Amazighs in Moroccan EFL textbooks: An integrated critical discourse analysis

Abstract This study is a critical appraisal of official knowledge about: (1) Amazigh ethnic groups in Morocco; (2) the socio-semantic resources for representing them; and (3) the interaction between Amazigh ethnic groups and the dominant Arab group in 33 EFL textbooks which have been developed, approved, and distributed by the Moroccan Ministry of Education, and have been required to be used in every school, public or private from the early 1980s up to the present. Examined though an integrated CDA model which is inspired by Van Leeuwen’s (1996) social actor analysis, Scott's (2018) sociology of the (un)marked, and Bank’s (1989) Ethnic Content Integration model, the findings demonstrated that Amazighs have received varying degrees of discursive representation, ranging from suppression, fixation, cataloguing and backgrounding to partial inclusion and fractional participation. The analyzed EFL textbooks were also found to promote an official stance that can be lexicalized in five main discourses about Amazighs. Such a stance, I argue here, is a clear instance of an exclusionary discourse whose impacts, the study recommends, should be well-adjusted by integrating more precise and wide-ranging ethnic knowledge in Moroccan EFL textbooks.


PUBLIC INTEREST STATEMENT
The ethnic composition of Moroccoa strategically North African Country-is made up of Arabs, the dominant social group, and Amazighs, the majoritarian minority.This multicultural composition has been a constant source of diversity and an asset to strengthen national unity.Notwithstanding this apparent socio-ethnic pride, Moroccohas always had to face the arduous task of "nation-building" that requires the fulfilment of a unified and continuing national narrative that would articulate a distinct Moroccan identity.This study positions official school textbooks in their context of microdynamics of power negotiation and critically appraises the portrayal of Amazighs in EFL textbooks.The analysis has shown the fragility of the discursive inclusion of Amazighs, limiting them to specific social activities that define them in terms of who they, permanently or unavoidably, are, which, in turn, leads to the objectification of Amazigh social actors through the fixation of their way of life.

Introduction
In Morocco, the linguistic, cultural, and ethnic background is both rich and complex.The ethnic composition of the country is largely made up of Arabs and Amazighs.This complex multi-ethnic and multilingual setting is replete with contradictions and disparities in the sense that "nothing is what it seems to be" (Ennaji, 2005, p. 1).While Arabs constitute the dominant ethnic group, both linguistically and politically, the status of Amazighs has never ceased to provoke debate.Key Amazigh grievances include under-representation and under-valuation.Amazigh activists militate for a formal acknowledgement of the linguistic and the cultural rights of Amazighs, thus, denying the deliberate reduction of their culture to folklore status (Maddy-Weitzman, 2012), denouncing the decline in the rate of Moroccans who use the Amazigh varieties which renders Amazigh speakers "a minority within the region" (Maddy-Weitzman, 2015a, p. 1), and, above all, promulgating "a redefinition of Morocco on the basis of its pre-colonial and pre-Islamic Berber heritage" (Hoffman & Miller, 2010, p. 86).In their forward-looking strategy, the Amazigh cultural and political movement calls for more social visibility in public life domains (Idhssaine & El Kirat, 2019), fair access to language and education rights (Gagliardi, 2019), joint right to the protection of Amazigh customs from "erosion and disappearance" (Jay, 2015, p. 68), and firm endorsement of their "ethnocultural demands as part of a broader democratic transformation of society and state" (Maddy-Weitzman, 2015a, p. 2499).
In such contexts, it is often assumed that official school textbooks strengthen the dominant national discourses while only partially considering the explicit views and outlooks of the various cultural and ethnic groups (Apple, 2000;Apple & Christian-Smith, 1991;Macgilchrist, 2018;Silverman, 1999;among others).Locally produced textbooks are argued to constitute social capital sites (Bourdieu, 1986) whose key tasks are to institutionalize, legitimize, and thereby naturalize the nation's master narratives through demarcating its borders and acculturating its different peoples to its selected traditions, values, and ideals of an imagined community (Anderson, 1991), or a collective identity (Apple & Christian-Smith, 1991;Williams, 2014;Yaqub, 2014).The ideological supremacy of textbooks is even more problematic in multi-ethnic countries, such as Morocco, where the ethnic content included in the mainstream curriculum is likely to be biased, fragmented, or uneven (Banks, 1993).Such countries have always had to face the arduous and convoluted task of "nation-building" that requires the fulfilment of a unified and continuing national narrative that would shore up a distinct national identity.
The present study grows out of the above concerns and sets out to provide a critical appraisal of ethnic content in officially produced EFL textbooks in Morocco over the last three decades.More specifically yet, I aim at researching the discursive recontextualizations of Amazigh ethnic groups in locally developed EFL textbooks, from the early 1980s up to now.The corpus came from 33 EFL textbooks, counting ancillaries such as teachers' books and workbooks (Appendix 1), which have been developed, approved, and distributed by the Moroccan Ministry of Education, and have been required to be used in every school, public or private.Following Said's (2019) periodization scheme of the development of EFL official textbooks in Morocco, the selected textbooks were distributed over three periods that match the historical development of EFL textbooks in Morocco: The Germinal Phase (from the early 1980s-to the late 1990s), the Critical Phase (from the mid-1990s to mid-2000), and the Take-off Phase (from mid-2000 up to now) This considerable time frame allows for a deep reflection and a reliable analysis of the official narratives that make up Morocco's selective vision of its ethnic components.The examined corpus, the study hypothesizes, records a snapshot of a Morocco moving from a traditional state towards a contemporary civic-state and therefore highlights "the process of pedagogising" (Zhao, 2014) the polytechnic citizens to a new form of social affiliation.And since all the investigated textbooks received the Ministry of Education approval, I believe that they are likely to offer one of the most comprehensive accounts to the evolution of the Moroccan sociocultural identity There are several important areas where this study makes original contributions to the study of minority representation in official school textbooks.The first important significance is purely theoretical as the study advances our understanding of how critical discourse analysis (CDA) as a multidisciplinary and multi-methodological approach for uncovering opaque "structural relationships of dominance, discrimination, power and control" (Wodak & Meyer, 2001, p. 2) can serve as ideology detector in educational settings.Hence, the critique of Moroccan EFL textbooks is also a critique of the state's normative ideology towards Amazighs.The current study aims to contribute to this growing area of research by exploring how official language textbooks could robustly become ideological tools for the dissemination of official values softly imposed by state's hegemonic apparatuses.Another critical significance of the current study is purely methodological.Research in Morocco has for a long time focused on specific textbook types, mainly history and social sciences textbooks (e.g., Hamid, 2015;Maye-Saidi, 2018).Often, the unstated argument has been that history textbooks, for instance, are more likely to bear visible traces of hegemonic coercive discourses.This visibility requirement has cast doubt on other data sources like language textbooks, which have, on the whole, been bypassed.The present study, however, sheds light on a rarely visited area of investigation (EFL textbooks).The choice to investigate language textbooks challenges the mainstream contention that language textbooks are likely to be mere "innocent", "pedagogical" artefacts.
For international comparative ethnic studies, the present study hopes to provide a fresh perspective on how ideological sets of sociosemantic categories are recontextualized in officially produced EFL textbooks in Morocco, and thus energizes internationally well-grounded comparative frameworks.This, it is argued here, would pave the way for new critical perspectives in language education that will challenge the current hegemony of the state's grand narratives .The results of the current study will be of great benefit to international scholars through contributing to knowledge growth about the discursive representations of ethnic groups in EFL textbooks from a critical perspective which views language as an ideological meaning-making potential.The study also has the potential merit of enriching the knowledge about the ways in which national identities are constructed.Therefore, further research in this field would be of great help in the Moroccan context .

