Next Article in Journal
Maintaining the Indigenous Udmurt Language beyond the Community: An Autoethnographic Analysis
Next Article in Special Issue
Language Management in Transnational Multilingual Families: Generation 1.5 Parents in Finland
Previous Article in Journal
Introducing the Special Issue Terminology in the Digital World
Previous Article in Special Issue
English Immersion as Family Language Policy in Spanish Households: Agentive Practices and Emotional Investments
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Albanian as a Heritage Language in Italy: A Case Study on Code-Switching within DP

by
Gloria Cocchi
* and
Cristina Pierantozzi
Dipartimento di Scienze della Comunicazione, Studi Umanistici e Internazionali, Università degli Studi di Urbino “Carlo Bo”, 61029 Urbino, Italy
*
Author to whom correspondence should be addressed.
Languages 2024, 9(9), 285; https://doi.org/10.3390/languages9090285
Submission received: 31 May 2024 / Revised: 10 August 2024 / Accepted: 12 August 2024 / Published: 23 August 2024
(This article belongs to the Special Issue Language Policy and Practice in Multilingual Families)

Abstract

:
In this pilot work, we are going to discuss several aspects concerning the Albanian language spoken in the Italian territory by immigrants of different generations. After an excursus on heritage languages in general, and Albanian as a heritage language in particular, we present the results of both a sociolinguistic and a linguistic survey conducted among some Albanian immigrants in Italy. The former aims at investigating the contexts of use of Albanian and Italian languages, the participants’ competence in both of them and their attitude towards code-switching. The latter is focused on the participants’ judgments of the acceptability of different types of mixed Italian–Albanian DPs, i.e., DPs where D and N are expressed in different languages, and the theoretical implications that emerge.

1. Introduction1

Italy and Albania are two countries that have a common history: they are divided only by a narrow strip of sea of approx. 75 miles and have a solid relationship. The contact between Italian and Albanian languages is ancient and well established, on both sides of the Adriatic sea. In ancient times, seven migratory flows led to the settlement of Albanian populations on the Italian territory, between the 15th and the 18th centuries (Mastroberti 2008). In current times, after the massive migration of the 1990s, many Albanian speakers live in Italy and are well integrated into the Italian social fabric; as a consequence, they have become more and more accustomed to the Italian language, sometimes to the detriment of Albanian, especially in regard to second-generation immigrants.
Hence, the present research aims at investigating, on the one hand, the sociolinguistic situation of the Albanian language spoken in Italy, as well as its relationship with Italian and the emergence of code-switching within the bilingual community; on the other hand, we will concentrate on some linguistic phenomena characterizing code-switching contexts, and in particular the acceptability of mixed DPs in bilingual discourse.
The first part of this work is devoted to sociolinguistic analysis. In particular, in Section 2 we will discuss the concepts of a heritage language and heritage speakers according to the most common definitions we find in the literature. In Section 3, we will summarize the history of the contacts between Albanian and Italian languages, while in Section 4 we will review some studies on Albanian as a heritage language. In Section 5, we will present our sociolinguistic survey, a questionnaire administered to 27 Albanian–Italian speakers, whose results will be outlined in Section 6.
The second part of this work is instead dedicated to linguistic analysis. First of all, in Section 7 we will discuss the concept and typologies of code-switching, as well as the main theories on this phenomenon. In Section 8, we will present the syntax of Italian and Albanian DPs, so that, in Section 9, we can introduce the phenomenon of mixed DPs and the possible D-N combinations which may emerge. In Section 10, we will outline our linguistic survey, an Acceptability Judgment Task administered to eight bilingual speakers, whose results will be presented in Section 11. Finally, in Section 12 we will discuss the results obtained by the two surveys and the interesting interplay of sociolinguistic and linguistic factors that has emerged from our research.

2. An Overview of Heritage Languages

In the literature, we find different definitions of heritage languages (HLs), each of which focuses on different aspects of the acquisition process, as well as different definitions of heritage speakers (HSs).2
One of the most comprehensive descriptions is given by Rothman (2009, p. 156), who defines a heritage language as a language spoken at home, or however is readily available to young children, which is not the dominant language of the society in which they live. According to the author, a heritage language is acquired through the interaction with naturalistic input, like monolingual L1 acquisition; unlike the latter, however, there may be quantitative and qualitative differences in the heritage language input, which, together with the influence of the dominant language of the society (also in formal education), may result in an arrested development of the HL in children and in cases of attrition in adult speakers.
The difficulty in defining a heritage speaker also relates to the great variability observed, which depends on different sociolinguistic sources of variation (Aalberse and Muysken 2013), among which are the age of acquisition of the two languages and, even more important, the domains of language use: whether the HL is used only at home or, rather, in a larger linguistic community (i.e., in the street, in shops, at work or even at school).
It is worthwhile underlining that the quantity of input of the HL is not the sole potential reason for what has been labeled “incomplete acquisition”. As Kupisch and Rothman (2018) point out, the opportunity to receive a formal education in the HL clearly leads to a different outcome of linguistic development. Crucially, bilingual children do not often have the possibility to benefit from schooling in their HL, due to the linguistic and educational policy of the host country, as well as to the prestige of the HL. As a consequence, their HL mastery in adulthood is incomplete or, to some extent, different from the baseline.3
The poor quality of the input and the lack of literacy training in the HL is also the main cause of one of the distinctive features of HSs reported in the literature (Benmamoun et al. 2013, p. 8), namely a discrepancy between receptive and productive command of the HL. The data reported by Carreira and Kagan (2011) are particularly relevant to this point: in self-assessment tasks, HSs rate their listening and speaking skills as stronger than reading and writing skills.
Furthermore, the families’ attitudes towards the HL may vary.; On the one hand, there are parents who always use it in interactions with their children, in order to preserve it for the future generations; on the other hand there are others who want to raise their children in the dominant language of the country because they aim at a better integration of them in the hosting country and, in the latter case, the HL will be easily understood but not fully mastered by the youngest generation.
Last but not least, we have to consider the prestige of the two languages involved; there are indeed situations where the HL is considered less prestigious and less important than the language of the hosting country, and this leads to so-called subtractive bilingualism (Lambert 1981), which may end up in the progressive loss of the HL over time. This is indeed what often happens to the Albanian language spoken in Italy, a country where the Albanian immigrant community is very large, due to the long-standing contacts between the two populations, and hence between the two languages, as we will see in the next section.

