNOLOGUE, Lewis, M. Paul (Org.). Ethnologue. Languages of the World. 16^th ed., Dallas, TX: SIL, 2009.

FERGUSON, C. Diglossia. Word, v. 15, n. 2, p. 325–340, 1959.

FERNANDES, F. A organização social dos tupinambá. São Paulo: Instituto Progresso Editorial, 1948.

FIGUEIRA, L. Arte da Lingua Brasilica. Lisboa: Manoel da Silva, 1621.

FIGUEIRA, L. Arte de grammatica da lingua brasilica. Edição facsimilar, por Julio Platzmann, da 2. ed. de 1687. Lisboa: Miguel Deslandes. Leipzig: B. G. Teubner, 1880.

FINBOW, T. D. “A formação dos conceitos de “latim” e “romance””. In: LAGARES; BAGNO, 2011, p. 89-120.

FINBOW, T. D. “As categorias metalinguísticas tupinambá/língua geral e latim/romance: alguns paralelos sociofilológicos”. In: ALFARO LAGORIO; ROSA, FREIRE, 2012, p. 1-244.

FINBOW, T. D. The emergence and nature of the Língua Geral Amazônica’ in accordance with Mufwene’s Language Ecology model. Revista do GEL. v. 19, n. 2, p. 75-112, 2022.

FINBOW, T. D. Old Nheengatu? A “lost” alignment from 19^th century Amazonia and its implications for the evolution of the Língua Geral Amzônica. Paper to Amazônicas XI, 5^th – 9^th June 2023. Universidad de los Andes, Bogotá, Colombia, 2023a.

FINBOW, T. D. “Uma reanálise de dois casos de mudança estrutural entre o tupi antigo e a língua geral amazônica/nheengatu tidos como exemplos de crioulização e a apresentação de um caso de mudança gramatical no nheengatu do século XIX por contato com o português”. In: LIMA-HERNANDEZ; SANTOS; ADRIANO, 2023b, p. ....

FINBOW, T., O’NEILL, P. “Koineization and language contact: the social causes of morphological change in and with Portuguese”. In: LEDGEWAY; SMITH; VINCENT, 2022, p. 381-412.

FISHMAN, J. Bilingualism with and without diglossia; diglossia with and without bilingualism. Journal of Social Issues, v. 23, n. 2, p. 29–38, 1967.

FISHMAN, J. (ed.). Focus on Diglossia. International. Journal of the Sociology of Language 157. Belin: Mouton de Gruyter, 2002.

FRANÇA, E. Ferreira. Chrestomathia da língua brasílica. Leipzig: F. A. Brockhaus, 1859.

FREIRE, J. R. Bessa. Rio Babel. A história das línguas na Amazônia. 2. ed. Rio de Janeiro: Atlântica, 2011.

FREIRE, J. R. Bessa; ROSA, M. C. Línguas Gerais. Política linguística e catequese na América do Sul no período colonial. Rio de Janeiro: EdUERJ, 2003.

GALÚCIO, A. V.; MEIRA, M.; BIRCHILL, J.; MOORE, D.; GABAS JR., N.; DRUDE, S.; STORTO, L., PICANÇO, G., REIS RODRIGUES, C. Genealogical relations and lexical distances within the Tupian linguistic family. Boletim do Museu Paraense Emílo Goeldi. Ciencias Humanas, v. 10, n. 2, p. 229-274, maio-ago. 2015.

GOEBL, H. “Quelques remarques relatives aux concepts Abstand et Ausbau de Heinz Kloss”. In: AMMON, 1989, p. 278-290.

GOMES, M. Carta que o Padre Superior Manoel Gomes escreveu ao Padre Provincial do Brasil. Anais da Biblioteca Nacional do Rio de Janeiro, v. 26, 1904.

GÖSCHEL, J.; NAIL, N.; VAN DER ELST, G. (eds.), Zur Theorie des Dialekts: Aufsätze aus 100 Jahren Forschung. Zeitschrift fur Dialektologie and Linguistik, Beihefte, n. F., v. 16. Wiesbaden: F. Steiner, 1976.

HAAS, W. (ed.). Standard Languages Spoken and Written. Manchester, 1982.

HARTT, C. F. Notes on the Lingoa Geral or Modern Tupi of the Amazonas. Transactions of the American Philological Association, v. 3, p. 58-76, 1875.