Amazighs
Historically, the Amazigh-speaking groups comprised ancestors of the pre-Arab inhabitants of North Africa around 3000 B.C (Sadiqi, 2011); they were the first residents of the land.Their geographical and historical geneses are numerous, hailing from the Mediterranean, the Nile Valley, and the Sahara, and resulting in a composite population (Maddy-Weitzman, 2015a, 2015b).Amazighs' encounters with the Arabo-Islamic civilization around the mid-seventh century have resulted in wide-ranging results across different countries, fluctuating between resistance and free willing participation (ibid).This has led to a Morocco "never been totally 'homogenized', 'Arabized' or 'Islamized'.Morocco is a Berber, Arab, Muslim, Mediterranean and African country", according to Sadiqi (2003, p. 2).Over a considerable period of time, Amazighs, it is argued, have stood up against waves of challenges to maintain a discrete socio-linguistic identity.The rise of Morocco as a modern nation-state signaled the beginning of an ongoing pull between the state and its ethnic Amazigh groups, with Morocco being more inclusive and progressive than other neighboring states hitherto thought to be more repressive and exclusionary (Kruse, 2013).This ebb and flow characterizing Amazighs' identity has resulted in shades of upshots, ranging from "a proud ethnic one, a dormant marginalized one, an insular tribal one and one struggling for its voice in society" (Kruse, 2013, p. 1).
Today, Amazighs seem to account for a substantial, yet statistically ambiguous, proportion of the total population.The relative size of their community cannot be taken as determinant of its socio-political status.According to Zaid and El Allame (2018), researchers tend to cast doubt on the real number of Amazighs in Morocco, claiming that the number is between 60 or 70 percent.Recently, Ahmed El Hlimi Al Alami, chairperson of Morocco's High Commission for Planning, has declared that 26.7 percent of Moroccans speak Amazigh, with 14 percent using Tashelhit, 4.1 percent speaking Tamazight, and 4.1 percent Tarifit.Castellino and Cavanaugh (2013) refer to this ambiguous situation of ethnocultural and religious minorities as majoritarian minorities, political minorities, and trapped minorities "who are numerically larger but excluded from sites of power" (p. 6).
The statistical ambiguity outlined above comes with another terminological opacity.There is not a common compromise among political activists and scholars as to which term (i.e., Amazigh(s), Berber or Imazighen) is most appropriate.Since the 1990s, the term "Amazigh" (or Amazigh language) has seen widespread use, gradually replacing "Berber" to the extent that it is accepted in academic discourse.People who speak "Berber," however, favor "Imazighen", in the singular "Amazigh", which means a "free (nobel) man" over the appellation "Berber".Some linguists, however, (e.g., Ennaji, 2005;Sadiqi, 2009;Youssi, 1990) prefer to retain the term "Berber" over other labels because of two reasons.First, the term "Berber" in the Anglo-Saxon world, does not carry the same negative implications.The second reason is that some linguists believe that the "term 'Berber' is neutral whereas 'Tamazight' cognates with the name of a specific variety of Berber, a loose taxonomy that might exclude Tashelhit, Tarifit, and the other tribes" (Sadiqi, 2009, p. 12).
In politics, the shift in Morocco's official policy towards Amazighs along with the adoption of a new language policy have resulted in the recognition of the Amazigh culture as an integral component of the state's mosaic ethnic make-up and with Tamazight as a national language.During the 2011 Arab Spring, the Amazigh language was officialized alongside Arabic in the New Constitution (art.5), which asserts that the Moroccan culture "is forged by the convergence of its Arab-Islamist, Berber and Saharan-Hassanic components, nourished and enriched by its African, Andalusian, Hebraic and Mediterranean influences" (The Moroccan Ministry of National Education, Vocational Training, Higher Education and Scientific Research, 1999).The new constitutional gains have had broad symbolic significance, and have been greeted with a massive approval from the Amazigh movements in Morocco (Madani et al., 2012), In education, the National Charter for Education and Training (NCET,1999) and the Royal Institute of the Amazigh Culture have helped in the standardization and promotion of the Amazigh language and culture.Equally, the establishment of the Royal Institute, largely referred to in French as "Institut Royal de la Culture Amazighe" (IRCAM), have responded to a long-standing demand by the Amazigh movement for better inclusion of Amazigh speakers into the national fabric.In 2003, and in partnership with IRCAM, the Moroccan Ministry of Education initiated the first formal teaching of Amazigh language.By so doing, 'Morocco was the only North African state, "and only the second core Arab League member state (after Iraq), in which Arabic was not the sole official language" (Maddy-Weitzman, 2015a, p. 5).
In its quest to correct "potential" ethnic imbalances and ensure a unified national narrative, the Moroccan government has turned to different educational policies and reforms to boost economic development and foster national unity.One of the main aspirations of the Moroccan educational system has been to promote ethnic harmony, national unity, gender equality and economic competitiveness.In this regard, NCET (1999), National Curriculum (The White Book), and The Strategic Vision (2015-2030) have advocated a solid and an indivisible national unity through the enhancement of the Moroccan cultural diversity where Amazigh cultural and social rights are central.The Strategic Vision, the last episode in the educational reforms in Morocco, for example, affirms that the "Amazigh language is an official language and a common denominator of all Moroccans without exception" (Strategic Vision, , p. 37).

School textbooks and the ideological construction of knowledge
A state educational system assumes an ideologically instrumental function, and it is incumbent upon this system, embodied in institutions like schools, educational artefacts, social capital sites and reproduction systems (Bourdieu, 1986), to institutionalize, legitimize, and thereby naturalize the nation's master narratives through demarcating its borders and acculturating its different peoples to its selected traditions, values and notions of an imagined (Anderson, 1991) or collective identity (Apple & Christian-Smith, 1991;Williams, 2014;Yaqub, 2014).In such contexts, official school textbooks constitute the state's official knowledge (M.W. Apple, 2000), moral history (Friedrich, 2014), and romanticized portrait of itself (Williams, 2014) and are, therefore, perceived as "vessels ripe for the embodiment and transmission of such state-envisioned histories, memories, and discourses of nation(hood)" (Yaqub, 2014, p. 222).In the study, school textbooks are understood as ideological sites of struggle, published within "the political and economic constraints of the market" (Apple & Christian-Smith, 1991, p. 2), and "mediated through the textual and visual worlds of the state's selective narrative" (Curdt-Christiansen & Weninger, 2015, p. 1).
Locally produced textbooks assume the function of naturalizing the official imaginations (Anderson, 1991) through intricate, and often opaque, national narratives that describe "how things were, what happened, and how they came to be the way they are now" (Williams, 2014, p. 1).Although official school textbooks often claim neutrality, objectivity, certainty, and factuality (Olson, 1989), from a critical perspective, it seems that these claims rest on flimsy grounds.In Venezky's (1992) terms, school textbooks are situated in an intermediate position between "latent curriculum" and the "manifest curriculum".This intermediate position allows official textbooks to convey the "selective traditions and ideologies of dominant social and cultural groups" (Yaqub, 2014, p. 222), and to create unquestionable common sense (Apple, 2000), especially with those nation-states standing out of colonialism and struggling "to revive particular national myths and narratives long repressed, obscured, and quite selectively edited" (Yaqub, 2014, p. 222).