3. The Historical Relationship between Italian and Albanian

The first contacts of the Albanian language with Italian can be dated around the 12th century, when Venice founded commercial establishments in the ports on the Albanian territory (Di Giovine n.d.).The first Italianisms (or better, Venetianisms) in Albanian date back to this era, and to the immediately following centuries. For example, the word rrugë, which in Albanian means ‘road/street’, is taken from ruga in the Venetian dialect of the 15th century (Devole 2022).
Around the first half of the 15th century, large groups of Albanians left their country, because of the advance of the Ottoman Empire in the Balkans, and arrived in Southern Italy. These groups gathered in communities and founded many villages in the hilly areas of various southern regions (Mandalà 2007; Dajko 2022). The isolation of these areas contributed to the preservation of the Albanian language and traditions in the Italian territory.
During the 16th and 17th centuries, after the Counter-Reformation, the widespread diffusion of Catholicism in Albania, especially in the northern area, led to the translation into Albanian of liturgical texts written in Latin or Italian. In the following two centuries, particularly in the 19th century, we observe borrowings belonging to an upper-bourgeois register, introduced through journalism or literature (Di Giovine n.d.).
In the twentieth century, the relations between the two countries became more intense, especially during the Italian occupation of Albania in the 1930s. The Albanian lexicon borrowed a lot of Italianisms of ideological origin, as well as referring to institutions and artifacts imported from Italy, like automobil(ë) ‘car’.4
The anti-Italian reaction coinciding with the end of the Second World War and the affirmation of a particularly rigid communist regime led to a slowdown in the loans from Italian, but only at an official level. Indeed, Italian was very widespread during the totalitarian Albanian regime, and a large part of the population spoke the Italian language, a custom influenced by tradition, historical ties and, since 1985, by Italian TV, which was illegally received in Albania. Italy was seen as a model of modernity; hence, it was considered very attractive (Devole 2022). This is the reason why larger and larger groups of people started moving towards Italy right after the fall of the communist government and the consequent opening of the borders in 1991.5
Hence, nowadays we may observe different instances of linguistic contact between Italian and Albanian. The old Albanian language—nowadays known as Arbëresh—which arrived in Southern Italy in the 15th century is still spoken as L1 in around 50 communities (Altimari 1986). The modern Albanian language is instead spoken by the approx. 420,000 Albanian citizens currently living in Italy (ISTAT 2022); they represent one of the largest immigrant communities which have fully integrated into Italian society, as confirmed by the high number of families and children, many of which were born in Italy (Pittau et al. 2009).

4. Albanian as a Heritage Language

With regard to Albanian as an HL, most of the existing research concentrates on Albanian immigrants in Greece (Gkaintartzi et al. 2014). For what concerns the speakers’ attitude towards the HL, a quantitative large-scale survey conducted nationwide recorded a higher competence in Greek in comparison to Albanian and a preference for the hosting country’s language among second-generation speakers of Albanian origin, especially when communicating among siblings and peers (Gkaintartzi et al. 2016).
Unfortunately, despite the widespread presence of Albanian communities, as said above, there is a lack of research on Albanian as an HL in Italy. Among the few studies, Dzeka (2023) conducted a sociolinguistic survey on 20 first- and second-generation Albanian speakers living in Italy, aimed at investigating, among other things, their ability in the two languages, their usage of them and their feelings towards their mother tongue and culture. Interestingly, only 20% of the participants were assumed to speak only Albanian at home, while the remaining 80% used a mixture of the two languages.6 All (100%) declared speaking Albanian with the family, 50% of them with friends and much lower percentages were recorded for the usage of Albanian at school or with colleagues. However, second-generation speakers self-evaluated as more proficient in Italian in comparison to Albanian in all of the four linguistic skills.
Mucaj (2018) conducted an online survey on 120 people of Albanian origin living in Italy, aged between 4 and 54. Interestingly, 78.3% deemed Albanian as their L1, but only 52.9% considered Albanian as the language in which they were most fluent; moreover, Italian was preferred for written communication and Albanian for oral, informal conversations. Finally, almost all (97.5%) admitted the frequent use of code-switching, with different functions: to fill lexical gaps, to be better understood, to express group identity and, last but not least, to speak of certain topics.
To sum up, the speakers’ attitudes appear to be rather variegated. The process of the integration of Albanians in the Italian territory, and the long-standing contact between the two languages discussed above, slowly led to the fact that people started mixing Albanian and Italian in informal speech; the progressive loss of competence in the mother tongue of second-/third-generation Albanians and the progressive fluency reached in Italian—also due to formal education—resulted in more and more frequent instances of code-switching in the daily speech of Albanian immigrants.7

5. The Sociolinguistic Survey: Research Questions

In order to gather more data on the situation of Albanian as an HL in Italy, we administered a sociolinguistic survey to some Albanian–Italian bilingual speakers. It was delivered online via the Limes Survey platform and was filled in autonomously by the participants.
The sociolinguistic survey contains 37 questions divided into two main sections. The first section collects information about the bio-data of the speakers (age, profession, the age of acquisition of both languages and the time of permanence in Italy) while the second section contains questions about the contexts of use of the two languages and the practice of code-switching. The second section also contains a self-assessment task for the four basic skills (listening, speaking, reading and writing) scored on a Likert scale from 1 (insufficient) to 10 (excellent).
The sociolinguistic survey was filled in by 27 bilingual speakers; they are of different ages and have been subdivided according to the Age of Onset (AoO), that is, the age at which they first came into contact with each of the two languages, as can be observed in Table 1:
For this sociolinguistic inquiry, we opened the following research questions:
  • RQ1: which are the main contexts of use of Albanian and Italian?
  • RQ2: what are the participants’ self-assessments of their own competence in the two languages?
  • RQ3: what are the participants’ opinions on code-switching and do they frequently resort to it in informal conversations?

6. Results of the Sociolinguistic Survey

In this section, we will present the results that we have provisionally obtained in this pilot work with the sociolinguistic survey, and answer our research questions in turn. Since some of these results intermingle with those obtained with the linguistic survey, we will postpone the discussion to the end of the paper.

6.1. RQ1: Which Are the Main Contexts of Use of Albanian and Italian?

As can be seen in Figure 1 below, both languages are widely used in everyday practice; however, we may underline a significant difference among formal and informal contexts, as expected. Since all of our participants presently live in Italy, it comes as no surprise that Italian is the sole language spoken at school or university. Italian is also predominantly spoken at work; however, since many Albanians work together with family or people from the same origin, the Albanian language is also used in this context.
Conversely, Albanian is predominantly used in informal contexts, in particular at home or with friends. These results are in line with what is assumed by Dzeka (2023). Interestingly, we also see a significant contrast between these contexts, which typically involve spoken language and social networks; though the latter is a far more informal context in comparison to school/university or work, the fact that it requires written language explains the higher percentage of the use of Italian.

6.2. RQ2: What Are the Participants’ Self-Assessments of Their Own Competence in the Two Languages?

If we look at Figure 2 below, the most important result which emerges is that, on average, all participants deem that they have a good competence in the two languages at all levels, with the highest rate for listening and the lowest rate for the writing skill (however, it is still above 8.40), independently of the language used. These data confirm Carreira and Kagan’s (2011) data, where receptive skills (listening and reading) were assessed as higher than productive ones (speaking and writing).
Furthermore, there are no significant differences in the speakers’ self-assessment of their own competence in the two languages, except for what concerns the speaking skill, for which they rate definitely higher their competence in Italian, which is the language they use most in everyday life.
However, far more interesting data emerge if we split the participants into four groups on the basis of the age of onset for the Italian language (Figure 3).
Those who came into contact with Italian before the age of 5—presumably most of them were born in Italy—deem their competence in Italian perfect (with a score of 10.00 in all of the four abilities); conversely, their competence in Albanian is rather good in receptive skills (8.00 for listening and 7.00 for reading), but not so good in productive ones, especially writing (5.50). These data, which confirm again what was assumed by Carreira and Kagan (2011), are a consequence of the fact that they use the Albanian language only orally, at home or with friends, and they have never received formal education in this language, but only in Italian. As discussed in Section 2 above, the importance of receiving formal education in a language is crucial for its complete development, as pointed out by Kupisch and Rothman (2018), and this will be confirmed by the data relating to the following groups.
Participants who came into contact with Italian from the age of 6 to the age of 10 had not lived in a bilingual environment in the first years of their life, which are fundamental for linguistic development. This is the reason why they judge high, but not perfect, their competence in Italian (from 9.00 to 9.50 in all skills). At the same time, their competence in Albanian is much higher in comparison to the previous group in all skills, especially in writing. This comes as a consequence of the fact that they presumably attended some school years in Albania before moving to Italy. Still, in all skills they deem their competence in Italian higher than in Albanian.
The situation is reversed in the third group (participants who came into contact with Italian from the age of 11 to the age of 15). Their competence in the two languages is quite balanced but, crucially, these speakers deem their competence in Albanian as higher in all of the four skills, though there is no significant difference except in the case of writing. This can be explained by the fact that they received at least their basic education (primary school) in the Albanian language and they have come into contact with Italian after the so-called sensible period for linguistic acquisition (Lenneberg 1967, and more recently; Abrahamsson and Hyltenstam 2009).
Finally, participants of the fourth group, who came into contact with Italian after the age of 16, definitely feel they are more competent in Albanian than in Italian at all levels; however, they do not judge their competence in Albanian as perfect in any of the four skills, speaking included, and this can be viewed as a result of the attrition process (Brehmer and Treffers-Daller 2020), due to the fact that they have been living in Italy for many years. Concerning their competence in Italian, it is considered very good for what concerns the oral skills, listening and speaking, but they assign a lower score to reading and especially to writing, as a consequence of the fact that they (mostly) completed their education in Albania.

