Abstract

ENG: English Tag Questions (henceforth TQ) are a complex linguistic phenomenon. This complexity derives from the fact that their meaning/function is determined by the interplay of different linguistic levels, namely syntax, pragmatics and phonetics – i.e. intonation. Besides, the formation of tags often undergoes various exceptions while speaking, so that on the basis of some sociological parameters, namely diatopic varieties, social class and cultural background, speakers use non-standard variants of tags, such as innit – mainly identified with London area – or other forms presenting a lack of concordance in number between operator and subject, such as weren’t it?. It has been observed that recent movies are characterised by quite a faithful reproduction of spontaneous face-to-face conversation (Taylor, 1999), since linguistic realism is necessary to attract the audience’s attention (Pavesi, 2005: 30), and as a consequence, many features of orality appear in films – i.e. hesitations, overlapping, fillers, ellipsis and also tag questions. In this paper, I would like to focus on non-standard variants of canonical tags, such as innit – functioning either as a proper TQ or as an invariant tag (Andersen, 2001) – and tags of the type of weren’t it, in some British movies, where much space is devoted to conversation in everyday language, namely Secrets and Lies (1996, M. Leigh), The Full Monty (1997, P. Cattaneo), Bend it like Beckham (2002, G. Chada) and Green Street (2005, L. Alexander). The predominant spoken variety is British English, even if other varieties are present, such as American, Scottish and Northern English spoken in Sheffield. Moreover, language also varies according to the social class to which the characters belong, ranging from the upper to the working class, and the ethnic group of provenance – i.e. see the presence of Indian and black characters. Indeed, generally non-standard tags are used by specific groups of characters, who belong to the working class or to determinate ethnic groups. Therefore, I will analyse non-standard variants on the basis of diastratic and diatopic differences. Since Italian does not display such varied and complex set of forms as English does, it is often very difficult to translate TQs into Italian. For this reason, I will compare the original versions of these movies with their dubbed versions in order to pinpoint how and whether these types of non-standard tags are translated into Italian, whether their indexical function is preserved and which strategies are employed to produce the same effect of non-standard forms.

Keywords

ENG.: tag questions, non-standard features, innit, ain’t, S-V disagreement, dubbing, pragmatic functions, English, Italian.

 

1. Introduction

The translation of dialect in multimedia contexts such as dubbing and subtitling is a theme of great interest. A linguistic phenomenon such as tag questions in English is worth studying from the perspective of translation, not only because the Italian language does not seem to have formal or functional equivalents, but also because of the complexity of tags. More specifically, various factors contribute to determine their meaning/function – i.e. paralanguage, prosodic features, etc. However, the task of transposing such forms into Italian is even more challenging in the case of Non-Standard tags, in which the main difficulty is to render the sociolinguistic aspects that are attached to them – i.e. the speaker’s cultural background, education, social status, ethnicity, etc. For this reason, the present paper deals with the translation of this specific type of English tags – i.e. Non-Standard tags – in Italian dubbing, mainly because film is a polysemiotic text in which all these factors come into play. This study is based on a small corpus of data, which includes four films (cf. infra 4.1.).

2. Tags in English: Types and Functions

The English language displays a complex system of tag questions (henceforth TQs), which can be generally defined as a declarative sentence to which a shortened form of question is appended (Bonsignori, 2007a: 6). As a matter of fact, TQs can be divided into two groups: tag questions stricto sensu and invariant tags.  The difference between these two types is mainly syntactic, in the sense that the former are created according to certain syntactic features that characterise the main sentence – or ‘host clause’ (Cattell, 1972) – to which the tag is appended and comply with the polarity requisite – i.e. TQs in the strict sense are inflectional – while the latter are completely independent from the main sentence. More specifically, invariant forms are expressed by single lexical items, which can be adverbial or interjectional particles such as right, ok, yeah, no, and eh (Algeo, 1988: 174), as can be seen in the examples below:

(1) Roxanne: So you work with mum, right? (Secrets and Lies)
(2) Mrs Paxton: I bet your room at home doesn’t look like this, eh? (Bend it like Beckham)

Conversely, the formation of TQs stricto sensu is governed by precise rules that concern the operator and the subject in relation to the main sentence (Quirk et al., 1988: 810). As a consequence, we can have either reversed polarity tags (Huddleston, 1970; Hudson, 1975; Aijmer, 1979), which are characterised by the inversion of polarity between the main sentence and the tag – i.e. a condition of grammatical well-formedness in generative-transformational studies (Akmajian and Heny, 1975: 203) – as in (3), or positive constant polarity tags [url=#n1][1][/url]  (Kimps, 2007), in which the main sentence and the tag share the same positive polarity, as in (4):

(3) Monica: It’s not the same thing, is it. (Secrets and Lies)
(4) Cynthia: What? Your mum and dad told you, did they? (Secrets and Lies)

However, the complexity of tags does not lie only in their analytical structure, but it also relates to the determination of their functions. More specifically, the meaning/function of TQs is determined by the interplay of different linguistic levels, namely syntax, pragmatics and phonetics – i.e. intonation. In fact, several factors intervene to build up the meaning of a tagged utterance: polarity on the syntactic level; the situational context of use – i.e. the speech event, the relationship between the participants, their social role, shared knowledge – the speaker’s communicative intentions and the illocutionary force, on the pragmatic level; and finally, intonation. In this sense, the same tagged utterance can be characterised by different prosodic patterns – i.e. it can have either a rising or a falling intonational contour on the tag portion – thus performing different functions (Quirk et al., 1985: 811).

The complex character of tags is also due to their multifunctionality (Bonsignori, 2007b).  Multifunctionality is here meant in two senses. On the one hand, tags can perform a different range of functions, so that even opposite functions may correspond to the same syntactic structure. More specifically, adopting Holmes (1982) and Algeo (1988, 1990, 2006) as reference models, tags are used as a strategy to turn a simple assertion into a real question, inviting the hearer to validate the truth of a certain proposition – i.e. informational function – or just to ask for confirmation – i.e. confirmatory function; or as a strategy for turn-allocation, thus inviting the hearer to participate in the conversation – i.e. facilitative function – or as a strategy to reduce the strength of an utterance that may appear threatening or disagreeable to the addressee, rendering it more acceptable and not offensive – i.e. softening function – and thus complying with politeness requirements. But conversely, there are specific situational contexts in which TQs are used with an opposite function, that is as ‘aggravators’, especially when appended to certain statements or imperative constructions, thus contributing to strengthen their illocutionary force and producing aggressive overtones – i.e. challenging function, also specifiable as peremptory and antagonistic function (Algeo, 2006). On the other hand, the multifunctional nature of tags also relates to the fact that different functions may coexist in the same tagged utterance (Holmes, 1982: 47; Eckert and McConnell-Ginet, 2003: 170). For instance, the facilitative and confirmatory functions may often overlap. 

3. Tags and Sociolinguistic variation: The case of Non-Standard Tags

In the previous section a general overview of the possible types of tags, together with the functions they can perform, was outlined. However, the formation of tags often undergoes various exceptions while speaking, so that, on the basis of certain sociological parameters, speakers also use non-standard variants of tags, which are the focus of the present study. Sociolinguistic variation can be observed in diatopic, diaphasic and diastratic varieties and more specifically it relates respectively to regional varieties (including the varieties of English and accents) register, and finally social class, age, gender, cultural background and ethnicity. In this work, three different types of non-standard tags are analysed, namely innit, ain’t and weren’t it?. In the following paragraphs, a brief description of each non-standard tag is outlined.