School textbooks and ethnic inequalities
Against the specific conceptualizations of school textbooks outlined above, a strong and diverse collection of studies using different lenses have examined state-produced textbooks in order to uncover potential ethnic inequalities (e.g., Altbach, 1991aAltbach, , 1991b;;Altbach & Kelly, 1988).The central argument has been that, in the process of selective representation, ethnic groups and their cultures are seriously underrepresented, marginalized (Chu, 2015), and/or left unimagined in the official narrative (Anderson, 1991).
Over the past two decades, a significant body of research has emphasized the use of use of CDA to examines ethnic inequalities in official school textbooks.The emphasis is even stronger in multiethnic countries such as Morocco where it is incumbent upon the educational system to harbor the pressures of diversity in a culturally dissimilar population.Scholars from diverse places such as China (Cheng, 2013;Chu, 2015;Gladney, 1994;Zhao, 2014), Japan (Taylor, 2017), Korea (Yim, 2003), Turkey (Çayır, 2015), Belgium (Zanoni et al., 2017), Sri Lanka (Tilakaratna, 2016), Sudan (Mugaddam & Aljak, 2013) and Egypt (Abdou, 2017) have examined the status of ethnic groups in the pedagogical context.Findings indicate that the majority of the reviewed studies reveal that official textbooks are generally biased, promoting one version of national identity where minority ethnic groups are often left unimagined.In other words, nation states have failed to properly handle ethnic heterogeneity.Chu (2015), for example, points out that textbooks in China seem to present dominant groups in terms of knowledge and values while downgrading other "minorities" who remain marginalized and almost invisible and are articulated from the perspective of the dominant ideology (p.469).Findings have also revealed that the portrayal of ethnic groups in teaching materials tend to focus on specific recurrent themes where ethnic groups are often linked to festivals, singing and dancing.Ethnicity, here, seems to conjure up images of primitivity, poverty, tradition, and nostalgia.The discursive construction of space is also significant as places that are linked to ethnic groups are often represented as poor, purely rural and natural (Chu, 2015;Slimi, 2009).Consequently, implications tend to urge governments, textbook designers, and various stakeholders to take urgent actions to ensure that ethnic groups are granted equal and fair representation.

Critical discourse analysis
This study privileges CDA as a multidisciplinary and multi-methodological approach for exposing structural relationships of domination, discrimination, and power in discourse (Fairclough, 1995;Van Dijk, 1993).Textbooks are understood as ideological sites of struggle, published within "the political and economic constraints of the markets" (Apple & Christian-Smith, 1991, p. 2), and orchestrated through the linguistic and pictorial modalities of the state selective narratives (Curdt-Christiansen & Weninger, 2015).
In their endeavor to unveil the interconnectedness of social problems, CD analysts choose "the perspective of those who suffer most, and critically analyze those in power, those who are responsible, and those who have the means and the opportunity to solve such problems" (Van Dijk, 1986, p. 4).According to Wodak et al. (2009), the central objective of CDA is "to unmask ideologically permeated and often obscured structures of power, political control, and dominance, as well as discriminatory inclusion and exclusion in language use" (p.8).
In this study, the privileged understanding of discourse is largely informed by Bernstein's (1996) notion of pedagogical recontextualizations and Van Leeuwen's (2008) extension of the term that all discourses, pedagogical or otherwise, are recontextualized social practices.Our understanding of discourse undertakes a critical language analysis of the verbal as well as the visual representation of the social practices in all social contexts.Given the centrality of discourses as recontextualized social practices, the study follows the chains of arguments that Van Leeuwen has developed in several writings and has finally culminated in his (Van Leeuwen, 2008) seminal work.The prime aspect of recontextualized social practices hinges on the hypothesis that "discourses not only represent what is going on, they also evaluate it, ascribe purpose to it, justify it, and so on, and in many texts these aspects of representation become far more important than the representation of the social practice itself" (Van Leeuwen, 2008, p. 6).

Social actor analysis
Social Actor Analysis (SAA) has the critical merits of combining linguistic and sociological insights to the analysis of how social actors are represented in discourse.Van Leeuwen (2008, 2009) has been credited for outlining "a sociological grammar" (Van Leeuwen, 2008, p. 56) by developing a "socio-semantic inventory" which outlines critically relevant connections among the various lexico-grammatical options available to represent social actors and therefore make salient their theoretical and critical relevance (Van Leeuwen, 2008).Van Leeuwen's socio-semantic inventory has been intensively drawn on by researcher to examine the officially produced textbooks (e.g., Dashti & Mehrpour, 2017;Roohani & Tanbakooei, 2012).
The Social Actors Approach is one of the major approaches to CDA.The approach broadly draws on sociological and linguistic insights, particularly the reciprocal relationship between action and social structure whereby discursive representation is eventually grounded in social practice.'SAA introduces sources from Malinowski to Parsons, and from Bernstein to Bourdieu.This idea of individual actors permanently constituting and reproducing social structure is linked with a Foucauldian notion of discourse, somewhat similar to DA and DRA' (Wodak & Meyer, 2009, p. 26).Although Van Leeuwen's (2008) model is a linguistically oriented conceptual framework, he did not, start out from linguistic operations such as nominalisation and passive agent deletion, or from linguistic categories such as the categories of transitivity, but instead seek to draw up a sociosemantic inventory of the ways in which social actors can be represented, and to establish the sociological and critical relevance.(p.23) The socio-semantic network which Van Leeuwen (2008) proposes involves four key types of transformation: addition, deletion, rearrangement, and substitution.It, therefore, reveals the extent to which social actors are excluded or included, assigned active or passive roles via rearrangement, substituted or spatialized.In the methodology section, I will explore the sociosemantic units of analysis which are relevant to the purposes of the current study.

Research questions
Privileging an integrated understating of CDA to the representation of social actors in school textbooks, this study attempts to provide a firmly grounded understating to the following questions: • How have Amazigh social actors been represented in Moroccan ELF textbooks?
• What knowledge and beliefs (discourses) about Amazighs have been built in Moroccan EFL textbooks?

