Sociolinguistics of Hindī An Analysis of TV-Mediated Spoken Hindī Features Through TV Programmes’ Language

This paper analyses varieties of Hindī language differentiated on the social ground, namely standard Hindī (the sanskritised language) and non-standard Hindī. The latter, which includes different varieties of spoken Hindī, was studied listening to two TV programmes: Delhi Crime and Satyamev Jayate. Several features were observed on every level, but here the focus is on syntax (Hindī word order) and lexicon (code-switching). Information structure theory and pragmatics are used to analyse TV-mediated spoken Hindī word order. The comprehension of these special features can be helpful to predict the changes towards which Hindī tends to move and to give examples of the spoken


Variability and Variety
One of the main (and most visible) ways through which the language change happens surely is the aerial-phonic medium or spoken language. The relations humans have with their counterparts are extremely complicated and several factors can influence the communicative events they experience with each other at the linguistic level. The utterances change according to factors like the languages which the interlocutors know, the situation in which they speak, their age, and so on. These assumptions are based on the linguistic work sketched with Bloomfield (1933) and carried out with the research of Labov 1 (1966) and Gumperz 2 (1971), who started analysing languages through the social point of view. Through the following decades, the knowledge of the discipline now known as sociolinguistics made us more and more conscious about how and in which ranges and dimensions changes in a language take place.
Here, some crucial concepts in sociolinguistics need to be mentioned, the nature of which has been discussed over the decades: 'variability' and 'variety'. The former can be described as the dynamic process of linguistic differentiation (Berruto 1980, 20-1;Meyerhoff 2006, 10) and the latter the result of this process. Since the nature and the direction of the processes of differentiation are heterogeneous, there are different criteria that can define what a variety implies: geographical difference (dialect), 3 social use (sociolect), situational use (register) (Biber, Finegan 1994, 4;Bussmann 1996Bussmann , 1261. Despite the origin of this differentiation process, a variety is: 1 In 1963 Labov firstly observed that there were some phonetic peculiarities in the English of the island of Martha's Vineyard, off the coast of New England, USA. These distinctive features were consequences of social changes that happened in the island, which during the years became a holiday destination after being fairly isolated. More precisely, Labov has shown that the vowel sound of words like house, loud had two different pronunciations, one low-prestige, old-fashioned and one more recent, found in a prestigious American accent, the former becoming exaggerated as a reaction of the native of the island towards the mass touristic invasion (Trudgill [1974(Trudgill [ ] 2000. Later, Labov (1966) increased the knowledge of the social stratification of the language of New York through his pioneering research. every set of different and specific ways to use a language, recognizable by a specific series of traits from all the levels of analysis or [from] some [of the levels of analysis] (phonology, morphology, syntax, lexicon, textuality) which qualify it and differentiate it from other sets of usage [of the language] and is provided with a certain degree of homogeneity of recurrence in conjunction with social traits and/or different classes of situation. (Berruto 1980, 25; translation of the Author) 4 As this assumption points out, the concepts of community and socially shared features are crucial to circumscribe the concept of variety.
What are, thus, the criteria used to delimit a specific variety? The material on this issue is huge, but there is little disagreement on the taxonomy of the factors triggering variation. In addition to the 'time' factor (the diachronic varieties or états de langue), the second class of factors concerns geographical diversification, namely the origin and the distribution of the speakers, which create peculiarities (at the regional or local level as well as national). The third class gathers factors about the speaker, their social and cultural background (level of education, job etc.). The last one, instead, concerns the different communicational situations in which the speaker might be. More precisely, the situational-contextual factors may be analysed through the relation of the speakers which can change 'register', see above, the topic of the conversation (technical language etc.), the medium (written/spoken language, e-mails etc.) (Berruto 1980, 27-9;Pistolesi 2016).
Nonetheless, one can argue that there is no easy way to delimit a variety. As for other concepts regarding language and linguistics -for example, the ones coined by Saussure at the beginning of the study of general linguistics ([1916] 1971) -it can be said that they are a simplification of more difficultly understandable phenomena. Let us take the concept of 'synchronic' approach as an example: since the language is constantly moving forward, one état de langue combines features from a past état de langue that are still used among the speakers as well as new features that are still not accepted by all the community of speaker (see Saussure [1916Saussure [ ] 1971. A synchronic approach thus takes a 'photograph' of a single moment of the history of a language and makes it static, regardless of the prior or successive changes. It can be said that this approach is a useful abstraction for the study of a language, of its feature and of the ones which are 4 The original quote in Italian is the following: "ogni insieme di modi diversi e determinati di usare una lingua, riconoscibile per una certa serie di tratti di tutti o di alcuni livelli di analisi (fonologia, morfologia, sintassi, lessico, testualità) che lo qualificano e differenziano da altri insiemi di modi, e dotati una certa omogeneità di ricorrenza in concomitanza con certi tratti sociali e/o diverse classi di situazioni" (Berruto 1980, 25).

286
Annali di Ca' Foscari. Serie orientale e-ISSN 2385-3042 57, 2021, 283-312 universally shared. As for états de langue, the difference from a variety to another is not as distinguishable as black from white, since there is a continuum in which the features of a single variety blend softly into another variety 5 (Berruto 2013, 22-30).