6.3. RQ3: What Are the Participants’ Opinions on Code-Switching and Do They Frequently Resort to It in Informal Conversations?

As can be seen in Figure 4 below, CS emerges as a normal everyday practice among bilingual speakers in informal contexts. Almost all of our participants admit they frequently hear and use mixed sentences, with very few exceptions. The majority of them declare they hear CS often or very often (21 out of 27) and use it often or very often (20 out of 27).
Concerning their personal judgment of the practice of CS (Figure 5 below),8 most of our participants regard it positively, i.e., as a fun or useful language practice, and even as a linguistic skill. This is in line with much recent literature on Translanguaging,9 which considers the languages known by bilingual/multilingual speakers as a unique repertoire, from which they can select the linguistic items that are most appropriate to each context. Hence, the knowledge of more languages, as well as the possibility of mixing them in the same discourse, is viewed as an extra skill which characterizes the bilingual speaker.
At the same time, some negative judgments are also reported: some participants, and in particular those who came into contact with Italian at a later age, judge it as a sign of weakness, a practice speakers resort to when they do not master (one of) the languages perfectly, i.e., when they lack a lexical item. Indeed, also in Section 4 above we have mentioned the fact that one of the main functions of CS is to fill lexical gaps.
This is the reason why we have decided to investigate in greater detail the phenomenon of CS within the Albanian–Italian bilingual community, and its directionality, i.e., if speakers prefer to insert Italian words into Albanian discourse, or vice versa. But before illustrating our linguistic survey, we will briefly summarize the main theories on CS that we adopt in our analysis.

7. Typologies of Code-Switching

Mixing languages is a common practice in bilingual communities, though it is often stigmatized and considered as a sign of poor proficiency in (one of) the languages involved. However, nowadays bilingualism studies have largely moved away from this simplistic and reductive analysis; the monolingual bias has been largely replaced by a more dynamic and holistic view of bilingualism and its typical linguistic phenomena. Code-switching (CS) is no longer considered a sign of weakness, but a sophisticated and rule-governed language practice, both pragmatically (Auer 1995) and grammatically (Poplack 1980; Myers-Scotton 2002; MacSwan 1999; López 2020 among the many).
When talking about CS, the first issue that emerges is which grammar is in use in mixed productions. The main CS theories diverge on this point. According to the Matrix Language Frame model (MLF),10 in CS contexts one of the two languages—the so-called Matrix language—is dominant; as a consequence, mixed productions should follow the sole grammatical rules of the Matrix language, which also establishes the linear order of constituents. In addition, functional words should be expressed in the Matrix language, while the other language—called the Embedded language—should only provide lexical material and its grammar is not active.
Conversely, in the Borrowing Hypothesis (BH),11 the grammars of both languages are considered active. In this regard, the Equivalence Constraint (EC) states that “code-switches will tend to occur at points in discourse where juxtaposition of L1 and L2 elements does not violate a syntactic rule of either language, i.e., at points around which the surface structures of the two languages map onto each other” (Poplack 1980, p. 586). Any violations of the EC will then result in the syntactic integration of the single word as if it were a loanword. In addition, if the EC imposes restrictions at the syntactic level, the second constraint, the Free Morpheme Constraint (FMC), defines the extent of the morphological integration of the two languages. In particular, the FMC forbids the switch of bound morphemes, below the word level, as in the Spanish–English example *eat-iendo ‘eating’ reported by Poplack (1980, p. 586).
The two core constraints of the Borrowing Hypothesis have been translated into Minimalist terms in the Bilexical Model advanced by MacSwan (1999 and subsequent work).12 In this model, any switch is admitted as far as no rule of the two languages in question is violated.
From a descriptive point of view, CS can be classified into three distinct typologies (Muysken 2000, 2013): Insertion, Alternation and Congruent Lexicalization. Each type of CS varies according to both the structural properties of the languages in contact and the sociolinguistic factors defining the bilingual setting. If the Insertion type is more frequent in the speech of first-generation immigrants, who are less proficient in the language of the hosting country, Alternation and Congruent Lexicalization require a higher fluency in both languages, hence we expect these types to be more present in second-generation immigrants.
In Insertion, one single lexical item/phrase is inserted within the structure of another language, i.e., the Matrix language in Myers-Scotton’s (1993) terminology, as in (1):
1.na’iish-crash
1:SG:pass out-crash EMPH
I am about to pass out
(Navajo/English; Muysken 2000, p. 5)
In (1), the English verb crash is inserted into a Navajo sentence. The switch occurs at a point where the two languages do not share a morphosynatctic structure; as a consequence, the inserted lexical item can be considered as a fully integrated loanword.13
However, if in (1) it is possible to identify the Matrix language of the clause (Navajo), in (2) below, which exemplifies Alternation, the contribution of one of the two languages is not limited to lexical material, since both languages provide the functional skeleton of the clause. Crucially, the switch occurs at a neutral point, i.e., at the main clause boundary. Consequently, no potential conflict arises between the grammars of the two languages.
2.Je telephone à Chantal, he, meestal voor commiskes te doen en eten.
‘I call Chantal, hm, mostly to go shopping and eat.’
(French/Dutch, from Treffers-Daller 1994, p. 213)
The absence of a potential conflict at the switch point also characterizes Congruent Lexicalization, exemplified in (3):
3.Weet jij waar Jenny is?
Do you know where Jenny is?
(Dutch/English, Muysken 2000, p. 5)
In (3), a single item is inserted, as in (1). However, in (3) the English item is not a lexical but a functional word, which is not integrated into the morphosyntactic structure of Dutch, since it retains its inflection. Thus, both grammars are considered to be active. In addition, unlike Insertion, Congruent Lexicalization is also characterized by the number of possible switches within the sentence. In (4), the speaker frequently switches, going back and forth between the two languages even across the clause border:
4.Bueno, in other words, el flight que sale de Chicago around three o’clock.
Good, in other words, the flight that leaves Chicago around three o’clock.
(Spanish/English, Pfaff 1979)
Among the several constraints proposed in the CS literature, the two core constraints assumed by Poplack’s Borrowing Hypothesis, i.e., the EC and the FMC discussed above, are particularly suitable to distinguish Insertion from Alternation and Congruent Lexicalization.
As we will see in the next section, Italian and Albanian present structural properties in the DP domain which are particularly suitable for testing the validity of the mentioned constraints.

