This article examines the mechanisms involved in the formation of regional Italian from the perspective of contact linguistics. Varieties of regional Italian containing elements of both local dialects and Standard Italian (SI) are spoken throughout Italy; this paper focuses primarily on Salento, a southern region characterized by a strong bilingual environment. The aim is to investigate the interaction between a dialect and a standard language, as well as the concrete linguistic mechanisms involved. The historical background of the acquisition of SI and its diffusion throughout the national territory are considered crucial moments in language transmission. Sociolinguistic characteristics of speakers are also analyzed, as are three types of constructions: change of verb transitivity, use of polyvalent complementizer, and gerunds. The analysis confirms that the spoken variety of Italian, regional Italian, is a result of contact-induced change of SI influenced by the local dialect, Salentino.
Interaction between the standard language and dialects is a key issue in the field of Italian linguistics and dialectology. The consequences of this interaction, including code-switching and mixing, ‘Italianization of dialects’ and reciprocal interference between Italian and dialects, are also of paramount importance; likewise is the argument of this paper: the definition of regional Italian and its evolution.
In this manuscript, I will focus my attention on the linguistic mechanisms involved in the formation of regional Italian and make some assumptions about its nature. In doing so, I do not intend to limit myself to the conventional dialectological approach, rather, I wish to enrich my analysis with vast theoretical reference to contact linguistics, including studies of pidgins and creoles.
A recent paper by Auer (2005) covers the complexities of the European linguistic situation, proposing a general classification of the dialectological condition and linguistic repertoires in Europe while simultaneously commenting on types of dialect/standard continuum.
For the purposes of this paper, I have decided to base the description and analysis of the formation of regional Italian (RI) in Salento – a region in the extreme south of Italy – and its linguistic features. I regard these features, which can be considered non-standard and deviating from the norm, as indexes of contact-induced change. I aim to give a more detailed account of three of the most salient variables which I classified as regional during the course of my prior studies of language use in Salento. Attention to European dialectology also suggested speculation on multiple contact situations, which were approached from the traditional dialectological perspective. Recently, an interchange between these two disciplines (contact linguistics and dialectology) began bearing fruitful results, through an analysis of regional variation from the contact perspective.
This paper consists of four main sections: an introduction to the problem; a theoretical overview of the Italian linguistic situation and contact phenomena; data processing; and a discussion considering theoretical and empirical data.
1.1 Italian linguistic repertoire
Throughout most of its history, Italian linguistics has mainly been concentrated on the representation of the architecture of the linguistic system and the classification of its varieties. This article will exclusively present data on RI, a spoken variety of a specific region of Italy, in this case, Salento. “Regional Italian is a variety of Italian, essentially oral, spoken by well-educated persons in a determined geographical area, and is characterized by its distance from the varieties of other areas, on the one hand, and from SI, on the other” (Tempesta, 2005, my translation).
It is worth noting that there is a certain structural distance separating RI from both Salentino and Italian1. Unlike Sabatini, Mioni, Sanga, Trumper, Trumper and Maddalon and others (see Berruto 1989, 10-12), I do not adhere to a system of multiple gradations of the repertoire, but rather conduct analyses on the concrete ‘new’ forms, identified over the course of data-processing and studied in light of the triple repertoire paradigm (standard - regional - local dialect2), to then classify linguistic change types. In my opinion, for the case in question, it is impossible to draw boundaries between multiple varieties within the linguistic repertoire. My objective is to understand which internal movement is or was in progress and how the interaction between the two linguistic systems has influenced the current state of spoken Italian.
One of the central issues when treating the Italian linguistic repertoire is the definition of the distance between Italian dialects and the Italian language (see 2.1. for an elaboration on this). Speakers of different Italian dialects cannot understand each other, especially if they come from very distant areas: a speaker of a northern dialect cannot understand Sicilian or another southern dialect, even though these languages are genetically close. It can definitely be asserted that Italian and Salentino are not mutually intelligible. It is possible, though not always easy, to distinguish which features are dialectal and which are Italian. In fact, as Cerruti (2011, 13) affirms: “Contemporary Italian is undergoing a restandardization process, caused by the mutual interrelation between spoken and written language […] and the consequent acceptance of previously non standard features into the standard ones”. In fact, it is necessary to note that a neo-standard variety of Italian permits regional differentiation, that is, “region-specific (standard RI) features are equally accepted and commonly used even by the most educated speakers and are embedded in a number of nationally shared linguistic traits, mainly concerning morphosyntax” (Cerruti, 2011, 13)
Before proceeding with the theoretical analysis, I would like to briefly introduce Salento in terms of its geographic and linguistic identity. Salento is a sub-region of Southern Italy often referred to as the extreme south of the country. Part of the larger region of Apulia, a border separates this peninsula from the rest of the region. This border is not a natural geographical barrier, but a line dividing the linguistic space of the region in two clearly marked zones: the northern dialect of Apulia (Barese), and the Salentino dialect. Belonging to separate dialectal groups, they notably differ in grammatical structure (Map 1).
Due to its geographic position, Salento has historically been a rather isolated community. The nationwide success of a number of Salentino-singing music groups (for example, Sud Sound System) in the 1990s instigated a widespread awareness of the region and its linguistic profile, particularly among the younger population of Italy.