Corpus
The corpus came from 33 officially produced EFL textbooks (counting ancillaries such as teachers' books and workbooks) which have been developed, approved, and Distributed by the Moroccan Ministry of Education, and have been required to be used in every school, public or private from the early 1980s up to the present.Chronologically, the selected textbooks have been located along a timeline continuum, with two salient phases which are separated by a transitory-critical phase: The Germinal Phase (from the early 1980s to mid-1995), the Critical Phase (from mid 1990s to mid-2000), and the Take-off Phase (from mid-2000 up to now).
Each phase is emblematic of the nation's socio-political tone of the time.The Germinal Phase witnessed the production and distribution of Steps to English, Further Steps in English, and the English in Life textbook series.This phase reflects the post-independence narrative of national revival with the prevailing pro-Arab stances and sentiments of nation-building and national reconciliation.
The Critical Phase was a short-lived period which witnessed the publication of Highway and Quick way textbook series, which were developed to fill a pedagogical gap and to pave the way for the Take-off Phase.The last period marks the high influence of internationalization and the pedagogization of the Millennium Development Goals (MDGs).This phase has also resulted in a comprehensive curriculum reform with radical changes in teaching approaches and methodologies, signaling a move from traditional to more communicative and task-based teaching/learning modes.
This considerable time frame (from the early 1980s up to the present) allows for a deep reflection and a reliable analysis of the narratives and discourses that make up Morocco's official vision of its cultural components.Similarly, the large corpus of EFL textbooks offers critical insights into the discursive representation of ethnicity along with its accompanying discourses.These textbooks, this study hypothesizes, offer a socio-historical portrait of a Morocco moving from a traditional state towards a modern nation-state and highlight the process of pedagogizing (Zhao, 2014) the polytechnic citizens to a new form of social affiliation.Finally, since all the investigated textbooks received the Ministry of Education approval, the study claims that they offer one of the most extensive accounts to the growth of the identity narrative of "the Moroccan" nation.

Theoretical units of the analysis
In CDA, units of analysis are not stable; they are often tailored according to the different research agendas that scholars set beforehand.Generally, researchers working within the CDA paradigm tend to focus on different units of analysis which are often situated between the clause and the whole texts or conversations (Babaii & Sheikhi, 2017).In this study, I adopted Van Dijk's (1981) "episode" as a meso-level discursive unit of analysis, which is situated between clause/sentence and the whole unit of conversation or text.According to him, "episodes appear to be linguistically and psychologically relevant units of discourse structure and processing" (p.192).

Amazighs vs. Arabs
Before deciding on the "appropriate" analytical categories and scheme needed for the analysis of the representation of Amazighs in Moroccan EFL textbooks, the precise description of who would qualify as Amazigh is compulsory.This category is particularly problematic because it has been so ill-defined, polysomic, and contested, raising the legitimate question as to who would(not) count as an Amazigh.The analysis proceeded in two complementary steps.
First, the differentiation between direct and indirect inclusion (be it linguistic or visual) of Amazigh social actors was established.Direct inclusion comprised instances where the discourse clearly designated certain social actors as being "Berber or Amazigh".Indirect inclusion, be it linguistic or visual, defined Amazighs by zooming on predefined sociocultural aspects inherent to them, such as language, location, dress, food, festivals, and ceremonies.Instances of analysis are critically relevant "insofar as they are critically nontrivial and can be coupled with distinct lexicogrammatical realizations" (Van Leeuwen, 2008, p. 57).Therefore, I considered to be relevant all linguistic or pictorial elements describing Amazigh, be they direct or indirect.
The second step places Amazighs in a dialectal relationship to Arabs, thought to be the dominant social group.This choice redefines ethnicity in Morocco in terms of two subcategories: Amazighs vs. Arabs.The most troublesome issue here, however, has been that due to the Arabs' prevailing status in Morocco, they have been portrayed as if they were simply "Moroccans".In other words, "Moroccanness" has usually been normalized and replaced by an unmarked cosmopolitan image (Chu, 2015) representing an unspecified "Moroccanness".
Inspired by Chu's (2015) analysis of Chinese school textbooks, and in order to capture this undistinguishable representation of the Arabs as "Moroccans" in general, a code of "Arab" was given to an episode when: (1) the ethnicity of the participants or ethnic features of "Arab" was clearly exposed in the texts or the description of the pictures; and (2) a reference to the features of Arabs was not openly told but implied by linguistic and/ or pictorial information.

Socio-semantic categories of analysis
Theoretically, the working units of analysis adopted in this study have drawn on the "Social Actor Analysis Model" (Van Leeuwen, 2008).However, the interpretation of the findings has been inspired by insights from the "sociology of the invisible/the unmarked" (Blumer, 1969;Brekhus, 1998;Scott, 2018), Abdou's (2017) analytical model to the analysis of Copts in Egyptian history textbooks, and Chu's (2015) three distinct yet related analytical categories used in the portrayal of ethnic groups in China's elementary textbooks, and Banks (1989Banks ( , 1993Banks ( , 2003) ) model of Ethnic Content Integration.The socio-semantic units of analysis include: (1) Inclusion (direct vs. indirect inclusion).Understood in terms of linguistic/visual visibility, inclusion was the first socio-semantic category which the analysis set out to investigate.
According to Van Leeuwen (2008), "representations include or exclude social actors to suit their interests and purposes in relation to the readers for whom they are intended" (p.28).This unit of analysis is instrumentalized in relation to the distinction between "marked vs.
unmarked" ethnic groups.The difference is sociological and is deeply rooted in what has come to be known as the "sociology of the invisible" (Blumer, 1969;Scott, 2018), or the sociology of the "unmarked" (Brekhus, 1998), where the choice between visibility or invisibility is purely ideological.Brekhus (1998) asserts that there is an epistemological disparity in our recognition of social groups where the dominant ones are invisible and therefore sociologically unmarked while the non-dominant groups often stand socially marked and are, hence, discursively visible.
(2) Unique vs. shared identity.This entails asking whether Amazighs are represented with regard to their exclusive identity or with regard to the characteristics and functions they share with others (Van Leeuwen, 2008).According to him, when social actors are individually nominated, they are often represented in terms of their unique identity.When categorized, on the other hand, they are represented in terms of identities and functions they share with others (p.52).If heavily categorized, the analysis should highlight the "identities" and/or "functions" that Amazighs are likely to share together.The distinction between "identities" and "functions" is critically relevant because the emerging discourses can either represent them in terms of "being" (identities) or in terms of "doing" (functions).Abdou (2017) notes that "some societal roles such as economic activities that a minority group is said to excel or specialize in provide insights into how subtly that group's characteristics, dispositions and inclinations might be essentialized" (p.8).
(3) Genericization' vs. "Specification".These are two closely related analytical categories that index power.In this very context, Amazighs can be included as classes, or as specific, identifiable individuals (Van Leeuwen, 2008).Again, the choice between generic and specific reference is an important factor in the representation of Amazigh social actors.Abdou (2017) refers to this as "internal homogeneity vs. heterogeneity".According to him, this distinction, "entails asking whether the group is presented as homogenous, 'unanimous group' or a heterogeneous group" (p. 7).
(4) Association and dissociation.Association is another significant socio-semantic inventory which allows Amazighs to be associated with other groups.Association, according to Van Leeuwen (2008), 'refers to groups formed by social actors and/or groups of social actors (either generically or specifically referred to) which are never labelled in the text (although the actors or groups who make up the association may, of course, themselves be named and/or categorized (p.39).Linguistically, the most common realization of Amazighs in terms of association is parataxis.Again, it is of critical relevance to examine the contexts where Amazighs are associated or disassociated with other groups.
(5) Personalization vs. Impersonalization.While "personalization" bestows the feature [+human] on the represented social actors, "impersonalization" designates social actors who do not have the feature of [+human].In the case of impersonalization, Van Leeuwen (2008) distinguishes between abstraction, objectification, and spatialization.Ideologically, impersonalization leads to the passivation of the propositional content associated with ethnic minorities.It suppresses, deagentializes, and backgrounds the social agents.

