Glocal Languages and Critical Intercultural Awareness: The South Answers Back 9780815395058, 9781351184656

This volume provides a new perspective on prevailing discourses on translanguaging and multilingualism by looking at ‘gl

223 109 1MB

English Pages [255] Year 2019

Report DMCA / Copyright

DOWNLOAD PDF FILE

Table of contents :
Cover
Half Title
Series Page
Title
Copyright
Contents
Acknowledgments
Introduction: Glocal Languages, the South Answering Back
SECTION I Glocal Languages—Theoretical Background
1 Glocal Languages, Coloniality and Globalization From Below
2 Glocal Languages Beyond Post-Colonialism: The Metaphorical North and South in the Geographical North and South
SECTION II Indigenous Languages as Glocal Languages
3 Glocalism Now and Then: The De-Colonial Turn of Guarani, Portuguese and Spanish
4 Reshuffling Conceptual Cards: What Counts as Language in Lowland Indigenous South America
SECTION III Portuguese as Glocal Language
5 The Imaginary in Portuguese Language Perceptions in Academia: (Mis)directions Between the Local and the Global
6 The Linguistic Atlas of Brazil Project: Contributions Towards Knowledge, Teaching and Disclosure of Brazilian Portuguese
SECTION IV Spanish as Glocal Language
7 Comparisons Between Spanish and Portuguese: Proposals for University Teaching
8 Multiculturalism and Glocal Languages: The Impact of Cultural Mobility in Spanish Teaching and Learning in Southern Brazil
SECTION V English as Glocal Language
9 English (Mis)education as an Alternative to Challenge English Hegemony: A Geopolitical Debate
10 Teaching English to Undergraduate Students in a Brazilian University: Thinking Glocally
Conclusion: Towards Globalization From Below
Contributors
Index
Recommend Papers

Glocal Languages and Critical Intercultural Awareness: The South Answers Back
 9780815395058, 9781351184656

  • 0 0 0
  • Like this paper and download? You can publish your own PDF file online for free in a few minutes! Sign Up
File loading please wait...
Citation preview

Glocal Languages and Critical Intercultural Awareness

This volume provides a new perspective on prevailing discourses on translanguaging and multilingualism by looking at ‘glocal’ languages, local languages which have been successfully ‘globalised’ and relocalised. Focusing on European languages recreated in Latin America, the book features examples from languages underexplored in the literature, namely Brazilian Portuguese, English and Spanish teaching in Brazilian universities, including Amerindian languages, as a basis for advocating for an approach to language education rooted in critical pedagogy and decolonial perspectives while countering hegemonic theories of globalisation. Being rooted in a discussion of the South, the book offers a fresh voice in current debates on language education that will be of broader interest to students and scholars across disciplines, including language education, multilingualism, cultural studies, and linguistic anthropology. Manuela Guilherme is Senior Researcher at the Center for Social Studies at the University of Coimbra, Portugal, and was a Marie SklodowskaCurie Research Fellow (2014–2017). Lynn Mario T. Menezes de Souza is Professor of Language Studies at the University of São Paulo, Brazil.

Routledge Studies in Language and Intercultural Communication Edited by Zhu Hua, Birkbeck College, University of London Claire Kramsch, University of California, Berkeley

Language and Intercultural Communication in the New Era Edited by Farzad Sharifian and Maryam Jamarani Reflexivity in Language and Intercultural Education Rethinking Multilingualism and Interculturality Edited by Julie S. Byrd Clark and Fred Dervin Researching Identity and Interculturality Edited by Fred Dervin and Karen Risager Online Intercultural Exchange Policy, Pedagogy, Practice Edited by Robert O’Dowd and Tim Lewis The Critical Turn in Language and Intercultural Communication Pedagogy Theory, Research and Practice Edited by Maria Dasli and Adriana Raquel Díaz Interculturality, Interaction and Language Learning Insights from Tandem Partnerships Jane Woodin Beyond Native Speakerism Current Explorations and Future Visions Stephanie Ann Houghton, Damian J. Rivers and Kayoko Hashimoto Screens and Scenes Multimodal Communication in Online Intercultural Encounters Edited Richard Kern and Christine Develotte Glocal Languages and Critical Intercultural Awareness The South Answers Back Edited Manuela Guilherme and Lynn Mario T. Menezes de Souza www.routledge.com/Routledge-Studies-in-Language-and-InterculturalCommunication/book-series/LICC

Glocal Languages and Critical Intercultural Awareness The South Answers Back Edited by Manuela Guilherme and Lynn Mario T. Menezes de Souza

First published 2019 by Routledge 52 Vanderbilt Avenue, New York, NY 10017 and by Routledge 2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN Routledge is an imprint of the Taylor & Francis Group, an informa business © 2019 Taylor & Francis The right of the editors to be identified as the authors of the editorial material, and of the authors for their individual chapters, has been asserted in accordance with sections 77 and 78 of the Copyright, Designs and Patents Act 1988. All rights reserved. No part of this book may be reprinted or reproduced or utilised in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the publishers. Trademark notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation without intent to infringe. Library of Congress Cataloging-in-Publication Data A catalog record for this book has been requested ISBN: 978-0-8153-9505-8 (hbk) ISBN: 978-1-351-18465-6 (ebk) Typeset in Sabon by Apex CoVantage, LLC

Contents

Acknowledgments Introduction: Glocal Languages, the South Answering Back

vii 1

MANUELA GUILHERME AND LYNN MARIO T. MENEZES DE SOUZA

SECTION I

Glocal Languages—Theoretical Background 1 Glocal Languages, Coloniality and Globalization From Below

15

17

LYNN MARIO T. MENEZES DE SOUZA

2 Glocal Languages Beyond Post-Colonialism: The Metaphorical North and South in the Geographical North and South

42

MANUELA GUILHERME

SECTION II

Indigenous Languages as Glocal Languages 3 Glocalism Now and Then: The De-Colonial Turn of Guarani, Portuguese and Spanish

65

67

FERNANDA MARTINS FELIX

4 Reshuffling Conceptual Cards: What Counts as Language in Lowland Indigenous South America JAMILLE PINHEIRO DIAS

90

vi

Contents

SECTION III

Portuguese as Glocal Language 5 The Imaginary in Portuguese Language Perceptions in Academia: (Mis)directions Between the Local and the Global

105

107

GESUALDA DOS SANTOS RASIA

6 The Linguistic Atlas of Brazil Project: Contributions Towards Knowledge, Teaching and Disclosure of Brazilian Portuguese

124

MARCELA MOURA TORRES PAIM AND SILVANA SOARES COSTA RIBEIRO

SECTION IV

Spanish as Glocal Language 7 Comparisons Between Spanish and Portuguese: Proposals for University Teaching

147

149

ADRIÁN PABLO FANJUL

8 Multiculturalism and Glocal Languages: The Impact of Cultural Mobility in Spanish Teaching and Learning in Southern Brazil

163

MARIA JOSELE BUCCO COELHO

SECTION V

English as Glocal Language 9 English (Mis)education as an Alternative to Challenge English Hegemony: A Geopolitical Debate

181

183

DANIEL DE MELLO FERRAZ

10 Teaching English to Undergraduate Students in a Brazilian University: Thinking Glocally

207

ALESSANDRA COUTINHO FERNANDES

Conclusion: Towards Globalization From Below

228

LYNN MARIO T. MENEZES DE SOUZA AND MANUELA GUILHERME

Contributors Index

242 245

Acknowledgments

This book owes much to the work carried out within the scope of the first phase (Oct. 2014–Oct. 2015) of the following project (Sept. 2014–Sept. 2017). ‘Glocal Languages’ and ‘Intercultural Responsibility’ in a postcolonial global academic world (GLOCADEMICS): Power relations between languages/cultures within and between research groups (www.ces. uc.pt/projectos/glocademics) funded by an individual grant awarded by the European Commission (Marie Sklodowska-Curie Actions) to Maria Manuela Duarte Guilherme, Principal Investigator and sole author of the proposal.

Project’s home institution (2014–2017): Centro de Estudos Sociais, Universidade de Coimbra General Coordinator of the Project: Boaventura de Sousa Santos Project’s home institution for the empirical study in Brazil: Departamento de Letras Modernas, Universidade de S. Paulo Coordinator of the Project in Brazil: Lynn Mario T. Menezes de Souza Therefore, the book editors are particularly grateful to the participants of Phase 1 of the project (2014–2015) who have accepted to author most of the chapters of this book and to make public some of their contributions to the project. I, Manuela Guilherme, am deeply grateful for the efficient and valuable support, even at a distance, provided by the staff at the Centro de Estudos Sociais, Universidade de Coimbra, namely its Executive Director, João Paulo Dias, the Project Manager, André Caiado, the Coordinator of the Project Management Section (GAGEP), Rita Pais, the ICT Technician

viii

Acknowledgments

Pedro Abreu and the CES librarians, Maria José Carvalho, Acácio Machado and Inês Lima, as well as for their solidarity in times of personal distress. I am also very thankful to the other project participants who also contributed to Phase 1 of the project, most of whom are collaborating in other publications, as well as those colleagues who facilitated the implementation of the project in their departments and deserve my highest appreciation, namely Lynn Mario T. Menezes de Souza, Sávio Siqueira and Clarissa Jordão. I felt welcome, both personally and professionally, and intellectually stimulated by all the colleagues who hosted me in their various academic events at the Departmento de Línguas Modernas at the Universidade de S. Paulo, Instituto de Letras at the Universidade Federal da Bahia, and Departamento de Línguas Modernas and Centro de Línguas e Interculturalidade (CELIN), at the Universidade Federal do Paraná. Last but not the least, I am most grateful to Lynn Mario T. Menezes de Souza (USP) and Sávio Siqueira (UFBA) for having hosted me in their research groups and, therefore, also to their doctoral and postdoctoral students.

Introduction Glocal Languages, the South Answering Back Manuela Guilherme and Lynn Mario T. Menezes de Souza This book aims to shed light on a particular perspective of the globalness and localness of some “glocal languages” (Portuguese, Spanish, English and Indigenous languages), having Brazil as its locus of enunciation and offering some highlights of language education in higher education there. The co-authors of this book are all involved in language education at the undergraduate level, and some also are in postgraduate programmes— most of them targeting teacher education—or have just completed their doctoral or post-doctoral research in any of the Brazilian universities participating in the “Glocademics” project, which was funded by a Marie Sklodowska-Curie fellowship, whose implementation in Brazil was coordinated by the book co-editors. The following texts emerge from within a study comprehending a small sample of language teaching (27 teachers of Portuguese, home and foreign language, Spanish and English as foreign or additional languages, and Indigenous languages, as additional languages) evenly distributed in three federal universities in Brazil.1 The study was based on curriculum analysis as well as direct written and oral statements by each participant teacher about their corresponding analysed syllabi and the theoretical background underlying their teaching activities and, finally, their conceptual frameworks. The following chapters aim to give public voice to the project participants who accepted this challenge. The numbers of language educators in the Brazilian higher education landscape are overwhelming; therefore, this book is far from aiming to provide a quantitatively representative view of the targeted field, but, by including the texts of scholars whose voice is acknowledged and highly respected by their colleagues, we hope to offer texts that qualitatively illustrate the thought and the practices of language education in the selected highly rated universities in Brazil. Brazil covers an immense territory, almost half of South America. It became independent in the beginning of the 19th century, and two main moments when the Portuguese language was more fiercely imposed through nationwide linguistic policies can be singled out. First, in the late 18th century, by laws issued by the Marquês de Pombal, who was the authoritarian Secretary of State to the Portuguese King D. José I

2

Guilherme and Menezes de Sousa

and who led the reconstruction of the city of Lisbon after the big 1775 earthquake as well as the political organisation of the territory. Second, during the 20th century, in the ’30s and ’40s, under the linguistic policies and the reinforcement of nationalism by Brazilian President Getúlio Vargas. Anyway, the Portuguese language had already become the language adopted by high society, to some extent also in rural areas, during the stay of the Portuguese Royal family and court in Brazil while fleeing from Bonaparte’s invasion in the first half of the 18th century. The representation of the Portuguese language had also increased exponentially throughout the 18th century due to a substantial inflow of Portuguese colonisers (Lucchesi 2015). However, when European sailors and missionaries arrived at the coastal areas of Brazil they found more than a thousand Indigenous languages, most of which belonged to the largest tupi branch, whose tribes had been expelling other tribes with different cultural matrices for the past two centuries (Ribeiro 1995: 28). Several African languages arrived later through the arrest and forced importation of slaves. The situation during the 15th and 16th centuries was therefore one of heavy multilingualism, not only Indigenous but also European, such as Dutch in the north-eastern Recife area for example. However, due to increased mobility and contact between the Indigenous populations and the immigrants, voluntary and involuntary, both from Europe and Africa, some “common languages” (língua geral) started to become popular in the state of S. Paulo, later in the north-east and then travelled the rivers into Amazonia. These common languages, línguas gerais, were heavily based in specific stronger Indigenous languages, “tupi” and “tupinambá” “pidginised” with Portuguese and African languages, and later known as língua geral paulista, língua brasílica, língua geral amazónica and the latter also called Nheengatu in the 20th century (Ribeiro 1995; Freire 2014). Such common languages—língua geral— were spread and registered by the colonisers (bandeirantes) and the Jesuit missionaries, the latter later expelled by the above mentioned Marquês de Pombal. Freire (2014) compares this process of creating common languages, in very wide regions, for religious, business, rural labour, social life, both in the private and public spheres, and still used for local government purposes, to the creation of “imagined communities” as proposed by Benedict Anderson. According to the same author, such “common languages” were widely used for “interethnic communication” in the states of São Paulo, Minas Gerais, Goiás, Mato Grosso and Pará and survived several prohibitions from the late 16th century to the late 19th century. In the late 18th century, the predominance of the língua geral also in large regions of inland on the Portuguese side in South America made difficult the separation between the lands owned by the Portuguese and the Spanish Crowns which were divided according to linguistic dominance (Freire 2014). However, it was due to this linguistic and cultural “plasticity” that the Portuguese colonisers, in too small numbers, managed to

Introduction

3

control the large territory of Brazil (Holanda 2016: 211–224). The 20th century was even more aggressive both towards Indigenous and immigrant languages (Italian, Dutch, Japanese, German, etc.), not only due to violently enforced linguistic policies for the use of Portuguese, both in the public and private spheres, but also due to the expansion of the education system. As a consequence, the Portuguese language definitely took the floor in almost half of South America. In the beginning of the 21st century, English and Spanish are also becoming important players in the economy and, therefore, in language education as well as in the internationalisation of education, the former pushed by globalisation and the latter by the Mercosul economic agreement (Asuncion Treaty Mercosur 1991). This book undertakes a reverse South-North perspective; assuming that North and South are more than geographical positions, they are used as metaphors which express symbolic and conventional cultural and epistemological baggage present both in the geographical south and north. “North” and “South” are used not as ontological or geographical reference points but as epistemological sites involved in hegemonic relations of power both regionally and globally. Following this rationale, in terms of epistemological production, there are nowadays local souths within a global North and local norths within a global South. However, historically, this metaphor is inspired by the colonisation crusade that did move from North to South, that is, from Europe to wherever the sailors, missionaries and armies, altogether, could find their way, which happened to be southwards across the Atlantic Ocean. This book, and the Glocademics project from which it emerges, were inspired by Sousa Santos’ theory on the Epistemologies of the South which is at the roots of what he calls the “ecology of knowledges” which demands an “intercultural translation” through which he argues that “it is imperative to start an intercultural dialogue among different critical knowledges and practices: South-centric and North-centric, popular and scientific, religious and secular, female and male, urban and rural, and so forth” (2014: 42). Promoting an ecology of knowledges does not, according to Sousa Santos, mean accepting relativism; on the contrary, it proposes “to reassess the concrete interventions in society and in nature that the different knowledges can offer” (205). Otherwise, it means that every knowledge, and every language, is incomplete, hence, in need to be reciprocally complemented. Furthermore, the “Epistemologies of the South” does not propose the “South-North” divide as another dichotomy; instead it claims for a dialogue in equity among their pluralities. For example, Estermann (2008) explains that the Indigenous person cannot be defined in an essentialist manner, rather as part of a dialogical procedure, although violently imposed, between different philosophical models and not from the point of view of the dominant philosophical model only or

4

Guilherme and Menezes de Sousa

otherwise (“No podemos, en fin, definir lo ‘indígena’ de manera essencialista sin caer con ello ya en una presuposición filosófica no articulada [la ‘mania clasificatoria’ del Occidente]. Lo ‘indígena’ tiene que surgir justamente como el resultado de un proceso dialógico entre diferentes modelos filosóficos, y no como un a priori de la filosofia dominante”) (101). Sousa Santos (2007) elaborates on this divide, to be overcome, in terms of an “abyssal line” that has separated, on the one side of the line, scientific knowledge from, on the other side of the line, “beliefs, opinions, intuitive or subjective understandings, which, at the most, may become objects or raw materials for scientific inquiry” (47). Within such a conventionally imposed divide, if so understood, there is the adoption of scientific standards and academic criteria that meet cultural frameworks from the other side of the abyssal line, and which makes ratings of excellence more distant to the other side of the line. Post-abyssal times require that the south talks back, which “does not mean discarding the rich Eurocentric critical tradition” (Sousa Santos 2014: 44), instead that the South is also entitled to add their voices and that these are not only respected but that they also do count for a sustainable and ecological balance of knowledges. Such a position entails the negation of what CastroGómez (2005) calls the zero-point hubris—la hybris del punto cero—the knowledge that presumes to be the centre, before which nothing existed, universal and neutral, uncommitted, objective and unlimited unless by itself, therefore, that refuses any particular locus of enunciation, in sum, that observes without being observed. Our approach to the term we are introducing and developing here— “glocal languages”—is also inserted in what some authors call a decolonial turn (“El giro decolonial”) (Castro-Gómez and Grosfoguel 2007). This means not only leaving political colonialism behind, by moving into postcolonial governance, but also undoing the supremacy of the colonial epistemic heritage, by endorsing the creativeness of mestizage as equally legitimate, as well as by revitalising the remains of the colonial “epistemicide”, the term used by Sousa Santos to name “the murder of knowledge” (2014: 92). According to Mignolo, a “de-colonial politics of knowledge” that entails a “knowledge-making for wellbeing rather than for controlling and managing populations for imperial interest shall come from local experiences and needs, rather from local imperial experiences and needs projected to the globe” (2009: 19). Elsewhere, Mignolo describes the decolonial thought as one that liberates and opens itself (“el pensamiento decolonial es, entonces, el pensamiento que se desprende y se abre”) and explains that decoloniality of thought started occurring in simultaneity with epistemic colonisation throughout the history of colonisation (2007: 27). Moreover, Mignolo also states that the genealogy of decolonial thought is pluriversal, not universal (“la genealogia del pensamiento decolonial es pluriversal [not universal]”) (ibid.: 45).

Introduction

5

Spivak (1999) in her Critique of Postcolonial Reason also concedes that “a certain postcolonial subject had, in turn, been recoding the colonial subject and appropriating the Native Informant’s position” (1999: ix), meaning that postcolonial studies had not eventually made room for the colonised to re-appropriate the legitimacy of their identity, heritage and voice, in direct speech, but that the colonised is still being reported and analysed in scientific terms. The dialectical relation between coloniality and decoloniality, the ontological and epistemological nature of the colonial and decolonial beings, expresses itself well when referring to the “native/non-native” cultural and linguistic essence and status. The concept of “nativeness”, no matter how mythical this concept is, is still today hierarchically dependent on the colonial matrices, e.g. the native Americans, the native languages, versus the native speaker of English, versus the native speaker of Portuguese or the native speaker of Spanish. We agree with Kumaravadivelu when he calls not only for “the unfreezing of the subaltern’s potential for thinking otherwise” (2014: 79) but also for them to “activate [their] latent agentive capacity” (ibid.: 81). Therefore, Kumaravadivelu (2014) claims that “a grammar of decoloniality, if it is to be useful and useable, has to be formulated and implemented by local players” even though “they can, of course, be guided by a broader framework” (81) to which, we add, they should also have a say. And this also goes for languages which are viewed as global, like English, and whose localness is also “lost in translation”. This is how we view global-local— glocal—reciprocal dynamics, with mutual unlearning, learning and relearning. Language education cannot escape this challenge. Our term “glocal languages” is inspired by the idea of “glocalization” put forward by Robertson to replace the terms “globalization” and “globality” which were, according to him, being loaded with notions with which he did not agree. Therefore, he adopted the term “glocalization”, which already existed, in order to highlight some features of the globallocal nexus that had been neglected, and which we also find relevant, such as that “what is called local is in large degree constructed on a trans- or super-local basis” (1995: 26) and that “globality at this point being viewed in terms of the interpenetration of geographically distinct ‘civilizations’” (ibid: 27). We are not, however, using the term “glocal” in order to clarify our understanding of the global dimension; instead, we believe that the impact of the “local” in the “global” is as strong as the reverse and, moreover, they are not in a dichotomous relationship but closely intertwining with each other. According to Urry, the permanent interaction between the global and the local takes the form of a “complex relationality”, as inspired by Marx and therefore between superstructure and agents, that “explains the ways in which local forms of information and action can result in the emergence of far-from-equilibrium system effects” (2005: 242). He adapts his analogy to globalisation by stating that “globalization

6

Guilherme and Menezes de Sousa

(or global capitalism) is the new ‘structure’, while nations, localities, regions and so on, comprise the new ‘agent’” (ibid.: 242) and we assume that one needs the other. Therefore, no matter the distance between the superstructure and the local agency, in globalisation, they are both very powerful, intense and tied together. However, capitalist globalisation viewed from the southern hemisphere seems even more invasive than from the perspective of the northern hemisphere because its language (English), its culture (technology and science) and its religion (secularism and atheism) have emanated from the North, so much that sometimes it is even forgotten by either side that, at this stage, it is colonising the geographical north as much as the south. Milton Santos (2000), a Brazilian geographer, calls it “globalitarianism” (“globalitarismo”), in order to connect it with authoritarianism. Further inspiration for the adoption of this terminology comes from Sousa Santos’ forms of production of globalisation, namely that of a “localised globalism”, that the author describes as “the specific impact of transnational practices and imperatives on local conditions that are thereby destructured and restructured in order to respond to transnational imperatives” (1999: 217–218). However, our focus on “glocal languages” aims to go beyond the simple impact of the global on the local and it concentrates on the response of “localised globalisms” back to globalising practices and meanings. Sousa Santos’ identification of the four forms of globalisation gives space for a “decolonial turn”, since it acknowledges the localness of globalness by pinpointing that globalisation starts with: (a) “globalised localism”, a localism that turned global but which, nevertheless, remains a localism, which homogeneous globalisation attempts to cover, and follows with the message that such “globalised localism” turns into a (b) “localised globalism”, that is, it eventually lies in the hands, minds and hearts of local agents. Furthermore, the two remaining forms of production of globalisation identified by Sousa Santos (1999) also encompass two potential forms of agency for localness in the circumstances of globality: (c) cosmopolitanism and (d) common heritage of humankind, both of which invite individuals and collectives, namely social movements, to join and struggle across borders for common ideals, for the preservation of local natural and cultural wealth, for different forms of development, for epistemologies made invisible, for insurgent cosmopolitanism and for an ecology of knowledges, where languages are included. The concept of “glocal languages”, while moving beyond the “localised globalism”, by undertaking a “decolonial” turn and by earning voice, therefore, implies claiming for the ownership and the revivification of colonial and colonised languages, that is, all of these implied in the different colonial matrices. The so-called Indigenous languages cease to be perceived as residuals of the past whose life and learning is circumscribed to the limits of their condition as endangered species to give

Introduction

7

way to general awareness of the linguistic, cultural epistemological and ontological value they have in providing us with different world visions (Cavalcanti and Maher 2017). The originally European languages must be regarded as they are in fact, having gained their autonomy and as standing on their own, of which Brazilian Portuguese is a good example. Considering them otherwise, still depending on their European roots and keeping them symbolically connected with colonialism, is a postcolonial perception of “glocalisations” whereas regarding them as linguae francae is putting them at the service of a neo-colonialist understanding of the globalisation of languages. Neither does keeping languages imprisoned in their past, at one end, nor throwing out that burden as if it had never existed, at the other end, offer a consistent solution. The notion of “glocal languages” rests on an understanding of “multiscalar” criss-criss-crossing globalisations and localisations, in the plural, but nevertheless historically constituted, economically driven, hierarchically placed and in constant turbulence (Blommaert 2015; Canagarajah 2013; Mignolo 2000; Sousa Santos 2006, 2014). In agreement with Pietikäinen and Kelly-Holmes, who argue that “the peripheries are rarely examined in terms of their contribution to globalization; instead, they are often seen to follow rather than lead” (2013: 5), we have proposed the idea of “glocal languages” that avoid the “interest in global and local languages (echoing the centre-periphery distinction)” (ibid.: 6) and promote a critical dialogue and interplay between both on equity and reciprocity terms. Based on Bakhtin’s (1981) tenets on the uninterrupted dynamics of heteroglossia, “glocal languages” profit from the centrifugal and centripetal forces of any living language, while taking meaning and shape at any moment in history in articulation with other moments of history and in simultaneity with other developments elsewhere. The perception of the “glocal languages” conceptual framework as enunciated above, and in the following chapters, for the purpose of language education, inserts it within the scope of a decolonial intercultural approach to a critical pedagogy of language and culture. We propose a critical pedagogy in language education, following Paulo Freire’s proposals, in that it addresses the political, social and cultural role of education and enhances the validation of the “glocal” critical and creative contribution of teachers and students to knowledge production (Guilherme 2002, 2012a, 2012b, 2015; Phipps and Guilherme 2004; Souza 2011). According to Torres, a connoisseur and a compagnon de route of Freire’s work and life, Freire was also knowledgeable and influenced by the Critical Theory put forward by the Frankfurt School (Morrow and Torres 2002; Torres 2014). However, he made a “decolonial turn” and presented his own theory on Critical Pedagogy clearly rooted in the geographical and metaphorical Souths (Freire 1970, 1974, etc.) that continues to inspire critical pedagogues all over the world.

8

Guilherme and Menezes de Sousa

In addition, the expression “critical intercultural awareness” put forward in the title of the book draws on Byram’s proposal of “critical cultural awareness” proposed as an axiom, a fifth savoir–savoir s’engager, of his theory about Intercultural Communicative Competence (1997, 2008) that he later developed as a foundational tenet for citizenship education in language education, simultaneously with his work as a consultant for the Council of Europe. Guilherme (2002), following this line, expanded the theoretical grounds of this concept “critical cultural awareness” around Freire’s works on Critical Pedagogy, also drawing on Frankfurt School’s Critical Theory and Postmodernism. Byram and Guilherme’s approaches differ, from the beginning, in that, Byram’s approach to critical awareness in language and culture education may be characterised as pragmatist, of a Deweyan kind, while Guilherme’s was inspired by the utopian work of Paulo Freire. Guilherme adopts Freire’s sense of utopia, understood as the accomplishment of the “inédito viável” that she translates into the “[viable unknown] the ‘not yet’ that is still deemed feasible” (Guilherme 2017a: 431, 2017b, 2018). Not that Byram’s and Guilherme’s proposals ultimately separate in what concerns their perceptions of active democracy or intercultural dialogue, but that their conceptions of the cultural and the intercultural clearly emanate from different cultural matrices, and therefore intercultural matrices. Byram’s focus emerges from the North, both the metaphorical and geographical one, and Guilherme’s unfolds from the South, while both meet up in a situation of “intercultural translation” and “diatopical hermeneutics” as defined by Sousa Santos (2014), that is, in open positions that understand, respect and complement each other. The term “critical intercultural awareness” is adopted here, instead of “critical cultural awareness”, precisely because this book intends to provide the readers with illustrations of a decolonial “intercultural translation” across and within language(s) and culture(s) rather than simply rely on the (post)colonial meeting of languages and cultures. Accordingly, we intend to respond to a particular cultural fabric of miscegenation that is evident in the Brazilian society and that responds to its specific colonial history. Our emphasis on a critical view of interculturality (a word that, in our work, only attempts to translate into English the term interculturalidad(e), both in Spanish and Portuguese) is also heavily grounded on the vast work by Walsh (2007, for example) throughout her long experience in Ecuador, about her concept “interculturalidad crítica” which is endorsed by the editors of this book. Finally, the chapters below, while dissecting the globalness and localness of Indigenous, Portuguese, Spanish and English languages, respond, in our view, to a critical and intercultural decolonial approach that is nowadays evident, both in the geographical and metaphorical South and North, in order to develop a “critical intercultural awareness” of language education that engages and commits teachers and students to issues of local, national and global citizenship. This book comprises a Section I

Introduction

9

entitled “‘Glocal languages’—theoretical background” that aims to provide a theoretical background for the concept of “glocal languages” and within which we find Chapter 1 by Lynn Mario T. Menezes de Souza, entitled “Glocal Languages, Coloniality and Globalization From Below”. This first chapter offers a critical analysis of the processes of globalisation and glocalisation as a background for the perception of the concept “glocal languages”. The author undertakes a critical analysis of hegemonic and non-hegemonic globalisation and of the various theories supporting the conceptual framework for the notion of glocalisation. This chapter then contextualises the idea of “glocal languages” in relation to other theories that have been developed in order to examine language within the perspective of world communication nowadays. Finally, this chapter presents a perspective from the South on linguistic issues, with its locus of enunciation in Brazil. Chapter 2, by Manuela Guilherme, is entitled “Glocal Languages Beyond Post-Colonialism: The Metaphorical North and South in the Geographical North and South” and starts with an introductory reflection on the role of language in the constitution of nationality and in the imposition of coloniality. It then proceeds by describing the state of the art in the different perceptions of the linguistic landscape of today, through the analysis of other contemporary theories that contextualise, support or demand a different conceptual framework such as that of “glocal languages” that is developed in this book. This chapter clarifies the use of the North-South metaphor in relation to the geographical north and south and also critically addresses concepts of synchronicity and diachronicity. It discusses various aspects of multilingualism and conceptions of language and diversity as put forward by other authors. Finally, the author summarises the results of a study of language education (Portuguese, first and foreign language, English and Spanish as foreign or additional languages and Indigenous languages as additional languages) in three public universities in Brazil. Section II addresses Indigenous languages as “glocal languages”. It starts with Chapter 3 entitled “Glocalism Now and Then: The Decolonial Turn of Guarani, Portuguese and Spanish” by Fernanda Martins Felix. This chapter deals with an “Intercomprehension” project between Guarani, Portuguese and Spanish which is perceived as a response to the decolonial turn movement. The author provides us with the historical background of the three languages and offers theoretical support for the decolonial approach undertaken by this project carried out in collaboration with European partners within the scope of Galanet and MIRIADi projects. Finally, it discusses some of their activities. Section II also includes Chapter 4, by Jamille Pinheiro Dias, entitled “Reshuffling Conceptual Cards: What Counts as Language in Lowland Indigenous South America”, that discusses communicative practices and potential notions of language in some Indigenous communities. Drawing mainly on applied linguistics and South American Indigenous ethnology, it calls

10

Guilherme and Menezes de Sousa

for a broader ontological and epistemological understanding of what counts as “a language”. Section III deals with Portuguese as “glocal language” and includes Chapter 5 entitled “The Imaginary in Portuguese Language Perceptions in Academia: (Mis)directions between the Local and the Global” by Gesualda dos Santos Rasia that questions identity ties formed in a language that increasingly inhabits the interstices between the local and the global. Its theoretical background builds upon discourse studies with a materialistic approach by Pêcheaux put in relation to Bourdieu’s theories in sociolinguistics which establish the ground for the analysis of the data collected through a study with some students of Brazilian Portuguese, both native and non-native. The title of Chapter 6 is “The Linguistic Atlas of Brazil Project: Contributions towards Knowledge, Teaching and Disclosure of Brazilian Portuguese” by Marcela Moura Torres Paim and Silvana Soares Costa Ribeiro. It describes the nationwide project ALIB—the Linguistic Atlas of Brazil that focuses on the immense variety of Brazilian Portuguese across its territories. It provides some examples of the data collected which give an impressive dimension of a “glocal language” within national borders. Section IV includes two chapters on Spanish as “glocal language”. It starts with Chapter 7, by Adrián Pablo Fanjul with the title “Comparisons between Spanish and Portuguese: Proposals for University Teaching” that deals with comparative research between South American Spanish and Brazilian Portuguese and its use in classroom activities. These courses have put together comparison topics which can be sorted into three main categories: the political delimitation of languages, discursivity and linguistic dynamics in its systemic aspects. Chapter 8, by Maria Josele Bucco Coelho, is entitled “Multiculturalism and Glocal Languages: The Impact of Cultural Mobility in Spanish Teaching and Learning in Southern Brazil” and deals with teaching of Spanish from the point of view of cultural mobility and multiculturalism. It aims to describe teaching strategies that better represent the plurality of cultural communities in Hispanic America by examining the curriculum proposal and teaching practices developed at the undergraduate level. The author considers this challenge as a local epistemic effort to let the experiences lived and suffered in Latin America shine through. Section V deals with English as “glocal language” and begins with Chapter 9, by Daniel de Mello Ferraz, entitled “English (Mis)education as an Alternative to Challenge English Hegemony: A Geopolitical Debate”. This chapter places critical perspectives on glob(c)alisation, (glocal) languages and language education in Brazil. It presents a brief state-of-the-art of globalisation and its connections to glocal relations and language education and discusses some perspectives regarding language education in Brazil through curricular analysis and, finally, through the voices of some pre-service teachers. Chapter 9, by Alessandra Coutinho

Introduction

11

Fernandes, with the title “Teaching English to Undergraduate Students in a Brazilian University: Thinking Glocally” undertakes an autobiographic and critical reflection upon the mythical “native speaker” of English and its role in Brazilian higher education, to follow with a critical analysis of the concept of globalisation as viewed from Brazil and its relation with the teaching and learning of English as it has been theorised by scholars from the geographical north and Brazilian ones as well. She finally adopts a “glocal” perspective and makes some proposals based on her own experience in English teaching at the Federal University of Paraná. All the three public universities, among the largest in Brazil, participating in the Glocademics Marie Sklodowska-Curie project (Universidade de S. Paulo, Universidade Federal da Bahia and Universidade Federal do Paraná) are represented in this book. The Introduction and Conclusion, by the Editors, aim to contextualise, both politically and theoretically as well as experientially, the contents of this book and wrap up the messages found more relevant.

Note 1. Universidade de S. Paulo, Universidade Federal da Bahia and Universidade Federal do Paraná.

References Asuncion Treaty Mercosur (1991) http://www.sice.oas.org/trade/mrcsr/mrcsrtoc. asp Bakhtin, M. M. (1981) The Dialogic Imagination: Four Essays by M. M. Bakhtin. Edited by M. Holquist and Translated by C. Emerson and M. Holquist. Austin: University of Texas Press. Blommaert, J. (2015) Chronotopes, scales and complexity: In the study of language in society. Annual Review of Anthropology, 44, 105–116. Byram, M. (1997) Teaching and Assessing Intercultural Communicative Competence. Clevedon: Multilingual Matters. Byram, M. (2008) From Foreign Language Education to Education for Intercultural Citizenship. Clevedon: Multilingual Matters. Canagarajah, S. (2013) Agency and power in intercultural communication: Negotiating English in translocal spaces. Language and Intercultural Communication, 13, 2, 202–204. Castro-Gómez, S. (2005) La Hybris del Punto Cero: Ciencia, raza e ilustración en la nueva granada (1750–1816). Bogotá: Pontificia Universidad Javeriana. Castro-Gómez, S. and Grosfoguel, R. (eds.) (2007) El Giro Decolonial: Reflexiones para una diversidade epistémica más allá del capitalismo global. Bogotá: Siglo del Hombre Editores. Cavalcanti, M. C. and Maher, T. M. (eds.) (2017) Multilingual Brazil: Language Resources, Identities and Ideologies in a Globalized World. New York: Routledge. Estermann, J. (2008) Si el Sur fuera el Norte: Chakanas interculturales entre Andes y Occidente. La Paz: Instituto Superior ecuménico Andino de Teologia (ISEAT).

12

Guilherme and Menezes de Sousa

Freire, J. R. B. (2014) A demarcação das línguas indígenas no Brasil. In M. C. Cunha and P. N. Cesarino (eds.) Políticas Culturais e Povos Indígenas (pp. 363–389). São Paulo: Editora UNESP. Freire, P. (1970) Pedagogy of the Oppressed. New York: Continuum. Freire, P. (1974) Education for Critical Consciousness. London: Sheed and Ward. Guilherme, M. (2002) Critical Citizens for an Intercultural World: Foreign Language Education as Cultural Politics. Clevedon: Multilingual Matters. Guilherme, M. (2012a) Critical language and intercultural communication pedagogy. In J. Jackson (ed.) The Routledge Handbook of Intercultural Communication (pp. 357–371). London: Routledge. Guilherme, M. (2012b) Critical pedagogy. In C. A. Chapelle (ed.) The Encyclopaedia of Applied Linguistics. Oxford: Blackwell Publ. Guilherme, M. (2015) Critical pedagogy. In J. M. Bennett (ed.) Encyclopaedia of Intercultural Competence (pp. 138–142). Thousand Oaks, CA: Sage. Guilherme, M. (2017a) Freire’s philosophical contribution for a theory of intercultural ethics: A deductive analysis of his work. Journal of Moral Education, 46, 4, 422–434. Guilherme, M. (2017b) Visões de futuro em Freire e Dewey: Perspectivas interculturais das matrizes (pós)coloniais das Américas. ECCOS, 44, 205–223. Guilherme, M. (2018) O diálogo intercultural entre Freire & Dewey: O Sul e o Norte nas matrizes (pós)coloniais das Américas. Educação e Sociedade, 39, 142, 89–105. Holanda, S. B. (2016) Raízes do Brasil. São Paulo: Companhia das Letras (1st ed., 1936). Kumaravadivelu, B. (2014) The decolonial option in English teaching: Can the subaltern act? TESOL Quarterly, 50, 1, 66–85. Lucchesi, D. (2015) Língua e Sociedade Partidas: A polarização sociolinguística do Brasil. São Paulo: Contexto. Mignolo, W. D. (2000) Local Histories/Global Designs: Coloniality, Subaltern Knowledges, and Border Thinking. Princeton, NJ: Princeton University Press. Mignolo, W. D. (2007) El pensamiento decolonial: Desprendimento y apertura (Un manifesto). In S. Castro-Gómez and R. Grosfoguel (eds.) El Giro Decolonial: Reflexiones para una diversidade epistémica más allá del capitalismo global (pp. 25–46). Bogotá: Siglo del Hombre Editores. Mignolo, W. D. (2009) Epistemic disobedience, independent thought and decolonial freedom. Theory, Culture & Society, 26, 7–8, 1–23. Morrow, R. A. and Torres, C. A. (2002) Reading Freire and Habermas: Critical Pedagogy and Transformative Social Change. New York: Teachers College Press. Phipps, A. and Guilherme, M. (eds.) (2004) Critical Pedagogy: Political Approaches to Language and Intercultural Communication. Clevedon: Multilingual Matters. Pietikäinen, S. and Kelly-Holmes, H. (2013) Multilingualism and the periphery. In Multilingualism and the Periphery (pp. 1–16). Oxford: Oxford University Press. Ribeiro, D. (1995) O Povo Brasileiro: A formação e o sentido do Brasil. São Paulo: Companhia de Bolso. Robertson, R. (1995) Glocalization: Time-Space and Homogeneity-Heterogeneity. In M. Featherstone, S. Lash and R. Robertson (eds.) Global Modernities (pp. 25–44). London: Sage. Santos, M. (2000) Por uma outra globalização: Do pensamento único à consciência universal. Rio de Janeiro: Editora Record (25th ed., 2015).

Introduction

13

Sousa Santos, B. (1999) Towards a multicultural conception of human rights. In M. Featherstone (ed.) Spaces of Culture: City, Nation, World (pp. 214– 229). London: Sage. Sousa Santos, B. (2006) Globalizations. Theory, Culture & Society, 23, 2–3, 393–399. Sousa Santos, B. (2007) Beyond abyssal thinking: From global lines to ecologies of knowledges. Review, 30, 1, 45–89. Sousa Santos, B. (2014) The Epistemologies of the South. Boulder: Paradigm. Souza, L. M. (2011) Para uma Redefinição de Letramento Crítico. In R. F. Maciel and V. A. Araújo (eds.) Formação de Professores de Línguas: Ampliando perspectivas (pp. 128–140). Jundiaí, São Paulo: Paco Editorial. Spivak, G. C. (1999) A Critique of Postcolonial Reason. Cambridge, MA: Harvard University Press. Torres, C. A. (2014) First Freire: Early Writings in Social Justice Education. New York: Teachers College. Urry, J. (2005) The complexities of the global. Theory, Culture & Society, 22, 5, 235–254. Walsh, C. (2007) Shifting the geopolitics of critical knowledge. Cultural Studies, 21, 2–3, 224–239.

Section I

Glocal Languages— Theoretical Background

1

Glocal Languages, Coloniality and Globalization From Below Lynn Mario T. Menezes de Souza

There is strictly speaking no single process called globalization; there are, rather, globalizations; bundles of social relations that involve conflicts and hence both winners and losers. More often than not, the discourse of globalization is the story of the winners as told by the winners. The victory appears so absolute that the defeated end up vanishing from the picture altogether. Boaventura de Sousa Santos (2002a) Our capacity to see is poor because our apparatus of knowledgeproduction is poor. We are tied up with our European inheritance and more recently, with an American empire, in the thinking of our Social Sciences. Milton Santos (2007)

In the context of current language scholarship, reflections on the relationship between language practices, policies and language ideologies are not new. It seems, however, that the ideologies within which these studies are framed need to be the object of more critical focus. As Joseph (2006) reminded us, language is a political construction. Makoni and Pennycook (2007) highlighted the inequality of power dimension by adding that language classification is a political construct instrumental for the control of variety and difference. Previously, Blommaert and Verschueren (1992) had also pointed out the controlling role of language ideologies in Europe but saw them as emerging from and pertaining to nineteenth century European nationalism. Makoni and Pennycook (2007) see both language and the metalanguages that conceive and categorize it as political constructs. As such, languages are inseparable from metadiscursive regimes that are not only representations of language but also social-institutional instances that produce knowledge about and control language. Makoni and Pennycook point to the need to not only see language and its study as inherently implicated in ideology and politics (local, national, colonial,

18

Lynn Mario T. Menezes de Souza

or Eurocentric) but to also understand the interrelationships that prevail in order to seek ways to reconstitute the constraining consequences of these interrelationships. More recently, Flores (2013) and Kubota (2014) discussed the interconnections between language ideologies and global neo-liberal ideology. Canagarajah (2017), building on these interconnections, reinforces the need to consider language ideologies more deeply rather than persist in the current focus in applied linguistics to describe and analyze strategies deployed by users in multilingual practices. Stroud (2015), also building on the interconnections between language, ideology and politics and the need for a reconstruction of these interconnections, introduces the concept of linguistic citizenship as “an invitation to rethink our understanding of language through the lens of citizenship at the same time that we rethink understandings of citizenship through the lens of language” (24). Given these interconnections between language and ideology, this chapter discusses the concept of glocal language, glocalization and globalization from a southern perspective in the context of decoloniality. These terms will be explained below.

Hegemonic and Non-Hegemonic Globalization Speaking epistemically from the margins of the hegemonic North, Santos (2002a), in an effort to examine emancipatory possibilities away from the negative effects of globalization and its accompanying cultural influences, proposes an analytical framework to apprehend globalization from a non-hegemonic, southern perspective. Santos prefers to see globalization as plural and sensitive to social political and cultural factors and always entailing localization. He differentiates between hegemonic and non-hegemonic forms of globalization. Whereas hegemonic globalization builds on and maintains established hierarchies and functions through an impetus for regulation, non-hegemonic globalization seeks horizontal collaboration and solidarity. When neo-liberal globalization and its penchant for de-regulation1 becomes hegemonic, de-regulation acquires a normative stance; by forcibly demanding de-regulation it maintains a hegemonic position and functions as a variant of regulation (the rule is to de-regulate). For Santos, there are two interconnected manifestations of hegemonic globalization: First, there is the successful globalization of a localism. The point here is that there is nothing that is transnationally globalized that does not originate locally somewhere; i.e. there is nothing that is not always already embedded in some specific culture. Given that all knowledge needs to be produced by someone, somewhere, there is, in social terms, nothing that is not already a localism. Through hegemonic expansion, certain localisms are globalized and acquire the aura of being universal. One consequence of the hegemony acquired by these globalized

Glocal Languages, Coloniality, Globalization

19

localisms is their capacity, not only to deem themselves universal, but also to deem all other localisms as merely and insignificantly local. For Santos (2010a), such is the case of what is considered to be scientific knowledge. As a monoculture of knowledge, modern science has ‘forgotten’ its local origins and development in Western Europe, and its implications in European philosophy, history and politics, and transformed itself into a global, universal yardstick that does not accept as science or as knowledge the multiple knowledges produced outside Europe. It is our contention in this chapter that hegemonic ‘language ideologies’ considered to be placeless, and hence ‘universal,’ may be following the exemplar of modern science. The second manifestation of hegemonic globalization, according to Santos, is localized globalism; this refers to the imposition on particular localities of elements originating in the hegemonic transnational ‘global.’ Given their hegemonic force, such elements cannot be easily resisted but can be recontextualized to suit the conditions and interests of the local. Santos exemplifies this with the case of North American cinema whose stars are seen as transnationally global whereas other national cinemas and their stars are seen as merely local. He also cites the case of the global spread of English becoming a lingua franca in several local contexts. Both forms of hegemonic globalization maintain their hegemony by establishing an abyssal line, which separates what is considered to be of value on this (hegemonic) side of the line and consequently produces the invisibility of whatever is located on the other side of the line. Modern science, for Santos, is a prime example of the abyssal line (see below for further discussion). Santos exemplifies non-hegemonic globalization with counter-hegemonic social movements and other forms of grassroots global exchanges; in these, collaboration and solidarity set the tone. Santos calls this subaltern cosmopolitanism. Given that established hierarchies of globalized power are unable to exclude the possibility of contact, exchange and collaboration between local (un-globalized) groups, it is this sphere of translocal exchange and solidarity that Santos defines as subaltern cosmopolitanism. Unlike liberal elite cosmopolitanism that emphasizes the individual and aims at the acquisition of social and cultural superiority through the accumulation of symbolic capital, subaltern cosmopolitanism aims at the creation of south-south dialogues and networks for mutual benefit and mutual emancipation. Two features that highlight the horizontal nature that characterizes such networks and dialogues are an ecology of knowledges and the importance of translation. As an effort to deconstruct the hegemonic abyssal line that purports to separate the ‘global’ from the ‘local,’ the concept of an ecology of knowledges is based on the presupposition that no knowledge is total, complete and capable of everything. As Santos (2007) says, knowledge and ignorance are interconnected; a community’s

20

Lynn Mario T. Menezes de Souza

knowledge of something implies an ignorance of other forms of knowing; thus, this same community’s ignorance does not invalidate the fact that it also knows. The concept of translation comes into the picture in the wake of the same reasoning: The need for translation implies the need to know what one does not know. Translation, like the knowledges interconnected horizontally in an ecology, rather than signifying a total transference of meaning, implies incompleteness and ignorance and the need to overcome both; translation refers also to the fact that overcoming both of these in order to attain the desire of completeness is beyond realization. However, these difficulties in translation do not indicate incommensurability; they do indicate the need for constant exchange and for the persistence in the continuous work of translation. These characteristics of a translocal ecology of knowledges and the unfinished work of translation add depth to Robertson’s (1995, 2015) and Pieterse’s (1995) descriptions (see below) of the complexity of glocalization, beyond the standard reductive dichotomies of local-global and homogeneous-heterogeneous. However, it is essential to note that unlike Pieterse and Robertson, Santos is self-consciously aware of his epistemic location in the global south. As we shall see below, the concepts of subaltern cosmopolitanism and the ecology of knowledge are relevant for a consideration of glocal languages from a southern perspective.

Glocalization As we have seen, globalization, according to Santos (2002a), is necessarily also about localization, not only in the sense of the global forcefully affecting non-global locations, but also in how it involves the expansion from one local to other locals; it also involves, where necessary, the forceful returning of its competitors-in-the-process-of-expansion to their original locations. Globalization is then about the play of asymmetrical relations. Glocalization, as we shall see, is about how all these actions of localization play out at relatively non-hegemonic levels in the process of hegemonic globalization. As my first epigraph says, a large part of the discourse of globalization is the story of the victors as told by them. If this is to change, it needs to go beyond substituting one story (the victor’s) for another (the loser’s). It needs to go beyond telling the story from a single perspective. Thus, besides troubling the singularity of the story, it is also necessary to “change the terms of the conversation” (Mignolo 2007) in order to ‘re-localize the global,’ mark the unmarked and in short, ‘provincialize’ the apparently universal. In this process, the unmasking of one’s locus of enunciation or that of other theorists and critics becomes increasingly significant.