8. The Syntax of Italian and Albanian DP: A Contrastive Analysis

As clearly emerges from the literature on code-switching, one of the most widely observed and studied switch points involves mixed DPs. This is a very interesting topic to investigate in Italian–Albanian CS contexts, since the two languages present many differences concerning the determiner system, which we will briefly present in what follows.
All Italian nouns exhibit a grammatical gender, masculine or feminine, which reflects the natural gender in (most) animate nouns and is assigned arbitrarily to inanimate/abstract nouns. The same grammatical gender is also assigned to the determiner or any other modifier which accompanies the noun.
The Italian language features six definite articles, which encode gender and number information; for masculine (singular or plural) nouns, two allomorphic forms co-occur, which are selected according to the initial sound of the nouns they precede. In addition, there are three indefinite, singular articles (again, two allomorphic forms for masculine); for plural indefinite nouns, partitive articles are employed, which are formed out of the preposition di + definite article forms, with regular morphological adjustments. Italian articles, as well as nouns, do not exhibit any morphological Case distinction.
Table 2 shows the complete scheme of Italian articles:14
The syntax of Italian articles is quite simple: both definite and indefinite articles are free morphemes, which always precede the nouns they accompany and agree with them in gender and number:
5.a.il ragazzo/un ragazzo (masc.sing)
‘the boy/a boy’
b.la ragazza/una ragazza (fem.sing)
‘the girl/a girl’
In Albanian, nouns are arbitrarily distributed into three gender classes (except for some animate nouns, which follow natural gender). However, there are few neuter nouns and they are generally derived from past participles or adjectives, always preceded by the free morpheme . All noun modifiers receive the same gender as the noun they accompany; indeed, in many cases it is the article employed which reveals the gender of a noun, as the morphological form of the latter is often ambiguous (Turano 2004, p. 25), unlike what is observed in Italian, where gender can generally be inferred by the final inflection of the noun.
The situation of Albanian articles is rather complex. Like Italian, this language also features two series of articles, definite and indefinite ones. Indefinite articles are free morphemes: there is only one form, një, which is used for all singular nouns and precedes them. Definite articles, instead, are bound morphemes which appear post-nominally, as a sort of inflection. Singular definite articles show a gender and number distinction, as in Italian, and there are allomorphic forms as well, depending on the final sound of the word to which the article affixes.
Plural definite articles do not show a gender distinction; the same forms are used for all genders. There are no plural indefinite articles.15
In Table 3 below, we find the scheme of Albanian articles (Turano 2004):
Unlike Italian, Albanian definite and indefinite articles do not occupy the same position: the former are enclitic bound morphemes, while the latter are pre-nominal free morphemes:
6.a.një djalë/djal-i (masc.sing.)
a boy the boy
b.një vajzë/vajz-a(fem.sing.)
a girl the girl
To complicate the picture even more, Albanian nouns, pronouns and adjectives are inflected according to five morphological Cases: nominative, accusative, dative, genitive and ablative. Cases are encoded into different morphological endings; hence, with definite nouns, Case inflections intermingle with definite articles.

9. Deriving Mixed DPs

Given the mentioned structural differences in the DP domain, the main focus of our linguistic investigation is, on the one side, the acceptability of Italian–Albanian mixed DPs, namely DPs where D and N are expressed in different languages, and, on the other, the directionality of CS, i.e., if participants prefer to insert Italian material within an Albanian sentence, or vice versa.
Most of the existing research on mixed DPs deals with languages which exhibit the same linear order, with D preceding N, in line with Poplack’s (1980) Equivalence Constraint discussed above. This is indeed the problem which arises in Italian–Albanian mixed DPs, which challenge the mentioned constraint, as D and N do not often exhibit the same linear order in the two languages (cf. (5) and (6)).
Hence, in deriving Italian–Albanian mixed DPs, we distinguish two main contexts according to the position of the determiner: the Post-N context (definite article) and the Pre-N context (indefinite article). The sentence pairs in (7)–(8) exemplify the two contexts, featuring a DP in the subject position in Albanian (7a–8a) and in Italian (7b–8b):16
7.a.Tenxher-ja është brenda furr-it
Pot-DEF.F.SG is inside oven-DEF.M.SG
‘the pot is inside the oven’
b.La pentola è dentro al forno
DEF.F.SG pot-F.SG is inside to-DEF.M.SG oven
‘the pot is inside the oven’
8.a.Një shishe u thye
INDEF bottle(F.SG) REFL is broken
‘a bottle has broken’
b.Una bottiglia si è rotta
INDEF.F.SG bottle-F.SG REFL is broken
‘a bottle has broken’
If we manipulate the sentences in (7a–b) by switching the nouns, we obtain the Insertion types in (9a–b) below. In these sentences, only one language provides all the functional categories of the clause and the other language simply provides a lexical element, in line with the MLF. Crucially, there is only one D in each DPsubject: the enclitic Albanian D (-ja) in (9a) and the proclitic Italian D (la) in (9b).
9.a.Pentol-ja është brenda furr-it
Pot-DEF.F.SG is inside oven-DEF.M.SG
‘the pot is inside the oven’
b.La tenxhere è dentro al forno
DEF.F.SG pot-F.SG is inside to-DEF.M.SG oven-M.SG
‘the pot is inside the oven’
The two patterns in (9) would both be problematic for the EC; according to this constraint, this type of switch would not be acceptable, since the relative position of definite D and N is different in Italian and Albanian. However, only (9a) violates the FMC, as it features a switch between a root and a bound morpheme (the enclitic article -ja); conversely, (9b) does not violate the FMC since the Italian article is a free morpheme and we have used the basic form of the Albanian N, that is, the form used with indefinite articles.
However, if we inserted into (9a–b) the inflected forms of N (i.e., N + enclitic article) we would obtain two other possible outputs, with the co-occurrence of two Ds, as in (10). In this case, both the EC and FMC would be violated; moreover, these sentences do not conform to the requirements imposed by the MLF, i.e., that functional words should all be expressed in the Matrix language:
10.a.La tenxher-ja è dentro al forno
DEF.F.SG pot-F.SG is inside to-DEF.M.SG oven-M.SG
‘the pot is inside the oven’
b.La pentol-ja është brenda furrit
DEF.F.SG pot-F.SG is inside DEF.M.SG
‘the pot is inside the oven’
With indefinite DPs, instead, the syntactic order of D and N is the same in both of the languages involved. Therefore, no conflicts arise if we exchange the nouns in sentences like (8a–b) above; neither the EC nor the FMC are violated.17 In (11), we exemplify the mixed patterns available with indefinite articles:
11.a.Një bottiglia u thye
INDEF bottle-F.SG REFL is broken
‘a bottle has broken’
b.Una shishe si è rotta
INDEF.F.SG bottle(F.SG) REFL is broken
‘a bottle has broken’
The same type of mixed patterns is also derivable by the manipulation of DPs in the object position. This case is more complex than the preceding one, as the Albanian article presents accusative Case morphology. In (12) and (13), we report the four combinations of the Post-N context (definite Ds), while in (14) we report the two combinations of the Pre-N context (indefinite Ds). Notice that the Albanian enclitic article in (12a), as well as in the cases of article reduplication in (13a–b), is inflected in the accusative:
12.a.Derdha gjithë olion mbi tavolinë.
I-poured all oil-DEF.M.SG.ACC on table-DEF.F.SG.
‘I poured all the oil on the table’
b.Ho versato tutto il vaj sul tavolo.
I-have poured all-M.SG DEF.M.SG oil-M.SG on-DEF.M.SG table-M.SG
‘I poured all the oil on the table’
13.a.Ho versato tutto il vajin sul tavolo.
I-have poured all-M.SG DEF.M.SG oil-DEF.M.SG.ACC on-DEF.M.SG table-M.SG
‘I poured all the oil on the table’
b.Derdha gjith l’olion mbi tavolinë.
I-have poured all DEF.M.SG oil-DEF.M.SG.ACC on table-DEF.F.SG.
‘I poured all the oil on the table’
14.a.Ndoshta nëse shkon tani gjen akoma ndonjë negozio të hapur.
Maybe if you go now you will still find some-M.SG shop-M.SG open
‘Maybe, if you go now, you will still find some shops open’
b.Forse se vai adesso trovi ancora qualche dyqan aperto.
Maybe if you go now you will still find some-SG shop-M.SG open
‘Maybe, if you go now, you will still find some shops open’
Table 4 summarizes the different predictions made by the patterns described above with respect to the BH and MLF constraints.