Data analysis and coding
Data analysis consisted of three stages.First, two recording sheets are used to examine and classify the content elements of textbooks into Linguistic episodes.I reviewed 33 EFL textbooks for any direct or indirect inclusion, mentioning, or allusion to "Berbers", "Amazigh(s)" or "Imazighen".As I examined the relevant episodes, I broke the data into manageable instances and then placed them under the corresponding units of analysis.The second stage was the quantitative analysis.Tables of frequencies and percentages summarizing linguistic representations were generated.Finally, while focusing on the key socio-semantic options defined above, during the analysis, however, remarkable discursive patterns emerged.I termed these patterns "emerging discourses" and analyzed them as recontextualized social practices.
To increase the inter-rater reliability of the analysis, 20% of the data was given to three raters and results were compared using Cronbach's alpha for each category.The inter-rater reliability was between [α = 0.71 and α = 0.83], which ranges from good to high agreement among the raters.It should be noted however, that that the agreement rate keeps fluctuating across the analytical categories without, however, falling below α = 0.70 .In order to facilitate future reference or replication by other researchers, the following order has been proposed in all textbook citations used in this study: Abbreviated textbook title, grade level, and page.Appendix 1 fully itemizes the examined textbooks according to the order proposed above.

Linguistic inclusion
Given the total linguistic inclusion of Arabs, the dominant social group, and Amazighs, the minoritized ethnic groups (Table 1), it seems that Amazighs were included significantly more than Arabs (63.49% vs. 36.50%).Besides, the results of the Chi-square test (p > 0.05) indicate that there is a significant difference between the two included groups of social actors (χ2 = 19.167p > 0.05).Table 1 also reveals an uneven pattern in the statistical distribution between the direct and the indirect inclusion of both groups.A comparison between both groups demonstrates the fact that in terms of "direct inclusion", Amazighs were included more significantly than Arabs (χ2 = 2.240 p > 0.05).However, in terms of "indirect inclusion" the difference is not statistically significant (χ2 = 2.240 p > 0.05).This means that the two groups were treated similarly in terms of direct inclusion Examples (1) and (2) illustrate the type of direct inclusion of both Amazighs and Arabs.Example (1) directly includes Amazighs in terms of nomination (Rabha Akka which is an Amazigh name), functionalization (journalist), spatialization (Middle Atlas which often designates places where Amazigh live), objectification (Berber carpet.Amazighs have been identified with objects such as traditional carpets), and temporalization (tenth century).The representation directly includes the social actors, their actions, time, place, and legitimation.The inclusion is total, and therefore, (1) was directly placed under "direct" inclusion.
Example 1: Rabha Akka is an Amazigh Journalist.She is from a small city in the Middle Atlas (AWW, p. 14).(2) establishes a direct inclusion of the Arabs who are nominated (Ibn Battuta who was a famous Moroccan Amazigh Islamic scholar and explorer), functionalized (traveler and explorer who traveled more than any other explorer in pre-modern history), spatialized (Tangier and Fez which are two major cities in Morocco.Fes, for instance, is considered to be the cradle of the Arabic and the Islamic traditions), appraised (greatest), and temporalized (1304).In other words, the representation includes all the elements required for the discursive inclusion of Arab social actors.
Indirect inclusion, on the other hand, was less straightforward although the representation did leave traces such as reference to language (Amazigh/ Arabic), figures and heroes, clothes, places, events, types of knowledge, physical descriptions, among other things.Example (3), for instance, indirectly includes Amazighs, and therefore the linguistic inclusion was judged to be partial.Three key narrative markers help index the above example as being about Amazighs: nomination (Benhamad Ali which is a common personal name for Amazighs), functionalization (Grocer.Amazighs are often referred to as "shopkeepers"), and spatialization (Essaouira is a Moroccan city where a considerable Amazigh population live).
Example 3: Mr. Benhamad Ali, 58, follows reports about locusts, in the newspapers and on TV, with the same horror today as 30 years ago.Mr. Benhamad is a grocer now, but until 1955 he was a prosperous farmer near Essaouira (EIL 3, p. 51, emphasis added).
Another important observation is linked to the chronological inclusion of Amazigh social actors.Table 2 reveals that the quantitative inclusion of Amazighs has witnessed remarkable fluctuations, going as high as (51.49%) during the Germinal Phase, and as low as (10.77%) during the Critical Phase.Table 2 also shows how the Take-off Phase has witnessed a revival of the linguistic inclusion of Amazighs.In order to capture this discursive fluctuation over time, Table 2 can be graphically represented as follows (Figure 1):

Nomination
The analysis reveals three patterns of nominating Amazighs, and therefore representing them in terms of their unique identity: Formal nomination (Title +Last name), Semi-formal nomination (Title + Given name + Last name), and Informal nomination (First name/ Given name only/ Nickname).Table 3 details the verbal variations in nominating Amazighs across the three phases that marked the evolution of Moroccan EFL textbooks.
Table 3 reveals that it is very hard to specify a unified statistical pattern whereby Amazighs were truly nominated and therefore represented in terms of their unique identity.The only textbook which seems to directly endow Amazighs with the power of "nomination" is Al Masar, a textbook designed for Common Core students and narrates the life of the Itrys family (An Amazigh family) both in Morocco and in London.The Germinal and the Take-off Phases display two contradictory patterns of nomination.While the EFL textbooks of the Germinal Phase predominantly represented Amazighs formally, EFL textbooks from the Take-off Phase opted for the opposite, with Amazighs being predominantly referred to informally (48.93%).