HEMMING, J. Amazon Frontier. The Defeat of the Brazilian Indians. 2 ed. London: Papermac, 1995.

HEMMING, J. Red Gold. The Conquest of the Brazilian Indians. 2nd ed. London: Papermac, 1987.

HERMAN, J. Spoken and Written Latin in the Last Centuries of the Roman Empire. A Contribution to the Linguistic History of the Western Provinces, in R. Wright (ed.), Latin and the Romance Languages in the Early Middle Ages, p. 29-43, 1996.

HOUAISS, A. O português no Brasil. Rio de Janeiro: Unibrade - Centro de Cultura, 1985.

HUTTON, C. Linguistics and the Third Reich: Mother-tongue Fascism, Race and the Science of Language. London: Routledge, 1999.

JENSEN, C. “Tupi-Guarani”. In: DIXON; AIKHENVALD, 1999, p. 125-163.

JENSEN, T. “Language change and metalinguistic change: Latin to Romance and other cases”. In: WRIGHT, 1996, p. 19-28.

Joyce, M. de L. Caderno da doutrina pella lingoa dos Manaos. Manuscrito do século xviii». Boletim de Etnografia e Língua Tupi-Guarani, v. 22, p. 1-99, 1951.

KAYE, A. Modern Standard Arabic and the Colloquial. Lingua, v. 24, p. 374–91.

KAYE, A. (1972). Remarks on diglossica in Arabic: well-defined versus ill-defined. Linguistics v. 81, p. 32–48, 1972.

KLOSS, H. Abstand languages and Ausbau languages. Anthropological Linguistics, v. 9, n. 7, p. 29-41, 1967.

KLOSS, H. “Abstandsprachen und Ausbausprachen”. In: GÖSCHEL; NAIL; VAN DER ELST, 1976, p. 301-322.

KLOSS, H. Die Entwicklung neuer germanischer Kultursprachen seit 1800. 2 ed., Schwann, Düsseldorf, 1978.

KREFELD, T., Über ‚Dächer‘, ‚Schirme und Diversität – Sprachsoziologie im kommunikativen Raum. Korpus im Text. Serie A, 2020 (kit.gwi.uni-muenchen.de).

LAGARES, X. C.; BAGNO, M. (org.). Políticas de norma e conflito linguístico. São Paulo: Contexto, 2010.

LEDGEWAY, A.; SMITH, J. C.; VINCENT, N. (eds.). Periphrasis and Inflexion in Diachrony: a View from Romance. Oxford: Oxford University Press, 2022.

LEE, M. K. Conversing in Colony: The Brasílica and the Vulgar in Portuguese America 1500-1759. Ph.D. Dissertation, John Hopkins University, 2005.

LEE, M. K. “Language and Conquest: Tupi-Guarani Expansion in the European Colonization of Brazil and Amazonia”. In: MUFWENE, 2014, p. 143-167.

LEITE, F. R. A Língua Geral Paulista e o “Vocabulário elementar da Língua Geral Brasílica”. Dissertação (Mestrado em Linguística), Instituto de Estudos da Linguagem, UNICAMP 2013.

LEITE, S. História da Companhia de Jesus no Brasil, t. IV, Rio de Janeiro, 1943.

LIMA-HERNANDEZ, M. C.; SANTOS, M. M. Soares; ADRIANO, P. Soma (orgs.), Estudos de aquisição, Mudança e Ensino-aprendizagem de línguas. VIIIº Simpósio Mundial de Língua Portuguesa (SIMELP 8) / IIIº Simpósio Interdisciplinar de Estudos de Linguagem (SINTEL 3), Brasil – Angola, 4 a 7 de outubro de 2022. vol. 6. São Paulo: Estige Editorial, 2023.

LLOYD, P. M. “On the names of languages (and other things)”. In: WRIGHT, 1996, p. 9-18.

MADUREIRA, J.M. A liberdade dos índios, a Companhia de Jesus, sua pedagogia e seus resultados. Rio de Janeiro: Imprensa Nacional, 1977.

MAIDEN M.; SMITH, J.C.; LEDGEWAY, A (eds.). The Cambridge History of the Romance Languages. Cambridge University Press, 2013.