The most interesting research on the form innit is by Andersen (2001), who states that it is a ‘highly noticeable feature of the London teenage vernacular and is one of the most outstanding elements of non-standard grammar to be found in the COLT[url=#n2][2][/url]  corpus’ (2001: 105). Andersen claims that innit has two possible origins: 1) it may derive from the standard English form isn’t it?, which has then undergone a phonological reduction; or 2) it may have developed by the non-standard verb ain’t (2001: 116). Moreover, it seems that the major users are the various ethnic minorities – i.e. West-Indians (Wright, 1981) and Jamaicans (Hewitt, 1986) – who live in London, who then have influenced the standard variety. From a syntactic point of view, innit can be used either as an invariant tag, regardless of the grammatical features of the main sentence, as in example (5), or as a real tag question, being the subject of the main clause it and the verb be in the present tense, as in example (6) below:

(5) Pinky: (on the phone) Yeah? Mum! No, no, nothing! I’m just at work, innit! Yeah, I know Polly’s coming to do the suits. So Jess can get the tube, innit! Yeah, all right, all right, I’ll pick her up. (Bend it like Beckham)
(6) Mel: It’s a bit backward, innit! (Bend it like Beckham)

As can be noticed from the examples above, innit generally occurs when the polarity of the statement to which it is appended is positive, no matter if when it is used as an invariant tag it undergoes ‘invariabilisation’ (Andersen, 2001), from a restricted to a more generalised use, therefore losing the semantic features that characterise its original form isn’t it.
Another widespread feature of English dialects is the use of ain’t[url=#n3][3][/url]  to negate be – both as a copula and auxiliary – and have with any subject. This non-standard contraction undergoes a levelling process which avoids the subject-verb agreement (Hudson, 2000) and it is somewhat more current in American English than in British English (Quirk et el., 1985: 129), even if it is rapidly becoming more and more common in British English (Algeo, 2006: 21). Actually, ain’t does not occur in Irish or Scottish English, but is rather part of the traditional dialect system of the Southeast of England (Anderwald, 2004: 186). In fact, its use and occurrence has been treated by Cheshire (1981) with reference to a specific area in England, namely the town of Reading – i.e. West of London – where ain’t is frequently used in the formation of tag questions, as in the following examples:

(7) I’m going out with my bird now, ain’t I?
(8) Cynthia: You’ve got a bed-sit, ain’t you, Paul? (Secrets and Lies)

Finally, the last type of non-standard tags considered in this study refers to those cases in which the verb in the tag is different in person and number from its subject, thus not complying with the rule of subject-verb agreement (Hudson, 1999), which can therefore be labelled as cases of subject-verb disagreement, as in examples (9) and (10):

(9) Dave: That were our Jean, weren’t it. (The Full Monty)
(10) Cynthia: You’realways too busy, isn’t you! (Secrets and Lies)
Such variations on subject-verb agreement actually characterise a large number of non-standard dialects of English (Trudgill, 1999).

4. Non-Standard Tags in Dubbing

In the previous section, non-standard variants of tags have been described. In the present section, I wish to analyse English non-standard tags from the perspective of a specific type of audiovisual translation – i.e. dubbing – in order to verify how and whether these non-standard forms are transposed in Italian. Recently it has been observed that, even though film language is a fictional spoken variety, it nonetheless attempts to portray everyday conversation, in order to draw the audiences into the film’s narrative (Pavesi, 2005). As a matter of fact, film language in recent films shares many features of spontaneous conversation (Taylor, 1999), thus mirroring – in Halliday’s terms (1985) – the so-called ideational, interpersonal and textual metafunctions of real life. Some examples are the use of hesitations, false starts, overlapping, jargon, slang, colloquial expressions and, of course, tag questions as well as non-standard tags.

The translation of English tags into Italian poses various problems. First of all, there is a linguistic difficulty in the transposition of a syntactic construction which is an inner and characterising feature of the English language and which conversely does not belong to Italian. In fact, English has proved to have a complex, organized system of TQs (cf. 2), while Italian has neither functional nor formal equivalents, but can count on a wide range of different solutions, such as the use of simple expressions, like no?, interjections, like eh?, and conversational routines of the type of è vero/non è vero? (Chiaro, 2000; Bonsignori, 2007b). The range of possible translation equivalents is quite wide, since in Italian there is no such unified phenomenon as in English.

In addition, in the translation of non-standard tags there is also a socio-cultural difficulty, since in the source language these forms express a certain social and cultural meaning, which hardly ever finds a perfect correspondence in the target language. Moreover, dubbing itself is an extremely complex process, because it deals with audiovisual texts and is therefore characterised by a set of predetermined aspects that cannot be modified, such as setting, gestures, facial expressions, sounds and the like. Furthermore, the linguistic code undergoes a series of constraints that are linked to the visual ones (Pavesi, 2005: 12) – i.e. synchronism.
For the purposes of the present paper, I have analysed four films of British production where much space is devoted to everyday conversation, namely Secrets and Lies (1996, M. Leigh), The Full Monty (1997, P. Cattaneo), Bend it like Beckham (2002, G. Chada) and Green Street Hooligans (2005, L. Alexander), comparing the original versions[url=#n4][4][/url] of these films with their dubbed versions in Italian, in order to pinpoint how and whether non-standard tags are translated in the target language, whether their indexical function is preserved and which strategies are employed to produce the same effect of non-standard forms. Before investigating the topic, a general overview of the non-standard tags in these films is in order.

4.1. The data

The chosen films for this analysis are of British production, so the predominant spoken variety is clearly British English. However, language varies diatopically, for instance there are characters who speak American English, such as Matt and Shannon in Green Street Hooligans, or Irish English – Joe in Bend it like Beckham – and also different accents like the Scottish accent – Monica in Secrets and Lies – or the northern accent spoken in Sheffield in The Full Monty, and of course the English spoken in London, where the majority of these films are set, with instances of Cockney and slang. In this sense, language varies also diastratically, according to the social class to which the characters belong, ranging from the upper to the working class, and the ethnic group of provenance – see the presence of Indian and black characters in Bend it like Beckham and Secrets and Lies. For instance, innit is used by the only Indian character in Secrets and Lies, thus becoming a marker of ethnicity, as in Bend it like Beckham, where innit is used by Indian immigrants living in London, both of first and second generation, even though the latter speak English with a British accent whereas the former speak Indian English. Innit is also used by black characters in London – Hortense in SL and Mel in BILB. Moreover, the use of these non-standard tags is also a matter of social class, in the sense that it is generally related to a low and informal register, used by characters who belong to the working class – see Cynthia in SL. Finally, age seems to play a role, since innit is used by youngsters in London – see Pete in GSH – who belong to a specific social group – i.e. hooligans. On the other hand, ain’t, being a non-standard form of negation, is only used by working class characters living in London, but it is not used by either Scottish or Irish speakers. Finally, the use of S-V disagreement is limited to working class characters, both in the London area – Cynthia in SL – and the Sheffield area – Dave and Gaz in FM.

4.2. Translating non-standard tags into Italian

4.2.1. The case of innit?

Of the four films analysed, the non-standard tag innit is present only in the three films that are set in London – i.e. Secrets and Lies, Bend it like Beckham and Green Street Hooligans.

Table 1: Translation options for innit in Italian dubbing

As can be noticed from the examples in Table 1 above, innit is translated in various ways. The solutions adopted in Italian range from the interjection eh?, to the simple expression no? and the conversational routine vero?, but also the direct question with a verb of belief like parere is possible. And this happens both when innit functions as a real TQ – i.e. examples (1) to (3) – and as an invariant tag – i.e. example (4). But in this way, the translating options chosen for the dubbed Italian version produce neutral effects and make the language levelled from the point of view of sociolinguistic variation. For instance, the use of innit by Pinky in BIBL is indexical of her cultural background and of the fact that she belongs to a specific ethnic group, namely the Indian community living in London. The same happens for the character of hooligan and cockney speaker Pete in GSH and for Cynthia in SL, whose use of innit, confirmed by her low, informal register, betrays her working class origin. However, the most striking feature to be noticed is that in the majority of cases, innit is not translated at all in the Italian version. See the examples in Table 2 below:

Table 2: When innit is not translated in Italian dubbing

Indeed, the fact that innit is not transposed in Italian in any way, not even with neutral solutions, causes an even more marked loss of the diatopic variation, together with the loss of social, cultural and ethnic connotations. However, some strategies for compensation are employed in order to maintain the same illocutionary force of the original tagged utterances in English and the low and informal register. In the first case, the use of emphatic adverbs, such as ma and ma tanto in (1) to (3), exclamative sentences in all instances, except in (5), and bad language in (4) are useful tools to express the same level of aggressive and sarcastic overtones. More specifically, in (1) the use of ma gives the right amount of emphasis to Mel’s comment, which retains the challenging function performed by the falling innit in the English version, leaving the debate open and expecting the addressee’s reply. Conversely, in (5) the confirmatory function of innit cannot be expressed by the statement chosen in the Italian dubbed version, where the speaker sounds much more assertive and less cooperative, so that the absence of the tag makes Hortense’s utterance aimed at closing off debate quite brusquely. Finally, the use of colloquial expressions, such as combinarsi in (3) or darla in (4) as well as bad language reproduce the same low register, typical of youngsters and of characters with a certain social background. Yet, these strategies are often not sufficient to express both diatopic and social connotations which are associated to certain characters in the original version of the films. In fact, not translating innit into Italian and not rendering the slangy expression of schlep[url=#n5][5][/url]  in (6) gives rise to quite impersonal and levelled effects.

4.2.2. The case of ain’t …?

The second non-standard tag to be analysed is the one in which the non-standard form of negation ain’t functions as the operator in the tag portion. This tag type occurs only in two films – SL and GSH – that are set in London, being a characterising feature of the South-east of England (cf. p. 4). Let’s analyse the examples in Table 3, which shows some translating options for this type of tag in Italian dubbing.

Table 3: Translation options for ain’t…? in Italian dubbing

As can be noticed, varied solutions have been employed, ranging from the conversational routine vero?, to the use of direct questions, as in examples (2), (4) and (6), and of the question with the verb of belief credere in (1). Generally, the function performed by the tag in the original English version is preserved in the dubbed version, yet with the loss of diatopic and diastratic variation. For instance, in example (2), the informational function of the rising tag in English is still expressed by the direct question ho indovinato? in the Italian version. At the same time, this translating option completely lacks the social value that the non-standard tag with ain’t communicates nonetheless, being a marker of a low and informal register. This is an example of how film language in dubbing tends to undergo a process of ‘standardisation’ (Pavesi, 2005: 21), in order to render the text more accessible to the audience, but at the risk of producing a neutral and artificial language, without any characterising personal connotation of the speaker. Another interesting example is the tagged utterance in (4), which is characterised by an operator ellipsis[url=#n6][6][/url], a clear marker of orality and informality that, instead, in the Italian dubbed version are rendered with the use of a colloquial register, of which the pronominal redundancy of ci – a solution adopted also in (1) with ti – in ci siamo parlate instead of the more standard option abbiamo parlato, is an example. Even though the same level of informality is kept, the non-standard tag is translated with the direct question eri tu?, in which the use of the correct form of the second person pronoun tu clashes with the previous choice of a low register, thus appearing incoherent on the whole context.
However, there are also cases in which the tag with ain’t is not translated into Italian, as can be seen in Table 4:

Table 4: When ain’t…? is not translated in Italian dubbing

When the tag is not translated into Italian, various strategies for compensation are adopted, such as repetition in (2) and (3) or the use of the demonstrative pronoun quello in (5), which accomplish the same function as the corresponding tagged utterances in the source text. In fact, the use of the tag is simply a ‘matter of emphasis, without either encouragement or discouragement of a reply by the addressee’ (Algeo, 2006: 299), and thus acting as a reinforcing formula with an expressing-phatic function. The alternative solution for the untranslated tags in the target text is on the diaphasic level, with the choice of colloquial expressions, as in example (1), and a low register with the use of the verb stare, in (4), which in some way balance and reflect the non-standard nature of ain’t.

4.2.3. The case of S-V disagreement

Finally, the last non-standard tag to be investigated is the tag with S-V disagreement. See the examples below:

Table 5: Translating options for S-V disagreement in Italian dubbing

In the dubbed version of the analysed films, this tag type is generally translated with eh?, no? and vero?. Such choices obviously produce neutral and levelled effects on the linguistic level, with the loss of diastratic variation, expressed by the S-V disagreement in the English tag, as in (1) and (5). However, in other cases, in order to portray the characters’ social and cultural background, some strategies are adopted, namely the use of pronominal redundancy in (4), where ci is the marker of a colloquial and low register, and the use of the emphatic adverb of negation mica in (2) which, together with the emphatic adverb pure in (3), represents a translating option that is marked diatopically in Italian. These two adverbs are in fact both typically used in central and southern areas of Italy. Another interesting translating solution in Italian is represented by the use of ancora in initial position in (6), which contributes to give emphasis to the whole tagged utterance, thus performing the challenging function successfully as in the original version. Finally, there is only one case in which the S-V disagreement tag is not translated into Italian, that is in example (7). In this case, the English tag performs the challenging and peremptory function, in the sense that the speaker firmly believes in the truth of the proposition contained in the anchor clause and tries to contradict the addressee’s assumptions. The falling tag is used to close off debate and has the force of an exclamation. In the dubbed version, the same function is performed by an exclamative utterance, even if the tag is not translated. Moreover, the use of the contracted form of the pronoun ci – i.e. c’hai – mirrors the use of a low and informal register that is expressed by the non-standard tag in the source text, even though the use of the correct form of the second person pronoun tu clashes with the whole tenor of the Italian utterance.

5. Conclusions

The analysis of the translation of English non-standard tags in Italian shows an evident linguistic difficulty in transposing a linguistic construction that does not belong to the target language, a difficulty that increases when non-standard features also have to be represented. As a consequence, various solutions are adopted from time to time. The most frequent translating options are the use of simple expressions, such as vero? and no?, or the interjection eh?; but also the verbs of belief, like credere, and of evidentiality, like parere, are employed to accomplish the same function. Moreover, often non-standard tags are not translated at all into Italian – for instance, even the use of direct questions is preferred – thus the speaker appears more assertive and less interactive or cooperative in the talk-exchange and indeed in this way, the dubbed version appears definitely more levelled and neutral, with the loss of both diatopic and diastratic variation. Actually, the choice of strategies of levelling out (Laviosa-Braithwaite, 1998) of linguistic variation is recognised as a universal in translation. One instance is represented by the levelling of the different accents spoken by various characters in the four films in English, who speak a standard Italian in the dubbed version (Pernigoni, 2005). Usually, diatopic and diastratic varieties are partly rendered through morphosyntactic and lexical means (Pavesi, 2005), as some strategies for compensation are adopted, ranging from the use of colloquialism, informal register, bad language, as can be seen in the dubbed versions of the four films that have been analysed. Actually, in the case of innit, a few attempts at reproducing ethnic variation may be identified, for instance in the fact that in BILB Indians of first generation speak Italian with an Indian accent. Conversely, in the same film, youngsters of Indian descent speak standard Italian, so that it is possible to understand the cultural gap between the two generations. However, it has to be noticed that whereas Pinky often uses the non-standard tag innit, which becomes a marker of cultural identity and tradition, on the contrary, her sister Jess never uses it, thus reflecting her attitude forwards traditions. Unfortunately, this difference cannot be identified in the dubbed Italian version.

We can conclude that, although the various strategies for compensation in Italian dubbing may appear quite efficient in conveying diastratic and diaphasic variation to some extent, they are less convincing for expressing diatopic variation and the values that are attached to it.