Functionalization
When not nominated, Amazighs were categorized, represented in terms of the social/occupational activities they shared with others.Linguistically, the functionalization of Amazighs has been realized by nouns derived from verbs (such as dancer, farmer, and shopkeeper) or by nouns derived from other nouns used to name tools that are associated with an activity (e.g, carpet weavers).Table 4 classifies the major activities (occupations) in which Amazighs were portrayed.Chronologically, one can notice a quantitative and a qualitative growth in the number of the occupational choices with which Amazighs have been identified.(a) Age: Amazighs were classified either as children or adults.Representing them as "youth/ teenagers" was very rare if not absent (Al-Masar textbook may be the only exception).
(b) Gender: Amazigh females were predominantly foregrounded especially as mothers, wives, daughters, singers, and dancers.This, along with the pictorial representation strengthens the claim of the "feminization of ethnicity".
(c) Class and wealth: Amazighs were most frequently represented as poor, orphans, illiterate, underprivileged and beneficiaries.Al-Masar may be the only exception, where the Itrys are represented as entrepreneurs running their project in London.
The last type of categorization represents Amazighs by reference to their personal kinship, or work relations.The analyzed textbooks revealed three types of relational identification for Amazigh social actors: (a) Tribalism: The examined textbooks seem to strengthen the tribal dimension of the Amazigh way of life.Excessive appeal to "tribalism" gives the impression that all Amazighs live in tribes.
(b) Work relations: Amazighs were often portrayed in specific work relations that require "teamwork or solidarity" such as developing small family projects together, fighting illiteracy together, weaving carpets, or dancing together.
(c) Family: Building strong family ties received considerable attention especially during the Take-off Phase, with a whole textbook (Al-Massar) being entirely devoted to an Amazigh family life, featuring its members, friends, work, daily routine, dreams and aspirations both in Morocco and in London.

Appraisement
Investigating the appraisement of Amazigh social actors throughout the three phases, it can be seen that they have been included as beautiful (example 4), heroic (example 5.), active (example 6), and colourful (example 7).However, this system of appraisement remains problematic unless it is critically examined within and across the various discourses that display Amazighs.

Association and dissociation
Examples ( 8), ( 9) and ( 10) create a discursive association between Amazighs and other social groups such as Phoenicians, Arabs, Jews, and Europeans.In this context, Van Leeuwen ( 2008) notes that "the group is represented as an alliance which exists only in relation to a specific activity or set of activities" (p.39).In (10), for example, "Berber" music is juxtaposed to "Al Ala", an Andalusian music with a strong cultural and political presence in Morocco.
Example 10: Al Ala, Al Gharnati, Al Malhoun, the songs of religious sects, popular songs of cities, Berber songs of the Atlas, of the Rif, of the Sous and the songs of the Grand South are tangible proof of the richness and diversity of this Moroccan musical Heritage (Quick way 2, p. 30).
Table 5 shows that association takes place 28 times and for different reasons: political, historical, and cultural, with the last one accounting for half of the total association rate.Again, the results of the Chi-square test indicate that the observed difference between the three categories is not significant.

Genericization and impersonalization
Overall, Amazighs have been genericized as being "Berbers", "Amazighs", "Atlas dwellers", "shopkeepers", and "naturally born singers and dancers".They have been represented as unspecified, "anonymous" individuals or groups, who have been metonymically represented by reference to something directly linked to their being or the activities they perform.For instance, "traditional carpets", "argan oil", "dates" and "goats" are recurrent ways of objectifying Amazighs.
Another type of impersonalization is spatialization whereby Amazighs have been replaced by a direct reference to the spaces to which they relate.Spatialization, if taken alone, however, may not constitute a reliable criterion for reference to Amazighs given that cities such as Agadir or Essaouira are not homogeneously Amazigh.In this context, the discourse combines other modes of inclusion such as names of the social actors, or the social activities they are most likely to engage in.Excessive spatialization, as in (11) gives the impression that Amazighs are mountain dwellers who barely go beyond the village: Ait Haddidou, Imilchil, and Atlas Mountain are essentially Amazigh locations which are excessively highlighted in Moroccan school textbooks.
Example 11: We visited the wedding festival of the Ait Haddidou near Imilchil, high in the Atlas Mountain (Horizons, p. 85, emphasis added)

Discussion of the results
There are several possible explanations for the results which can be framed within five emerging discourses.The discussion of the findings depends on the results obtained from the recording sheets devised according to research questions and the predetermined main socio-semantic categories and subcategories (Van Leeuwen, 2008).All the episodes (sentences, phrases, clauses, separately and in conjunction with the neighboring phrases, clauses and sentences in reading passages, dialogues, and /or language tasks) were extracted and analyzed in light of Van Leeuwen, 2009, 2008) social actor analysis, Scott's (2018) sociology of the (un)marked, and Banks's (1989Banks's ( , 1993) Ethnic Content Integration model.

Visible yet unseen
The statistically significant inclusion of Amazighs creates a critical moment of wonder which can be lexicalized as follows: How can we explain the significant "visibility" of Amazighs compared to Arabs?While the question remains linguistic, the answer is rather sociological and is deeply rooted in what has come to be known as the "sociology of the invisible" (Blumer, 1969;Scott, 2018), or the sociology of the "unmarked" (Brekhus, 1998), where the choice between visibility or invisibility is purely ideological, re-shifting the sociological radar from focusing on the visible to focusing on the invisible.Brekhus (1998) convincingly notes that there is an epistemological disproportionateness in our recognition of social groups where the dominant ones are invisible and, therefore, sociologically unmarked while the non-dominant groups often stand socially marked and are, hence, discursively visible.
Again, while it is very hard, if not impossible, to locate images of other social groups that make up the Moroccan cultural identity, Amazighs were easy to identify and mark both linguistically and visually.Being visually marked, they, unlike the other social groups, cannot circumvent the sociological radar.This explains why different raters were able to identify images of scenery, villages, objects, and people as being Amazigh.
Bringing those observations under the theoretical apparatus of the sociology of the invisible, we can put forward three significant comments.First, Amazighs are directly included more than Arabs because they are sociologically "marked", and in being so, they can easily be recognized, identified and ultimately ideologically standardized.Quoting Scott (2018), sociological markedness has rendered Amazighs "empirically unusual, politically salient, ontologically uncommon or morally critical . . .we train the sociological gaze upon the extreme, exotic, unusual and deviant" (p.4).Value judgements, therefore, may be built into what Scott (2018) calls the "extreme", the "exotic" and the "unusual".
Second, according to this line of reasoning, Amazighs constitute the unnaturalized part of the Moroccan cultural identity, marking a discursive point of reference for cultural identification and signifying the difference between the figure and the ground and between the mundane and the exceptional.Indeed, being "marked", Amazighs can be examined, scrutinized, and therefore studied.Excessive highlighting of the taken-for-granted rules and routines of Amazighs, in Garfinkel's (1967) words, represents them as the "epistemological blind spot" who are "seen but unnoticed" (Garfinkel 1967, as cited in Scott, 2018, p. 2).
Third, the investigated textbooks seem to display what Apple (2019) designates as "the most conservative form of multiculturalism: 'mentioning'" (p. 195) where the ethnic content related to Amazighs "have the status of add-ons about the culture and history of 'the Other.'" (p. 195), or what Banks (1989) considers the lowest level of ethnic recognition: "addition".This leads to the understanding that, more than often, Amazighs were not included as much as "mentioned", or "added" to the mainstream official narrative.According to Banks (1989), addition "often results in the trivialization of ethnic cultures, the study of their strange and exotic characteristics, and the reinforcement of stereotypes and misconceptions" (p.18).Babaii and Sheikhi (2017) argue that mentioning is not free of bias and has a tinge of "Othering" (p.13) social actors along with the different forms of knowledge with which they are associated.In our context, the "Othering" of Amazighs has been rendered visible by the careful emphasis on their ethnic markers.
The Arab component, on the other hand, is very hard, if not impossible, to spot; this ethnic component appears to be quite invisible.Arabs are represented as "unmarked", "ordinary", and "mundane" and ergo remain unexamined, taken for granted as normal.Unmarkedness, at this juncture, seems to identify with invisible dominance and soft power.Using Scott's (2018) analogy, Arabs are (re)presented in terms of absences, emptiness, invisibility, nothing, and silence.Sociological unmarkedness establishes hegemony through silence, which helps dominant social groups pass under the sociological and the CDA radar.Similarly, Janks (1997) argues that "ideology is at its most powerful when it is invisible, when discourses have been naturalized and become part of our everyday common sense" (p.341).There is power in the iteration of silence, echoing the Simon and Garfunkel classic, "The Sound of Silence," which criticizes silence that maintains domination.Finally, it seems that the conclusions reached thus far consolidate the findings of various studies which share the same critical concern (Chu, 2015;Naseem et al., 2016;Slimi, 2009;Zhao, 2014).Similarly, Naseem et al. (2016) reach the conclusion that textbooks, as ideological state apparatus, 'become a key site for normalizing the authority of one group while also normalizing the rest as the "other".This "other" need not only be a person, a group, or a rival nation.Whole knowledge systems and ways of knowing, thinking, doing, and being are "othered" (p.9).
Chronologically, during the Critical Phase, Amazighs were almost excluded from the official EFL discourse both linguistically and visually.This can be explained in light of the dominant postindependence narratives, which tend to include images/texts of scientists (often Arabs), national heroes, figures, dates and events, whose task was to enhance the national unity and strengthen the national pride.Striking, however, is the fact that the Critical Phase was the period during which Morocco witnessed a heated debate on the question, place, and role of Amazighs in the overall socio-political architecture.Instead of responding, textbooks chose to categorically "avoid" Amazighs, both linguistically and visually.This multimodal exclusion was compensated for by an excessive presence of "themes" and thematic structures which, compared to textbooks from the other phases, were predominantly about "minorities".The Critical Phase seems to employ what Wodak et al. (2009) refer to as the topos of "avoidance" and "evasiveness" which translate into a laissez-faire attitude favoring in a way a deliberate non-interference as long as it "happens" "there" therefore "we are not concerned".
The Take-off Phase has been characterized by a marked increase in visual and linguistic inclusion, laying the cornerstone for a multimodal revolution, with Amazighs being visually included 51 times and thus accounting for half of the total visual inclusion.Visually as well as linguistically, it seems that this phase represents a special socio-political moment where the Amazighs' "elite" were given the opportunity to represent themselves instead of being represented.The textbooks that belong to this phase could best be described as a pedagogical outcry or voices from below.