MAMIANI, L. V. Arte de grammatica da lingua brazilica da naçam Kiriri. Lisboa: Miguel Deslandes, 1699.

MAMIANI, L. V. Arte de grammatica da lingua brazilica da naçam Kiriri. Lisboa: Miguel Deslandes, 1699., 2a. edição (com notas introdutórias de Baptista Caetano de Almeida Nogueira). Rio de Janeiro: Bibliotheca Nacional, 1877.

MAMIANI, L. V. Catecismo da doutrina christãa na lingua brasilica da nação Kiriri. Lisboa: Miguel Deslandes, 1698.

MARTIUS, C. F. P. von. Beiträge zum Ethnographie und Sprachkunde Amerikas zumal Brasiliens. Glossarium linguarum Brasiliensium. Leipzig: Friedrich Fleicher, 1867.

MASON, J. Alden. “The languages of South American Indians”. In: STEWARD, J. H. (ed.), Handbook of South American Indians, v. 6. Washington, D. C.: Government Printing Office, p. 157-317, 1950.

MATTOS E SILVA, R.V. Ensaios para uma sócio-história do português brasileiro. São Paulo: Parábola, 2004.

MATTOS E SILVA, R. V.; OLIVEIRA, K.; AMARANTE, J. (org.). Várias navegações: português arcaico, português brasileiro, cultura escrita no Brasil, outros estudos. Em homenagem a Therezinha Barreto. Salvador: EdUFBA, 2012.

MELO, A. A. Souza; KNEIP, A. Novas evidências linguísticas (e algumas arqueológicas) que apontam para a origem dos povos Tupi-Guarani no leste amazônico. Literatura y lingüística, v. 36, p. 299-312, 2017.

MÉTRAUX, A. La réligion des Tupinamba et ses rapports avec celle des autres tribus Tupi-Guarani. Paris: Librairie Ernest Leroux, 1928.

MÉTRAUX, A. “The Tupinambá”. In: STEWARD, J. H. (ed.), Handbook of South American Indians, v. 3: The tropical forest tribes. Smithsonian Institution, Bureau of American Ethnology, Bulletin 143. Washington: Government Publishing Office, p. 193-196, 1948.

MÉTRAUX, A. A religião dos tupinambás. São Paulo: Companhia Editora Nacional; Editora da Universidade de São Paulo, 1979.

MICHAEL, L. D.; CHOUSOU-POLYDOURI, N.; BARTOLOMEI, K.; DONNELLY, E.; WAUTERS, V.; MEIRA, S.; O'HAGAN, Z. A Baysian Phylogenetic classification of Tupi-Guarani. LIAMES, v. 15, n.2, p. 1-36, 2015

MONSERRAT, R. “O Tupi do século XVIII (tupi médio)”. In: FREIRE; ROSA, 2003, p. 185-194.

MONSERRAT, R.; BARROS, C. de, MOTA, J. Comparação entre dois diólogos de doutrina jesuíticos tupi: João Filipe Bettendorf (1687) e José Vidigal (1740). I: Anais da XIII Jornadas Internacionais Missões Jesuíticas, Dourados, 2010.

MONSERRAT, R.; BARROS, C. de. Fontes manuscritas sobre a língua geral da amazônia escritas por jesuítas “Tapuitinga” (século XVIII). Confluência, v. 49, p. 236-254, 2015.

MONTEIRO, J. M. Os negros da terra: índios e bandeirantes nas orígens de São Paulo. São Paulo: Companhia das Letras, 1994.

MONTOYA, A. Ruiz de. Arte de la lengua guarani. 1640. Asunción: CEPAG, 2011.

MUFWENE, S. S. The Ecology of Language Evolution. Cambridge: C.U.P, 2003.

MUFWENE, S. S. Language Evolution. Contact, Competition and Change. London & New York: Continuum, 2008.

MUFWENE, S. S. (ed.). Iberian Imperialism and Language Evolution in Latin America. Chicago: University of Chicago Press, 2014.

MULJAČIĆ, Ž., “Über den Begriff Dachsprache”. In: AMMON, 1989, p. 256-277.

MÜLLER, K.-L. Zum Fortleben von LATINU und seinen Verwandten in der Romania”. Zeitschrift für romanischen Philologie, v. 79, p. 38-73, 1963.