10. The Linguistic Survey: Research Questions

In order to investigate the acceptability of the different types of mixed DPs among Italian–Albanian heritage speakers, we administered an Acceptability Judgment Task, scored on a 6-point Likert scale,18 to 8 of the 27 Albanian–Italian bilingual participants who had participated to the sociolinguistic survey.19 The AJT was delivered online via the Limes Survey platform, but it was filled in in person with the support of an Italian–Albanian bilingual speaker who pronounced the sentences to the participants.
In Table 5, we can find information on the participants of the AJT:
The AJT contains 218 sentences, encompassing a range of different linguistic aspects, as can be observed from Table 6 below: 48 sentences with a mixed DP in the subject or in the object position, 28 monolingual controllers (14 grammatical sentences and 14 ungrammatical sentences), 63 mixed sentences with genitive constructions, and 64 sentences with mixed words (compounds and mixed verbal forms). In this study, we will focus our attention only on the 48 test sentences (TSs) with a mixed DP in the subject and object position, leaving the discussion of other aspects to future research.
The 48 test sentences are derived from 15 nouns with the same gender in the two languages: 8 are masculine and 7 feminine; 24 are singular and 24 plural. The 218 sentences were distributed across four tests of approximately 60 sentences each and administered separately.
For the linguistic survey, we opened the following research questions:
  • RQ4: In Italian–Albanian CS contexts, is it possible to have mixed DPs? If so, will speakers prefer combinations of Albanian D + Italian N, or vice versa?
  • RQ5: Will mixed DPs be accepted only with indefinite articles, which are pre-nominal in both languages?
  • RQ6: With definite articles, would it be possible to have a noun accompanied by two determiners, one pre-nominal and one post-nominal?
  • RQ7: Will there be differences in the acceptability of mixed DPs according to their syntactic role (subject vs. object)?

11. Results of the Linguistic Survey

Since our participants assigned a score to the acceptability of mixed sentences, we calculated the frequency percentage of the assigned scores.
First of all, participants were divided into two groups according to their AoO of the Italian language: group A contains five speakers who acquired Italian from age 6 to 10, and group B contains three speakers who acquired Italian after the age of 16. The two groups exhibit a disparity in the number of participants; in particular, there are only three participants who acquired Italian after 16, a fact which consequently diminishes their impact on the calculated acceptance rates. Nevertheless, the distributions of ratings demonstrate the presence of distinct behaviors among the two groups. In order to confirm this tendency, we also conducted a second analysis by comparing the mean of the percentages of acceptance of each group.
In Figure 6 below, we report the frequency percentages of the scores provided by the eight Italian–Albanian bilinguals for the 16 test sentences containing mixed DPs with a definite article, i.e., the CS type reported in (9) above. As we can observe, there is a clear preference for the insertion of an Italian N into an Albanian sentence rather than the other way around. The insertion of an Italian N reaches 44% acceptability in sentences with the mixed DP in the subject position, and 47% in sentences with the mixed DP in the object position. In contrast, the acceptance rate of the eight test sentences where an Albanian N is inserted into an Italian sentence does not exceed 20%, either for the subject or the object position.
Interestingly enough, the four TSs with a mixed DP with an Albanian N in the object position were not accepted by all of the three speakers who acquired Italian after 16. Two of them judged only the sentences in (15) below more or less acceptable, but not completely grammatical:
15.?Passami la fshesë
pass-me DEF.F.SG bottle-F.SG
‘pass me the bottle’
In (15), the inserted Albanian N has the basic form (indefinite) that does not match the definite interpretation of the DP. It is worthwhile noticing that the insertion of the inflected form of the Albanian N in the accusative Case (16) causes an increase in the acceptance rate:
16.Passami la fshesën
pass-me DEF.F.SG bottle-DEF.F.SG.ACC
‘pass me the bottle’
Something similar takes place in the sentence with a masculine Albanian N in the subject position (28%). Also in this case, the double D pattern is preferred by group B, as can be seen in Figure 7, where we report the results of the 16 TSs with a double D.
In all of the other cases of doubleD, expectedly, the acceptance rate decreases. In particular, the combinations featuring an Italian N are almost always rejected, especially by group A participants.
In Figure 8 below, we report instead the results of the 16 TSs with indefinite articles. As can be observed, this pattern obtains positive scores.
In sentences with a mixed DP in the subject position, the acceptance rates reach 63% with an Italian N (17a), and 38% with an Albanian N (17b):
17.a. Disa domande ishin shumë të vështira
some-PL questions-F.PL were too difficult
‘some questions were too difficult’
b. Alcune pyetje erano troppo difficili
some-F.PL questions-F.PL were too difficult
‘some questions were too difficult’
We also recorded a high acceptance rate (72%) for the insertion of an Italian N in the object position (18a). Conversely, the acceptability of sentences with an Italian indefinite D + Albanian N (18b) reached only 44%, which is, however, a positive score in comparison to what was observed for definite DPs.
18a.Do më tregosh ndonjë borsa në Zalando.
Show me some-INDEF bag-F.SG on Zalando
‘show mw some bag on Zalando’
b.Mi fai vedere qualche çanta su Zalando
Show me some-INDEF bag-F.SG on Zalando
‘show mw some bag on Zalando’
Finally, in Figure 9 and Figure 10 below we report the acceptability rates of the various types of mixed DPs placed in, respectively, the subject and the object position. Crucially, an interesting difference emerges between the two syntactic contexts, especially for what concerns group A participants:
After this long presentation of the data, we are now able to provide an answer to the research questions set above.

11.1. RQ4: In Italian–Albanian CS Contexts, Is It Possible to Have Mixed DPs? If So, Will Speakers Prefer Combinations of Albanian D + Italian N, or Vice Versa?

The data presented above confirm that it is certainly possible to have mixed Italian–Albanian DPs, and the acceptance rates are altogether positive, though not all available D-N combinations obtain the same scores. Overall, we have recorded a clear preference for the insertion of an Italian N into an Albanian matrix clause rather than the other way around: ca. 45% (29/64) vs. 17% (11/64) with definite determiners, and 67% (43/64) vs. 25% (25/64) with indefinite determiners.20

11.2. RQ5: Will Mixed DPs Be Accepted Only with Indefinite Articles, Which Are Pre-Nominal in Both Languages?

Mixed DPs are not accepted only with indefinite articles; however, the latter is by far the most widely accepted option: 53% (68/128) vs. 31% (40/128) for definite DPs.
Indeed, in the case of indefinite DPs, Albanian and Italian share the same linear order (D-N), and hence the switch causes no violation of the EC; moreover, the FMC is not violated either, as indefinite articles are free morphemes in both languages (see Table 4 above). In addition, in line with Myers-Scotton (1993, 2002), the language in which D—i.e., the functional head—is expressed represents the matrix language, and the other language provides only lexical material.
With definite DPs, the combination of an Albanian N and an Italian D causes a violation of the sole EC, the Italian D being a free morpheme, while the combination of an Italian N and an Albanian D violates both constraints, the Albanian D being a bound morpheme (see Table 4 again). Nonetheless, the latter option obtains higher acceptance rates, by virtue of the fact that participants prefer the insertion of an Italian N into an Albanian clause, as seen in 11.1 above.