Hindī Varieties
Once these preliminary issues are taken into consideration, the main topic of this paper can be introduced. For the field of Indo-Aryan languages, the variability among the languages -the totality of varieties -is particularly visible. Whether socially (distinguishing 'sociolects') or geographically (distinguishing 'dialects'), India and Pakistan are extremely dense as far as language richness is concerned. In this sense, we cannot avoid mentioning the work carried out by George Abraham Grierson with the Linguistic Survey of India, who collected information and described 179 languages and 544 dialects of the Indian Subcontinent, whether about the genetics of these languages or the peculiarities of every single language. 6 Despite Grierson's research and other scholars' work in the field of Indo-Aryan languages, there are few works on Hindī sociolinguistics, varieties and registers and how these are intertwined in their contemporary forms -whereas there are several works concerning the English, the Italian or more in general on the situation of the European languages. 7 The main goal of this paper thus is to give an outline to the social varieties of contemporary Hindī and then to focus on the TV-mediated spoken variety of language. Since it was impossible to be in 5 The linguistic continuum can have different shapes, depending on the distribution of the varieties in the analysed community. For example, Mioni, Trumper (1977, 330) assumed that, based on the Italian situation, there are two easily identifiable poles and between them there are all the other varieties that blend into another. Given the real nature of the continua, it is difficult to incorporate in the analysis more than one dimension of variation (temporal, geographical, social, situational) (Berruto 1998, 24-5). Since it is beyond the scope of this paper to discuss the real nature of linguistic continua, whether in general or applied to the Indian context, we will simply assume that there is not a sharp division between the varieties, but the traits blend from one into another.
6 The journal named Linguistic Survey of India was published to present a summary of the results of the Linguistic Survey of India, a research conducted from 1894 to 1928 by Grierson. It has been the "culmination of the machinery for data collection" (Pandit 1975, 76).
7 For the sake of information, some scholars need to be mentioned as well as their works: Nespital (1990), Masica (1991) focused on the relations between Indo-Aryan languages and dialects; Cardona and Jain (2003) gave attention to the sociolinguistics of the Indo-Aryan languages; Shapiro (2003) focused on the general information about the history of Hindī language; Montaut, then, in her several works about Hindī language, gave attention to grammatical issues as well as sociolinguistic ones, like bilingualism and linguistic diaspora (1991; 2001; 2004; 2014). the Hindī-speaking area, the approach to collect linguistic information, traits and features of Hindī -namely of spoken Hindī -was different from the classical in-field research. Thus, a Hindī Netflix series, Delhi Crime, and some episodes of Satyamev Jayate, a famous Indian TV programme that uses Hindī/Urdū as medium were used to gather linguistic material. Listening to these episodes, some peculiar traits did not match with the standard language, namely the prescriptive language of the grammars and the language that is taught (at the University, in Hindī teaching schools etc.). For this reason, the Hindī variety analysed here shows the features of a TV-mediated language -and it is just one of the spoken Hindī varieties. From this starting point, then, the whole work was built: the first variety that the foreigners run into is surely the standard one.