NANTES, B. de. Katecismo indico da lingua Kariris. Lisboa: Officina de Valentim da Costa Deslandes, 1709.

NANTES, B. de. Catecismo indico da lingua Kariris. Edição facsimilar, por Julio Platzmann, da primeira edição (1709). Leipzig: B. G. Teubner, 1896.

NAVARRO, E. de Almeida. Dicionário de tupi antigo. A língua indígena clássica do Brasil. São Paulo: Editora Global, 2013.

NAVARRO, E. de Almeida. Método novo de tupi antigo. 3. ed. São Paulo: Editora Global, 2008.

NIKULIN, A. Proto-Macro-Jê: um estudo reconstrutivo. Tese (Doutorado em Linguística), Instituto de Letras, Universidade de Brasília, 2020.

NOLL, V. Das brasilianische Portugiesisch. Heidelberg: Universitätsverlag Winter, 1999.

NOLL, V. O português brasileiro: formação e contrastes. São Paulo: Globo. Tradução: Mário Eduardo Viaro, 2008.

NOLL, V.; DIETRICH, W. (org.). O português e o tupi no Brasil. São Paulo: Contexto, 2010.

OLIVEIRA, M. S. D. de; ZANOLI, M. de L.; MODOLO, M. O conceito de “Língua Geral do Brasil” revisitado à luz da linguística de contato. Journal of Ibero-Romance Creoles, v. 9, n. 1, p. 306-333, 2019.

PÉREZ PUENTE, L. La creación de las cátedras públicas de lenguas indígenas y la secularización parroquial. Estudios de historia novohispana, v. 41, p. 45-78, jul./dez. 2009.

PRAZERES, Fr. F. de N.S. dos. Poranduba maranhense, ou relação historica da provincia do Maranhão [...] com [...] um dicionario abreviado da lingua geral do Brazil. Revista Trimensal do Instituto Historico e Geographico Brazileiro, Rio de Janeiro, v. 54, pt. 1, p. [4]-277, 1891.

QUEIROZ, J. M. Correia de. Aspectos da fonologia Dzubukuá. Dissertação (Mestrado em Linguística). Universidade Federal da Paraíba, 2008.

QUEIROZ, J. M. Correia de. Um estudo gramatical da língua Dzubukuá, família Karirí. Tese (Doutorado em Linguística). Universiade Federal do Pernambuco, 2012.

RAIOL, D. A. Catechese de índios do Pará. Belém: Annaes da Bibliotheca e Archivo Público do Pará, v. 2, 1900.

RAMOS PÉREZ, D. La crisis Indiana y la junta magna de 1568. Jahrbuch für Geschichte Lateinbamerikas/Anuario de historia de América Latina, v. 23, p. 1-61, 1986.

REICH, U. “Mudança sintática e pragmática na Língua Geral Amazônica (LGA): marcação de caso e sistema pronominal”. In: FREIRE; ROSA, 2003, p. 167-184.

RIBEIRO, E. Rivail. On the inclusion of the Karirí family in the Macro-Jê stock: additional evidence. Paper presented at SSILA 2011 (Pittsburgh), January 7, 2011.

RODRIGUES, A. Dall’Igna. Classification of Tupi-Guarani. International Journal of American Linguistics, v. 24, n. 3, p. 231-234, 1958.

RODRIGUES, A. Dall’Igna. Contribuição para a etimologia dos brasileirismos. Revista Portuguesa de Filologia, v. 9, p. 1-54, 1958/59.

RODRIGUES, A. Dall’Igna. Relações internas na família Tupi-Guarani. Revista de Antropologia v. 27/28, p. 35-54, 1984/1985.

RODRIGUES, A. Dall’Igna. As línguas brasileiras. Para o conhecimento das línguas indígenas. São Paulo, Loyola, 1986.

RODRIGUES, A. Dall’Igna. As línguas gerais sul-americanas. Papia: Revista Brasileira de Estudos Crioulos e Similares v. 4, p. 6-18, 1996.

RODRIGUES, A. Dall’Igna. “Karirí”. In: DIXON, R. M. W.; & AIKHENVALD, A. Y. (orgs.), The Amazonian languages. Cambridge: Cambridge University Press, p. 164-206, 1999.