Contrary to what was expected by the majority of linguists advocating the dominance of Italian over local varieties, Mirko Grimaldi (2004) claimed that in the year 2000 dialects were far from dead or dying. Rather, they were regenerating and acquiring new forms, something Grimaldi refers to as ‘neodialettalità’ or ‘neo-dialectality’. This encompasses an active competence of both Italian and a local dialect, where the dialect does not serve as an additional source for speakers to express themselves when faced with the inability to produce a proper sentence in the Italian language. Rather, in informal contexts, it is a ‘preferred alternative’ to the national language (see the following section for a discussion on the type of repertoire defined as dilalia).
2. General issues
2.1 Italian dialects and SI
There is a fairly large structural distance between SI and the majority of Italo-romance dialects, as well as between the individual Italo-romance dialects themselves, so much so that they must be considered true separate systems rather than mere varieties of the same linguistic system, namely the Italian language. Thus an Italian dialect, aside from Tuscan and perhaps Roman, cannot be considered a variety of Italian. This somewhat paradoxical observation is of great importance in order to grasp the peculiarity of the Italian situation” (Berruto, 1989, 7).
Contrary to Auer’s observation that the structural distances between German and Italian dialects are the same or similar, the treatment of local dialects as independent linguistic systems, and not as varieties of Italian, has already gained firm ground in Italian dialectology. Berruto states, “The structural distance between the standard and the dialects in Italy is much greater; Italo-romance dialects have their own history, many of them have a (notable) literary tradition” (Berruto, 2005, 82). Italian dialects, according to the scheme proposed by Coseriu (1980 cited by Berruto, 2005), “belong to primary dialects.” Nevertheless, the linguistic repertoire (lingua cum dialectis) is characterized by the presence of tertiary dialects, i.e. regional varieties of Italian. Furthermore, the factor of mutual intelligibility supports the separate status of these systems.
As a consequence of this strict distinction between the two systems, we have “a continuum with two subcontinua; one on the side of the dialect and the other on the side of Italian. In certain cases, this continuum can resemble well-known creole repertories with an acrolect and many basilects, whereas in other cases, it appears quite like a gradatum with fairly clear-cut borders between the different varieties” (Berruto, 1989, 8). Based on Berruto’s assumptions, the main difference between Italian and other linguistic repertoires in Europe can be characterized: “regional Italians are therefore the true ‘dialects of Italian’, since the Italian dialects, as mentioned above, are at least partially separate linguistic systems” (Berruto, 1989, 8-9). The ‘standard’ Italian may be identified as an ideal form of Italian, legitimized by grammar reference books and mainly with no (or very few) native speakers. I chose to consider regular prescribed norms of Italian that are accepted throughout the entire country as the standard in my research. Italy represents a decentralized model where the capital does not exert a major linguistic influence on the language use of the population3. Following the scheme proposed by Auer (2005, 22), we can state that the present situation in Salento is similar to diaglossia or repertoire Type C (see the detailed discussion in 5).
Dilalia was the name chosen by Berruto (1989) for the type of repertoire established in Italy from the 1960s onward, in order to define the expansion of Italian into different domains and its overlapping with dialects in informal communication spheres (Dal Negro and Vietti, 2011, 72). Variation in Italy is traditionally referred to in four dimensions: social (diastratia), situational (diaphasia), geographical (diatopia) and that of means of communication (diamesia). As it is widely known, the geographical variation “is considered the primary dimension of variation” (Cerruti, 2011, 19). In fact, I will observe features that can be classified as diatopical in the RI of Salento, but, due to the type of data I dispose of and to my particular interest in the mechanisms of formation of the regional variety, will not refer to the other three dimensions. The data I analyze requires theoretical application, thus, first and foremost, a definition of a contact variety is necessary. As Thomason states, referring to phenomena such as creole formation and the genesis of pidgins, “a contact language is a language arising over the course of the historical process other than by means of a regular transmission. In other words, a contact language is comprised of a grammatical and a lexical system which may not all be traced back to a single parent language” (1997, 74-75). In this paper, I deal with a ‘second branch’ of contact studies related to contact-induced language change and consequent formation of varieties of language (in this case formation of the regional variety of Italian) under the influence of another language (in this case a local dialect).
RI is a geographical variety of the standard in which several innovations and ‘simplifications4’ of Italian are represented, as are “fossilized dialect interferences” (Cerruti, 2011, 15). This paper treats a variety of Italian with only ‘selected’ traits transferred from one variety to another (‘replica constructions’) as a consequence of the process of “substratum interference” (Cerruti, 2011, 13). One of my main goals is to discover which mechanisms are involved in the transfer process from one variety to another. During my search for insight into common linguistic mechanisms, I also turned to literature on creole and pidgin formation. In fact, my assumption echoes Winford’s statement that “the processes of change observed in creole formation are not different in kind from those found in other cases of language contact where one language is dominant over the other” (Winford, 2008).
Most authors agree on the extremes of the continuum or gradatum, but as Berruto underlines “there appears to be a remarkable amount of uncertainty concerning its intermediate zone” (1989, 11). Over the two decades separating us from Berruto’s (1989) paper, much research has been conducted (see Cerruti, 2011). We can confirm that RI not only adopts the direct dialect heritage, but, as a result of ‘grammatical replication’, also other features (reanalyzed features, neo-standard features and so on). “Grammatical replication’ is a process whereby speakers of a language – the replica language – create a new grammatical structure on the model of a structure of another language – the model language” (Heine and Kuteva, 2005). I intend to use a theoretical framework based on the findings of Winford (2008), Heine and Kuteva (2005), and Matras and Sakel (2007).