Amazighs: a nation of singers, dancers, and shopkeepers
The linguistic representation reveals three significant patterns of inclusion for Amazighs in terms of functionalization.First, chronologically, a noticeable quantitative growth in the number of the occupational choices with which Amazighs have been identified can be clearly located.Second, considering the quality of the representational choices, the findings also show a qualitative clear upward-trend in the "quality" of the activities or the occupations in which Amazighs have been depicted.For example, when contrasted with the Germinal Phase or the Critical Phase where Amazighs were predominantly farmers, singers or dancers, the newly emerging discourse (Takeoff Phase) displays them as businessman/women, journalists, students, cooperative managers, doctors, political activists, and football players.Third, it seems that some activities or occupations (singers, dancers, farmers, carpet weavers) defy time and persist throughout the three phases, and thus become defining properties of the Amazigh identity.
Remarkably, it seems that there is a considerable difference between referring to Amazighs and Arabs in terms of functionalization where the relationship is strikingly asymmetrical.For example, during the Germinal Phase, Arabs were predominantly functionalized but in activities and occupations which are related to knowledge productions (scholars, teachers, scientists).In other words, Arabs were not represented in terms of "doing" but in terms of "being" (e.g., scientists and scholars).Amazighs, on the other hand, were predominantly portrayed in terms of "doings".

Amazighs: nominated yet genericized
When it comes to nominating Amazighs, the Germinal and the Take-off Phases display two contradictory patterns of nomination.While the EFL textbooks of the Germinal Phase predominantly represented Amazighs formally (53.12%),EFL textbooks from the Take-off Phase opted for the opposite, with Amazighs being predominantly referred to informally (48.93%).The two modes of nomination betray different ways of perceiving Amazighs.Formal modes of nomination are likely to breed, apart from formality, unfamiliarity, which may ultimately result in "Othering" Amazigh social actors.Opting for informal ways of address, on the other hand, places Amazighs face-to-face with the readers, wiping out any traces of unfamiliarity, minimizing social distance and depicting them as people with whom readers are familiar.
Overall, Amazighs have been genericized as being Berbers, Amazighs, Atlas dwellers, shopkeepers, and naturally born singers and dancers.They are less often specified or individualized.Once genericized, they are discursively distanced, removed from the reader's immediate context and cast in general terms.This generic reference to Amazigh social actors, according to Bourdieu (1986, as cited in Van Leeuwen, 2008, p. 35), is related to the concept of reality perception, where the working-class access to reality is mediated through the "generic".Amazighs have been represented as unspecified, "anonymous" individuals or groups.In other words, genericization anonymizes and devoices them, treating their identity as collective and eradicating individual differences.According to Van Leeuwen (2008), genericization blurs access to specific reality, and "once genericized, they are symbolically removed from the readers" world of immediate experience, treated as distant "others" rather than as people with whom "we" have to deal in our everyday lives ' (p. 36).Genericization also entails collectivization, where the latter may serve a political end.Chu (2015) observes that "when referred to collectively, ethnic minorities in textbooks are almost always used to support the idea, both historically and contemporarily, that China is a unified multiethnic country" (p.85).

Amazigh: a fixed timeless Essence
Identification renders Amazighs fixed, temporal and, therefore, ready to be recognized everywhere and every time.Excessive classification, for example, leads to the creation of a specific Amazigh profile.Amazighs were classified either as "children" or "adults/ grown-ups".Representing them as "youth/ teenagers" was very rare if not absent (Al-Masar may be the only exception).The pictorial analysis strengthens this observation.Classified as children, they have been often included either in the discourse of "dropping out of school" or leading a difficult life in the countryside or the villages, herding goats, or riding donkeys in search of water.As grown-ups, however, Amazighs are either portrayed dancing and singing, weaving carpets, fighting illiteracy, or working in cooperatives.
Impersonalization has been instrumental in exposing "hidden" aspects of the Amazigh portrayal in Moroccan EFL textbooks.In this regard, Bortoluzzi (2010) highlights the discursive implications of impersonalization and its power: "a) to build prior knowledge about social roles and identities of the participants, b) to grant authority or power to an activity or quality of the participant, and c) to add positive and/or negative connotations to an activity or utterance made by the participant" (p.529).