RODRIGUES, A. Dall’Igna. 2001. Flexão relacional no tronco Macro-Jê. Boletim da ABRALIN v. 25, p. 219-231, 2001.

RODRIGUES, A. Dall’Igna. “Tupi, tupinambá, línguas gerais e português do Brasil”. I: NOLL; DIETRICH, 2010a, p. 27-47.

RODRIGUES, A. Dall’Igna. Some cases of regrammaticalization in Tupí-Guaraní languages. Revista Brasileira de Linguística Antropológica, vol. 2, n. 2, p. 65-74, Dez. 2010b.

RODRIGUES, A. Dall’Igna. Karirí como família linguística Macro-Jê no Nordeste do Brasil. Revista Brasileira de Linguística Antropológica, v. 11, n. 1, p. 47-52, 2019.

RODRIGUES, A. Dall’igna; CABRAL, A. S. Arruda Camara. “Revendo a classificação interna da família Tupí-Guaraní”. In: CABRAL; RODRIGUES, 2002, p. 327-337.

RODRIGES A. Dall’Igna; CABRAL, A. S. Arruda Camara. A contribution to the linguistic history of the Língua Geral Amazônica. A.L.F.A, v. 55, n. 2, p. 613-639, 2010.

____. Descrições missionárias de língua geral nos séculos XVI-XVII: que descreveram? Papia, v. 1, p. 85-98, 1992.

SANTOS, B. Machado dos. As cartas do padre Manoel Gomes e suas representações em torno da incipiente missão jesuítica no Maranhão. Paper presented at XVI Jornadas sobre Alternativas Religiosas en América Latina. 2011. Accessed on 24/10/2023 at: https://ptdocz.com/doc/424876/as-cartas-do-padre-manuel-gomes-e-suas-representa%C3%A7%C3%B5es

SANTOS-GRANERO, F. Vital Enemies: Slavery, Predation and the Amerindian Political Economy if Life. Austin, TX: University of Texas Press, 2009

SARAGOÇA, L. De Feliz Lusitânia aos Confins da Amazónia 1616-62. São Paulo: Cosmos, 2000.

SILVA, J. R. Araújo da (Guyraakanga Potiguara); ARAGON, C. Coelho; BARBOSA, D. da Silva (org). Tupi potiguara kuapa. Conhecendo a língua Tupi Potiguara. Mamanguape, PB: editora do autor, 2024.

SILVA NETO, S. da. Introdução ao estudo da língua portuguesa no Brasil. Rio de Janeiro: Presença, 1951.

SOUSA, G. Soares de. Tratado descritivo do Brasil em 1587. São Paulo: Editora Hedra, 2010.

SCHMIDT-RIESE, R. “Condições da mudança em nheengatu: pragmática e contatos linguísticos”. In FREIRE; ROSA, 2003, p. 147-166.

TEYSSIER, P. História da língua portuguesa. 3. ed. São Paulo: Martins Fontes. Tradução: Celso Cunha, 2007.

TEJEDO-HERRERO, F. Algunas reflexiones en torno al término latín en la documentación alfonsí. Romance Quarterly, v. 56, n. 1, p. 4-12, 2009.

VIEIRA, M. M. Damaso. A Manifestação de Tópico e Foco em Línguas da Família Tupi-Guarani. D.E.L.T.A., v. 30, p. 659-683, 2014.

WILEY, T. G. Heinz. Kloss Revisited: National Socialist Ideologue or Champion of Language-minority Rights?, International Journal of the Sociology of Language, v. 154, p. 83-97, 2002.

WOOLARD, K. A.; GENOVESE, E. N. Strategic bivalency in Latin and Spanish in early modern Spain. Language in Society, v. 36, n. 4, p. 487-509, 2007.

WRIGHT, R. Speaking, reading and writing Late Latin and Early Romance. Neophilologus v. 60, p. 178-89, 1976.

WRIGHT, R. Late Latin and Early Romance in Spain and Carolingian France. Liverpool: F. Cairns, 1982.

WRIGHT, R. “Versatility and vagueness in Early Medieval Spain”. In: MACKENZIE; MICHAEL, 1993, p. 207-223.

WRIGHT, R. (ed.) Latin and the Romance Languages in the Early Middle Ages. Reprint 1996: University Park, PA: Pennsylvania State University Press, 1996.