11.3. RQ6: With Definite Articles, Would It Be Possible to Have a Noun Accompanied by Two Determiners, One Pre-Nominal and One Post-Nominal?

The acceptance rate of nouns accompanied by two Ds (one Italian, pre-nominal, and one Albanian, post-nominal) is expectedly quite low, in line with Myers-Scotton’s assumption that functional heads should be provided only by one language (the Matrix language).
However, unlike what was attested for definite and indefinite articles, the lowest scores were reported for Italian nouns inserted into Albanian clauses (acceptance rate below 14%, 9/64). There were slightly more positive scores for (some) Albanian nouns inserted into Italian clauses (19%, 12/64). This can be explained by the fact that, in definite contexts, an Albanian noun should never appear in the basic form, but always inflected by means of an enclitic morpheme which signals not only definiteness but also Case; vice versa, an Italian noun is never inflected for Case and always shows up in the basic form.

11.4. RQ7: Will There Be Differences in the Acceptability of Mixed DPs According to Their Syntactic Role (Subject vs. Object)?

Finally, if we compare the acceptance rates of mixed DPs in the subject and object positions, we notice an interesting difference. With definite Ds, the acceptance rate of mixed DPs in the subject position is overall higher than in the object position; specifically, the insertion of an Italian N in the subject position reaches 55% for group A and 25% for group B, and, respectively, 31% and 16% in the object position. The asymmetry is even more evident in sentences where an Albanian N is inserted in an Italian clause. In this case, the acceptance rate of mixed DPs in the subject position is 20% for group A and 8% for group B. Interestingly enough, only the first group accepted an Albanian N in the object position (19%).
An analogous asymmetry is also available in indefinite contexts, where the percentages of grammatical mixed DPs in the object position is systematically lower than those recorded for sentences with a mixed DP in the subject position, regardless of the language of N.

12. Discussion

Though we are aware that much more data would be necessary to draw more definitive conclusions, the results of the two surveys carried out in this pilot work have outlined some interesting tendencies.
From the answers to RQ1, we infer that our participants mainly use Albanian in oral, informal contexts. This is confirmed also by the answers to RQ2: participants’ self-assessed fluency in Albanian decreases in the skills relating to written language (reading and especially writing). Moreover, people of the youngest generations feel they are 100% competent only in Italian, but, crucially, even people who came into contact with Italian at a later stage do not assess as perfect their competence in Albanian. This comes as a consequence of the attrition process undergone by the Albanian language, caused by the daily use of Italian, or even of a mixed language. Indeed, the practice of code-switching is deemed frequent and is generally well regarded (RQ3).
Our linguistic survey confirms these judgments. Indeed, from the answers to RQ4 we infer that not only are mixed DPs generally deemed acceptable—a fact which endorses that the CS practice is far from stigmatized and mixed DPs naturally emerge in informal conversations—but the majority of speakers prefer the insertion of an Italian N into an Albanian sentence rather than vice versa. In short, the data show a definite directionality in the switching process. The heritage Albanian language seems to be more prone to be permeated by Italian lexical items than the other way around; this means that HSs, when speaking Albanian, tend to ‘forget’ some words and replace them with the Italian equivalents, which are more commonly heard in everyday conversations. Conversely, in contexts where they speak Italian, they do not feel the urge to insert Albanian words as much.
Regarding the answers to RQ5, i.e., the acceptability of different patterns of mixed DPs, it comes as no surprise that participants score mixed DPs with indefinite articles the highest. Indeed, these combinations do not violate either the Equivalence Constraint or the Free Morpheme Constraint, since indefinite articles are pre-nominal free morphemes in both languages. These combinations are also in line with the restrictions imposed by the Matrix Language Frame model, as the functional words (the determiners) are always expressed in the dominant language of the clause, while the other language only provides lexical items.
Though they are acceptable for the MLF, like the previous combinations, mixed DPs with definite articles represent a violation of the EC, since the relative order of D and N is different in the two languages; furthermore, in case of Albanian D + Italian N, we also observe a violation of the FMC, since the Albanian D is a bound morpheme. These violations explain the lower acceptance rates of these combinations in comparison to those with indefinite Ds.
We also tried to test mixed DPs with two Ds (RQ6), one pre- and one post-nominal. Expectedly, this combination is not highly accepted. Moreover, it represents a strong violation of the MLF, since there is a functional word (one of the two Ds) which is not expressed in the Matrix language of the clause.
The last element which emerges from our data is the higher acceptance of mixed DPs in the subject rather than in the object position (RQ7). This is a very interesting result that would certainly deserve a more detailed investigation. A partial explanation of this asymmetry might be found in the fact that Albanian enclitic articles do not only signal definiteness, but also Case, as discussed above. Hence, since the accusative is morphologically marked in comparison to the nominative, this would explain the higher acceptance of less marked forms, i.e., those in the subject position, with definite DPs (or double Ds). Unfortunately, this explanation cannot be extended to indefinite DPs, since indefinite Ds do not carry Case distinction. Hence, we leave this question open for future research.
To conclude, it is worthwhile pointing out that we recorded different behaviors among the two groups of participants to the AJT, i.e., people who have a different level of fluency in the two languages provided different judgments. This becomes more evident if we take into consideration the percentages of acceptability of the different types of mixed DPs recorded within the two groups: see again Figure 9 and Figure 10 above.
The five participants of group A generally exhibit higher degrees of acceptability in all types of switching. This means that these people, who acquired Italian at an early age (from 6 to 10), consider the two languages as part of a unique repertoire, from which they can freely select items and combine them, in line with the mentioned recent hypotheses of Translanguaging (Vogel and Garcia 2017 among others). Nonetheless, their preferred type of switch involves the insertion of an Italian N; this confirms once more the attrition undergone by the Albanian language among second-generation bilinguals, also evidenced by their answers to RQ2, illustrated in Figure 3.
Conversely, the three Albanian bilinguals of group B, who acquired Italian as adults, generally accept less mixed DPs in comparison to group A; the sole context where their judgments rate higher is the combination of an Italian N and a double D, either in the subject or in the object position. This can tentatively be explained by assuming that these people, whose competence in Albanian is higher (see again Figure 3 above), feel the need to add to N—even to an inserted Italian N—an inflection which signals definiteness and Case, in accordance with the grammatical requirements of their dominant language, Albanian, in line with the MLF.

Author Contributions

G.C., Section 1, Section 3, Section 4, Section 5, Section 8 and Section 11; C.P., Section 2, Section 6, Section 7, Section 9, Section 10, Section 12. All authors have read and agreed to the published version of the manuscript.

Funding

This research received no external funding.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Informed consent was obtained from all subjects involved in the study.

Conflicts of Interest

The authors declare no conflict of interest.