“In any scheme we should first recognize the existence of regional Italian (or even a standard regional Italian: we have, in fact, a series of ‘regional standards’ constituting social norms within each region) as opposed to SI, the latter existing in scarcely more than a theoretical sense” (Berruto, 1989, 11). To illustrate the coexistence and contact of Italian with Salentino, I will apply the terms proposed by Heine and Kuteva (2005), that is, ‘model language’ for dialect and ‘replica language’ for RI. I avoid using terms such as ‘dominant language’ for Italian, in line with Winford when he states, “it is simply not the case that a socially dominant language is necessarily the linguistically dominant for the speaker who initiates contact-induced change” (Winford, 2008, 126). In the case of Salento, the hierarchy of languages is not so easy to reconstruct due to its situational dependency (we cannot affirm that Italian is always considered the most prestigious variety).
2.3 Historical account of the Italian linguistic situation: internal migration
The Italian situation is characterized by a relatively new bilingualism due to the recent co-existence of two languages (local dialect and Italian language) in the speakers’ repertoire. It is fair to assume that the escalation of national language use was spurred by the massive northbound internal migration. “The so-called internal migration, that is, the moving, especially in the period after World War II, of millions of people from all over Italy toward the northwestern area known as the ‘industrial triangle’”(Berruto, 1989, 13). This process corresponded with the state policy for the promotion of education and the elimination of illiteracy. This process was very important in Salento as there was only an outward-bound migration. This permitted the region to maintain a rural and marginal linguistic landscape and not lose its dialect as a primary means of communication (as was the case of most southern Italian areas, Dal Negro and Vietti, 2011, 73).
In the 50s and 60s, many young Italians learned the national language in school, while the only language spoken out of school was the local dialect. The following generations were raised understanding the necessity of the national language – notions of which had already been passed down by their parents. The first generation of varieties induced by contact between dialects and the Italian language included those spoken by immigrants from areas of the south of Italy in the industrial areas of the north. The result is something referred to as italiano popolare – a variety of Italian strongly influenced by dialects – defined by Cortelazzo (1972) as a “type of Italian imperfectly acquired by people who have dialect as their L1”. At the present moment, as attested by national census, bilingualism “has considerably increased in the last 20 years, especially at the expense of dialect monolingualism” (Dal Negro and Vietti, 2011, 72). This phenomenon holds true, in varying degrees, for the entire country. In Salento, most are bilingual and use both varieties in everyday life. The tendencies observed over the past few years display an apparent stabilization and harmonious co-existence of the varieties, but, through my observations, it can be concluded that the spoken language, alongside the local dialect, is RI. This observation is valid not only for Salento but also for the rest of the country (see Cerruti, 2011, Dal Negro and Vietti, 2011).
Italiano popolare can be called a transitional variety (being always a variety of Italian) creating a ‘bridge’ between dialects and Italian language. In fact, scholars were ‘expecting’ the general disappearance of local dialects and their substitution by regional varieties. Dal Negro and Vietti attest the two most important processes effecting dialects: language shift scenario due to the reduction of linguistic functions and Italianization (2011, 72-73). Therefore, though not immense, the structural distance between RI and SI remains as “local dialects not necessarily converge towards Italian but also towards dialect koines” (ibid. 78).
In this article, I treat the evolution of the most salient features of the regional variety of Italian, evaluating it alongside the aforementioned italiano popolare. A study of this phenomenon requires perceiving italiano popolare as born from the need to create a variety fit for communication between speakers throughout the country who did not have a shared variety. The following generation – children born from parents who adopted italiano popolare – studied Italian in school, and in this passage we can observe an abnormal language transmission5. The parents of these children were not native speakers of Italian, yet, due to the altered social and cultural context, they were obliged to speak Italian and recognize it as the socially dominant language. These conditions did not exactly stimulate the development of the pidginized variety, rather, they instigated its so-called ‘de-creolization’. I use this term to metaphorically illustrate its proximity to Italian, that is, the development of this variety in the direction of Italian. The ‘imperfect’ transmission of Italian was definitely one of the factors that influenced the appearance of new features in RI. Other important factors for the understanding of the current linguistic situation are the “effects of a process of language shift from dialects toward Italian (a process that is still underway)” (Cerruti, 2011, 11).
3. Data Analysis
The data analyzed in this research was collected in Salento between 2007-2009, during my PhD research, through semi-structured interviews which were recorded and then fully transcribed. I interviewed eighteen craftsmen involved in pottery production, with each interview lasting approximately one hour. The questions were related to the production process and techniques adopted by the craftsmen. The focus of the research hinged on a study of how craftsmen describe what they do with their hands and their evaluation of changes in traditional pottery production.
3.1 Change of verb transitivity
The change of subcategorization properties of some verbs, particularly verbs of motion, is one of the most salient and well-known traits characterizing ‘southernness’ throughout Italy. This type of change represents the insertion of the dialectal verb/replication of the dialectal construction to the Italian utterance and the consequent change.
I would like to examine the verb ‘ssire ‘exit’ which is of central interest for dialectologists and linguists working on southern Italy and dealing with the use of monorhematic verbs instead of syntagmatic verbs. This verb changes transitivity properties in Sicily (Amenta, 2007) and in northern Puglia, but not in Salento.
To express the utterance “I bring the dog out [for a walk]” we have the following examples:
Porto fuori il cane – SI.
Bring-1SG out ART-DEF dog
‘I take out the dog’
In this case we have a syntagmatic construction following Talmy’s scheme presenting verb + satellite construction (2000).
Cacciu lu cane – Salentino dialect.
Take out- 1SG ART-DEF dog
‘I take out the dog’
In Salento, contrary to other southern areas, the verb ‘uscire’ is not used transitively, and thus, in this Regional Italian, the regional construction typical of northern Apulia cannot occur:
Esco il cane
Exit-1SG ART-DEF dog
‘I take out the dog’
The omitted personal pronoun ‘I’ plays the role of the subject, and the dog, who is the patient or experiencer of the action, is the direct object. For the RI of Salento, the typical construction would use the verb ‘cacciare’, meaning ‘take out or take away’. This is the case of replication of the dialectal verb, meanwhile, the same verb exists carrying a different meaning in the Italian language. It is interesting to note that some native speakers of the Salentino dialect affirmed that expressions like (2) and (4) are equally grammatical and possible:
sta portu ddra fore lu cane
stay-1SG bring-1SG out ART-def dog
‘I bring out the dog’
This example shows the reconstruction and replication of the SI construction in the dialect, along with the Italian language-influenced change from the monorhematic verb ‘cacciare’, typical of the dialect, to the syntagmatic verb widely used throughout the country, especially in the north (Masini and Jacobini, 2009). The informants who deemed this example both possible and grammatically correct are young and hold high levels of education and mobility. Thus, this may present one of the possible directions of change occurring in the dialect and may be widespread among the younger generations under the influence of SI.
In my data, the verb ‘exit’ is always used in constructions with manipulated agentivity: the direct objects act as Subjects. In the following example (5), changes of the subject in two parts of the utterance can be verified: in the first, ‘pezzi grandi’ is the subject and, in the second, the subject is the omitted pronoun ‘loro’, craftsmen.
Essane strozzi crandi e poi li ‘mpastane – Salentino
Escono pezzi grandi e poi li impastano – SI
Exit-3PL piece-PL big-PL and then they pug-3PL.
‘Big pieces are taken out and then they are pugged’
In this case, it is logically clear that the object it taken out by the craftsmen, but the construction permits two interpretations. The following, similar examples are typical of the Salentino dialect:
su ‘ssuti li cornetti?
‘Did the croissants come out?’
È ‘ssutu lu pane?
‘Did the bread come out?’
These constructions present situations as though the ‘cornetto’ or ‘pane’ themselves were acting and not being acted upon – taken out of the oven – by baker. These are two examples with unclear distributions of agent/patient roles among semantic and grammatical roles. There are various ways of underlining the action in the Salentino dialect. For example, continuing with the example of bread, a question may be:
Lu hai cacciato lu pane?
ART-DEF have-AUX take-PTCP out ART-DEF bread?
‘Did you take out the bread?’
Amenta (2007) also describes this case when she states that, in Sicilian RI, motion verbs used transitively are semantically characterized by a dislocation and pronominalization of the object depending on the transitively-used verb. This is the case of the dislocation of the clitic pronoun to the left.
A ‘classical’ case of transitivity change, equally common in RI, is the substitution of the verb, that is, the use of a monorhematic verb instead of a syntagmatic construction typical of the SI:
Io scendo la valigia – RI
I go down-1SG ART-DEF bag
Scindu la valigia – Salentino
Go down-1SG ART-DEF bag
Io porto giù la valigia – ST. It.
I bring down-1SG ART-DEF bag
The general structure of the sentence remains the same: Agent – Verb – Direct Object The phenomenon observed is the transfer of the more economical dialectal verb to the Italian structure. This is the case when the dialectal verb is monorhematic and is preferred to the dialectal Italian structure using syntagmatic verbs. It can usually be found in semantic couples, such as ‘enter - exit, climb - descend’.For example:
Nchianame la posta - Salentino
Climb-me the mail
Sali la posta - RI
Climb the mail
Porta su la posta – SI
‘Bring up the mail’
In these cases, the speaker uses a more ‘suitable’ verb with a corresponding translation in the Italian language, but the semantic and syntactic characteristics remain those of the model language. This is one of the examples where bilingual speakers attempt ‘to align the structures’ (Matras and Sakel, 2007, 834) and, as a consequence, “the syncretization of processing operations in the two languages, allowing speakers to apply similar mental organization procedures to propositions in both languages of their repertoire” (ibid, 835) occurs. It is perhaps appropriate to adhere to the idea expressed by Bolonyai (1998), and mentioned in Matras and Sakel (2007, 849), that bilingual speakers, when changing the structure of one of the languages, use both languages simultaneously. Another explanation of this phenomenon is proposed by Winford (2008, 139) who considers the imposition6 as a “type of cross-linguistic influence in which production processes of a dominant source language are transferred to the production of a recipient language in which speakers are less proficient”.
The task of lemma access, or activation at the level of the Formulator, is to “select appropriate items and integrate them into a syntactic structure or grammatical form” (Bierwisch and Schreuder, 1993, 26 cited by Winford, 2008). When speakers need to express newly acquired or enlarged syntactic structures “they tend to fall back on the lemmas associated with semantically equivalent items in their L1, to supply the information necessary for them to initiate more complex syntactic procedures” (Winford, 2008, 139). In our example of RI, we can see that the speakers take lemmas from the dialect and apply them in actual Italian – a detail with a historical background7. These verb constructions in RI maintained properties of the dialectal verbs, and this choice may be accounted for by factors such as sentence economy and ‘simplification’ of the structure in RI compared with the Italian sentence. The example quoted by Winford (2008, 140) explains that in such a situation, “the subcategorization properties of substrate motion or transfer verbs […] are imposed on superstrate lexical items”.
I would like to refer to an example, from a study by Silva-Corvalan (1998) on LA Spanish speakers, which falls into the category referred to by Winford as the ‘situation of attrition’, where speakers “transfer the argument structure and subcategorization properties from an L2, which have become dominant, over an L1”.
The example is as follows:
LA SpanishYo gusto eso
I like-1s that
Gen SpanishA mi me gusta eso.
To me pro please-3s that
‘I like that’
Gen Spanish:Experiencer – Experience – Theme
Ind. Object – Verb - Subject
LA Spanish:Experiencer – Experience – Theme
Subject - Verb - Object.
In the case of RI in Salento, we are not dealing with language attrition, but with the introduction of the change (imposition or replication of L1 lemma to L2 item) and its gradual diffusion and acceptance among speakers.
3.2 Polyvalent che ‘that’
This feature is typical of informal spoken Italian and popular Italian, as defined by Berruto (1987). In the first written record of this new informal feature, it was described mainly as a general complementizer introducing several subordinates (mostly causative and explicative) and was recurrent in non-educated and non-observed speech. Currently, there is an extension of the uses of this feature and its growing stabilization in the language, not only of Southern Italy, but also throughout the entire country. This feature is widespread and, in Italian linguistics, is usually referred to as a feature of neo-standard variety8. It is one of the most diffused features throughout the country, though not directly related to the influence of a particular dialect, thus bearing no regional connotation. Due to its simplification of a complex system of subordinate clauses, this feature continues to spread rapidly.
It is widely used in place of the explicative conjunction ‘perché’ (because): the word drops the first, not the final syllable. In this example, che introduces the argumental (completive) clause depending on the nominal head ‘necessità’ with ad sensum agreement:
erano le necessità di quel periodo che l’uomo potesse
be-PST- PL ART-PL necessities of-GEN that period that ART-DEF man could-SBJV
avere degli strumenti vari - RI
have ART-INDEF instruments various
‘it was the necessities of the period why the man could have different tools’
In constructions of this kind, we are dealing with the desemanticization of the form and extension of its grammatical meaning to the general complementizer. This is one of the parameters of contact-induced grammaticalization, including desemanticization, decategorization, and erosion (Heine and Kuteva, 2005, 15). In my corpus, there are examples of the usage of ‘che’ in place of the indirect pronoun ‘in cui’, acting as an invariable complementizer:
dei giorni che non lavora - RI art-INDEF-PL days that not work-3SG
Giurni ca quiddru fatica – Salentino
Day-PL that he works-3SG
Giorni in cui non lavora – ST. It.
Day-PL in which not work-3SG
‘Days when he does not work’
It may be affirmed that, in Salentino, this role of the general complementizer is also reinforced by the general use of ‘ca’ which introduces different types of subordinates. The extension of the complementizer function also applies to the types of clauses not documented in italiano popolare. The mechanism is analogous to the verbs described in the previous paragraph. Speakers identified parallel constructions in the two languages and then extended the uses of the complementizer in Italian to those possible in Salentino. As I previously mentioned, the parent-to-child transmission of the imperfectly acquired variety adopted the more ‘comfortable’ construction. This widely occurs with the verb ‘to be’ used as a focalization, and the ellipsis of the introductive part of the clause in the following examples can be observed: È ovvio, È chiaro it is obvious; credo, penso I believe, I think:
che c’è una bella differenza tra il dialetto quello leccese,
that there’s a beautiful difference between ART-DEF dialect that leccese
diciamo, e quello cutrofianese – RI.
say, and that cutrofianese.
‘There is a big difference between the dialect from Lecce and, let’s say, that from Cutrfofiano’.
As in many other cases ‘che’ is used in the beginning of the clause without any particular syntactic or semantic meaning, thus it is sometimes used as a desemanticized particle and no longer as a conjunction:
Che ho diversi forni, e alla fine vorrei usare soltanto quello - RI
that have-1SG different oven-PL and finally want-SBJV use only that
‘I have different ovens but I would like to use only that one’.
There are cases when ‘che’ is used to introduce some comment or specification to the information that was given in the first part of the clause:
Sbattere l’argilla, sbattere che è ovviamente il movimento.
Bang art-DEF clay, bang that be-3SG obviously art-DEF movement
‘Bang the clay is obviously a movement’
In this case, the first clause is commented on by the speaker explaining the verb. The explanation presents a cleft construction with the verb put in the first position and connected to the rest of clause by ‘che’. Here, ‘che’ functions as a connector between the verb and the rest of the clause containing an irregular word order. It is possible to reconstruct this clause in the following manner:
Sbattere l’argilla, sbattere ovviamente è il movimento.
‘Bang the clay, bang obviously is a movement’
The insertion of ‘che’ in the clause serves to focus attention on the pragmatic level and is used within the specification structures. ‘Che’ as a clause constituent does not bear any syntactic or semantic functions: it can be eliminated, as is shown in the reconstructed clause, and is fully desemanticized, both as a pronoun and as a complementizer. Here, ‘che’ is not used as a conjunction linking two clauses – one which can be called ‘main’ and the second ‘dependent’ on the first –, typical of the functions of ‘che’ as a subordinating conjunction. Rather, a non-standard use of ‘che’ is presented, shifting in its syntactic and semantic functions. I assume this change to be due to processes of grammaticalization, in particular, the bleaching of the semantic meaning of a functional word and the extension of its grammatical functions within the clause. This may also be observed in the following example:
Sono sull’autobus che sto arrivando
Be-1sg on ART-DEF bus that stay-1sg arrive-GER
“I am on the bus and I am arriving”
In this sentence (16), both verbs express parallel actions: one is progressive, the other is static, and though ‘che’ roughly presents a consecutive-explicative function, there is no clear relation of subordination between them. ‘Che’ functions as a connector introducing a clause with a generic eventive value. The use of ‘che’ here may also be explained as a tool for focusing attention, thus it predominantly holds a pragmatic rather than a syntactical or semantic meaning.
Summarizing the analysis of polyvalent ‘che’ uses in my corpus, I conclude that this conjunction functions mainly as a desemanticized marker of syntactic relationship between clauses and as a multifunctional particle within the clause. Therefore, the tendency is the extension of its use in a more generalized manner, as well as its acquisition of new functions. It is neither socially limited nor geographically detectable, and represents one of the main innovations currently identifiable in the spoken language. This feature may occur in speech for different reasons. The diffusion of this feature can also be explained by the social factor of prestige: it is oftentimes heard in informal speech on television and radio. Another advantage of this feature is its (growing) suitability for any type of construction. As shown, it is no longer restricted in speech by a concrete type of subordinate clause, but is extended to different types of subordination, greatly simplifying the construction of these. This change may be explained by the ongoing process of grammaticalization.
Through my data, I found new uses of the gerund. In some utterances, speakers replace embedded clauses with a gerund. Constructions expressing habitual action and requiring verbs in the indicative mood, or expressions not typical of the local dialect, can be performed in the RI of Salento with the help of a gerund. The following example shows how a causal clause is achieved through the use of a gerund:
Questo lo immergevano e si riempiva, e poi chiudevano,
This it immersed-3PL and filled up-3SG, and then closed-IPRV-3pl
e poi mancando la pressione non scendeva.
and then lack-GER art-DEF pressure not descend-IPRV-3SG
‘It was immersed and filled up, then they closed it and since/as the pressure was lacking, it was not descending’.
The pragmatic construction is not regular, but the operation of replacement allows the speaker to avoid a subordinate clause and to thus proceed with a linear phrase formation without creating a hierarchical relation within the clause.The next example shows how the gerund is used to substitute an expression not typical of the Salentino dialect (col passar delle ore with the passing of time):
Più andava, più si buttava passando delle ore – RI
More go-IPRV more throw-IPRV-3sg pass-GER art-indef hours
Più andava (il forno), più si buttava (la legna) col passar delle ore – ST. It.
More was going (oven), more we threw (firewood) with-art-def pass of-art-indef-GEN hours
‘More the oven was going, more wood was thrown, while the time was passing’
Another use of the gerund typical of the Salento area is in the expression of a future perfective action. In the following example, the Italian periphrastic construction stare + gerund acquires a perfective aspect and no longer represents, as in SI, an atelic action. This use is related to the interference of a local dialect and the formation of the present and future using the construction stare-1SG + verb-1SG in the Salentino dialect:
Sabato sto andando al cinema – RI
Saturday stay-1SG go-GER to the cinema
Sabato sta bau a lu cinema – Salentino
Saturday stay-1sg go-1sg to the cinema
Sabato andrò al cinema - SI
‘Saturday I will go to the cinema’
In brief, it can be concluded that the gerund in the RI of Salento has several functions in the sentence: the replacement of subordinate clauses for the preservation of linear construction of the utterance, the substitution of the indicative mood for future perfective actions, and the substitution of expressions lacking in Salentino.
4.1 Italiano popolare – regional Italian, comparative approach
The first studies of italiano popolare, by De Mauro (1970) advancing his thesis on its uniform character throughout the entire territory of Italy, were based on written texts, and consequently, the drawback of these studies lies in their neglect of the spoken language. This is all the more serious since italiano popolare is mainly a spoken variety and, hence, phonetics and prosody are of an elevated importance, contributing to a specific type of italiano popolare for each region. Even syntactic change is conditioned by the structures of the model language, as shown in the previous examples. Thus, replication is different in each region or macro-region, consequently reflected in the regional character of RI (for a comment on this see p.766, “Lexikon der Romanistischen Linguistik (LRL), volume IV). Table 1 uses the description of italiano popolare from LRL.
These features of italiano popolare display a general reduction of morphology and a shift of the functional charge to the word order and lexicon (Berretta, 768 in LRL, 1988). In fact, some authors recognize elements of pidgins and creoles (for example, Mioni, 1983) in these types of ‘simplifications’. Though obviously just an analogy, this can assist the understanding of the entire phenomenon, and particularly, additional trends of the development of RI as a ‘consequence’ of italiano popolare. What can be seen is a diversification of non-standard features, divisible in three main categories: those already described as typical features of italiano popolare, reanalyzed features of italiano popolare (which have acquired new uses and new dimensions over recent years), and the new forms, not yet described and representing a new formation.
In figure 1, the speakers are organized in decreasing order based on the quantity of non-standard constructions they used in the interviews. I counted the variables rather than the frequency of their uses. In total there were eighteen variables, which I divided in three groups:
Blue - features of italiano popolare;
Red - features descending from italiano popolare which had undergone modifications, especially as extensions and changes;
Yellow - formations and constructions new to the Italian language.
The two speakers who used the highest number of non-standard features belong to the same age group – 52 and 58 –, but their education level is considerably different. In the middle, we find three speakers with the same results: they used twelve of the twenty-three non-standard features considered in this research. Their socio-demographic characteristics vary: their relative ages were 27, 69 and 57, with the youngest and the oldest craftsmen representing the highest and lowest levels of education – university degree and fifth grade, respectively.
Also, the lowest quantity of non-standard features was produced by those with a relatively low level of education, which leads to the conclusion that there is no direct relationship between the level of education, age, and the quantity of non-standard features used in speech. Thus, these variables are not socially dependent. I view this data as a sign of the changes occurring in the spoken norms of the Italian language, as well as the transformations, either in progress or already introduced, in the RI.
From figure 1, it can be inferred that features of italiano popolare, as well as their extensions and transformations, are predominantly used. The latter present changes since the first written records of this variety. Though both the corpus and the quantity of features examined are limited, they are nonetheless highly demonstrative. A constant use of non-standard features, with a relatively moderate variation within the group, can be noted and may be accounted for by the status of RI in the linguistic repertoire. This is the variety acquired by the speakers and it is how the Italian language appears in Salento. Used on a daily basis and transmitted from parent to child, its features are stable in the speech of all of its speakers. In this context, the prestige of a written and imposed language is lower than that of a local variety spoken by the resident population. The tendencies of the changes in Salento can be described as a nearing of the standard without quite approaching it. The structural distance established by the close contact with the dialects remains.
I can therefore conclude that RI is a variety of Italian that is both dialect- and Italian-adjacent, being always a variety of Italian. This variety is formed as a consequence of contact-induced change and the parallel, but not balanced, co-existence of two linguistic systems. RI can be defined as a diatopic (geographical) variety of Italian, as well as a secondary or tertiary dialect in Coserian terms.
Since dialect speakers and speakers of italiano popolare aim for the superstrate and strive to learn the standard variety (in the process of acquisition of the national variety) – widely shared by the population – the formation of RI should be viewed from the contact perspective. However, the general tendency is not the use of a superstrate and a moving away from the dialect, but rather the active application of mechanisms that Siegel refers to as transfer, mixing, leveling, and further simplification (Winford, 1997, 14). Table 2 is the approximate model of the generational scheme which summarizes the presented interpretation of the transmission of RI.
During the first stage of the substantial acquisition of Italian by adults throughout the country, most of the subjects had dialect as their L1 and Italian as their L2. Not having mastered Italian, they relied on the use of corresponding lemmas for some syntactic structures. This happened at the formative stage of italiano popolare, when L2 Italian speakers identified parallel structures in both languages and projected some constructions of the model language onto the replica language. Parents transmitted structures, crystallized in their varieties, to their children9, the use of which, for a number of factors, could no longer be justified as incompetence in Italian. First of all, either Italian had become the L1 of the children or they were bilingual. Secondly, these uses were caused by the transmission of stabilized structures remaining after what I metaphorically term ‘de-creolization’. And finally, italiano popolare was brought nearer to the standard variety through nativization. Winford (2008, 139) claims that “if they have continuous and adequate access to L2, they gradually acquire the relevant lemma information and learn how to reproduce syntactic procedures”. This statement is true for the situation of the SLA. However, in our case, the acquisition presented ruptures and, in particular, the L2 was gradually becoming the L1, within which some syntactic features, which had remained stable solid structures, were rapidly incorporated.
Rapid nativization is a factor exerting a great influence on the formation of the new variety and its characteristics. As a process, nativization provides the possibility for the emergence of innovations in speech arising from children. In our case, innovation, was ‘proposed’ by adults who aspired to adapt to the new linguistic situation, whereas the children accepted it, brought it through the phase of nativization, and stabilized it in their L1. According to Siegel, “in nativization, the most common variants are the ones acquired by the children” (1997, 129). Thus, in Siegel’s view, (1997, 138) the substrate influence normally consists of transfer, performed by adults and then reinforcement, completed, by children.
I consider abnormal transmission to be one of the major factors affecting the current linguistic situation: the formation of italiano popolare and its influence on RI in the form of indirect transmission of the language between generations. Siegel affirms that the stages corresponding to the development of contact varieties are due, firstly, to the creation of the L2 variety of the superstrate, secondly, to the stage of leveling, and thirdly, to the stage of vernacularization (Winford, 1997, 14). In the case in question, this model can be used as it explains the stages of the creation of italiano popolare as a result of contact, followed by its leveling and the further vernacularization of Italian, and, as a consequence, the development of RI. I am not dealing with the entire process, but only focusing on the development involving selected features of the variety, a partial vernacularization expressed in replica constructions; “… a stage in which adults of various language backgrounds create second language varieties of the superstrate, and later, at the stage of leveling10, when vernacularization occurs. Substrate transfer occurs at the first stage, with reinforcement taking place at later stages. The vernacularization stages involve koineization, that is, mixing, leveling and simplification” (Siegel, 1997).
This study also reveals that there is no direct relationship between the demographic characteristics of the speaker and the quantity of non-standard features used by him in speech (see figure 1). This can be accounted for by the status of RI in the linguistic repertoire: it is a variety spoken by the majority of speakers in the region (at times, the only one available as a spoken variety), and the features I defined as ‘non-standard’ (based on the rules of the SI) are not actually non-standard for the RI of Salento, but are stabilized features of the variety. Years of language contact and a changing linguistic situation have brought about a reanalysis of the constructions and its adaptation to the speaker’s uses and needs. The most frequently used features were grasped by the population, took root and, thus, were no longer recognized as non-standard. In light of the conclusions drawn in this article, I would like to discuss the scheme proposed by Auer (2005) with respect to the situation in Southern Italy (figure 2).
Following the scheme proposed by Auer (2005: 22), it is reasonable to state that the present situation in Salento is similar to diaglossia, or repertoire Type C. This type of repertoire is characterized by the ‘intermediate variants between standard and (base) dialects’ (2005: 22). Auer speaks of leveling and advergence to the standard as key ‘moves’ within the repertoire. Furthermore, other processes are quite frequent, such as step-by-step dialectisation and standardization: hybridization, simplification, and innovation (Auer, 2005: 25). The changes described in this paper are mainly expansion, desemanticization and grammatical replication.
In concurrence with these kinds of changes, I would like to refer to the work of Nilsson and Svahn (2009), claiming that, in West Sweden, the two extremes of the continuum have disappeared and what remains are regiolects or dialect. This study is very interesting and useful for the observation of ongoing changes in other parts of Europe and as a suggestion of a possible approach to the creation of a scheme of the repertoire. Nilsson and Svahn (2009) arrived at the conclusion that nobody speaks either a pure dialect or a pure standard – a similar situation to that of Italy: of all the speakers I interviewed, none showed complete absence of non-standard features. What can be observed, both in Italy and Sweden, is a large intra- and inter-speaker variation.
Nilsson and Svahn’s (2009) conclusion concerning the repertoire in Western Sweden is that the dialect change in progress has reduced the repertoire (without the standard and base dialect). However, considering that new variants are possible, the repertoire is not as small as it may first appear. Their scheme for the Swedish dialect situation is as in figure 3.
To adapt this scheme to the Italian situation it must be divided in two separate poles (Italian and dialect), while including the intermediate variety (RI) between them (see figure 4).
This scheme illustrates the Italian repertoire as having two main constituents: the Italian language and a dialectal system. The intermediate variety is the connecting nexus influenced by and influencing the superstrate and the substrate. The process of formation of RI is certainly very complex. It first involves the acquisition of the second language (stage one – acquisition of italiano popolare) with the imposition of features typical of the L1 onto the newly acquired language, and then the formation of the contact variety and its contact-induced change and stabilization (italiano popolare > RI). This paper does not wish to answer all the questions related to RI, but it undoubtedly constitutes a first step in the analysis of the Italian linguistic situation in Salento from the viewpoint of contact linguistics and mechanisms of contact-induced change.
An earlier version of this paper was presented on the occasion of ICLaVe6 – held in Freiburg and organized by FRIAS. In particular, I would like to thank prof. Peter Auer and all the participants of this conference for their useful comments. I am grateful to Jenny Nilsson, Maria Mazzoli and prof. Immacolata Tempesta, who read and commented on this paper during its elaboration. Furthermore, two anonymous reviewers provided useful and precise comments which allowed me to develop and improve my paper. All remaining errors and imperfections are mine.
Nilsson Jenny and Svahn Margareta. 2009. What can the dialect situation in West Sweden tell us about the concept of dialect and dialect change? Paper presented at the 5th International Conference of Language Variation in Europe Copenhagen 25-27 June 2009.
1 RI being a variety of Italian.
2 Salento, as demonstrated through the research “Lingua nazionale e dialetto in Italia all’inizio del Terzo Millenio”, represents a repertoire with “l’italiano da un lato, il dialetto dall’altro, e un’enorme area mediana, con ampia e pervasive presenza di fatti intermedi riconducibili a varietà di italiano dialettizzate e a varietà di dialetto italianizzate. In un certo senso, repertorio tripartito” (Berruto, 2006, 11). In this paper, I will not distinguish between standard RI (Berruto, 1987), ‘folk’ RI and ‘folk’ Italian (Cerruti, 2011).
3 See Berruto (2003) regarding the notion of ‘native speaker’ in Italian, as well as the differing opinions on grammaticality of some constructions.
4 I place the term simplification in quotation marks in order to convey its relative significance, not to express the ‘wrong or deviant construction’, but a new construction ‘born’ from the contact of two varieties.
5 Thomason calls normal transmission ‘complete and successful transmissions, by native speakers to child or adult learners, of an entire language, i.e., a complex interlocking set of phonological, morphological, syntactic, semantic, and lexical systems (Thomason, 1997, 74).
6 Winford proposes the use of the single term ‘imposition’ for the cross-linguistic influences previously referred to as ‘substratum influence’, ‘interference via shift’, ‘transfer’ and ‘convergence’.
7 This is one of the central features of the process of generational transmission of language.
8 Often heard on informal radio and television programs, social diffusion confers prestige to this feature, while others, such as the previously discussed change of transitivity, are highly stigmatized because they are typical of southern areas.
9 Transmission and nativization, i.e. acquisition of native speakers, is an important step since. As Mufwene (2007) notes, this is the step in which the vernacularization of the creole occurs; the phase in which it becomes the primary means of communication, as it did for the second generation of speakers – children of speakers of italiano popolare.
10 Thomason defines leveling as ‘a process by which dialect differences are eliminated through replacement of one’s dialect’s features by those of another dialect’ (1997, 85).
See Berruto (2003) regarding the notion of ‘native speaker’ in Italian as well as the differing opinions on grammaticality of some constructions.