Amazighs and the world: a temporal alliance and a romanticized space
The examined textbooks tend create a discursive association between Amazighs and other social groups such as Phoenicians, Arabs, Jews, and Europeans.In this context, Van Leeuwen (2008) notes that "the group is represented as an alliance which exists only in relation to a specific activity or set of activities" (p.39).This alliance, in the examples above, serves to legitimize the existence of an "ancient" Moroccan nation (example 8), to justify multiculturalism (example 9), and to defend music plurality (example 10).In other words, the association of Amazighs with other groups helps create a cohesive national narrative at the political, historical, and cultural level.Association here helps create a solid national narrative based on a shared political destiny, religious unity, and cultural diversity.
However, it should be noted that the constructed alliance has only a "temporary benefit" (Van Leeuwen, 2008, pp. 38-39).Once the benefit is achieved, dissociation takes place.In many texts, associations are formed and unformed ("dissociation") as the texts unfold, and it is of critical interest to us to investigate the different discourses which display dissociation.The episode on marriage rituals in Imilchil festival centers on Amazighs' unique, yet outlandish, marriage traditions.However, when it comes to "hospitality", dissociation takes place, rendering this value a purely Moroccan trait available for all and to all.
In addition, it seems that tribalism and rurality accompany the representation of Amazighs under both folklorized and non-folklorized discourses.Three critical insights are worth discussing at this juncture.First, the excessive ideological construction of Amazighs as essentially rural leads to their portrayal as "virtuous children of nature" (Anderson, 1991, as cited in Slimi, 2009, p. 86), who are continuously set in a background of a natural landscape containing hills, plants, and grass.This formulaic depiction not only tokenizes them but removes the complexities of the daily lives, depicting them as "unspoiled ethnic Other" (Varutti, 2014, p. 183).Second, being essentially rural, a sense of "nostalgia" might be built in the way Amazighs are portrayed as "reservoirs of stillextant authenticity" (Schein, 1994, p.72, as cited in Zhao & Postiglione, 2010), or a counterweight to modernity.Amazigh social actors are often portrayed as having a "traditional" life in an environment that is close to nature; thus, lending more authenticity and exoticism to the representation.Slimi (2009) points out that rural depiction of ethnic groups helps draw a clear line between them and "mainstream" society, setting them apart "from the urban metropolises, from realms of education, employment and other activities of civility" (p.87).Third, it seems that the discourse not only depicts Amazighs social actors as "virtuous children of nature", but also specifies the types of activities which are supposed to strengthen this representation.The most common occupation associated with Amazighs in this discourse is that of animal husbandry, depicting them as shepherded both linguistically and visually.

Conclusion
In this article, I attempted to analyze the place of the Amazigh ethnic groups in Moroccan EFL textbooks since 1980s.I have demonstrated that Amazighs have not been included as much as "mentioned".Mentioning, here, can be taken as the lowest level of the integration of ethnic content in school textbooks.The analysis has also revealed the centrality of specific sociosemantic categories for the representation of Amazighs.These categories have been instrumental in revealing the extent to which Amazighs have been categorized, genericized and collectivized, which ultimately, lead to the backgrounding of Amazighs and to the suppression of their individuality.Association and dissociation have also undermined the manner in which Amazighs have been included in a "temporary" alliance with other social groups.The analysis has shown the fragility of such alliance which, soon, gives way to dissociation.Related to this are "functionalization" and "identification", which, on the whole, have revealed intriguing patterns of representation which have tended to limit Amazighs in specific social activities that define them in terms of what they do or in terms of who they, permanently or unavoidably, are.Finally, referring to Amazighs in terms of impersonalization, such as spatialization and abstraction, leads to the objectification of Amazigh social actors through the fixation of their way of life.The examined textbooks include nothing about Amazighs contributions to modern Morocco.I do believe that the conclusions I have reached thus far can help us set the following two overarching recommendations.First, in Morocco, the national curriculum is centrally administered by the Ministry of Education which is accountable for the design and development of school textbooks.Consequently, the Moroccan Ministry of Education should play a dynamic role in the enhancement of ethnic content in school textbooks.It should also establish new guidelines to evaluate the quality of textbooks in terms of ethnic equity.In addition, the Ministry needs to consider the improvement of instructional materials to meet the needs of the diverse Moroccan population.Second, it is incumbent on the Ministry of Education to provide training programs for textbook designers, ELT supervisors and teachers, in addition to other departments under the Ministry of Education, for instance, the Curricula Directorate.The training programs should be "engineered" (Arrif, 2017) in such a way as to help the various stakeholders and officials gain awareness of ethnic content in school textbooks.This will help enhance their awareness of the latent harm in ethnic biased textbooks.
There are also implications for textbook designers.It is conventional wisdom that textbook designers are central to the learning process.Most teachers often rely on the textbook content and material to deliver their teaching.Thus, Moroccan textbook designers need to (1) recognise the changing role of Moroccan women and Amazighs and their contributions to society at various levels (beyond the economic and the folklorized perspective), (2) be aware of the research on ethnic and gender portrayals in textbooks as well as the research on the effects of gender and ethnic-bias on learners and utilize that knowledge to publish textbooks that provide positive images for women and ethnic groups, and ( 3) be open to critical insights from national and international research on the integration of ethnic content in EFL textbooks This research study was limited in different ways.These limitations emanate from two interrelated sources: data and methodology.First, this study was limited to officially produced EFL textbooks in Morocco where the primary focus has been on students' textbooks.The examined instructional materials have been limited to those approved and circulated by the Moroccan ministry of education.There are, however, other commercially produced textbooks that have a wide appeal in the private sector.Allocating due attention to these sources has been beyond the scope of the present study.Second, one of the methodological limitations of the current study is its scanty account of the historical dimension of the representation of Amazigh social actors across the three phases.It is recommended that further research be undertaken in this area.

Appendix: Analyzed Moroccan EFL textbooks used in the study
Figure 1.The chronological inclusion of Amazigh social actors.

Example 4 :Example 5 :Example 6 :Example 7 :
A lot of beautiful girls in colourful dress with enormous traditional ornaments (EIL1, p108) Khettabi proved a most effective coordinator of the Maghreb office (EIL3, p. 46) It was just four months since the heroic Riff War had ended (EIL3, p. 45) What beautiful carpets!p. 91) Author(s).This open access article is distributed under a Creative Commons Attribution (CC-BY) 4.0 license.This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http:// creativecommons.org/licenses/by/4.0/),which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.The terms on which this article has been published allow the posting of the Accepted Manuscript in a repository by the author(s) or with their consent.and dialog with, expert editors and editorial boards • Retention of full copyright of your article • Guaranteed legacy preservation of your article • Discounts and waivers for authors in developing regions Submit your manuscript to a Cogent OA journal at www.CogentOA.com

Table 4 . The Functionalisation of Amazighs Phases Functions
Amazighs were not only categorized in terms of what they do (functionalization), but in terms of what they permanently, or unavoidably are; that is, in terms of identification.Two types of identification were examined: classification and relational identification.Under classification, Amazighs have been represented in terms of and in relation to age, gender, class/social status, and ethnicity.The carried-out analysis reveals the following patterns of classification: