Academia.eduAcademia.edu

Figure 12 – uploaded by Arvind Iyengar

See full PDFdownloadDownload figure
Figure 5.2. Triscriptal Sindhi-language signboard in Mumbai = VY Sn = Similar variation prevails with regard to the vowel signs for /e/ and /9/. Th ‘ommonly used forms for these vowels are (.¢) and (93), respectively, but (¢¢¢: ind (§:) are also seen (Lekhwani, 1996, p. vi). Figure 5.2 shows a signboard a in intersection in the Khar suburb of Mumbai, named after an Indian Sindh ‘ducationist. The signboard displays the name of the intersection transcribe nto Devanagari, Perso-Arabic and Roman. In Perso-Arabic, the Sindhi wor {65k°] ‘square, intersection’ has been spelt (<j9>) (or (43 5>), were vowe liacritics to be included). In contrast, Mewaram (1910, p. 211) and Lekhwan 1996, p. 57) both spell the same word (43 95).

Figure 5 2. Triscriptal Sindhi-language signboard in Mumbai = VY Sn = Similar variation prevails with regard to the vowel signs for /e/ and /9/. Th ‘ommonly used forms for these vowels are (.¢) and (93), respectively, but (¢¢¢: ind (§:) are also seen (Lekhwani, 1996, p. vi). Figure 5.2 shows a signboard a in intersection in the Khar suburb of Mumbai, named after an Indian Sindh ‘ducationist. The signboard displays the name of the intersection transcribe nto Devanagari, Perso-Arabic and Roman. In Perso-Arabic, the Sindhi wor {65k°] ‘square, intersection’ has been spelt (<j9>) (or (43 5>), were vowe liacritics to be included). In contrast, Mewaram (1910, p. 211) and Lekhwan 1996, p. 57) both spell the same word (43 95).

Related Figures (51)

Figure 1.1. Banner in Pune for the Chett Chand" festival, with Sindhi written in Perso-Arabic (centre) and Devanagari (right)
Figure 1.1. Banner in Pune for the Chett Chand" festival, with Sindhi written in Perso-Arabic (centre) and Devanagari (right)
Figure 2.1. Extent of the Indus civilisation at its peak Harappan Phase”; see Figure 2.1), and was highly advanced for its time. 3-4). The civilisation reached its pinnacle between 2600 and 1900 BC (“Mature Harappan Phase”; see Figure 2.1), and was highly advanced for its time. 3-4). The civilisation reached its pinnacle between 2600 and 1900 BC (“Mature
Figure 2.1. Extent of the Indus civilisation at its peak Harappan Phase”; see Figure 2.1), and was highly advanced for its time. 3-4). The civilisation reached its pinnacle between 2600 and 1900 BC (“Mature Harappan Phase”; see Figure 2.1), and was highly advanced for its time. 3-4). The civilisation reached its pinnacle between 2600 and 1900 BC (“Mature
in rural areas of Sindh did native Sindhi speakers form a clear majority (92 percent). Of the urban population of Sindh, only 25.8 percent was native Sindhi-speaking. In contrast, the proportion of native Urdt speakers in urban Sindh was almost 42 percent. In Karachi, the proportion of native Urdu speakers was slightly higher, at 48.5 percent, but that of native Sindhi speakers was less than eight percent (Blank, Clary, & Nichiporuk, 2014, pp. 18-19). Consequently, Sindh has seen several instances of Sindhi-Muhdajir tensions erupting into violence, not least over the issue of language (Shackle, 2014). For an overview of ethnolinguistic conflicts in the recent past in Sindh and Pakistan, see Rahman (1995, 1999), Ayres (2009) and Siddiqi (2012).
in rural areas of Sindh did native Sindhi speakers form a clear majority (92 percent). Of the urban population of Sindh, only 25.8 percent was native Sindhi-speaking. In contrast, the proportion of native Urdt speakers in urban Sindh was almost 42 percent. In Karachi, the proportion of native Urdu speakers was slightly higher, at 48.5 percent, but that of native Sindhi speakers was less than eight percent (Blank, Clary, & Nichiporuk, 2014, pp. 18-19). Consequently, Sindh has seen several instances of Sindhi-Muhdajir tensions erupting into violence, not least over the issue of language (Shackle, 2014). For an overview of ethnolinguistic conflicts in the recent past in Sindh and Pakistan, see Rahman (1995, 1999), Ayres (2009) and Siddiqi (2012).
From Google Maps, n.d. (http://maps.google.com). Copyright 2013 by Google.
From Google Maps, n.d. (http://maps.google.com). Copyright 2013 by Google.
Figure 2.4. Classification of Sindhi within Indo-Aryan
Figure 2.4. Classification of Sindhi within Indo-Aryan
Figure 2.5. Distribution of Sindhi dialects in Sindh and environs
Figure 2.5. Distribution of Sindhi dialects in Sindh and environs
Table 2.1. Phonemic inventory of consonants in standard Sindhi Sounds enclosed in parentheses in Table 2.1 are those listed by Nihalani (1999) as independent phonemes, but by Cole (2001) as phonemically ambiguous. Cole notes that it is difficult to claim independent phoneme status for [m4], [n*], [n*], [1*], [r*] and [v8], since they do not contrast with a sonorant+/h/ cluster. Knubchandani (2007) too does not list the aforementioned sounds as independent phonemes in Sindhi. This is also the perspective reflected in the Perso-Arabic Sindhi orthography, in that each of these sounds is written with two separate characters. In addition, [,] and [;*] freely alternate with [cr] and [r*] in certain speakers’ idiolects (Nihalani, 1978, p. 103). Khubchandani also lists a voiceless uvular stop /q/ as a marginal phoneme, albeit mentioning that it is only found in the “formal speech of Persian-oriented speakers” (2007, p. 689). Allophones are shown enclosed in square brackets. [n] is an allophone of /n/ preceding dental stops in consonant clusters, as in /dando/ [dand®] ‘tooth’. [f] is an intervocalic allophone of /n/, as in /man*u/ [majf*u] ‘man’. This is reflected in the observation by early grammarians that /n/ was occasionally interchangeable with /r/ (Trumpp, 1872, p. 16). [w] and [w®] tend to be the realisations of /v/ and /v4/ before back vowels, although this is subject to high idiolectal variation.
Table 2.1. Phonemic inventory of consonants in standard Sindhi Sounds enclosed in parentheses in Table 2.1 are those listed by Nihalani (1999) as independent phonemes, but by Cole (2001) as phonemically ambiguous. Cole notes that it is difficult to claim independent phoneme status for [m4], [n*], [n*], [1*], [r*] and [v8], since they do not contrast with a sonorant+/h/ cluster. Knubchandani (2007) too does not list the aforementioned sounds as independent phonemes in Sindhi. This is also the perspective reflected in the Perso-Arabic Sindhi orthography, in that each of these sounds is written with two separate characters. In addition, [,] and [;*] freely alternate with [cr] and [r*] in certain speakers’ idiolects (Nihalani, 1978, p. 103). Khubchandani also lists a voiceless uvular stop /q/ as a marginal phoneme, albeit mentioning that it is only found in the “formal speech of Persian-oriented speakers” (2007, p. 689). Allophones are shown enclosed in square brackets. [n] is an allophone of /n/ preceding dental stops in consonant clusters, as in /dando/ [dand®] ‘tooth’. [f] is an intervocalic allophone of /n/, as in /man*u/ [majf*u] ‘man’. This is reflected in the observation by early grammarians that /n/ was occasionally interchangeable with /r/ (Trumpp, 1872, p. 16). [w] and [w®] tend to be the realisations of /v/ and /v4/ before back vowels, although this is subject to high idiolectal variation.
Table 2.2. Examples of voiced implosive stops in Sindhi
Table 2.2. Examples of voiced implosive stops in Sindhi
Table 2.3. Phonemic inventory of vowels in standard Sindhi Certain environments can cause vowels to change their quality; for instance, the sequence /afi/ is realised allophonically as [eh®] (Grierson, 1919, p. 22; Trumpp, 1872, p. x). RS See Pee Tee ey, Pe ee An oft-cited feature of Sindhi is its vowel-finality. Some authors assert that all Sindhi words are vowel-final (Bughio, 2006, p. 98; Cole, n.d.-b; Grierson 1919, p. 22; Hardwani, 1991, p. iii). Others state that indigenous Sindhi words are vowel-final, while certain consonant-final loanwords that have not yet been phonologically indigenised remain as they are (Khubchandani, 2007, pp. 691. 701). Conversely, if a loanword that is consonant-final in the source language is considered to have been assimilated into Sindhi, it may have a lax vowel suffixed. This word-final lax vowel manifests in a ‘reduced’ form (see Table 2.4).
Table 2.3. Phonemic inventory of vowels in standard Sindhi Certain environments can cause vowels to change their quality; for instance, the sequence /afi/ is realised allophonically as [eh®] (Grierson, 1919, p. 22; Trumpp, 1872, p. x). RS See Pee Tee ey, Pe ee An oft-cited feature of Sindhi is its vowel-finality. Some authors assert that all Sindhi words are vowel-final (Bughio, 2006, p. 98; Cole, n.d.-b; Grierson 1919, p. 22; Hardwani, 1991, p. iii). Others state that indigenous Sindhi words are vowel-final, while certain consonant-final loanwords that have not yet been phonologically indigenised remain as they are (Khubchandani, 2007, pp. 691. 701). Conversely, if a loanword that is consonant-final in the source language is considered to have been assimilated into Sindhi, it may have a lax vowel suffixed. This word-final lax vowel manifests in a ‘reduced’ form (see Table 2.4).
Table 2.4. Assimilated loanwords in Sindhi with suffixed reduced vowels
Table 2.4. Assimilated loanwords in Sindhi with suffixed reduced vowels
Table 2.5. Masculine and feminine nouns in Sindhi with typical final vowels
Table 2.5. Masculine and feminine nouns in Sindhi with typical final vowels
Table 2.6. Selection of masculine and feminine noun cases in Sindhi
Table 2.6. Selection of masculine and feminine noun cases in Sindhi
Table 2.7. Declinable and indeclinable adjectives in Sindhi Of late, a few morphological changes have emerged in Sindhi usage in India. In the literary variety, this primarily involves replacing vernacular Sindhi adjectival declensions with Sanskritic adjectival suffixes. Lekhwani (2011, p. 34) illustrates this phenomenon with the example of the Sindhi word [samadz°) ‘society, community’. While this word has traditionally been adjectivised as [samadzi] ‘societal’, recent Indian practice is to use the Sanskritised adjectival form [samadzrk®]. In the spoken variety, morphological changes are most evident in the speech of the younger generation. For instance, Parchani (1998, p. 20) observes that the negative copula [nahe] ‘is not’ may be realised by younger Sindhi speakers as [na ahe], where the negative particle [na] ‘not’ is simply juxtaposed with the copula [ahe] ‘is’, rather than phonologically merged with it. Such changes are the morphological counterparts to the intergenerational phonological changes described earlier (§ 2.2.3).
Table 2.7. Declinable and indeclinable adjectives in Sindhi Of late, a few morphological changes have emerged in Sindhi usage in India. In the literary variety, this primarily involves replacing vernacular Sindhi adjectival declensions with Sanskritic adjectival suffixes. Lekhwani (2011, p. 34) illustrates this phenomenon with the example of the Sindhi word [samadz°) ‘society, community’. While this word has traditionally been adjectivised as [samadzi] ‘societal’, recent Indian practice is to use the Sanskritised adjectival form [samadzrk®]. In the spoken variety, morphological changes are most evident in the speech of the younger generation. For instance, Parchani (1998, p. 20) observes that the negative copula [nahe] ‘is not’ may be realised by younger Sindhi speakers as [na ahe], where the negative particle [na] ‘not’ is simply juxtaposed with the copula [ahe] ‘is’, rather than phonologically merged with it. Such changes are the morphological counterparts to the intergenerational phonological changes described earlier (§ 2.2.3).
Figure 4.1 summarises the distribution of participants across age groups. Figure 4.1. Number and gender of participants across three age groups PELOU-AALANIC OUILOLL, WELC SUNUNVUTTIALLY Ledlistd. A total of 50 participants—31 males and 19 females—were interviewed. Th youngest participant was 28, and the oldest 85. Based on their age, participant were categorised into the youngest (ages 28-44), middle-aged (ages 45-64) an oldest (ages 65-85) generations. This three-generation classification was baset on a similar concept adopted by Daswani and Parchani (1978) in their classi sociolinguistic study on Indian Sindhi. Of the participants interviewed, 20 wer of the youngest generation, 12 were middle-aged, and 18 were of the oldes veneration. Among the oldest generation, 16 were born in pre-Partition Sindh Figure 4.1 summarises the distribution of participants across age groups.
Figure 4.1 summarises the distribution of participants across age groups. Figure 4.1. Number and gender of participants across three age groups PELOU-AALANIC OUILOLL, WELC SUNUNVUTTIALLY Ledlistd. A total of 50 participants—31 males and 19 females—were interviewed. Th youngest participant was 28, and the oldest 85. Based on their age, participant were categorised into the youngest (ages 28-44), middle-aged (ages 45-64) an oldest (ages 65-85) generations. This three-generation classification was baset on a similar concept adopted by Daswani and Parchani (1978) in their classi sociolinguistic study on Indian Sindhi. Of the participants interviewed, 20 wer of the youngest generation, 12 were middle-aged, and 18 were of the oldes veneration. Among the oldest generation, 16 were born in pre-Partition Sindh Figure 4.1 summarises the distribution of participants across age groups.
Of the 50 participants, 44 had spent all their childhood years in India, or in Sindh followed by post-Partition India. Of these, 25 had spent the majority of their childhood in large metropolises. These included 22 who had grown up in Bangalore, Delhi, Mumbai or Pune, and three older participants who had spent some or all of their school years in Karachi. In contrast, 19 participants had grown up in smaller cities, including Ulhasnagar, Gandhidham, Ajmer and Hyderabad (Sindh). Five had split their childhood between India and another country, and one had grown up entirely in Malaysia. The distribution of participants’ childhood locations is shown in Figure 4.2. Figure 4.2. Locations where participants spent their childhood years
Of the 50 participants, 44 had spent all their childhood years in India, or in Sindh followed by post-Partition India. Of these, 25 had spent the majority of their childhood in large metropolises. These included 22 who had grown up in Bangalore, Delhi, Mumbai or Pune, and three older participants who had spent some or all of their school years in Karachi. In contrast, 19 participants had grown up in smaller cities, including Ulhasnagar, Gandhidham, Ajmer and Hyderabad (Sindh). Five had split their childhood between India and another country, and one had grown up entirely in Malaysia. The distribution of participants’ childhood locations is shown in Figure 4.2. Figure 4.2. Locations where participants spent their childhood years
2000) can also be partly attributed to its high levels of education. The exceptior to this rule was the oldest participant, who was 85 years old at the time of the interview. He had dropped out of school in Sindh after Year 2, due to economi constraints, but had gone on to become a wealthy businessman. Mos‘ importantly, he was literate in both Perso-Arabic and Devanagari Sindhi. Those participants who did not specify their levels of education were all over the age of 70, and were either retired or working part-time. Figure 4.3 shows the relative levels of education among participants. Despite the high levels of education, only 19 of the 50 participants had received any education in Sindhi, either as a medium of instruction or a school subject. Of these participants, 17 were 65 years of age or older. This was indicative of the decline in demand for the language in education and its consequent availability in schools. In other words, the only participants to have received any formal education in Sindhi were those of school-going age during the Partition era, before the mushrooming of private English-medium schools in Indian cities. Significantly, the two younger participants who had some Sindhi-medium education were in fact lecturers of Sindhi in universities in Mumbai and Pune. Both had Master’s degrees in the language. The other 31 participants had received no formal or informal education in Sindhi. ee oe: Oe i: i i ; . . , oer ae ee , , : i ie |
2000) can also be partly attributed to its high levels of education. The exceptior to this rule was the oldest participant, who was 85 years old at the time of the interview. He had dropped out of school in Sindh after Year 2, due to economi constraints, but had gone on to become a wealthy businessman. Mos‘ importantly, he was literate in both Perso-Arabic and Devanagari Sindhi. Those participants who did not specify their levels of education were all over the age of 70, and were either retired or working part-time. Figure 4.3 shows the relative levels of education among participants. Despite the high levels of education, only 19 of the 50 participants had received any education in Sindhi, either as a medium of instruction or a school subject. Of these participants, 17 were 65 years of age or older. This was indicative of the decline in demand for the language in education and its consequent availability in schools. In other words, the only participants to have received any formal education in Sindhi were those of school-going age during the Partition era, before the mushrooming of private English-medium schools in Indian cities. Significantly, the two younger participants who had some Sindhi-medium education were in fact lecturers of Sindhi in universities in Mumbai and Pune. Both had Master’s degrees in the language. The other 31 participants had received no formal or informal education in Sindhi. ee oe: Oe i: i i ; . . , oer ae ee , , : i ie |
as ‘scholars’ in this study. Nine of the scholars belonged of the oldest generation. Figure 4.4 shows the distribution of participants’ claimed Sindhi fluency on a scale of 1 to 5, where 1 represents bare minimum knowledge and 5 fluency. Figure 4.4. Breakup of participants according to Sindhi fluency CHIE LEGALS UWVU OLLIUIAGLO WELLE LAL ULILY VVALEE LLLe LALLEL. In terms of Devanagari Sindhi abilities, ascertaining participant abilities was problematic. All but one participant claimed to be able to read Devanagari. However, their Devanagari ability was largely due to literacy in Hindi. In fact, 14 participants were not aware that Sindhi was also written in Devanagari. Seven were aware of the existence of Devanagari Sindhi, but had never read any. Of the 28 that had had exposure to Devanagari Sindhi, only four had received formal education in it, at least as a school subject. Seven were fluent in Devanagari Sindhi despite not having studied in it, due to extensive reading and writing on a personal level. 17 indicated that they had some reading experience in Devanagari Sindhi, in some cases extremely limited. 14 participants were not aware that Sindhi was also written in Devanagari.
as ‘scholars’ in this study. Nine of the scholars belonged of the oldest generation. Figure 4.4 shows the distribution of participants’ claimed Sindhi fluency on a scale of 1 to 5, where 1 represents bare minimum knowledge and 5 fluency. Figure 4.4. Breakup of participants according to Sindhi fluency CHIE LEGALS UWVU OLLIUIAGLO WELLE LAL ULILY VVALEE LLLe LALLEL. In terms of Devanagari Sindhi abilities, ascertaining participant abilities was problematic. All but one participant claimed to be able to read Devanagari. However, their Devanagari ability was largely due to literacy in Hindi. In fact, 14 participants were not aware that Sindhi was also written in Devanagari. Seven were aware of the existence of Devanagari Sindhi, but had never read any. Of the 28 that had had exposure to Devanagari Sindhi, only four had received formal education in it, at least as a school subject. Seven were fluent in Devanagari Sindhi despite not having studied in it, due to extensive reading and writing on a personal level. 17 indicated that they had some reading experience in Devanagari Sindhi, in some cases extremely limited. 14 participants were not aware that Sindhi was also written in Devanagari.
Consonants are represented as simple base graphemes. A consonant that is not followed by a vowel is simply shown as a bare consonant grapheme. If the absence of a vowel needs to be explicitly indicated—for instance, in primers— then the diacritic (:4) is placed atop the consonant grapheme. However, there is disagreement among Sindhi grammarians on the possibility of vowelless consonants in word-medial and final positions. This is explained further in Section 5.1.3.
Consonants are represented as simple base graphemes. A consonant that is not followed by a vowel is simply shown as a bare consonant grapheme. If the absence of a vowel needs to be explicitly indicated—for instance, in primers— then the diacritic (:4) is placed atop the consonant grapheme. However, there is disagreement among Sindhi grammarians on the possibility of vowelless consonants in word-medial and final positions. This is explained further in Section 5.1.3.
Table 5.2. Representation of vowels in Perso-Arabic Sindhi Graphemes for Sindhi-specific phonemes have been created by adding or repositioning dots and diacritics on graphemes used in Arabic or Persian (Salomon, 2007, p. 110; see Table 5.3).
Table 5.2. Representation of vowels in Perso-Arabic Sindhi Graphemes for Sindhi-specific phonemes have been created by adding or repositioning dots and diacritics on graphemes used in Arabic or Persian (Salomon, 2007, p. 110; see Table 5.3).
Table 5.3. Selection of Sindhi-specific graphemes derived from pre-existing Arabic graphemes
Table 5.3. Selection of Sindhi-specific graphemes derived from pre-existing Arabic graphemes
Table 5.4. Arabic and Sindhi phonemic values of certain Perso-Arabic graphemes A logical extension of graphemic retention is orthographic retention. This refers to the retention of source-language orthography for loanwords in the target language, when both the source and target languages use the same script. Share and Daniels (2016) refer to this phenomenon as ORTHOGRAPHIC INERTIA. Sindhi is a noteworthy case of a language that displays orthographic inertia both in its Perso-Arabic and Devanagari scripts. An example of orthographic inertia in Perso-Arabic Sindhi is that of the Arabic loanword (aulsi) ‘education’ In standard Arabic, this word is pronounced [ta‘li:m]. However, in Sindhi, its pronunciation is indigenised to [telim®]. In general, loanwords from Arabic and Persian retain their original orthography in Sindhi, apart from minor adaptations to Sindhi morphophonology. This retention of source orthography sometimes serves a disambiguating purpose, as can be seen in Table 5.5.
Table 5.4. Arabic and Sindhi phonemic values of certain Perso-Arabic graphemes A logical extension of graphemic retention is orthographic retention. This refers to the retention of source-language orthography for loanwords in the target language, when both the source and target languages use the same script. Share and Daniels (2016) refer to this phenomenon as ORTHOGRAPHIC INERTIA. Sindhi is a noteworthy case of a language that displays orthographic inertia both in its Perso-Arabic and Devanagari scripts. An example of orthographic inertia in Perso-Arabic Sindhi is that of the Arabic loanword (aulsi) ‘education’ In standard Arabic, this word is pronounced [ta‘li:m]. However, in Sindhi, its pronunciation is indigenised to [telim®]. In general, loanwords from Arabic and Persian retain their original orthography in Sindhi, apart from minor adaptations to Sindhi morphophonology. This retention of source orthography sometimes serves a disambiguating purpose, as can be seen in Table 5.5.
Table 5.5. Homophonous lexemes in Perso-Arabic Sindhi
Table 5.5. Homophonous lexemes in Perso-Arabic Sindhi
Table 5.6. Homographic lexemes in Perso-Arabic Sindhi It is evident that the onus is on the reader to disambiguate homographs from the context. Such disambiguation necessitates fluency in the language. It may be argued that there is a semblant advantage to omitting lax vowel diacritics in Perso-Arabic Sindhi orthography. Since reduced lax vowels are present in old variety Sindhi but largely absent in the new variety (§ 2.2.3), omitting lax vowel diacritics in the orthography allows the reader to mentally add reduced lax vowels according to their own idiolect. That is, it acts as a hypodifferentiated standard (Cerron-Palomino, 1991, p. 36) that allows for words to be pronounced according to the old or new variety chronolect, while maintaining a cross-chronolectal orthographical standard. However, this is effectively a moot point, since most Sindhi new variety speakers in India are nonliterate in the Perso-Arabic script.
Table 5.6. Homographic lexemes in Perso-Arabic Sindhi It is evident that the onus is on the reader to disambiguate homographs from the context. Such disambiguation necessitates fluency in the language. It may be argued that there is a semblant advantage to omitting lax vowel diacritics in Perso-Arabic Sindhi orthography. Since reduced lax vowels are present in old variety Sindhi but largely absent in the new variety (§ 2.2.3), omitting lax vowel diacritics in the orthography allows the reader to mentally add reduced lax vowels according to their own idiolect. That is, it acts as a hypodifferentiated standard (Cerron-Palomino, 1991, p. 36) that allows for words to be pronounced according to the old or new variety chronolect, while maintaining a cross-chronolectal orthographical standard. However, this is effectively a moot point, since most Sindhi new variety speakers in India are nonliterate in the Perso-Arabic script.
Figure 5.1. Variable vowel graphemes in Perso-Arabic Sindhi orthography The question of representing reduced vowels also surfaces with regard to medial consonant clusters. Since there is great idiosyncratic variation in the pronunciation of consonant clusters, there is corresponding variation in the orthography. This uncertainty regarding orthographic representation of consonant clusters occasionally results in variant orthographic practices by the same author and within the same text. Figure 5.1 offers an example of such practice. This is an image from the website of the Karachi-based Sindhi media group Awami Awaz, advertising a programme entitled [trk™ya miya dal™ji], with the name spelt in Perso-Arabic as (7920S ¢;ci0 7)§.S3).
Figure 5.1. Variable vowel graphemes in Perso-Arabic Sindhi orthography The question of representing reduced vowels also surfaces with regard to medial consonant clusters. Since there is great idiosyncratic variation in the pronunciation of consonant clusters, there is corresponding variation in the orthography. This uncertainty regarding orthographic representation of consonant clusters occasionally results in variant orthographic practices by the same author and within the same text. Figure 5.1 offers an example of such practice. This is an image from the website of the Karachi-based Sindhi media group Awami Awaz, advertising a programme entitled [trk™ya miya dal™ji], with the name spelt in Perso-Arabic as (7920S ¢;ci0 7)§.S3).
Note: Collation order based on Khubchandani (2007, p. 698)
Note: Collation order based on Khubchandani (2007, p. 698)
Table 5.9. Representation of vowels in Devanagari Sindhi Unlike Perso-Arabic Sindhi, Devanagari Sindhi requires all vowels to be explicitly represented. Only the inherent vowel, when following a consonant, remains unwritten and implicit.
Table 5.9. Representation of vowels in Devanagari Sindhi Unlike Perso-Arabic Sindhi, Devanagari Sindhi requires all vowels to be explicitly represented. Only the inherent vowel, when following a consonant, remains unwritten and implicit.
Table 5.10. Devanagari Sindhi graphemes for plosives and implosives The Devanagari Sindhi graphemic inventory is inspired by that of Hindi, which itself is based on the graphemic inventory as used for Sanskrit. Graphemes for phonemes absent in Sanskrit have been created by adding an underdot diacritic to existing graphemes. For instance, both standard Sindhi and Hindi have the phonemes /z/ and /f/, which classical Sanskrit does not. Hence, graphemes for these phonemes have been created by adding an underdot to the graphemes for /dz/ and /p/, respectively (see Table 5.8). Graphemes for the Sindhi implosives have been created by adding an underline diacritic to the corresponding plosives (see Table 5.10).
Table 5.10. Devanagari Sindhi graphemes for plosives and implosives The Devanagari Sindhi graphemic inventory is inspired by that of Hindi, which itself is based on the graphemic inventory as used for Sanskrit. Graphemes for phonemes absent in Sanskrit have been created by adding an underdot diacritic to existing graphemes. For instance, both standard Sindhi and Hindi have the phonemes /z/ and /f/, which classical Sanskrit does not. Hence, graphemes for these phonemes have been created by adding an underdot to the graphemes for /dz/ and /p/, respectively (see Table 5.8). Graphemes for the Sindhi implosives have been created by adding an underline diacritic to the corresponding plosives (see Table 5.10).
Table 5.11. Representation of [?] in Devanagari Sindhi orthography It should be noted that the Sindhi new variety pronunciations tend to be closer to the Hindi pronunciations shown in Table 5.11. oncerns the representation of the reduced lax vowels [®], [‘] and [°], which are argely absent from the phonemic inventories of most other languages that use Jevanagari.’® As mentioned earlier, the diacritic () 1s placed below the base ‘rapheme if the inherent vowel sound /a/ needs to be muted and a pure ‘onsonant sound is to be represented. However, in Hindi, the vowel /a/ does not yccur word-finally (Gumperz, 1958, p. 216), except in some Sanskrit neologisms. ‘or this reason, the diacritic (*:) is typically not used in word-final position in Jevanagari Hindi orthography. Rather, the reader ignores all orthographic inal schwas, and interprets such words as consonant-final. In Sindhi, on the ther hand, the vowel /a/ does occur word-finally, usually in the reduced orm [?]. Thus, in Devanagari Sindhi, if a final schwa exists in the orthography, t is pronounced. Table 5.11 shows a few words common to both Sindhi and lindi. In both languages, they have the same orthography and meaning. iowever, they differ in pronunciation in terms of the final schwa. However, they differ in pronunciation in terms of the final schwa.
Table 5.11. Representation of [?] in Devanagari Sindhi orthography It should be noted that the Sindhi new variety pronunciations tend to be closer to the Hindi pronunciations shown in Table 5.11. oncerns the representation of the reduced lax vowels [®], [‘] and [°], which are argely absent from the phonemic inventories of most other languages that use Jevanagari.’® As mentioned earlier, the diacritic () 1s placed below the base ‘rapheme if the inherent vowel sound /a/ needs to be muted and a pure ‘onsonant sound is to be represented. However, in Hindi, the vowel /a/ does not yccur word-finally (Gumperz, 1958, p. 216), except in some Sanskrit neologisms. ‘or this reason, the diacritic (*:) is typically not used in word-final position in Jevanagari Hindi orthography. Rather, the reader ignores all orthographic inal schwas, and interprets such words as consonant-final. In Sindhi, on the ther hand, the vowel /a/ does occur word-finally, usually in the reduced orm [?]. Thus, in Devanagari Sindhi, if a final schwa exists in the orthography, t is pronounced. Table 5.11 shows a few words common to both Sindhi and lindi. In both languages, they have the same orthography and meaning. iowever, they differ in pronunciation in terms of the final schwa. However, they differ in pronunciation in terms of the final schwa.
Table 5.12. Representation of ['] and [°] in Devanagari Sindhi orthography
Table 5.12. Representation of ['] and [°] in Devanagari Sindhi orthography
Table 5.13. Sanskrit and Sindhi phonemic values of certain Devanagari characters
Table 5.13. Sanskrit and Sindhi phonemic values of certain Devanagari characters
Table 5.14. Realisation of orthographic final (i) in Hindi and Devanagari Sindhi
Table 5.14. Realisation of orthographic final (i) in Hindi and Devanagari Sindhi
Table 5.15. Realisation of orthographic final (u) in Hindi and Devanagari Sindhi In summary, retaining Sanskritic spellings in Devanagari Sindhi ensures a common orthography for these words across Sindhi and Hindi. This has potential advantages for a Hindi-literate reader or learner of Sindhi. However, Sanskritic spellings often conflict with Devanagari Sindhi orthographic rules, requiring such spellings to be treated as exceptions. This results in an increased learning curve.
Table 5.15. Realisation of orthographic final (u) in Hindi and Devanagari Sindhi In summary, retaining Sanskritic spellings in Devanagari Sindhi ensures a common orthography for these words across Sindhi and Hindi. This has potential advantages for a Hindi-literate reader or learner of Sindhi. However, Sanskritic spellings often conflict with Devanagari Sindhi orthographic rules, requiring such spellings to be treated as exceptions. This results in an increased learning curve.
Table 6.1. Differences between Griersonian and Romanized Sindhi orthographies As seen in Table 6.1, the RST’s orthography uses additional symbols as free- standing diacritics, which can be termed POSTLITERAL diacritics. This differs from the practice of CIRCUMLITERAL diacritics, that is, those positioned above or below the base letter, as is traditionally the practice. In principle, usage of postliteral free-standing diacritics is not uncommon in Roman orthographies worldwide. For instance, the apostrophe symbol is widely used as a postlitera free-standing diacritic to indicate ejective and palatalised consonants, among others (Baker, 1997, p. 103). The colon symbol has been attested as a postlitera tone diacritic in a few languages of Papua New Guinea (Priest & Constable, 2006, p. 7). That said, the symbols used by the RST—the circumflex and tilde— seem to be rarely used as postliteral diacritics in Roman orthographies. Rather, these two symbols appear more frequently as circumliteral diacritics. Thus, while the principle of using postliteral diacritics is not unconventional, the chonira nf the RCT’e eyyzmhoale ic
Table 6.1. Differences between Griersonian and Romanized Sindhi orthographies As seen in Table 6.1, the RST’s orthography uses additional symbols as free- standing diacritics, which can be termed POSTLITERAL diacritics. This differs from the practice of CIRCUMLITERAL diacritics, that is, those positioned above or below the base letter, as is traditionally the practice. In principle, usage of postliteral free-standing diacritics is not uncommon in Roman orthographies worldwide. For instance, the apostrophe symbol is widely used as a postlitera free-standing diacritic to indicate ejective and palatalised consonants, among others (Baker, 1997, p. 103). The colon symbol has been attested as a postlitera tone diacritic in a few languages of Papua New Guinea (Priest & Constable, 2006, p. 7). That said, the symbols used by the RST—the circumflex and tilde— seem to be rarely used as postliteral diacritics in Roman orthographies. Rather, these two symbols appear more frequently as circumliteral diacritics. Thus, while the principle of using postliteral diacritics is not unconventional, the chonira nf the RCT’e eyyzmhoale ic
Table 6.2. Global and Organising themes from the data Of the above, the entire theme of orthographic aspects is devoted to data on participants’ reading of the Sindhi-language sample texts shown to them (see Appendix C), in the scripts they were literate in. In addition to interview excerpts, this section also includes numerous examples of nonstandard participant pronunciation of certain words from the sample texts. It is emphasised that the reading exercise was an unstructured one, intended to gauge participants’ overall feel and reaction to Devanagari and Roman Sindhi orthography. The reading task was not designed for a quantitative analysis of pronunciation patterns. Besides, several participants, especially those with limited spoken Sindhi skills, stopped reading one or more of the texts after the first paragraph. Two participants of the oldest generation were unable to read the sample texts due to age-related eyesight issues. All these lacunae make a statistical or quantitative approach to the subject matter unsuitable. Hence, the
Table 6.2. Global and Organising themes from the data Of the above, the entire theme of orthographic aspects is devoted to data on participants’ reading of the Sindhi-language sample texts shown to them (see Appendix C), in the scripts they were literate in. In addition to interview excerpts, this section also includes numerous examples of nonstandard participant pronunciation of certain words from the sample texts. It is emphasised that the reading exercise was an unstructured one, intended to gauge participants’ overall feel and reaction to Devanagari and Roman Sindhi orthography. The reading task was not designed for a quantitative analysis of pronunciation patterns. Besides, several participants, especially those with limited spoken Sindhi skills, stopped reading one or more of the texts after the first paragraph. Two participants of the oldest generation were unable to read the sample texts due to age-related eyesight issues. All these lacunae make a statistical or quantitative approach to the subject matter unsuitable. Hence, the