Standard Hindī Variety
Some have the misconception that the standard variety 8 is the purest and closest to the original speech of native speakers, but this is not necessarily true. Every national and linguistic situation is surely different, but there are some features that the standard language varieties have in common: a standard language variety is a variety that has been designated as such and for which a set of forms has been identified and codified in dictionaries and grammars. (Finegan, Besnier 1989, 496) and: a standard language is a code in which the separating, solidarity and prestige functions of language are optimally operative, and purism […] is one of the possible rational responses to these three functional criteria. (Thomas 1991, 115-16) There are some concepts that thus need to be discussed here: these definitions of 'standard' highlight that a standard variety is the result of the process of standardization […] one of the main agents of inequality.
[…] Standard languages do not arise via a 'natural' course of linguistic evolution. (Romaine 1994, 84) 8 The term 'standard language variety' or 'standard variety' is more correct as compared to 'common language' (Berruto 1980, 34) as is clear from the very process of creation of the standard that will be displayed in the following lines. There are thus some criteria which operates in the creation of the standard variety. In general, scholars listed some properties that standard varieties have, such as: • being codified through grammars and/or institutions, • being stable/stabilised, • being unifying (for different varieties) and dividing (a national state from another), • being the model of language that can be referred to; • having prestige (Garvin, Mathiot 1956;Ammon 1986;Joseph 1987).
The codifying process is what makes a variety of language the standard variety. Nevertheless, there must be a need for a prestigious variety which can identify a precise and delimited community, conscious of its existence: there needs to be an 'Other' (Brass [1974(Brass [ ] 2005. In the Indian situation, an 'Other' appeared when the Muslim community started being considered dangerous for the sprouting Hindū identity in the middle of 19th century due to the British policy-making. In fact, in 1832 (in Bombay and Madras) and in 1837 (in Bengal) the British substituted the official language, Persian, with Urdū, which was used by both Hindū and Muslim and at the same time was used for Islam-related literary purposes 9 (King 1994, 54;Consolaro 2003, 12). This event culminated in the creation of a conscious community and (religious) identity, which required also two 'national' purified languages in the event of Partition of 1947: Hindī and Urdū. 10 9 Since this event is just the visible end of a longer process, to examine in depth the common origin of Hindī and Urdū, the uses of the languages during the centuries and then the standardisation and nationalisation of the two languages, see Rai 1984;Shackle, Snell 1990;King 1994;Dalmia 1997. 10 Since the process of codification and nationalisation of the language is not the main topic of this paper -even if I agree on the extreme importance of its objective analysis -some of the main points this issue raises will be rough out here. First, the legitimation and codification process of Hindī (and Urdū) was made through different channels and in a considerable amount of time. One of the most visible ways was the choice of the script, the devanāgarī, while for Urdū the nastaʿlīq script was chosen. The discourse through which this change was made is carefully analysed by Ahmad (2012). Other channels were the academies and organisations which promoted a Sanskritisation of Hindī, mostly by means of the lexicon, like Nāgarī Pracāriṇī Sabhā -founded in 1893 -and the Hindī Sāhitya Sammelan -founded in 1910. Through these organisations a choice was made to create and/or use words originated from Sanskrit rather than from Persian or Arabic ('Urdū' words), or considered 'regional', 'dialectal' or 'wrong' (Bhatia 1987, 178-9). This whole process did not last just a couple of decades, but it continued during all the first part of the 20th century, culminating in the Partition of 1947 and continuing after it with several politics. In 1965, with the Official Languages Act, Hindī became the official language of India with English. Through time, and with several (violent) protests, other languages became official, since the plurality of languages, cultures and histories in India makes it impossible and unjust to have just Hindī and English as official languages. The Eight Schedule thus lists 22 official languages recognised by the Constitution. To analyse in depth this topic see: for the Through the process of Sanskritisation, standardisation and purification of Hindī, śuddh hindī ('pure Hindī') was outlined. Even if it is impossible to describe every feature of this variety, some assumptions are necessary. As mentioned before, śuddh hindī, or Hindī standard variety, has a highly Sanskritised lexicon and several neologisms -developed by the Board of Scientific and Technical Terminology (founded in 1951) on the basis of existing words of Sanskrit. It is an improvised language from a 'mixed' basis, whose regional words were eliminated, being considered 'vulgar' by the leaders and founders of the organisations for the standardisation of the language (namely, Nāgarī Pracāriṇī Sabhā and Hindī Sāhitya Sammelan, see note 10). Also, among the different forms that existed in the spoken language, only some of those were chosen for the standard: the least 'regional', 'dialectal', 'Urdū'. 11 All the processes of using this variety in education, newspapers, media, by the Government, in the politics, in the grammars, led to the effects described by Shapiro: at the beginning of the twentieth century, and even as late as the 1950s and 1960s […] most so-called Hindi speakers were actually native speakers of one or another regional dialects of Hindi, but with some degree of competence in the standard language learned through formal education. At present the effects of a half century of effort by the Government of India have clearly been felt. There are now tens of millions of people, including many living in geographical areas which would have been thought of as the heartlands of Braj, Avadhi, Bhojpuri etc. whose native language is some variety of MSH [Modern Standard Hindi]. In addition, the massive spread of modern technologies of communication […] have had a standardizing effect, and brought heavy exposure of MSH to significant portions of the population in north India. (Shapiro 2003, 282) theory of the creation of the 'Other' and the process of nationalisation of language: Brass ([1974] 2005); for the dynamics of the construction of Hindū consciousness: King (1994, chs. 2-3), Dalmia (1997); for the academies and organisations for the Sanskritisation of Hindī: Kachru (1989), King (1994), King (2001), Shapiro (2003); for the post-Partition issues in India: King (2001); and in Pakistan: Schmidt (2003), Rahman (2006).
11 An example of these 'vulgar' or 'regional' forms from which śuddh hindī was purified is narrated by Barannikov (1936). In 1804-1810 Lallūjī Lāl wrote the Prem Sagar, considered the first literary work in Hindī. Literary critic Jagannāth Prasād Śarmā (born in 1905) assumed, as well as other scholars: "his grammar is not stable and has no standard; for instance, he uses several variations for the same form; in order to express the Conjunctive Participle he uses such forms as: kari, karke, bulāy, bulāykari, bulāykar, bulāykarike" (Barannikov 1936, 386). This example serves to highlight how the standardisation process worked, and how the importance of personalities such as Śarmā had a visible impact on the language. The politics used by the Government to standardise the language produced also changes in the registers or styles of Hindī used in different contexts and functions, dictating which form of Hindī must be used throughout the country, namely a Sanskritised variety. Even if in formal situations the Sanskritised variety is used, in various public spheres there has been the tendency towards "a less Sanskritic and more heterogeneous register" (Shapiro 2003, 282).
All the features of this language are, as one might think, several and rather complicated to explain in just few pages. As Caracchi states: one of the main difficulties one can encounter while writing a Hindī grammar is due to the proteiform nature of this language, [which is] spoken in a really extended area and which, not only in the lexicon and in the pronunciation, but also, even if in a lesser extent, in grammatical structure, experiences variations determined by the place, by the social and religious context, by the proximity with other linguistic areas. Thus [,]  For this reason, I will recommend the most commonly used books to the ones who are not familiar with the grammar rules of Hindī, namely Teach Yourself Hindi: Complete Course by Snell and Weightman (2003), Hindi by Kachru (2006) and the Italian Grammatica hindī by Caracchi (1992). Also, the standard variety form of the sentence or of a word will be displayed above the non-standard forms when needed. Thus, in the following paragraphs, some of the features of the non-standard varieties that were analysed through the means of the TV series/programme will be described.