WRIGHT, R. Early Ibero-romance: Twenty-One Studies on Language and Texts from the Iberian Peninsula between the Roman Empire and the Thirteenth Century. Newark, DEL: Juan de la Cuesta, 1994.

WRIGHT, R. Reading a will in twelfth-century Salamanca. Latin vulgaire – latin tardif, v. 5, p. 505-516, 1999.

WRIGHT, R. The assertion of Ibero-Romance’, Forum for Modern Language Studies, v. 36, p. 231-240, 2000a.

WRIGHT, R. El Tratado de Cabreros (1206): estudio sociofilológico de una reforma ortográfica. London: Department of Hispanic Studies Queen Mary and Westfield College, 2000b.

WRIGHT, R. A Sociophilological Study of Late Latin. Turnhout: Brepols, 2002.

WRIGHT, R. Sociophilology and twelfth-century Spain. Medioevo romanzo. v. 27, n. 2, mag./ago., p. 257-274, 2003.

WRIGHT, R. The Monolingual Latin Glossaries of the Iberian Peninsula: Can They Help the Romanist?”. In: Van Acker, M.; Van Deyck, R.; Van Uytfanghe, M. (eds.), Latin écrit-Roman oral, De la dichotomisation à la continuité. Turnhout: Brepols, p. 137-158. 2008.

ZAVALA, S. El castellano, lengua obligatoria? SEP, Coordinacion de publicaciones, Promoción Editorial y Bibliotecas, 1977.

Review

DOI: https://doi.org/10.25189/2675-4916.2025.V6.N1.ID806.R

Editorial Decision

EDITOR 1: Joe Salmons

AFFILIATION: University of Wisconsin–Madison, Wisconsin, Estados Unidos.

-

EDITOR 2: Josh Brown

AFFILIATION: University of Wisconsin-Eau Claire, Wisconsin, Estados Unidos.

-

EDITOR 3: David Natvig

AFFILIATION: University of Wisconsin–Madison, Wisconsin, Estados Unidos.

-

DECISION LETTER:

Rounds of Review

REVIEWER 1: Josh Brown

AFFILIATION: University of Wisconsin-Eau Claire, Wisconsin, Estados Unidos.

-

REVIEWER 2: Zachary O’Hagan

AFFILIATION: University of California, California, Estados Unidos.

-

ROUND 1

REVIEWER 1

2025-01-29 | 03:16 PM

This manuscript provides an analysis of Língu Geral and how modern conceptualizations of languages do not neatly map onto historic ones. It shows that earlier ideas about the naming of varieties are incorrect.

This manuscript is well written and uses appropriate evidence. There are several areas where the writing could be tighter and the visualization of data in tables would be better than in prose. The author convincingly shows problems in modern analyses of varieties that do not fully capture the historic conceptualization. My major concern is the use of terminology and heuristics from Heinz Kloss in a manuscript on indigenous linguistic research, as Kloss was variously employed in institutions of the Third Reich (see Hutton 1999). This should be acknowledged somewhere in the paper.

-

REVIEWER 2

2025-04-30 | 04:03 AM

I commend the authors for tackling a difficult set of nomenclatural issues in the colonial literature and subsequent interpretation thereof. However, the article as it is is extremely densely written. Considerable familiarity with the historical and linguistic literature is assumed by the authors, as well as frameworks and resultant terminologies from many scholarly traditions, to a degree that I feel the chapter cannot be published without significant revisions that warrant resubmission. I've attached a number of comments in the PDF, some trivial, others indicating where I feel the authors can significantly expand various discussions. In general, I feel it is of paramount importance to continually remind the reader throughout the chapter what the claims of the authors are, and for those claims to be stated in the simplest possible terms. For example, are we not dealing with a situation of a standard and a non-standard plus diglossia? Of course it is not simple to demonstrate that, but even the authors' conclusions are somewhat opaque to me (e.g., the notion of a macrolanguage is not clear). Separately, the chapter could more generally benefit from more explicit framing (roadposting) and ways for the reader to digest summaries of complex historical reviews (e.g., in the form of tables). Changes of this nature could easily make the chapter fifty percent longer, but I feel it would be worth it. I encourage the authors to pursue revision of the paper, as I feel there are very important issues to be dealt with.