Al-mihna wa’l-huwiyya. Motivational Dimensions of Arabic Heritage Learners in a Professionally-Oriented Education Setting

This study investigates motivational and professional orientations and identity issues of students of Arab descent majoring in Arabic in a teaching setting oriented to develop vocational and professional language skills. Their motivations for learning the target language were investigated and compared with those of their colleagues of non-Arab descent through a questionnaire that explored motivational factors, job expectations and education history. Apparently, students were prompted to take Arabic classes because of employment concerns and they proved keen on undertaking most of the proposed professions, providing that they were connected with the Arab world. It is also possible to highlight how the study of the language of origin was determined by thoroughly intrinsic motives. Cultural interests and especially identity needs provided the strongest push for the respondents, even in the surveyed education setting, where instrumental motivation was clearly expected to play a major role. Implications can be drawn as regards curriculum design and classroom practice.


Introduction
Patterns of immigration and globalisation, as well as social and political developments, have long been recognised to have an impact on language learning trends (Van Deusen-Scholl 2003;Temples 2010). This was also the case for Italy, that, since the 1990s, has started to shift from being a country with a low rate of foreign immigrants to becoming a society exposed to increasing migration flows. One of the most representative cities in this regard is possibly Milan, with its vast outskirts: being considered the most relevant area for business and industry in Italy, it has attracted a large number of immigrants that officially accounted for 18.8% of the populace at the end of 2016. Among them, the subjects coming from Arab countries were quite high in number: Egypt, for example, was the second most represented foreign country of origin (Comuni italiani 2019). As a consequence, over the past two decades, first or second generation immigrant students have started to attend Arabic courses at the highest education level, so both the composition of the classes and the reasons for studying the language have undergone relative change.
Recognition of the heritage learners' (HLs) motivational inputs is fundamental in order to help devise language teaching policies consistent with their actual needs and requirements. Existing literature has often scrutinised motivations of students of Arabic in the US. It has more rarely examined the Arabic heritage learners' professional orientations, the distribution of their motivation, and the way this is peculiar to them rather than other groups of learners, especially in contexts different from North America. This study surveys a group of 40 heritage learners studying Arabic at a major university in Milan in a professionally-oriented education setting. It is aimed at casting light on their motivational and professional dimensions and the way they potentially differentiate themselves from non-heritage learners (NHLs) of Arabic. The role identity plays in their academic choice is also attentively considered.

Background
The concept of heritage learners has been debated especially as regards the subject's ethnic, cultural or linguistic identity and the minimum language level he/she should possess to be considered a heritage learner. A widespread opinion is that a heritage learner is a subject who perceives him/herself having a cultural connection to a specific language (Van Deusen-Scholl 2003), despite his/her actual fluency, or sometimes even despite the absence of any fluency. More specifically, the debate has also attained the concept of heritage learners of Arabic. Contrasting positions are not completely im-mune from ideology (Leeman 2015). Temples's (2010) opinion in this regard stands quite in isolation. Her survey compares learners of Arabic of different types: language heritage learners (HLLs), religious heritage learners (RHLs) and NHLs) aged from 10 to 14 at a public American school. According to the researcher's findings, although the religious values are a powerful motivator for the parents of both the subgroups of language and religious HLs, the religious HLs' proficiency does not resemble that of language HLs: In the RHL group, 90% had some proficiency in Arabic prior to studying it at this public school, but nevertheless they do not have exposure to communicative Arabic at home, as do traditional HLLs. (Temples 2010, 115) Her conclusion is that the three groups of students should have three different kinds of tracks according to their different levels of proficiency, prior language exposure and motivations.
Arabic and Muslim heritage learners are more frequently hardly disjunct in literature, thus following the general definition given above. It is Husseinali's (2006) opinion, for example, that all religious HLs of Arabic (i.e. Muslim students), due to their identity, can safely fall into one and the same group, regardless of their Arab or non-Arab descent. In other words, from his perspective, all heritage students of Islamic religion and those of Arab descent are comparable -at least from a motivational point of view -provided they display convergent "cultural and historical ties to Arabic" (Husseinali 2006, 396) and independently of their different initial level of Arabic proficiency. Engman's (2015) conclusions proceed in the same direction. In line with the sensibility and positions also expressed in other previous studies (e.g. Sehlaoui 2008;Shah 2006;Suleiman 2003), Engman analyzes the role of religion in Arabic language learning for young Muslim learners in the United States in their enactment, resistance, and co-construction of myriad identities for themselves […]. Their language (both content and form) suggests that exposure to Islamic religious discourses permits access to salient cultural resources for Muslim students as language learners. (2015,223) Although this refers to the concept of identity more than to wide language achievements, some learning objectives are also concerned, such as linking the language to (the promotion of expected Islamic) behaviours and the use in class of reading skills and strategies learned in an external religious context. The researcher's explicit conclusion is that the benefits of an ideologically inclusive classroom must not be ignored and educators should be free to resist the total separation of Arabic from Islam. Even in mixed classes, at public institutions, I advocate not only 'allowing' heritage language learners to draw on their religious-cultural resources, but also encouraging them to do so. The ideological space for this sort of expanded repertoire already exists within the language itself, but teachers need dispensation to create spaces for these heritage learners in the classroom. (Engman 2015, 237) Other studies face motivation, identity or peculiarities of the Arabic HLs from different points of view and according to various methodologies. Bale (2010) offers a wide picture of heritage Arabic in an objective social and educational perspective drawing on archival research and statistics analysis. He sheds light on the history of the Arab immigration to the US, the socio-economic composition of the local Arab communities, and the language effort dispended in order to teach and maintain this heritage language alive within these communities, in the form of schools, courses and programs that are supported by private or public funding. Provided this great variety of conditions, he gets to the conclusion that "there can not be a single approach to Arabic heritage language instruction" and stresses the importance of designing programs that "are sensitive to the immense amount of linguistic diversity within the community" (Bale 2010, 148). Contrarily to Bale, Seymour-Jorn (2004) bases her survey on questionnaires or interviews and focuses on a specific local community, that of the Arab-American students taking courses of Arabic at the University of Wisconsin in Milwaukee. The researcher enquires about their reasons for studying Arabic and the way this relates to their sense of cultural identity, belonging to the community, and need to stay in contact with the enlarged family of origin.
Another more 'personal' contribution to the subject matter is Sehlaoui's (2008), that relies on the author's own experience as a heritage learner, father and teacher. His article is mainly aimed at offering advice for (parents and) teachers at lower education levels. It highlights the importance of native language literacy also for L2 acquisition and the value of the affective dimension in heritage language teaching starting from home. The role played by families in children's acquisition of Arabic was especially investigated in the form of narrative research (Gomaa 2011;Al-Sahafi 2015;Turjoman 2013Turjoman , 2017. Exposing kids to language in a natural way, setting explicit family language policy, establishing co-ethnic contacts, and providing materials to enhance Arabic literacy are among the most relevant actions taken in this context. Parents' language ideologies and ethno-linguistic aspirations have an important impact on language policies of families of Arabic descent, in the per- Marco Aurelio Golfetto Al-mihna wa'l-huwiyya: Motivational Dimensions of Arabic Heritage Learners 159 spective of both developing bilingualism and resist cultural colonisation (Yazan, Ali 2018). In general, parents do hold expectations for their children's acquisition of the heritage language since Arabic is perceived as an important identity marker: it operates as a base for networking and provides an aspect of communality beyond the different countries of origin (Alsahafi 2018).
The HLs' specificity has also been focused more recently in Zarabah (2016). The author enquires into the relevant issue of whether HLs should better stay in separate classrooms or share the same classes with NHLs. In doing so, she examines the expectations, language exposure and proficiency levels of a group of incoming college-level heritage students and concludes that HLs' needs should be given attention in order to reduce the attrition rates. Finally, Kamel Hasan (2018) also participates in the debate by examining aspects of the HLs' learning experience at American University in Cairo (AUC) in order to understand what could help sustain and stimulate their motivation.

Participants
The total participants to this survey consisted of 205 students who were majoring in Arabic in late 2018 and included both heritage and non-heritage learners. All the students were enrolled in BA or MA degree courses in Language and Cultural Mediation at the Università degli Studi of Milan. Besides Arabic, the participants were learning another language, either a European or non-European one, as a requirement and in parallel with the former throughout their entire career. They also were studying the cultures related to the two target languages and a number of other subjects relevant for developing professional skills, in fields of geography, economics, law, sociology, history, professional translation, and so forth. Out of the overall surveyed population, the HLs were 19.5% (40). The remaining 80.5% (165) were NHLs of different provenance and mostly Italians. Heritage learners are meant here as language HLs, i.e. respondents who chose the option "Arabic" or "both Arabic and Italian" as a language/s spoken in their family (Q41 1 ). The remaining participants (i.e. those who responded "Italian" or "another language") were counted as NHLs. More in detail, the HLs who revealed that they only spoke Arabic in their family were 45% (18), whereas those who declared "both Arabic and Italian" as family languages were 55% (22). Only rarely (3 cases) did students point out that they spoke an "Arabic dialect" (or another variant) instead of "Arabic". Students who spoke Arabic in their family did not indicate any other language other than Italian as a second family language. As concerns the students' birthplace (Q42), 50% of the subjects (20) were born in an Arab country: 8 in Morocco, 7 in Egypt, 2 in Tunisia and Kuwait, 1 in Lebanon. One more student was born in Chad. The other HLs were all born in Italy.
As for the participants' gender (Q38), 85% (34) were female and 15% (6) were male. 33 participants were enrolled at BA level, and 7 at MA level (none of the respondents was a 1 st -year MA HL). The ages of the subjects ranged from 19 to 44 (Q39). As can be seen in [table 1], at BA level the medians of the HLs' age were higher than those of NHLs for all years.

Instruments
The materials that were used in this survey consisted in a questionnaire worded in Italian and composed of three sections for a total of 51 items. In the first part of the questionnaire, 19 items related to the reasons for studying Arabic were to be answered based on a 7-point Likert scale, in which 1 expressed total disagreement and 7 expressed total agreement. The battery of items was selected based on literature on language learning and psychology studies on motivation (see in particular Husseinali 2006; see also Dörnyei, Taguchi 2010;Noels et al. 2000). The batteries of items that had been used in previous questionnaires were made less redundant and modified in order to be applicable to students who were majoring in Arabic. The following scales and subscales were included: extrinsic (7 items in all, divided into circumstantial and instrumental), intrinsic (7 items, related to the stimulation of the language and perception of self-efficacy, desire of knowledge and understanding), and integrative (5 items,

161
pertaining to relation, affiliation or identity). All items were preceded by the opening phrase "I am presently studying Arabic because…".
Another section of the questionnaire investigated the learners' professional orientations. The 15 options proposed were selected according to previous informal interviews with students and stakeholders. All items were preceded by the opening phrase "As a student of Arabic, I would choose to work…", followed by a professional field or specific job. The items were given in random order but they can be grouped in five broader sectors. Answers were graded according to the same scale as the previous section. The Crombach's α test confirmed that both section 1 and 2 of the questionnaire were acceptable (0.767 and 0.749 respectively).
Section 3 elicited information on personal background such as gender, age, class, language/s spoken in the family, country of birth. Another array of questions in this section enquired about the participants' learning experience: attended high school, curricular study of other foreign languages (FLs), second FL studied at university, attained FL certifications, (estimated or objective) achieved levels of proficiency, and so forth, in order to sketch out the HLs' history of language learning. This was studied separately (Golfetto 2020).

Procedure
Students participating in this survey answered the questionnaire at the beginning of the academic year 2018-19. The widest audience possible was reached through repeated sessions during regular Arabic classes. Students were asked to indicate their immediate reactions and be as truthful as possible. They were also informed of the voluntary basis of their participation in the survey, the confidential nature of their answers and their right to refrain from answering at any time should they find any item objectionable.
After the answers were encoded, descriptive statistics were used to describe the group of HLs that were a part of the largest population of university students taking Arabic. As for inferential statistics, Mann Whitney tests were used to evaluate the difference in the results between HLs and NHLs, while Fisher's exact tests were used to investigate the relation between the heritage students and categorical variables. In this regard, non-parametrical tests seemed more appropriate than the parametrical ones (such as T test).

Professional Orientations
As the main reason for students to enroll in degree courses in Mediation is to pursue appropriate employment and to develop specific language skills in order to be able to face future professional challenges, 2 the first part of this study is dedicated to detecting the HLs' most desired professions and highlighting in which way these were peculiar to their own rather than to their non-heritage peers. This will be helpful in two ways: having initial insights into the learners expectations about the target language and highlighting future directions for developing policies in the teaching of Arabic for Specific Purposes. Professional motivation of the heritage students (especially those of Arabic) is a topic that has hardly received attention in literature so far. As previous studies are defective, the items of the present section (whether types of jobs or professional fields) were pre-selected according to interviews with different stakeholders. They can be broadly grouped into the following sectors: mediation at international and local level, business, tourism, translation, and others. However, professional orientations must be considered as independent from each other: each career requires a different and specific variety of Arabic; knowledge of specialised vocabulary, language functions, and textual types; higher proficiency in oral/written or receptive/productive skills, and so forth.
Within the group of jobs related to mediation, all the professions at either international or local level were highly rated by the surveyed subjects. Orientation towards working in the field of information for the press, television, or the Internet (Q22) was good at 4.78 out of the maximum score of 7.0. Work in the field of international politics and diplomacy (Q23) was even more appealing, at 4.9. However, HLs showed the greatest interest in professions related to migration and humanitarian support. Being employed in "national/international institutions or organisations that operate in Arab countries" (Q26), indeed, scored very high at 5.84. The standard deviation (SD) of this item was also the lowest (1.38) of the entire set, meaning that there was very little discrepancy in the respondents' enthusiasm in this regard.
At the local level, mediation "for State institutions (municipalities, settlement visa offices, and so forth)" (Q28) was the most appealing profession, with the highest rate (5.87). It is noteworthy that this kind of institutions is a regular venue for migrants throughout their stay in the country; thus, working in this context is perceived as some- thing familiar and being useful for their community of origin. Employment "as a cultural mediator in schools, refugee centers, and charities" (Q29) in the local context scored only a little lower (5.3). Finally, working in the field of healthcare mediation in hospitals (Q27) was rated at 4.95. All these are greatly needed jobs that have particularly flourished over recent years with the so-called 'migrant crisis' that has involved Italy (and Greece) since 2013, but are now suffering a severe setback due to the current changes in government policies and to closed borders. Nevertheless, awareness of restriction in related funding does not seem to have an excessively negative impact in the rating of this specific professional sector.
A position in commerce was also desired by the respondents. In fact, a job "in companies trading with Arab countries (as a sales man/ woman or representative abroad)" (Q30) was rated at 5.63. The result of the item related to "the promotion of halāl goods or goods for families of Arab origin living in Italy" (Q31) was a bit lower (4.34). Being something that affects the daily life of many individuals within local Arab and Islamic communities, one might expect the score to be higher. However, it could have been hindered because the sector of halāl certification, although on the rise, has not been thoroughly developed in the Italian food industry yet.
HLs were also highly motivated towards the two professions related to tourism. Working as a "tour guide for visitors from and to Arab countries" (Q25) received a rather good score (4.95). This is clearly connected with high appreciation for both travelling to Middle-East and North Africa (MENA) (Q7) and meeting new and different people (Q14) in the next session. A position in the sector of "promotion of tourism" (Q24) was also well evaluated, with a rate of 4.9.
Among the remaining professions, there was a good general agreement rate for working "as an interpreter, translator, or for a publishing house" (Q32), as the mean recorded in this case was 5.42, with a relatively low SD at 1.6. Other options were less interesting for HLs, and were evaluated below the medium. A role as financial consultant or similar (Q35) scored 3.29, but opinions were divergent and the distribution of the responses for this item was the widest of the section, with a SD of 2.29. That is, in fact, the highest SD of the entire questionnaire. A job "in the armed forces or security" (Q34) was very little agreed upon (2.95), although the responses were quite heterogeneous again, with a relevant SD of 2.2. Finally, it was a position "in the field of energy and oil" (Q33) that very strikingly revealed to be the least appealing job among all (2.86) and was better appraised by the NHLs than the HLs. See [  When it comes to statistically significant differences among the groups, according to the Mann Whitney test, these were found in 9 items out of 14. This means that HLs' and NHLs' perspectives and expectations on employment are potentially quite divergent. In particular, the "promotion of halāl goods" was the job for which the fork between the two groups was the widest (4.34 vs 2.68), and the difference was, at the same time, the most statistically significant (p <.0001). Similarly, the related "trading with Arab countries" also presented a rather relevant divergence (5.63 vs 4.53), with a significant statistic difference between the two groups (p .0006). Another relevant divergence concerned working in the "promotion of tourism", with HLs being clearly more motivated to find employment in this field (4.9 vs 3.57) and a statistically significant difference of p .0001. The same difference was found for the other related profession, i.e. that as a tour guide, for which the variation between the two groups (4.95 vs 4.01) was statistically significant again (p .0045). See [ fig. 1

Motivational Orientations
This other session of the study examines a broader set of motives at the base of the participants' decision to major in Arabic at university. Studies on motivation and orientations -mainly developing from the field of psychology -have put forward different theoretical constructs, from which different scales of motivation were advanced. The present survey focused on three main scales: extrinsic, intrinsic and integrative. A total of 19 items were selected to cover the three scales.

Extrinsic Motivation
The first group of items pertained to extrinsic motivation. Within it, the first subscale was related to circumstantial motivation. Two items were proposed: "I had to study a (second) FL as a degree requirement and had no choice [other than Arabic]" (Q1) and "Arabic is easy for me, as I am already familiar with it" (Q3). The two options had completely different agreement rates: 1.58 out of 7 for the former one, and 5.55 for the latter. In addition to this, the first item also recorded the lowest SD (1.44) of the section. Five other items of this set were connected to instrumental motivation. The answers to this subscale ranged high (from 4.98 to 5.53). The participants demonstrated being very motivated to study Arabic in order to benefit from both more job opportunities (Q5, score 5.5) and higher wages (Q6, score 4.98; in this case the SD was very low, at 1.49). When it came to items pertaining to the specific fields in which they reputed Arabic to be most useful, HLs stated that they decided to learn Arabic because they thought it was relevant in the fields of international politics (Q10, score 5.53) and world economy (Q11, score 4.97). The only item that scored lower was related to migration and terrorism (Q9, rated at 3.83). However, upon comparing this result with those of other items (especially Q27, 28, 29), it becomes clear that this result was due to the way the statement was phrased: "[so that] I can be considered an expert in migration or terrorism" (Q9). In fact, HLs did not deny being interested in working in the field of migration; rather, they were less interested in participating in theoretical debates on this subject.

Intrinsic Motivation
The 7 items pertaining to the intrinsic motivation scale were also highly rated. In this case the range was wider, between 4.51 and 5.8. One of the most common justifications for studying Arabic resulted being "I think this language is fascinating, exotic and less banal than others" (Q2), which scored as high as 5.73. It is clear that Arabic exerts a strong attraction within Arabic-speaking communities (Bale 2010) and, upon rating this reason, heritage students clearly put more stress on their fascination for the language rather than its exoticism. Only slightly less convincing, with a good rate of 4.8, was "because I like learning a challenging language and being effective" (Q4). As for the motives related to the second subscale, i.e. desire of knowledge, the respondents were quite convinced of studying Arabic because they thought it was useful "to understand our time's events" (Q12, score 5.43). Awareness that Arabic "is useful to understand other people's perspective of the world" (Q13) had an even higher impact on their choice (rated at 5.8). Such a high rating might appear sur-

167
prising, especially because the item was mainly phrased for NHLs, inasmuch the word 'other people' implies separateness between the respondents and Arabic-speaking people.
Three other items explored the desire for knowledge. Among them, it was interest in news, newspapers and other sources of information (Q17) that scored the highest (5.38). Literature (Q15), philosophy and religion (Q16) were just slightly less motivating factors for the HLs to major in Arabic (4.8 and 4.51 respectively). When comparing the two groups of respondents, HLs were also more motivated to learn about Arabic literature and (Islamic) philosophy/religion than NHLs, and this difference was statistically significant (p .016 and .001 respectively).

Integrative Motivation
The third and last scale was that of integrative motivation and explored the respondents' willingness to learn about the target culture and immerse into the related human environment. The item that rated highest in this context was connected to religion and culture: "It is useful in order to know my family cultural or religious background" (Q19). The item was clearly aimed at learners of Arabic descent, and was somehow opposed, or complementary, to Q13, for, if the latter referred to the idea of 'otherness', the former activated the engine of self-identification. The item scored as high as 6.28, recording a moderate SD (1.53) and a statistically very significant difference compared to the NHL subgroup (p <.0001). More in general, all the statements of the scale of integrative motivation were highly evaluated (above 5 out of 7). HLs were very open to interaction with both their geographical area and human environment of origin, although they were not oriented to leave the country in which they presently lived. In particular, respondents agreed on their appreciation for "travelling to North Africa and the Middle East" (Q7), with a rate of 5.08. Also in this case, there was a statistically significant difference with the other group (p. 0059), although the values were not very discordant in absolute terms (5.08 vs 4.45). HLs were also very positive towards "talking to Arabic-speaking friends or relatives" (Q18) and "meeting and confronting new and different people" (Q14), which both scored 5.83. The statistic difference from the NHL group was significant in this case (p .006 and .0459 respectively). The only motive that did not spark outstanding enthusiasm was "[because] I want to settle in an Arabic country" (Q8), which scored a lower rate, at 4.18. The responses to this item were quite divergent anyway, as the SD was the highest for this set of items (2.27). The country of origin played a relevant role in this regard. In general terms, however, the HLs proved to be more open towards this op-     (Q5,6,9,10,11) 4.96 4.97 .9361 Extrinsic motivation (total;Q1,3,5,6,9,10,11) 4.56 3.91 <.001* Intrinsic motivation (Q2,4,12,13,15,16,17) 5.18 5.12 .3282 Integrative motivation (Q7,8,14,18,19) 5.44 4.1 <.001* In the end, the respondents were also provided with a list of items enquiring after the reasons for choosing a BA or MA degree in Language and Cultural Mediation rather than other degrees offering Arabic as a FL (Q51). They were now requested to select the three strongest motives that led them to make their decision and discard the others. Once again, the distribution of the HLs' motives was quite broad but the responses were consistent with those collected in the previous session. The relevancy of professional aims was confirmed by: 1. the interest in studying specific subjects such as law, economics, and so forth (selected by 50% of the students) and 2. the type of academic curriculum explicitly oriented to professions (32.5%). The opportunity to learn about the heritage culture was the second incentive in absolute terms for majoring in Arabic, and was selected by 47.5% of the respondents. Studying specific subjects connected with the traditional Arabic heritage, such as linguistics and literature, were selected less frequently, but in both cases they resulted stronger motivations for HLs than for NHLs (35% vs 20%). Finally, the chance of simultaneously studying two languages (one European and the other non-European) was the least motivating factor (23% vs 31% for the NHLs). See [ fig. 3].

Discussion
In the surveyed education setting, when choosing their degree courses, the HLs of Arabic gave priority to their professional goals, seeking acquisition of theoretical knowledge and language skills (especially varieties of Arabic for Specific Purposes necessary for the desired job). Their instrumental motivation appeared very strong, as most of the proposed professional options scored highly. Upon comparison, their motivational rate was even higher than that of the NHLs in all options, with the only exception of working in international politics and diplomacy. In short, HLs appeared to be very motivated to find a job, rather than a specific job, providing that this was linked in any way to the Arabic language. The students' decision to major in Arabic was partially connected to circumstantial motivations. Familiarity with Arabic (and thus higher expectations of easiness and 'comfort') admittedly played a role in the choice. However, given the means and SDs recorded, it also appears that they were not forced into their language decision by the circumstances or by their relatives (as found in other research), but rather led to it by a conscious, real and personal involvement. When it comes to the intrinsic motivational scale, one can safely assume that the HLs' motivation was as strong as that of the NHLs in quantitative terms: with p value .6 there was no statistically signifi- cant difference between the two groups. Nevertheless, this type of motivation was different in qualitative terms: Arabic was actually not studied because it was perceived as a learning challenge, as might be the case for NHLs, but rather because of the feeling of effectiveness that stems from studying an already familiar language within a formal education context. Consequences may be drawn in terms of different expectations about the task and workload to face in the study program: HLs usually feel they already know 'Arabic' and expect to easily succeed in this regard. Therefore, instructors should pay attention to use appropriate tools to prevent motivation loss, frustration, and attrition by focusing on their actual needs, rights and duties and organizing class activities accordingly.
Besides the desire for achieving professional competence for a future career, the questionnaire results have highlighted the relevancy of cultural and identity issues at the base of the HLs' motivation. Three figures prove it: the HLs rated highly most of the items connected to the integrative motivation scale; the statistic differences between the HLs and NHLs were ubiquitous; the single items of the integrative subscale scored the highest of the entire questionnaire in absolute terms. To put it simply, when identity was singled out, the heritage students' reaction appeared heartfelt. As in another, less specific situation (for example Seymour-Jorn 2004), the HLs were strongly oriented to study and use Arabic in order to foster communication with peer native speakers and family, learn about their perspectives on the world, and re-appropriate their own cultural and religious heritage. Provided the collected data, it can be safely assumed that, however disguised, the major set of motivating factors that led HLs to learn Arabic was connected to identity, even in the present context where one would expect professional education to play the most powerful motivating role, as it is in fact for NHLs.
Identity and affective factors should thus be distinctively focused in language planning and teaching for HLs. As a general rule, and especially under specific circumstances (such as conditions of social or educational disadvantage, see Engman 2015), the teaching personnel should wisely value and care for the HLs' status in class, keeping in mind the implications that studying Arabic has on their selves (Seymour-Jorn 2004). Due recognition should be given to the HLs' cultural identity, by presenting and discussing their personal cultural values and worldviews (Engman 2015). It was noticed that the heritage students' resistance and negative attitude towards the teacher can be rooted in the latter's failure to acknowledge their perceived 'true' identities (Helmer 2013). Furthermore, their language identity, as self-perceiving users of Arabic (Bale 2010), should also be given relevance, by allowing the native language variants to be used in class. Since identity implies recognition, the HLs should feel free to express their linguistic foreknowledge, rework it correctly and con- sciously, and evolve naturally towards proficiency in MSA. At an early stage, specific attention should be paid to HLs' affective dimension, which is frequently connected to their home and family's memories. As also Carreira puts it, pedagogical practices must focus on […] validating the learner's prerogative to define him/herself in terms of their language and culture of ancestry no matter how remote or insignificant the connection to this ancestry may seem to native speakers of the HL or to anyone else. (2004,8) Attention to such an affective dimension might mean initially giving HLs tasks that are aimed at collecting or recalling vocabulary related to their childhood, home environment and family history, also by presenting them in personal autobiographies or family/community tales. This seems to be appropriate especially in the initial part of the track for many reasons. Firstly, it immerses "the learner into a world that is personal in a very profound sense" (Sehlaoui 2008, 288). Secondly, this respects the principle of proximity in the initial phases of language learning in formal instruction, according to the CEFR guidelines (as for their specific application to the teaching of Arabic as a L2, see Giolfo, Salvaggio 2017;Soliman 2018). Thirdly, it enhances and fosters a communicative approach to language by promoting its use in natural communication and real dialogic contexts within the HLs' closest social network, which is made up of family members and friends. Finally, in classes where different local variants are represented, it facilitates a contrastive approach to language, in both a diaglossic and pluri-linguistic perspective. In doing so, positive repercussions will be perceivable not only in the earliest stages of formal learning, but also in the long term, helping students to get ready to face real-world challenges.

Conclusion
To conclude, in the enthusiasm displayed by the surveyed students for most of the professional options proposed in the questionnaire, two main directions can be identified. The first one is the international dimension. Heritage students of Arabic are naturally projected beyond any geographical borders and would be keen on being employed in international organisations and supranational institutions. They feel at ease in the role of bridging across different worlds, cultures and languages and consider that their privileged position can also be positively capitalised in specific sectors, such as business and tourism, provided that they imply connections with the Arab countries. Additionally -and this is the second direction -, they feel close-ly linked to their 'original' identity, something that implies loyalty to the group of provenance. Indeed, their very positive inclination touches also those professions that allow them to play a helpful role for their own community in the local area. Being useful in the mediation process between individuals and State institutions is somehow perceived as a call of duty to which they favourably respond. Consequently, this study confirms that a double motivation leads learners of Arabic descent to study Arabic. On the one hand, they admittedly have professional expectations, which are linked to the need of developing specific language varieties and skills. On the other hand, the strongest incentive to majoring in Arabic is connected to intrinsic factors, among which identity plays the most vital role. HLs undeniably perceive studying Arabic as a way to re-appropriate what they feel as their own heritage culture and an opportunity to foster their relations with the community of origin, real or perceived as it might be. Therefore, identity issues at the base of Arabic HLs' education choice must be held as relevant and recognition of their identity and foreknowledge is recommended in any teaching settings, with no exception for formal professionally-oriented instruction contexts.