Non-Standard Varieties of Hindī
As quoted from Caracchi (1992), there are different features of the same language -Hindī -which coexist under the same 'label' of 12 The original Italian quote is the following: "una delle difficoltà principali che si incontrano nella stesura di una grammatica hindī è dovuta al carattere proteiforme di questa lingua, parlata su una superficie vastissima e che, non solo nel lessico e nella pronuncia, ma anche, sebbene in misura assai minore, nella struttura grammaticale, subisce variazioni determinate dal luogo, dal contesto sociale e religioso, dalla vicinanza con altre aree linguistiche. Così talvolta quella che alcuni considerano acchī hindī può esser ritenuta da altri una forma colloquiale, locale, desueta o addirittura scorretta" (Caracchi 1992, 7). Hindī. The spoken form is the main way in which humans communicate with each other, which also include the peculiarities of the varieties of the same language (in this case, Hindī along with its standard and non-standard varieties).

Sociolinguistics of Hindī
Even if space-related change cannot be considered purely a 'social' variation of the language -as mentioned in the first paragraph -the first change that stood out for the scholars was related to geography. Grierson firstly dealt with Hindī dialects: in the 6th volume of the Linguistic Survey of India he divided Eastern Hindī in three varieties (Avadhī, Baghelī and Chattīsgaṛhī), while in the 9th he lists the Western Hindī varieties (Hindostānī, Bāngarū, Brāj Bhāṣā, Kanaujī and Bundelī). More recent distinctions and additions among the varieties of Hindī are made by Nespital (1990, 5-6), Masica (1991, 9-12) and Kachru (2006). Nespital makes a clearer and wider list: 1. regional dialects in a narrower sense, divided in a) Western (Hariyāṇī, Dehātī Khāṛī Bolī or Kauravī, Brāj Bhāṣā, Kanaujī, Bundelī), b) Eastern (Avadhī, Baghelī, Chattīsgaṛhī); 2. regional dialects in a broader sense, divided in a) Pahāṛī with Gaṛhvālī and Kumāonī, b) Rājasthānī 3. with a number of dialects, and c) Bihārī, used as a cover term for Bhojpurī, Maithilī and Magahī); 4. urban dialects or forms of town speech within the Hindī region, a) old ones, such as the dialect of Old Delhi, and the forms of town speech of cities, like Varanasi or Allahabad, and b) new ones, like the spoken variety of Hindī resulted from the 1947 afflux of Pañjābī people to Delhi.
Kachru adds a Southern dialect, the Dakhinī (which is mentioned also by Nespital but only as a dialect outside the Hindī 'belt', or Hindīspeaking area; see Nespital 1990, 5). As one can expect, these varieties differ from each other in some features even if they are mutually intelligible: for example, Dakhinī Hindī does not show the ergative construction, while the other varieties do (Kachru 2006, 5). An important feature of the dialectal variation is the interference and the presence of loan words or loan grammatical features from other languages cohabiting in the area or spoken in the contiguous area. For example, in Dakhinī Hindī -centred in Āndhra Pradeś but spoken also in other urban areas of the Deccan Plateau with a significant number of Muslim people -there are traces of the influence of Telugu (the Dravidian language spoken in Āndhra) in the present tense, in which the participle and the auxiliary verb are blended: ātūṃ is the Dakhinī form for ātā (M.SG. present participle of ānā 'to come') + hūṃ (1st SG. present of honā 'to be') (Masica 1991, 426;Montaut 2004, 102). There are also other possibilities concerning different levels of language analysis, like the nouns or the pronouns. Nespital maintains, for example, that in Eastern dialects of Hindī (such as Hariyāṇī or Kauravī/Khāṛī Bolī but also in the speech of Old Delhi) the genders of many nouns deviate from the one they have in standard Hindī (Nespital 1990, 9).
As seen before, there are fewer works that deal with the social dimension of variation, which creates styles and registers, formal and informal ones, with prestigious influences or more rural ones. Gumperz, talking about the languages of India, assumes that all […] have two styles, the colloquial and the literary. The latter is used in formal speech-making and in writing, and is taught in schools. It is often quite different from the colloquial. Only educated people are familiar with it. (1971,(4)(5) This "literary" variety thus has the role of standard variety of the previous paragraph. Gumperz indicated then conversational styles for Hindī, like Hindustānī, Khaṛī Bolī, Dakkhinī Hindī, Dakkhinī Urdū, Bāzār Hindustānī, 13 as well as the already signalled Hindī and Urdū with their peculiar, standardised features and their respective dialects (Gumperz 1971, 48). Another classification of Hindī varieties is made by Nespital: 1. standard Hindī a) in its written form and b) in its spoken form; 2. regional or local dialects in the Hindī 'belt'; 3. dialects outside the Hindī 'belt' but genetically and historically related to Hindī (as is already mentioned Dakhīnī); 4. pidgin-like speech varieties of Hindī, as the one spoken in Bombay or Calcutta (Nespital 1990, 5).
In another interesting classification made by Nespital, instead of grouping geographical or conversational features of Hindī, he grouped Hindī speakers: 1. speakers of standard Hindī as their mother-tongue, limited in number; 2. speakers of Hindī whose primary language is a certain Hindī dialect, the majority of Hindī speakers; 3. speakers of Hindī whose primary language is another Neo-Indo-Aryan language (Pañjābī, Gujarātī etc.), a very large number; 13 For what concerns Hindustānī (as British called their era's lingua franca ;Schmidt 2003, 318), it is a variety with a native lexical core and several words originating both from Sanskrit and Perso-Arabic and it is written in both devanāgarī and nastaʿlīq scripts. Lahiri assumes that Hindustānī is more "a distinctive set of words, phrases and stylistic conventions" than "a set of grammatical rules and lexical definitions" (Lahiri 2015, 73). All the features of the language are pracalit, 'in use', mostly for the formative years of "work of Progressive writers, and Urdu writers more generally" (Lahiri 2015, 76) but it is very far from Hindī and spoken Hindī. 4. speakers of Hindī whose primary language is a Dravidian language (Tamil, Telugu etc.), a very large number, as for the previous group; 5. native speakers of Urdū; 6. speakers of Hindī whose primary language is English (a variety of standard English or of Indian English), relatively few (Nespital 1990, 3-4).

Sociolinguistics of Hindī
An important point Nespital adds in this classification -in addition to the geographical or circumstantial variations -lies in the knowledge of the speakers. Most of the speakers in the groups classified by Nespital are bilingual, and this surely has an impact on their linguistic production and in the spoken Hindī varieties (see § 2.2.4). He maintains that the impact of the languages spoken by Hindī speakers -other than Hindī -have a massive interference on the lexical, morphological and syntactic level of language (Nespital 1990). This paper aimed to collect information about characteristics of the spoken language and the social, dialectal or circumstantial reasons which make them possible. Since the classifications above do not consider the medium used to communicate, there needs to be an addition to further the analysis. As already mentioned, the collection of linguistic traits and features was made through TV programmes and thus the variety then delineated, and this will be presented in the following paragraphs which can be labelled as TV-mediated, spoken and non-standard Hindī. If these features are common to the more generally labelled 'spoken Hindī variety', it will be discussed at the end of this paper.

Phonetic Features of Spoken Hindī
Starting with order with the phonetic level of analysis, one phenomenon that has to be mentioned concerns the pronunciation of sounds which derive from the phonetic system of Urdū (and indirectly from Persian), namely /f/ and /z/. In written Hindī, in devanāgarī script, 'autochthone' syllabic bases are used, namely फ /ph/ and ज /ʤ/, to which usually a dot is added, resulting in फ़ /f/ and ज़ /z/. Pronunciations for Urdū, and even English, words are thus sometimes: /ˈʤa:datər/ (mostly), /ˈʤa:da/ (more), /te:ʤ/ (fast), /ʤaˈru:ri/ (important), /dərva:ʤa/ (door), /pʰaiv/ (five). 14 The presence of these sounds shows research for a structural homologation at the phonic level. In the analysed TV programmes, the change /f/ > /ph/ and /z/ > /ʤ/ were present when 14 The standard pronunciations of these words are: /ˈzja:da/, /te:z/, /zaˈru:ri/, /dərva:za/, /faiv/. the speakers (or the fictitious speakers) had little knowledge of Urdū/ English and the sounds of these languages, and/or when they were not educated.
If, on one hand, the use of such sounds may be due to the education of the people, on the other hand it has to be noticed that the Hindī Śabdasāgara by Śyāmasundara Dāsa of Nāgarī Pracāriṇī Sabhā (1965-1975 and updated in December 2020) consider the use of /ʤ/ in the previous examples as the correct orthography. The nationalistic ideas of this institution must be kept in consideration, since in other dictionaries (Caturvedi's A Practical Hindi-English Dictionary and Bahri's Learner's Hindi-English Dictionary) the original /z/ is quoted. Since the government of India has had nationalistic leanings for several decades, the theory that this kind of phonic change can be a push from outside forces cannot be rejected without further analysis. Let us now switch to morphological features of spoken Hindī, which along with the syntactic ones -which were analysed more deeply than the phonetic ones.

Morphological Features of Spoken Hindī
Some interesting features observed at the morphological level of analysis mainly concerns agreements. In standard language, honorific pronoun āp is followed by the third person plural of verb 'to be', haiṃ. An example follows: The standard form of this sentence would be āp sahī kah rahe haiṃ, with subject in first position (see § 2.2.3) and the agreement between the honorific pronoun and the auxiliary verb 'to be'. The phenomenon which concerns the pronoun āp followed by the second person plural of 'to be' (ho) is observed also by Singh (2016, 290-2) like a new unofficial configuration of honorific, in the middle between a formal and an informal expression. The reasons this is happening are explained by Singh: regional language influence, popularization through mass media and ongoing linguistic and cultural change in Hindi speaking communities. It is also possible […] a combination of these factors. The 15 The episodes of Delhi Crime cited in this paper can be seen on Netflix.

295
Annali di Ca' Foscari. Serie orientale e-ISSN 2385-3042 57, 2021, 283-312 dynamics of media penetration by cinema, TV and radio into rural culture creates a mixing of urban elite language usages with rural language usages. (2016,292) The penetration of cinema into rural culture and language thus explains this peculiar linguistic trait, as well as it can help understanding the phenomenon of code-switching (see paragraph 2.2.4).
Other peculiarities of the pronouns concern the pro-drop possibility of Hindī language. If in standard Hindī the pronoun should always be cited, in spoken language, where the context makes it understandable who/which is the subject of the sentence, the pronoun can be tacit: There are other grammatical traits of Hindī which are described in grammars, but which are treated as 'rarely possible' in standard language. In spoken language, these traits are more present and can thus be observed. One of these concerns the genitive particles, namely kā (M.SG.DIR), kī (F.SG/PL) and ke (M.SG.OBL/M.PL/HON), which agree with the possessed object. As Caracchi highlights, possess can be expressed -rarely -with the oblique particle ke (1992,94). Nevertheless, in spoken language this can happen more frequently, especially since the speakers pay less attention to the accuracy and grammatical precision of what they say. It is the same happening with the verb 'to be' honā conjugated at the third person singular and at the third person plural, which in prescriptive grammar would be respectively hai and haiṃ (the latter with nasalisation). In spoken language the verb hai (third person singular) is used indiscriminately for singular or plural. This can be due to the influence of Hindustānī, which does not differentiate the verb honā in its singular and plural form.
From these few examples from the morphology field (to which surely more examples can be added) it is clear that the spoken language shows some difference from the prescriptive one -the grammar and the standard language. This is due to the fact that spoken language -in this case spoken Hindī -is the real actualisation of the language, it is used by the speakers and thus put in different context where there are different communicative circumstances and different interlocutors. In the following paragraph, concerning the syntactic features of spoken Hindī -the one the Author was more focused on -this tangled relation between speaker and language will be analysed even more deeply.

Syntactic Features of Spoken Hindī
For what concerns the syntax, namely the order of constituents, a premise must be visited about word order and the information that peculiar word order can convey. Hindī standard word order is SOV (subject/object/verb) and as in other languages it is flexible (such as Italian, French, English). All the patterns which differ from this one, though, convey a slightly different sense from the standard one. These non-standard syntactic orders are mostly used and 'fixed' in the spoken language. In the spoken situations, in fact, the speaker uses the language in the most 'pragmatic' way with all the means they possess to convey the highest amount of information during the event of communication. Let us, thus, introduce the theory of information structure, developed in the domain of pragmatics, 18 to explain this topic in a more precise way.

Information Structure
As quoted above, there is a standard word order -in Hindī and in every other language. Nevertheless, 97% of the world languages presents a SOV-SVO pattern. Nevertheless, other patterns are possible, such as VOS, OVS, OSV -at least theoretically. 19 The predominance 18 'Pragmatics' is the discipline that relates signs (specifically linguistic signs) and speakers. It thinks of the language and its products as actions, thus inescapable from the 'doers' and their context. The first contributions to pragmatics are Austin's speech act theory (1962) and Grice's implications (1975). Through pragmatics, the cognitive processes which make the communication possible are analysed through what is said (discourse particles, like deixis) and on what is not said (word order etc.). Here, information structure will be analysed, which is an important field of pragmatics. For a detailed study of pragmatics, see Morris 1938;Carnap 1942;Austin 1962;Grice 1975;Levinson 1983;Neale 1992;Bublitz, Norrick 2011;Ehrhardt, Heringer 2017. 19 The third most used standard word order, VSO (verb, subject, object), is adopted by less than 10% of the languages of the world (for example by Arabic, Hebrew, Celtic languages). VOS accounts for only 3% of the world's languages, OVS for 1% and OSV for just one known language, Warao, spoken in Venezuela, Guyana and Suriname (Grandi 2003, 24-6). of SOV/SVO order is due to the information every member of the sentence conveys (namely subject, object and verb).

Sociolinguistics of Hindī
To help understand how and why this happens, information structure theory is useful. According to Roberts, information structure is: generally characterized as a variation of sentential structure along certain parameters to modulate the presentation of the information imparted by the sentence in such a way as to relate that information to prior context. (Roberts 2012, 2) Here, 'information' is understood as the result of the relation between what comes from outside the listener (people participating in the communicative event, context) and what is inside them (linguistic expressions and their meanings, but also linguistic representations created during the communication). In this place, information is created in the minds of the interlocutors, which can be different from the real meaning of the utterance -every interlocutor's information differs one from the other since it is created from different starting points (Lambrecht 1994, 37-43). At this point the concepts of given/new, topic/comment and focus/background can be added to infer the way information structure works. 20 Firstly, one interlocutor can possess some information, namely the 'given', but not the others, for whom it is the 'new'. For this reason, what is considered as a 'given' information is presented to the listener during the communicative event to help them understand the 'new' information better (Lambrecht 1994, 51). Another way the speaker introduces the issue they are talking about is the 'topic': the thing which the proposition expressed by the sentence is about (Lambrecht 1994, 118); what is spoken of in the sentence, the psychological subject, the element around which the predication is built (Berretta 1995, 127, translation of the author); 21 the entity that a speaker identifies about which then information, the comment, is given. (Krifka 2008, 40) 20 There are many ways through which the "presentation of the information" is "modulated", such as prosody, grammar particles etc. (Lambrecht 1994, 12;Roberts 2012, 2). In this paper only the syntactic constituents order will be presented through the use of topic/comment, given/new and focus/background concepts.

21
The original Italian quote is the following: "ciò di cui si parla nella frase, il soggetto psicologico, l'elemento attorno al quale è costruita la predicazione" (Berretta 1995, 127 The 'comment' is thus what is added to the topic, a piece of 'new' information. In most of the languages of the world there is not a formal/ grammatical indication of the topic -as in Chinese, where the topic is always at the beginning of the sentence. In a standard sentence such as Mary ate an apple it is impossible to understand which member is the topic unless the context in which the sentence is uttered is also known. This sentence can be the answer to different questions, and each answer to these questions would present a different topic (which is italic in the following examples): i. What did Mary do? Mary ate an apple. ii. Who ate an apple? Mary ate an apple. iii. What happened? Mary ate an apple. 22 Lastly, the 'focus' is parallel to the topic: that portion of a proposition which cannot be taken for granted at the time of the speech. It is the unpredictable or […] non-recoverable element in an utterance. The focus is what makes an utterance into an assertion. (Lambrecht 1994, 207) Thus the 'background' is what is taken for granted and is predictable: the 'given' information.
All these parts of information -'given'/'new', 'topic'/'comment', 'focus'/'background' -have their own standard position in the sentence. The topic (which corresponds to the 'given' information) is usually in the first position, since it introduces the sentences and gives the interlocutor an idea of what they are talking about -in a standard basic sentence it is represented by the subject. The newest information and the focus are in the central or final position (depending on the structure of the language -OV or VO), corresponding usually to the object or other complements. The position fulfilled by the verb (second in VO and third in OV) is of background: the intonation of the sentence as well gives less emphasis to this element (see Patil et al. 2008, 64). The position of all these parts of the information structure of a sentence conveys different information depending on the position they have in the utterance. These positions and, thus, the word order depend on which information the interlocutor(s) already possess(es) and which one the speaker wants to convey. The information structure of the sentence can thus change the word order. With the following paragraph some spoken Hindī word order and the information structure they display will be examined.
22 This last sentence, where there seems to lack a topic, can be considered thetic. In this kind of sentences all the information conveyed are new and thus a topic cannot be identified by the interlocutor (Lambrecht 1994, 222).

Word Orders in Spoken Hindī
To see how the order of the elements changes in spoken language, an example of the standard language is useful.
1. maiṃ āj keval bhāt aur dahī khāūṃgī maiṃ (S, 1SG) āj (ADV, today) keval (ADV, only) bhāt aur dahī (O, boiled rice and yogurt) khāūṃgī (V, to eat, FUT, 1SG) 'Today I will eat just boiled rice and yogurt' (from Caracchi 1992, 28, translation of the Author) In example 1, the subject corresponds to the topic, opens the sentence and the object is the focus. The verb closes the sentence. One of the most observed non-standard word order in the TV programmes was the post-verbal subject (thus with VS order). 23 Here, just some examples will be considered: 2. kal phone kartā hūṃ maiṃ kal phone kar-t-ā hūṃ maiṃ ADV.tomorrow call to do-PRES-M.SG AUX S.1SG 'Tomorrow 24 I will call' (literally 'tomorrow call I'll do') (from Delhi Crime, ep. 3) In example 2, the subject has the least importance in the sentence, and is thus positioned at the end, where there is usually the verb. All the elements to whom the speaker gives more importance in the context are in first and second position, in this case 'tomorrow ' and 'will call'. 25 23 The topic of post-verbal subject has huge relevance in recent studies, mostly for helping understand the real nature of 'subject', for centuries considered a static and monolithic category of grammar. See for the notion of subject: Keenan 1976;Cole et al. 1980;Van Valin, LaPolla 1997;Drocco 2008;Montaut 2014; see for post-verbal subject : Masica 1991;Cardinaletti 1997;Pinto 1997;Leonetti 2018. 24 The words in italics in some of the examples are from the Author. The intention and the reason to use this written device -italic -is to translate in English the communicative intention and emphasis of the sentence. Since in English some kinds of dislocation and word order cannot be represented merely by the syntactic structure of the sentence (because of syntactic restrictions of the English language), the use of other devices is made (such as emphasis, intonation, periphrasis etc.). Thus, the strategies English-speaking people would use to convey the same meaning of the Hindī examples are mainly based on intonation and emphasis, which cannot be written down and will be displayed using italics.
25 Differently from English, Hindī is a pro-drop language, that means that the verb can be separated from the subject and the latter can also be implied. A sentence like example 2 cannot exist in English. am', which is put in first position, and in 6 there is an entire syntagm, pulīs ke sāth. Lastly, in example 7, it is the verb that is uttered first.
All these elements which are positioned at the very beginning of the sentences have peculiar meanings, which can be different from one example to another. In sentences 4, 5 and 6 (with paṛhnā likhnā, enter exam and pulīs ke sāth) the element in first positions are topics. Being topics, they are in subject-position to create a 'Common Ground', a 'scene-setting' (see Krifka 2008). For this reason, the subject of the sentence is shifted in another position that does not have the prominence the first one has.
For what concerns examples 2 and 3, instead, it is important to highlight that those first uttered elements (kal and kuch nahīṃ) have special emphasis. The peculiar communicative context of sentences 2 and 3 concerns the interrogatory of two different people in two different moments, but both were investigated by the police for a crime. Both the answers to the police's questions have as first element something that can save them from the pressing interrogatory: 'Tomorrow I will call!' and 'I did nothing!'. In a real-life context the use of such a device -putting in focus the important element that can save someone from the police -can be understood better.
As for a more 'television' language, example 7 can help understand another reason for a word order change. In this sentence, in fact, the verb -which is usually the last element in Hindī -is in first position, turning upside-down the syntax of the entire utterance. The focus of the sentence, in other words the new information given to the listeners, is kept as the last element apne agle mehman se. With this formulation of the utterance, the attention of the interlocutor -in this case the public -is maintained high until the end.
From the analysis of these first examples, it is clear that putting an important element of the sentence in first position -and at the same time putting the one which should be in first position at the end -gives prominence to that same element. In other cases, when the prominent element of the sentence is put at the end of the sentence the attention of the listeners is kept high. In addition to this, it is clear that every kind of element -whether it is an object, an adguage, but in past participle sentences the subject is followed by the ne postposition, which highlights the agent of the action described by the verb -and is thus ergative. The object thus becomes the grammatical subject and agrees with the verb. The ergative subject is considered a 'non-canonical subject' since the subject is traditionally believed to have some properties, like being in first position, agreeing with the verb etc. In ergative sentences the object, and not the subject, agrees with the verb. The sentences with dative subject as well are considered 'non-canonical' since also here the object agrees with the verb, as the subject is followed by the dative postposition ko. Usually, these sentences are used to convey the meaning of preference ('I like…', 'I do not like…' etc.) and of ability ('I can do…', 'I cannot do…' etc.). For a further examination of the topic of non-canonical subjects see Keenan 1976;Drocco 2008;Montaut 2014. verb, a verb or a complement -can thus be dislocated in first position, or, on the contrary, uttered at the end. While there are some restrictions to word order in some languages (like English, as can be seen from the translations of the examples above), there are no syntactic restrictions to word order in Hindī.
Here These above-mentioned examples have a word order which is definitely similar to the previous examples, but they highlight a topic that still needs to be examined. The last elements of the sentence ('uska', 'hameśā aisā', the interrogative 'kyā' and the syntagm 'is cīz kā') are added even though what the sentence wanted to convey was already clear from the context -in other words, that same element had already been presented in the conversation and was implied in the sentence. For the examples 10 and 11, the element examined is kyā, which is an interrogation mark (not a pronoun, as in example 12). Kyā as interrogation mark (which does not have a translation in English) is usually at the beginning of a sentence, even if it has great mobility within the sentence. 27 For these utterances, specifically, the fact that it was a question would have been clear from the intonation of the utterance even if the interrogative mark were missing. For Montaut, this dislocation gives it the communicative status of an afterthought after the complex utterance […] a delayed topic […] a 'post-rhème '. (2004, 276) The elements in this position can have the meaning, thus, of a later addition, of a reiterated information added to give more precision to something which is presumed to be already understood by the interlocutor.
These last elements are thus in this position for two reasons, and one does not exclude the other: firstly, it is the position they get when an element considered more prominent (the topic or the focus) is shifted to first position. The other ones consequently shift in the other direction (and become background). Secondly, those elements could have been tacit, but they are mentioned to make the sentence clearer and more understandable for the interlocutor.
From this brief excursus just some of the possible dislocations were examined: there are other examples of other elements of the sentence which can be dislocated at the beginning or at the end of the clause -different adverbs, different syntagm, different kinds of objects etc. Here just some of the main ones were brought to light. One issue shall be, nevertheless, clear: there is a great possibility of movement for the different elements of the sentence in Hindī. In the following paragraph, the level of analysis of the lexicon will be examined through examples from the same corpus.
27 Differently, kyā as a pronoun (as in example 12) is always in the position of the element it is referring to. If it is referring to a subject it would thus be at the beginning of the question, while if it is referring to an object in the second position. As well as kyā as interrogation mark, other interrogative particles (like kaisā 'how' and kyoṃ 'why') have great mobility within the sentence. The reason why these elements can be in different positions in the sentence is due -as for other kinds of elements -to the emphasis they have and to their role in the information structure.

Code-Switching and Code-Mixing in Hindī
One of the most distinctive characteristics of Hindī from the TV programmes that were listened, and more in general in the spoken Hindī (see Nespital 1990), was the use of English words and sentences. This widely spread phenomenon of code commutation is usually divided in two main actualisations: 'code-switching' is when there is an entire part of the sentence in another language; 'code-mixing', instead, is when there is a juxtaposition of two languages words and syntagms. Usually, a code-mixed sentence is entirely constructed in a language's grammar and morphology (with pronouns, prepositions/postpositions) but the nominal parts, such as nouns, adverbs, or the nominal parts of the verbs, are in the other language. The diffusion of code-switching and code-mixing is due to the central position of English in India for what concerns politics, but also for economic and social reasons (Si 2010, 390;Abbi, Sharma 2014, 107-10). Moreover, the prestige of English for medium-high classes (Si 2010, 390) and the wider distribution of the language through Bollywood (Dey, Fung 2014, 2410 increased the general use of English also through the phenomenon of code commutation. 28 An example from the analysed corpus for code-mixing is the following: 1. sir, maiṃ pulīs kī high court meṃ represent kar rahā hūṃ is inquiry meṃ Sir maiṃ pulīs k-ī high court meṃ represent kar Sir 1.SG. police GEN.-F. high court LOC. represent fare rah-ā hūṃ is inquiry meṃ to stay-PERF.M.SG. to be.AUX-PRES.3.SG. OBL.3.SG. inquiry LOC. 'Sir, I am representing the police in the high court in this inquiry' (from Delhi Crime, ep. 5) As can be seen from this short example, the English words are nouns (which usually are 'head' 29 or adjectives) and the first part of a verb (as in represent karnā where karnā is the general verb 'to do' usually used to create compound verbs), 30 while the core grammar (pronouns, postpositions) of the sentence is Hindī (meṃ, is, kī).
28 In Hindī also the contrary phenomenon in which English is switched to Hindī exists, as well as code-switching with other Neo-Indo-Aryan languages (see Kachru 1978, 108).
29 The term 'head' concerns the syntactic role a linguistic element can have in a complex structure which either "(a) is in a morphologically marked relationship of coreference with the preceding or following coreferential elements or (b) is modified semantically by these coreferential elements as attributes" (Bussmann 1996, 502).
30 These kinds of verbal compounds exist because of the limited number of simple verb lexemes in Hindī (as well as in other Neo-Indo-Aryan languages), around 1200. Differently from Sanskrit, but also from other Indo-European languages (such as Neo-Latin, Slavonic languages or German) Hindī and the Neo-Indo-Aryan languages do not use prefixes to create new verbs with new meanings. The linguistic strategy used to do this lies in these verbo-nominal expressions, namely syntagms created with the jux-