Notes

1
This work is the result of the collaboration of the two authors in all respects. Nevertheless, Gloria Cocchi takes responsibility for Section 1, Section 3, Section 4, Section 5, Section 8 and Section 11, and Cristina Pierantozzi for Section 2, Section 6, Section 7, Section 9, Section 10, Section 12.
2
See the overviews offered by Polinsky (2018) and Aalberse et al. (2019) for greater detail.
3
See in this regard Kupisch and Rothman’s (2018) epistemological paper and the debate around the keynote paper of Benmamoun et al. (2013) for a detailed discussion.
4
Ibidem.
5
We can divide the migration into three main stages: 1991, 1997 and 1999, during the Kosovo war (Dajko 2022).
6
50% had Albanian as the dominant language, and 30% had Italian. Nobody declared to speak only Italian.
7
We wish to thank our many Albanian students at Urbino University who have personally provided us with information on this subject.
8
Participants could give more than one answer to this question.
9
See among the many Garcia (2009), Garcia and Wei (2014), Vogel and Garcia (2017).
10
See Myers-Scotton (1993, 2002) and related work.
11
12
However, see López (2020) for a different position.
13
However, see Myers-Scotton (2002 and related work) for a radically different analysis.
14
When definite articles lo and la, and the indefinite feminine una, precede a noun starting with a vowel (or semi-vowel) sound, the elided forms l’ and un’ are employed.
15
However, for the few neuter nouns which can be inflected in the plural, there is also an indefinite form with a final inflection, in addition to the pre-nominal particle të.
16
For the Albanian examples see also Dajko (2022).
17
Remember that indefinite Ds are pre-nominal free morphemes in both languages.
18
The judgment scale goes from 1—totally unacceptable, to 6—totally acceptable and natural.
19
Unfortunately for this pilot work we have not been able to collect linguistic data from more participants until now (including the remaining 19 who filled the sociolinguistic survey). It is our intention to proceed with data collection in order to be able to draw more definitive conclusions in future work.
20
Here and in what follows, percentages are calculated with respect to the total number of sentences of the relevant type (see Table 6 above) multiplied by the number of participants (8).