Glocal Languages, Coloniality, Globalization

21

Marking the unmarked and changing the hegemonic terms of the conversation is the strategy of epistemic reconstitution that various Latin American thinkers propose under the terms decoloniality and coloniality. It is a strategy of response to the singularity of perspective and to the predicament that the Brazilian social scientist Milton Santos (2007) describes in my second epigraph above. Grosfoguel (2013) calls this predicament epistemic racism;2 Santos (2014) calls it epistemicide.3 Given the significance of power relations and their unequal distribution in globalization, the manner in which globalization and glocalization are understood is inseparable from the locus of enunciation of the theorist or critic. This concurs with Santos’ concept, mentioned above, that one’s embedding in a specific local is inescapable. Similarly, in theorizing about language ideologies, whether to identify them, describe the practices and strategies through which they are manifested, or to undo their effects and transform them, one’s locus of enunciation is crucial. It is in general what will situate the theorist on this side or the other of the abyssal line of modern science; it is what will deem certain knowledges as ideology and others as scientific fact. For two theorists of glocalization, Robertson (1995) and Pieterse (1995), for whom identifying their loci of enunciation is not significant, globalization is portrayed as a dynamic process more complex than that of a unified and unifying homogenizing force. They describe it as involving varying degrees of contact between the so-called global and the so-called local, including moments of convergence and divergence, interaction, interconnectedness and resistance. These complex moments of contact and their consequences are defined as glocalization. Reading Pieterse and Robertson critically, Roudometof (2015) contrasts on the one hand Pieterse’s conception of glocalization as an integrative process of degrees of homo- and heterogeneity convergently integrating the global and the local in ever-changing patterns with, on the other hand, Robertson’s conception of glocalization as not only creating new, homogeneous units (albeit constructed from heterogeneous elements) but also initiating non-convergent processes of fragmentation. Robertson, in a later discussion (2015), now in a post 9/11 frame, calls for a critical global consciousness and claims that a focus on globalization as mere interconnectedness has helped to hinder this consciousness. According to Robertson, such a global consciousness is of dire importance in light of some of the deleterious effects of glocalization. An effect that he cites is related to one possible process of glocalization as non-convergent fragmentation, often called indigenization; here, as in the localized globalism of Santos, elements absorbed from transnational global origins are integrated into local systems. However, their integration is forgotten and the local systems are seen as remaining unchanged, untouched by external influence and authentically local. As a conservative reaction of resistance to exogenous intrusion from the global, indigenization is seen

22

Lynn Mario T. Menezes de Souza

to propitiate cultural and religious fundamentalisms and foster violence and unrest. For Robertson, this kind of reaction highlights the critical importance of calling attention to the processes of recontextualization and resignification in glocalization, lest they be forgotten. In general then, for both Pieterse and Robertson the process of glocalization involves external (‘global’) elements coming into contact with preexisting local elements; these external elements, originating in different and distant contexts, are recontextualized as they are accommodated into local cultures, knowledges or beliefs. Glocalization necessarily involves the re-recontextualization and resignification of both external and local elements. This is not a simple top-down phenomenon in which the external global element remains intact globally and is only recontextualized and resignified locally; it is not simply a process of an external element imposing homogeneity and eliminating pre-existent heterogeneity or a reverse process of unleashed hybridization. As we shall see below in the case of perspective and locus of enunciation, in the construction of knowledge, as Santos reminded us above, the ‘global’ is always someone’s ‘local.’ In relation to the case of fundamentalism cited by Robertson (2015), though indigenization might, in practice, involve a large degree of recontextualization and resignification of extraneous elements absorbed into a local culture, in order to become ‘fundamentalist’ it must involve a simultaneous denial and rejection of such extraneous influence and adaptation. This denial and rejection are often a posteriori and mask the already occurred resignification of the new elements (now recontextualized and resignified as ‘old’ or pre-existing). In light of this, Robertson proposes what he calls a critical perception and understanding of the processes of glocalization—recontextualization and resignification—as possible correctives not only to prevent fundamentalism in such cases but also to minimize the homogenizing force of globalization in cases that don’t involve fundamentalism but where there is little local resistance to the extraneous elements of the global. Roudometof (2015) calls attention to the fact that in much of globalization theory, specific considerations of glocalization remain largely silent and are rarely engaged with. On many occasions glocalization is dismissed as being simplistic and dichotomous (Pennycook & Otsuji 2013), amalgamating an allegedly simple homogeneous ‘global’ with an allegedly simple homogeneous ‘local.’ In the work of the theorists of glocalization mentioned above, it is clear that though they alert one to the complexity of globalization, they do not distinguish between hegemonic and non-hegemonic forms of globalization. They seem to be analyzing globalization from the perspective of ‘the winners’ of the global north and not from its margins. Their concept of fundamentalism and the need for a critical approach to glocalization hence become doubly flawed. First, what they identify as fundamentalism may be a legitimate (as seen from a counter-hegemonic perspective)

Glocal Languages, Coloniality, Globalization

23

attempt to resist hegemonic globalization from a non-hegemonic locus. Second, the categories of analysis they propose—recontextualization and resignification—do not distinguish the complexity of epistemological perspectives involved and seem to analyze the phenomena from the purportedly neutral perspective of modern science on the hegemonic side of the abyssal line. ‘Fundamentalism’ or unquestioned truth-value could easily be attributed to such theorizing that delights in calling attention to the ignorance of others but not to its own ignorance. In Santos’ terms, rather than simply focusing on global-local contacts in terms of convergence or fragmentation, glocalization could benefit from a more complex analysis of globalization involving notions of emancipation and resistance, a multiplicity of knowledges and an ongoing process of translation. The work of troubling the singularity of the narrative, changing the terms of the conversation, re-localizing the global, marking the unmarked and in short, provincializing the apparently universal must go on.

Glocal Languages Theorizing language practices from the perspective of Santos’ theories of hegemonic and non-hegemonic globalization, and situating such practices in the context of glocalization, Guilherme (2014, 2018) defines ‘glocal languages’ as products of processes related to Santos’ second type of hegemonic globalization—localized globalisms. Like Santos in relation to globalization, Guilherme sees glocal languages as complex and promising phenomena going beyond simple local-global and homogenizinghybridizing dichotomies. They are complex because they involve active counter-hegemonic resistance to hegemonic globalized languages; they are promising because they do not simply reject but also transform global languages according to their own, local necessities. These transformations thus occur at the interface between the non-hegemonic local languages and the globalized hegemonic languages. Glocal languages tend to be more commonly associated with hegemonic European languages now used in practices unforeseen in their hegemonic ‘original’ variants. However, non-European non-hegemonic languages may potentially also become glocal if transformed in contact with hegemonic localized-global languages. Guilherme calls attention here to the complex role of scale in such contacts: ‘global’ and ‘local’ languages may also exist and interrelate at the level of region or nation. An important aspect of glocal languages is the role that local (nonhegemonically global) agency plays in their construction. In this sense, by ‘answering back,’ glocal languages represent a form of non-hegemonic resistance and blur the lines that separate Santos’ terminology of hegemonic globalization and non-hegemonic globalization, globalized localism and subaltern cosmopolitanism.

24

Lynn Mario T. Menezes de Souza

In Brazil an example of how ‘the’ global language, English, has become a glocal language is how it is portrayed in local official documents (Brasil 2006)—the Orientações Curriculares do Ensino Médio (OCEM), a proposed national curriculum for the teaching of English as a Foreign Language at Secondary Level in the public school system. The OCEMs recontextualize and resignify English as a global language by making a distinction between the teaching of English as a subject in the school system and the teaching of English as a language for native-like competence in private language institutes. Whereas, in the latter case, it is language competence that is the objective of teaching, in the former case, the objective is educational and does not include the aim of achieving language competence. The educational objective of English as a foreign language is stipulated as that of a means towards understanding and respecting difference. By appreciating how speakers of the foreign language use, for example, their bodily apparatus differently to produce different sounds, different intonations and different body languages; by appreciating how the foreign language organizes time in verb tenses differently to the mother tongue (Portuguese); by appreciating how written language is organized differently in English and in Portuguese, it is hoped that students will appreciate and respect differences in expression not only of foreign-speakers of English, but also of speakers of any language, including the mother tongue. Thus, from a global language with a globalized demand for native-speaker proficiency, the English language in the Brazilian school system is transformed into a subject that attends more adequately to local educational demands. Though in both cases teachers may be adamant that they teach English, and students may insist that they learn English, in fact, as a glocal language, it is used in the official school system more as a pedagogical device than as a natural language. In both cases the reason for this is the hegemonic and global nature of English in Brazil in relation to and in contact with the national language—Portuguese. In terms of Santos’ classification of globalization, we see in this example a localized globalism being transformed through local agency to attend to local demands. As a glocal language, English in this case serves the need for creating new knowledge, of greater local value—knowledge that can propitiate a respect for difference—rather than a reproduction, in natura, of a globalized localism that the English language effectively is in relation to Brazil. As a glocal language and through localized globalism it is ‘translated’ and, generating new knowledge through contact, it contributes to the ecology of knowledges characteristic of non-hegemonic globalization and characteristic of subaltern cosmopolitanism. A second example of glocal languages in Brazil is the case of the interface between indigenous languages and Portuguese among indigenous communities in the Western Amazon in Brazil. Maher (1998, 2018) points towards the inseparability of the indigenous languages from Portuguese

Glocal Languages, Coloniality, Globalization

25

in these communities and calls attention especially to the identity issues that emerge. She sees both the indigenous languages and Portuguese in these communities as emergent, as are the identities of these communities. Though their proficiency in Portuguese may not be that of an urban non-indigenous speaker, these communities, whether they maintain their indigenous mother tongues or not, see no clear separability between the languages. Those who have maintained their indigenous mother tongues recognize the importance of using Portuguese in official contacts with government officials, even though their proficiency in Portuguese may be limited. This perception of the necessity of Portuguese for their survival leads to their recontextualizing and resignifying of the language by inhabiting it as theirs, to be used as they wish and when necessary, without being limited to or hampered by non-indigenous ‘standard’ language norms. Furthermore, according to Maher, those who have lost their mother tongues but symbolically still identify with them see themselves as fully fledged members of an indigenous community with an indigenous language, even if they are not fluent in it. In these cases, both the ‘global’ national language, Portuguese, and the local indigenous languages are transformed in contact with each other. By claiming ownership of the indigenous mother tongue even when it is no longer spoken, their use of Portuguese has undergone ideological transformation: Even if they are de facto monolingual speakers of Portuguese, it is neither their mother tongue, nor are they ideologically monolingual. At the scale of the nation Portuguese as a localized globalism has been recontextualized and resignified, and the force of local indigenous agency marks this transformation as counter-hegemonic. As was mentioned above, Santos (2002a) cites the example of English as a localized global phenomenon of hegemonic globalization. As a hegemonic global language, English undergoes resignification and recontextualization in non-hegemonic locations when it is used between speakers for whom it is not a first language. We have seen in the example above how this occurs in Brazil. Portuguese in Brazil is another example of a localized global phenomenon not only because it is, like English, a former language of foreign colonialism, but also because in relation to indigenous communities within Brazil, as the national language of Brazil, it moves from being a former colonial globalized localism to function in the context of contact between indigenous communities and the Brazilian state, as a localized globalism. Portuguese here suffers significant recontextualization and resignification in these various contexts to the point of undergoing syntactic, morphological and pragmatic changes, beyond the more commonly expected phonological changes (see Chapter 6, this volume). The indigenous languages also undergo change through this contact, albeit at the pragmatic level of use and at the ontological level of coming into being as written languages.

26

Lynn Mario T. Menezes de Souza

Although much of current research in Applied Linguistics on languages in contact resulting from globalization are conducted under the labels of superdiversity (Blommaert & Rampton 2011), metrolingualism (Pennycook & Otsuji 2015), translanguaging (Canagarajah 2011; Garcia & Wei 2015), polylanguaging (Møller 2008) or, more recently and more descriptively, language practices in combinatorial spaces (Arnaut et al. 2017), most of these remain silent to the term glocalization or glocal languages, and few are aware of their loci of enunciation, and therefore of their role in constituting or maintaining hegemonic language ideologies. Many of these theorists of language seem to agree that in the context of globalization, language study has moved away from a focus on languages as structured, independent units, linked to specific cultures and nations, to a focus on language seen as a set of resources of linguistic and semiotic features that can be assembled and reassembled into registers or repertoires. In terms of Santos’ terminology, this change of focus should be expected to reflect a change of perspective from that of regulation to that of solidarity or in other words, a move away from the perspective of the rule-maker to that of the user of language. If language is now seen as an open-ended set of linguistic and semiotic resources, one would expect that this change of perspective reflects a change in access to these resources previously controlled by the hegemonic rule-makers of language. However, the fact that only a few theorists of language (Makoni et al., Makoni & Pennycook, Canagarajah, Kubota, Flores, etc.) are selfconsciously aware of their own language ideologies—and hence their loci of enunciation—seems to indicate that a change may have occurred at the level of conceptions of language but perhaps not entirely at the level of metadiscursive regimes; the latter may hide the subjectivity of the theorist from him/herself and hence maintain his/her genealogical attachments and entanglements with hegemonic metadiscursive regimes beyond critique and consequently hinder effective change. In short, whereas access to linguistic and semiotic resources by the user of language may have occurred, access to epistemological resources that theorize language and its ideologies may have not yet occurred. Blommaert (2014) signals the importance of the issue of ideological perspective in relation to glocalization (though he does not use the term) when he once more recognizes the relevance of linguistic ideologies and identifies a paradigm change in language study in contexts of globalization. Besides taking into account the above-mentioned move from a focus on languages to one on language, he sees this change as involving processes of mobility and complexity in language use in spaces of global-local contact: “Taking mobility as a paradigmatic principle of sociolinguistic research dislodges several major assumptions (of mainstream sociolinguistics) and invites a more complex, dynamic, multifaceted view” (8).

Glocal Languages, Coloniality, Globalization

27

In this paradigm change where language is now taken to be a set of resources, mobility seems to refer both to people moving across the globe and also to features of language structure and use—the linguistic and semiotic resources—that undergo change under conditions of globallocal contact. Blommaert further stresses the need to complexify previous notions of context in language analysis under these conditions and the need to take an ethnographic stance in the analysis of language use in superdiversity. This appears to be a methodological preoccupation for analyzing language under these complex conditions and approximates the kind of ideological preoccupation of perspective as imbricated in power and history that characterizes coloniality and is discussed below. Related to Blommaert’s proposal of an ethnographic stance, Arnaut et al. (2017) propose, in their analyses of language use in the combinatorial spaces (where global-local contact occurs, spaces of glocalization) afforded by globalization, the concept of the poiesis-infrastructures nexus; this is described as “the double process of emergent normativities and sedimentations on the one hand and the creative and material production processes unsettling these, on the other hand” (15). The poiesis part of the nexus is clearly related to the processes of contextualization and resignification or to non-hegemonic resistance discussed above. However, the infrastructures part, said to refer to “myriad forms of organisation and institutionalisation” (17) could benefit from more reflection; though akin to Santos’ concept of regulation that characterizes hegemonic action, in the theorization of Arnaut et al., references to metadiscursive regimes, power relations and the institutional context, though present, appear to function as explanatory devices for language use rather than as an object of critique.

Towards a Southern Perspective Apart from significant exceptions, some of which have been already mentioned (Makoni et al. 2003; Makoni and Pennycook 2007, Flores 2013, Kubota 2014, Canagarajah 2017) much of the theorizing on language in globalization, and especially in global-local contacts, continues to take place from a zero-point perspective (Castro-Gomez 2005); such a perspective is one in which the locus of enunciation of the theorist and its genealogical entanglements with hegemonic regimes is seen as irrelevant and where the knowledge thus produced is considered to have universal valence. This seems to indicate that in studies of language in globalization and more specifically in glocalization, a concern with epistemic or cognitive justice as emphasized by Santos, Grosfoguel, Dussel, Quijano and other theorists of coloniality is not a priority. From my perspective, located epistemically in the global south, it is not without irony that I wonder: In which contexts and from whose perspectives are the concepts of local and global defined or dismissed?

28

Lynn Mario T. Menezes de Souza

From whose perspectives are heterogeneity and homogeneity, recontextualization and resignification, said to accompany glocalization, defined or dismissed? If, as Santos says in my epigraph, globalization involves conflicts, winners and losers, on which side are the theorists that portray globalization and glocalization as if from a distant neutral standpoint? As we have seen, even when echoing Blommaert’s concept of a paradigm change in language study and his proposal for greater ethnographic focus, many applied linguists often seem not to have noticed the need to critique metadiscursive regimes. Even when the agency and subjecthood of users is taken into account, it is largely as subjects of local relevance for the study of the data at hand. Their complex socio-historical constitution as subjects entangled in wider historical, ideological, racist, cultural and linguistic issues often tends to be reduced to generalizing notions of globalized mobility. It seems that a globalized epistemic posture, possibly involving the “story of the winners as told by the winners,” may have escaped the critical gaze of many thinkers of globalization and glocalization. In spite of their sensitivity to context, the question as to where they are thinking or speaking from in relation to the rest of the world seems not to be an object of relevance or critique. There are historical reasons for this disembodied, un-self-conscious and universalizing gaze, and they have to do with globalization and global-local contacts; they pertain, however, to a much older phase and locus of globalization. Denouncing the restricted universalizing gaze that is blind to and actively (though not necessarily consciously) engaged in perpetuating its own privilege, Connell (2007 p. 226) critiques what she calls Northern social theory. She claims that, given its hegemonic position, such necessarily located social knowledge refuses to locate or contextualize itself and hence uncritically assumes universal value for its claims. By Northern theory, Connell refers to theory produced in the Eurocentric metropolitan West (or Global North)4 and self-portrayed as mainstream: On close examination, mainstream sociology turns out to be an ethno-sociology of metropolitan society. This is concealed by its language, especially the framing of its theories as universal propositions or universal tools [. . .] its theorising is vitiated whenever it refuses to recognise its ethno-sociological being—[. . .] its situation in the world and its history in the world. Against this universalizing gaze, in this chapter we engage in what Connell (207) describes as dirty theory or theory intimately connected to specific situations: “The goal of dirty theory is not to subsume, but to specify, not to classify from outside but to illuminate a situation in its concreteness.” This involves, not slimming down, but multiplying theoretical ideas and the local sources of one’s thinking.

Glocal Languages, Coloniality, Globalization

29

Theorists like Connell and the previously mentioned Santos, Grosfoguel and Mignolo are critically aware of their locus of enunciation, or the historical and ideological context from where they are speaking and from where their knowledge is being produced. They see themselves as epistemically located on the margins of hegemonic (‘metropolitan,’ ‘northern’, ‘Eurocentric’, etc.) centers of economic, cultural and knowledge production, defined as the South. It is crucial to note that this location need not be geographical or literal; it may be metaphoric, but it is always epistemic. In terms of the processes of recontextualization and resignification as characteristic of contact through globalization and the emergence of glocal languages, the calling of attention to one’s epistemic location on the margins of the supposedly universal hegemonic episteme demands that all language and meaning-making seen as transparent, neutral and universal be recontextualized and resignified. Metadiscursive regimes need to be disclosed, and their roles in hegemonic or non-hegemonic regimes and their affordances for regulation and/or solidarity need to be considered. Grosfoguel (2011), epistemically located in the South, defines fundamentalism as the premise that there is one sole epistemic tradition from which to achieve truth and universality. Fundamentalism may be hegemonic or marginal. Most of the theorizing of globalization that does not critically take into account its own epistemic location and relate it to epistemically different accounts of the phenomenon ends up performing what Santos called the globalization of localisms. According to the definition of Grosfoguel, such theorizing would be hegemonically fundamentalist in nature. Speaking from a distinctly Southern perspective, on the margins of hegemonic globalization and located epistemically in Latin America, Grosfoguel emphasizes the problematic insufficiency of speaking of the south and from the south. He sees neither of these as guarantees that one is speaking epistemically from the south. To do so, Grosfoguel suggests that one speak from and with the south. Such is the perspective pursued in this chapter. Speaking then from a Southern perspective, and avoiding the risk of falling into the trap of either marginal or hegemonic fundamentalism, Grosfoguel and Castro-Gómez (2007) propose the critical strategy of border thinking from the perspective of decoloniality. This does not amount to simply reaffirming the tiresome adages that “knowledge cannot be separated from social context and the ideologies inherent in the latter,” or that, as a consequence of this, “knowledge is always partial in both senses of the word.” The focus of decolonial border-thinking, according to Grosfoguel, involves engaging with and attributing to oneself and to others with whose epistemologies one engages, the crucial significance of the epistemic locus of enunciation. This identification of one’s locus of enunciation is defined by Grosfoguel—and other Latin

30

Lynn Mario T. Menezes de Souza

American thinkers, such as Quijano (2000), Mignolo (2007) and CastroGomez (2005)—as the geopolitical and body-political location of the subject that speaks. For Grosfoguel (2007, 2009, 2011, 2013) in the hegemonic, Cartesian, Western tradition, theoretical discussion tends to focus on what is said and not on who is speaking. This in general holds true as long as the speaker speaks from an unmarked (from the Western perspective5) locus, as white, male, heterosexual, Christian and as a speaker of a hegemonic language. The unmarkedness of the speaker affects the import of what is said, permitting it to be understood in positive and authoritative terms (e.g. true and of universal value). This tradition of presupposing the unmarkedness of the speaker and the resulting focus on what is said characterizes what the thinkers of decoloniality refer to as an ego-politics of knowledge. This ego-politics is enacted by hiding the subject that speaks and hence his (yes, historically, male) body-political (the ascription of hegemony to race, gender, sexuality) and geopolitical (the ascription of hegemony to an epistemic location in the North) characteristics. For Grosfoguel, this is the basis on which Western science and modernity are constructed. Like Santos, as we have seen above, Grosfoguel connects the purported neutrality of modern science with the ideological project of modernity. Globalization is seen as a continuation of this hegemonic onslaught of the West and its knowledges disguised behind apparently perspective-less theorizing unanchored from their loci of enunciation. According to Castro-Gomez (2005), it is through the ego-politics of knowledge of concealing the speaking subject that the zero-point hubris is also enacted. Like the ego-politics of knowledge, the zero-point refers to a point of view that ‘zeroes’ or conceals itself as a point of view in order to create the illusion of an objective, neutral, all-seeing, ‘god’s eye’ point of view. The zero-point hubris refers to the hegemonic colonial insistence on not only propagating the illusion that there is no other knowledge but Western knowledge but also by hiding all influences from and exchanges with other epistemic traditions that Western knowledge may have had. From the Southern perspective located in Latin America, both the egopolitics of knowledge and the zero-point hubris are integral parts of the epistemic continuation of the colonial experience in the region; this persists long after the declared end of colonization. Current globalization, then, is seen as the continuation of the colonial process that began in the sixteenth century. Quijano (2000) uses the term coloniality of power to describe the continuing persistence of the episteme first established by the European colonizers in the sixteenth century in Latin America. On contact with the Amerindian native, the white Christian Iberian male instituted a selfimage as modern and superior and constructed a discourse of superiority founded on the initial perception of the race of the Amerindian as being inferior. This initial racial superiority founded a semiotic economy of

Glocal Languages, Coloniality, Globalization

31

inferiority that productively extended beyond race to include religion, culture, gender, language, sexuality and modes of production, and characterized all ensuing relations until today. This produced the rhetoric of modernity, the logic of coloniality and epistemic racism. Coloniality contains no small degree of irony:6 As the founding myth on which the rhetoric of European modernity is established, the Amerindian and his deemed inferiority, says Quijano, are constitutive of and necessary for the maintenance of coloniality, the ego-politics of knowledge and the zero-point hubris; without the primitivity or color ascribed to the native, the European cannot be modern or white. This may be a prime example of the complexity of the dynamics of glocalization, often simplistically reduced to dichotomic local-global, homogeneous-heterogeneous or authentic-hybrid dualities. In the case of the coloniality of power each of the parties (hegemonic and marginal) in contact are constitutive of each other. A perception of the mutual interconnectedness between the superiority of the European and the inferiority of the Amerindian is critical for the strategies of border-thinking or the decolonial option as a means of transforming the coloniality of power and avoiding the trap of falling into either a marginal or a hegemonic fundamentalism that affirms one or other element of the hegemonic-marginal coupling that holistically constitutes coloniality. Border thinking involves, first, taking stock of one’s epistemic locus and the multiple discourses that constitute it and, second, working through the limitations of each of these discourses in order to transform them into something more productive. In terms of using border-thinking as a critical strategy for glocalization, the entanglement of multiple epistemes in one’s epistemic locus of enunciation may require no small amount of critical creativity. Similar to and related to border thinking is the decolonial option. Mignolo (2017) defines this as to detach from that overall structure of knowledge in order to engage in an epistemic reconstitution. Reconstitution of what? Of ways of thinking, languages, ways of life and being in the world that the rhetoric of modernity disavowed and the logic of coloniality implement. This means questioning the terms of the conversation, that is, “questioning the structures of knowledge and subject formation (desires, beliefs, expectations) that were implanted in the colonies by the former colonizers” and continue till the present. For Grosfoguel, like Mignolo, the decolonial option involves the decolonization of Western epistemological canons. This involves constructing a broader canon of thought than that imposed by the Western canon; it involves working towards a ‘truly universal’ perspective; not one which

32

Lynn Mario T. Menezes de Souza

is abstractly universal (a globalized local) such as that of the Western canon, but one that may arise from a critical dialogue between diverse critical epistemic/ethical/political projects; finally, the decolonial option requires the consideration of epistemic perspectives located in and thinking from the global south. The objective of the decolonial option for Grosfoguel, the need to dialogue with and relate to other perspectives and those located in the global south is to constitute a pluriversal and not a universal world. The relevance, for a Southern perspective of glocalization, glocal languages and the concepts of the coloniality of power, its ego-politics of knowledge and zero-point hubris, is to raise questions such as: Located as I am, in Latin America, within the rhetoric of modernity and the logic of coloniality, how does globalization relate to me and my context? How does it affect what I consider to be global and local? How does it affect how I see myself? Am I local or global? For whom? How does it affect what I consider to be knowledge and what isn’t? How does it affect the descriptions and analyses I read about globalization and glocalization? How does it affect what I consider to be language and what isn’t? Santos (2007) further contributes to the debate of interconnectedness, mutual constitution and attempts at mutual exclusion by means of the already mentioned concept of abyssal thinking. Santos describes the existence of an epistemic abyssal line, originating in colonial times and persisting under the current conditions of globalization. On this (hegemonic) side of the line, lie the values, cultures, languages and knowledges of metropolitan societies, whilst on the other side of the line lie the values, cultures, languages and knowledges of colonial societies. Abyssal thinking refers to a hegemonic form of thinking that values and makes visible what is on this side of the line and intentionally produces the invisibility of what lies on the other side: Today as then, both the creation and the negation of the other side of the line is constitutive of hegemonic principles and practices. Today as then, the impossibility of co-presence between the two sides of the line runs supreme. (Santos 2007 p. 53) Like the complex mutual constitution and exclusion of the subjects of coloniality as seen by Quijano, the knowledges of both sides of the abyssal line constitute each other. However, the exclusion of one by the other is not symmetrical. The hegemonic this side of the line constructs more actively the invisibility of the other side and could be associated with the strategies of the ego-politics of knowledge and the zero-point in the sense that it seeks to declare itself as unmarked. As a rejection of, and a desire to control difference and impose itself, abyssal thinking closely resembles what Robertson, in relation to

Glocal Languages, Coloniality, Globalization

33

glocalization, called the fundamentalism present in indigenization, where newness is either rejected as undesirable and reduced to invisibility or translated into the already known and declared to be the same as what there was before. Glocal languages as parts of an ongoing process of resistance to and transformation of hegemonic imposition, as examples of the unforeseen use of unauthorized non-hegemonic agency, are not only examples of actions indicating a decolonial option but also examples of a quest to establish co-evalness between both sides of the abyssal line. Glocal languages from this southern perspective are products of hegemonic globalization forcibly submitted to its canons; at the same time glocal languages are indications of the transformation of a purportedly universal and imposed monoculture into a diversality or pluriversality of knowledges. A concept of justice that separates Southern perspectives from metropolitan or hegemonic ones is cognitive or epistemic justice. Whereas, from the metropolitan hegemonic perspective, social justice based on largely economic and political considerations tends to be the goal, from the Southern perspective of both coloniality and abyssal thinking, the quest for critique goes beyond social justice and aims at epistemic and cognitive justice (Grosfoguel & Castro-Gomez 2007, Mignolo 2011, Santos 2007, 2010b, 2010c), which value knowledge, culture and language as being equally necessary for social well-being. Besides the transformative strategy of border-thinking that Grosfoguel proposed above, Santos proposes the strategy of post-abyssal thinking or thinking beyond the divide. Like border-thinking, which takes stock of the constitutive epistemic complexity of the subject in order to go beyond established exclusions, post-abyssal thinking values what Santos calls the ecology of knowledges as a means to dissolve the abyssal line; this consists of opening up established knowledge systems to new ones, thus introducing dynamic incompleteness and plurality without aiming at an ultimate complete totality. According to Santos (2007 p. 10), this move in post-abyssal thinking “defends the fact that the understanding of the world by far exceeds the Western understanding of the world and that therefore our knowledge of globalization is much less global than globalization itself.” This opening up to a plurality or ecology of knowledges, located in but not limited to the margins of hegemonic globalization, aims at promoting epistemic plurality that in fact constitutes, but is unrecognized by, the hegemonic episteme: “The more non-Western understandings of the world are identified [. . .] it becomes more evident that there are still many others to be identified and that hybrid understandings, mixing Western and non-Western components, are virtually infinite” (ibid.). For Santos, this infinity of the epistemological diversity of the world, at the interface of contact between ‘both sides of the abyssal line,’ or ‘across the border of border thinking,’ or in the various global-local

34

Lynn Mario T. Menezes de Souza

interconnections of glocalization, is a form of globalization from below in the sense that it challenges any attempt at establishing divisive lines or totalizing shapes imposed by hegemonic forces. In this sense, glocal languages represent a move that may also be called ‘globalization from below.’ The multiplicity of the ecology of knowledges was always there, but the rhetoric of modernity, Western science and the logic of coloniality, like the ‘story of the winning told by the winners,’ imposed singularities. Once again, the alleged simplicity or dichotomic nature of the globallocal encounter in glocalization does not hold. The concept of the existence of an ecology of multiple knowledges made invisible by the abyssal line of the rhetoric of modernity becomes clear when glocalization is seen to refer not just to the encounter between the (singular) global and the (singular) local but to several globals (e.g. the current global, the historic, colonial global, a regional global etc.) and several locals. As in the case already mentioned of global English, what of the use of Portuguese in indigenous communities or in official indigenous educational policy in Brazil?

A Southern Perspective: A Brazilian Indigenous Contribution In the wake of the Southern strategies of decoloniality, border-thinking and post-abyssal thinking, and their proposed dynamics for challenging metropolitan or hegemonic impositions, two indigenous cultural tropes from Brazil have the potential to contribute to an understanding of the global-local dynamic of glocalization and its processes of recontextualization and resignification: the trope of familiarizing predation (Fausto 1997) and the trope of equivocal translation (Viveiros de Castro 2004). Both tropes refer to the fact that a large part of indigenous cultures in Amazonian Brazil value otherness and newness; both of these aspects are seen as essential for the survival of a community. In other words, for these communities, knowledge-production is inseparable from a firm and conscious anchoring in one’s context; the curious aspect here, relevant to our consideration of glocalization and glocal languages is that in spite of the priority and significance given to one’s local context, it is still seen as insufficient. One’s local context can only exist and survive if in the context of other local contexts. However, probably in order to preserve themselves from self-extinction, the openness to and quest for alterity of these indigenous cultures is highly codified culturally and linguistically. This codification involves processes of resignification and contextualization. In the trope of familiarizing predation, in order to maintain the psychic and physical health (and therefore develop the personhood) of the male individual, this individual must engage in a war of predation that

Glocal Languages, Coloniality, Globalization

35

involves putting oneself at risk and capturing a life-threatening enemy. This predation of the enemy involves capturing the animating force or ‘soul’ of a powerful enemy, thus appropriating the lethality of the enemy and reducing him (culturally it has to be male) to the status of a slave; hence the familiarizing or taming of the enemy. The predation and familiarizing of the enemy is done ritually and symbolically through codified dreams and chants. The enemy is encountered and captured in a ritualized dream. If the enemy is captured and familiarized, he delivers a chant to his captor; the symbolic death of the captured enemy is resignified into the production of a chant. This chant is seen to have an intense healing power that may be used to heal or strengthen the captor or any other member of the community. However, once the healing force of the chant has been used, both the enemy and the chant cease to exist, and a new enemy has to be captured in the exterior to guarantee the well-being of the male member of the community. Thus, male members of the community have to continuously engage in ritual predation and familiarizing. This trope can be fruitfully read in the context of glocalization and glocal languages as relating to the contact between the exterior and the interior, the unknown and the known, the global and the local, framing this contact as necessary, life-giving and fruitful. Though necessary, the contact, if not controlled and codified, is also threatening. In other words, in terms of glocal languages, this trope warns that the new threatens continuity if it is not familiarized or made recognizable. Familiarization here involves recontextualizing and resignifying the new into something apparently already known, even though it is the category (the familiarized, the chant) that is known and not the content. Thus in glocal languages, even though transformation occurs and changes the language from its original form, the appearance of formal continuity needs to remain in order for the glocal language to be considered part of or related to the hegemonic global language from which it emerged. In short, relevance of this trope for the dynamic of glocalization and the emergence of glocal languages lies in the complexity involved in 1) desiring the new; 2) the new, in order to be desirable, has to be lifethreatening; as such, it has to signal discontinuity (in order to be new) at the same time as it offers the promise of continuity and stability through resignification and recontextualization; and 3) the paradox between the new and the old, continuity and discontinuity, life and death, is resolved only formally through the process of familiarization but maintained in practice. Given that only the strange can be made familiar, the ambiguity and irony is maintained beneath the illusory appearance of reducing the new to the old (chants as a category are pre-existing and ‘old’ even though each specific chant has to be captured anew and is therefore ‘new’). The self-esteem and sense of stability of the community needs to be preserved: Nothing has changed; nothing is therefore under threat; continuity

36

Lynn Mario T. Menezes de Souza

prevails; yet all has changed. With the introduction and processing of the new input, nothing remains unchanged; what was before is now no longer, yet the illusion of stability and continuity is maintained. Connections may be clearly and productively drawn between this indigenous cultural trope of familiarizing predation and the concepts of coloniality of power, a body-politics of knowledge, a zero-point hubris and post-abyssal thinking. The most important aspect of this trope for our purposes is that it refers to cross-epistemic contact that is seen as simultaneously necessary and dangerous, and performed through a strategy of border-thinking as explained above. Continuing with our process of dirty thinking (Connell 2007), the second trope, of equivocal translation (Viveiros de Castro op. cit), also originates in the indigenous cultures of Amazonian Brazil and refers again to the high value attributed to the new and the unknown in these cultures and the necessary interconnectedness between these. In the so-called perspectival indigenous cultures of Amazonian Brazil, each species of living being is seen to be culturally (where culture refers to the perceived capacity to think and communicate) equal to every other species. However, each species occupies a different nature or form (human, animal, plant, etc.). Though there may be contact between different natures or forms, their difference, if erased in contact, results in the symbolic ‘death’ of the previous form and possibly in physical demise. What one sees, knows or says depends on where one is located in relation to one’s interlocutor. Thus, while a human may see a jaguar drinking the blood of its prey, the jaguar sees itself as a human drinking beer and sees the human as a jaguar. If however, the jaguar sees itself and the human as jaguars, this indicates an erasure of formal difference between the species and is an indication that one or both have ‘died’, that is, one or both have abdicated from or lost their original form. In this sense, though necessary, all communication across species can only be equivocally translated because there can be no shared perspective from which one can see ‘the same thing’: “Equivocal translation is not to find a ‘synonym.’ [ . . . ] Rather, the aim is to not avoid losing sight of the difference concealed between our language and that of the other since we and they are never talking about the same things” (Viveiros de Castro 2004 p. 7). In relation to glocal languages, this trope suggests that as sites of contact between different languages and knowledges, these languages need to mark their difference from both the hegemonic and the local languages from which they emerge. As in the case of the trope of familiarizing predation, equivocal translation refers positively to the need for contact and exchange between different communities, seen to be different at one level (the cultural) but similar at another (the natural). Because of this, the exchange can never result in convergence, in total translation. The contact or translation will always be partial or equivocal. In

Glocal Languages, Coloniality, Globalization

37

other words, though the same and the different need to attempt approximation and contact, their difference will persist in keeping them apart. Resignification and recontextualization of elements in glocal languages in situations of contact can never occur to the point of eliminating the markedness of these elements. In terms of the abyssal line proposed by Santos, equivocal translation clearly separates the parties involved but does not produce the invisibility of the Other. In fact, in relation to the concept of the abyssal line, equivocal translation, in emphasizing the equivocal, maintains symmetry between both sides of the abyssal line. Therefore, there is and there is not an ‘other side of the line,’ if by ‘other side of the line’ is meant ‘all that is seen as being of lesser value.’ In this case, the ‘invisibility of the other’ is constantly and necessarily attempted rather than actually produced. Neither side eliminates de facto the other; both sides of the line remain clearly visible to each other; it is the line itself that gains prominence as uncrossable and unerasable. The relevance of this trope to global-local contacts in glocalization is that, though total convergence is deemed to be impossible in such contexts, convergence nonetheless needs to be attempted. Furthermore, it signals situations in which the resignification of elements of contact will always remain marked.

Concluding In a world saturated by globalization, the age of the phenomenon suggests that it may no longer be worthy of interest; the work of critique, however, must persist. Here critique was not intended as a quest for a purported ‘true’ or ‘right’ way of knowing. As the discussion above attempted to show, such a concept of unchallenged truth only makes sense in fundamentalism. As academics, we know that knowledge arises from and cannot be separated from social context and that, because of this, knowledge is necessarily ideological and partial. However, as in the Brazilian expression pimenta nos olhos dos outros é refresco (pepper in someone else’s eyes is cooling), unless one has felt the brunt of an ego-politics of knowledge, the import of these expressions remains innocuous. Unless one has been told that the language one speaks is not language or that the knowledge one has inherited from generations before is fiction, an egopolitics of knowledge that empowers such statements by transforming them into science seems little more than a phrase invented to confound. If glocalization is concerned with global-local contact and if by globalization one means the localizing of a phenomenon in all the senses (including the localizing of globalisms) this chapter began with, one needs to be critically aware of the import of one’s locus of enunciation. The perception of ‘global’ and ‘local’ and the need or not for contact and exchange are inseparable from one’s histories and one’s epistemologies.

38

Lynn Mario T. Menezes de Souza

If glocalization involves seeking out and understanding change, we have attempted to show that there is no simple process of change: It can be desirable or not, ethical or not, regulatory or solidary. Hence the necessity of confronting the loci of enunciation of all those involved, not only of those undergoing change but also of those perceiving and reflecting on the change. Working through these loci, reading them against their histories, their epistemologies and the power relations in which they are immersed, and confronting them with one’s own histories, epistemologies and power relations is what decolonial border-thinking is all about. Having one’s perspective and knowledges marginalized and learning to see oneself as Other is what makes us South and apparently impotent; hence the critique that we need to engage in. This includes a reappraisal of glocal languages formerly seen, like knowledges and peoples from the South, as lesser phenomena. The critique that we have sought to develop in this chapter consists, then, of unlearning learned ignorance. I began with an epigraph from Santos (2002a p. 41) and will end with words from Santos (2010a p. 116), where he defines the ignorance of one’s locus of enunciation and the consequence of this ignorance: “The less a given way of knowing knows the limits of its knowing about other ways of knowing, the less aware it is of its own limits and possibilities.”

Notes 1. The de-regulation of neo-liberalism is self-serving and refers to the interests of the dominant, not to the interests of the subaltern who must forcibly submit to hegemonic interests. 2. In Dussel’s (1995) and Quijano’s (2000) considerations of coloniality in the Americas, they identify race as primordial and the basis on which other aspects of the hierarchy of coloniality are constructed. Once alterity has been racialized, all other social aspects of that alterity—culture, language, class, gender and sexuality—are subsequently classified as lacking. Thus, in the specific case of the Americas, where the Other was indigenous, the cultures, languages, class, gender and sexuality of the indigenous Other would always be worth less than that of a white. 3. Santos is here referring to his concept of the hegemonic abyssal line on which modern science is established and on which the invisibility of other knowledges and sciences is produced. 4. For the purposes of this discussion, whose intention is to give importance to one’s locus of enunciation, the difference between the use of Eurocentric, Western, metropolitan, northern, North or Global North is of little significance. 5. Markedness or unmarkedness is always a question of presupposition and perspective. 6. Several post-colonial thinkers, such as Bhabha (1994), define irony in terms of its duplicitous “janus-faced” ambiguity. In order to be irony, it must refer to two things at the same time, one deemed to be more visible than the other. Irony however depends on previous knowledge. If the observer of irony does not identify both things referred to the user of irony, the force of the intended irony is lost. In the case of coloniality, the native victim of coloniality is as important to its maintenance as the white perpetrator of racism.

Glocal Languages, Coloniality, Globalization

39

References Arnaut, K., Karrebaek, M. S., Spotti, M. & Blommaert, J. (eds.). 2017 Engaging Superdiversity: Recombining Spaces, Times and Language Practices, Bristol, Multilingual Matters. Bhabha, H. 1994 The Location of Culture, New York, Routledge. Blommaert, J. 2014 From Mobility to Complexity in Sociolinguistic Theory and Method. Tilburg Papers in Culture Studies, paper 103, Tilburg University, Tilburg. Blommaert, J. & Rampton, B. 2011 Language and Superdiversity. Diversities/ UNESCO 13: 2, pp. 1–21. Blommaert, J. & Verschueren, J. 1992 The Role of Language in European Nationalist Ideologies. Pragmatics 2: 3, pp. 355–376. Brasil. 2006 Orientações Curriculares do Ensino Médio: Linguagens, códigos e suas tecnologias, Brasília, Ministério da Educação, Secretaria de Educação Básica. Canagarajah, A. S. 2017 Translingual Practices and Neoliberal Policies, Cham, Springer. Canagarajah, A. S. 2011 Translanguaging in the Classroom. Applied Linguistics Review 2, pp. 1–28. Castro-Gomez, S. 2005 La Hybris del Punto Cero, Bogota, Universidade Javeriana. Connell, R. 2007 Southern Theory: The Global Dynamics of Knowledge in Social Science, Cambridge, Polity Press. Dussel, E. 1995 The Invention of America; Eclipse of ‘the Other’ and the Myth of Modernity, trans. Michael Barber, New York, Continuum. Fausto, C. 1997 A Dialética da Predação e Familiarização entre os Parakanã da Amazônia Oriental: Por uma Teoria da Guerra Ameríndia, Doctoral Thesis. Rio de Janeiro, PPGAS/Museu Nacional/UFRJ. Flores, N. 2013 The Unexamined Relationship between Neoliberalism and Plurilingualism: A Cautionary Tale. TESOL Quarterly 47: 3, pp. 500–520. Garcia, O. & Wei, L. 2015 Translanguaging, New York, Palgrave Macmillan. Grosfoguel, R. 2013 The Structure of Knowledge in Westernized Universities: Epistemic Racism/Sexism and the Four Genocides/Epistemicides of the Long 16th Century. Human Architecture: Journal of the Sociology of SelfKnowledge 11: 1, Article 8, pp. 73–90. Grosfoguel, R. 2011 Decolonizing Post-Colonial Studies and Paradigms of Political Economy: Transmodernity, Decolonial Thinking and Global coloniality. Transmodernity, Journal of Peripheral Cultural Production of the LusoHispanic World 1: 1, pp. 1–38. Grosfoguel, R. 2009 A Decolonial Approach to Political Economy: Transmodernity, Border-Thinking and Global Coloniality. In Epistemologies of Transformation, Kult 6 Special Issue, Roskilde, University of Roskilde, pp. 10–38. Grosfoguel, R. & Castro-Gomez, S. (eds.). 2007 El giro decolonial Reflexiones para una diversidad epistémica más allá del capitalismo global, Bogota, Universidade Central. Guilherme, M. 2018 ‘Glocal Languages’: The ‘Globalness’ and the ‘Localness’ of World Languages. In Coffey S. & Wingate, U. (eds.) New Directions for Research in Foreign Language Education, New York, Routledge, pp. 79–96. Guilherme, M. 2014 ‘Glocal’ Languages and North-South Epistemologies: Plurilingual and Intercultural Relationships. In Teodoro, A. & Guilherme, M. (eds.) European and Latin American Higher Education between Mirrors; Conceptual Frameworks, and Policies of Equity and Social Cohesion, Rotterdam, Sense Publishers, pp. 55–72. Joseph, J. 2006 Language and Politics, Edinburgh, Edinburgh University Press. Kubota, R. 2014 The Multi/Plural Turn, Postcolonial Theory, and Neoliberal Multiculturalism. Applied Linguistics 33: 1–22, August, pp. 474–494.

40

Lynn Mario T. Menezes de Souza

Maher, T. 2018 Shifting Discourses about Language and Identity among Indigenous Teachers in Western Amazonia in the Wake of Policy Change in Cavalcanti, M. & Maher, T. (eds.) Multilingual Brazil: Language Resources, Identities and Ideologies in a Globalized World, New York, Routledge, pp. 41–56. Maher, T. 1998 Sendo índio em português. In Signorini, I. (ed.) Língua(gem) e identidade: elementos para uma discussão no campo aplicado, Campinas, São Paulo: Mercado de Letras, pp. 115–138. Makoni, S. & Pennycook, A. (eds.). 2007 Disinventing and Reconstituting Languages, Clevedon, Multilingual Matters. Makoni, S., Smitherman, G., Ball, A. & Spears, A. (eds.). 2003 Black Linguistics: Language, Society and Politics in Africa and the Americas, New York, Routledge. Mignolo, W. 2017 Interview with Alvina Hoffman. E-International Relations, January 17. www.e- ir.info/2017/01/17/interview- walter- mignolopart- 1activism-and-trajectory/ (accessed January 5th 2018) Mignolo, W. 2011 The Darker Side of Western Modernity: Global Futures, Decolonial Options, Durham: Duke University Press. Mignolo, W. 2007 Delinking. Cultural Studies 21: 2, pp. 449–514. http://dx.doi. org/10.1080/09502380601162647 (accessed January 18th 2017). DOI: 10. 1080/09502380601162647. Møller, J. 2008 Polylingual Performance among Turkish-Danes in Late-Modern Copenhagen. International Journal of Multilingualism 5: 3, pp. 217–236. Pennycook, A. & Otsuji, E. 2015 Metrolingualism: Language in the City, New York, Routledge. Pennycook, A. & Otsuji, E. 2013 Unremarkable Hybridities and Metrolingual Practices. In Rubdy, R. & Alsagoff, L. (eds.) The Global-Local Interface and Hybridity: Exploring Language and Identity, Cleveland, Multilingual Matters, pp. 83–99. Pieterse, J. 1995 Globalisation as Hybridization. In Lash, S., Featherstone, M. & Robertson, R. (eds.) Global Modernities, London, Sage, pp. 45–68. Quijano, A. 2000 Coloniality of Power, Eurocentrism and Latin America. Nepantla 1: 3, pp. 533–580. Robertson, R. 2015 Beyond the Discourse of Globalization. Glocalism: Journal of Culture, Politics and Innovation 2015: 1, pp. 1–14. Robertson, R. 1995 Glocalization: Time-Space and Homogeneity-Heterogeneity. In Lash, S., Featherstone, M. & Robertson, R. (eds.) Global Modernities, London, Sage, pp. 25–54. Roudometof, V. 2015 Mapping the Glocal Turn: Literature Streams, Scholarship Clusters and Debates. Glocalism: Journal of Culture, Politics and Innovation 2015: 3, pp. 1–21. Santos, B. de S. 2014 Epistemologies of the South: Justice against Epistemicide, Boulder, CO, Paradigm Publishers. Santos, B. de S. 2010a A Non-Occidental West: Learned ignorance and an ecology of knowledges. Theory Culture Society 2009: 26, pp. 103–123. Santos, B. de S. 2010b A Gramática do Tempo: para uma nova cultura política, São Paulo, Cortez Editora. Santos, B. de S. 2010c Para além do pensamento abyssal: das linhas globais a uma ecologia de saberes. In Santos, B. de S. & Meneses, M. P. (eds.) Epistemologias do Sul, São Paulo, Cortez Editora, pp. 31–83. Santos, B. de S. 2007 Beyond Abyssal Thinking: From Global Lines to Ecologies of Knowledge, Review XXX-I-2007, pp. 45–89.

Glocal Languages, Coloniality, Globalization

41

Santos, B. de S. 2002a Towards a Multicultural Conception of Human Rights. In Hernandez-Truyol, B. (ed.) Moral Imperialism: A Critical Anthology, New York, New York University Press, pp. 39–60. Santos, B. de S. 2002b A Critica da Razão Indolente: contra o desperdício da experiência, São Paulo, Cortês. Santos, M. 2007 Encontros, São Paulo, Saraiva. Stroud, C. 2015 Linguistic Citizenship as Utopia. Multilingual Margins 2: 2, pp. 22–39. Viveiros de Castro, E. 2004 Perspectival Anthropology and the Method of Controlled Equivocation. Tipití: Journal of the Society for the Anthropology of Lowland South America 2: 1, Article 1, pp. 3–22.

2

Glocal Languages Beyond Post-Colonialism The Metaphorical North and South in the Geographical North and South Manuela Guilherme Metodicamente o explorador examinou com o olhar a barriguita do menor ser humano maduro. Foi neste instante que o explorador, pela primeira vez desde que a conhecera, em vez de sentir curiosidade ou exaltação ou vitória ou espírito científico, o explorador sentiu mal-estar. É que a menor mulher do mundo estava rindo. (Methodically, the explorer looked at the belly of the smallest adult human being. It was in this moment, for the first time since he had met her that he, instead of feeling curiosity or excitement or victory or scientific interest, felt discomfort. The smallest woman in the world was laughing. [my translation]) Clarice Lispector, A menor mulher do mundo (1920–1977)

Language, Nationality and Coloniality Language official status has been awarded by nation-statehood since the idealisation of this political entity, the nation-state, in France and its first accomplishment in the United States of America, although the latter has never had an official language per se. The expansion of this political model—the nation-state—all over the world mainly resulted from the independence of previous colonies that, after all, tended to reproduce the political model of the colonial power. Hence, it becomes clear that the end of colonialism, in political terms, did not mean the end of coloniality, in epistemological and cultural terms (Quijano 2000, Grosfoguel 2007), and, therefore, language status in the new global era still replicates the hierarchies previously established in modern and colonial times. In addition, at the intra-national level, the link between the nation and the state has been mutually reinforced through the national language and the national system of education. ‘Glocal languages’ are not necessarily national languages, although most languages that have become global across the Atlantic Ocean, which is our focus here—besides the Pacific and

Glocal Languages Beyond Post-Colonialism

43

the Indian oceans—were originally and have become national languages. However, there were indigenous languages, before and during colonisation, that covered immense regions, although they have not crossed the oceans unless by immigration, for example, in South America, Central America, China and India, and which have resisted or survived external colonisation and internal nation-statehood, even though in residual circumstances in some cases. At the international level, on the one hand, elite multilingualism, leading to elite cosmopolitanism, has been composed by linguistic and cultural literacy in distinct national languages whose value is ensured by their power in the global economy and finance, business and politics, while deficit multilingualism brings in the so-called minority-in-power languages, those which have survived both colonialism and coloniality across times and spaces (Guilherme 2007). Such a divide mirrors what Mignolo calls “the geo-politics of epistemology” to describe “the uneven distribution of knowledge” (2005: 44), not that knowledge is unevenly distributed in the world but, instead, he is talking about the uneven distribution of epistemological recognition, across the ‘abyssal line’ as identified by Sousa Santos (2001, 2007, 2014). According to Sousa Santos, such “abyssal thinking . . . consists of a system of visible and invisible distinctions, the invisible ones being the foundation of the visible ones” (2007: 45) in such a way that the former, the invisible, not only supports the latter, the visible, but also gives it the strength of the unquestionable. Such a power system is also expressed, extrinsically, in the relations between distinct languages, as discrete entities, as well as, intrinsically, in the hierarchy of linguistic registers and, finally and more subtly, in the cosmopolitan translanguaging, the hypothesis offered more recently by several authors to be discussed below.

The North and the South—Geography and Metaphor Hence, the idea of plurilingualism, together with the concept of multilingualism, does not only entail an enticing promise of harmony but also evidences the impact of world multiple stratifications whose focus here lies on the North-South metaphor (Sousa Santos 2014) that determines the power relations between peoples, languages, cultures and epistemologies across the Atlantic, namely between and within the Americas, Europe and Africa both in the inter- as well as in the intra-national contexts. As Mignolo puts it, “Theories travelling from the South have the colonial difference inscribed in their luggage” (2000: 183), however, this can also be understood as a metaphorical South that can be found in the globalised societies of both the geographical north and south. The different perceptions of plurilingualism and multilingualism and of the various statuses of plurilingual or multilingual individuals or communities are relative to each social, cultural, political and epistemological context,

44

Manuela Guilherme

and corresponding global capital and are not necessarily numerically determined, that is, by the number of languages one speaks/writes or are spoken/written in a community. Instead, different perceptions of plurilingualism are determined by their status and the power relations between the languages, as entangled with the epistemological distinctiveness of the context that underlies the predominant notion of language and culture in there. It is within this context that we place ‘glocal languages’, meaning a context where languages and cultures cohabit with each other, at different stages, in complex tugs-of-war, whose motivations are not innocent or disinterested and whose results are never definitive. This understanding requires a different perspective of difference that reminds me of a reflection made by Souza about “the indigenous philosophy of perspectivism”, on accounts of his experience, within the scope of a literacy programme, with the Kashinawa people of Amazonia, where he concludes that, in their world: “Difference is qualitative and not quantitative, contextual and never decontextualized” (2007: 160). Thus inspired, difference between distinct conceptions of plurilingualism and multilingualism is then more of a qualitative nature, depending on the languages and cultures involved, rather than to be merely describable in a quantitative manner. Here Souza is inspired by Castro, a Brazilian anthropologist, who argues that, for an indigenous philosophy of perspectivism that he generalises as ‘Amerindian cosmologies’, “personhood and ‘perspectivity’—the capacity to occupy a point of view—is a question of degree and context rather than an absolute, diacritical property of particular species” (2004: 470). The world view of ‘Amerindian cosmology’, according to Castro, unsettles the order and the rationality established by European modernity for which difference is but dichotomous and can be countable. Souza then elaborates on indigenous perspectivism for the purpose of critical linguistic and intercultural literacy supported by Castro, who explains that “Amerindian myths speak of a state of being where self and other interpenetrate, submerged in the same immanent, presubjective and preobjective milieu” (464). Both argue that there other possible ‘perspectives’ rather than that of one independent Self and one independent Other and, furthermore, that such ‘perspectives’ are not only synchronic but also diachronic, not only material but also spiritual.

Plural Multilingualism or Plural Monolingualism? It is important to understand that concepts, although reaching a universal conceptual dimension and common features in practice, such as plurilingualism and multilingualism that exist in many languages, are, at the same time, embedded in particular languages and cultures that do not flow apart from each other or remain enclosed within themselves. Languages indeed follow their lives, even migrate, but neither do languages and cultures become a hybrid mishmash when in contact, nor do they simply or

Glocal Languages Beyond Post-Colonialism

45

totally transcend themselves of themselves, as language and as culture, unless viewed from the perspective of a mythical centre. This can be the case of some theories that have attempted to describe language and culture, in general, as following the waves of hegemonic globalisation and to focus on diversity from where the flow ends and not from where the flow starts. This can be the case of English as a Lingua Franca (ELF), that for some reason only goes for the English language, and of some extreme perceptions of ‘translanguaging’ and ‘transculturalism’. Neither this book, nor this chapter, are only about English teaching; on the contrary, they aim to provide a view of some languages and cultures (teaching/learning of Portuguese, English, Spanish and indigenous languages) from a decolonial South-North perspective—in Latin America, that is, from a particular locus of enunciation, Brazil, despite the fact that data collected cannot be considered as representative of the general language education landscape in that immense territory. Nevertheless, it is relevant to give beforehand some account of the idea of ELF, since this terminology is gaining some ground in all the ‘circles’ identified by Kachru, in a way that the ‘inner circle’ feels relieved that the ‘expanding circle’ is designated as speaking a lingua franca, not English stricto sensu, while the ‘expanding circle’ feels compensated for the fact and legitimated. No matter how un-proficiently in English a native-speaker can perform, s/he will not be allowed in the ELF club, whereas a highly proficient non-native communicator remains qualified as an ELF performer by the ELF supporters themselves. Such a divide between native and non-native English performers is too simplistic and, furthermore, ends up validating this divide and supporting corresponding bias even among non-native speakers themselves who will certainly assume any other arbitrary and immediate criteria for that discriminatory purpose. Moreover, academic arguments for ELF have been published in native-like linguistic proficiency of the ‘inner circle’ and far from the ELF it proclaims. While the argument for EFL aims for plural monolingualism, the argument in this chapter suggests plural multilingualism through the possibility to look at languages and cultures from a ‘glocal’ point of view that encompasses plural loci of enunciation for the same language, both through its own experience in different loci of enunciation and with other languages in the same locus of enunciation. Therefore, this ‘glocal’ position calls for fortis lingui in the plural, not for a lingua franca in the singular. Finally, there is no such thing as a lingua franca at all, since every language is loaded with heavy luggage, the more powerful and dominant the less free the zone is. Not that the role of English has not been important in circumstances of world communication, in those of advantage and in those of disadvantage, however, it must be subject to everyday negotiation with other languages since there is no one language that is sufficient to describe the whole world (Guilherme 2018). The concept of lingua franca is itself acritical, if not uncritical, because it does not entail

46

Manuela Guilherme

critical reflection upon power issues at the deeper layers of communication, instead it targets the superficial and apparent achievements of communication, the top of the iceberg only. Moreover, this is an issue of a North-centric English-speaking elite, both native and non-native, not an issue for most English speakers or relevant for the rest of the population in the world. Jenkins calls English “the primary global multilingua franca”, supported by terminology such as Canagarajah’s ‘plurilingual English’ and Pennycook’s ‘translingua franca English’, while admitting that until recently “multilingualism has on the whole been restricted to underdeveloped references to ELF’s multilingual nature”, therefore, she “finally decided the time had come to work on a reconceptualization of ELF in relation to multilingualism” (2018: 68–76). In this excerpt, Jenkins justifies the interest ELF research work should take in the multilingual background of ELF speakers by condemning the fact that “the learning of additional languages therefore is the learning of their monolingual versions” and by emphasising the need to pay attention to the multilingual nature of ELF speakers and resorting to “ELF users’ multilingual resources”. As I see it, ELF’s multilingualism appears to reside in some kind of linguistic Darwinism, where the stronger grow at the expense of the weaker, and not through some ecologically sustainable development that includes reciprocity and mutual negotiation of permeable linguistic units. The latter, nonetheless, remain as independent, both subjective and objective, entities capable to take decisions at every step without being dispersed in some apparently non-ruled world where the weaker will finally perish. ELF supporters are therefore seeking support in ‘translanguaging’ and ‘transcultural’ theories that have, to a larger or shorter extent, been questioning the existence of languages as discrete systems. From this point of view, ELF carries, in its hubris, the danger of uncritically reinforcing the idea of English as an ambiguous and haunting “Hydra” (Bunce et al. 2016). Indeed, education both in the English language and in the other languages with which it has to negotiate life in this complex world deserve profound disquiet, vigilance and prudence, and that one dives into the deep waters to watch the lower levels of the iceberg.

Globality and Locality Across Synchronicity and Diachronicity Blommaert highlighted the importance of the relationship between text and context when he wrote his critical introduction to discourse. In this regard, he discussed the conceptualisation, dynamics and implications between each other, namely amongst ‘context’, ‘contextualisation’ and ‘entextualisation’, whose discussion adds to the argument of this chapter. The author, to start with, argues that “context is local as well

Glocal Languages Beyond Post-Colonialism

47

as translocal” (2005: 45) and, in order to clarify the intrinsic dynamics of the notion of ‘context’, in its discursive (trans)locality, he ties it up with ‘contextualisation’ which he describes as ‘not unproblematic’ as “all kinds of things can go dramatically wrong” (ibid.: 42). And this is where we get to the nitty-gritty of the problem, all kinds of things can indeed go wrong, or right to be positive, for a number of reasons beyond individual deliberation. However, “Context and contextualisation are dialogical phenomena” (43) and discourse cannot happen but in the process of such a contextualising endeavour, to which another concept concurs, that of ‘entextualisation’ which “refers to the process by means of which discourses are successively or simultaneously decontextualized and metadiscursively recontextualised” (47). Since context always involves constraints, discourse happens within “the interplay between creativity and determination that accounts for . . . the connection between agency and structure, or micro-events and macro-relations and patterns in society” (99). In this book, Blommaert widens the scope of his understanding of context, that he seems to narrow later in his work when he concentrates more on ‘superdiversity’ and ‘translanguaging’, to be discussed below. Still in his critique of ‘Critical Discourse Analysis’ (CDA), Blommaert calls for a ‘higher-level of situatedness’ that is “large, general, supraindividual, typical, structural, and higher than the single society” (2005: 67), however, he later narrows this scope to the ‘here’ and ‘now’, aiming at a sociolinguistics restricted to a globalised urban space. In this respect, Pavlenko and Mullen (2015) give his work as an example of the lack of ‘diachronicity’ in current studies of ‘linguistic landscapes’ and claim that “the interpretation of signs is intrinsically linked to the preceding signs and to related signs elsewhere and is thus diachronic in nature” (114) and only then do such studies “examine linguistic landscapes as a site of social, political and economic changes” (129). Although Blommaert contested this critique elsewhere (2016), by reminding that his ethnographical studies target “traces of multimodal communicative practices within a socio-politically structured field which is historically configured” (2), both discussions about ‘context’ and ‘diachronicity’, as well as about ‘synchronicity’, add to the argument of ‘glocal languages’. This concept is located at the crossings of ‘synchronicity’, of ‘here’ and ‘now’, with ‘diachronicity’, of ‘there’ and ‘then’, therefore, across multiple spaces and times, since ‘glocal languages’ evolve simultaneously either in large regions or across the oceans and have been developed across different historical moments, contexts and geographies. In any way, they live, have survived, expanded and dominated other languages and have been dominated, and eventually remain in close relations with other languages, being ‘contextualised’, ‘recontextualised’ and ‘entextualised’ in social, cultural and political sites, both at home and abroad, in different spaces and times. They have been ‘territorialised’, ‘reterritorialised’ and

48

Manuela Guilherme

‘deterritorialised’, both at home and abroad, in play with different social, political and cultural events.

Superdiversity Requires Superdemocracy Blommaert adopted the description of ‘superdiversity’ for the context, the linguistic landscape, of his ethnographic studies, a term that was put forward by Vertovec to describe the ethnic landscape of London, and that he defines as “a multidimensional perspective on diversity” (2007: 1025), which, besides, it cannot simply be reduced to ethnicity or country of origin, according to him. Vertovec points out the “widely differing statuses within groups of the same ethnic or national origin” (1039) and “enhanced transnationalisation” (1042) of migrants as aspects not to be disregarded and that result in “new patterns of inequality and prejudice” that he enumerates (1045). This concept is further described and dissected in categories by Meissner and Vertovec (2015), and the idea was adopted by several authors while, at the same time, it has responded to earlier concerns expressed by others, as discussed below. These authors suggest new categories and new approaches for a new situation—the intensification of migration to the northern hemisphere, mainly to Europe, that has changed the social and cultural landscape, although the political landscape has not changed that much. If we observe the political and economic leaders of the countries in the northern hemisphere that have been invaded by an explosion of ‘superdiversity’, we must conclude that the circles of power are not ‘superdiverse’ yet. Therefore, we must also conclude that change in numbers is not equivalent to change in power and that inequality remains inequality, despite a few escapes through some net holes. Furthermore, such a state of ‘superdiversity’, with intensive diversified immigration, has been familiar, for centuries in the past, to urban spaces, and beyond, in most southern hemisphere colonies, whereas the reverse was in fact exceptional. Such a dominating north-south movement was general, although with different historical, political and social contours, from the geographical north to the south, in every continent. Even the movements to the east and inland were, to some extent, characterised by this colonial pattern, although each also depending on specific colonial matrices—Portuguese, British, Spanish, etc.—prevalent in each colonised region, and which can be seen as reflected in the hosting patterns to immigration in Europe and in former colonies, with remaining colonial institutional frameworks, as well as in other large powers in the northern hemisphere (China, Russia, Japan). The metaphorical North-South relationship, within previously colonial and colonised societies, seems therefore adequate to describe the situation of diverse societies in a far from completed ‘decolonialisation’ process. With the social, political, cultural and economic implications of such a backdrop, the idea of

Glocal Languages Beyond Post-Colonialism

49

‘super-diversity’ misses the conundrum that such diversity entails, despite its sudden intensiveness in specific spaces where ‘superdiversity’ was not previously allowed in, directly or indirectly, and was until now made invisible. Nowadays, ‘superdiversity’ is piling up at European doors and, more than before, the ‘superdiverse’ ones are needed due to many factors and not only due to travel costs but also to demographic needs of aged societies as well as ‘immigrants’ and ‘refugees’ higher levels of education and, in a few cases, financial status. The concept of ‘glocal languages’ aims to address the linguistic landscape of the same transnational engagements, both personal and professional, that many authors today call ‘translanguaging’ in ‘superdiversity’ (Arnault et al. 2016), but which our work in this book intends to ‘localise’, though in translocal spaces bound by globalisation. In this discussion, we generally bring into play several concepts with different prefixes, e.g. trans-, inter-, multi- and super-, as if they were of common understanding and ready-made universal abstractions, which they aren’t (Guilherme and Dietz 2015). By questioning the prefixes, we also question the nouns, and none is static, neither the trans- nor the interor the multi-, and certainly not -cultural, nor -languaging or diversity. Blommaert finds “much argumentation on postcolonial views highly problematic when it presumes, without much substantive proof, that certain academic discourses are ‘clearly’ locked into one or another culture” (2015a: 22). In fact, the more we pretend that we do not have ties, or better that we are not tied up, the more hidden and inaccessible those ties remain and therefore, the more unquestioned, the more powerful or weaker those who tie or are tied become. If ‘superdiversity’ implies that we don’t acknowledge ‘who’s who’, then there is no diversity at all, because it remains fuzzy and blurred; we can see it but we cannot identify its fabric or its backdrop. For we all know, in academic discourse, there have been some who have been expected to teach and some who have been expected to learn and they both have been “‘clearly’ locked into one or another culture” and, moreover, into one or another language. The conundrum cannot be solved by ignoring the historical-cultural-ethnic-political-social fossils that lie in our way, by remaining blind and moving beyond—by simply ‘translanguaging’, ‘transculturalising’ or ‘superdiversifying’. As Blommaert (2015b) recalls, based on his study of Bourdieu’s work and with regard to Blommaert’s notion of ‘orders of indexicality’, “Recognition as (identity X) is a socially regimented effect that demands recognizability within a frame of intersubjectivity” (8). My argument above goes precisely in this direction, that recognition demands recognisability and, for this purpose, difference should be made evident both synchronically and diachronically, across space and time, while acknowledging mental hierarchies that have become crystallised and established into ‘regimes of truth’, in Foucault’s terms.

50

Manuela Guilherme

There have been certain arguments lately for ‘transcending’ language as well as the idea of multilingualism because the former implies the concept of ‘a’ single language and the latter is viewed as an addition of distinctive linguistic codes; therefore, codeswitching is perceived as jumping from one code to another, both remaining untouched (Jorgensen et al. 2011). Translanguaging, currently adopted by many sociolinguists (e.g. Canagarajah, Blommaert, Pennycook, Wei, etc.), highlights the creative potential of ‘language’ use, in the context of linguistic diversity, without being attached to codes, that is, ‘language’ without ‘languages’, an explosion of linguistic swirls. This is poetics. And poetics used to demand a higher command of the linguistic codes, but “one can extend to all discourse what has been said of poetic discourse alone, . . . the effect which consists in awakening experiences which vary from one individual to another” (Bourdieu 1991: 39). The average ‘citizen’ is thus assumed to enjoy the same discursive liberty, however, only once ‘superdiversity’ entails ‘superdemocracy’ and when full polycentricity is a given and not yet a mirage or if this is not made a hidden detail in the discussion.

Beyond Language and Diversity Pennycook (2017) mentions a general ‘translingual turn’, whose first proposal he credits to Canagarajah and Blommaert, and where he includes a ‘translanguaging turn’, which he assigns to Garcia and Wei. Along the discussion, Pennycook concludes that “there are several strands to this shifting landscape, but they are all unified by an emphasis on language practices”, by which he means “local practices [that] need to be understood in relation to local language ideologies . . . and to other social and cultural practices” (p. 135). Our proposal of a perspective that we call ‘glocal languages’ meets this ‘local’ focus; however, it is our understanding that neither the global nor the local social, cultural or linguistic practices are territorially circumscribed, static or independent from each other, since the two (or multiple) levels live in close interaction with each other. In sum, how local are local practices in the 21st century? Each communication situation in diverse societies—any of them nowadays—is involved in such complexity of intervening features at various levels and of different nature that to reduce it to singularity always lags behind the elastic whole in which it evolves. However, unanimity about the ‘trans-lingual/languaging’ concept is more apparent than real and some contradictions can be found between and even within the various arguments. Wei concedes that, by proposing ‘translanguaging’ as a ‘practical theory of language’, he thought of “a term that better captures multilingual users’ fluid and dynamic practices” in the sense that “multilinguals do not think unilingually in a politically named linguistic entity” (2017: 18). Therefore, Wei accepts the existence of multiple and named languages, although the ‘translanguaging’

Glocal Languages Beyond Post-Colonialism

51

practice entails border crossing on which he, nevertheless, puts the focus. Likewise, his challenge is for ‘language’ to be rethought as a ‘multisensory’, ‘multimodal’ and ‘multi-scalar’ collection of events, therefore, also looking beyond the local nexus. Interesting to our argument is also his description of a “Translanguaging Space . . . where language users break down the ideologically laden dichotomies between the macro and the micro, the societal and the individual, and the social and the psychological through interaction” (23), despite our refusal to believe that these spaces of interaction and border breaking result in an ‘ideologically free’ translanguaging space. This is where the notion of ‘glocal languages’ sets apart from the idea of ‘translanguaging’ practices, as such, and remains suspicious of the ‘lightness’ of the ‘trans-world’ some authors proclaim, in spite of the inspiration that the notion of ‘glocal languages’ may draw from other afore mentioned aspects of the description. Jorgensen et al., for example, offer a ‘polylanguaging norm’ that contradicts a ‘multilingualism norm’, the former consisting of “whatever linguistic features are at their [language users’] disposal to achieve their communicative aims as best they can, regardless of how well they know the involved languages” (2011: 34). Such more radical understandings of ‘translanguaging’ and ‘polylanguaging’ raise questions about language teaching and learning, which is our main interest in this book. If, according to a ‘polylanguaging norm’, language users are expected to employ whatever linguistic features they have at their disposal, we can ask where such linguistic knowledge was acquired at first hand. School? Home? Media? And, furthermore, what is the balance between the different features? Who establishes the ‘polylanguaging norm’? How should it be implemented in language classes, either at university or at school? Say, classes of Danish in Denmark, named as such in the curriculum? First, second, foreign, additional language. Can one imagine that there can ever be no legitimate interference of other languages in any case? Can this be considered a highly probable expectation by language teachers? And what about students and their parents/tutors’ expectations? The authors claim an impossibility, since “learning ‘a language’ is then, with the statements we have made so far, of course impossible in a purely linguistic understanding” (p. 30). Agreed, with or without their statement (Byram 2008, Guilherme 2002). And Jorgensen et al. clarify that “one can learn a number of features associated with a specific sociocultural construction, for instance, ‘Spanish’” (ibid.). One can raise many questions about the statements above. What does the ‘polylanguaging norm’ eventually establish for the identification, breadth and evaluation of such features? What should be the purpose of the ‘polylanguaging’ formal learning from the perspective of the learner? What worries me most is not so much the proposal of ‘polylanguaging’ as the assumptions about the ‘state-of-theart’ in language education and in sociolinguistics that it implies. Wei also briefly mentions the educational implications of the concept for which

52

Manuela Guilherme

he claims, “Translanguaging has proven to be an effective pedagogical practice in a variety of educational contexts”, and he even associates it with a critical pedagogy in that “translanguaging empowers both the learner and the teacher, transforms the power relations” (2017: 15). Still, we remain unknowledgeable about the different perceptions of ‘translanguaging’ that guide such practises, that is, how much you need to ‘translanguage’ in language education in order to be ‘translanguaging’ in a way that meets the theory of language practice put forward as above. The idea of ‘glocal languages’ does not aim to present linguistic theory but to address a critical pedagogy and decolonial curriculum development under the umbrella themes of language education and teacher education. ‘Glocal languages’ require a critical, both synchronic and diachronic, approach to language education enlightened by a political and cultural understanding of language. This approach entails the critical examination of the ‘globalness’ and the ‘localness’ of language use without overlooking the power relations and the subtleness of pressures that they convey on language users and that cannot be ignored while teaching and learning languages (Guilherme 2018). Languages are in fact displayed as discrete entities in the curricula, but they should be regarded and treated like porous and malleable material in the classroom, reflecting social language practice, respecting learners’ linguistic ‘capital’ and heritage while promoting ecological knowledge for sustainable societies. It is evident that we are talking in general about very diverse learning contexts, of which the above references are only a few examples. In sum, language education cannot go without basing itself upon local linguistic practices and enhancing their impact into the global understanding of the language in question, without neglecting the global pressures upon them and, above all, the learners’ potential to agency, their purposes and expectations while engaging in such a task. Makoni and Pennycook stated that “languages, conceptions of languageness and the metalanguages used to describe them are inventions” (2007: 1). Do they mean from scratch? Do they mean that such a thing does not exist at all? The authors clarify that they suggest us to ‘reconstitute’ and ‘reinvent’ the idea of languages as it has developed in the last centuries, that is, to ‘disinvent’ the descriptions of language as they were ‘reinvented’ and schooled in ‘modernity’ and ‘colonial times’ by “rethinking the ways we look at languages and their relation to identity and geographical location” (p. 3). The authors admit that their notion of ‘invention’ is not far from Anderson’s idea of ‘imagined community’ through which he described the formation of nationality (1983). Likewise, Pavlenko and Norton (2007) use the same image, also quoting Anderson, to figure out “the relationship between second language learning and identity”, with a focus on English language education, and language learners’ expectations to be part of respective linguistic communities. However, they expand and make the idea of ‘imagined community’

Glocal Languages Beyond Post-Colonialism

53

plural by pointing out ‘five identity clusters’ to be kept in mind in language education, namely “postcolonial, global, ethnic, multilingual and gendered identities”. The concept of ‘glocal languages’, here put forward, also pluralises the idea of ‘imagined communities’ both beyond and within the ‘nation-state’ entity and identity as well as for the relations between nation states that have intensified due mainly to market globalisation, transnational governance and people’s mobility. The implications of a critical pedagogy of ‘glocal languages’ concur with Makoni and Pennycook’s statement as below: The ideology of invention serves as a critique of language imposition or linguistic imperialism . . . in the sense that the imposition lies in the ways in which speech forms are constructed into languages, and particular definitions of what constitutes language expertise are construed and imposed. (2007: 30) Above all, ‘global’ and ‘local’ concepts also need to be reinvented. Anything that is global has been local and has become global due to unbalanced power relations. More equity, reciprocity and dialogue are necessary in order to achieve such balance in the recognition of the actual entanglement between the local and the global. And this should be done, according to Canagarajah’s, “by taking greater account of the local and respecting its value and validity” (2005: xiv). Furthermore, the ‘local’ has often been delocalised, although not necessarily globalised, through diaspora communities, but remains generally ‘imagined’ as powerless and situated in a static past, while the ‘global’ is connected with progress and the future (Canagarajah 2005). Finally, neither ‘global’ nor ‘local’ is unidimensional; nor are they mutually exclusive layers or is there ‘first-hand global’; ‘global’ is ‘local’, to some extent, imposed and adopted but always recycled locally. ‘Glocal’ aims to encompass such dynamics and ‘glocal languages’ to focus on the dialogicity between and within languages. Canagarajah also discusses the issue of scales that interferes directly with the idea of ‘global’ and ‘local’ and, therefore, the ‘glocal’ scale as well. The author highlights the ways “how micro and macrolevel relations interpenetrate the spaces of globalization and intercultural communication” (2013: 204) that contribute to describe the ‘glocal’ spaces and languages that ‘people’ (as a verb) both the global and local layers of interaction and communication. Canagarajah also notes that: scales cannot be considered as static and predefined. The translocal scale may have been associated with privileged varieties of native speaker English in the past. However, this scale is being redefined with new language ideologies and norms presently. (2013: 221)

54

Manuela Guilherme

Furthermore, the ‘translocal’ scale is also ‘glocalised’ while territorial size is also being made more visible in this process. The view taken by the ‘centre’ of itself shrinks as it starts a dialogue with the peripheries in the process of localising themselves in globalisation (e.g. Western Europe vs. South America or Austral Africa or southeast Asia). The potential that this ‘relocalising’ process may bring to the epistemological wealth in our planet is immense, not only by bringing in some ‘exotic’ raw materials that the metaphorical North keeps recycling or by appreciating the ways in which the metaphorical South has been recycling the colonising epistemologies but also by reshuffling scales, layers and limits without forgetting the remaining social and mental constraints.

Features of Decolonial Language Education in Brazilian Higher Education The empirical study I am describing here was carried out in three federal universities in Brazil,1 lasted for one year (October, 2014–2015) and comprised the participation of 27 university language teachers (Portuguese, English, Spanish, evenly distributed across the three universities, indigenous languages and studies at USP and UFPR only and Afro-Brazilian Studies at UFBA only2). The study was carried out within the scope of the broader aims3 of the Glocademics project, accomplished under the auspices of a Marie Sklodowska-Curie grant (2014–2017), whose aim was to find out, through curriculum analysis, about the respective language’s global and local features that teachers prioritised and enhanced and single out the critical and decolonial approaches they had meant to undertake.4 With regard to Portuguese language education, both classes of Portuguese as mother tongue and as foreign language were taken into account and, therefore, contents depended on this circumstance. However, there are common aspects in both cases such as the focus on Brazilian Portuguese, as it should be expected, and likewise geographical varieties and historical evolution were taken into account. Besides, there is evidence of a strong interest in other varieties of Portuguese and creole categories mainly derived from research carried out in African Lusophone countries on the other side of the Atlantic, together with an understanding of Portuguese as a transnational language. European Portuguese is analysed as one reference point, among others, for comparative studies versus the versions mentioned above, since Portuguese is also understood mainly as a language of contact, in that it has developed throughout its history between other languages: some more powerful languages, other European languages, and also in contact with indigenous languages, as well as together with its creole versions and dialects throughout its multiple territorialities scattered throughout the world. Power relations within the Portuguese language’s diverse

Glocal Languages Beyond Post-Colonialism

55

performances are also approached, as revealed by the data collected, by means of its stereotypes, hierarchy of patterns, related prejudice, semantics, subjectivities and alterities. This led me to conclude that the study of the Portuguese language in Brazilian universities, within the limits of the data collected, enhances language varieties more than the study of English or Spanish, understandably due to its predominant status as mother tongue and official language. The data collected about the teaching/learning of Spanish, according to the sample, give a heavy focus on comparative studies between Latin American Spanish and Brazilian Portuguese in higher education language classes as well as on research, especially in one particular department. According to Fanjul and González, such comparative studies have moved “from inventories of forms to a comparison of functioning” (“de uma comparação de inventários de formas para uma comparação do funcionamento”) (2014: 20, my translation, the author’s emphasis). Participant teachers mentioned that most students decide to learn Spanish because they think that it is easier for Portuguese speakers; however, their learning experience results “in a mismatch, for the Spanish speakers, and, for the Brazilians, in the ratification of an illusion” (“para os falantes de espanhol um desencontro e para os brasileiros a ratificação de uma ilusão”, Fanjul and González 2014: 9). Furthermore, Spanish lessons also pay attention to different textual genres both in the written and spoken language. In addition, a few syllabi also analyse the importance of frontier language contact in Brazil, namely with Paraguay and, in this context, the influence of indigenous languages in the border. All in all, Spanish language education has increased in Brazil and grown more and more focused on Latin America, especially after the Mercosul (Assunção Treaty) was established in 1991. The teaching/learning of Spanish in pre-university education was established, although it still remains a long way behind English. This also coincides with a growing academic focus on Spanish from a local perspective, based in South America, and an increasing awareness of Spanish as a global language competing with English. And the same goes for Brazilian Portuguese— there is increasing awareness of its status as a global language competing with English, both in teaching/learning as well as in research activities and outcomes. In the same line, some academic attention is being given to ‘portunhol’, a mixture of Portuguese and Spanish, which is interestingly an endeavour undertaken mostly by Portuguese speakers whereas the reverse effort is less frequent. ‘Portunhol’ can be considered an example of ‘translanguaging’ that, in any case, can express language status inequality since in Latin America it can occur due to the territorial and economic superiority of Brasil, which leads Brazilians to dare speak ‘portunhol’ with their Latin American peers. The reverse can also happen with Brazilians speaking ‘portunhol’ with Argentinians or Portuguese with Spaniards, for example, besides individual examples and

56

Manuela Guilherme

circumstances. In any case, a greater flexibility, in general, of Portuguese speakers to dare ‘translanguaging’ and to be prone to adventure in other languages should also be taken into account. As far as English is concerned, on the whole, the syllabi contents address the language from a transcultural, nonetheless hegemonic, point of view, even though it is examined through different methodologies, either based on distinct categories of discourse analysis, or on multimodality and critical literacy types, or on intercultural approaches. Often, the understanding of English as a Lingua Franca, more or less explicit, underlies mixed-theory methodologies that sometimes appear to be merely eclectic or indeed paradoxical (namely when it claims for ELF in conjunction with Critical Discourse Analysis or Critical Pedagogy). Actually, it was noticeable that, in general, a different perspective was taken by the participant teachers according to each department of modern languages where this study was carried out as to the role that English is perceived to play in the contemporary world and how it should be approached in language education. This means that it was visible that each department developed a particular endemic school of thought for each language, notwithstanding that each participant kept some degree of independence in their particular approaches to teaching and research. As for Portuguese and Spanish, it was evident that their teaching and research work was more independent and, to some extent, individualistic, perhaps because they were neither perceived as to be foreign to Latin America nor so powerfully imposing from abroad. However, all participant teachers, for every language, maintained their individual international networks and corresponding research fields. To be more precise, while some participant English teachers, in two departments, put a stronger focus on English as an international language, by endorsing English as a Lingua Franca, although often in ambiguous terms as mentioned above, by questioning the supremacy of the native-speaker model and either concentrating on discourse analysis or discussing cross-cultural relations, in another department, they were more concerned in examining “language as the locus of socio-cultural processes in the production of meaning” and in undertaking a “meta-theoretical and critical approach” to language (my translation). In relation to the different approaches mentioned above, in each department, other topics were addressed that evidenced such difference, respectively globalisation and multimodality in one case, intertextuality and interdiscursivity, in another case, and finally technology, science and travelling. Although the teaching/learning of Portuguese and Spanish also revealed different approaches and methodologies in each department, they did not imply distinct perceptions of the roles played by the language in reference. Both Portuguese and Spanish were fundamentally taking an intercultural view departing from the local to the global, which seldom happened with English that generally remained at the transcultural and

Glocal Languages Beyond Post-Colonialism

57

transnational global level, even though Spanish also represents a foreign language in Brazil and students, on the whole, are not expected to travel more frequently to other South America neighbouring countries, or host travellers, than to or from English-speaking countries. Nevertheless, Spanglish in North America was also an issue addressed by one participant teacher. While all participant teachers of English were Brazilian, and speakers of Portuguese as their first language, some of the participant teachers of Spanish were originally either from Argentina or from Spain, but equally staff of their respective universities and at different stages of their career, as were all the participant teachers in this study. Every participant teacher gave priority to the use of authentic materials, not produced for pedagogic purposes, in many cases issued out by Brazilian media, therefore in Portuguese, in order to stimulate discussion of local issues in the foreign language classroom. Despite the fact that Brazilian policy-makers tend to replace the ‘foreign’ terminology by that of ‘additional languages’, both English and Spanish languages were indeed handled as ‘foreign’ by the participant teachers, due to the fact that “English is not used for communication neither within the country (as it seems to be the case when the words ‘second’ and ‘additional’ are used) nor inside the classrooms where it is taught/learned” (Jordão 2011: 39). Moreover, there is an academic need to acknowledge that research on Spanish or Portuguese as foreign or second, or even additional, languages, and corresponding teaching methodologies or comparative analysis, is much more recent, especially in Latin America where linguistic studies, when international, were more directed at dominant European languages, above all of previous epistemological, not necessarily political, colonial powers. Although the study’s focus was curriculum analysis, the participant teachers’ voices were offered the main stage, not only through two 60 minute-long meetings but also in a written semi-structured inquiry, to which almost all participants generously, committedly and fully responded. It seems worth noting that, during this last step of data collection, which consisted of a written ‘final statement’, all participant teachers, no matter the language they taught, expressed a similar language concept, which means that, at the abstract level, they coincided in the same conceptual framework about the language they were teaching. On the whole, they view the language to be taught as political and social practice, a means to communicate ideas and values, a linguistic tool inseparable from its cultural construction, changeable and diverse, responsible for the interaction between the self and the world, impregnated by historicity and power relations. All participant teachers place their scientific field, with regard to their conception of language and language teaching, in sociolinguistics and applied linguistics. When questioned about the linguistic and cultural representations that they privileged in their classes, all the participant teachers agreed

58

Manuela Guilherme

that theirs was an approach focused on linguistic and cultural diversity. However, it was my understanding that the perspective of the English language was one from above, that is, having hegemonic international globalisation in mind, with an eye on North American and British patterns, spreading geographically North-South. On the contrary, Portuguese and Spanish languages were setting their foundations at the local/regional levels, nonetheless diverse, while keeping an eye on globalisation and the growing influence worldwide, spreading geographically South-North. Awareness of the volatility and variability of linguistic and cultural representations, in class and in their social context, was steady. Despite the fact that all participant teachers’ approaches to language were prone to diversity, questioning the fictional model of the nativespeaker, but not necessarily its existence, discussing power relations, hierarchical positions, stereotypes and prejudice, and hesitating in the normative idea of ‘error’, they all were understandably providing, and to some extent requiring from, their university students with an ‘academic’ register of written and oral linguistic performance. This was justified by the fact that they needed to respond to higher levels of mutual intelligibility, communication efficacy and recognition and that this could even be considered as more praiseworthy. According to participant teachers, this is what higher education students are generally looking for, nonetheless considering that they also benefit from critically reflecting upon contextualised linguistic and cultural representations about languages and cultures to which they are having access, both at home and abroad, while increasing awareness about the relativity of centre and periphery relations and corresponding epistemological hierarchies. It is worth remembering that this study was limited to data collection of individual participant teachers’ oral and written statements, the latter through syllabi and final statements, without any classroom observation. Such statements were both theoretical and practical in that most participant teachers reported results of their own research and described the conceptual frameworks that they thought were guiding their pedagogical practices. Hence, when questioned about their conception of ‘interculturalidade’ and how they developed it in their practice, none of the participant teachers showed any perplexity about this terminology, idea or feasibility. All of them referred to a recognition of and respect for linguistic and cultural diversity, no matter the language they were teaching, and to the need of renegotiating meanings and perceiving cultural encounters and fluxes, be they due to the new intensiveness of Brazil’s voice in international relations or the recent increase of voluntary immigrants and refugees, the latter fleeing from war (Syria) or environment cataclysms (Haiti). Not to be disregarded are also the intra-national and, until recently, ‘invisible and voiceless’ groups (e.g. indigenous and quilombolas5). Some of the teacher participants (of Portuguese) described ‘interculturalidade’ as encompassing “new possibilities of

Glocal Languages Beyond Post-Colonialism

59

human relations . . . in complex historical, multilingual, multi-ethnic and multicultural contexts”, while others of English mentioned that it entailed “to problematize difference . . . revise hierarchies . . . and a real opportunity for recognising the Other”, and those of Spanish as “the possibility to recognise ourselves as belonging to a place, although always in dialogue with other places . . . as well as the recognition of differences and power relations that permeate those representations [linguistic and cultural] and social interactions” (my translation). The participant teachers also expressed their concern about addressing citizenship education at its different levels (local, national and global) in their university language classes. In all three languages, Portuguese, Spanish and English, the relation between the global-local and the localglobal world levels was evidenced, however, the starting point was obviously different; while it emerged mainly from the ‘national-local’ direction to the ‘global’ in Portuguese classes, it was highlighted from a ‘regional/local-global’ viewpoint in Spanish classes and from a ‘globallocal’ perspective in English classes. With regard to indigenous languages and Afro-Brazilian studies the direction appeared to be ‘(intra-)national/ local-global’, based on data analysis provided by the respective syllabi and individual meetings only. As for Portuguese as ‘second’, ‘foreign’ or ‘additional’ language, the view taken on citizenship education depended on the student groups, whether classes of incoming or outgoing students or for adult immigrants and refugees. In first language Portuguese, participant teachers were more focused at the intra-national level, on the variety of the peoples of Brazil, and, at the global level, on the lusophone (Portuguese-speaking) peoples, e.g. in Africa. Although this was not their straightforward or explicit objective, the participant teachers, when questioned directly, admitted that all the above would help their students, still at the undergraduate level, to be better prepared to carry out plurilingual and intercultural research, both at the intra-national and transnational levels. This would occur, most particularly, if they had learned how to critically analyse difference, how to place themselves in a position that gives priority to South-South relations (Spanish), how to reflect upon language from its political dimension and to analyse the historical path of colonial dominations (Portuguese language classes), to pinpoint ethnocentric attitudes and to discuss the role of discourse in the maintenance/transformation of power relations. Finally, the participant teachers were requested to comment on the power relations between languages in their departments and to describe their ‘horizon’ as researchers, meaning the limits they had established for the impact of their research work, that is, at the local, regional, national or global levels. Therefore, we may conclude that, within the intra-national scope, Portuguese is naturally stronger in language departments but still exceeding the limits of what one might expect since, in these modern language departments, the publication rate is higher in Portuguese, which

60

Manuela Guilherme

is also the leading language adopted even in international conferences in Brazil, in opposition to English. Interestingly enough is that there has been a reverse tendency in the natural sciences departments with increased numbers of publications in English. Nevertheless, a participant teacher reminds us that it is only the standard form of Brazilian Portuguese, with an academic register, that is awarded more power as opposed to its vernacular varieties, namely Brazilian Portuguese spoken by AfroBrazilian, indigenous or disadvantaged immigrants and refugees, as in society in general. Spanish is gaining ground both in the language departments, mainly after its learning/teaching in pre-university education was enforced by law and, therefore, created an exponential need for many more teachers of Spanish, as well as in the labour market after the Mercosul market was established, although it is still not considered a prestigious language in the Brazilian society according to one of the participant teachers of Spanish. As far as English is concerned, its power has strongly increased not only in the labour market but also in the academy due to the process of globalisation and internationalisation of Brazilian universities and, therefore, countering the tendency to monolingualism in the Brazilian society and education, resulting in an intensive and steady growth in the amount of students and teachers. This overwhelming presence of the English language is also causing conflict in the Brazilian academy, according to one of the participant teachers. However, almost all of them describe their professional ‘horizon’ as ‘glocal’, that is, in the confluence of the ‘local’ and the ‘global’. On the whole, the learning/teaching of indigenous languages and cultures at the undergraduate level in Brazilian universities is almost nonexistent, despite the fact that there are around 200 indigenous languages in Brazil still struggling for survival and that research carried out on indigenous studies in Brazilian universities is proliferating intensively and becoming rather relevant, sometimes excessive from the indigenous communities’ point of view. In this study, I could find and include data about language classes of indigenous languages at the Universidade de São Paulo, both of Ancient Tupi and Modern Tupi (Nheengatu or ‘general language’), from the Amazonian civilisations, and an introduction to the Karitiana language, whose communities refused to mix from the onset of colonisation. Ancient Tupi is the old family tree comprehending the language varieties spoken in the coastal areas of Brazil when the Europeans arrived and, therefore, by then it was called the língua brasílica meaning the language from Brazil, the famous wood—pau-brasil—which inspired the name given to the territory (Navarro 2006). It was still used for the two following centuries, until the 17th century. After the second half of the 17th century, Ancient Tupi developed into Modern Tupi (Nheengatu in the Amazonian territories), which was predominant in Brazil, not only

Glocal Languages Beyond Post-Colonialism

61

among indigenous peoples but also among European colonisers and missionaries and African slaves and, therefore, earned the status of língua geral, common language, later forbidden by the Portuguese government in the 19th century when the Portuguese language was imposed. Nevertheless, another língua geral paulista survived, still for quite some time, inland the S. Paulo State (Navarro 2016). It should not be forgotten that Brazil comprises an immense territory that more than doubles the area of the European Union, which justifies the use of terminologies such as ‘global’ and ‘local’ inside its borders. Moreover, the remaining 200 languages were the ones that survived the “epistemicide” (Sousa Santos 2014) of the 1,500 found by Europeans in the current territory of Brazil, and they are spoken by only a minority of indigenous descendants who have remained in rural areas and in indigenous communities. The Nheengatu still exists in Amazonia where it once was more popular than Portuguese. This language did not exist before colonization; it was developed through language contact among indigenous languages based on the heritage language called Ancient Tupi (Navarro 2016) and pidginised with Portuguese, through Jesuit missionaries, and African languages. Brazilian Portuguese displays a rich heritage of words borrowed from the língua geral and specific indigenous languages, namely in Brazilian toponymy and vocabulary related to food. As far as the Karitiana language is concerned, belonging to the Aikém branch, in the Tupi family, it is spoken by a community of around 400 speakers, in the Rondonia state, in the northwest of Brazil (Storto and Rocha 2015). At the Universidade de S. Paulo, students of such Indigenous languages abound not only raising the interest of students from the Department of Modern Languages but also of some students of Physics and Biology, for example. At the Universidade Federal do Paraná, teaching/learning of Indigenous languages has been less successful even though there was a previous attempt to offer it at the Language Centre (CELIN). The Universidade Federal da Bahia does not offer Indigenous languages courses at the undergraduate level or other. However, it offers courses of Afro-Brazilian literature that focus on Brazilian writers of Afro-Brazilian descent and take both a local/national perspective (Bahia/Brazil) and a global viewpoint with discussion of racism and ethnicity in general and also by providing examples of literary works other than Brazilian ones.

Conclusion This chapter has provided a theoretical discussion of the state-of-the-art in the ‘globalness’ and ‘localness’ of linguistic landscapes and language education. It has made a brief contextualisation of the concept of ‘glocal languages’ with the support of theories on globalisation, coloniality and (trans)nationality. Therefore, it has proposed a decolonial epistemological

62

Manuela Guilherme

framework based on the North-South metaphors for the discussion of the conceptual framework of the notion of ‘glocal languages’. Furthermore, it has addressed this concept in relation to ‘translingual’ and ‘translanguaging’ theoretical arguments in sociolinguistics that have developed with regard to contemporary societies described as in a state of ‘superdiversity’. By introducing the term ‘glocal languages’, this chapter engaged in a discussion about a critical pedagogy of language education through questioning the concepts that have been describing and theorising linguistic practices in our times. Finally, this chapter illustrates the ideas discussed beforehand with the analysis of data collected from a small sample of language teaching in three universities in Brazil, namely Portuguese, both first and foreign language, Indigenous languages, Spanish and English as foreign or additional languages. On the whole, this chapter attempts to counter remainings of coloniality in language practice and education since, in agreement with Souza “even though language, literacy, and culture are multiple and heterogeneous, they are also inseparable from the epistemologies, knowledges, and socio-historic conditions that produce them” (2017: 261). Pequena Flor respondeu-lhe que ‘sim’. Que era muito bom ter uma árvore para morar, sua, sua mesmo. Pois—e isso ela não disse, mas seus olhos se tornaram tão escuros que o disseram—pois é bom possuir, é bom possuir, é bom possuir. O explorador pestanejou várias vezes. (Little Flower answered ‘yes’. She said it was really good to have a tree to live in, to call one’s own, one’s very own. Since—this she didn’t say, but her eyes became so dark that they said it instead—it’s so good to call something one’s own, to call something one’s own, one’s very own. The explorer’s eyes blinked several times. [my translation]) Clarice Lispector, A menor mulher do mundo (1920–1977)

Notes 1. Universidade de S. Paulo, Universidade Federal da Bahia, Universidade Federal do Paraná. 2. (USP) Universidade de S. Paulo; (UFB) Universidade Federal da Bahia; (UFPR) Universidade Federal do Paraná, respectively in the centre, the northeast and the south of Brazil. 3. The main aim of this project, and of its two-year empirical study in Brazil, was an interdisciplinary study of the implications of plurilingualism and intercultural epistemological negotiation throughout the research tasks of five research groups (Linguistics, Political Science, Indigenous studies, Biology/ Ecology/Philosophy of Science and Nutrition) at the Universidade de S. Paulo, Universidade Federal da Bahia and Universidade Federal do Sul da Bahia. 4. The general methodology was the following: (1) first 60-minute meeting with each participant teacher in order to find out about subjects they taught at undergraduate level, priorities with regard to teaching objectives, curriculum contents and methodologies, and finally to jointly select the two subject syllabi on which data analysis was going to be focused; (2) data categorization

Glocal Languages Beyond Post-Colonialism

63

according to contents, perspective, approach and bibliography; (3) second 60-minute meeting meant to discuss completed syllabi analysis and to clarify the aims of the final written statement to be produced by each individual participant teacher; (4) final statement delivery and analysis. 5. Descending from ancient African slaves who fled from their masters and occupied small farms or formed villages called quilombos. This word originates in tupi-guarani and means “that who flees”. These populations often mixed with indigenous tribes who sheltered them. There are still more than 2,000 quilombos, whose land was theoretically made legally owned by the 1988 Federal Constitution. They are scattered mainly in the northeast and centre of Brazil, in the regions where slavery was once more present.

References Anderson, B. (1983) Imagined Communities: Reflections on the Origin and Spread of Nationalism. London: Verso. Arnault, K., Blommaert, J., Rampton, B. & Spotti, M. (eds.) (2016) Language and Superdiversity. London: Routledge. Blommaert, J. (2005) Discourse: A Critical Introduction. Cambridge: Cambridge University Press. Blommaert, J. (2015a) Commentary: ‘Culture’ and superdiversity. Journal of Multicultural Discourses, 10, 1, 22–24. Blommaert, J. (2015b) Pierre Bourdieu and Language in Society. Working Papers in Urban Language and Literacies, 153, Tilburg University. Blommaert, J. (2016) The conservative turn in Linguistic Landscape Studies. https://alternative-democracy-research.org/2016/01/05/the-conservative-turnin-linguistic-landscape-studies/ Bourdieu, P. (1991) Language and Symbolic Power. Malden, MA: Polity Press. Bunce, P., Phillipson, R., Rapatahana, V. & Tupas, R. (2016) Why English? Confronting the Hydra. Bristol: Multilingual Matters. Byram, M. (2008) From Foreign Language Education to Education for Intercultural Citizenship: Essays and Reflections. Clevedon: Multilingual Matters. Canagarajah, A. S. (2005) Introduction. In A. S. Canagarajah (ed.) Reclaiming the Local in Language Policy and Practice (pp. xiii–xxix). Mahwah, NJ: Lawrence Erlbaum Associates. Canagarajah, A. S. (2013) Agency and power in intercultural communication: Negotiating English in translocal spaces. Language and Intercultural Communication, 13, 2, 202–204. Castro, V. De (2004) Exchanging perspectives: The transformation of objects into subjects in Amerindian ontologies. Common Knowledge, 10, 3, 463–484. Fanjul, A. P. & González, N. M. (2014) Políticas do saber e (re)descoberta das línguas. In A. P. Fanjul & N. M. González (eds.) Espanhol e Português Brasileiro: Estudos Comparados (pp. 7–25). São Paulo: Parábola. Grosfoguel, R. (2007) The epistemic decolonial turn. Cultural Studies, 21, 2–3, 211–223. Guilherme, M. (2002) Critical Citizens for an Intercultural World: Foreign Language Education as Cultural Politics. Clevedon: Multilingual Matters. Guilherme, M. (2007) English as a global language and education for cosmopolitan citizenship. Language and Intercultural Communication, 7, 1, 72–79. Guilherme, M. (2018) ‘Glocal languages’: The ‘globalness’ and the ‘localness’ of world languages. In S. Coffey & U. Wingate (eds.) New Directions for Research in Foreign Language Education (pp. 79–96). London: Routledge.

64

Manuela Guilherme

Guilherme, M. & Dietz, G. (2015) Difference in diversity: Multiple perspectives on multi-, inter-, and trans-cultural conceptual complexities. Journal of Multicultural Discourses, 10, 1, 1–21. Jenkins, J. (2018) No English but English-within-multilingualism. In S. Coffey & U. Wingate (eds.) New Directions for Research in Foreign Language Education (pp. 65–78). London: Routledge. Jordão, C. M. (2011) New designs for new identities: Are we going anywhere? Critical Literacy, 5, 1, 36–44. Jorgensen, J. N., Karrebaek, M. S., Madsen, L. M. & Moller, J. S. (2011) Polylanguaging in superdiversity. Diversities, 13, 2, 23. Makoni, S. & Pennycook, A. (2007) Disinvented and reconstituting languages. In S. Makoni & A. Pennycook (eds.) Disinventing and Reconstituting Languages (pp. 1–41). Clevedon: Multilingual Matters. Meissner, F. & Vertovec, S. (2015) Comparing super-diversity. Ethnic and Racial Studies, 38, 4, 541–555. Mignolo, W. (2000) Local Histories/Global Designs: Coloniality, Subaltern Knowledges, and Border Thinking. Princeton, NJ: Princeton University Press. Mignolo, W. (2005) The Idea of Latin America. Malden, MA: Blackwell. Navarro, E. de A. (2006) Método Moderno de Tupi Antigo: A língua do Brasil dos primeiros séculos. São Paulo: Global Ed. (3ª ed.). Navarro, E. de A. (2016) Curso de Língua Geral (Nheengatu ou Tupi Moderno): A língua das origens da civilização amazónica. São Paulo: Paym Gráfica e Editora (2ª ed.). Pavlenko, A. & Mullen, A. (2015) Why diachronicity matters in the study of linguistic landscapes. Linguistic Landscape, 1, 1/2, 114–132. Pavlenko, A. & Norton, B. (2007) Imagined communities, identity, and English language learning. In J. Cummins & C. Davison (eds.) International Handbook of English Language Teaching (pp. 669–680). Dordrecht, Netherlands: Springer. Pennycook, A. (2017) Language policy and local practices. In O. Garcia, N. Flores & M. Spotti (eds.) The Oxford Handbook of Language and Society (pp. 125–140). Oxford: Oxford University Press. Quijano, A. (2000) Coloniality of power and Eurocentrism in Latin America. International Sociology, 15, 2, 215–232. Sousa Santos, B. de (2001) Toward an epistemology of blindness: Why the new forms of ‘ceremonial adequacy’ neither regulate nor emancipate. European Journal of Social Theory, 4, 3, 251–279. Sousa Santos, B. de (2007) Beyond Abyssal thinking: From global lines to ecologies of knowledges. Review, 30, 1, 45–89. Sousa Santos, B. (2014) Epistemologies of the South. Boulder: Paradigm. Souza, L. M. T. M. de (2007) Entering a culture quietly: Writing and cultural survival in indigenous education in Brazil. In S. Makoni & A. Pennycook (eds.) Disinventing and Reconstituting Languages (pp. 135–169). Clevedon: Multilingual Matters. Souza, L. M. T. M. de (2017) Multiliteracies and transcultural education. In O. Garcia, N. Flores & M. Spotti (eds.) The Oxford Handbook of Language and Society (pp. 261–279). Oxford: Oxford University Press. Storto, L. & Rocha, I. (2015) Estrutura argumental na língua Karitiana. In L. Storto, B. Franchetto & S. Lima (eds.) Sintaxe e Semântica do verbo em Línguas Indígenas no Brasil (pp. 17–41). Campinas, São Paulo: Mercado de Letras. Vertovec, S. (2007) Super-diversity and its implications. Ethnic and Racial Studies, 30, 6, 1024–1054. Wei, L. (2017) Translanguaging as a practical theory of language. Applied Linguistics, 39, 1, 9–30.

Section II

Indigenous Languages as Glocal Languages

3

Glocalism Now and Then The De-Colonial Turn of Guarani, Portuguese and Spanish Fernanda Martins Felix

Introduction This chapter integrates a theoretical background on historical conditions about Portuguese, Spanish, and Guarani languages, and a concept of a workshop designed to put into practice the discussed theory. Guarani, Portuguese, and Spanish are languages of significant power in Latin America and have been sharing territories for more than 500 years. The first part of this chapter intends to contextualize their battles for space, voice, and power in the world, and the dynamics that have led to a fusion of languages and cultures. Also part of this fundamental moment is a positioning on the glocality of those three languages and how, by so, they have overcome colonialism in certain ways. The struggles for a “de-colonial turn”, making use of the concept by Sousa Santos (2004), have found some tools in contemporary flows, and the pedagogical approach of Intercomprehension among related languages, is here presented as one of them. Seeking discourse practices where interlocutors can use their own repertoire, while being able to comprehend the languages of the others because of their lexical, morphological, and phonological similarities comes up as an instrument for tackling hegemonic mentalities on language contacts, policies, and teaching/learning. In the second part, this chapter announces a proposal of experience based on Intercomprehension among related languages approaches, taking the three languages as subjects. Even though the Guarani language has a completely different root from the Neo-Latin languages Portuguese and Spanish, they have been in contact for so much time that each of them is deeply permeated by the others. This level of relatedness might be perceived as a space in which Intercomprehension experiences could take place (Felix 2016). Making use of the accessible social media Facebook, this practical part aims to the structuring of workshops, university subjects, language teaching/learning courses, etc. From a perspective aligned with plural pedagogy studies, this practical proposal has been entitled “Awaking to what is ours”, while attempting to reduce the “waste of experience” (Sousa Santos 2004), and to celebrate knowledge that lingers around. This practical

68

Fernanda Martins Felix

proposal’s fluid structure calls to collectivity and engagement. What is presented here does not compose a rigid method. Quite the opposite, it is a first attempt to brainstorm on the search for experiences that are contextualized with nowadays society and struggles for the validity and legitimacy of cultural, linguistic, and knowledge diversity. The hundreds of languages that exist in South America are, unquestionably, a common heritage to humankind. Not only should they be recognized as such and preserved, but also this diversity should become a fountain from which we can learn and prosper. Giving voice to languages that have been historically underestimated is more than recognizing the importance of languages within cultural identities construction; it is an opportunity to all—as we benefit from the cultural knowledge that coconstructs every language. Throughout history, indigenous languages share experiences of contact with non-indigenous languages, such as Portuguese and Spanish. They experience subjugation, imperialism, resistance, transformation of themselves and the cultural societies intrinsic to them. As global contacts intensified, speakers have found a wave of opportunities. What was once a period of obliteration of many languages now is much more adapted to diversity. Documenting languages that linger through oral traditions has never been so achievable. In addition, with the advent of the Internet, many minoritized languages have conquered more space on the globe. Even though efficient digital resources are far from evenly diffused, indigenous languages such as Guarani are growing on the web. The significance of the Guarani language presence reflects its history as a language with wide powers in ancient Latin America, a language that has resisted through more than 500 years of colonization. Living today with languages considered “glocal”, Guarani is experiencing phenomena quite similar to becoming a glocal language itself. Widespread and powerful during a pre-colonial time, the Guarani language already had a glocal existence: structured by many local practices and yet covering extensive lands. Nevertheless, the age of the Internet intensified contacts to a point where the glocality of the Guarani language is a growing process, making through the savage decimating from colonialism. The communicative resources provided by a globalized society can rearrange language roles and the way speakers relate in their cultural society or outside it. For the purposes hereby exposed, digital platforms such as educational projects Galanet1 and MIRIADi2 or even social media network Facebook3 come to hand as significant tools for the process.

Historical Conditions on Guarani, Portuguese, and Spanish: How Their Glocality Reflects an Overcoming of Colonialism The dynamical character of languages is a long-lasting subject of interest. Despite this, not all speakers ask themselves where their language came

Glocalism Now and Then

69

from or why one language is so similar to another; many are the historical and comparative studies that allow us to have a glimpse on how language development works. Languages that nowadays play important roles in worldwide communication and that, sometimes, even get in conflict with one another for political reasons are frequently connected by a common ancestral link. That is the case of Portuguese and Spanish, languages of specific interest in this chapter. Deriving from the imperialistic Latin language, more specifically its vulgar variety, they have progressively grown in Roman Hispania territories. Latin kept its hegemony until the Roman Empire ended in the fifth century. The barbarian states that followed the disintegration of the Empire, such as Hispania, configured themselves as spaces of innovative phonetics and grammars, of new linguistic constructions, until the Latin hegemony was definitively broken and the local differences became official languages. The precise period when the Roman conquest began in the Iberian Peninsula and the social basis that settled in that territory produced specificities in the languages that originated there. Not only the fact that when the peninsula was conquered, the Latin language had not yet acquired its classical form, but also the particular cultures of the colonizers that have arrived to the new settlements have been effective components of a linguistic development that culminated with the organization of the Portuguese and Spanish languages (Fernández Jaen 2006). The expansionist policy of Iberian kingdoms involved new spaces in the practices of Portuguese and Spanish. From the sixteenth century on, the Romanic languages have established themselves permanently over the Americas and until the eighteenth century, Portuguese was the communication language in India and Southeast Asia harbours as well (Teyssier 1982). The peculiarities of these diverse places implied a constructive variation of linguistic production. Nevertheless, the political actions amongst the lusophones and hispanophones territories keep directed towards a still shallow recognition of the varieties, focusing on the maintenance of centres of reference, which remain conveniently connected to the colonizers. This still keeps the mentality of centralization of power over the language and the restrictive normative objectives. There are nine territories that unite by sharing Portuguese as an official language. They are Portugal, Brazil, Angola, Cape Verde, Guinea Bissau, Macau, Mozambique, São Tomé and Príncipe, and Timor-Leste. Apart from these nations, emigrants all over the world are speakers of this language, which makes around 260 million speakers. It is the eighth language in numbers of speakers in the world, and the third when we consider only the Western world (Peralta 2008). It is also one of the official languages of the European Union since 1986, and of the MERCOSUR. The Spanish language takes second place in Western languages with the biggest amount of speakers. Worldwide, it is the fourth, only behind

70

Fernanda Martins Felix

Mandarin, English, and Hindi, according to the Instituto Cervantes (2016). The number of speakers of Spanish, which is believed to be around 472 million, is increasing, with predictions that assume the mark of half billion speakers by 2030. This allows the deduction that the contemporary practices of sharing and strengthening of the Spanish language have been effective on their purposes. It is a greatly spread language, sustaining twenty-one Hispanic speaking countries. They are: Argentina, Bolivia, Chile, Colombia, Costa Rica, Cuba, Dominican Republic, Ecuador, Equatorial Guinea, El Salvador, Guatemala, Honduras, Mexico, Nicaragua, Panama, Paraguay, Peru, Puerto Rico, Spain, Uruguay, and Venezuela. In Central and South America, Portuguese and Spanish share a great territory conquered by the European colonial entrepreneurship. Because of MERCOSUR policies, both of them are sustained to be taught, on regular education in Brazil, Argentina, Paraguay, Uruguay, and Venezuela, the official members of the organization. Even though, for many, they are the languages of Latin America, they are actually only two of the hundreds that co-exist in this “cultural-linguistic continent”. They share lands and social interactions with a large number of other languages. Immigrants and native languages, which have been lingering minorities and minoritized, have been in contact with these two Romance languages for centuries. The relationships amongst them are configured as intricate dynamics, which have been historically favouring Portuguese and Spanish in detriment of the entire variety. Only in Brazil, where Portuguese remains as the official language of the nation and Spanish is established as foreign language in regular basic education, there are more than 200 other languages actively in production across the country (Lewis et al. 2013). This is, surprisingly, a relatively small number, compared to what existed before the arrival of the European colonizers. The number of indigenous languages enduring over Brazilian lands, nowadays, turns around 170 languages (De Oliveira 2009), what is believed to be only 25% of what once existed (Moore et al. 2008). Even though the majority of them have perished along the way, some of them remain vivid and linger through cultural groups that struggle to keep them alive. The situation of the Brazilian minority languages, amongst themselves, is obviously heterogeneous. There are languages in high risk of extinction with less than ten speakers alive, like some from the Macro-Jê linguistic branch, and others, which have not been documented yet. They persist on oral traditions, although, depending on the community, this is no longer a wide habit (Guirardello 1993). Fortunately, some of them remain quite successful, with a large amount of speakers. That is the case of the Guarani language, for instance. The genealogy of the Guarani language goes back to what is nowadays considered to be a whole linguistic family, but, just like Latin,

Glocalism Now and Then

71

had been once a language itself. Guarani is a Tupi-Guarani language, derived from what comparative linguistics refers to as the proto-language Tupi (Rodrigues 1993). This branch holds together languages like the Nheengatu, the Kamayurá, and all significant indigenous languages in existence. On the arrival of the European colonizers, the Tupi language was used to instrument intercultural contacts. It was known as “língua geral” and “lengua general” (Lagorio and Freire 2014: 574). The convenience of adopting languages of wide communication took over extensive geographical areas, “many times formed by ‘empires’ that have now disappeared”4 (ibid 575). Not only the Tupi language, but also the Quéchua, Aimará, Náhuatl, and Chibcha all have sustained intercultural contacts as línguas gerais. The concept of línguas gerais, like the current perception of English as a lingua franca, is aligned with the understanding of the global practices of languages over extensive areas of land. From the perspective of globallocal characteristics in languages, it can be relevantly noticed that those powerful ancient languages owned a diversity of local practices, while stringing together and impersonating growing linguistic empires. As happens to languages during war and colonialism, there is a revolution when politics realize how power is connected to language, when languages are perceived as one of the supreme constitutional spheres of human living. The European groups arriving to colonize immediately recognized the wide use of the Tupi by different ethnic groups in ancient South America. There were even documented perceptions, exemplified by the appointment of Pr. Antonio Ruiz de Montoya about a language “so universal that domains both seas, the one from the South, all over Brazil, and the two great rivers, the Plata and the Maranhão”5 (Melià et al. 1987: 102). This area supposedly corresponds to an extension bigger than India, for example. The Jesuits elaborated the first written documentations of this language, and the adoption of the Latin alphabet constituted a proposal of comprehension, absorption, and standardization, all precise strategies of colonial politics of expansion (Lagorio and Freire 2014). Even though, paradoxically, many of those documents play, nowadays, the main role on languages maintenance and expansionism. Even though the dynamics on that time involved a larger number of monolingual speakers of the língua geral rather than bilinguals, to Ayron Rodrigues (2000: 11), the línguas gerais did not reach the “stability that would allow them to expand in space and survive for a long time”.6 By the eighteenth century, the línguas gerais no longer sustained their vivacity, and Portuguese initiated to gain control over the national territory. Not only the immigrants who came for the gold and diamond mines, many of them speakers of Portuguese, but mainly the political actions have started an effective obliteration of the native languages. As an example, the Directory established by the Marquis of Pombal, in

72

Fernanda Martins Felix

1757, prohibited the use of the línguas gerais and officially obligated the use of the Portuguese language. From this moment on, Portuguese would have barely eliminated the linguistic concurrence, reducing them to sparse lexis, integrated in local vocabularies. It also was during the eighteenth century that the first documents engaged in characterizing the specific traces of Brazilian Portuguese; there was a realization about the specificities from the different places of practices. The Tupi language inevitably permeated Brazilian Portuguese, even though this language was no longer recognized as an official characteristic of the nation and its citizens. Moreira (2005) defends that, because of strate phenomena suffered by Portuguese and Spanish languages in their conquering of the South American territory, these languages went through the natural tendency of receiving lexis influences from the native languages, especially from the Tupi. In a power game, the Tupi language, powerful and widespread as it was back then, injected many semantic contributions that remain preserved in proper names, in fauna and flora names, in food names and represent a rich, yet neglected, corpus in South American Portuguese and Spanish etymology studies. The Guarani group is one of the main survivors from the Tupi peoples. There are wide linguistic groups in Brazil, Paraguay, Argentina, Uruguay, and Bolivia. The far-reaching Guarani territory is believed to advance, nowadays, over one-fifth of South American lands (Silvetti and Silvestri 2015). Because of the lack of investments in the language, data are quite inconclusive when it comes to the total number of speakers. Appointments estimate from 5 to 8 million speakers of Guarani (Lewis et al. 2013; Silvetti and Silvestri 2015), being more than 4 million of them Paraguayans and between 30,000 and 50,000 Brazilians. According to the latest Brazilian Demographic Census (IBGE 2010), 67,523 people within the Brazilian territory declared or considered themselves belonging to a Guarani ethnic group. From this number, around 50% spoke their language at home. Guarani has held the status of official language in Paraguay, along with Spanish, since 1992. Lagorio and Freire (2014) appoint that 85% of the country’s population speak Guarani, either exclusively or combined with Spanish. This country has been successful in maintaining its investments over this native language, mainly because of strategic movements during war periods. It has been politically favourable, along with socially interesting, to keep the population of Paraguay bilingual (Felix 2016). It is also because of this country that Guarani remains as official language of the MERCOSUR. Even though, differently from Spanish and Portuguese, it does not benefit from the same amount of investments or interests within the group. Guarani has become official in the province of Corrientes, in Argentina and in some Brazilian cities, such as Dourados, in the state of Mato

Glocalism Now and Then

73

Grosso do Sul. Apart from Paraguay and the Argentinian province of Corrientes, where Guarani is taught in the regular education systems, no other places with significant Guarani groups have yet implemented effective language policies for the maintenance and promotion of this language (Dietrich 2010). Because of the colonial dispersion and the cultural habits of relative nomadism, different varieties compose the Guarani group, while they have kept their individualities and a common sense of unity. Specially defined by Silvetti and Silvestri, “It possesses a distinctive particularity among American native languages: it is not only spoken by indigenous communities but by all groups and social classes: it is the only pre-Columbian language spoken by a large non-indigenous community” (2015: 3). Reduced in numbers of speakers and spaces of occurrence, after the colonial attempts of obliteration, the Guarani peoples have been demonstrating notable practices of resistance, which have allowed, even under huge adversity, the survival of one of the most representative indigenous people in the Americas and in the world. With the current globalization flows in intensive scale, Guarani speakers have spread all around the globe. Not only are there representations of Guarani members in migration currents, but also virtual linguistic contacts have put this language in a privileged position amongst minority languages. The amount of data available in and about Guarani language presents an unquestionable fact that Guarani is a powerful enough language to fight colonialism and to spread over many places. Such optimistic information has only earned sparse governmental initiatives and, more often, some private actions. Despite concluding that, compared to other minority languages, Guarani holds a position of prominence; when we compare it to widely promoted languages such as Portuguese and Spanish, investments are far from substantial or satisfactory. As pointed out by Guilherme (2014: 56) [T]he hierarchy of languages in multilingual, postcolonial settings, between European colonized languages and indigenous languages, and even amongst themselves, reveals the complex relationship that multiculturalism and multilingualism has shared throughout history. This proves the extent to which the hierarchy that comes from the dynamics of colonial systems is still a strong residual in contemporary post-coloniality. In fact: “[m]uch of the post-colonial theory (Spivak 1990; Bhabha 1986; Gandhi 1998; Mignolo 2000, 2007) has pointed to the continuity and not the break implied in the prefix post-, where elements of colonial hegemony persist long after the departure and end of official colonialism” (De Souza 2014: 36). This hegemonic form of reason that persists and needs to be seen as shooting ourselves in the foot has unequally valued the different languages

74

Fernanda Martins Felix

and knowledge that co-construct societies. When we speak of (non-)validated knowledge, we are talking about a “waste of experience”, as proposed by Sousa Santos (2004), when we ignore “the alternative knowledge and practices that exist” (De Souza 2014: 43) and maintain ourselves in a limited row of cosmologies that impoverishes our knowledge repertoires. This comes in accordance with a group of scholars who, in Guilherme’s words (2018: 4): have been theorizing about a “de-colonial turn” (el giro decolonial) that Mignolo defines as an epistemic decolonization, a new way of thinking that imprints an unthinkable fracture in the imperial genealogy of modernity, a kind of thought that releases and opens a critical border thinking (2007). (. . .) [H]e also claims that “we need to think seriously about the processes by which languaging and the allocation of meanings to groups of people presumed to have common features (e.g., ‘ethnic culture’, ‘national culture’, etc.) are being relocated and how linguistic maps, literary geographies, and cultural landscapes are being repainted”. These assumptions bring a lot into discussion. They suggest that not only should we detach ourselves from the colonial ways of thinking, which transfers languages and knowledge from “North to South”, but to rethink all the spatial and social aspects of languages and cultural groups. With the unfolding of colonialism and globalization, which have constituted different phases of a continuing process led by capitalism at both its early and later stages, North and South have, to some extent, been de-territorialized and therefore permeated every society in both hemispheres. (Canclini 2005 cited in Guilherme 2014: 54) The “North and South”, so historically dichotomic and imprinted with so many symbolic values, have been facing innumerous cracks that are punctually located and head unpredictably around the globe. This brings the necessity to look from many perspectives and allows languages and their speakers to move intricately. What Sousa Santos (2004) brings in the terms of “cosmopolitan reason” is an opening of possibilities, a different model of rationality that expands that of the Western colonist reason. Explained by Lynn Mario T de Souza (2014: 43): This cosmopolitan reason [. . .] would seek to expand the present and contract the future, re-signifying the concepts of time-space in order to make visible the co-extensive complexities and diversities (ecology of knowledges) existent today so that the knowledges and

Glocalism Now and Then

75

social practices once produced as absent and invisible,—wasted— may emerge, be faced and signified in all their complexity. As languages, knowledge, and culture are interlaced, the demand for spaces where languages can prosper and “from which we all have to reciprocally learn the most” (Guilherme 2018: 5) is a right to be fulfilled. Leaning on globalization resources, reorganizing the uneven state of access and investments, languages may be empowered, as some of them have been. A main aspect of these reflections is that languages have been surviving “between and across the global and the local” (Guilherme 2014: 53) and their characteristics are not frozen. The sophisticated development of languages demands updates on how we conceptualize its components. It is not new that several authors have been theorizing about the relationships between the local and the global aspects of language dynamics. According to Sousa Santos’ remarks (1999 cited in Guilherme 2018), modes of production that compose globalization, more specifically globalized localisms and localized globalisms, have been happening on languages developments and present their own characteristic results in languages contacts, languages developments, and in how speakers relate to their language(s) and the language(s) of the others. The impact of the connection between the local and the global has brought us to a point where analyzing them from a perspective of an intrinsic bond is a strong way of validating their current flows as well as of objectifying dynamics that need to be understood in order to pursue a counter-hegemonic strategy for languages and their speakers. It is taken into consideration that “the dichotomy of global scale versus local setting is false, for we live in a globalized world, we live it through local circumstances, and the terms global/local are necessarily linked” (Collins et al. 2009: 1 cited in Guilherme 2014). Consequently, the definition of languages like Portuguese, Spanish, and even Guarani as glocal languages seems the most accurate perception of how they have organized themselves throughout time and across global borders. Not only have Portuguese and Spanish ranged all the way from localized empowered heirs of the Latin language to globalized, widespread phenomena, they have also been hosting impacts on local instances. More than that, the Guarani language, as a language of long-lasting resistance and significance of diffusion, is perceptibly a glocal language as well. All of them have been printing their traces on each other as well as re-signifying and reconstructing themselves while in contact with many other languages of the world. For this reason, I have called them “glocal” languages, using the composite word introduced by Robertson (’995), keeping in mind his statement that “the ‘global-local’ is more complex than an

76

Fernanda Martins Felix ‘action-reaction’ relationship”, since they have become global, deterritorialized, and again been reinvented locally. “Glocal” languages are, therefore, confronted with issues of power, as they compete with each other and subjugate other languages, at the intra-, inter-, and trans-national levels. (Guilherme 2014: 65)

Guarani, Portuguese, and Spanish have been pluralizing themselves all the way from their local starting points. They have been flowing across boundaries of space and time, accumulating epistemologies, knowledge, along the socio-historic conditions that co-produce them (De Souza 2014), and also undergoing dislocations, fractures, and re-significations. This dialogical, intensively dynamic process impacts not only across colonized and colonizer spaces, but also across and within nations. [Glocal languages] “translate” simultaneously different cultural baggage related to both native users as well as others. [T]hey refer to a plurality of equally valid native speaker’s models and encompass diverse non-native speakers’ legitimate performances. (. . .) Therefore teaching/learning of a “glocal” language implies a critical intercultural pedagogy that calls for a critical and conscientious use of linguistic tools and offers plentiful opportunities for critically active cosmopolitan citizenship (Guilherme 2002, 2007), while also making room for expansive cross- and inter-cultural savoirs (Byram 1997). (Guilherme 2018: 9–10) Speaking about diversity does not imply to speak only to those who are considered a majority or a minority. [Addressing society as a whole] does not imply a holistic concept of culture; on the contrary, it suggests that the intercultural character of life in society is not a matter only for those who are different, [. . .] but rather, it is the cultural diversity in society as a whole. In sum, it is everyone’s issue; that is, everyone is someone else’s Other. (Guilherme 2014: 62) That way, when speaking about languages teaching/learning, we might need to consider the “symbolic nature of transcultural competence” as it entails “a risky circulation of values across historical and ideological time scales, the negotiation of nonnegotiable identities and beliefs”. It demands “reflecting on the way that our and the Other’s realities mutually construct each other through symbolic systems” (Kramsch 2012 apud Guilherme 2014: 63). Dealing with linguistic relationships, policies, imperialism, and diversity is not simply analyzing and validating different linguistic status and

Glocalism Now and Then

77

positions in societies. It alludes to taking into consideration the symbolic experiences that surround every linguistic use, as speakers and their languages are constantly dwelling in the symbolic values of their linguistic performances. Therefore, seeking approaches that enable experiences of mutual validation, of meaning confrontation, that are structured over collaborative negotiation of meanings, and are constantly reinforcing the intrinsic and non-derogatory relationships amongst languages, is a challenging endeavour towards overcoming colonialism.

Intercomprehension and Numeric Platforms: Tools for the Battle Comprehending various languages, while keeping whenever possible your own language(s) to express yourself, has been a proposal of Intercomprehension approaches. As an area of studies, it finds its origins in Contrastive Linguistics. During the 1990s, it shifted towards language pedagogy, spreading and becoming a complex subject of interest, research, and action (Felix 2016). Doyé conceptualizes it as a “form of communication in which each person expresses in one’s own language and comprehends that of the other” (2005). Experiences of Intercomprehension inherently involve socio-political relations. With educational, emotional, or economic purposes, understanding the language of the Other may have implicit attitudes towards the opening for the Other, the valuing of different knowledges, the interest in relating to others. In the same way, making yourself comprehensible in your own language also presents relational, intercultural phenomena, as noticed, for example, in the efforts to conceive ways of making yourself clear, accessible (Felix 2016). In accordance to a critical framework proposed by Guilherme (2014) when dealing with intercultural speakers, the processes of Intercomprehension may well motivate “a reflective, exploratory, dialogical and active stance towards cultural knowledge and life that allows for dissonance, contradiction, and conflict as well as for consensus, concurrence, and transformation” (64). With the intention of defining Intercomprehension both in a direct and overarching way, it is valuable to make use of Degache’s (2006) formulation: “comprehend the language of the Other and make yourself comprehensible in your language[s]”.7 Carola points out the reason why this is fundamental (2015: 31): It covers two essential aspects to the notion of intercomprehension: the (written and oral) comprehension and the interaction—the latter here understood not only in the context of an encounter of two or

78

Fernanda Martins Felix more subjects, but in the context of the encounter of the subject with acknowledged languages or those in terms to be revealed, and in the sense built in this encounter.8

Intercomprehension practices may well be understood as possibilities to overcome the “waste of knowledge” (Sousa Santos 2004), as it offers an enabling atmosphere to the development of linguistic and social skills. Not only are we offered the possibility of being in touch with and learning (from) different languages, but also of rearranging our cultural and linguistic paradigms while re-signifying others and ourselves. As a communicative strategy, Intercomprehension settles over this quality or property of languages. A fundamental concept about Intercomprehension states that it benefits from interlinguistic contacts amongst related languages, successors of a common root (Éloy 2004). Éloy (2004) presents a series of categories for languages’ distance and proximity. He argues that complete mutual comprehension can occur— level zero—amongst the differentiations of the same language, the varieties of a language widely perceived as a single one. An example is the varieties of Portuguese, as in Brazilian Portuguese, European Portuguese. In addition, it refers to varieties within the “same language”, such as regional practices of Brazilian Portuguese, for example. Amongst different languages, it is also possible to find categories of proximity, as in languages from the same family. Also, according to the researcher, this closeness begins to fade as historical distance rises. This notion could be interpreted as in the distance between classic and contemporary Spanish, for example, or as in the distance between Portuguese and Sanskrit, on the other hand, languages that remotely share the Indo-European and have grown apart along history. Perspectives like this assume that because of the common origin, a speaker from a Romance language, for example, would be able to comprehend other languages that have derived from Latin. Accordingly, the resemblances and analogies that exist amongst languages from the same family are so numerous and profound that the differences do not break the fundamental unit that string them together. Languages are considered susceptible to Intercomprehension by the use of common lexis, by phonetic similitudes, by similar morphological and syntactic constructions, by sharing cultural matters (Felix 2016). Santos (2010) conceives this bond as either a real concept amongst them or a concept perceived by individuals. That is, languages may be perceived as proximate, as related, because they share a common ancestor, as Portuguese and Spanish, from Latin. as Also, their similarities and bonds may be an individual’s perceptive phenomenon. In education, Intercomprehension comprises locating the language apprentices in the centre of the learning process, where they engage in organizing, reorganizing, sharing, and accumulating knowledge, where

Glocalism Now and Then

79

they value the linguistic and cultural expertise of those involved and relativize the complexity of languages. Evidencing not only the mental processes, the cognitive activities, the purely linguistic knowledge, but also the valuing, the construction and the renewal of the knowledge from all the individuals, assigning an inexorable social character to Intercomprehension practices.9 (Felix 2016: 28) While working with related or non-related languages, Intercomprehension experiences intend to overcome the traditional teaching/learning process, in which previously known languages are underestimated pillars for the learning of a new one. It allows spaces of simultaneous learning and values different levels of abilities. The collaborative dynamics intrinsic to many Intercomprehension experiences are a solid track to emphasizing a “horizontal” hierarchy for speakers. As for languages, the validation of their similitude and the active approach to them as related and equally valuable also promotes this horizontality amongst languages. Attributing equal value to the use of languages—quite different from many language teaching/learning processes, when the use of a specific language is deliberately suppressed—practices of Intercomprehension aim a more equalized linguistic hierarchy. Bringing to attention the possibilities of closeness, the obscurities and many of the symbolic values ingrained in languages dissipate, offering an environment where languages can be unveiled. For involving a plural practice of language teaching/learning, Intercomprehension experiences frequently promote the specific development of one or a few linguistic proficiencies, prevailing the work with reading and writing on most of the European projects so far (Planet 2011). This approach promoted by Intercomprehension practices does not need to be seen as weak. There is no sense in saying one would remain a shallow knowledge holder if confronted by a plural and simultaneous languages pedagogy. Such approach goes beyond a fragmented knowledge of independent languages; it might as well indicate a perspective of fluid configuration of languages. Intercomprehension proposals usually make meaning on languages dynamics, boundaries, and their juxtapositions. They might as well be heading towards a notion of indeterminate languages, touching theories with translanguaging10 or even the complete subjectivity of languages (Pennycook 2006). The collaborative creation of spaces and moments of discovery reaches to comprehend bonds of similarity amongst languages, to get aware about your own linguistic and cultural knowledge as representative and useful for embracing a wider, new linguistic repertoire. This integrant proposal of most Intercomprehension experiences is understood within the terms of “éveil aux langues”, an “awareness of languages”, although

80

Fernanda Martins Felix

personally I prefer “awakening to languages”. Alluding to a clearing of the fog that surrounds languages, which, after all, are connected and yet have so many peculiarities. Many educational practices, and quite significant Intercomprehension practices, have been benefiting from the boom of the Internet as a space of diffusion and encounters for collective works. It allows the emergence of projects that, profiting from the web resources, produce new experiences in languages teaching/learning. An example is the development of numeric platforms directed towards practices of Intercomprehension by projects maintained by governmental and private initiatives (Felix 2016). The Galanet and the MIRIADi are two large projects that aim at the diffusion of Intercomprehension as a basis for the teaching and learning of languages. They have built virtual spaces organized to sustain, materially and theoretically, experiences of Intercomprehension amongst, not exclusively, speakers of different Romance languages. Based on the principles of hybrid education they had been proposed, primarily, to students and adults with competences on at least one Romance language, not necessarily initiated in any other. Specifically, for working with Intercomprehension among speakers of different languages, the local proximity of the participants of practices is usually a challenge. Initial works with hybrid pedagogy seem to offer a way so people from different locations can virtually meet, also inciting presence encounters and live group work. It conceives online and offline moments, both theoretically designed for profiting the most from each circumstances. Numeric platforms such as Galatea, Galapro, MIRIADi, and, why not, some social media, are instrumented with digital resources for language teaching/learning and can take place in both formats. The Galanet project began in 2001, and the platform remained available for work until 2015. It has been organized by the Socrates Lingua Group and financed by the European Commission. According to Capucho, such were the prerogatives of the platform: This platform [ . . . ] allows to speakers of different Romance languages the practice of intercomprehension, namely, a form of plurilinguistic communication where each one comprehends the languages of the others and expresses him/herself in the language(s) that one knows; this way developing the knowledge of such languages in different levels.11 (2011: 26) The MIRIADi project—Mutualisation et Innovation pour um Réseau de l’Intercompréhension à Distance—aimed to offer a wider prospect on language teaching/learning and on Intercomprehension training, and it would succeed the extinct Galanet project. It is also structured on

Glocalism Now and Then

81

Sessions, where groups of speakers arrange themselves and work collaboratively around activities directed to the discovery of languages’ similarities, to cultural encounters, and to the reformulation of paradigms. Although both platforms have a pedagogical emphasis and contemplate a wide range of activity possibilities, the MIRIADi platform was thought to overcome some of the proposals of the Galanet. With a more dynamic structure, in terms of contextualizing the organization of the Sessions and its resources specifically to the needs of the context in which it is being used, MIRIADi has been conceived as a sharp tool. It is composed by virtual pedagogical resources, like forums, chat rooms, data basis, groups, and personal profiles, which are intended to assemble phases of different interactions in a process of “awakening to languages”. Activities such as self-presentation, debates on themes to be discussed along the Session, sharing of pertinent data, and collective work for the production of linguistic and cultural materials are some of the possibilities of organization. MIRIADi has come up with a dynamic that allows the organizers of the Session to arrange its own flow, while Galanet had a less flexible structure, previously defined for the fulfilling of a Session’s four phases. Selma Martins (2014: 20) understands some of the objectives and possibilities of the Galanet project, which still underlie the MIRIADi project basis: It favors interactions amongst the apprentices of languages from distinct cultural contexts, allowing the development of the knowhow-to-do and the know-how-to-be in a cooperative and collaborative dynamic of learning. This kind of collaborative on-line learning reinforces the active participation of its actors, representing a space of cognitive and social development through the mediation amongst the subject, as one is responsible for the comprehension of the other, what provokes a feeling of safety and equality, favoring the motivation and engagement for learning.12 Most experiences shared on the platforms have been mainly about speakers of Romance languages, even though not exclusively. That requirement is quite delicate to impose. Since the recognition of a language as being accessible to oneself is not immediately evident, the width of intercomprehensibility has been stretched only up to a certain point. For educational purposes, the need to be familiar with at least one Romance language has been cautiously ensured, so that the organized activities could depend the least on deepening metalinguistic discussions. Documented experiences have, so far, reached a point where participants were able to work with non-related languages, stringing with shared lexis as a basal connection. That is the case of the InterRom Project, from

82

Fernanda Martins Felix

the Universidad Nacional de Córdoba—Argentina, with Romance languages and English, and the Guaranet Project (Felix 2016), where the Guarani language has been discussed by speakers of Portuguese, Spanish, and French. Nevertheless, the teaching/learning of a non-related language through Intercomprehension experiences situated on hybrid programs is still very incipient in terms of amount of experience and research. Far from fading the importance of actions on that direction, this shallowness reveals the exact opposite necessity—that of more attempts to reach the similar aspects amongst languages, to a point where they can be as comprehensible as their closeness allows. Finally, this presentation of the pedagogical platforms is a requirement for offering meanings to an understanding of how a social media network such as Facebook can be a pedagogical resource as well. The accessibility of this contemporary device is not only attractive because of its broad use, but for its allowance to a deliberate use for teaching/ learning uses: organizing a profile on Facebook, where people can meet virtually and engage on transcending experiences, facing themselves and the others, valuing their own knowledge and that from the others, putting in question all this knowledge and disposing for new paradigms on speakers contacts, on languages status, on teaching/learning, on world borders. The fluidity and the intrinsic diversity of experiences within such a space declare a pronunciation of the glocality of languages. Connected along time and crossing through space, they are always taking place from a local perspective, from a punctual moment and location. These concepts are, in a social media platform, easily perceived as dialogical, as the locality of phenomena in symbiosis with their globalism. Things that happen from a local perspective, which have already been touched by globalism, happen to flow gigantically and affect other localities, and so on. This dynamics detailed in Sousa Santos remarks (1999 cited in Guilherme 2018) can be evidently perceived in social media praxis.

Awaking to What Is Ours: A Practical Experience on Guarani, Portuguese, and Spanish This section will be organized in aligned phases. The structure is based on the idea of a Session, from Galanet and MIRIADi experiences, where, for a certain period, people get to work together through different, oriented but yet dynamic and flexible activities. The encounters can be exclusively virtual or hybrid, when people can both meet face-to-face, engaging on different dynamics, while simultaneously using the platform. An awakening to, a sensitization for, ultimately, the teaching/learning of “glocal” languages, namely Portuguese, Spanish, and Guarani, might be pointed as main objectives for such an experience. As well as with

Glocalism Now and Then

83

the duration, specific objectives will match the context of each singular implementation. They may vary according to available time, to the level of acquaintance among the participants, to their needs or interests in working for more or less time in one phase or another, among other context specificities. The development of specific abilities of linguistic performance, an endeavour aimed to discover resemblances or differences amongst those languages in question, or critical reflection on the languages in question, are some examples of what might be identified as a guideline for activity and discussion proposals. As for this particular experience, hypothetically located in South America, detailed objectives are focused on reflections on the glocality of the languages in question and the points in which they touch each other. However, they are re-presented, along the work plan and when pertinent, by their localization. The public is also rather broad. A group of students from a university degree, a group of teachers in a teacher training programme, high school students, groups of speakers of the selected languages from different sites. It will depend in which context it is to be applied. This text will focus on a group of language students participating in a workshop. They are speakers of Romance languages with the intention of getting in contact with Portuguese, Spanish, and Guarani. It might be also particularly interesting to have speakers of Guarani amongst them. As work phases do not depend on specific timings, they can be estimated and rearranged, depending on the possibilities of each specific scenario. The workshop hereby proposed was intended to last two weeks, be completely virtual, and be developed in synchronized or asynchronized patterns. The creation of an interest group in Facebook is one first step in the use of this platform as a place of work. It will unfold a series of resources available for fulfilling the purposes already mentioned. Furthermore, the creation of a group for inbox messaging is another possibility for more fluid communication. As it is not sectioned, which means that messages line up and can be posted simultaneously, it can work as a less structured forum, when compared to posts on the page’s main wall. Therefore, on the wall, all those who are allowed in create messages and they are organized in posts that can be subdivided, creating niches of debates. In addition, the possibility of video hangouts is easily available and allows participants to engage in live conversations, each one from one’s local point on the globe. On the wall or by inbox messaging, many kinds of digitalized material, such as texts, videos, pictures, images, links, and external files, can be shared. That is one of the main aggregated values for using Facebook for educational purposes. A very wide range of material can be shared in a very accessible, easy way.

84

Fernanda Martins Felix

After the creation of the page on the platform, the following steps can take place: Phase 1: Ice-Breaking Objectives: Self-presentations; objectives discussion and first agreements. Discussions: By making use of the communication resources provided on the platform, participants use their own languages to present themselves, to discuss the structure proposed and therefore begin the first linguistic contacts. A “linguistic biography” can be an activity that is intended not only to present the linguistic background of each participant, but be a way of one realizing more effectively all the linguistic knowledge already possessed. It can be a simple form, structured previously, that attempts to cover all relevant linguistic contacts made by the participants or an open demand for reflecting on and sharing information about the participants’ personal linguistic experiences. Phase 2: Brainstorming Objectives: Discussing the main theme; discussing sub-themes; and organizing group(s) of work. Discussions: This is usually one of the phases with the longest duration, as linguistic contacts begin but also are already put in use for negotiation of ideas, contents, and meanings. Also making use of wall messages, of inbox mail, or video calls, participants engage in using their languages to discuss a main theme for them to work on as the experience unfolds. The use of their own languages will already begin to appear as an exercise for all the purposes discussed along this chapter and will reflect on how the participants work for choosing a main theme. This main theme will guide the following of the discussions and will chain together the inter-communications. For this example of experience, “The glocality of Portuguese, Spanish and Guarani” will be specified as the main theme for the workshop and it can be sectioned into “global reaches and local contexts”, “peculiarities of each language or speakers participating”; “resemblances”; and “how socio-historical contacts have merged them”. The whole group can discuss these themes, or they can be delegated for sub-groups, which will engage in discussing them within their own borders. All that will depend on the contextual objectives and possibilities. Participants will be active in discussing the topics scheduled but will also be constantly dwelling on their linguistic transactions. They will be constantly trying to make themselves comprehensible, while trying to

Glocalism Now and Then

85

understand the others. That is a very intricate process, which cannot be predicted. Phase 3: Collecting Data Objectives: Collecting and sharing linguistic and thematic materials. Discussions: Although linguistic data and information on the themes have been shared already, this phase is usually when the discussions go deeper, and linguistic experiences are expected to go along with other thematic learnings. The sharing of information with the intention of learning the language of the other through the discussion of a theme is the thick moment of the process. While the group, or groups, searches many source materials to be shared, put into debate, questioned, reconfigured, analyzed, the theme in question is explored and the participants remain on the quest for Intercomprehension. Previously elaborated activities can be an example of shared material. Information about the languages, exercises on language comprehension, on language production and even as a form of evaluation, all this can be prepared in advance and proposed to the participants, in a most pedagogically structured moment. Digitalized materials on language teaching/ learning can provide language support for more clearness in language comprehension, while simultaneously, collective, organic productions give form to the process. Phase 4: Press Pack Objectives: Production of collaborative material, aimed to summarize the experiences of the participants. Discussion: The allusion to the press pack is for emphasizing the action of producing collective material, capable of representing a little bit of what has been apprehended along the workshop and that is posted, presented, and discussed. They are produced by groups or by the whole group; usually it is when participants generate text documents and multi-media materials that are a collection of linguistic and information production. Summarizing their experiences, participants elaborate a collective document, usually multilingual and negotiated in every aspect.

Conclusions From a factual point of view, Guarani, Spanish, and Portuguese all have overcome colonization. They had localisms and globalisms, in different scales, reconfiguring them, mainly after their first contacts. This indigenous language and both Romance languages in question, especially in Latin America, but definitely worldwide, have permanently enlaced

86

Fernanda Martins Felix

themselves. None of them went through colonialism, followed by postcolonialism, immune from each other or from other world languages; that would be a radical yet existing purpose in coloniality. The social positions of the Tupi language right before European colonizers arrived were powerful, even to a point where we could discuss its “globality” way before its contact with other imperialistic languages. It was powerful enough that it has not only survived but also deeply incorporated itself into languages such as Portuguese and Spanish. Guarani, as a significant heir of the Tupi language, exhibits an outstanding ability to maintain and expand its events of practice and all the knowledge along them, which happens, as well, with Romance languages, Portuguese and Spanish. This is a reference to the observation of the glocal dynamics of these languages, which, in constant move, reconfigure and sustain themselves. Finally, working along non-hegemonic forms of reason can allow speakers of different languages to enrich themselves linguistically and culturally, offering experiences in which we can confront, negotiate, and overcome paradigms. Intercomprehension as a socio-pedagogical proposal, directed to language teaching/learning, more specifically glocal languages, reveals one of many ways through which we can provoke experiences such as the ones described above. As for the use of social media as educational tools, it has been widely perceived that these globalized phenomena have plenty to offer.

Notes 1. www.galanet.net 2. www.miriadi.net 3. www.facebook.com 4. Translation by the author. “[. . .] muitas vezes formadas por ‘impérios’ agora desaparecidos”. 5. Translation by the author. “tan universal que domina ambos mares, el del Sur por todo el Brasil, y los dos grandes ríos, el de la Plata y el gran Marañón”. 6. Translation by the author. “a estabilidade que lhes permitiria expandir-se no espaço e sobreviver por longo tempo”. 7. Translation by the author: “compreender a língua do outro e se fazer compreender na sua língua”. 8. Translation by the author: “ela aborda dois aspectos essenciais à noção de intercompreensão: a compreensão (escrita e oral) e a interação—esta última compreendida aqui não somente no contexto do encontro de dois ou mais sujeitos, mas no contexto do encontro do sujeito com as línguas conhecidas ou em fase de descoberta, e no sentido construído neste encontro”. 9. Translation by the author. “[E]videnciar não apenas os processos mentais, as atividades cognitivas, os conhecimentos puramente linguísticos, mas também a valorização, a construção e a renovação dos conhecimentos de todos os indivíduos, atribuindo um caráter inexoravelmente social às práticas de Intercompreensão”. 10. Canagarajah (2011, 2017); Pennycook and Makoni (2006); De Souza (2014). 11. Translation by the author: “Essa plataforma [ . . . ] permite aos locutores de diferentes línguas românicas a prática da intercompreensão, a saber, uma

Glocalism Now and Then

87

forma de comunicação plurilíngue onde cada um compreende as línguas dos outros e se exprime na ou nas línguas românicas que conhece, desenvolvendo assim em diferentes níveis o conhecimento dessas línguas”. 12. Translation by the author: “favorece interações entre os aprendizes de línguas de contextos culturais distintos, permitindo o desenvolvimento do saber-fazer e do saber-ser em uma dinâmica de aprendizagem cooperativa e colaborativa. Esse tipo de aprendizagem colaborativa online reforça a participação ativa de seus atores, representando um espaço de desenvolvimento cognitivo e social por meio da mediação entre os sujeitos, uma vez que cada um é responsável pela compreensão de outrem, o que provoca um sentimento de segurança e igualdade, favorecendo a motivação e o engajamento pela aprendizagem”.

References Bhabha, H. K. (1986) Introduction to the English edition of Black Skin White Mask. In Fanon, F. & Markmann, C. L. (eds.) Black skin, white masks. London: Pluto Press. Byram, M. (1997) Teaching and assessing intercultural communicative competence. Clevedon: Multilingual Matters. Canagarajah, S. (2017) Translingual practices and neoliberal policies. Switzerland: Springer International Publishing. Canagarajah, S. (2013) Translingual practice: Global Englishes and cosmopolitan relations. Routledge: London. Canagarajah, S. (2011). Translanguaging in the classroom: Emerging issues for research and pedagogy. Applied Linguistics Review, 2, pp. 1–28. Canclini, N. G (2005) Hybrid Cultures: Strategies for entering and leaving modernity. Minneapolis: University of Minnesota Press. Capucho, F. (2011) L’intercompréhension: l’inovation decline au passé, au présent et au futur. REDINTER-Intercompreensão, Chamusca, Edições Cosmos/REDINTER, 3, 15–36. Carocha, C. H. (2015) Práticas de Intercompreensão entre línguas românicas: desenvolvimento da competência de leitura plurilíngue em estudantes do Ensino Médio de uma Escola Técnica (ETEC) de São Paulo. Master’s Thesis. Universidade de São Paulo. Collins, J., Slembrouck, S. & Baynham, M. (2009) Introduction. In J. Collins, S. Slembrouck & M. Baynham (eds.) Globalization and Language in Contact: Scale, Migration and Communicative Practices (pp. 1–16). London: Continuum. Degache, C. (2006) Didatique du plurilinguisme. Travaux sur l’intercompréhension et l’utilisation des technologies pour l’apprendissage des langues. In Synthèse de la activité de recherche. Grenoble: LIDILEM/Université Stendhal Grenoble III. De Oliveira, G. M. (2000) Brasileiro fala português: Monolinguismo e Preconceito Linguístico. In Silva, F. L & Moura, M. M. (eds) O direito à fala: a questão do preconceito linguístico (pp. 83–92) Florianópolis: Editora Insular. De Souza, L M. T. M. (2014) Epistemic Diversity, Lazy Reason and Ethical Translation in Post-Colonial Contexts: The Case of Indigenous Educational Policy in Brazil. Interfaces Brasil/Canadá, 14, 2, 36–60. Dietrich, W. (2010) O Guarani Paraguaio. Presentation Website. Available in: www.etnolinguistica.org/lingua:guarani-paraguaio Doyé, P. (2005) Intercomprehension: Guide for the Development of Language Education Policies in Europe: From Linguistic Diversity to Plurilingual Education. Reference Study. Strasbourg: Council of Europe.

88

Fernanda Martins Felix

Éloy, J. (2004) Langues proches: que signifi e de les enseigner. In ÉLA-Accès aux langues proches et aux langues voisines (pp. 393–401). Paris: Didier Érudition. Felix, F. M. (2016) Guaranet: Experiências de contato e intercompreensão em Guarani, Português, Espanhol e Francês. Master’s Thesis. Universidade Federal do Paraná. Available in: http://calvados.c3sl.ufpr.br/bitstream/ handle/1884/45946/R%20-%20D%20-%20FERNANDA%20MARTINS %20FELIX.pdf?sequence=1&isAllowed=y Fernández Jaen, J. (2006) El latin em Hispania: la romanización de la península Ibérica. Alicante: Biblioteca Virtual Miguel de Cervantes. Available in: www. cervantesvirtual.com/obra/el-latn-en-hispania-la-romanizacin-de-la-pennsulaibrica-el-latn-vulgar-particularidades-del-latn-hispnico-0/ Gandhi, L. (1998) Postcolonial theory: A critical introduction. Crows Nest: Allen & Unwin. Guilherme, M. (2018) ‘Glocal Languages’: The ‘Globalness’ and the ‘Localness’ of World Languages. In Coffey, S. & Wingate, U. (eds.) New Directions in Foreign Language Education (pp. 79–96). London: Routledge. Guilherme, M. (2014) ‘Glocal’ Languages and North-South Epistemologies: Plurilingual and Intercultural Relationships. In Teodoro, A. & Guilherme, M. (eds.) European and Latin American Higher Education between Mirrors: Conceptual Framework and Policies of Equity and Social Cohesion (pp. 55–72). Rotterdam: Sense Publishers. Guilherme, M. (2007). English as a Global Language and Education for Cosmopolitan Citizenship. Language and Intercultural Communication, 7 (1), 72–79. Guilherme, M. (2002). Critical Citizens for an intercultural world: Foreign language education as cultural politics. Clevedon: Multilingual Matters. Guirardello, R. A. (1999) Reference Grammar of Trumai. Doctoral Thesis, Linguistic Department. Houston: Rice University, 1999. Instituto Brasileiro de Geografia e Estatística-IBGE. (2010) Censo Demográfico: Características Gerais dos Indígenas (pp. 1–245). Rio de Janeiro: IBGE. Instituto Cervantes. (2016) El Español, uma lengua viva. Departamento de Comunicación Digital del Instituto Cervantes (Ed.). Available in: www.cervantes.es/ imagenes/File/prensa/EspanolLenguaViva16.pdf Kramsch, C. (2012) Theorizing Translingual/Transcultural Competence. In G. S. Levine & A. Phipps (eds.) Critical and Intercultural Theory and Language Pedagogy (pp. 15–31). Boston, MA: Heinle. Lagorio, C. A. & Freire, J. R. B. (2014) Aryon Rodrigues e as Linguas Gerais na historiografia linguística. D.E.L.T.A., 30 especial, 571–589. Lewis, M., Simons, G. F. & Fennig, C. D. (eds). (2013) Ethnologue: Languages of the world. Dallas: SIL International. Makoni, S., & Pennycook, A. (2006) Disinventing and Reconstituting Languages. Clevedon: Multilingual Matters. Martins, S. A. (2014) A intercompreensão de línguas românicas: proposta propulsora de uma educação plurilíngue. Revista Moara, 42, 7–12. Meliá, B.(1987) O Guarani. Uma bibliografia etnológica. Santo Ângelo: Fundames. Mignolo, W. (2007) Coloniality and modernity/rationality, Cultural Studies, 21, 2–3, 155–67. Mignolo, W. (2000) Coloniality of power, ethnocentrism, and Latin America. Nepantla, 1 (3), 533–580. Moore, D., Galucio, A. V. and Gabas, JR., N (2008) O desafio de documentar e preservar as línguas amazônicas. Scientific American (Brasil), 3, Amazônia (A Floresta e o Futuro), 36–46.

Glocalism Now and Then

89

Moreira, C. M. (2005) A influência do Tupi na formação do português do Brasil. In Mesa-redonda, II CLUERJ. Rio de Janeiro: CiFEFiL. Pennycook, A. (2006) Uma linguística aplicada transgressiva. In Lopes, L. P. M. (org.) Por uma linguística aplicada Indisciplinar (pp. 67–84). São Paulo: Parábola Editorial. Peralta, J. J. (2008) Princípios e Diretrizes do Pacto Lusófono Mundial. Disponível em: www.portaldalusofonia.com.br/pactoprincipios.html Planet, M. T. (2011) Les perspectives multimodales de l’intercompréhension aujourd’hui, ou différentes manières de ‘[partir] de sa langue [et de as culture] pour aller vers celle[s] des autres’. REDINTER-Intercompreensão, Chamusca, Edições Cosmos/REDINTER, 3, 51–62. Rodrigues, A. D. (1993) Línguas Indígenas: 500 anos de descobertas e perdas. Ciência Hoje, 16, 95. Rodrigues, A. D. (2000) Panorama das Línguas Indígenas da Amazônia. In Queixalós, F. & Renault-Lescure, O. (orgs.) As línguas amazônicas hoje (pp. 15–28). São Paulo: IRD/ISA/MPEG. Santos, B. S. (1999) Towards a multicultural conception of human rights. In Featherstone, M. (ed.) Spaces of Culture: City, Nation, World (pp. 214–229). London: Sage. Santos, B. S. (2004) A Critique of Lazy Reason: Against the Waste of Experience. In Wallerstein, I. (ed.) The Modern World-System in the Longue Durée (pp. 157–197). London: Paradigm Publishers. Santos, L. (2010) Defining Intercomprehension: Is a Consensus Essential? REDINTER-Intercompreensão, Chamusca, Edições Cosmos/REDINTER, 1, 29–46. Silvetti, J. & Silvestri, G. (2015) Shaping the Guarani Territory. Harvard Review of Latin America. DRCLAS, 14, 3, 2–6. Spivak, G. (1990) The post-colonial critique: Interviews, strategies, dialogues. New York: Routledge. Teyssier, P. (1982) História da Língua Portuguesa. São Paulo: Martins Fontes.

4

Reshuffling Conceptual Cards What Counts as Language in Lowland Indigenous South America Jamille Pinheiro Dias

Introduction This chapter discusses communicative practices and potential notions of language in lowland Indigenous South America by taking two key dimensions into account: that of intersemiotic relations involving image, words, and music, and that of affect that takes place beyond the human/ non-human divide. Drawing mainly on applied linguistics and South American Indigenous ethnology, it provides a review of examples coming from ethnographic studies in an effort to emphasise how the co-presence and simultaneous engagement of different verbal and non-verbal components by Indigenous peoples in the region—as well as their particular dynamics of affect between people, animals, plants, spirits, and a range of other characters that may not be humans for us, but that are human for themselves—call for a broader understanding of what counts as “a language,” primarily understood as a signifying system used by a specific group of people, and “language,” generally seen as the model of signifying systems. The intention here is not to give an exhaustive exposition of the topic, but to draw attention to its importance and to work out possible grounds for further investigation. Considering that the very idea of “language”—identified and delimited in Western ideologies of linguistic standardization—corresponds to an invention arising from seventeenth- and eighteenth-century coloniality and nationalism, the assumption that such a notion, as a preconceived, independent object, is readily transposable to all locations and populations, is questionable from the very start. In this sense, I resort to the notion of “root metaphor” as discussed by anthropologist Marilyn Strathern (1988: 134) in order to address the category of “language” that is widely used as an operator of our relationship with multiple communicative practices. Strathern’s approach to root metaphors can help us understand how, from our exogenous frame of reference, we project the notion of “language” onto Indigenous peoples, entailing assumptions that interfere with and limit our understanding of their communicative practices. By drawing on Strathern’s perspective, I propose that a

Reshuffling Conceptual Cards

91

critical and self-reflexive dialogue between applied linguistics and South American Indigenous ethnology, operating at both local and global levels, involves being aware of, and calling into question, root metaphors—such as “language”—that underlie the modes of thinking and convictions of each discipline. In other words, it calls for a “reshuffling of our conceptual cards,” to quote from Eduardo Viveiros de Castro (2002/2003: 348).

Towards a More Inclusive Approach to Language Reducing language to a communication device, a vehicle for messages expressible in an ideally abstract verbal code, or a system for naming or reflecting a reality which supposedly precedes it, rules out an array of local communicative practices that resonate closely with Deleuze’s approach to language as comprising, among other aspects, the sensorial, the kinesic, and the verbal, as it engages the entirety of an affective framework. As we will see in this chapter, if one is to speak of language among non-modern humankinds such as the Amerindians, working towards a more inclusive approach to what language is or, rather, to what language does, is inescapable. The notion we inherited from Aristotle1 that man alone, among all living beings, is endowed with the capacity for speech that makes political life possible lies ingrained in the Western historical presumption of human exceptionalism and superiority. We are also heirs to the Cartesian agenda of rationality and its ascription of language to human beings.2 Considering the underpinnings of this view of language, Robert B. Louden says that From at least the time of Aristotle, philosophers and scientists have repeatedly pointed to the faculty of language as one of the primary differentia between humans and nonhumans. Language, perhaps because its presence or absence seems more easily detectable than other alleged differentia such as rationality or consciousness, is, as Mary Midgley remarks, “possibly our favourite human distinguishing mark.”3 This leads us to one of the cornerstones of Giorgio Agamben’s wellknown investigation of the way humankind was dignified and privileged over other species on the basis of the presumption that language is the defining attribute of the properly human mode of existence.4 The construction of humanity through the acquisition of language, as he shows, extracted the human out of animality, constituting what he calls the anthropological machine—an apparatus that divides life into human and non-human. The ethical implications of the perception that language exists, and that the human is the living being who has it, highlight the need to consider what counts as language. What might sound

92

Jamille Pinheiro Dias

indiscernible and like inarticulate noise to some might count as language to others, especially if we are to conceptualise language, in a DeleuzoGuattarian way, as affect rather than information or representation. That said, this leads us back to our case in point, and to question, then, what counts as language in lowland Indigenous South America. Before outlining a review of pertinent ethnographic studies, though, let us summarise some of the ways applied linguistics has been examining and questioning hegemonising conceptions of language.

Disinventing Language as a Root-Metaphor The cultural specificity of the notion of “a language” was highlighted by Peter Mühlhäusler (2000, 2002) in his discussion of the issue of Western linguistic imperialism in the Pacific. It is problematic to reify “a language,” as Mühlhäusler demonstrates, since this gesture abstracts languages from the complex ecology of forms of communication in which they are embedded. Also, his criticism draws attention to the association between the notion of “a language” and “the rise of the European nation states and the Enlightenment,” and to how referring to “a language” “makes little sense in most traditional societies” (2000: 358). In his study of the development of the Tok Pisin pidgin in Papua New Guinea, for instance, Mühlhäusler has shown how it would reunite a wide array of linguistic varieties under the same name; how it would borrow lexical items from German or English; and how neighbouring languages would borrow lexical items from it. Cases such as that of Tok Pisin indicate that the neutrality of a supposedly countable phenomenon called language must be called into question. Along similar lines, Richard Bauman and Charles Briggs explain how languages “came into being” (2003: 7) through their conceptualization and division into bounded entities, detachable from their environment, by the modernist metadiscursive regimes used to describe them. The deprovincialisation of Euro-American5 linguistic ideologies naturalised certain metadiscursive regimes (30), which were determinant to the widespread construction of languages as unified systems—giving rise, as Bauman and Briggs argue, to “a powerful means of creating social inequality” (9). In particular, they discuss the ways in which words of Others were approached from the early nineteenth century onwards, through the collection, management, and publication of their “oral traditional texts” (15). If this appropriative intervention was key to how modernity was symbolically shaped, we are now in a position of not allowing “constructions of language and tradition masquerade as cartographies of the real” (317). In the spirit of Bauman and Briggs, Sinfree Makoni and Alastair Pennycook state that “languages do not exist as real entities in the world and neither do they emerge from or represent real environments; they

Reshuffling Conceptual Cards

93

are, by contrast, the inventions of social, cultural and political movements” (2007: 2). They point out the need for unpacking the metadiscursive regimes that have underpinned the classification, naming, and invention of languages. Makoni and Pennycook support the view that if languages were invented, we need to imagine ways of “disinventing” and “reconstituting” them through an acknowledgement of fluidity in the concrete workings of language at the local level. Their suggestion for challenging an ideology of languages as fixed and enumerative is to embrace actual communicative practices. Bearing that in mind, they argue that “we need to understand the interrelationships among metadiscursive regimes, language inventions, colonial history, language effects, alternative ways of understanding language and strategies of disinvention and reconstitution” (4). Makoni and Pennycook (2012) argue that languages are discursive and social constructions that foster inequalities and serve political purposes, being frequently used as instruments of domination that have material effects on the living conditions of people. This theoretical framework, therefore, has emphasised the problem of hegemonising and colonialist conceptions of language. Makoni and Pennycook’s critical historiographical approach demonstrates that languages were invented through a process of classification and naming (1). For instance, they highlight how languages such as “Bengali” e “Assamese” were constructed as “new objects” (10). That said, since local knowledge is crucial to our understanding of language, this leads to a call for more research allowing us to develop a more localised understanding of the relationships between what people think about their language (or about the language of other people), their situated forms of speech, and the material—social, economic, environmental—effects of such perspectives and uses. In this respect, Makoni and Meinhof (2006) argue that it is essential that conceptions that speakers have of their own language are taken into consideration, even if those conceptions seem to contradict scientific data. In their study of Applied Linguistics in Africa, they argue for the importance of verifying how discourses about “languages” are understood by the ones who use them, in so far as African experiences with language and linguistic practice in the content may diverge from the imposed notion of language as a distinct entity (196). Since Makoni and Meinhof share the understanding that “the notion of ‘language’ as a marker of social identity did not exist prior to colonisation and the introduction of Christian evangelism and literacy in Africa,” the “outside” perspective of linguists may not necessarily coincide with the “inside” perspective of speakers (208). Taking the “inside” perspective of speakers seriously poses the challenge of putting aside the tendency to disqualify the specificity of other modes of existence on the basis of a supposed scientific universalism.

94

Jamille Pinheiro Dias

Additionally, Makoni and Mashiri (2007) suggest that instead of developing linguistic policies that isolate languages hermetically, we should describe the use of vernaculars that come into contact and influence each other in order to understand the social realities of their users. They argue that proposing a world in which plurality is privileged over singularity requires rethinking conceptions based on notions of uniformity, and giving preference to those based on diversity. Placing an emphasis on the speaker, not on the code or the system, is decisive for this argument, since for many people the question is not whether they are monolingual or multilingual, but whether they actually use languages. Thus, Makoni and Pennycook provide a powerful questioning of assumptions about how languages are encapsulated in regimes that have become naturalised by the colonial project. They suggest that if a supposed scientific neutrality of linguistic knowledge characterises native speakers as idealised and language as an autonomous entity, engaging in a more symmetric dialogue with the users of multiple minority languages leads us to a necessary disinvention of “language” as a root-metaphor that guides our thinking about communicative practices. As such studies in applied linguistics remind us, linguistics has long served the “interests and politics of missionaries and colonial administrators” (Makoni 2003: 136), and it certainly runs the risk of reinforcing power asymmetries through its practices. But this is not always necessarily the case today. Linguistics can also be a locus of resistance, empowerment, creativity, and positive change. Indigenous peoples and their allies can mobilise linguistics against coloniality. Think of Indigenous linguists like Mutuá Mehinaku (2010), for instance, who are interested in studying and documenting their languages for the best interest of their communities (Franchetto 2004: 43). Nowadays, we are even in the position of mobilising linguistics against linguistics: in other words, of meta-methodologically turning it against its colonial legacy in a way that allows for a deep look into questions of power, dominance, privilege, and ideological and religious bias in language.

The Flesh of Language As I write, a railway cuts through portions of Awá Guajá lands in the state of Maranhão. Once again, Awá Guajá men will fail to hunt howler monkeys. The hunting relationship depends on sonic features: hearing, echoing, and deceiving the monkeys. In other words, the Awá Guajá hunt by mimicking. This is constantly threatened by the noise coming from the railway. It might be that no game will be brought home by night.6 This relates to how, in Indigenous Amazonia, “sound is the flesh of language,” to borrow from Charles Bernstein (1998). From the “strictly” linguistic perspective, though, some sounds count more than others: those associated with articulate language and humanness. At

Reshuffling Conceptual Cards

95

all events, there are multiple instances in which sound mimicry, echo, repetition, and reiteration make sense and make life in Indigenous sonic regimes in Amazonia. Numerous accounts and comparative analysis of the sonic regimes of Indigenous peoples in South America put flesh on this statement. As musicologists Bernd Brabec de Mori and Anthony Seeger say (2013), Indigenous Amazonian songs play a central role in rituals involving nonhuman knowledge, since they are a means of communication between humans and non-humans. Also, even if non-humans usually might not understand speech, they do understand songs. People with a trained ear can listen to non-humans and bring their knowledge into their own music. These are people who have mastered “world hearing,” as Brazilian ethnomusicologist Rafael José de Menezes Bastos—who has been working with the Kamayurá in the Upper Xingu region, in the state of Mato Grosso, Brazil, since the 1960s—calls it. Over the course of decades of fieldwork, Menezes Bastos has studied this topic in depth. In 1981, he was crossing Ipavu Lake by canoe with Eweka Kamayurá, an Indigenous master of “world hearing”—as Menezes Bastos would later describe him. Menezes Bastos recalls (2013a: 287): We talked, rowing, under a beautiful sunset. Suddenly Eweka stopped speaking and rowing, fell silent and asked me to be quiet too, gesturing towards the bottom of the lake. Whispering he told me to listen to what was coming from down below. Despite my best efforts, I heard nothing from the watery depths. He said to me insistently: “Can’t you hear the fish singing? Listen, listen . . . ”. I heard nothing. This went on for several minutes. Later, back in the village, I concluded that Eweka had experienced some kind of hallucination, a fit of poetic inspiration or holy ecstasy, the whole event just a flight of imagination. I recall that some days later he simply told me that I needed to train my hearing. It was not until years later that Menezes Bastos became acquainted with recordings of the “acoustic behaviour” of fish in an exhibition at the Federal University of Santa Catarina (288). This was how he realised how skilful Eweka was as a listener, to the extent of being able to perceive fish songs. By sharing this account, Menezes Bastos drew attention to another aspect related to how music is accessed among Indigenous societies in lowland South America: the way in which they become audible, which has to do with the development of an expertise in “world hearing.” In this sense, the Kamayurá term “anup,” which evokes both “hearing” and “understanding,” indicates that “world hearing” demonstrates an astounding ability for comprehension. In his turn, Yanomami shaman and leader Davi Kopenawa, in his elaborate autobiography The Falling Sky—Words of a Yanomami Shaman,

96

Jamille Pinheiro Dias

co-produced with French ethnologist Bruce Albert, tells of how he began to sing by listening to the voices of the shamanic spirits known as xapiri (2012: 89–90): At the moment the xapiri finally reveal their voices, your fear vanishes and you experience an intense bliss. (. . .) This is how I began to sing, despite all my fears! (. . .) I decided to answer the xapiri’s voices by echoing them. (. . .) Their voices seemed perfectly discernible to me. Satisfied, I applied myself to imitating their sound and their words, again and again, without stopping. Seeing my efforts, the xapiri came to my assistance. They told themselves: “He probably doesn’t hear us well! Let’s start over! What can we do so our songs become audible to him?” Then they start to sing again, making our voices stronger. This is how I finally heard them and began to sing like them. If we try hard to answer the spirits, (. . .) they lend us their throats and reinforce our tongues. This way, the words of the xapiri’s song rapidly increase within us in a tape recorder. Having mastered how to echo xapiri spirits, Yanomami men are ready to engage in a ritual that starts when one man begins chanting softly and asks another man to repeat his words. They take turns, chanting very rapidly and echoing each other, sometimes all night long, until the sun comes up.

“Altership” and Affect Beyond the Human/ Non-Human Divide Indigenous specialists on world hearing such as Eweka Kamayurá and Davi Kopenawa have mastered the practice of listening to apparently inaudible nonhuman voices that remain unknown or ignored within distributions of the sensible proper to Euro-American frames of reference. In The Politics of Aesthetics and Aesthetics and its Discontents, French philosopher Rancière defines the distribution of the sensible as both “the system of a priori forms determining what presents itself to sense experience” (2008: 13) and the way in which “the practices and forms of the visibility of art . . . distribute spaces and times, subjects and objects, the common and the singular” (2009: 25). Taking seriously that listening to fish songs and to the xapiri is possible puts into question a priori distributions of the sensible brought about the human/non-human divide in relation to communicative practices. This awareness calls into question the tendency to disqualify the specificity of Indigenous communicative practices with derogatory and colonially condescending common places such as “It’s their belief,” which imply there’s one reality, and then true and untrue versions of it, the untrue versions being the ones that should be merely tolerated.

Reshuffling Conceptual Cards

97

Affect beyond the human/non-human divide also takes place among the Yanesha of Peruvian Amazonia, as Fernando Santos-Granero reports: “It is only after having heard the same song ten times that an apprentice must begin to repeat it and learn it, which consists on attracting the animal that owns the song, making it into a friend and spiritual protector” (2006: 113). In addition, Deise Lucy Oliveira Montardo (2002: 45) tells us about how Guarani songs are not created by them but come from elsewhere: The Guarani do not consider themselves masters of songs. Even individual songs personally received by each one of them in dreams are received by merit, as a gift. They are not composed by the person. The person listens to them. Their idea is that music already exists elsewhere. This has to do with the idea of “altership” proposed by Marc Brightman, Carlos Fausto, and Vanessa Grotti (2016: 20) in order to emphasise how humans in Indigenous Amazonia are not exactly creators in the sense of an individual author, but rather alterers, “capable of othering themselves and switching perspectives in order to appropriate new songs and new names” (21). By presenting that proposal, the ethnologists are interested in making explicit the contrast between the individuation of property in Western property relations and the multiplicity of alterities that constitutes forms of ownership in Indigenous Amazonia. This is not the same as saying that a notion that could be related to a property regime does not exist among Indigenous groups in the region (also when it comes to the complex relationships they establish with the State and market economy), but that the specificity of the idea of Western property as the institutionalisation of ownership, predicated on assumptions of individualism and possession, needs to be acknowledged—in other words, its presumed universality has to be challenged. It cannot be simply generalised to lowland Indigenous South America, where “personhood extends far beyond the human,” “the subject-object distinction is by definition fuzzy (even inapplicable)” (11), and ownership is above all a relational phenomenon which has more to do with rights over relationships with others—be them persons, parts of persons, or things— than with people with respect to things and ideas (8). The appropriation of new songs and new names among lowland Indigenous South American peoples is often made possible through a rigorous process of physical and intellectual learning, as well as through negotiations with nonhuman beings, song owners,7 and master spirits from different realms of the cosmos. If an idea of “authorship” is to emerge here, it will differ radically from the “gene r ic image of primitive collectivism and from some of the other commonplaces brought by Western thought,” as argued by Pedro de Niemeyer Cesarino (2010: 150),

98

Jamille Pinheiro Dias

who has been conducting a long-term ethnographic research among the Marubo, a Panoan people from the Javari Valley in Western Amazonas, Northwestern Brazil. Among Marubo shamans, what real l y matters is an ability to make connections with a “virtual field of knowledge” (162) that is neither individual nor internal; what really matters is mediating, with effectiveness and eloquence, words that come fr o m elsewhere— for example, recursively embedding voices of differe n t persons within one another. The web of relationships from which a shamanic utterance unfolds is an important indication that Amerindian m u ltivocal “altership” distances itself from the image of individual creativity that permeates much of our repertoire in literature and other arts.8

Intersemiotic Relations In lowland Indigenous South America, ritual songs r e late to manifestations that involve non-verbal features such as mu s ic, gesture, dance, graphic patterns, among others, as Pierre Déléage explains (2012: 114). The association with modes of expression distinct from the song itself, of which Déléage talks about, is even more revealing of the reductions and constraints that such modes of expression undergo when approached in a strictly linguistic manner: Numerous rituals add various modes of expression to the songs, which are distinct from them: instrumental music, gesture, dance, artefacts, the organization of the space and chronology of the ritual etc. Each of these modes of expression comes from a particular form of learning, sometimes temporally distinct from that of the songs, but which in any case requires capacities that differ widely from those required for the memorization of speech. However, they are all intersemiotically connected to the ritual songs. Menezes Bastos showed not only that ritual makes communication between different groups in the upper Xingu possible, but also that Kamayurá music is a “machine that transforms words into bodies” (1999b: 53). A similar dynamic to that of Kamayurá music can be verified in rituals of several peoples of Indigenous America, says Menezes Bastos, and it consists precisely on the role of music as the pivotal element of what he calls an “intersemiotic chain of ritual,” in which verbal, choreographic, and plastic-visual systems connect. Menezes Bastos refers to this dynamic as translation and to music as the Kamayurá “translating machine.” Similar intersemiotic connections were observed by Angelika GebhartSayer among the Shipibo-Conibo, a Panoan people living in Peruvian Amazon, between songs and graphic patterns. For the Shipibo-Conibo, therefore, reading the graphic patterns would mean to read music—in

Reshuffling Conceptual Cards

99

other words, to translate between modes of expression: “In the shamanic ritual of this people, songs are the sonorous, reversible translation of pictorial motifs. One might even say that the latter would be the visual transcription of the former, which would be, therefore, their musical score,” explains Menezes Bastos (2007: 298) as he recalls Gebhart-Sayer’s study. The importance of the extralinguistic dimension highlights a decentralisation of the verbal code, which finds resonance in the context of studies on multimodality.9 In particular, Lynn Mario T. Menezes de Souza (2009) explains the multimodal tradition of the Kashinawá, another Panoan people, who live in the state of Acre, Brazil. Kashinawá multimodality, according to Menezes de Souza, integrates various systems of symbols—words, figures, icons, colours, and spatial dispositions— contrasting with the verbocentric bias propagated worldwide by the colonising legacy of Europe, which led to the suppression of other modes of expression.

Effectiveness Rather Than Referential Meaning Thinking about communicative practices among Indigenous peoples in lowland South America also leads us to question the assumption of referentiality that is deeply rooted in Euro-American ways of conceiving of language. For instance, let us consider the limitations of referential semantics for approaching techniques of enunciation in shamanic songs. If, as Carlo Severi (2007) reported, the therapeutic effectiveness10 of Kuna healing songs does not depend on the understanding of the message by the person who receives care—as in the case of women who give birth without understanding the meaning of the words sung by the shaman who assists her—what actually matters is the production of sound images through the proliferation of parallelisms, rather than the intelligibility of narration. Severi’s account echoes the way Graham Townsley (1993: 458) describes the “elliptical” language of shamanic Yaminahua songs: among this Panoan people of Peruvian Amazon, shamanic practice [is] intended to construct a particular type of visionary experience in the shaman himself and a communication, not with other humans, but with the non-human yoshi11 who populate that visionary experience. The clue to this is given by the fact that most Yaminahua can barely understand the songs. Many shamanic songs are almost totally incomprehensible to all but other shamans. As we see, the songs’ effectiveness prescinds being understood by those who are “sung” by them. This is related to the argument that Yaminahua shamanism is an “ensemble of techniques for knowing,” not “a system of knowledge”; that it is “a way of constituting” a discourse, not a

100

Jamille Pinheiro Dias

“constituted discourse” (452). And if singing, in turn, constitutes a path of verbal images to be traversed, traversing it in a skilful way is a capacity that is acquired through a process of transformation of the “condition of the body and its perceptions” (456). As to Marubo shamanic songs, Cesarino (2008) talks about how parallelistic intensity in the healing songs mediates ritual agency and controls the shamanic processes of duplication which are involved in neutralising agents of illness. Through initiation rituals, shamans learn to master this parallelistic intensity by gradually transforming their bodies and empowering their speech. Instead of being directed towards the patient as an audience, the reiterative formulae actually seek to directly approach doubles/spirits as helpers. In other words, patients do not need to interpret them in order to be healed. Once again, effectiveness, rather than semantic referentiality, plays a prominent role.

Conclusion By way of conclusion, I want to point out that we are colonised by certain ways of conceiving of language, but there is space for a disinvention and reconstitution of reified approaches to language through a consideration of the non-verbal and nonhuman complex ecology of forms of communication. In this chapter, I suggested that a critical approximation and shared commitment between applied linguistics and South American Indigenous ethnology may contribute to calling into question colonising ways of conceiving of language, as well as to permanently decolonise thought (Viveiros de Castro 2014: 40) and, as such, to help improve “the conditions of the ontological self-determination of the collectives” (43). As we call for a reshuffling of conceptual cards in relation to what counts as language among lowland South American Indigenous peoples, it is important to elucidate the cosmological contrast between other worlds and our modern, Euro-American world. In the decolonising agenda that we need to constantly articulate and re-articulate along global and local dimensions, there is also the possibility of “post-abyssal thinking” (Sousa Santos 2007: 97)—a systematic, collective effort towards self-reflexivity that may give rise to the “epistemological diversity of the world.” Postabyssal thinking emerges from the idea that there is endless diversity in the world and that this diversity still lacks an adequate epistemology, which still needs to be constructed (84). What needs to be avoided in the construction of epistemological diversity is the epistemology of the “point zero” or the “point zero hubris”— in other words, the epistemic coloniality of a subject of enunciation that does not consider the situatedness of his point of view, presupposing a “point zero” from which it would be possible to mask both the speaker

Reshuffling Conceptual Cards

101

and the location from where he speaks (Castro-Gomez and Grosfoguel 2007). Finally, it is important to remember that translating across cultural differences does not lie in consensus, nor in a unifying perspective, but in an ethical attitude of acceptance in face of the incompleteness of one’s own knowledge—one that is capable of allowing “a mutual intelligibility between different experiences of the world,” as Boaventura de Sousa Santos puts it (2005: 16).

Notes 1. Aristotle (2015) Politics. London: Aeterna Press. 2. Descartes, R. (2003) Discourse on Method and Meditations. Translated by Elizabeth Haldane and G. R. T. Ross. New York: Dover Publications. 3. Louden, R. B. (2007) ‘Language: Who/What Has It? (And Were Aristotle and Descartes Right?)’. History of Philosophy Quarterly 26, 373–387 (373). 4. See, among others, Agamben, G. (2002) The Open: Man and Animal. Translated by Kevin Attell. Stanford: Stanford University Press and Agamben, G. (1993) Infancy and History: On the Destruction of Experience. Translated by Liz Heron. London and New York: Verso. 5. Let me clarify at the outset that when I say “Euro-American” or “we,” I don’t refer to a reified, unchanging, monolithic construct, but to discursive traditions that dwell in North American and Western European categories, vocabularies, and frames of reference, regardless of being located in Europe or in the Americas, and whose perspectives are to a large extent informed by Greco-Roman and Judaeo-Christian lenses or filters—even if not in stable, homogenous ways. Moreover, this clarification is important to prevent us from falling into the trap of a Manichean dichotomy between a readily familiar Euro-America versus a presumably unfathomable non-Euro-America. 6. For further information on the railway cutting through portions of Awá Guajá land, see “Indigenous tribe occupies railroad in protest” (http://plus55. com/brazil-culture/2016/06/indigenous-tribe-occupies-railroad-in-protest, last accessed: August 3, 2017). 7. For an in-depth discussion of the category of “owner,” see Carlos Fausto (2008). In a nutshell, it designates “a generalized mode of relationship which constitutes Amazonian sociality and characterizes interactions between humans, between humans and nonhumans, and between persons and things” (333), playing a fundamental role in Amerindian cosmological and political relations. 8. As Cesarino notes, this certainly does not mean that Indigenous peoples do not claim they possess forms of “knowledge such as songs and graphic patterns” (184). 9. As Kress and Van Leeuwen observe, multimodal texts are those “whose meanings are realized through more than one semiotic code” (1996/2006: 177). 10. In “The Effectiveness of Symbols” (1949/1963), Claude Lévi-Strauss attributed a physiological and psychological “effectiveness” to the shaman’s ritual symbols, which would be able to induce healing. Carlo Severi’s argument offers a revision of that theory. 11. “Yoshi” is defined by Townsley as spirit or animate essence (452). The greatest the shaman’s ability to deal with a yoshi is, the more powerful the former will be.

102

Jamille Pinheiro Dias

References Agamben, G. (1993). Infancy and History: On the Destruction of Experience. Translated by L. Heron. London and New York: Verso. Agamben, G. (2002). The Open: Man and Animal. Translated by K. Attell. Stanford: Stanford University Press. Aristotle (2015). Politics. London: Aeterna Press. Bauman, R. and Charles, L. B. (2003). Voices of Modernity: Language Ideologies and the Politics of Inequality. Cambridge: Cambridge University Press. Bernstein, C. (1998). Close Listening: Poetry and the Performed Word. New York: Oxford University Press. Brabec de Mori, B. and Seeger, A. (2013). Introduction: Considering Music, Humans, and Non-Humans. Ethnomusicology Forum, 22, 3, 269–286. Castro-Gomez, S. and Grosfoguel, R. (2007). El giro decolonial: Reflexiones para una diversidad epistémica más allá del capitalismo global. Bogotá: Siglo del Hombre. Cesarino, P. (2008). Oniska: A poética do mundo e da morte entre os Marubo da Amazônia ocidental. Ph.D. Dissertation in Social Anthropology. Rio de Janeiro: Museu Nacional, Federal University of Rio de Janeiro. Cesarino, P. (2010). Donos e duplos: relações de conhecimento, propriedade e autoria entre marubo. Revista de Antropologia, 53, 1, 147–197. Déléage, P. (2012). Transmission et stabilisation des chants rituels. L’Homme, 203–204, 103–138. Descartes, R. (2003). Discourse on Method and Meditations. Translated by E. Haldane and G. R. T. Ross. New York: Dover Publications. Fausto, C. (2008). Donos demais: Maestria e domínio na Amazônia. Mana, 14, 2, 329–366. Franchetto, B. (2004). Línguas indígenas e comprometimento lingüístico no Brasil: situação, necessidades e soluções. Cadernos de Educação Escolar Indígena, 3, 9–26. Grotti, V., Fausto, C. and Brightman, M. (eds.) (2016). Ownership and Nurture: Studies in Native Amazonian Property Relations (pp. 186–209). New York: Berghahn Books. Kopenawa, D. and Bruce, A. (2012). The Falling Sky: Words of a Yanomami Shaman. Translated by N. Elliott and A. Dundy. Cambridge, MA: Harvard University Press. Kress, G. and Van Leeuwen, T. (1996). Reading Images: The Grammar of Visual Design. London and New York: Routledge. Lévi-Strauss, C. (1949). The Effectiveness of Symbols. Structural Anthropology (pp. 186–205). New York: Basic Books. Makoni, S. (2003). Black Linguistics: Language, Society, and Politics in Africa and the Americas. London: Routledge. Makoni, S. and Mashiri, P. (2007). Critical Historiography: Does Language Planning in Africa Need a Construct of Language as Part of Its Theoretical Apparatus? In S. Makoni and A. Pennycook (eds.). Disinventing and Reconstituting Languages (pp. 62–89). Clevedon: Multilingual Matters. Makoni, S. and Meinhof, U. (2006). Lingüística aplicada na África: Desconstruindo a Noção de Língua. In L. P. Moita Lopes (ed.). Por uma Lingüística Aplicada Indisciplinar (pp. 191–213). São Paulo: Parábola Editorial. Makoni, S. and Pennycook, A. (2007). Disinventing and Reconstituting Languages. In S. Makoni and A. Pennycook (eds.). Disinventing and Reconstituting Languages (pp. 1–41). Clevedon: Multilingual Matters. Makoni, S. and Pennycook, A. (2012). Disinventing Multilingualism: From Monological Multilingualism to Multilingual Francas. In M. Martin-Jones,

Reshuffling Conceptual Cards

103

A. Blackledge and A. Creese (eds.). The Routledge Handbook of Multilingualism (pp. 439–453). New York: Routledge. Mehinaku, M. (2010). Tetsualü: pluralismo de línguas e pessoas no Alto Xingu. MA Thesis. Rio de Janeiro: Universidade Federal do Rio de Janeiro. Menezes Bastos, R. J. de (1999a). Apùap World Hearing: On the Kamayurá Phono-Auditory System and the Anthropological Concept of Culture’. The World of Music, 4, 1, 85–99. Menezes Bastos, R. J. de (1999b). A Musicológica Kamayurá: para uma Antropologia da Comunicação no Alto Xingu. 2nd ed. Florianópolis: UFSC. Menezes Bastos, R. J. de (2007). Música nas sociedades indígenas das terras baixas da América do Sul: estado da arte. Mana, 13, 2, 293–317. Menezes Bastos, R. J. de (2013a). Apùap World Hearing Revisited: Talking with ‘Animals’, ‘Spirits’ and Other Beings, and Listening to the Apparently Inaudible. Ethnomusicology Forum, 22, 3, 287–305. Menezes Bastos, R. J. de (2013b). A festa da jaguatirica: Uma partitura crítico interpretativa. Florianópolis: Editora da UFSC. Menezes de Souza, L. M. T. (2009). The Ecology of Writing among the Kashinawá: Indigenous Multimodality in Brazil. In A. S. Canagarajah (ed.). Reclaiming the Local in Language Policy and Practice (pp. 73–95). Mahwah and London: Lawrence Erlbaum. Montardo, D. L. O. (2002). Através do mbaraka: música e xamanismo guarani. Ph.D. Dissertation in Social Anthropology, Universidade de São Paulo, São Paulo. Mühlhäusler, P. (2000). Language Planning and Language Ecology. Current Issues in Language Planning, 1, 3, 306–367. Print. Mühlhäusler, P. (2002). Linguistic Ecology: Language Change and Linguistic Imperialism in the Pacific Region. London: Routledge. Rancière, J. (2008). The Politics of Aesthetics. Translated by G. Rockhill. New York: Continuum. Rancière, J. (2009). Aesthetics and Its Discontents. Translated by S. Corcoran. Cambridge: Polity. Santos-Granero, F. (2006). Vitalidades sensuais. Modos não corpóreos de sentir e conhecer na Amazônia indígena. Revista de Antropologia, 49, 1, 93–131. Severi, C. (2007). Le principe de la chimère: une anthropologie de la mémoire. Paris: Musee du Quai Branly/Editions Rue d’Ulm. Sousa Santos, B. de (2005). The Future of the World Social Forum: The Work of Translation. Development, 48, 2, 15–22. Sousa Santos, B. de (2007). Para além do Pensamento Abissal: Das linhas globais a uma ecologia de saberes’. Revista Crítica de Ciências Sociais, 78, 3–46. Strathern, M. (1988). The Gender of the Gift: Problems with Women and Problems with Society in Melanesia. Berkeley, Los Angeles and London: University of California Press. Townsley, G. (1993). Song Paths: The Ways and Means of Yaminahua Shamanic Knowledge’. L’Homme, 126–128, 449–468. Viveiros de Castro, E. (2002). A inconstância da alma selvagem. São Paulo: Cosac Naify. Viveiros de Castro, E. (2014). Cannibal Metaphysics: For a Post-Structural Anthropology. Edited and translated by P. Skafish. Minneapolis: Univocal.

Section III

Portuguese as Glocal Language

5

The Imaginary in Portuguese Language Perceptions in Academia (Mis)directions Between the Local and the Global Gesualda dos Santos Rasia

Introduction In a documentary by Victor Lopes, entitled Língua, vidas em Português (Language: Lives in Portuguese) (Lopes 2002), the writer José Saramago argues: “There is not a Portuguese language, there are languages in Portuguese.” This statement puts us face to face with the paradoxical uniqueness-plurality of the language of Camões and Machado de Assis. It also raises another question about the issue that we aim to investigate in this study: How are identity ties formed in a language that increasingly inhabits the interstices between local and global? Furthermore, another correlated question emerges from the latter: How is the Imaginary in the Portuguese language formed by language students, having in mind that previously they also become acquainted with theoretical assumptions that motivate these questions? This analysis is based on the discourse studies of a materialistic approach, as proposed by Michel Pêcheux (1988), focusing specially on the notion of the Imaginary as developed in the “Imaginary Formations.”1 This approach concerns the fact that meanings are of a social character, and they are produced from the positions that the subjects attribute to each other mutually. In the same way, indexes of value are attributed via the material through which we speak: language. These discursive postulates are observed in relation to the studies on Sociolinguistics2 and Sociology developed by Pierre Bourdieu, which grounded the notions of norm, identity, and power relations. The approximation established between both of these fields of knowledge is justified by the relation between the values that are socially attributed to the different linguistic registers and to the production of a political division on the very core of the language. This is outlined in its global instance and also in its internal borders, which need to be overcome by subjects in their daily exchanges. The hybridization movement consists of a coming and going between linguistic identity processes that are situated in a localized instance, in which there is a differential marking and a global marking, which is

108

Gesualda dos Santos Rasia

responsible for the effect of the universalizing norm. This finding allows us to problematize how, in the case of Brazilian Portuguese, centripetal and centrifugal forces act in the power relations concerning the recognition/legitimation processes of the subjects and their linguistic registers. The empirical dimension of this research consists of interviews with language professors in the undergraduate program of the Federal University of Paraná. The students were selected from different semesters and degrees and have attended the Sociolinguistics and/or the Discourse Analysis subjects in the last three years. The aims of the research plan consisted in: a) mapping the representations these students have about Portuguese as a language that moves between the local and the global; b) pointing out how theoretical contributions studied in academia provide subsidies for the confrontation of linguistic dissensions in daily exchanges; and c) realizing how Portuguese is spoken and placed in relation to foreign languages also studied in this program. Given that the condition of the language of Camões and Machado de Assis is that of a plurality in its uniqueness, we are concerned with mapping how the imaginary that forms representations about this language constitutes itself from different points of view—woven by internal and external forces—as well as from its multiple developments.

Theoretical Assumptions: The Multiple Faces of Brazilian Portuguese’s Constitution of Identity Heterogeneous as any language, the differences in the Portuguese language in all the different countries/continents where it is spoken are not smaller in its Brazilian territory, which is our starting point for this study. The interplay and clashes between diversity and uniqueness have their historical roots in the conditions of production in which the establishment of this language, originated beyond the Atlantic, is inserted in the Brazilian historicity. Its double foundation, which has given the language a paradoxical character, as previously affirmed, of diversity in unity, beckons us to approach the language issues from a historical and cultural perspective tightly influenced by aspects of political order. Orlandi (2009) refers to the local memory that has been weaving the language through the elaboration of its own linguistic instruments, which contemplated aspects that have been distinguishing Brazilian Portuguese from European Portuguese since the colonization process. It is worth emphasizing that these instruments are not limited to grammars and dictionaries, which are historically endorsed as a source of language legitimization. Nonetheless, the instruments have, in a special way, given visibility to the regional registers, in a clear eulogy to the difference. However, these only started finding their formal place in the early nineteenth-century production named Brasileirismos, idiotismos, curiosidades verbaes (Brazilianisms,

The Imaginary in Academia

109

idiomatisms, verbal curiosities) (Orlandi 2009: 80). The author above nevertheless based his argument by providing legitimacy to the margins of what is generally considered the norm. The same norm that played, from its original institution, with what the author called “false bottom” in which the “same” has, however, an “other,” a historical “different” that constitutes itself even in the resemblance of the same. The Portuguese and the Brazilian languages cover each other as if they were the same language; however, they are not. They produce different discourses; they have different meanings. (Orlandi 2009) While attempting to constitute an identity of this language, the relation of the Brazilian memory with the Portuguese memory produced its effects in the determination of what is the grammatical norm and of what is “to use the language.” Consequently, it generates a double linguistic reality, defined by Orlandi (2009) as the imaginary and the fluid language, in which the latter would be the real, spoken language. The fluid language is the language that cannot be constrained since it is in a state of constant motion. The imaginary language would be the idealized form, which is installed in a fixed system of rules. This Imaginary establishes not only the external border but also the multiple internal borders of the language, determined in an axial mode by the imaginary language. This occurs in a continuous and permanent attempt to paralyze the multiplicity of other possible and real forms, the fluid language. This relation is paradoxical due to the fact that the same movements towards the stiffening, the immutability, nonetheless confirm that of the becoming. From this tension derive the dissents that are not infrequently present in the media—for example, in situations where we witness attempts of regulation of what is imposed as correction patterns, defined by normalization manuals as “writing and speaking well.” In addition, there is—not absolutely disconnected from the previous situation—the political division of the speakers. The farther from the imaginary language, the more distant are the subjects from the possibility of being recognized as legitimate speakers of the language. This fact, in Bourdieu’s terms, is an attribution of symbolic value of the subjects’ social status through their respective linguistic registers. Hence, it is a relation of symbolic forces that is not, according to the author, “only defined by the structure of existing and properly linguistic competences, and the properly linguistic dimension cannot be separate from the linguistic productions” (Bordieu 2008: 11–12). Marcellesi and Gardin already affirmed that “the groups effectively do not exist if not through the collective society, and the collective society through the groups” (Marcellesi and Gardin 1975: 19). Especially

110

Gesualda dos Santos Rasia

if these groups distinguish themselves by their antagonisms and not by their similarities. From this point of view, the authors propose the study of linguistic activity by contrast, because “the various social classes have a historical role to consider.” This occurs as it starts from the following assumptions: linguistic activity is determined by an exteriority that is contingent in itself; the elements of culture are also active in linguistic activity, and not as an external and independent entity; the linguistic problems of the modern societies must be considered in its social contradictions. This study does not perceive the external variables in an objective relation to the linguistic phenomena, but considers them in a historical-symbolic perspective. Besides some of the questions reported above and to be discussed below about Brazilian Portuguese regarding the refraction of the mother tongue, there are still more recent linguistic relations. Marked by globalization, the twentieth century was the century of economic and political exchanges. However, we know that these two dimensions are not only overdetermined but also unequally related to each other and, furthermore, they affect linguistic issues directly. Thus, the influences on Brazilian Portuguese have been increasing due to the enlargement of the economic exchanges, to people flow, and, more particularly, to the importing of technology. The presence of the other languages reveals itself through morphic segments, words, and, finally, the limit of syntactic constructions. The incorporation of elements from outside the language inevitably incites confrontations of a political-ideological order—some ascribed to nationalistic positions—and also from the idea that it is possible to control the movements of the language and its uses through legal regulation.3 Both internal frontiers and external borders, woven in the multiple territorialities of the Portuguese language, refer to constructions that derive from political and economic orders. This fact ends up producing effects in the imaginary relations as the subjects project their own places and the place of the Other on society. However, the functioning of the Imaginary reaches the concrete relations, and it has not avoided leaving traces of symbolic violence. For example, the recent case of a Brazilian doctor who mocked a patient on social networks, because the patient, in an appointment, pronounced the pulmonary infection “pneumonia” as “peleumonia” (Victal 2016). The doctor made this jocular comment: There is no “peleumonia.” Nonetheless, the conditions of production in which the word has emerged were situated in an interactive context, spoken, in which the doctor understood his patient very well. At that interval, the word existed. Not only at that interval, because in that moment the notion of a certain social segment identity was already marked. Such marking is not related to a causal relation between the social segment and the phonetic-phonologic occurrences, but to a place of resistance against the spoken, well represented in the case mentioned.

The Imaginary in Academia

111

Also emblematic is the case of an advertisement exhibited in the streets of a Brazilian state capital in 2013, at public transportation stops, in upper-middle class neighborhoods, which read, among other information, the following as a main slogan: Curso de Português para brasileiros. Não tropece no Português e cresça na carreira. (Portuguese course for Brazilians. Do not stumble on your Portuguese and move forward in your career). The text says much more than one may think at a first glance, as soon as the effects of evidence are undone. Starting with the core assumption that Brazilians supposedly do not know their own language when they should already be proficient in it and, moreover, if we consider the image in the advertisement of a teenager and an adult. Then, we can understand that Portuguese is a language that presents difficulties, unveiled by the idea of “stumbling,” and that can be understood if we analyze it in terms of the ever-present dichotomy between the stiff norm and the flexible uses, which has already been discussed. Last, the relation between language mastery and professional success. This aspect has its share of veracity in the world order. Nonetheless, the problem lies in the gaps of what has not been said in the advertisement, and even among what could be said considering the historical conditions in which the advertisement had been produced. The advertisement is from a language school, which implies that it is the mother tongue taught from the perspective of the Other. Hence, what does “do not stumble on” mean? A better qualification on reading and writing a text? In this case, we believe that, in fact, this could qualify the subjects not only for a professional career, but also for their lives. However, if it is something restricted to normative knowledge, the effectiveness of this promise remains in doubt. Up to this point, we have discussed some aspects that constitute Brazilian Portuguese identity, whether in its internal or external dimension, as well as in its historical and contemporary dimension. The fact that it is the language of colonization, which has obtained its own statute, not forgetting this memory, which is nonetheless re-signified, brings some elements that direct us to the next point of discussion, the local and global dimension of the language.

Agreements and Disagreements Between the Local and the Global Nowadays, the Portuguese language lies in the sixth position of the most spoken languages in the world, with 244 million speakers (Camões: Instituto da cooperação e da lingual). This fact, on the one hand, allows this language a political-cultural position and, therefore, an economic

112

Gesualda dos Santos Rasia

position; on the other hand, it brings about a discussion on the meanings implied in its realization in this wide range of countries and continents where its speakers are spread. Saramago’s statement above, “There is not a Portuguese language, there are languages in Portuguese,” emphasizes the unity/diversity paradox. This paradox is based on the questions recently discussed around the identity of the Brazilian language and can be put in terms of diversity in unity. Pêcheux’s notion of imaginary formations, which refers to the places that subjects attribute to themselves and to others within the discursive processes, may help us in this matter. These places emerge from the projected image regarding the position they occupy in the social spectrum, and also from the projected image about the referent. In this case, the language of which they are speakers: Brazilian Portuguese. Following Pêcheux (Gadet and Hak 1993: 83), a discursive process implies the anticipation, from the subject who speaks, based on the representations resulting from the interplay of images, which impose the following questions mutually: “Who am I to speak to you in this manner?” “Who is he for me to speak with him in this manner?” as well as the different ways these questions respectively unfold. Such interrogations enable the understanding, for example, of the meaning implied in the previously discussed language school advertisement about the Portuguese language. The subject of discourse who makes an utterance there, speaks from a realm of anticipation, crystallized in the already-spoken, which affirms the difficulties of learning the mother language. At this point, the gap between the notions of fluid language and imaginary language is installed. In the name of such supposed difficulty, the subject of discourse situates him/herself in the site of arrival, of waiting for the subject-Other, the one addressed by the advertisement, promising the dissipation of all and any possibility of “stumbling.” Therefore, the relation with the refereed is somewhat ambiguous, due to its double standing in a double position of inaccessibility to its own speakers. This is parallel to the possibility of “disclosure” strangely assigned to the Other’s place. This advertisement is very emblematic in this discussion by demonstrating some aspects of the relation between the local and the global since that is where the zones of contradiction become materialized. Orlandi (2012: 6–18) approaches the notion of globalization also from the perspective of an imaginary construction that establishes the idea of an “apparent global unification” that, however, “erases deep disparities and, by doing so, reinforces the differences, both in space and social levels” (my translation). Such imbalance materializes itself both in the discourse of expropriation of the language by its own speakers and in the incoherence between the national language and the foreign language school. The question of value, as pointed out by Bourdieu (2008), emerges once again in the materialization of difference amid the discourse of universalism.

The Imaginary in Academia

113

Nevertheless, the compensation, the instruments of resistance, and the forms that make possible the agreement between the local (or the locals) and the global emerge while giving visibility to the multiple cultural, social, and linguistic forms, in the different spaces, being they local or global. However, it is important to highlight, as emphasized by Orlandi (2012: 5) that it is: “the capacity to get in touch with different languages that allow the local actors to impose themselves at an international level. And not the monolingualism, I would add, that deprives the subject from their pluridimensional and interchangeable subjectivities” (his emphasis). This repositioning of the subject in relation to his/her language, when considered in the spectrum of other languages in the global scenario, can be conceived in a special manner if we consider the historicity of the Portuguese language. Its memory domains must also be considered as well as the internal and external relations of forces that derive from it and still produce their effects. These questions have immediate implications in the sphere of international organizations, which regulate the linguistic policies. Their effects are not infrequently sensed/experienced by tourists, (im)migrants or students who every day move across borders. These are not those who make political decisions because they are in the condition of language users, some of whom are language scholars. These will be our focus from now on.

The Brazilian Portuguese Language in Transit: In and Between the Glances of University Students As far as the non-universalization of the Portuguese language in the world context is concerned, its speakers are now in a much more global condition than in previous decades. And, in the Brazilian case, more specifically, the last decade experienced a maximization of this transit in the academic sphere, as a result of public policies that aimed at research and internationalization. As a consequence, the subjects were called to interact with other languages from the starting point of their mother and official language. As an effect of globalization and its economic movements, many young students came to Brazil seeking cultural and knowledge exchanges. Here they also put themselves at the linguistic border, in which they produced meaning about and to their mother languages having the Brazilian language as a parameter and, dialectically, produced meaning toward it, although based on their own language’s otherness in relation to Brazilian Portuguese, and what they mean to the participants as their speaking subjects. Therefore, these relations, which constituted this study’s material of analysis, will be the focus below. The subjects of this research were: three native speakers of Brazilian Portuguese without experience of living abroad; three native speakers of Brazilian Portuguese with experience of living at least six months abroad; and three speakers of Portuguese as a foreign language who were

114

Gesualda dos Santos Rasia

undergraduate students in exchange programs at the Language department. It is important to note that the first three students without experience abroad are also studying other languages, most of them English. The research questions conducted the analysis excerpts presented in the sections below as discursive blocks. This notion helps us give a certain discursive unity throughout the analysis since the various particular “sites of voice,” by which we mean their particular contexts, will not be given a focus here. 3.1. The Brazilian Portuguese Language From Inside: The Vision of Academics With No Experience Abroad Identities are always forged from the glance of the Other, a relation of otherness that configures from the outside how the subjects are seen so that they can speak about themselves. We are discussing the domain of the Imaginary Formations, in Pêcheux’s terms (Gadet and Hak 1993: 82), and when the first question was proposed, “Which aspects define/ distinguish you as a Brazilian?” we understood that the answers are not only projected on the horizon of a distinguishable Other, but also on an imaginary constructed Brazilian-ness. De1: The first of the aspects that I consider is to have been born and have always lived in Brazil. The understanding of the Brazilian culture and the use of the Portuguese language as my mother language. De2: The fact that I have the Brazilian Portuguese as my mother language and all the cultural aspects of the society where I live, which were incorporated by me. Language and culture converge as elements of identification in both discursive sequences. It is important to stress that language emerges in its mother status with regard to the constitution of identity, the meaning of which can be perceived beyond the mere scope of designation; it is circumscribed to that which cuts/divides the space of politics in language between the dimension of the State language, with its injunctions and the language that is brought from home, used for the daily exchanges, and is submitted to institutionalization at school. Another aspect that also deserves attention is the reference to the language, which is, however, stated in the quotes above: Portuguese language, in De1, and Brazilian Portuguese in De2. The subjects reproduce, in the very space of the utterance, the double filiation, determined by different historical conditions of production, which was discussed in the initial part of this study. Concerning the second question, “What does ‘to know the Portuguese language’ mean to you?” it can be stated that the answers point out to the internal dimension of the language: the skill of making themselves

The Imaginary in Academia

115

understood among their pairs. On the one hand, this seems to be obvious but, on the other hand, it is in fact a discourse that deconstructs utterances about the hermetism of the language, of a supposed incomprehensibility for the speakers themselves. De3: Being able to make other speakers of the language understand what is been said in Portuguese. De4: To know the Portuguese language is not only to be able to communicate in the language, but also to understand the nuances of the language. 3.2. The Brazilian Portuguese Language Across the Border: The Imaginary Constructed About and From the Language The point of view of academics who had studied abroad was undoubtedly partial since it is a small sample within a significantly larger universe. However, this is not a quantitative research—for this reason, we consider the quotes provided below as representative of some perceptions produced by academics with experience across borders. The constitution of social memory is not woven, by default, of the subjects’ constitution of identity when talking about their culture or about their language. The academics who participated in this study are from a public University of the State of Paraná, in southern Brazil, and are all descendants of European immigrants.4 The question proposed in this research was: “Are/were you easily and immediately identified as a Brazilian, in your contacts with other people abroad? In this case, in what aspects specifically?” Invariably, all answered that initially they were not recognized as Brazilians, especially because of their physical traits. A few of them were identified as foreigners “through their first words,” but not necessarily as Brazilians. However, what calls our attention in their answers is the projection of an imaginary distant from the Brazilian identity, provoked by the look of the foreign-Other and legitimated at once by the former: De5: “No, not even my accent was identified as one of a Brazilian, people seem very surprised when they know about my origin.” De6: “In the first moment, I was not identified as a Brazilian, on the contrary, they even thought I was Spanish (. . .) before my first words, of course, they thought I was Spanish.” De7: “I was easily identified as foreigner, but they only knew that I was Brazilian after I said that.” The legitimation, which we referred to, is given through the symbolic, constituted by the surprise element in De5. In this moment, it can be said that

116

Gesualda dos Santos Rasia

there is a desire of approximation with the European identity, observed in the double negative: “not even my accent was identified.” Furthermore, the “confusion” made by the Other’s perception when the subject is not himself/herself attached to their language, as is apparent in De6. In this “arrival” to the Other-language, the subject presents him/herself at distance from possible elements of identification with their homeland identity, materialized by the intensifiers on the contrary and even. De7 follows the same perspective, in which the adversative structure has a strangely made inversion: in a first moment, there is a false assertive, which is, however, presented as true: “I was easily identified as foreigner.” Only in the second segment of the sentence the true proposition related to her nationality is presented. The expected, in terms of the construction of the utterance, from the point of view of a desire of recognition of their Brazilian identity, could perhaps be stated as: “I was not recognized immediately as Brazilian, but only after I said I was not a foreigner.” Another question proposed to the interviewees was, “Are there written materials in Portuguese in the country you lived in? In this case, in which contexts and which kinds of materials?” In addition: “Was it in Brazilian or European Portuguese? Besides written materials, did you perceive the presence of other aspects of Brazilian culture (song, art, food, etc.)?” These three questions were used as subsidies to the study tools so that the research participants could reflect on the following question: “What is your evaluation about the space that the Portuguese language occupies in the country you lived in?” De8: A very small number. At the university (Lyon 2) there is an undergraduate course in Languages—Portuguese. I know that there is a Portuguese center in the city. In general, the Portuguese language is associated to Portuguese people, because there was a great migratory flux at some time. The Portuguese occupied jobs notably in civil construction. De9: I opted to circulate in a space where there were many speakers of Portuguese. Because of the music, all musicians interested in samba, bossa nova and MPB ended up learning Portuguese. Then, considering this space specifically, I felt that there were many exchanges with the Brazilian culture being passed on through the language. On the other hand, in more institutionalized spaces there was nothing in Portuguese. Even in some museums, with guides in several languages, I realized that we could not find it in Portuguese at all times. Or when Portuguese was presented, it was marked with the flag of Portugal. Audio guides were always in European Portuguese too. De10: I think that the interest in the Portuguese language is very specific. I got in touch with Portuguese classes, and the majority

The Imaginary in Academia

117

of the students were learning the language because they had some kind of relationship with Brazilians or Portuguese (they were dating, married, had grandchildren or friends in Brazil/Portugal). The interest in learning Spanish is much higher too, which in a certain way “overshadows” the teaching of Portuguese. De11: In Munich especially, one can find a lot about Brazilian culture. I found some places, as hairdressers, Brazilians food stores, nightclubs with the same music and operation as in Brazil, forró events, cultural meetings with Brazilians stalls and many Brazilians. The Germans I have met loved caipirinha and soccer, they always reminded me of both. Especially of soccer because they won the last World Cup. The answers were almost consensual since academics practically did not notice the free circulation of Portuguese language items. The considerations about this absence of linguistic awareness deserve our attention. The ways in which the subjects answered allow us to perceive the production of meanings about the language and the subjects themselves. Due to the European context, it is understandable that Portuguese, if found to be present in any environment, has its European variant as a dominant, as reported by the student who mentions the Portuguese flag. However, even this presence is sporadic and restricted to touristic spaces and in writing. Another feature of the Portuguese presence, besides tourism, is the specificity of learning the language as it is determined by the particular interests of the individuals themselves. Even if this interest emerges from interpersonal relationships, in fact it is also determined by the actual exchanges. However, the question resides in which ways these actions are considered, ranging from a political perspective, which aims to the institutionalization of a space in which language affairs, identity, and political dimension are considered constitutive. The relation between language and its cultural dimension also called our attention. It is known that Brazil is recognized by its music, especially samba and MPB (Brazilian popular music), by its festivities in Carnaval, and by soccer. These characteristics are not absent of internal criticism, because we see in this a recognition of a stereotypical form of our representation. This stereotype obliterates other materialities of our cultural multiplicity, besides overestimating the economic-touristic dimension. Then, considering this criticism, the music sphere works as a link for linguistic approximation and also for learning, as expressed in De9. If in the French context, the “meeting” with the language occurred through cultural elements that happened due to the initiative and the movement of Brazilian residents, it seems not to be the case of a German context in Munich, as represented by De11. The emergence of the Brazilian presence through labor occurs in an interesting manner. On the one hand, it seems to maintain a niche aiming at Brazilians, especially for the maintenance/

118

Gesualda dos Santos Rasia

stabilization of the language. On the other hand, a “showcase” bias of tropical culture exoticism is created. Also, the rupture generated by Germany’s victory over Brazil by the score of 7–1 in the last World Cup seemed to be an inevitable topic of discourse. These questions, by being circumscribed to the cultural scope, wrongly seem to be disconnected from the language field. However, the insertion of the few students who participated in this research demonstrates that there is a connection that cannot be ignored, especially by the ways in which the subjects’ relations are created with their identity links. Probably, the academics in the context of studying/researching abroad were exposed to the critique of the stereotypes built around the image of Brazilians abroad. However, in a situation of fragility/scarcity of linguistic contact, no other possibility is left for them than to let themselves be integrated in their social context by earning any kind of recognition, even without feeling they fit in it completely. When a language has a universal status in relation to the global dimension, its functioning is fragmented, not to say weakened. The heterogeneity that composes Brazilian culture would demand the presence, beyond-borders, of food, dancing, rhythm, and representative songs spread from north to south, east to west, of the country with their peculiarities and idiosyncrasies. However, the hegemony effect is imposed and sometimes it is established as an effect of evidence. Why is it like this? Tau Golin, Brazilian historian, analyzed the indignation of the gaúchos5 due to the fact that its culture, as well as of other minorities, was not represented during the opening ceremony for the 2016 Rio Olympic Games in Brazil. In an article that circulated in press and online (Costa 2016), Golin stated that the predominance of the Carnaval image presence and of the northeastern culture refers to an absence that is also internal. Golin points out, based on anthropological studies, that the richness of the cultural heritage of different regions is invisible even to Brazilians themselves. This invisibility involves their absence in the consecrated arts, such as cinema, theater, and music, through which they could reach national representativeness. The invisibility hypothesis can be dimensioned in political and economic terms, since the space that Carnaval and the culture of the Northeast have in tourist terms is unquestionable. Furthermore, this is where the invisibility of all the other cultures begins and from where the idea that even the Brazilians have identified themselves with the idea of an homogeneous cultural expression cannot be ruled out, and leads them to ignore the peculiarities and particularities of their surroundings as deserving equal relevance in the national and international scenarios. Another related question emerges from the previous one, lying at the same level: How are internal relations between differences positioned in relation to the unity? And besides, how does this spectrum of singularities stand in the global scenario?

The Imaginary in Academia

119

3.3. The Brazilian Portuguese Language in the Eyes of the Other: The Exchange Students The number of undergraduate and postgraduate students from Brazil going abroad has increased in recent years, and the entrance of foreign students into the country has also increased significantly. Perhaps not in the same proportion, but it is still relevant. The Federal University of Paraná (UFPR), in which this study was conducted, received around 250 exchange students in the last three years, besides refugees who came from different continents.6 Of these students, we selected three in this research because they met the criteria mentioned initially. Hence, we selected one student from Belarus, one from Haiti, and one from Japan. The Belorussian is entirely proficient in Portuguese due to the fact that he already studied the language before coming to Brazil and has lived in the country for a longer period. The Haitian presents a very good level on spoken language and intermediate level of writing. The Japanese, due to the fact that she lived only a year in Brazil and also by being accustomed to a completely different writing system, can be considered to have a very good oral level and above average writing skill. Coming from different continents, the reasons that brought them to this country are also diverse. The Haitian came in the position of a refugee, a condition that imposed the choice of the country and, by extension, of the language. The condition of the Japanese student is entirely different; she comes from a middle-class family that allowed her to choose a country with a completely different culture, in order to expand her range of knowledge. “The Brazilian way of being: expansive, receptive, in contrast with the oriental seriousness” demonstrated to be an attractive factor, according to her statement. The Belorussian student affirmed, in an informal conversation, that he left his country seeking better jobs and opportunities to study, combined with his willingness to build his future and pursue education in another country. In Brazil, he established affectionate relationships and consequently took up residence in this country. The fact that these participants are not Western Europeans provides another perspective to the analysis of the facts, which is not based on contrast, but on other domains of memory and other relations of knowing, of power, and of identity. Some questions are equal or similar to the previous groups, such as this one: “Do you feel you are easily identified as a foreigner in Brazil? If yes, by what aspects?” However, the answers, different from the Brazilian students, indicated their languages provided a locus of recognition of their identity: De12: “Yes, I am learning Portuguese, so I do not speak it well as Brazilians. I think it is difficult to be identified only by the appearance, but by the speech it is easily recognizable.” De13: “Each time less. The accent, the style.” De14: “By the accent, and also the color and style.”

120

Gesualda dos Santos Rasia

The common aspiration of “dominating” the language of the Other does not prevent them, as foreigners, to keep their identity affiliation. It is important to highlight here the value different accents take in the culture Imaginary with regard to the foreigner, which is here presented as the celebrated Other. It is in contrast with the foreigner in France, for instance, who may be regarded as a “poor speaker” when displaying an inappropriate accent. Regarding the presence of the Portuguese language in the participants’ countries of origin, it is even weaker than in Western European countries, which is explained by the scarcity of political and economic relations. However, the space that Brazilian Portuguese has earned in Eastern European countries called our attention, though to a relatively small extent, due to the export of soap operas. This aspect would require a specific study, in the sense that a country’s interest in the Portuguese language tends most likely to emerge through the imaginary representations built by the culture of Brazilian soap opera culture, even though it is marked by version/translation elements. The last two questions were about the study participants’ projections of what the knowledge of Portuguese, in theory, would enable them to do. The questions were structured as follows: What uses have you made of the Portuguese you have learned (in personal life, work . . .)? In your opinion, which opportunities are made open by learning Portuguese? De15: I applied for admission in UFPR to improve my knowledge of the language. My job is teaching Russian to Brazilians where my knowledge of Portuguese helps me very much. I also work as a translator and interpreter. The greatest opportunities are the new horizons that the Portuguese language opened for me, the new people I have met, and the teaching job I have nowadays. De16: I believe I can have a job opportunity due to my knowledge of Portuguese. Nowadays, many Japanese companies are expanding to Brazil, and I can obtain a chance of working here. In addition, I can study Latin languages easier than before because these languages are similar. Then I believe that the Portuguese will open ways for me to learn other languages. De17: I use the Portuguese language to communicate in daily life and also at the job. I use it in academic activities and tasks. Portuguese is more widely spoken than French, and this is already a great opportunity. In addition, I can work as an interpreter in a French-Brazilian and Haitian company even outside my area.

The Imaginary in Academia

121

The quotes above bring two essential aspects about the Portuguese language: an economic dimension and a dimension that gives it a vehicular character. On the one hand, Portuguese does not have the same status of the English language in the global scenario. On the other hand, Portuguese is a possibility for Haitians refugees, due to its similarity to French, which offers advantages in the global scenario in quantitative terms, as expressed by the student. In addition, the origin of the Portuguese language as a Romance language offers the possibility of a bridge for Oriental students to learn other languages. Regarding the economic dimension, the Portuguese language has attracted the interest of a niche that is not South America, even though Brazil has suffered a decline in its image of an emerging country in the last years. Portuguese keeps itself as a possibility of accessing the job market, from different interfaces regarding foreign languages. Through Portuguese it is possible to teach Russian to Brazilians; Portuguese could be used to work as a mediation tool between French and Portuguese for the Haitian whose mother language is none of these. About the job market that the Japanese student mentioned, it is a niche centralized in automotive and computer technology areas. In these niches the language occupies an interesting space. For instance, mastering the English language would, hypothetically, make the knowledge of Japanese superfluous to the youth seeking to be employed at those companies. Thus, there remains the question about which interfaces can be built, at the level of a projection, from the links that the native speaker of Japanese weaves from his/her language to ours, such as when the participant talks about products of consumption idealized in her culture that become objects of desire in our environment. However, this is another question, not absent from the mediation of language, and from the relations that are simultaneously local and global.

Conclusions This study started questioning, among other things, the agreements and disagreements between the local and global dimensions of the Portuguese language. Such questioning means to specifically discuss the identity that this language has been composing in its Brazilian historicity. Nonetheless, it does not occur without ruptures, without dissent, without loss, without assimilation, without other agreements and strangeness. This movement is characteristic and constitutive of any and every language; the condition of non-enclosure in relation to the local establishes an inside that is intrinsically constitutive of its exterior. Repeating Marcellesi and Gardin (1975), “The groups effectively do not exist unless in the realm of the global society, and the global society in that of the groups.” In the same way, language is only achieved in the encounters established locally and

122

Gesualda dos Santos Rasia

globally. The students’ views, constituted by different perspectives, provided elements upon which to reflect about the status of these relations. First is the fact that the political divisions established in the languages, which are theoretically developed in the two syllabus components mentioned in this research, are represented in the academics’ answers. They talk in a different manner about the Portuguese language in its Brazilian specificity. They reach the comprehension that the constitution of identity is achieved through cultural aspects; however, since the individuals are also overridden by their level of unconsciousness, sometimes they do not realize how much the linguistic materiality plays with the meanings required to be situated between the local and the global. Second, “to know the Portuguese language” pointed to the inherent paradox of the internal dimension of the language: that it is oftentimes incomprehensible to the speakers themselves. This is not caused by a structure that makes the language hermetic, but by the discourses woven in an order imaginatively unattainable by ordinary speakers. Third, the Brazilian Portuguese located in the speakers themselves, in such a space of contradictions, when mentioned by the foreign-Other, is seen as a place of possibility, from which it might recover, even if timidly, its global perspective. Thus, we reaffirm the inseparability of this inside-outside contiguity similarly to the Möbius strip, which inseparability of interior and exterior does not allow us to determine the beginning and the end of a trajectory traversed by it. The relationships of language, never absent in cultural, political, and economic relations, cannot be reduced to isolation in an order of globalization. However, the space of resistance will not cease to exist, in which the universalizing norm has its effects made relative.

Notes 1. The notion of Imaginary Formations comes from the Lacanian concept of Imaginary. Pêcheux circumscribes it to the discourse processes in which the Imaginary Formations “determine the place where A and B attribute to each other the image they create of their own place and the other’s place” (Gadet and Hak 1993: 82–83). 2. The point of view adopted here is that addressed by Marcellesi and Gardin, within the field of Sociolinguistics founded during the 1930s in the Soviet context. See Marcellesi and Gardin (1975: 40–101). 3. I refer here to the Law Project number 1,676 from 1999, authored by Federal Deputy Aldo Rebelo (Dep. Federal Aldo Rebelo) from the Communist Party of Brazil—Partido Comunista do Brasil (PcdoB). This project proposed the promotion, protection, defense, and use of the Portuguese language, restricting the use of foreign words in audiovisual writings, official electronic documents, national public events, mass media, and publicity, among others. 4. The presence of black and mixed-race Brazilians in Public Brazilian Universities, especially in southern Brazil, is not expressive yet. Nevertheless, policies of racial quotas that have been implemented in the last years have promoted some modification of this scenario in Brazil. See Pensador Anônimo 2016.

The Imaginary in Academia

123

5. Inhabitants of the State of Rio Grande do Sul, southernmost Brazil. 6. Data from UFPR’s International Mobility Agency: http://internacional.ufpr.br/ site/.

References Bordieu, P. (2008). A economia das trocas linguísticas: O que falar quer dizer [The economy of linguistic exchange: What speaking means]. 2nd ed. São Paulo: Editora da USP. Camões: Instituto da cooperação e da língua. Accessed November 15, 2016. www.instituto-camoes.pt/. Costa, R. R. (2016). “Por que a cultura do Sul ficou de fora do retrato do Brasil na Olimpíada? [Why was Southern culture left out of Brazil’s presentation in the Olympics?]”. Gazeta do Povo. www.gazetadopovo.com.br/cadernog/g-ideias/por-que-a-cultura-do-sul-ficou-de-fora-do-retrato-do-brasil-naolimpiada-39kp1nc56n98jnjyropdhpxqw. Gadet, F. and Hak, T. (eds.). (1993). Por uma análise automática do discurso: Uma introdução à obra de Michel Pêcheux [For an automatic analysis of discourse: An introduction to the works of Michel Pêcheux]. Campinas: Editora da Unicamp. Lopes, V. (2002). Língua: vidas em Português [Language: Lives in Portuguese]. Documentary. Directed by Victor Lopes. Rio de Janeiro: TV Zero/Sambascope/ RioFilme. Marcellesi, J. B. and Gardin, B. (1975). Introdução à sociolinguística: A linguística social [Introduction to Sociolinguistics: The social linguistics]. Lisbon: Aster. Orlandi, E. P. (2009). A língua brasileira e outras histórias: Discurso sobre a língua e ensino no Brasil [The Brazilian language and other stories: Discourse on the language and its teaching in Brazil]. Campinas: Editora RG. Orlandi, E. P. (2012). “Espaços Linguísticos e seus desafios: convergências e divergências [Linguistic spaces and its challenges: Convergences and divergences]”. Rua 2, 18: 6–18. https://periodicos.sbu.unicamp.br/ojs/index.php/ rua/article/view/8638282. Pêcheux, M. (1988). Semântica e discurso: Uma crítica à afirmação do óbvio [Semantics and discourse: A critique of the affirmation of the obvious]. Campinas: Editora da Unicamp. Pensador Anônimo. (2016). “Negros quase triplicam no ensino superior no Brasil em 10 anos [Amount of Blacks in Brazilian higher education nearly tripled in 10 years]”. CEERT. www.ceert.org.br/noticias/educacao/10307/ negros-quase-triplicam-no-ensino-superior-no-brasil-em-10-anos. Victal, R. (2016). “Médico debocha de paciente na internet: ‘Não existe peleumonia’ [Physician mocks patient on the Internet: ‘No such thing as peleumonia’]”. G1. http://g1.globo.com/sp/campinas-regiao/noticia/2016/07/medico-debocha-depaciente-na-internet-nao-existe-peleumonia.html.

6

The Linguistic Atlas of Brazil Project Contributions Towards Knowledge, Teaching and Disclosure of Brazilian Portuguese Marcela Moura Torres Paim and Silvana Soares Costa Ribeiro

Introduction The history of the Brazilian Portuguese language reveals that its linguistic characteristics were from the early stages identified and recognized, becoming the object of study under different perspectives and focused on specific spaces or regions. The three main aspects, responsible for the linguistic setting of the country—indigenous, Portuguese and African— offer a broad range for study, not only in the perspective of the theories directly related to the languages, but also in relation to the history of the Brazilian nation. This chapter examines some of the aims, methodologies and data that have led and been put forward by the ALIB project (Atlas Linguístico do Brasil—Linguistic Atlas of Brazil) in that they may support the teaching of the Portuguese language, at every level, while perceived as a “glocal language”, a term defined in the introducing chapters of this book. This project aims to document, describe and interpret the reality of the Brazilian Portuguese language, with evident interface with different fields of organized knowledge, resulting from the fact that the history of a language is the history of the people who speak the language.

Brazilian Geolinguistics and the Importance of the Linguistic Atlas of Brazil Project for Describing the Portuguese Language Spoken in Brazil The particularities of the Brazilian Portuguese language, in relation to European Portuguese, were singled out since an early stage as shown by the first studies registered on this matter, even before independence of the colony. Dom Jerônimo Contador d’Argote (1725, apud Silva Neto, 1975: 561–564), after defining the dialect as a “diverse manner of speaking the same language”, referred to the “overseas dialects” mentioning India and Brazil, characterizing them as those with “many terms of the barbarian languages, and many words in ancient Portuguese”.

The Linguistic Atlas of Brazil Project

125

The first reference as a concrete example was made by Friar Luís de Monte Carmelo, in the Compendio de orthographia (Orthography Compendium) (1767, apud Teyssier, 1982: 77), pointing out the nondistinction between the open and closed pretonic vowels that were used in European Portuguese, for etymological reasons, and not verified in Brazilian Portuguese. In 1822, Jerônimo Soares Barbosa refers to this same fact. In the beginning of the twentieth century, Leite de Vasconcelos considers the Brazilian Portuguese language as a dialect—as he does in relation to the Portuguese language of Trás-os Montes (Portugal) and of other areas of the European continent—included under the category of “overseas dialects”, for which he traces characteristics in the fields of phonology, morphology, syntax and lexicon; he recognizes the plurality of the uses in Brazil alone, explaining that “due to its extension and variety of races that populate the country, there are dialectical differences” (1987: 134). The social history of Brazilian Portuguese1 (PB) indicates, in its origins, a tripod, basis of its linguistic conformation that assumes in the national territory of Brazil the Portuguese Language (PL): the transplanted Portuguese language—the European Portuguese language (EP); the indigenous languages existing at the time of the discovery, which still exist to date; the African languages brought over and the vestiges of these that still remain. The three strands responsible for the basic linguistic configuration2 of the country—indigenous, Portuguese and African—offer a wide field for study, not only from the perspective of the theories related directly to the languages, but also in relation to the history of the Brazilian nation. According to Souza (see Chapter 1 of this volume), in Brazil, the case of the interface between indigenous and Portuguese languages among the indigenous communities of the Western Amazon is an example of glocal languages. In these communities, speakers who have maintained their indigenous mother tongue recognize the importance of using the Portuguese language in official government contacts, even with limitations in Portuguese proficiency. This understanding of the need of Portuguese for its survival leads to its recontextualization and re-signification of the language by inhabiting it as its own, to be used as it wishes and when necessary, without being limited or prejudiced by non-indigenous “standard” linguistic norms. With reference to the Portuguese language, the fact that it is a language transplanted from another continent, establishing varied contacts, signals the need for reflections in the stricto sensu field of linguistic studies, involving: i. mechanisms of linguistic diffusion; ii. the conservative or innovative nature that the language assumes; iii. the continuity, or not, of a drift that accompanies the language from its origins;

126

Moura Torres Paim and Soares Costa Ribeiro

iv. the contribution received from languages with which there is coexistence in Brazilian territory and the reflections of this contact in the variety or in the varieties of use that developed; v. social and political implications of the different modalities that characterize the users of the Portuguese language, added to the potential offered, such as live documentation, the paths of the sedimentation of forms and uses that, when traced, offer relevant elucidations of for the social history of Brazil. The importance of the knowledge about the Brazilian linguistic situation, based on empirical data rigorously collected and exhaustively analyzed, led the Brazilian Government, in 1952, Decree No. 30. 643 of March 20, (Brasil 1952) in relation to the Portuguese language, to officially express itself about the need to produce a linguistic atlas of Brazil, recognizing in this type of scientific production an adequate and efficient manner of addressing the demands of the area. Various reasons hindered the realization of this wish of Brazilian linguists, despite governmental interest. After almost 50 years, the idea was once again resumed (i) due to the urgency of describing Brazilian Portuguese before data and facts capable of clarifying aspects of the linguistic history of the country are lost and (ii) due to the relevant role that, at present, Linguistic Geography has obtained in linguistic studies (according to Teles, Paim and Ribeiro 2015). For this purpose, the Linguistic Atlas of Brazil Project (ALiB Project)3 was conceived; the bases of this were presented in the seminar Caminhos e Perspectivas para a Geolingüística no Brasil (Paths and Perspectives for Geolinguistics in Brazil), held in 1996, in Salvador (BA—Brazil), with a representative number of Brazilian researchers of the area and with the presence of Prof. Dr. Michel Contini, of the Centre de Dialectologie de Grenoble, Director of the Atlas Linguistique Roman and member of the Executive Committee of Atlas Linguarum Europae. The Linguistic Atlas of Brazil Project (ALiB Project) is in essence a linguistic project because it aims to document, describe and interpret the reality of Brazilian Portuguese. It has, precisely in view of this characteristic, an evident interface with different fields of organized knowledge, due to the fact that the history of a language is the history of the people who speak the language. This characteristic that shrouds the ALiB Project has two evident implications: on one hand, it inspires and substantiates its conception of the plurality of knowledge; on the other, it permits that from these results the pedagogy of Portuguese may be benefited, both as a mother tongue and as a second language. Fourteen Brazilian university institutions integrate the ALiB Project, articulated through formal agreements. The coordination of the ALiB Project is under the responsibility of a National Committee, composed of

The Linguistic Atlas of Brazil Project

127

13 members. The ALiB Project offers, due to the nature of the data it proposes to collect, an interface with other fields of science that grant it its multi- and interdisciplinary characteristic. This is demonstrated below, in a purely illustrative manner. In relation to the type of collection foreseen, the data shall make evident different forms of linguistic behavior correlated to the type of discourse. The linguistic posture that the speaker assumes, depending on the nature of elocution, may offer to the studies, in the field of psychology and sociology, material for the analysis of knowledge on human behavior. The answers not given and the restrictions that are often clear in the speech of the informants, as well as the resource of using metaphors and circumlocution, permit reflections in the field of cultural studies, in general, indicating existing taboos, constructed through the course of history and motivated by differentiated impulses, as shown by Atlas Lingüístico Diatópico y Diastrático del Uruguay (ADDU), published in 2000. The set of data that the linguistic atlas reflects, in its broader range, can (i) demonstrate coordinates followed in the settlement of the country, unraveling questions about routes of penetration or offering elements of evidence of movements of people fixed in these locations or passing through them; (ii) assign the role of geographical accidents in the diffusion of linguistic habits—as can be seen, for example, in the role of rivers—or the isolation of phenomena that are kept behind mountains or incrusted in valleys; (iii) offer particular elements for specific studies such as in the field of medicine, with the names of diseases, diagnoses and cures from folk wisdom and which arise during questions in this direction, or in the field of geology, with the characterization and denominations of the types of land, for example, or even in the form of designating the elements of the biosocial world, an ample field for psychoanalysts. In education, the relation with the ALiB Project is highly significant. Results can offer an improved balance between teaching and learning and the reality of each region. Once the peculiarities of each area are described in relation to the dominant variety of the use of the language, it is possible to construct a more efficient model for teaching the vernacular. This can be perceived, for example, in relation to the pretonic mid-vowels, as exposed by Nascentes (1953) in relation to the dialectal division of Brazil, considering the open or closed realization of the pretonic mid-vowels as separating a linguistic Brazil of the north from a linguistic Brazil of the south. Such a situation is being confirmed with the analysis of the ALiB Project data, whose results could help the teaching and learning of Portuguese, contributing towards a relation between the teaching process and the linguistic reality of a particular area; for example, a profile could be presented of the realization of the “e” and “o” vowels, in a pretonic position, in relation to the rising, opening and closing of these vowels in words such as “cebola” (onion) and “tomate” (tomato), and contribute thus to discussions about linguistic discrimination.

128

Moura Torres Paim and Soares Costa Ribeiro

In relation to these aspects, of which there is no intention of exhausting the indication of possibilities of the interdisciplinarity of the Project, it is necessary to emphasize the implications of such an atlas for linguistic studies especially in their different aspects—semantics, lexicology, syntax, morphology, phonetics/phonology, pragmatics and discourse. The implementation of the ALiB Project, in 1996, fostered the resumption of discussions on methodology related to work of a Geolinguistic nature and to the preparation of work instruments adequate to the necessities for collecting empirical data (according to the Comitê Nacional do Projeto ALiB. (2001)). Such important facts for the development of Geolinguistics in Brazil make it possible to consider the ALiB Project as a milestone of a new phase in Brazilian Dialectology; it would be the fourth phase, as defined by Mota and Cardoso (2006), should the first three phases previously proposed by Ferreira and Cardoso (1994) be admitted, in extension to the two formerly proposed by Nascentes (1952, 1953). The ALiB Project, due to the nature of the data it conveys, can bring a contribution to the understanding of Brazilian Portuguese in its glocal aspect, permitting the recognition of forms of expression specific to the Portuguese language in Brazil when conveying new forms previously unknown. Such examples occur with denominations such as “carapanã” (mosquito), musse (jelly) and bolita (marble). In addition, there is the need to consider the plurality of uses of language in formal education, both for native speakers and for foreigners, through the presentation of such uses totally devoid of value judgment. Certainly, these are good points that can be listed as satisfactory results of the Project. The immediate result expected from the ALiB Project is, evidently, the production of the atlas itself, of which the initial volumes, Introduction and Linguistic Maps 1 were published in 2014: Volume 1—Introduction Volume 1, organized by Cardoso, Mota, Aguilera, Aragão, Isquerdo, Razky and Margotti (2014a), with 212 pages, presents the trajectory of the ALiB Project and describes the methodological steps taken. Added to this part is the reproduction of the methodological instruments used and appendices that complete the information. The methodological instruments used comprise: (i) linguistic questionnaires, presented in their initial version since, in the course of the research some alterations, motivated by field experience, were introduced; (ii) data annotation sheets of the locality and of the informant; and (iii) a survey control sheet, a tool that allows the research assistant, when marking the answers not obtained, to evaluate, immediately, the efficiency of the inquiry due to the percentage of answers given and thus allowing the inquirer to validate it or not.

The Linguistic Atlas of Brazil Project

129

In the Appendices there are the network points, a list of ALiB researchers, a list of inquirers and assistants and a list of Scientific Initiation and Technical Support Fellows, associated to different official research funding programs. Volume 2—Linguistic Maps 1 Volume 2 presents, in its 368 pages, a first set of linguistic maps that contemplate results, related to state capitals, in the fields of phonetics, lexicon and morphosyntax, bringing, in some cases, apart from the diatopic viewpoint, a diagenerational, diageneric and diastratic4 focus. In the form of an introduction, there are ten maps supplying information of a general nature on the political and geographical aspects of the country, detailing the geographical regions where the network points are identified. Subsequent to these come the linguistic maps themselves: • •



phonetic maps that address six facts described and analyzed in a set of 46 maps; lexical-semantic maps, primarily onomasiological, but including two semasiological maps, in a total of 106 maps contemplating eight of the 14 semantic areas included in the Lexical-semantic Questionnaire and focusing on the data in a general perspective—general diatopic maps—and with an indication per region—regional diatopic maps; morphosyntactic maps, totaling seven, with data related to the inflection of number and gender, distribution of pronouns of treatment and the use of the verb “ter” (to have) indicating existential value (there is/are).

Most of the maps are accompanied by notes with comments by the informants and remarks of the inquirer or of the party responsible for the preparation of the map, with the purpose of elucidating aspects considered as relevant. Data of this kind, added to the maps, are a source for other types of studies over and above exclusively linguistic studies, and may be consulted by non-linguistics.

Dialectology and (Under) Graduate Education The first question raised when studying linguistic variations is of establishing one’s own concept of variation. In relation to this Moreno Fernández (1998) comments that many scholars work with wide definitions and others with more restrictive definitions, but he prefers to visualize varieties as a set of elements or of linguistic standards associated to external factors, either situational, professional, social or geographic contexts. In this sense, when exposing these considerations about language and the phenomenon of variation, the questions that arise almost immediately

130

Moura Torres Paim and Soares Costa Ribeiro

from the students of the first semester of the undergraduate course in Languages at the Universidade Federal da Bahia, are: Why? How did it originate? The answers to these questions require the aid of disciplines such as Dialectology or Sociolinguistics because it is usual to have extralinguistic variables implied in the variation: aspects such as geography (geographical variation), history (historical variation), society (social variation) or the communicative situation, in its broader sense (stylistic variation). All of these factors may be responsible for or explain many cases of variation. Given this situation, the following question is possible: What is sought when studying linguistic variation? First, it could be said, as presented by Moreno Fernández (1998), that linguistic variation aims to explain the alternate use of certain forms of language, with the same truth value, under specific linguistic and extra-linguistic conditions: these can be units of different Geolinguistic origins present in a community, or of greater or lesser formal styles, among other possibilities. At the same time, the aim is to identify the characteristic possibilities of use (variants) of the different social groups: age group lexicon, profession lexicon, etc. As demonstrated by Moreno Fernández, the difficulties for analyzing linguistic variation exist and this is possible to visualize when discovering which social or stylistic variables explain the use of certain phonic, lexical or morphosyntactic variants of the same variable. To carry out a research on linguistic variation, there are various methodological itineraries. One of them is ethnographic fieldwork study: through continued coexistence within a social group or through direct observation of their speech, such as works based on the Third Wave in Sociolinguistics (according to Eckert, 2005, 2006, 2012). This procedure is of great interest, especially if the intention is to perform a qualitative analysis, in other words, to determine which are the lexical items that characteristically appear in each social group. Another methodological possibility, for the study of linguistic variation, is the interview. This can be used by the researcher to induce or provoke samples of variations in the various levels of analysis of the language. According to Labov (1972), the starting point for a sociolinguistic analysis is language use, the vernacular. For Labov the vernacular is the spoken language, casual talk without the concern of the prescriptive how to. Based on this, he mentions the importance of avoiding the paradox of the observer in data collection, which consists of providing the most suitable environment possible for an interviewer to approach an informal conversation without letting the concerns of the researcher with the linguistic fact interfere in the handling of the interview. This type of data collection guarantees the appearance of linguistic units in a determined quantity and, therefore, is more satisfactory for quantitative studies. Based on such a methodology, the possible objects of study are limited: it is possible to analyze the preference of one or more

The Linguistic Atlas of Brazil Project

131

groups of a community for certain linguistic forms according to the type of the interlocutor, and according to the situation (formal-informal style); it is possible to analyze the greater or lesser presence in certain social groups of ancient or modern, standard or non-standard forms (according to Chambers and Trudgill [s.d.]). Thus, when considering language as a social, historical and cognitive activity, it is possible to admit, as repeated by Marcuschi (2004), that it is susceptible to analysis and observation. In this manner, to understand is always to understand in the context of a relation with the other located in a culture and in a historic timeframe and this relation is always marked by an action. In this perspective, there is no direct relation between language and the world; there is a social act designating the world through a symbolic system whose semantics is constructed situationally. From the considerations drawn until now, it is possible to perceive that language as a social activity is used by all speakers, at all times in which they interact, either through oral means or through written means. As evoked by Cardoso (2008) and Aguilera and Romano (2016), the phenomenon of multi-dialect speech cannot be attributed to the present social stratification and, therefore, must be treated as a specific phenomenon of the modern world, typical of the present day. It is possible that it has become more evident today, motivated, at least, by factors such as the greater transit existing between different segments of the society as well as the displacement of rural masses to large urban agglomerations. The coexistence of many dialects, of many forms of using the language, either in its special diversity, or in the variety of occurrences distributed socially, spanning history, are linked to cultural diversity in its most different aspects. Faced with these questions, it is possible to reflect on the possibility of considering teaching as an intricate mixture of variation in the use of the Portuguese language, because, as Souza mentions (see Chapter 1 of this volume), when assessing how written language is organized differently, students are expected to appreciate respecting differences in expression. As highlighted by Razky, Lima and Oliveira (2006), a linguistic atlas has an important role in the basic education classroom. The authors, when discussing the diatopic variations in the phonetic level (117), bring examples of phenomena, such as the alternation between /l/ and /r/ in the /pl, kl/ groups in “planta” (plant) to “pranta” and “claro” (light) to “craro”, and the apocope of /r/ and /s/ in the absolute final position, in “beijá” for “beijar” (to kiss) and “costa” for “costas” (back) and the metathesis of the /r/ as in “partileira” for “prateleira” (shelf) documented in various linguistic atlases. These could be widely used in Portuguese language classrooms, using linguistic maps that “facilitate the reading of the linguistic complex and confirm, reaffirm or indicate changes of a phonetic nature”.

132

Moura Torres Paim and Soares Costa Ribeiro

In this line, the importance of bringing the discussion on linguistic diversity to the higher education classroom is evident, mainly in the considerations about language, culture and variation as well as about linguistic changes. As can be observed, the projection of a language in the contemporary world is directly related to the language of knowledge and innovation. Such is the language that is used by science, by digital platforms, and used to promote knowledge. An example of such use of language in knowledge and innovation is the word “deletar”, in Portuguese (from the English word delete), in the sense of “to erase”; it is used in general situations and not only in the context of Information Technology. Such a transposition has occurred in curricular components of the undergraduate course in Languages in various programs,5 as can be observed in the table below. Code

Name

Program

LETA13

Introduction to the study of the Portuguese language Introduction to dialectal studies

Study of the Portuguese language as a scientific object of analysis, from the perspective of linguistic diversity

LETC12

LETB12

LETC16

LETC15

Introduction to the study of Portuguese as a foreign language Diatopic variation in the Portuguese language Variation and teaching of the Portuguese language in Brazil

Presentation of the theoretical foundation for the development of studies in the field of Dialectology Study of the acquisition of Portuguese as a foreign language

Analysis and systematization of diatopic variations of the Portuguese language in Brazil and/or of the Portuguese language of Portugal A study of the sociolinguistic variations in Brazil focused on the social evaluation of the linguistic variables and the construction of a standard rule for teaching the Portuguese language

Source: Academic Superintendence UFBA—programs. Adapted. Available at: . Accessed on 30 January 2017.

In a similar manner in the graduate studies (master’s and doctorate) linked to the Graduate Program in Languages and Culture of the Universidade Federal da Bahia, it is possible to observe the presence of a range of curricular components that also approach the description of the complex relation between language, culture and linguistic variation (diatopic and diastratic variation). This is revealed by the curricular contents LET667 (space variation of the Portuguese language of Brazil), whose

The Linguistic Atlas of Brazil Project

133

program is about the “description and analysis of the diatopic variation observed in Brazilian Portuguese, with the purpose of identifying phenomena specific to regions and/or to define dialectal areas of Brazil”. The same happens with LET676 (social variation of the Portuguese language of Brazil), whose program is about “the variation and changes in the progress of the Brazilian Portuguese language, observable in studies of social dialects”. In the programs of the curricular components of the Language courses (in their different qualifications with the purpose of training Portuguese Language teachers), variation is discussed, and how and why it is present in natural languages. The objective is to lead the students to perceive the conflict that is established between linguistic variations and the standard, which normally characterizes most diverse societies, even pre-literate societies. It is expected that the student understands that language cannot be reduced to a standard rule and that, for linguistic studies, all varieties are legitimate and susceptible to investigation. Both in undergraduate and in graduate studies, it is possible to confirm the students’ understanding of language as a heterogeneous phenomenon and their awareness that words such as right, wrong, correct, incorrect, ugly and pretty should be removed from the linguistic repertoire and replaced by “adequate” or “inadequate”, according to the communicative context in which the speaker is inserted (according to Paim 2011). As for teachers of Portuguese, it is important to demonstrate that the student is part of a linguistic community and that the phenomenon of linguistic variation remains permanent and, therefore, it is possible to opt to behave linguistically in an informal manner in the community, in colloquial contexts, if the objective is not to set oneself apart from the context. However, if the situation of communication is different, such as a job interview, the student should perceive that it is necessary to use the rule closest to the standard language, since a more formal linguistic performance shall be required in this situation. Thus, this change in attitude, this awareness-raising among the students, will most likely prevent linguistic discrimination.

Dialectology and Linguistic Atlases: Contributions for Observing the Portuguese Language as a Local or Glocal Language Linguistic atlases excel for their social contribution and for the contribution they can bring to economy in the teaching-learning process of the Portuguese language, allowing knowledge of the particular context of the domain of the Portuguese language, to explain the divergence and convergence registered in the national territory, by relating dialectal areas to socio-cultural areas and by offering a set of linguistic data to improve the teaching of the Portuguese language.

134

Moura Torres Paim and Soares Costa Ribeiro

The initial published volumes of the Linguistic Atlas of Brazil pinpointed the understanding of the linguistic variation, in order to eliminate social prejudice and discrimination based on the reality of the language. They demonstrated how differences and convergences can coexist while recognizing, nevertheless, the validity of the existence of a cultured standard necessary for official communication, for teaching, for formal discourse and still providing options for different groups of speakers. Furthermore, volumes 1 and 2 of the atlas can supply elements for the construction of the history of the Portuguese language in Brazil, either though the paths followed or due to the nature of the changes made, by means of the establishment of particular linguistic, social and geographical categories, or also through contacts with other languages and consequent borrowings. Accordingly, the atlas promotes the updating of dictionaries of the Portuguese language and can help the elaboration of a Brazilian Portuguese grammar. The contribution of the Linguistic Atlas of Brazil for the teaching of the main official language in this country can be made evident with regard to the work of its scholars and to other areas of linguistic studies, to research in related areas and to its pedagogy (grammarians, authors of text books for elementary and secondary school, and teachers), in order to increase teaching and learning as well as an awareness of the Brazilian multidialectal character. This illustrates, as Guilherme presents (see Chapter 2 of this book), plural multilingualism, through the possibility of looking at language and culture from a “glocal” point of view that encompasses the plural loci of enunciation for the same language, through their own experience in different loci of enunciation. The published volumes of the atlas address variations of the Portuguese language, focused on different levels—phonetic-phonological, semanticlexical and morphosyntactic—and in varied social perspectives—age differences, gender and schooling—as illustrated in some of the linguistic maps presented below. a.

Depending on where the speaker of the language is located or depending on the social situation (diagenerational or diageneric), it is possible to verify the documented denominations of the words “ruge”, “blush” or “carmim”.

As can be visualized in Map L26 (Figure 6.1) of the Linguistic Atlas of Brazil, three lexical items were observed to denominate the product women spread on their face, on their cheeks, to become rosier: blush, ruge and carmim. The collected data are laid out, on this map, in such a manner as to permit the information of the diatopic distribution and preference of use, through the quantitative indication of the data through pie charts. The denominations verified in the state capitals of Brazil reveal that the variant “ruge” is the most popular, occurring with greater

Source: Cardoso (2014b: 331)

Figure 6.1 Representative diatopic denominations for “ruge”—Map L26—state capitals

136

Moura Torres Paim and Soares Costa Ribeiro

frequency, in Brazilian capital cities. This demonstrates that the denomination of French origin, diatopically, is maintained as preferable, which reveals, from a cultural viewpoint, the effects of a period with a greater relation between Portuguese and French. In relation to the diagenerational, in the map below, only the variants “ruge” and “blush” were considered, as illustrated in Figure 6.2, since the denomination “carmim” occurred only in two capital cities, Campo Grande and Rio de Janeiro. Map L26G, in Figure 6.2, presents the distribution of the uses of “ruge” and “blush” according to the preference by age group. It demonstrates that from a diagenerational viewpoint, the denomination “ruge” is the most frequent, in Brazilian capital cities, in the discourse of the older informants, according to age group. None of these older individuals recognized the use of the older term and made comments such as: “I find the world ruge strange, I believe it is older” (Manaus, woman, age group 1 and university level) and “When I was a child, it was ‘ruge’, now it is ‘blush’” (Curitiba, woman age group 2, university level). The generational choice for one or another of the forms demonstrated in this Atlas map a chronology of these variants that is linked to different moments of political-cultural influence; in other words, younger people use “blush”, which reveals the influence of the English language, while the elderly still maintain a link to a previous moment of the greater presence of French in the Portuguese language due, also, to political-cultural matters (according to Paim 2015). In relation to the diageneric factor (Figure 6.3), emphasis is given to the fact that the denominations were obtained more easily in the speech of the female informants since this is an issue more related to the female universe. Since regarding women’s make-up, the answers “maquiagem”, “pó” and “base” as denominations of a more general character were mentioned by male informants who, after clarification and insistence by the inquirer, were able to answer what was being asked. b. Depending where the speaker of the language is located or of the lexical choice (diatopic variation), it is possible to use the denominations “mandioca”, “aipim” or “macaxeira” (regarding a vegetable of Brazilian origin, cassava, etc.). Map L08 demonstrates the diatopic distribution of the documented variants for the root that is white inside and covered by a dark brown peel, which has to be cooked to be eaten. This food item is well known and used throughout the country and, when a photograph is observed, it has different denominations in each region. The country can be subdivided into three great dialectal areas: Area A—in the North and Northeast we have the variant “macaxeira” as preference of use; Area B—in the Center

Source: Cardoso (2014b: 335)

Figure 6.2 Representative denominations for “ruge”—Map L26G—state capital cities—diagenerational variation

Source: Cardoso (2014b: 337)

Figure 6.3 Representative denominations for “ruge”—Map L26S—state capital cities—diageneric variations

The Linguistic Atlas of Brazil Project

139

of Brazil, the variant used is “mandioca”; and Area C—in the Southeast and South, the choice lies with “aipim”. c.

Depending on where the speaker of the language is found, it is possible to hear a “bom dia” in different forms: in João Pessoa (PB) as “bom [dia]” or in São Paulo [dAia]. d. Depending on the informant, the plural of the word “degrau” can vary from the normative standard—Degraus—altering to other forms with Degrais and Degrau (steps). In relation to a morphosyntactic fact, Map M01 reveals the diatopic distribution of the variants documented for use in the plural of certain words ending with the vowel “u”. We have the standard form “degrau × degraus” (steps), which is in use in all the capital cities of Brazil, coexisting with the variant forms “degrau × degrau”—zero phonetics (with the inclusion or not of a number indicator in the determinant) and “degrau × degrais”—through analogy with “avental/aventais” (apron/aprons), since “avental”, in the singular, is pronounced with a semivowel [w], which resulted from the semi-vocalization of the ending. As it is possible to visualize, the linguistic atlas exhibits the visualization of the Portuguese language, at a given moment and in a specific geographical area. At this stage, the synchrony of the language is registered, which can, in turn, when exhibiting its variation, also outline paths for its change. It is important to emphasize that the work continues; there are other studies in progress. When comparing the empirical work, already presented in scientific articles, it is possible that other phenomena will be brought to light. In this respect, the relationship of the language with different areas of knowledge is also made visible through distinct behaviors, through people’s lives. After all, it is through language that people (i) spatialize— whether they are from São Paulo or from Pernambuco, they buy aipim or macaxeira to cook (as previously seen in Figure 6.4) or someone from Paraiba or from São Paulo saying “bom dia” (Figure 6.5); (ii) are placed socially, revealing their social class, e.g. when looking for a job (“trabalho” or “trabaio”); and (iii) use time without losing the temporal bond—people buy “blush” in the “shopping center”, but, at home, they continue spreading “ruge” on their faces. The same speaker can reveal the normative rule of the school and a higher level of education when using “degrais” or correcting himself using “degraus” (standard form), when in a formal situation of linguistic monitoring. As shown in Figure 6.6, there were variations in the speech of the informants with university level (capital cities) while the “standard form” is still based on the grammar learned in the school as the “only” form possible.

Source: Cardoso (2014b: 185)

Figure 6.4 Representative diatopic denominations for “aipim”—Map L08—state capitals

Source: Cardoso (2014b: 123)

Figure 6.5 Consonants “t” and “d” before the vowel “i”—Map F06 C 1—state capitals

Source: Cardoso (2014b: 345)

Figure 6.6 Use of the plural for the word “Degrau”—Map M01—state capitals

The Linguistic Atlas of Brazil Project

143

On the whole, the linguistic atlases can point out directions for new findings about the language, offering elements for the teaching and learning of the mother tongue as adapted to the reality of each region, allowing the recognition of the linguistic features of each area and its correlation to the adoption of methodologies for teaching the vernacular. The relation of Dialectology with the user society is explained through specific means of contribution, by describing the particular context of the Portuguese language. By making evident the differences and convergences registered in Brazilian territory, ALiB enables the discussion with students about the linguistic variation considered as a distinguishing phenomenon of any language. This study also helps to eliminate discrimination related to value judgments that denote erroneous notions of “right”, “wrong”, “ugly” and “pretty”, and social prejudice caused by the everyday performance of the language. Hence, the work with the linguistic atlas in the classroom provides a sample of differences and convergences that coexist and the recognition of the validity of the existence of a formal standard that is necessary for official communication, for teaching and formal speech as well as offering samples to different groups of speakers. The Linguistic Atlas of Brazil can significantly contribute to the teaching of the language by supplying elements for the construction of the history of the Portuguese language in Brazil, as well as by providing abundant examples of the occurrences of linguistic variation documented in loco. Furthermore, it can make feasible the updating of dictionaries of the Portuguese language, once it has within its objectives the project “Dicionario Dialetal Brasileiro (DDB)”6 (Brazilian Dialectal Dictionary), and allows a grammar based on the context of the Brazilian Portuguese language, aiming, among other objectives, at the teaching of the mother tongue with a perspective that highlights linguistic varieties. As far as the DDB is concerned, this is a project in progress that uses the ALiB corpus as a database and extensively presents all the documented variables. This project as linked to the Linguistic Atlas of Brazil Project will hopefully enable the dissemination of not yet registered forms in the general dictionaries of the Portuguese language and provide a wider circulation of what has already been.

Conclusion and Suggestions of Possible Paths for Future Research Improvement of education is an agent of progress and one that plays an important role in the quality of production and improvement of the qualified workforce; the contribution of Geolinguistic studies is of fundamental value in order to reach this goal. The work with the linguistic

144

Moura Torres Paim and Soares Costa Ribeiro

atlas in the classroom stimulates teaching and learning that is more suited to the reality of each region. The singularities of each area are described, and the dominant variety of the language use at the location is characterized. Therefore, a more efficient model of teaching the vernacular is made possible with the use of linguistic data that allow the adjustment of the pedagogical material to the linguistic reality of each region and the understanding of the multidialectal nature of Brazil. The reflections presented, with the examples of illustrative maps, which do not intend to be exhaustive, had the purpose of raising awareness in relation to multidialectalism, by emphasizing the different realities of the Brazilian Portuguese language. The importance of its recognition as a means of expression responds to the need for observing the pluralism of linguistic use in formal education. With this in mind, the contribution of the ALiB Project, and its outcomes, displays the Portuguese language spoken in Brazil and reflects cultural identifications and hybridizations that require a reflection on the subtleties of intercultural interactions among users of the same language. At present, Dialectology has an important role in the reflection on the Portuguese language, linking theory and practice so that students and teachers can find the ways for further developing the teaching and learning of the mother tongue, taking into consideration its variety. In this respect, the use of the linguistic atlas in the classroom contributes to an understanding that the language must always be an instrument of socialization that benefits history and sources of knowledge and, above all, promotes the humanization of all and any speaker in daily communication against discrimination or stigmatization.

Notes 1. With reference to the social history of Brazilian Portuguese and a broader approach, consultation to the site of the Brazilian Portuguese language history (PHPB), available at: . Accessed on: 27 January 2017. 2. In relation to the basic linguistic configuration, it is observed that the immigration languages, indigenous languages and African languages shall not be approached in this chapter, concentrating the information exclusively on information of the Brazilian Portuguese language. 3. The ALiB Project is presided over by Prof. Suzana Alice Cardoso and has as executive director Prof. Jacyra Andrade Mota, both from Universidade Federal da Bahia. 4. Diatopic—in different places and regions of the linguistic area, different dialects are spoken. Diagenerational—type of variation conditioned by the age group of the speakers. Diageneric—type of variation conditioned by the gender/sex of the speakers. Diastratic—in different social groups (according to degree of schooling, profession, etc.), different sociolects are used.

The Linguistic Atlas of Brazil Project

145

5. Undergraduate and Graduate Curricula of UFBA may be consulted on the official page of Universidade Federal da Bahia—Academic Superintendence— courses. Available at: . Accessed on: 30 January 2017. 6. The “Dicionário Dialetal Brasileiro (DDB)” (Brazilian Dialectical Dictionary) project is under the coordination of Américo Venâncio Machado Filho, professor of Universidade Federal da Bahia.

References Aguilera, V. and Romano, V. (2016). A Geolinguística no Brasil: caminhos percorridos, horizontes alcançados. Londrina: Eduel. Atlas Lingüístico Diatópico y Diastrático del Uruguay (ADDU). (2000). Directed by Thun, H. e Elizaincín, A. Fasc. A.1. Kiel: Westensee. Brasil (1952). Decreto n. 30.643, de 20 de março de 1952. Institui o Centro de Pesquisas da Casa de Rui Barbosa e dispõe sobre seu funcionamento. Diário Oficial [da] República Federativa do Brasil, Senado Federal, Subsecretaria de Informações, Brasília, DF, 20 mar. Cardoso, S. A. M. (2008). Sociedade pluridialetal, variação e ensino da língua materna. In E. Mendes and M. L. S. Castro (eds.). Saberes em português: ensino e formação docente (pp. 11–25). Campinas: Pontes. Cardoso, S., Mota, J. A., Aguilera, Vanderci A., Aragão, Maria do S. de S, Isquerdo, Aparecida N., Razky, Abdelhak, and Margotti, Felício W. (2014a). Atlas Linguístico do Brasil. Londrina: Eduel. Cardoso, S., Mota, J. A., Aguilera, Vanderci A., Aragão, Maria do S. de S, Isquerdo, Aparecida N., Razky, Abdelhak, and Margotti, Felício W. (2014b). Atlas Linguístico do Brasil. Londrina: Eduel. Chambers, J. K. and Trudgill, P. (s.d). Dialectology. Cambridge: University Press. Comitê Nacional do Projeto ALiB. (2001). Atlas Lingüístico do Brasil. Questionários. Londrina: Eduel. Eckert, P. (2005). Variation, convention and social meaning. Paper Presented at the Annual Meeting of the Linguistic Society of America, Oakland, CA. Eckert, P. (2006). Communities of practice. In B. Keith and A. H. Anderson (eds.). Encyclopedia of Language and Linguistics (pp. 683–685). Oxford: Elsevier. Eckert, P. (2012). Three waves of variation study: The emergence of meaning in the study of sociolinguistic variation. Annual Review of Anthropology, 24, 87–100. Palo Alto. Ferreira, C. and Cardoso, S. (1994). A dialetologia no Brasil. São Paulo: Contexto. Labov, W. (1972). Sociolinguistic Patterns. Philadelphia: University of Pennsylvania Press. Marcuschi, L. A. (2004). O léxico: lista, rede ou cognição social? In L. Negri, M. J. Foltran and R. P. de Oliveira (eds.). Sentido e Significação (pp. 263– 284). São Paulo: Pontes. Moreno Fernández, F. (1998). Principios de sociolingüística y sociología del lenguaje. Barcelona: Editorial Ariel. Mota, J. and Cardoso, S. (2006). Sobre a dialetologia no Brasil. In J. A. Mota and S. A. M. Cardoso (eds.). Documentos 2. Projeto Atlas Lingüístico do Brasil (pp. 15–26). Salvador: Quarteto. Nascentes, A. (1952). Études dialectologiques du Brésil. ORBIS: Bulletin International de Documentation Linguistique, Louvain, 1, 1, 181–184. Nascentes, A. (1953). Études dialectologiques du Brésil. ORBIS: Bulletin International de Documentation Linguistique, Louvain, 2, 2, 438–444.

146

Moura Torres Paim and Soares Costa Ribeiro

Paim, M. M. T. (2011). A variação lexical nos campos semânticos corpo humano e ciclos da vida: o que revelam os dados do Projeto Atlas Linguístico do Brasil. Diadorim, 8, 143–160. Paim, M. M. T. (2015). A variação semântico-lexical e a identidade social de faixa etária nas capitais do Brasil. In K. J. Kragh and J. Lindschouw (eds.). Les variations diasystématiques et leurs interdépendances dans les langues romanes (pp. 253–264). Strasbourg: Éditions de linguistique et de philologie. Razky, A., Lima, A., and Oliveira, M. (2006). Atlas linguísticos: contribuições para a educação básica. In J. A. Mota and S. A. M. Cardoso (eds.). Documentos 2. Projeto Atlas Lingüístico do Brasil (pp. 109–126). Salvador: Quarteto. Silva Neto, S. da. (1975). História da língua portuguesa. 3rd ed. Rio de Janeiro: Presença. Teles, A. R. T. F, Paim, M. M. T and Ribeiro, S. S. C. (2015). Les atlas linguistiques de Sergipe: ALS et ALS II, In Contini, M. and Lai, J.-P. (eds.). Revue Géolinguistique. Centre de Dialectologie. Grenoble, Univ. Grenoble Alpes, 15, 83–100. Teyssier, P. (1982). História da língua portuguesa. Tradução de Celso Cunha. Lisboa: Sá da Costa. Vasconcelos, L. (1987). Esquisse d’une dialectologie portugaise. Lisboa: Instituto Nacional de Investigação Científica / Centro de Linguística da Universidade de Lisboa.

Section IV

Spanish as Glocal Language

7

Comparisons Between Spanish and Portuguese Proposals for University Teaching Adrián Pablo Fanjul

Introduction This chapter is about a series of teaching and research projects carried out in the last ten years at the University of São Paulo in the Spanish Language and Literature programme, specifically using the compared approach between this language and Portuguese. These initiatives have produced a series of specific undergraduate courses, which have established rich interactions with the research work. We intend this chapter to be viewed as propositional, that is, rather an invitation for reflection than an opportunity for reporting. Indeed, the dialogue that we have initiated with Brazilian higher education groups based on the discussions and the publications where we promote our work has generated the development of similar experiences in the country, as well as the reorientation of our own practices. In this text, we will engage in a dialogue with the “glocalization” concept (Guilherme 2018), which forges the identity of this book, as we will deal with a historical and particular view of what is perceived as the world extension of each language, thus promoting a practice in which what tends to be global and what we can consider as local interact and impact each other and themselves. This chapter will be organized in four sections. In the first one, comparison as practice will be questioned as well as the issues that emerge when dealing with historical languages. Then will be outlined a brief history of situations and transfers in comparative practices developed in Brazil when comparing Spanish and Portuguese. The third section will explain the features of the new courses, which are the main focus of this text, by distinguishing three levels of problems involved in the comparative work in higher education which tries to connect the linguistic with the sociohistorical dimension. Finally, in the fourth section, some practices developed in the course will be described, while appraising their achievements and risks. In the final considerations section, I shall comment on perspectives and projections for this work at several levels.

150

Adrián Pablo Fanjul

A Discursive, Politically Determined, Practice It seems adequate to start with a topic that we usually approach when making a comparison: How do we perceive “language”? Shouldn’t this questioning precede the comparison between two languages? Based on which features may the identity of a specific language be considered a given fact? For that purpose, we believe that a distinction between the linguistic dynamics of language and its delimitation is necessary. On the one hand, the linguistic dimension is located, according to the theoretical approach that is adopted, in the verbal exchange of oral and written speech between speakers, or in the system underlying it, according to each adopted theoretical framework. On the other hand, the perception of each language as specific—“Spanish”, “French”, “Portuguese”—emerges from a historical series of political actions that name it as such, by stabilizing it with grammaticalization (Auroux 1994) and standardization tools, which institutionalize it and influence the speaker’s behaviour. Such dialectics keeps reaffirming the existence of such language and the identity perceptions to which it is related. This process, which has been happening in the world with special recurrence and emphasis since the appearance of national states at the beginning of the modern age and which is reaffirmed in several demonstrations of national ideologies (Hobsbawn 1990), is what makes us consider the linguistic identity mainly as a political issue. And, therefore, it is constantly reformulated and reviewed by description of actions, a premise of all comparative activities. It is also necessary to state that comparison itself is a discursive practice, hence socially and historically determined. In linguistics, this practice is constituted through discourses with requirements of organicity and systematicity that, however, don’t refrain from inserting a subject and a non-coincidence between the real object and the knowledge object (Henry 1977), as every classification and taxonomy necessarily sets aside other possible views for the linguistic phenomenon. The comparison between languages also happens in other fields of human activity, such as teaching and learning spaces, literary and artistic expression, means of communication, and even in non-institutionalized conversations, mainly in situations where there is contact with speakers of other languages or their cultural productions. Such practices, despite being much more permeable to stereotypes and to common sense than the pedagogical or scientific discourses, are part of the contexts in which two or more languages are contrasted. In sum, the comparison between languages is here perceived as a discursive practice about different political views on a constitutively heterogeneous linguistic reality. In the following section, reflection upon practices regarding the Spanish language in Brazil will draw special attention to some historical transfers.

Comparisons Between Spanish and Portuguese

151

Comparative Research and the Case of the Spanish Language in Brazil Some of the first linguistic tools for Spanish teaching in Brazil already used to make comparative observations. The Spanish grammar published by Antenor Nascentes (1920) emphasized what it called a large “similarity” between this language and Portuguese, concentrating the differences on the morphology and on the lexicon, mainly in a different meaning distribution for the same items. The syntax was presented as the level in which less “differences” were found. These premises consisted of a real tripod (big similarity between languages, syntax with no differences, lexicon as the main concern), which will bring long-term consequences for Spanish teaching and research practices in Brazil. Thus, advertising and teaching tools, as well as, in a general way, the discourse about the relations between Portuguese and Spanish, relied, throughout all the 20th century, on “easiness” representations, which reduced the learning process to finding “difficulties”, whose mastery would enable a “control” over the other language. As explained in Fanjul and González (2014), a combination of factors since the 1980s would bring about some changes in this scenario. We understand these factors in two levels: the economic and symbolic asset exchanges and the language perception based on determined spaces of knowledge production. On the one hand, there is a considerable increase in the activities which involve economic exchange and people transfers between the countries in South America, even before the MERCOSUR treaty was officialized in 1993, which may be seen as a superstructure expression of this change. Between 1985 and 1994, the internal trade in the region between Brazil, a country where Portuguese prevails as the official language and has a generalized use, Argentina and Uruguay, countries where most people are Spanish speakers, increased by 22% per year (Fausto and Devoto 2004: 506). This growth was followed by a concomitant increase in the circulation of art and culture, as well as tourism, between these countries. They are also the periods immediately after the military dictatorships in the region; in Brazil it was their final retreat, favouring exchanges between the countries and a growing interest in their neighbours. These new production and circulation conditions were expected to bring more of a disposition to learn each other’s language, and it effectively happened. However, in order to understand the specific ways in which this new regard was conveyed, it is important to take into consideration the other level of factors mentioned: the new practices regarding the linguistic knowledge. Some changes started to happen in the delimitation of languages as geopolitical constructions, mainly with the perception of “Brazilian Portuguese” as a subject to be studied in the field of language sciences. Actually, the development of specific linguistic approaches in Brazilian universities favoured questionings about the particularities of the

152

Adrián Pablo Fanjul

Portuguese language in Brazil, in a process which assigned to this specific item a status of something that deserves to be studied, starting to overcome what Faraco (2011: 271) defines as “a history of, at least, one and a half century in which the Brazilian society built a depreciative image about its language”. Brazilian Portuguese stopped being seen as a group of “deviations” from an origin and started a quest to recognize and explain the system in its own dynamics. One of the pioneer works in this area was Fernando Tarallo’s thesis (1983), which, based on the description of three possible modes for the relative clauses in Brazilian Portuguese, paved the way for the perception of important differences between Brazilian Portuguese and the European Portuguese regarding the occurrence of personal pronouns. Other aspects of the dynamics of Portuguese in Brazil started to be researched at that time in other studies, some of them in the diachronic perspective. On the other hand, in the field of linguistics, the discourse analysis approaches which emerged in the country at the same time developed ideas about a founding split of Portuguese in Brazil as a sign of the colonization relationship, “a birthmark which is worked out in several different ways throughout its history” (Orlandi 2002: 28). These conditions have enabled the comparison with the Spanish language to be thought of in different contrastive terms: Brazilian Portuguese with a syntax that presents some characteristics that strongly distance it not only from the European Portuguese but fundamentally from the “standard norm” which comes from an imaginary representation of “pure” Portuguese. Observed in the Portuguese effectively spoken by the educated Brazilians and taking into account the common traces within its heterogeneity, the differences with Spanish could not be reduced anymore to lexicon, like in the representation that used to be hegemonic. A wide field was opened to study the processes which, as it was becoming clear, largely involve the syntax and central semantic categories such as determination and thematic roles. Meanwhile, developments in discursive studies started to produce specific reflections. In 1994, two studies coincided as pioneers in the field of comparative studies between the languages discussed here. Neide González’s doctoral thesis (1994) analyzed the acquisition of the Spanish language by Brazilians focusing on personal pronouns, and it set a landmark in comparative observations between Brazilian Portuguese and Spanish, languages which are characterized in the thesis as “inversely asymmetric” regarding the occurrence of tonic and atonic pronouns. The Brazilian Portuguese tendency to repeat referents with full pronominal subjects and to produce null objects, opposite distribution to Spanish in all its variations, appeared as a productive counterpoint space which enabled a series of studies about this and other aspects of the linguistic dynamics related to it. On the other hand, Serrani’s article (1994) showed an experimental study in which Brazilians and Argentines produce negative and

Comparisons Between Spanish and Portuguese

153

refusal enunciations based on a supposed situation. Its disclosure showed a route to comparatively question the articulation between language and sociohistorical factors avoiding stereotypes, which had an impact on later studies and on teacher formation in both languages. In sum, new conditions for the circulation of people and cultural assets, along with changes in the way to perceive languages and linguistics, gave birth to comparative research practices guided by approaches which take into account the intrinsic heterogeneity of languages. It was in this process that the main focus of this chapter was developed: a plan for comparative studies in Languages and Literature in Brazilian universities, in the field of relations between Portuguese and Spanish languages. The description in the following sections about how this proposal was put into shape in specific courses will develop along with an explanation about the particular way that we conceive the proximity between Portuguese and Spanish in South America today.

The Development of Comparative Studies Courses The University of São Paulo was one of the central spaces in the academic area for the changes described above. The consolidation of several research projects started to give birth in 2004 to the first courses about comparative studies between Brazilian Portuguese and Spanish in the Languages and Literature programme. In their most recent forms, these courses have put together comparison topics which can be sorted into three main categories: the political delimitation of languages, discursivity and linguistic dynamics in its systemic aspects. They will be considered in this sequence. Glotopolitical Dimension As previously explained, in the first section, about the languages as institutionalized constructions, we consider it necessary for the comparative studies approach to include a component that we may call “glotopolitical”. In a wide interpretation, such as the one in Arnoux (2008: 10–20), glotopolitics is defined as a field that is interested in governmental interventions in languages and also in representations and fantasies about them in the social formation, as well as in discourses where the ideologies about languages are materialized. Besides, a glotopolitical perspective is useful not only to destabilize the ideological naturalization of what is considered at a given historical period as “one language”, supposedly homogeneous, but it also introduces something comparable regarding the space of each contrasted language. How was each of them introduced in colonization and conquest in historical contexts? Which policies were confronted in each case regarding the original people’s languages? Which institutional device followed the development of written and oral

154

Adrián Pablo Fanjul

discursive practices in each—“Hispanic” or “Luso”—1 colonial space? What may comparative observations of these processes tell us? In which way did the independence and the formation of national states modify and/or reproduce the relations between languages? How were these relations affected by the migration processes and by the policies adopted in face of these population movements? How has the linguistic planning been made regarding the “official” languages, the ones from minority sectors, the foreign languages and, in each territory, the “other language” out of the two we compare? What is the place for the Brazilian Portuguese and the particularities of Spanish in America in the power and force relations which determine the political management of these languages in the world? Such questions guide the observation of languages as glotopolitical entities and, in our opinion, they should not be separated from the strictly comparative reflections; on the contrary, they must frame them. In Latin America, the presence of both Portuguese and Spanish is a result of colonization processes. Their implementation peculiarities were different but not so much as to show homogeneous standards, which allow us to affirm that this process happened qualitatively and in a constantly different way in the realms of both languages. Both in one space and in the other some phenomena happened in unequal ways, such as the adoption by the colonizers of “general languages” with an Amerindian base for catechization (Lagorio 2011; Mariani 2003), the presence of African-Iberian speeches and creolization phenomena (Lipsky 1994; Mattos and Silva 2004). In the current context in South America, the Portuguese and Spanish spaces are affected by the regional integration process. But this scenario is not at all limited to these two languages, once it appears crossed by other ones. On the one hand, there is the English language. The mode and the political meaning of this crossing is determined by the position of English as the most far reaching international language, but also, in particular, by its relations with centres of political and cultural power which act in the region and which generate similar representations in the spaces of the languages that are discussed here. On the other hand, it is necessary to consider, in Latin America, an intrinsic contradiction pointed out by Souza (2010: 290), which says that the national state is “simultaneously post-colonial—in relation to the old colonial power—and neocolonial— in relation to the indigenous cultures” and, we could add, in relation to the heterogeneity within the “national language” itself. There are, in the countries that form this region of the world, diglossia situations regarding different Amerindian languages. There were also political oscillations around a more autonomous or more Eurocentric management of the official language itself. In this respect, countless critical studies written about the standardization conflicts about Portuguese in Brazil or about Spanish in Hispanic-American countries are easily identifiable. A less popular study, but necessary for the type of formation

Comparisons Between Spanish and Portuguese

155

that we provide, is Lagares (2016), which comparatively approaches the conflict situations in the delimitation and normalization of Spanish and Portuguese. Comparison, Discursivity and Proximity We believe it is necessary to briefly present how we see Portuguese and Spanish, as proximate languages, when they are analyzed within the discourse level, in which the linguistic aspects interact with history. In a recently published book which synthesizes a great part of our research about this issue, we defend the need to approach the comparative discursive studies in different languages by drawing upon a sociohistorical specification which cannot coincide with all the international representation of “one language”: Can this sociohistorical dimension considered in the proximity be the same for the same pair of languages in qualitatively different regions in the world precisely in the sociohistorical dimension? Can we consider the proximity between Portuguese and Spanish in Europe and in Latin America as the same? Would it be viable, when it comes to regions of the world that differ from each other on the development of recognizably distinct social formations, and where in each one of them the languages have had very different implementations and contact situations, because of, precisely, the particularities of this development? It seems that the proximity/distance should be specified: specific languages in a specific space, in a specific sociohistorical formation. (Fanjul 2017: 31–32, italics in the original)2 Considering the linguistic aspects as the materiality where discursive traces that form a sociohistorical memory are inscribed (Pêcheux 1988), the hypothesis that has conducted our work is the one that, in contemporary Brazilian Portuguese and Spanish in South America, there are memory spaces3 which are partially shared for discursive sequences in certain domains, which enables the operation of implicit items and paraphrases transversally crossing the two languages. Therefore, when a series of texts are observed in these languages in the light of categories of analysis which belong to the discourse itself, what we find is a “sway” of approaching and distancing which we call “contradictory proximity”. Indeed, dozens of comparative research studies have been produced in the last two decades considering some enunciation modes or the construction of discourse objects in several genres as contrast criteria. Only to mention a few of them, Lima (2012) about apologies in media interactions, Menón (2016) about the construction of the “famous” character in interviews, Menezes (2012) about marginal and

156

Adrián Pablo Fanjul

disciplinarian enunciators in samba and tango lyrics, Gallardo (2013) about politeness in corporate emails, Russo (2013) about polyphony in soccer goal narrations on the radio, Zoppi Fontana and Celada (2009) about subjectivation processes in dialogues related to the imaginary representation of rights and duties, or Pizzutiello (2017) about acting and voices in propaganda documentaries in military dictatorships in the 1970s. When comparing the ways that themes, objects and pictures are articulated in interdiscursive relations, we realize the effect of social formations with analogue inequalities, but in very different degrees and extensions, also with analogue contradictions in the sociocultural identification processes and in the occurrence of symbolic violence, and with political histories that present many—sometimes almost mirrored—parallels, but with institutional materializations and non-congruent timings. The presentation and discussion of these research studies and their results aim at incorporating some tools to the initial formation of the Language and Literature professional for him or her to perceive relations between linguistic and historical aspects using the recognition of convergent and divergent paraphrastic processes. As Celada states (2010: 117–118), between Portuguese and Spanish there are regularities that act “remembering, recalling, pointing out, evoking, insinuating or simply referring to ways of telling about the dynamics of the other”. The search for these oscillating regularities in a shared discursive memory in specific fields, such as politics, arts, education, sports and others, helps to achieve the sociocultural implementation expected in the Language and Literature formation and in the work with linguistic diversity. It also destabilizes and questions the homogenizing representations about each language and promotes the perception of what is local in its dialectics with what is non-local. Linguistic Dynamics Framed in the two dimensions we have already detailed, glotopolitical and discursive, the comparative studies do not set aside, in the planning that we propose, the linguistic dynamics regarding its systematic aspects. The different annual courses which we discuss have also always included comparative topics in the linguistic description area. There is a good quantity and diversity of syntactic problems and semantic categories which have been comparatively researched in a relatively recent time, and which can be issues for a course.4 We will mention some among many, knowing that we may be unfair to a large production. We have found comparative studies about state change constructions (Correa 2009), past verbal forms (Fonseca 2007), temporal and final structures (Oliveira 2013), demonstratives (Ramalho 2016; Stradioto 2012; Moreira 2014), passive and impersonal forms (Araújo Jr. 2016),

Comparisons Between Spanish and Portuguese

157

relative structures (González and Castaldo 2014), null objects (Simões 2010) and anaphoric objects (Yokota 2007). The way we approach in the classes the presentation of these studies and results attempts to prevent the comparison from bringing as a correlate the consolidation of this imaginary identity of “one language” as a closed system, always definable and autonomous. For that, one of the first preventions is the attention to the heterogeneity in each of the linguistic spaces at stake, also in relation to the item that we are comparing. And, above all, when dealing with proximate languages, many times what is presented as “different” between them is not so different in a specific variety that was not taken into account in either of the two languages. Another important aspect in this direction is the localization of the linguistic phenomena in change processes which are related to specific discursive traditions and to standardizing pressures, building, thus, a bridge to the other two dimensions that we discuss in the courses.

The Classroom: Horizons and Risks A great part of the research references aforementioned are study material in the courses we talked about. But the biggest challenge may be planning the activities that follow up and help deepen the reflections when handling textual series with a pedagogical, rather than investigative, purpose. This is what we will discuss in this section. The observation of translations and versions of texts from several genres and origins, including film subtitling or the transposition of musical compositions with lyrics, has been a fruitful space to tackle issues, mainly in the field of linguistic dynamics and of the representation of its heterogeneity. However, we believe that the richest aspect regarding the general purposes of what we discuss here is the classroom work with texts in Brazilian Portuguese and Spanish which do not come from the translation or transposition of each other, but rather from production conditions that, at some points, enable the comparison. For instance, media news reports about a fact with international repercussion in Brazil and in another Hispanic-speaking country, or text genres with a strongly stabilized rhetoric, such as the oral joke about certain issues or, more recently, the meme. Or, like in the practice described below, where we developed in the first class of the course a textual series arranged by several comparable factors. We took to the class a series of readers’ comments from popular daily news sites from Brazil and from Argentina related to reports about income distribution policies for the needy social sectors. It was the same discourse genre (readers’ comments) with the same support (online daily news sites) and the same conditions for the interlocutors. There were

158

Adrián Pablo Fanjul

also, among the comments, coinciding argumentative orientations, generally contrary to affirmative action. We comparatively find some coinciding premises, such as a “lack of awareness” among poor people about paternal responsibilities. Despite the coincidences, these premises were demonstrated in different narratives in each country. We have also found a similar conceptualization in the construction of the “social help” discourse object, conceived in both cases as a cost for those who “work” to favour those who “do not work”, but with the introduction of intermediary political entities in the narrative produced by the Argentinians, a difference that seems to be a sign of the “sways” of proximity and distance mentioned in the previous section. In both countries and within the “same” language, some comments supported its prejudiced argumentation in terms of linguistic heterogeneity, such as imitating the supposed speech of the social help beneficiaries. However, while in the Brazilian texts this mocking was directed at a social stratification stereotype by the language, in the Argentine comments it was more related with a supposed foreign aspect originated in neighbouring Hispanic-speaking countries. In the semantic determination field, which directly sets the relations between the discursive with the linguistic dynamics, the different modes to produce proverbial enunciations—argumentative resource which tends to show an assertion as something “known” by everybody—seemed to be productive for the comparison. In Spanish, they were mainly represented by impersonal pronominal constructions and a defined determination of the noun phrase. Otherwise, in Brazilian Portuguese, there were many formulations such as “criança virou moeda de troca e dá para imaginar bem o resultado disso”,5 supported in the generic value that the absence of determiners produces in this language and in a non-pronominal impersonality form (“dá para” + infinitive).6 The articulation of these several observation levels creates an opportunity to visualize the type of proximity which we postulate between languages when studied from a sociohistorical perspective. It also suggests lines for research and/or pedagogical planning for the teacher in training, regarding those levels of linguistic and discursive dynamics which are interesting for them. The direction of their questionings will depend on an individual developmental history, this example being just a part of it, but this will encourage him or her not to lose sight of possible relations between their interests and other aspects of verbal interactions in specific social formations. Regarding an evaluation of the developed practices, we believe that the main formative risk that should be faced is one that every comparison may hold because it is intrinsic to that: the imaginary closing of the universes confronted by the comparison, the reinforcement of a homogeneous view which we have already mentioned before. With respect to that, we coincide with Serrani (2005: 25) in the necessity, for every curriculum planning in

Comparisons Between Spanish and Portuguese

159

Languages education, whatever level or scope in which it develops, to operate with the notion of “contradiction” when studying the sociohistorical diversity in language. The perceptions of “one in the other”, of myself in what is not me, of what is never equally distant, and what may go further in one aspect while it goes closer in another one may be the best antidote against a stereotyped view about the comparison terms. It is because of this phenomenon that the most productive type of comparative practice in the classroom is the one that, like the example given in this section, integrates the different levels of dynamics.

New Pathways and Projections Before finishing, we consider it important to say that, besides the sociopolitical and epistemological factors that we mentioned in the second section, another very significant factor for the conception of this type of comparative perspective is our insertion in one area such as the Spanish Language and its Literatures at the University of S. Paulo, which, in the field of literary studies, has also been devoting special attention to the relations between the cultural spaces of both languages and to their contact and hybridization processes. The political-linguistic and discursive approach for the relations between Portuguese and Spanish presented above has already created projections in the university services, in the pedagogical planning on a national level, and in the production of educational materials for other fields, mainly elementary education. The various continuing education programmes for teachers which our area performs take this perspective in these segments. Several documents that guide our school education in foreign languages started to take into account many comparative aspects as a way to foster an education for citizenship that pays attention to alterity and for which the study of a proximate language offers specially instigating possibilities of recognition and unfamiliarity. Finally, this has been an attempt to argue for an approach to languages in higher education which, using comparative practices in research and in teaching, encourages the questioning of given universalities. We aimed at setting, in their heterogeneity and in their conflict relations, the presence of two undoubtedly hegemonic languages in the South American space, by establishing a relation between their historical position in colonization and resistance processes, the discourses that produce some materiality in themselves, and all that approximates and distances them in their dynamics.

Notes 1. Relating to Portugal. (Translator’s note) 2. Essa dimensão histórico-social a ser considerada na proximidade pode ser a mesma para o mesmo par de línguas em regiões do mundo qualitativamente

160

Adrián Pablo Fanjul

diferentes precisamente no sociohistórico? Podemos pensar a proximidade entre o português e o espanhol na Europa e na América Latina como a mesma proximidade? Seria isso viável, quando se trata de regiões do mundo que se diferenciam entre si pelo desenvolvimento de formações sociais reconhecidamente distintas, e em cada uma das quais as próprias línguas em questão tiveram e têm implantação e situações de contato muito diferentes, por causa, precisamente, das particularidades desse desenvolvimento? Parece-nos que a proximidade / distância deve ser especificada: a de determinadas línguas em determinado espaço, em determinada formação sociohistórica. (Fanjul 2017: 31–32) (Translator’s note) 3. Also based on Pêcheux, we understand “memory spaces” in a text as the remittances that it triggers to other texts or discursive series; a form of what, in studies of this author and of other ones in the same approach, is called “interdiscourse”. 4. There are also, of course, very little explored areas, mainly because a great part of what is researched is still questioned based on the specialization areas of the first ones who developed this interest. For example, comparative studies about prosody and, in general, about the phonetic dimension are still missing. Lexicology is also a little explored area. 5. Children have become currency and one can very well imagine the result of that. (Translator’s note) 6. “one can imagine”—impersonal pronoun.

References Araújo Jr., B. (2016). Impersonal SE and passive SE constructions in Spanish and Brazilian Portuguese. Signos ELE 9, 1–20. Arnoux, E. (2008). Los discursos sobre la nación y el lenguaje en la formación del Estado (Chile, 1842–1862). Estudio glotopolítico. Buenos Aires: Santiago Arcos. Auroux, S. (1994). La révolution technologique de la grammatisation. Liège: Mardaga. Celada, M. T. (2010). Entremedio español/portugués—errar, deseo, devenir. Caracol 1, 110–150. Correa, P. (2009). Interlanguage atributtive structures and the syntactic organization of Brazilian Portuguese and Spanish. Signo y Seña 20, 115–132. Fanjul, A. (2017). A pessoa no discurso. Português e espanhol: novo olhar sobre a proximidade. São Paulo: Parábola. Fanjul, A. and González, N. (2014). Políticas do saber e (re)descoberta das línguas. In Fanjul, A. and González, N. (eds.) Espanhol e português brasileiro. Estudos comparados (pp. 7–25). São Paulo: Parábola. Faraco, C. (2011). O Brasil entre a norma culta e a norma curta. In Lagares, X. and Bagno, M. (eds.) Políticas da norma e conflitos linguísticos (pp. 259– 275). São Paulo: Parábola. Fausto, B. and Devoto, F. (2004). Brasil e Argentina. Um ensaio de história comparada (1850–2002). São Paulo: Editora 34. Fonseca, M. C. (2007). Semantics and pragmatics in the English present perfect x past simple and the Spanish Pretérito Perfecto x Pretérito Indefinido comprehension. Doctoral thesis in Languages. São Paulo: Universidade de São Paulo. Gallardo, I. (2013). The presence of politeness found in corporate Emails: Brazilian Portuguese and Peninsular Spanish: Similarities and contrasts. Master’s degree dissertation in Languages. São Paulo: Universidade de São Paulo.

Comparisons Between Spanish and Portuguese

161

González, N. (1994). Cadê o pronome? O gato comeu. Os pronomes pessoais na aquisição/aprendizagem do espanhol por brasileiros adultos. Doctoral thesis in Linguistics. São Paulo: Universidade de São Paulo. González, N. and Castaldo, I. (2014). As construções relativas: parte das inversas assimetrias? In Fanjul, A. and González, N. (orgs.) Espanhol e português brasileiro. Estudos comparados (pp. 73–91). São Paulo: Parábola. Guilherme, M. (2018). ‘Glocal languages’: The ‘globalness’ and the ‘localness’ of world languages. In Coffey, S. and Wingate, U. (eds.) New directions for research in foreign languages education (pp. 79–96). New York: Routledge. Henry, P. (1977). La mauvais outil: langue, sujet et discours. Paris: Klincksieck. Hobsbawn, E. (1990). Nations and nationalism since 1780: Programme, myth, reality. New York: Cambridge University Press. Lagares, X. (2016). Dinámicas normativas del español y del portugués. In Arnoux, E. and Lauría, D. (eds.) Lenguas y discursos en la construcción de la ciudadanía suldamericana (pp. 283–298). Gonnet: UNIPE Editorial Universitaria. Lagorio, C. (2011). Norma e bilingüismo no espanhol americano. In Lagares, X. and Bagno, M. (eds.) Políticas da norma e conflitos linguísticos (pp. 193– 214). São Paulo: Parábola. Lima, F. (2012). Looking god X Being fair: The apology in image management in the midia interactions. Master´s degree dissertation in Languages. São Paulo: Universidade de São Paulo. Lipsky, J. (1994). Latin American Spanish. New York: Longman Group Limited. Mariani, B. (2003). O Estado e a Igreja na questão da língua falada no Brasil. In Arnoux, E. and Luis, C. (eds.) El pensamiento ilustrado y el lenguaje (pp. 17–37). Buenos Aires: Eudeba. Mattos e Silva, R. (2004). Para uma socio-história do português brasileiro. São Paulo: Parábola. Menezes, A. (2012). Among homelands, tambourines and bandoneons: The clash between disciplinary and marginal voices in compositions of samba and tango (1917–1945). Doctoral thesis in Languages. São Paulo: Universidade de São Paulo. Menón, L. (2016). The real life in the make-believe world: Na enunciativediscursive analysis of the magazines Gente and Contigo. Doctoral thesis in Languages. São Paulo: Universidade de São Paulo. Moreira, G. (2014). As séries de demonstrativos. Mais assimetrias. In Fanjul, A. and González, N. (eds.) Espanhol e português brasileiro. Estudos comparados (pp. 95–111). São Paulo: Parábola. Nascentes, A. (1920). Grammática da língua espanhola para uso dos brasileiros. Rio de Janeiro: Livraria Drummond Editora. Oliveira, B. (2013). Subordinadas temporais e finais em português e espanhol: questões de contraste e efeitos para a tradução. In Fanjul, A., Martín, I. and Santos, M. (eds.) VII Atas do VII Congresso Brasileiro de Hispanistas (pp. 190–197). São Paulo: Associação Brasileira de Hispanistas. Orlandi, E. (2002). Língua e conhecimento lingüístico. São Paulo: Ed. Cortez. Pêcheux, M. (1988). Discourse: Structure or event? Chicago: Ilinois University Press. Pizzutiello, A. (2017). Propagandistic documentary films of the military dictatorships of Brasil (1964–1985) and Argentina (1976–1983): A discourse and enunciation analysis. Master´s degree dissertation in Languages. São Paulo: Universidade de São Paulo. Ramalho, V. (2016). Sistema de Demonstrativos no Português Brasileiro e no Espanhol Mexicano sob a Perspectiva das Tradições Discursivas: Gêneros

162

Adrián Pablo Fanjul

Notícia e Romance. Doctoral thesis in Linguistics. Belo Horizonte: Universidade Federal de Minas Gerais. Russo, M. (2013). The voice of soccer fans (torcedor/hincha) in the description of soccer goals in matches between Argentinian and Brazilian teams. Master’s degree dissertation in Linguistics. São Paulo: Universidade de São Paulo. Serrani, S. (1994). Analysis of discursive resonances for the study of culturallinguistic identity. Trabalhos em Linguística Aplicada 24, 79–90. Serrani, S. (2005). Discurso e cultura na aula de língua. Currículo, leitura, escrita. Campinas: Pontes. Simões, A. (2010). Clitic, null objet or lexical pronoun? How much and however the variation change in the 3rd person accusative object pronominal realization in Brazilian Portuguese reflects on Spanish adquisition/learning for Brazilian learners along generations. Master’s degree dissertation in Linguistics. São Paulo: Universidade de São Paulo. Souza, L. M. (2010). Culture, language and dialogic emergence. Letras & Letras 26–2, 289–306. Stradioto, S. (2012). Deixis na Romania nova: o lugar dos demonstrativos no português de Belo Horizonte e no espanhol do México. Master’s degree dissertation in Linguistics. Belo Horizonte: Universidade Federal de Minas Gerais. Tarallo, F. (1983). Relativization strategies in Brasilian Portuguese. Doctoral thesis in Linguistics. Filadelfia: University of Pennsylvania. Yokota, R. (2007). What I say can not be written, but does anyone say what I write? The variation in use of the anaforic direct object for oral and written productions by Brasilian learners of Spanish. Doctoral thesis in Linguistics. São Paulo: Universidade de São Paulo. Zoppi Fontana, M. and Celada, M. T. (2009). Sujeitos deslocados, línguas em movimento: identificação e resistência em processos de integração regional. Signo y Seña 20, 159–181.

8

Multiculturalism and Glocal Languages The Impact of Cultural Mobility in Spanish Teaching and Learning in Southern Brazil Maria Josele Bucco Coelho

Introduction Cultural communities are currently experiencing intense processes of displacement that are becoming established not only in the territorial, physical sense, but also in movements of ontological and symbolic origins. These mobilities break from the traditional ethnic, linguistic and national references, which are usually responsible for the sense of belonging to a community. They create imaginary communities and raise awareness of transcultural, transnational and even post-national identities. Based on this premise, this study seeks to understand how Spanish, considered a glocal language, that is, one of the languages that “are surviving in-between the struggles between and across the global and the local” (Guilherme, 2017), incorporates such assumptions and deals with the knowledge that multiplies its potential in terms of multi, inter and transcultural relationships. It is possible to assume that the glocal tension started in the sixteenth century1 with the colonization of the Americas. It was maintained over time by the innumerable types of cultural mobilities experienced in this territoriality (migrations, exiles, diasporas) and intensified in the last decades as a result of the process of globalization. The perception that the glocalization of Spanish is based on cultural mobilities and that the linguistic, phonetic, syntactic and pragmatic characteristics resulting from these relationships correspond to a kind of “localized globalism” (Santos, 2002), that is, a local response to the global calls and frictions, guide this study. To Guilherme, New directions for language learning in the 21st century have to be “contextualized” in the global and local contexts (both transnationally and intra-nationally) in which they circulate and in which they are learnt, but from different perspectives (in the plural) from that (in the singular) in which they have been learnt, not only by taking into account language varieties in a de-colonial approach but also in terms of intra-national diversity. (Guilherme, 2017)

164

Maria Josele Bucco Coelho

For the author, the global-local relationship requires a new perception of languages—in relation to both use and teaching—that may inevitably result in an approach that responds to globalization as well as to how it affects the localisms and to how it is also affected by them. In this sense, this approach should presuppose what the author calls “a critical intercultural awareness”, focused on cultural exchanges. Regarding Spanish teaching in Brazil, it is important to highlight the strong peninsular influence, established by the turn of the twenty-first century. Such an effect is embodied in the predominant adoption of a single variant—the so-called “standard Spanish”—and in the stark silencing of all linguistic and cultural plurality arising from the Americas. This movement toward homogenization has been reinforced by the repertoire of teaching materials that were available in Brazil and which have worked, over time, as one of the most effective mechanisms for maintaining this hegemony. For Quijano, Latin America, “by its historically-structurally dependent constitution within the current pattern of power, has been [from its beginnings] and [during] all this time, constrained to be the privileged space of the exercise of the coloniality of power” (Quijano, 2000, p. 10). For this reason, cultivating critical intercultural awareness—as proposed by Guilherme—requires the formulation of decolonial strategies that result in actions that are not only presented as echoes of the global calls, but which reveal dynamics that meet the experiences lived in the local communities. In the meantime, we first seek to understand how Spanish teaching can incorporate cultural mobilities and multiculturalism, which are two paradigms of action that affect the localized globalism. We further seek to understand how this teaching can implement strategies that better represent the plurality of cultural communities of Hispanic America. Next, we will analyze the curriculum proposal and teaching practices developed for the undergraduate degree in Spanish literature from the Federal University of Paraná in order to understand and scrutinize how this pedagogical training includes these multicultural networks that reinvent a being-in-the-world that exceeds the notion of language, culture and nationhood.

Inter, Trans and Multiculturality: Addressing the Heterogeneity of the Americas According to Guilherme and Dietz (2015), the indiscriminate use of the terms interculturality, transculturality and multiculturalism occurs in a context of expanding mobilities that engender diasporic and transnational communities. Such concepts, however, “differ according to whichever perspective we take—the geographical, historical, cultural, political, ideological, sociological” (Guilherme & Dietz, 2015, p. 2).

Multiculturalism and Glocal Languages

165

With regard to the cultural communities of the Americas, it is necessary to point out that, long before the globalizing momentum, which propels the use of such concepts, the discourse of hybridity and miscegenation had been around for some time. This is because the colonization process, the formation of nation-states and the beginning of modernity materialized as an extension of the issue of identity and by the tension that the cultural plurality generated. In this sense, much has been thought about the Brazilian hybrid character and several narratives have been disseminated over time with the emphasis on cultural clashes, such as Freyre’s Casa-Grande & Senzala (1933), and Raízes do Brasil (1936), by Buarque de Holanda. Invariably, the perception of hybrid identity and the search for the appeasement of alleged conflicts prevail. The transcultural element,2 that is, the idea of synthesis permeates these narratives, easing the wounds left by coloniality. For Walsh, from 1990 onwards, it became common to employ the term “interculturality” to refer to both the emergence of demands for recognition, rights and social transformation that come from the originating/ ancestor’s communities vis-à-vis a decolonizing process, as well as its inclusion as a tool in the global market. They are, therefore, distinct conceptions—the former has a critical character and the latter, a functional character: La interculturalidad functional roots en el reconocimiento de la diversidad y differential culturales, con goals in the la inclusion de la mishma al interior de la estructura social ruled. [ . . . ] Critical interculturality is not part of the problem of difference itself, but of the structural-colonial-racial problem. That is to say, from a recognition that difference is built within a colonial structure and matrix of racialized and hierarchical power. (WALSH, pp. 77–78) Thus, critical interculturality is established at a deeper level of understanding as it seeks to elucidate how difference is created and disseminated. As stated by Walsh, it is a decolonial political practice, which aims at the construction of a “very distinct path, which is not limited to the political, social and cultural spheres, but is instead also crossed by knowledge, being and life itself” (Walsh, p. 89). Multiculturalism,3 as Kymlicka points out, is as old as humanity itself, since “different cultures have always found ways of coexisting, and the respect for diversity was characteristic of various empires, as the Ottoman Empire, for example” (our translation). However, multiculturalism currently refers to a more specific historical phenomenon, first emerged in Western democracies in the 1960s (Kymlicka, 2012, p. 131), as part of the post-war transformation movements and part of the struggle for

166

Maria Josele Bucco Coelho

human rights, which involved ethnic and racial diversity. As Kymlicka explains, While multiculturalism for immigrant groups clearly differs in substance from that for indigenous peoples or national minorities, each policy has been defended as a means to overcome the legacies of earlier hierarchies and to help build fairer and more inclusive democratic societies. Therefore, multiculturalism is first and foremost about developing new models of democratic citizenship, grounded in human-rights ideals, to replace earlier uncivil and undemocratic relations of hierarchy and exclusion. [ . . . ] Multiculturalism is precisely about constructing new civic and political relations to overcome the deeply entrenched inequalities that have persisted after the abolition of formal discrimination. (2012, p. 8) It is thus understood that the multicultural perspective is not only the celebration of difference. It is an approach that has as a purpose the formulation of civil, political and theoretical-practical strategies and resources that help overcome the entrenched inequalities—originating from coloniality, in our case. The existence of a majority group and other minority groups requires, according to the multicultural perspective, policies that guarantee the rights and protection of this diversity Obviously, as one may suppose, this stance has received much criticism. After all, multiculturalism does not presuppose a complete dissolution of the relations of domination. Rather, it is a possible and a viable alternative to the guarantee of heterogeneity rights while society fails to create other more effective mechanisms for assuring equality. Unlike universalism and cosmopolitanism, which are invariably impositions of a group, the multicultural perspective recognizes the difference and the need for the respect of the rights of the Other. Perhaps this is the reason for all the appeal that this perspective holds—because it represents a possibility of a political and theoretical action that is conciliatory and can be deployed at the present time. The multicultural perspective is grounded in a practical sense, as it allows the empathy and the sensitivity towards the Other to be transformed into real effective mechanisms that guarantee rights, representation and effective participation in communities. With these transient practice and purposeful traits, multiculturalism promotes awareness about the realities of oppressed groups, contributes to the understanding of how subalternity is formed and how power relations are established between groups, and encourages reflection on difference. Taken not only as a tool to describe the composition of contemporary cultural communities, multiculturalism has, therefore, as its main

Multiculturalism and Glocal Languages

167

characteristic the potential to become a feasible action, as an intervention and transformation of social dynamics. As explained by Candau and Moreira, it is “a cultural and political project, a means of dealing with cultural relations in a society and conceiving public policies under the perspective of the radicalization of democracy, as well as a designer of pedagogical strategies that correspond to this perspective” (our translation) (2013, p. 20). Regarding the relationship between multiculturalism and educational practices, according to Candau and Moreira it is important to note that the inclusion of this perspective in the initial training courses for educators is still very recent, and that this inclusion, sporadic and unsystematic, occurs “to the taste of personal initiatives of some teachers” (our translation) (2013, p. 19). This is because multiculturalism, which has had its locus of production in the political sphere at first, has been slowly incorporated into the university research field. However, the multicultural perspective assumes a preponderant role, since “there is no education that is not immersed in the cultural processes in which it is located” (our translation) (Candau & Moreira, 2013, p. 13). According to a study by Moreira (2001), research on initial teacher training and multiculturalism advocates a reflective teacher who is multiculturally competent. The studies also suggest alternatives to achieve this desired development, such as the inclusion of courses specifically focused on multicultural education, new content, procedures and values within the existing pedagogical courses, and contact of the teachers-in-training with different cultural realities. (2001, p. 70) Thus, based on the assumption that contemporary societies are affected by cultural mobilities that engender hybrid identity scenarios where the inter, trans and multicultural encounters are favored, the educational practices need, necessarily, to meet this demand with a deep reflection on the homogenizing and monocultural character which, in most cases, dominates the teaching-learning process.

Cultural Mobility and Multiculturalism in Teaching Practices The cultural mobilities that have comprised the Americas over time justify the importance the role multiculturalism takes on in order to understand Latin American reality. Forged following the clash between the different cultures of the continent and the European settlers during the period of conquest and colonization, this reality has been gradually

168

Maria Josele Bucco Coelho

incorporating migrations/mobilities derived from their social-political processes (such as wars, conflicts, dictatorships, exiles, settlement policies, etc.). According to Clifford, wherever we look, the processes of movement and human encounters are complex and ages-old. Cultural centers, delimited regions and territories do not precede contacts but are strengthened by their intermediation and, in this process, take ownership of the tireless movements of people and things, and discipline them. (1997, p. 14) The relationships established in contemporary times therefore originate from the tension of local and transnational forces, in a scenario where identity constitution is continuously configured/deconfigured/ reconfigured in contact zones, in border areas. This intensity of linguistic, territorial and identity-related encounters engenders processes in which one does not solely find “losses, deletions, or appropriations, but also the creation of new cultural products” (Bernd, 2003, p. 18). In the interim, “purity and permanence affirm themselves—creatively and violently— against historical forces of movement and contamination” (Clifford, 1997, p. 18). Therefore, the disturbance generated by the mobilities cannot be ignored when observing the ways societies organized/organize their assumptions regarding cultural identities. For Clifford, there are three interconnected global forces that intensified these migration processes: the continuous legacies of empires, the unprecedented effects of the world wars and the global consequences of the destructive and restructuring actions of industrial capitalism (1997, p. 18). Taken as the “ability of the subject to move from one domain to another” (our translation) (Bernd, 2010, p. 14), the mobilities assume different shades that compel them to break from the traditional ethnic, linguistic and national references, which are generally responsible for the sense of belonging to an “imagined” community. The mobilities also arouse questions about how the displaced subjects shape new networks and reinvent the sentiment of being-in-the-world that surpasses the notion of culture and nationhood. In Dicionário de Mobilidades Culturais, Bernd (2010, p. 18–23) uses a typology that seeks to encompass the various nuances that such displacements assume in contemporary times. The author nevertheless reminds us that the way these nuances appear does not need to be absolutely precise, since they are often inter-linked. However, she highlights five distinct categories: 1. Transcultural migratory mobility: refers to the migrations that are forced, experienced and undertaken by cultural communities during immigration and emigration processes, and implies cultural transfer.

Multiculturalism and Glocal Languages

2.

3.

4.

5.

169

This type of mobility may also occur within the cultural community when individuals are compelled to choose marginality. In addition, this category covers the movements of drift, displacement, deterritorialization, wandering, nomadism, course and transnation. Mobilities, memorials and the inter-subjective: designate the movements that conjure up, according to Bernd (2010), “the equations of one’s own memory where the oblivion is an integral aspect of the mnemonic process” (2010, p. 18). In this space of interstices and subjectivities, which gathers autofiction, memory and imagination, cultural traces are generated. They determine, on the one hand, work, duty and memory overuse, and on the other hand, forgetfulness, the unsaid, silence, mechanisms of memory activation and reinvented imageries. Transactional mobility: refers to the movements that carry the idea of “overtaking, going beyond, of passing through” and that involve transpositions of meaning that “operate on a symbolic level, [ . . . ] the result of numerous trickeries and negotiations with the language” (our translation) (Bernd, 2010, p. 21). Transactional mobilities (metaphor, linguistic mobility, translation, transportation and variation) mark the experience of the individuals. As they are in constant transit, they experience the lack and difficulty of finding an identity and seek new forms of expression that can portray this condition. Spatial mobility (of the metropolises imageries): designates the urban circulations and the flâneur experience. These mobilities refer specifically to spatial mobility and depict the experience of moving within urban chaos, of being part of the crowd and of getting constantly lost in the flow of big cities. Deviant mobility: designates the “mechanisms of norm transgression and the resistance that is motivated more by cunning than by force” (our translation) (Bernd, 2010, p. 22). These mobilities (braconagem, deviance and fluidity) take place amid prohibited and interdicted territories, and refer to the cunning mechanisms used as stratagems of transgression in the cultural communities.

This typology proposed by Bernd helps us understand how the movements—inherent to the experiences of the cultural communities— are responsible for the establishment of subjectivities that are constituted in a multi-territorial, geographical and symbolic way, revealing processes of hybridization, mestizaje and transculturation. They indicate, as pointed out by Hall, that societies are “formed and transformed continuously in relation to the ways we are represented or addressed in cultural systems” (2009b, p. 597). Being in the genesis of the discussion about the identity in/of the Americas, such processes of inter, multi and transculturation are responsible for the appearance of individuals heterogeneous, hybrid and marked

170

Maria Josele Bucco Coelho

by cultural difference, forged in the proliferation of all manner of passages, movements and displacement. In the meantime, some underlying notions around knowledge are necessarily deconstructed about/of Latin America that were based, in large part, on a Eurocentric and homogenizing view. The way societies are dealing with the consequences of the mobilities and dislocations in the contemporary world is very diverse. Imbert asserts that “there are many theories”, but that these do not necessarily imply “a process of cultural homogenization because the basic globalization process de favors movement and convergence” (Imbert, 2013, no page). It is in the recognizing that these meetings are responsible for the establishment, maintenance and transformation of differences that one may perceive the importance of the multicultural perspective, because, as Imbert reminds us, One must undo the perspective of the European colonial, which the Americas will achieve through reflective theories and policies that introduce multicultural and transcultural concepts of social, economic, and cultural relations. These concepts lead to the understanding that the children of earth are the products of interethnic and intercultural meetings that have arisen more out of coincidence than from the effects of long-established temporalities or transferred from Europe, and which govern the choices of life. (2011, p. 25) Multiculturalism and transculturality are, therefore, possible approaches to embrace cultural plurality and to avoid the erasure of communities that are subaltern. Although they find resonance in contemporaneity as social practices, they are perspectives that are confused with being-in-the-world.

Spanish as a Glocal Language: Challenges to Pre-Service Teacher Training With regard to the teaching and learning of Spanish in Brazil—and, consequently, to the process of initial training of educators—it should be noted that, since 1990, there has been some concern in terms of the cultural aspects.4 However, these remained subordinated to communicative competence; that is, the inter, trans and multicultural issues debuted in the formative teaching processes in a position submissive to the communicative presuppositions. According to Paraquett (2005), this mistaken perception that the knowledge of the culture of the Other served only to guarantee immediate communication is evinced by the body of textbooks available in the Brazilian market that took the cultural context, “as a mere excuse for

Multiculturalism and Glocal Languages

171

the acquisition of expressions that expedite the repetition of a specific cultural model, taken as hegemonic, without considering the ultra-vast panorama of Hispanic culture” (2005, p. 301). The rupture, in this case, “coincides with the (delayed) Brazilian understanding that Spanish is a language that is spoken by many Latin American peoples” (Paraquett, 2005, p. 301). The teaching of Spanish in Brazil was, as seen in Paraquett’s description, anchored in a Eurocentric perspective. This perception is extremely important because “multiculturalism without the critique of Eurocentrism risks becoming a shopping mall of world cultures, while criticism of Eurocentrism without multiculturalism runs the risk of simply reversing existing hierarchies rather than to rethink them in a profound way” (Shohat & Stam, 2006, p. 474). Having emerged as a sort of justification for colonialism, Eurocentrism is characterized as a way of thinking that permeates contemporary practices and representations—even after its official termination—and rules the hierarchical relations and power “obscured by a type of epistemology hidden” that is so inexorable in the center of our everyday lives, we barely perceive its presence. The residual traces of centuries of axiomatic European domination shape the common culture, the language of everyday life and the media, engendering a fictitious feeling of the superiority of European cultures and peoples. Eurocentrism bifurcates the world in “the West and the rest” and organizes the language of everyday life into binary hierarchies that implicitly favor Europe: our nations, their tribes; our religions, their superstitions; our culture, their folklore; our art, their crafts; our manifestations, their tumults; our defense, their terrorism. (Shohat & Stam, 2006, pp. 20–21) For Boaventura de Souza Santos, this epistemic primacy can only be touched and decolonized from a work on objects, “either in the sense of transforming them into recognizable objects of knowledge within the framework of what already exists, or in the sense of its redefinition as part of a more general redefinition of the spaces of knowledge” (Santos, 2010, p. 138). It is, therefore, considering the existence of a Eurocentric discourse implicit in the erasure of the heterogeneity of Latin American cultural communities in the practices of initial teacher training of Spanish teachers that one can think of and propose devices of inclusion and representation of the plurality and diversity that constitute them. Likewise, pre-service teacher training of Spanish teachers within Brazilian universities needs to be attentive to de Souza’s exhortation on the characteristics of colonization in the Americas that “intensified and complicated intercultural contacts to the point of making nations

172

Maria Josele Bucco Coelho

like Brazil simultaneously post colonial -and neo-colonial in relation to the pre-colonial indigenous cultures that remain alive in their territory” (Souza, 2010, p. 290 my translation). The questions that arise from this complex entanglement refer to the type of strategies and devices that are used to respond to the demand generated by Eurocentric criticism. In addition, bolstered by the multicultural perspective, questions emerge about the strategies undergraduate/ teacher training courses develop to assure the representativeness of Latin American cultural communities and how the “decolonial turn” is established in the training practices developed. These concerns guide the didactic-formative practices within the UFPR Spanish language undergraduate degree, in the subjects of Basic Spanish and Spanish Language 1—both taught in the first year of the Spanish Language major (“habilitação”), and cement both the organization of content and the production of teaching materials and other didactic practices. From these formative experiences, what follows is the description of how these courses operate and present some of the strategies and practices that materialize the glocality of Spanish that can be used to negotiate different systems of power, regulation and emancipation in the process of teaching and teacher education at different levels of identity and citizenship. Course description: taught in two weekly meetings of 3 contact hours, making up a total of 90 course hours. They are sequential, that is, students must take the Basic Spanish course first and, in the following semester, Spanish Language 1. The classes are taught in the language labs of UFPR and have multimedia equipment, with a maximum of 18 students per class. For the most part, the Brazilian students had not had contact with the Spanish language and started the process of language acquisition from a perspective that considers, at the same time, initial teacher training. Each of the courses has a syllabus that, in UFPR, is designated as “Ficha 2”, in which all relevant information is described (the nature of the class, hours, objectives, contents, methodology, assessment and bibliographic references, both required and complementary). The organization and operationalization of this curricular programming is quite open, since the teacher has complete autonomy to structure the programming of content, objectives and didactic and evaluative procedures, thus making for a bureaucratic-institutional device that can fully absorb the demands of Eurocentric criticism and at the same time incorporate pedagogical and formative practices based on multicultural ideals. This autonomy, in this case, is taken from a positive perspective. However, circumstantially, it may be a complicating factor, since any change of subject instructor also means the reworking of the Ficha 2. To minimize the

Multiculturalism and Glocal Languages

173

impacts on the education process, the Spanish faculty chose to prepare the Ficha 2 jointly. Among the teaching strategies that can be incorporated into initial teacher training that subscribe to Spanish from the perspective of a “glocal language”, we highlight some of those that are being incorporated, systematized and improved within these courses at UFPR: Representing the plurality of Hispanic cultural communities: considering the criticism made by Paraquett (2005) about the diffusion in Brazil of mostly peninsular methods and the homogenizing and monocultural character that prevails in them, the option of not adopting a Spanish teaching method (textbook) becomes a very effective action. This choice seals the intention of providing contact with Hispanic heterogeneity (Cornejo Polar 1996a),5 allowing the contact with discursive, linguistic, artistic and cultural practices coming from diverse territorialities. This occurs to the extent that the teacher, in producing the didactic material, has the autonomy to explore this plurality, without being confined to monocultural representations. Diversity, in this process, is not taken as an appendix or as simply as an “extra” that is opposed to hegemonic language/culture. Linguistic variety and plurality of social and cultural practices that make up the communities are seen and incorporated as the main component. Development of teaching material: by giving priority to the production of didactic sequences and the recommendation of support material centered on the experience of acquiring Spanish from the Brazilia6 experience, a local response to global appeals is brought forth. The refusal to use a method/book which, despite some efforts, remains representative of only one cultural community becomes a very fruitful device and is well received by the students. This is because the organization of the didactic sequences is thematic, obeying the thematic-cultural content proposed in the syllabus of the discipline, but that reaches different hues in each class, since it incorporates the discussions, news and problems that arise at the time of production. Thus, for example, addressing the theme of “family” and its developments in the lexical, linguistic and functional realms, the students can also discuss political issues on the subject. In addition, it seeks to demonstrate that there is a plurality in relation to the way cultural communities organize and respond to this appeal. Textual selection and discursive genres: each cultural community has, at its core, hegemonic devices for the production and dissemination of cultural, linguistic and discursive products. In the case of the teaching of Spanish—which includes numerous

174

Maria Josele Bucco Coelho cultural communities—many times, even when guided by multicultural dynamics, the use of texts that belong only to the spheres of majority groups predominates in the classroom. This occurs because there are certain communities that are not even represented. Thus, the insertion of a text that truly represents already seems to be an instrument of great value. However, it is necessary, to the extent possible, to select texts that belong to minority groups. And the didactic exploitation of texts originating from minority groups is a very plausible inclusive action due to the affordances of the Internet. The selection of texts, from this perspective, functions as a very rich device for the explication of the “ecology of knowledge” (Santos, 2010) that comprises the plurality of these communities. Returning to Spivak (2010), who questioned whether the subaltern could speak, this action seeks to give heed to those who found a space of enunciation, but who remain marginalized. Tension of neocoloniality: the critique of Eurocentrism cannot be separated from a critical perception about the other processes of colonialism that have established themselves internally in the Americas in relation to the native communities and to the blacks who later added themselves to the role of the subordinates. It is interesting to think how the conformation of the hybrid and transcultural identities described by such theorists as Ortiz, Rama and Canclini reveals the erasure of these communities and denounces their exclusion in social spheres and practices. This erasure takes place, also, in the practice of teaching. Therefore, from a multicultural perspective, it is necessary to gradually incorporate these communities into the discussions proposed by the didactic sequences revealing not only the memories of resentment for the violence suffered, but mainly, the contributions of these identities fractured by neocolonialism. This posture dates back to preColumbian America, through syncretism, hybridity and reaching the cross-cultural and phagocytic7 elements that make us who we are (García Canclini 2000). Consideration of cultural mobilities: the representativity of Hispanic cultural communities is traditionally treated in the teaching of Spanish and in the initial teacher training from a nationalist bias. Meanwhile, communities that are formed from contemporary displacements and that do not necessarily rest on physical and geographical boundaries are excluded. Such is the case of migrant communities and border regions. In this way, to stress the networks of belonging that are organized from symbolic and linguistic ties is to give the opportunity for these individuals to feel represented and to be integrated as part of these communities. This fracture generated by displacement is usually a source of identity-related

Multiculturalism and Glocal Languages

175

wounds. Much of the artistic and literary production of these subjects does not find a place of acceptance. Therefore, embracing this production thinking of its uniqueness, that is, as part of an in-between place, a third margin is an integrative political action. Establishment of the cross-cultural promise: cross-cultural awareness (Imbert, 2011)—as discussed earlier—emerges when the ability to see in the Other is exercised, when difference, responsible for identity recognition, stops frightening and becomes a space of empathy and respect. By presenting the plurality of Hispanic communities and reflecting on the colonial relations that make up the imaginary and the knowledge of the Americas, and by stressing geographical borders and assuming neocolonial violence, a portion arises where it is possible to inscribe the cross-cultural promise. For Imbert, this promise that transculturality ends shows itself as an “act of performative language that leads to the creation of relationships that are less conflictive, more careful, more attentive, through the very act of thinking in terms of its dynamics” and “develops a multiple personality which explores the contexts and plays with them” (2011, p. 34). Perceiving oneself as a multiple, hybrid and transcultural subject favors and provides a distinct sense of belonging that can invariably contribute to the acceptance of minority groups belonging to cultural communities with different values. To see oneself as a transcultural subject becomes, therefore, this kind of promise of recognition—and respect—for plurality and diversity.

Conclusion In the context of the formative experiences carried out in the UFPR Spanish language degree program, as argued elsewhere in this paper, the inter, trans and multicultural perspective and the criticism of the Eurocentric vision are being brought to the fore, initially, by the deliberate choice not to adopt specific didactic material, but also being reinforced by didactic practices that favor the insertion and representativeness of minority communities based on the strategies described. These actions are gradually being incorporated into teacher education and contribute to the development of a decolonizing critical consciousness based on the critique of Eurocentrism. It is these formative experiences that gradually create cracks and operate an epistemic turn. In this sense, the proposed reflection on cultural mobility, its characteristics and impacts constitute a local epistemic effort that lets the experiences lived and suffered within the framework of the Latin American cultural communities shine through. Recognizing the processes of hybridity, inter, trans and multiculturality resulting from them opens the way for the need to articulate theoretical and practical strategies that

176

Maria Josele Bucco Coelho

can, at that moment, respond to this demand. It is a gradual process of recognition that allows the proposition and achievement of actions that reinforce the glocal character of Spanish which, as Guilherme points out, “aims to provide a political and (inter) cultural, theoretical and practical, approach to language/culture education, to revive a critical pedagogy” (2017).

Acknowledgments The author would like to thank the Academic Publishing Advisory Center (Centro de Assessoria de Publicação Acadêmica, CAPA—www.capa. ufpr.br) of the Federal University of Paraná for assistance with English language editing.

Notes 1. For the purposes of analysis, the present study considers Spanish from the fifteenth century, a period in which the language was spread due to colonization processes in the Americas. However, it does not disregard that, as a Romance language, Spanish has lived other global and local tensions on the Peninsula. The choice of considering Spanish from its period of spread is due to the fact that, before 1492—the year in which the first grammar of Spanish was written by Antonio de Nebrija and that coincides with the arrival of the Europeans in America—the Catholic Kings had not yet defined Castilian Spanish as the official language of Spain. 2. The transcultural paradigm, based on the idea of synthesis of different cultures that constitute ourselves, was proposed in 1942 by Fernando Ortiz, in the emblematic text Contrapunteo del tobacco y del azúcar. Later, taken up again by the Uruguayan Angel Rama, it was extended to the sphere of narrative. In Brazil, the transcultural nuance was related to the anthropophagic movement, which, in 1922, pointed to the need of “incorporating the other”, making them part of oneself. This brief history of how the term transculturalism was incorporated in South America has the purpose of demonstrating that the synthesis of the cultural clashes is fundamental to the discussions on identity and difference raised in this work. 3. According to Patrick Imbert, multiculturalism in the Americas has its roots in Canada and in the struggle against colonialism and separatism. In 1964, Trudeau, prime minister at the time, made an option for bilingualism, which then became a state policy aimed at receiving and integrating immigrants into Canadian territory. In a second period, starting in 1980, multiculturalism became the front line in the fight against racism and exclusion. Today, it is characterized by the search for accommodation of the new, the subalternized groups of people who are building a place of enunciation and need to be heard and represented. In the Americas, in addition to Canada, Colombia is also noteworthy, for it recognized the heterogeneity of its population, comprised of over 80 ethnic groups and 64 different languages spoken nationwide, and enacted in 1991 a constitution based on recognition of ethnic, cultural and regional diversity. There is a big difference between the reasons why Canada and Colombia went the multicultural route. Migratory forces, welcomed and sought by the Canadian government, aim to accommodate the diversity

Multiculturalism and Glocal Languages

177

that enters the country. In Colombia, on the other hand, multiculturalism is implemented as a policy that recognizes the diversity of the original peoples that had always existed. 4. The cultural approach—advocated by authors such as Almeida Filho (2002), Brown (2000, 2001), Damen (1987), Klein and BellSantos (2006), Kramsch (1998, 2004), Mendes (2004, 2007), Moran (2001), among others—is happening gradually, but still quite slowly. 5. The concept of cultural heterogeneity was proposed by Cornejo Polar in the 1980s. Although it is a transculturation, it presents itself as a theoretical device focused on the perception of the conflicts and the difference that persist in the cultural encounters. Cornejo Polar bases his ideas on the understanding that there is a tense coexistence between diverse cultures whose heterogeneity is established due to the segmented participation in the systems of production. 6. By way of example, the works listed in the required references can be cited: Durâo, Adja Balbino de Amorim Barbieri. Análisis de Errores en la interlengua de braslileños aprendices de español y de españoles aprendices de portugués. Londrina: EDUEL, 2004; Fanjul, Adrian. Gramática de español paso a paso. São Paulo: Moderna, 2005; Milani, Esther. Gramática de Espanhol para brasileiros. São Paulo: Saraiva, 2006. 7. The term transculturation designates the preponderance of subjectivities built in a multi-territorial way—geographically and symbolically. It was proposed by Ortiz in 1940 to designate cultural transits. It was then used by Angel Rama in 1970 from a narrative perspective to elucidate the characteristics assumed by the Latin American literary expression fused to the project of modernity. In Brazil, Oswald de Andrade proposed the term “cannibalism” to designate these new additions. More recently, we can cite Canclini who employs the term “hybridity”—focusing more on the issue of mestizaje and Walter Mignolo who suggests, based on these meetings, the concept of “border thinking” (Mignolo, 2012). The phagocitation is a concept proposed by Rodolfo Kusch, Argentine theoretician, in 1963. Contrary to the way cultural elements are normally incorporated—the European element usually predominates—Kusch argues that there are many elements of the original peoples that were incorporated by the European (Kusch, 2000). This is an upside down acculturation. In Kusch’s Americanist logic, Eurocentrism is reversed insofar as it is the native elements that are strongly imposed on the constitution of relational and transcultural identity.

References Almeida Filho & José Carlos P. Língua além de cultura ou além de cultura, língua? Aspectos do ensino da interculturalidade. In CUNHA, Maria Jandyra C. & SANTOS, Percília. (Orgs.). Tópicos em Português Língua Estrangeira. Brasília: Editora Universidade de Brasília, 2002, pp. 209–215. Bernd, Zillah (ed.). Dicionário de mobilidades culturais: percursos americanos. Porto Alegre: Literalis, 2010. Bernd, Zilá. Americanidade e transferências culturais. Porto Alegre: Editora Movimento, 2003. Brown, H. Douglas. Teaching by Principles: An Interactive Approach to Language Pedagogy. New York: Longman, 2001. Brown, H. Douglas. Principles of language learning & teaching. New York: Longman, 2000. Candau, Vera Maria & Moreira, Antonio Flavio. Multiculturalismo e práticas educativas. Petrópolis: Editora Vozes, 2013.

178

Maria Josele Bucco Coelho

Clifford, James. Las culturas del viaje. In Revista de Occidente. Madrid: Arce, n. 170–171, 1997, pp. 45–74. Cornejo Polar, Antonio. Una heterogeneidad no dialéctica—sujeto y discurso migrante en el Peru moderno. Revista Iberoamericana, v. 176–177, 1996a. Available at: http://revistaiberoamericana.pitt.edu/ojs/index.php/Iberoamericana/article/viewFile/6262/6438. Accessed on: 12 May 2014. Damen, Louise. Culture learning: The fifth dimension in the language classroom. Cambridge: Cambridge University Press, 1987. Freyre, Gilberto. Casa-Grande & Senzala. São Paulo: Global Editora, 2005 1st ed. 1933. García Canclini, Néstor. Culturas Híbridas. São Paulo: EDUSP, 2000. Guilherme, Manuela. ‘Glocal languages’: The ‘globalness’ and the ‘localness’ of world languages. In Coffey, S. & Wingate, U. (org.). New Directions for Research in Foreign Language Education. Abingdon: Routledge, 2018, pp. 79–96. Guilherme, Manuela & Dietz, Gunther. Difference in diversity: Multiple perspectives on multicultural, intercultural, and transcultural conceptual complexities. Journal of Multicultural Discourses, 2015. no. 10:1, pp. 1–21. Hall, Stuart. A Identidade cultural na Pós-Modernidade. Trad. Tomaz Tadeu da. Silva. Rio de Janeiro: DP&A, 2009a. Hall, Stuart. Da diáspora: identidades e mediações culturais. Trad. Adelaine La Guardia Resende. Belo Horizonte: UFMG, 2009b. Holanda, Sérgio Buarque de. Raízes do Brasil. 26 ed. São Paulo: Companhia das Letras, 1995 (1st ed. 1936). Imbert, Patrick. El multiculturalismo en Canadá. Pontificia Universidad Católica del Perú. . Available at: http://puntoedu.pucp. edu.pe/entrevistas/multiculturalismo- canada- entendido- como- extensionderechos-humanos/. Accessed on: Jan. 2017. Imbert, Patrick. Multiculturalism in the Americas. Ottawa: University of Ottawa, 2011. Klein, Christian P. C. & Bell–Santos, Cynthia A. O papel da atitude cultural de alunos brasileiros imersos em cursos de inglês como segunda língua nos Estados Unidos. Revista Horizontes de Linguística Aplicada. Brasília: UNB, 2006, v. 5, n. 2, pp 59–85. Kramsch, Claire. Context and culture in language teaching. Oxford: Oxford University Press, 2004. Kramsch, Claire. Language and culture. Oxford: Oxford University Press, 1998. Kusch, Rodolfo. La seducción de la barbarie. In Obras completas. Tomo I. Córdoba: Editorial Fundación Ross, 2000. Kymlicka, Will. Multiculturalism: Success, Failure, and the Future. Washington, DC: Migration Policy Institute, 2012. Mendes, Edleise. A Perspectiva Intercultural no Ensino de Línguas: uma relação “entreculturas”. In Ortiz Alvarez, Maria Luisa, & Silva, Kleber A. (orgs.). Linguística Aplicada: múltiplos olhares. Campinas: Pontes Editores, 2007, pp. 119–139. Mendes, Edleise. Abordagem comunicativa intercultural: uma proposta para ensinar e aprender língua no diálogo de culturas. 2004. Tese (Doutorado em Lingüística Aplicada) - Campinas: Instituto de Estudos da Linguagem, Universidade Estadual de Campinas. Mignolo, Walter. Estéticas y opción decolonial. Bogotá: Universidad Distrital Francisco José de Caldas, 2012. Moran, Patrick. Languaje and culture. In Teaching culture: Perspectives in practice. Boston: Heinle & Heinle, 2001, pp. 34–47

Multiculturalism and Glocal Languages

179

Moreira, Antônio Flávio Barbosa. A recente produção científica sobre currículo e multiculturalismo no Brasil: avanços, desafios e tensões. Revista Brasileira de Educação. Universidade Federal do Rio de Janeiro, Faculdade de Educação, 2001. Available at: www.scielo.br/pdf/rbedu/n18/n18a07. Accessed on: Jan. 2017. Quijano, Aníbal. Colonialidad del poder, eurocentrismo y América Latina. In: Lander, Edgar (ed.) La colonialidad del saber: eurocentrismo y ciencias sociales. Perspectivas Latinoamericanas. Buenos Aires: CLACSO, Consejo Latinoamericano de Ciencias Sociales, julio 2000, p. 246. Available at: http:// bibliotecavirtual.clacso.org.ar/ar/libros/lander/quijano.rtf. Accessed on: 5 Jun. 2013. Santos, Boaventura de Sousa. A Gramática do tempo. Para uma nova cultura política. Porto: Edições Afrontamento, 2010. Santos, Boaventura de Sousa. Modernidade, identidade e a cultura de fronteira. In Tempo Social. São Paulo: USP, v. 5, 1993, pp. 31–52. Santos, Boaventura de Sousa. Toward a Multicultural Conception of Human Rights. In Hernandez-Truyol, Berta (ed.) Moral Troika Of Rhetorical Rituals: A Critical Anthology. New York: New York University Press, 2002, pp. 39–60. Shohat, Ella. & Stam, Robert. Crítica da imagem eurocêntrica. Trad. Mário Soares. São Paulo: Cosac Naif, 2006. Souza, Lynn Mario T. Menezes de. Cultura, língua e emergência dialógica. In Revista Letras e Letras. Uberlândia: v. 26, no. 2, jul–dec 2010, pp. 289–305. Spivak, Gayatri Chakravorty. Pode o subalterno falar? Belo Horizonte: Editora da UFMG, 2010.

Section V

English as Glocal Language

9

English (Mis)education as an Alternative to Challenge English Hegemony A Geopolitical Debate Daniel de Mello Ferraz

Introduction The world we live in now seems rhizomic (Deleuze and Guattari 1987), even schizophrenic, calling for theories of rootlessness, alienation and psychological distance between individuals and groups on the one hand, and fantasies (and nightmares) of electronic propinquity on the other. Here, we are close to the central problematic of cultural processes in today’s world. Appadurai (2008)

In Duboc and Ferraz (2018) we discussed our feelings of frustration, anguish, and exhaustion in relation to recent neoliberal neoconservative political and educational times. In this chapter, the debate goes on surrounded by several headlines: “An impenetrable wall will be built between Mexico and the USA”, affirmed President Donald Trump as soon as he was elected. “Paint it grey!” This is the decision of São Paulo’s mayor, João Doria, when he decided to cover all the creative and colorful arts graffiti spread in street walls and buildings. In order “to beautify” the city, the mayor ordered the painting of the open-air works of art in grey. “Brazilian judge approves ‘gay conversion therapy’, sparking national outrage”: Under the name of Gay Cure Project, psychologists are allowed to initiate treatment in order to treat or reverse homosexuality in Brazil. “More than R$ 3.000,00 a month”: This is how much it costs to study in the most “efficient and successful” schools in São Paulo. “Escola sem Partido: por uma lei contra o abuso da liberdade de ensinar”1 (School without Party: a law against the indoctrination in teaching): an educational law that attempts to convince society that teachers should not be expressing/discussing their own political, religious, and educational views (and their opinions, ideologies) during classes. These snapshots depict some of the recent happenings in Brazil and in the world. Even though they might be staggering to many of us— language educators who struggle for more dialogical relations amongst

184

Daniel de Mello Ferraz

beings (Haraway 2016)—many others believe they are the right things to do. If for Appadurai above the world seems rhizomic and schizophrenic, these snapshots—and so many others taking place around the globe— make a case for the schizophrenic side, showing that the abyssal thinking (Sousa Santos 2007) and its consequences are more present than never. Still according to Duboc and Ferraz (2018), the complex binary world of “us” (better, developed, owners of the truth, “knowledgeable”, who come from the epistemic center of the world) and “them” (worse, underdeveloped, ignorant, who belong to the epistemic periphery of the world) is at stake when we acknowledge the aforementioned news. This problematic relation certainly reverberates in education, politics, languages, cultural, and glocal relations. These snapshots are good examples of the relationship of “us” against “them”. In the case of the USA-Mexico wall, Trump’s supporters claim that Mexicans must be avoided for they “steal” American citizens’ jobs and negatively influence USA language and culture. For mayor Doria’s supporters, graffiti cannot be considered art, and grey is a much “better and neutral” colour. Religious followers, especially the supporters of Brazilian federal deputies Bolsonaro and Malafaia, amongst many other evangelical devotees, truly believe that it is possible “to cure” a gay person. For the Brazilian elite, a good school is an expensive one, not a public one: This is why parents are willing to pay R$3,000.00, sometimes much more, to have their kids educated. Finally, supported by religious beliefs, senator Magno Malta and his evangelical compatriots have insisted on the School without Party project in which teachers should be neutral in relation to all school subjects, as if this were even possible. Cultural practices are certainly influenced by—and influence—this relationship/separation “us” and “them”. The ways we conceive of culture help us problematize this dichotomy “us” versus “them” as culture is oftentimes defined in binary oppositions such as high versus low culture in which high culture comes from the epistemic centers (Europe, USA) and low culture is relegated to the rest of world. We face stiff challenges in contemporary times: violence of all kinds, symbolic (e.g. homophobia in schools), physical (e.g. the killing of police officers in Brazil or the violence of assaults experienced by so many in the country), and ideological (e.g. a political and religious backlash, extremely dichotomous and potentialized by mass media). As we have seen, in cultural terms, the dichotomy “we” (we have our culture and religion) versus “them” (their culture and religion are unacceptable) has generated intolerance, wars, and mass migrations of humans from their own countries. In Brazil, debates rage among religious institutions and are aggravated by the fact that religion and politics increasingly intersect (e.g. in the National Congress, many of the deputies are religious leaders). As a result, evangelical politicians have fomented homophobic bills and prevented the discussion about sexuality

English (Mis)education vs English Hegemony

185

and homosexuality to get to schools (e.g. the School without Party law). In this sense, Sousa Santos (2002: 40) clarifies that “to speak of culture and religion is to speak of differences, boundaries, particularity. How can human rights be both a cultural and global politics?” Though these challenges have apparently nothing to do with language education, they certainly have ripple effects on it. In relation to education, Apple (2004, 2009), Biesta (2000) and Ferraz (2015b) affirm that education faces an era of measurement and certification, based upon growing neoliberal and neoconservative waves. For Apple (2004: 13), “We need to make closer connections between our theoretical and critical discourses on one hand and the real transformations that are currently shifting educational policies and practices in fundamentally rightist directions on the other hand”. According to Biesta (2000: 12), “The rise of measurement culture in education has had a profound impact on educational practice, from the highest levels of educational policy and supra national level down to the practices of local schools and teachers”. It is therefore likely to assume that education has practically been forced into concentrating on measurement and effectiveness (what Brazilian philosopher Saviani calls the pedagogy of technicism). Effectiveness and measurement might be important educational values, but the problem is that “effectiveness is an instrumental value, a value which says something about the quality of processes and about the ability to bring about certain outcomes in a secure way (. . .) this is why effective education is not enough” (Biesta 2000: 3). Ferraz (2015a) argues that in the case of language education in Brazil, neoliberalism and education have been put together to evoke effectiveness (e.g. teaching and learning focused on achieving higher linguistic levels, such as native-like pronunciation), competitiveness (e.g. knowing English means getting better jobs), and instrumentality (e.g. learning the four habilities in order to communicate effectively, regardless of the context the language is being taught and learned). Complementing the debate, Giroux (1999) contends that the political and ideological times do not seem favorable for teachers at the moment. However, still according to Giroux, these very same times offer us educators the challenge of joining the public debate with our own critique as well as the opportunity to engage in much-needed self-critique regarding the nature and purpose of teacher education and of dominant forms of schooling. This can be an opportunity for educators to really integrate in decision-making in public school reforms. This chapter places critical perspectives on glob(c)alization, (glocal) languages and language education in Brazil. In order to do so, it is divided into three sections. This introduction briefly discussed the complex relations of the world of “us” versus “them”. The second section presents a brief state-of-the-art of globalization, and its connections to glocal relations and

186

Daniel de Mello Ferraz

language education. Section three investigates some perspectives regarding language education in Brazil by means of the analysis of two syllabi of two disciplines I have taught in the Letras course of a public university. Complementing this curricular analysis, the voices of the pre-service teachers who attended the courses are brought to the fore. To (in)conclude, I suggest three orientations in relation to language education in Brazil.

Globalization(s), Glocal Relations, and Language Education The epistemological relations between the geographical North and South have been those of colonial supremacy of the former upon the latter, perpetuating for a long time beyond the end of political colonialism. However, epistemological relations between the North and the South, particularly on the Atlantic Ocean axis, that is, western Europe, the Americas and Africa, have been changing while the geographic South, for a number of reasons and in different ways, has started to talk back. (Guilherme 2018)

Over the past decades, societies (especially urbanized ones) have been under fast-paced transformations when one thinks of communication, digitality, and consequently, social relations. In this sense, Kingwell (2000) contends that contemporary transformations happen within the time lapse a mobile phone drops from our hands. Globalization epitomizes these transformations. For Suárez-Orosco and Qin-Hilliard (2004: 14), “Globalization will continue to be a powerful vector of worldwide change. We need better understanding of how education will be transformed by globalization and how it, in turn, can shape and manage the course or courses of globalization”. In other words, we need “a major research agenda to examine how education most broadly defined can best prepare children to engage in a global world. We need better theoretical understandings of globalization’s multiple faces—economic, demographic, social, and cultural” (Suarez-Orosco & Qin-Hilliard (Ibid.). Globalization is a (saturated?) keyword that has been fueled by (post)modern post-industrial societies. According to Suárez-Orozco and Qin-Hilliard (2004), globalization defines (post) modern society. For Robertson (1995 apud Beck 2002), global and local (Glocalization) coexist and necessarily intertwine. Nevertheless, Appadurai (2008) draws attention to hegemonic views of globalization, defending that “grassroots globalization” is the one that moves in the opposite direction of top-down approaches and fosters a process from below, from the local, from the popular. By the same token, Guilherme (2018: 79) critiques this top-down view of globalization, “as if it fell down upon us, a kind of fate we cannot escape”. Ribeiro (2005) also wrestles with the concept of (hegemonic)

English (Mis)education vs English Hegemony

187

globalization by presenting an alternative view: cosmopolitanism, which is a “notion that epitomizes the need social agents have to conceive of a political and cultural entity, larger than their own homeland, that would encompass all human beings on a global scale”. In this sense, Sousa Santos (2002: 41) asseveres that “there is strictly speaking no single process called globalization; there are, rather, globalizations; bundles of social relations that involve conflicts and hence both winners and losers”. Thus, globalization can be interpreted as a singular, hegemonic process, or the story of the winners (Sousa Santos 2002; Menezes de Souza 2011) or pluralized—globalizations—with multifaceted and contextual meanings and consequences (Menezes de Souza 2011; Sousa Santos 2006; Appadurai 2006). For Suárez-Orozco and Qin-Hilliard (2004), globalization defines our era: “In recent years, globalization has come into focus, generating considerable interest and controversy in the social sciences, humanities, and policy circles and among the informant public at large” (SuárezOrozco & Qin-Hilliard 2004: 1). The same authors outline that we can approach globalization as the processes of change structured by four inter-related formations: (1) postnational forms of production and distribution of goods and services—fueled by growing levels of international trade, foreign direct investments, and capital flows; (2) information, communication, and media technologies that facilitate exchanges and instantaneously connect people across vast geographies; (3) growing levels of worldwide migration; and (4) the resultant cultural transformations and exchanges that challenge traditional values and norms in both sending and receiving countries. These four main characteristics comprise very positive aspects of globalization (production and distribution; connection; migration; cultural transformation and exchange). Although they may seem plausible to economists, they need to be problematized so that globalization is not only told as if it were a story solely about the winners (those who can afford and benefit from the transformations aforementioned), as Sousa Santos and Menezes de Souza have highlighted. Brydon and Coleman define globalization as referring to “processes, specifically the spread and growth of transplanetary connections between people. This growth might take place in economic, political, cultural, migration, military, and other realms” (Brydon & Coleman 2008: 8). Alternatively, Appadurai (2008) expands these definitions of globalization within a sociological perspective: “Globalization is certainly a positive buzzword for corporate elites and their political allies. But for migrants, people of color, and other marginals (the so called South in the Earth) it is a source of worry about inclusion, jobs, and deeper marginalization” (Appadurai 2008: 35). For Jefferess (2008: 33), “Global poverty is a ‘distant concern’ (Dower 2003: 137). The ‘evil’ of poverty must be ‘tackled’ by those in positions of privilege”.

188

Daniel de Mello Ferraz

Appadurai (2006: 2) questions the very notion of globalization by inquiring, “What is the hidden dowry of globalization?” In other words, the author questions Suarez-Orozco and Qin-Hilliard’s notions provided that the idea that globalization “is about a world of things in motion somewhat understates the point. The various flows we see—of objects, persons, images, and discourses—are not coeval, convergent, isomorphic, or spatially consistent”. For the author, the flows are in “relations of disjuncture” (Appadurai 2006: 4). By the same token, Sousa Santos (2006) problematizes globalization in two perspectives, a lot similar to Appadurai’s notions: (1) local globalism in which “the specific impact of transnational practices and imperatives on local conditions that are destructured and restructured in order to respond to transnational imperatives” (Sousa Santos 2006: 397); (2) global localism in which it is an illusion to think of something global for what we really experience are localities that connect to each other in order to give the impression—illusion—of globalness. Robertson (1995) coined the term glocalization in which localities and the “global” not only influence each other, but also do not survive without each other. Robertson’s idea of glocalization suggests that the global and local are not out there, but here (Robertson apud Beck 2002: 23). Global and local coexist and intertwine in complex ways, not in vectors going back and forth in opposite directions (local >>