References

  1. Aalberse, Suzanne, Ad Backus, and Pieter Muysken. 2019. Heritage Languages: A Language Contact Approach. Amsterdam: John Benjamins. [Google Scholar]
  2. Aalberse, Suzanne, and Pieter Muysken. 2013. In Perspectives on Heritage Languages. Radboud University, Faculty of Arts & Centre for Language Studies: Position paper. Amsterdam: Nijmegen. [Google Scholar]
  3. Abrahamsson, Niclas, and Kenneth Hyltenstam. 2009. Age of onset and nativelikeness in a second language: Listener perception versus linguistic scrutiny. Language Learning 59: 249–306. [Google Scholar] [CrossRef]
  4. Altimari, Francesco. 1986. Gli arbëreshë: Significato di una presenza storica, culturale e linguistica. In I Dialetti Italo-Albanesi. Studi linguistici e Storico-Culturali Sulle Comunità Arbëreshe. Edited by Francesco Altimari and Leonardo Maria Savoia. Roma: Bulzoni, pp. 9–30. [Google Scholar]
  5. Auer, Peter. 1995. The pragmatics of code-switching a sequential approach. In One Speaker-Two Languages: Cross-Disciplinary Perspectives on Code-Switching. Edited by Leslie Milroy and Pieter Muysken. Cambridge: Cambridge University Press, pp. 115–135. [Google Scholar]
  6. Benmamoun, Elabbas, Silvina Montrul, and Maria Polinsky. 2013. Heritage languages and their speakers: Opportunities and challenges for linguistics. Theoretical Linguistics 39: 129–81. [Google Scholar] [CrossRef]
  7. Brehmer, Bernhard, and Jeanine Treffers-Daller. 2020. Lost in Transmission: The Role of Attrition and Input in Heritage Language Development. Amsterdam: John Benjamins. [Google Scholar]
  8. Carreira, Maria, and Olga Kagan. 2011. The Results of the National Heritage Language Survey: Implications for Teaching, Curriculum Design, and Professional Development. Foreign Language Annals 44: 40–64. [Google Scholar] [CrossRef]
  9. Dajko, Francesco. 2022. Italiano e Albanese: Il Loro Rapporto in 500 Anni di Storia. Master’s dissertation, Università di Urbino, Urbino, Italy. [Google Scholar]
  10. Devole, Rando. 2022. Cosa è Successo Alla Lingua Italiana in Albania? Available online: https://www.albanianews.al/sociale/lingua-italiana-albania (accessed on 15 February 2024).
  11. Di Giovine, Paolo. n.d. Dal dukát all’investitór: Nove secoli di italiano in Albania. Available online: https://www.treccani.it/magazine/lingua_italiana/speciali/nazioni/digiovine.html (accessed on 15 February 2024).
  12. Dzeka, Sulejman. 2023. Between Heritage Language and Native Language: The Albanian Speaking Community in Vittorio Veneto. Bachelor’s dissertation, Università di Padova, Padova, Italy. Available online: https://thesis.unipd.it/handle/20.500.12608/60452 (accessed on 12 March 2024).
  13. Garcia, Ofelia. 2009. Bilingual Education in the 21st Century: A Global Perspective. Malden: Wiley-Blackwell. [Google Scholar]
  14. Garcia, Ofelia, and Li Wei. 2014. Translanguaging: Language, Bilingualism and Education. London: Palgrave Macmillan. [Google Scholar]
  15. Gkaintartzi, Anastasia, Angeliki Kiliari, and Roula Tsokalidou. 2016. Heritage language maintenance and education in the Greek sociolinguistic context: Albanian immigrant parents’ views. Cogent Education 3: 1155259. [Google Scholar] [CrossRef]
  16. Gkaintartzi, Anastasia, Aspasia Chatzidaki, and Roula Tsokalidou. 2014. Albanian parents and the Greek educational context: Who is willing to fight for the home language? International Multilingual Research Journal 8: 291–308. [Google Scholar] [CrossRef]
  17. ISTAT. 2022. Stranieri Residenti al 1° Gennaio—Cittadinanza. Available online: http://dati.istat.it/Index.aspx?DataSetCode=DCIS_POPSTRCIT1# (accessed on 20 October 2022).
  18. Kupisch, Tanja, and Jason Rothman. 2018. Terminology matters! Why difference is not incompleteness and how early child bilinguals are heritage speakers. International Journal of Bilingualism 22: 564–82. [Google Scholar] [CrossRef]
  19. Lambert, Wallace E. 1981. Bilingualism and language acquisition. Annals of the New York Academy of Sciences 379: 9–22. [Google Scholar] [CrossRef]
  20. Lenneberg, Eric H. 1967. Biological Foundations of Language. New York: Wiley. [Google Scholar]
  21. López, Luis. 2020. Bilingual Grammar. Cambridge: Cambridge University Press. [Google Scholar]
  22. MacSwan, Jeff. 1999. A Minimalist Approach to Intrasentential Code Switching. New York: Garland. [Google Scholar]
  23. Mandalà, Matteo. 2007. Mundus Vultdecipi: I Miti Della Storiografia Arbëreshe. Palermo: Mirror. [Google Scholar]
  24. Mastroberti, Francesco. 2008. Le colonie albanesi nel Regno di Napoli tra storia e storiografia. Annali della Facoltà di Giurisprudenza di Taranto 1: 241–51. [Google Scholar]
  25. Mucaj, Blerina. 2018. Unpublished manuscript. Università di Urbino, Urbino, Italy. [Google Scholar]
  26. Muysken, Pieter. 2000. Bilingual Speech. A Typology of Code-Mixing. Cambridge: Cambridge University Press. [Google Scholar]
  27. Muysken, Pieter. 2013. Language contact outcomes as the result of bilingual optimization strategies. Bilingualism: Language and Cognition 16: 709–30. [Google Scholar] [CrossRef]
  28. Myers-Scotton, Carol. 1993. Duelling Languages: Grammatical Structure in Code-Switching. Oxford: Clarendon Press. [Google Scholar]
  29. Myers-Scotton, Carol. 2002. Contact Linguistics: Bilingual Encounters and Grammatical Outcomes. Oxford: Oxford University Press. [Google Scholar]
  30. Pfaff, Carol W. 1979. Functional and structural constraints on syntactic variation in code-switching. Language 55: 297–318. [Google Scholar]
  31. Pittau, Franco, Antonio Ricci, and Giuliana Urso. 2009. Gli albanesi in Italia: Un caso di best practice di integrazione e sviluppo. REMHU—Revista Interdisciplinar da Mobilidade Humana 17: 153–73. [Google Scholar]
  32. Polinsky, Maria. 2018. Heritage Languages and Their Speakers. Cambridge: Cambridge University Press. [Google Scholar]
  33. Poplack, Shana. 1980. “Sometimes I’ll start a sentence in Spanish Y TERMINO EN ESPAÑOL”: Toward a typology of code-switching. Linguistics 18: 581–618. [Google Scholar] [CrossRef]
  34. Poplack, Shana, and Marjorie Meechan. 1995. Patterns of language mixture: Nominal structure in Wolof-French and Fonge-French bilingual discourse. In One Speaker-Two Languages: Cross-Disciplinary Perspectives. Edited by Leslie Milroy and Pieter Muysken. Cambridge: Cambridge University Press, pp. 199–232. [Google Scholar]
  35. Rothman, Jason. 2009. Understanding the nature and outcomes of early bilingualism: Romance languages as heritage languages. International Journal of Bilingualism 13: 155–63. [Google Scholar] [CrossRef]
  36. Sankoff, David, and Shana Poplack. 1981. A formal grammar for code-switching. Research on Language & Social Interaction 14: 3–45. [Google Scholar]
  37. Treffers-Daller, Jeanine. 1994. Mixing two Languages, French-Dutch Contact in a Comparative Perspective. Berlin: Mounton de Gruyter. [Google Scholar]
  38. Turano, Giuseppina. 2004. Introduzione Alla Grammatica Dell’Albanese. Firenze: Alinea. [Google Scholar]
  39. Vogel, Sara, and Ofelia Garcia. 2017. Translanguaging. In Oxford Research Encyclopedia of Education. Edited by George Noblit and Luis Moll. Reproduced by permission of Oxford University Press. Available online: http://education.oxfordre.com/view/10.1093/acrefore/9780190264093.001.0001/acrefore-9780190264093-e-181 (accessed on 10 February 2024).
Figure 1. Contexts of use of the two languages.
Figure 1. Contexts of use of the two languages.
Languages 09 00285 g001
Figure 2. Self-assessment.
Figure 2. Self-assessment.
Languages 09 00285 g002
Figure 3. Self-assessment with respect to age of onset (Italian).
Figure 3. Self-assessment with respect to age of onset (Italian).
Languages 09 00285 g003
Figure 4. CS practice frequency.
Figure 4. CS practice frequency.
Languages 09 00285 g004
Figure 5. CS practice evaluation.
Figure 5. CS practice evaluation.
Languages 09 00285 g005
Figure 6. Mixed DPs with a definite article. ? means ‘acceptable but not natural’, ?? means ‘degraded’, */** mean ‘unacceptable/completely unacceptable’.
Figure 6. Mixed DPs with a definite article. ? means ‘acceptable but not natural’, ?? means ‘degraded’, */** mean ‘unacceptable/completely unacceptable’.
Languages 09 00285 g006
Figure 7. Mixed DPs with a double definite determiner. ? means ‘acceptable but not natural’, ?? means ‘degraded’, */** mean ‘unacceptable/completely unacceptable’.
Figure 7. Mixed DPs with a double definite determiner. ? means ‘acceptable but not natural’, ?? means ‘degraded’, */** mean ‘unacceptable/completely unacceptable’.
Languages 09 00285 g007
Figure 8. Mixed DPs with an indefinite article. ? means ‘acceptable but not natural’, ?? means ‘degraded’, */** mean ‘unacceptable/completely unacceptable’.
Figure 8. Mixed DPs with an indefinite article. ? means ‘acceptable but not natural’, ?? means ‘degraded’, */** mean ‘unacceptable/completely unacceptable’.
Languages 09 00285 g008aLanguages 09 00285 g008b
Figure 9. Mixed DPs in the subject position.
Figure 9. Mixed DPs in the subject position.
Languages 09 00285 g009
Figure 10. Mixed DPs in the object position.
Figure 10. Mixed DPs in the object position.
Languages 09 00285 g010
Table 1. Participants of the sociolinguistic survey.
Table 1. Participants of the sociolinguistic survey.
Number27
Age range18–66
Mean Age41
Age of Onset (AoO)ItalianAlbanian
Since birth226
From 3 to 510
From 6 to 1081
From 11 to 1540
After 16120
Table 2. Italian articles.
Table 2. Italian articles.
Italian articlesmasc.sing.fem.sing.masc.pl.fem.pl
definiteil, lo (l’)la (l’)i, glile
indefiniteun, unouna (un’)dei, deglidelle
Table 3. Albanian articles.
Table 3. Albanian articles.
Albanian
articles
masc.singfem.sing.neut.singmasc.pl.fem.plneut.pl
definite-i, -u-a, -jatë+-t/-it/-të-t, -të-t, -tëtë+-t/-të
indefinitenjënjënjë të të + -a
Table 4. Violations of CS constraints.
Table 4. Violations of CS constraints.
Equivalence ConstraintFree Morpheme ConstraintMatrix Language Frame Model
Post-N Context
(definite article)
One D9a–b     12a–b9a    12anone
Double D13a–b  14a–b13a–b  14a–b 13a–b   14a–b
Pre-N Context
(indefinite article)
One Dnonenonenone
Table 5. Participants of the Acceptability Judgment Task.
Table 5. Participants of the Acceptability Judgment Task.
Number8
Age range23–62
Mean Age40
Age of Onset (AoO)ItalianAlbanian
Since birth08
From 3 to 500
From 6 to 1050
From 11 to 1500
After 1630
Table 6. Structure of the AJT.
Table 6. Structure of the AJT.
AJT
Post-N context: N-InsertionItalianAlbanian
Subject position44
Object position44
Post-N context: Double DItalianAlbanian
Subject position44
Object position44
Pre-N context: N-Insertion
Subject position44
Object position44
TOT2424
Other type of sentences
Insertion of Monolingual DP88
Monolingual Sentences1414
Mixed Words3132
Genitives2835
TOT105113
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Cocchi, G.; Pierantozzi, C. Albanian as a Heritage Language in Italy: A Case Study on Code-Switching within DP. Languages 2024, 9, 285. https://doi.org/10.3390/languages9090285

AMA Style

Cocchi G, Pierantozzi C. Albanian as a Heritage Language in Italy: A Case Study on Code-Switching within DP. Languages. 2024; 9(9):285. https://doi.org/10.3390/languages9090285

Chicago/Turabian Style

Cocchi, Gloria, and Cristina Pierantozzi. 2024. "Albanian as a Heritage Language in Italy: A Case Study on Code-Switching within DP" Languages 9, no. 9: 285. https://doi.org/10.3390/languages9090285

APA Style

Cocchi, G., & Pierantozzi, C. (2024). Albanian as a Heritage Language in Italy: A Case Study on Code-Switching within DP. Languages, 9(9), 285. https://doi.org/10.3390/languages9090285

Article Metrics

Back to TopTop