Table 6.3. Nonstandard vowel pronunciations in Devanagari text Inconsistent lax versus tense vowels in Roman. Overall, pronunciation of reduced vowels was more consistent when reading the Roman Sindhi text. In accordance with Grierson’s (1919) convention, unstressed vowels in the Roman Sindhi text (see Appendix C) were indicated with superscripts. Before reading the text, participants were also instructed to pronounce these superscript vowel letters very lightly. The iconic representation of reduced vowels as small- sized superscripts, coupled with new variety speakers’ tendency to drop these vowels, likely helped them avoid overpronouncing these vowels. On the other hand. the distinction between lax and tense vowels was not as Inconsistent lax versus tense vowels in Roman. Overall, pronunciation of reduced
Table 6.3. Nonstandard vowel pronunciations in Devanagari text Inconsistent lax versus tense vowels in Roman. Overall, pronunciation of reduced vowels was more consistent when reading the Roman Sindhi text. In accordance with Grierson’s (1919) convention, unstressed vowels in the Roman Sindhi text (see Appendix C) were indicated with superscripts. Before reading the text, participants were also instructed to pronounce these superscript vowel letters very lightly. The iconic representation of reduced vowels as small- sized superscripts, coupled with new variety speakers’ tendency to drop these vowels, likely helped them avoid overpronouncing these vowels. On the other hand. the distinction between lax and tense vowels was not as Inconsistent lax versus tense vowels in Roman. Overall, pronunciation of reduced
Table 6.4. Nonstandard pronunciations of vowels in Roman text A plausible explanation for the pronunciation of (jhalainda) and (gainda) as [dz‘ulenda] and [genda], respectively, is the ignoring of diacritics. That is, the
Table 6.4. Nonstandard pronunciations of vowels in Roman text A plausible explanation for the pronunciation of (jhalainda) and (gainda) as [dz‘ulenda] and [genda], respectively, is the ignoring of diacritics. That is, the
Table 6.5. Nonstandard pronunciations of /z/ in Devanagari text The words [zaruri] and [mazbut] also exist in Hindi, and are often realised with the nonstandard affricated pronunciations [dzaruci] and [madzbut]. Hence, the affricated pronunciations of these Sindhi words could also be attributed to Hindi interference. However, [uzve] has no common parallel in Hindi. Therefore, the affricated pronunciation of this word can only be attributed to the grapheme (9) /z/ being interpreted as (st) /dz/ due to visual similarity. Indeed, of all words in the text containing (3) /z/, [uzve] was affricated most frequently by participants.
Table 6.5. Nonstandard pronunciations of /z/ in Devanagari text The words [zaruri] and [mazbut] also exist in Hindi, and are often realised with the nonstandard affricated pronunciations [dzaruci] and [madzbut]. Hence, the affricated pronunciations of these Sindhi words could also be attributed to Hindi interference. However, [uzve] has no common parallel in Hindi. Therefore, the affricated pronunciation of this word can only be attributed to the grapheme (9) /z/ being interpreted as (st) /dz/ due to visual similarity. Indeed, of all words in the text containing (3) /z/, [uzve] was affricated most frequently by participants.
Table 6.6. Nonstandard pronunciations of stops in Roman text the initial retrotlex stop In [[Irengo} with the dental stop [Tt]. The case of [dz‘endo] being realised as [dz*endo], featuring similar replacement of retroflex stops with dental ones, is a curious one. Again, this Sindhi word is very similar to its Hindi counterpart, /dz*anda/, differing only in the final vowel. Hence, the substitution of retroflex stops with dental ones cannot be attributed to Hindi influence in this case. Similar is the case of [{re ~ te] being pronounced [te], since the phonetic shape of this Sindhi word is quite different from that of its Hindi equivalent /tin/. Hence, the confusion of retroflex and dental stops in [dz*ando] and [tre ~ te] could be ascribed to the graphemic indistinctiveness of retroflex (t) and (d) from dental (t) and (d). This phenomenon was more commonly observed in nonfluent participants’ speech, likely since their Sindhi knowledge was not advanced enough to correct perceived orthographic ambiguities from context. Inconsistent retroflex versus dental stop distinction in Roman text. Visual indistinct-
Table 6.6. Nonstandard pronunciations of stops in Roman text the initial retrotlex stop In [[Irengo} with the dental stop [Tt]. The case of [dz‘endo] being realised as [dz*endo], featuring similar replacement of retroflex stops with dental ones, is a curious one. Again, this Sindhi word is very similar to its Hindi counterpart, /dz*anda/, differing only in the final vowel. Hence, the substitution of retroflex stops with dental ones cannot be attributed to Hindi influence in this case. Similar is the case of [{re ~ te] being pronounced [te], since the phonetic shape of this Sindhi word is quite different from that of its Hindi equivalent /tin/. Hence, the confusion of retroflex and dental stops in [dz*ando] and [tre ~ te] could be ascribed to the graphemic indistinctiveness of retroflex (t) and (d) from dental (t) and (d). This phenomenon was more commonly observed in nonfluent participants’ speech, likely since their Sindhi knowledge was not advanced enough to correct perceived orthographic ambiguities from context. Inconsistent retroflex versus dental stop distinction in Roman text. Visual indistinct-
Table 6.7. Hindi interference in Devanagari Sindhi pronunciations In the first two examples in Table 6.7, it is possible that misinterpretation of the orthographic final lax vowel as a tense vowel encouraged Hindi interference. That is, the word was misunderstood as a Hindi word with a final tense vowel [i], which also had a similar pronunciation and meaning to the Sindhi word. This resulted in the mental retrieval of the Hindi word instead. The retrieval of a Marathi false friend was also attested from one participant, who pronounced (#fe) [kéh* ~ kéh] as [kahil] (cf. Marathi (#Tél) /kahi/ ‘any, some’). A similar phenomenon was seen in the Roman Sindhi text. Roman Sindhi words that were orthographically similar to English words or Roman Hindi
Table 6.7. Hindi interference in Devanagari Sindhi pronunciations In the first two examples in Table 6.7, it is possible that misinterpretation of the orthographic final lax vowel as a tense vowel encouraged Hindi interference. That is, the word was misunderstood as a Hindi word with a final tense vowel [i], which also had a similar pronunciation and meaning to the Sindhi word. This resulted in the mental retrieval of the Hindi word instead. The retrieval of a Marathi false friend was also attested from one participant, who pronounced (#fe) [kéh* ~ kéh] as [kahil] (cf. Marathi (#Tél) /kahi/ ‘any, some’). A similar phenomenon was seen in the Roman Sindhi text. Roman Sindhi words that were orthographically similar to English words or Roman Hindi
Table 6.8. Hindi and English influence on Roman Sindhi pronunciations Certain participants, especially those nonfluent in Sindhi but fluent in English, also read (agast®) and (janivari) as [ogast] and [dzzenja(a)ri] respectively. It could be surmised that the lexical recognition of the words as English loans likely made them retrieve the Indian English pronunciations from their mental lexicon, superseding any attempts to decode the words ohonologically. Indeed, the Hindi or English pronunciations of the words in question may well be the standard ‘Sindh? pronunciations of these words in Sindhi for new variety speakers. Hence, no new variety pronunciations have been listed in Table 6.8.
Table 6.8. Hindi and English influence on Roman Sindhi pronunciations Certain participants, especially those nonfluent in Sindhi but fluent in English, also read (agast®) and (janivari) as [ogast] and [dzzenja(a)ri] respectively. It could be surmised that the lexical recognition of the words as English loans likely made them retrieve the Indian English pronunciations from their mental lexicon, superseding any attempts to decode the words ohonologically. Indeed, the Hindi or English pronunciations of the words in question may well be the standard ‘Sindh? pronunciations of these words in Sindhi for new variety speakers. Hence, no new variety pronunciations have been listed in Table 6.8.
Table 6.9. Roman Sindhi spellings of English loanwords spellings were refiective Of thelr OWN pronunclatlons Of these Words. The Roman Sindhi text, too, contained several English loanwords, albeit commonly used in Sindhi. Two were names of months, (August) and (January), whose pronunciations have been discussed earlier. The others were the words ‘school, ‘college’ and ‘office’, but in their oblique or declined forms, that is, with the suffix [an‘] appended (§ 2.2.4, Table 2.6). The Roman Sindhi spellings of these forms and their assimilated Sindhi pronunciations are shown in Table 6.9. participants fluent in English typically read the root words with their Indian
Table 6.9. Roman Sindhi spellings of English loanwords spellings were refiective Of thelr OWN pronunclatlons Of these Words. The Roman Sindhi text, too, contained several English loanwords, albeit commonly used in Sindhi. Two were names of months, (August) and (January), whose pronunciations have been discussed earlier. The others were the words ‘school, ‘college’ and ‘office’, but in their oblique or declined forms, that is, with the suffix [an‘] appended (§ 2.2.4, Table 2.6). The Roman Sindhi spellings of these forms and their assimilated Sindhi pronunciations are shown in Table 6.9. participants fluent in English typically read the root words with their Indian
dle 6.10. Selection of Roman-orthography conventio: for distinguishing retroflex and dental stops As seen from Table 6.10, strategies for distinguishing retroflex stops from dental ones include adding diacritics, introducing new letterforms, and creating digraphs. Diacritics in Roman, as already mentioned, were not a
dle 6.10. Selection of Roman-orthography conventio: for distinguishing retroflex and dental stops As seen from Table 6.10, strategies for distinguishing retroflex stops from dental ones include adding diacritics, introducing new letterforms, and creating digraphs. Diacritics in Roman, as already mentioned, were not a
Table 6.11. Indexical associations of Sindhi scripts en eee ee ee en Ce NN Nt Ee —_—_——JF Ores Tre Oo 4 J a pe Jr On the other hand, the association of Perso-Arabic with “Muslim” by some participants categorically implied that the other two scripts were not Muslim. Similarly, the implication that Devanagari was the primordial Sindhi script automatically excluded the other scripts from this label. Thus, these values carry a specific polarity. Table 6.11 provides a consolidated overview of scriptal indexical associations, along with polarities where applicable. Question marks against a particular value mean that a clear indication of participant opinion was not obtained from the data. automatically excluded the other scripts from this label. Thus, these values particular context (Bender, 2008). For instance, if a script is used in an informa
Table 6.11. Indexical associations of Sindhi scripts en eee ee ee en Ce NN Nt Ee —_—_——JF Ores Tre Oo 4 J a pe Jr On the other hand, the association of Perso-Arabic with “Muslim” by some participants categorically implied that the other two scripts were not Muslim. Similarly, the implication that Devanagari was the primordial Sindhi script automatically excluded the other scripts from this label. Thus, these values carry a specific polarity. Table 6.11 provides a consolidated overview of scriptal indexical associations, along with polarities where applicable. Question marks against a particular value mean that a clear indication of participant opinion was not obtained from the data. automatically excluded the other scripts from this label. Thus, these values particular context (Bender, 2008). For instance, if a script is used in an informa
Table A-1. Comparison of various Sindhi scripts and orthographies lkennnn Aw Tweenies [14079 «wen POA POT),
Table A-1. Comparison of various Sindhi scripts and orthographies lkennnn Aw Tweenies [14079 «wen POA POT),
Table A-2. Simplified romanisation scheme modelled on Grierson (1919, pp. ix-x)
Table A-2. Simplified romanisation scheme modelled on Grierson (1919, pp. ix-x)
Supplementary figures Figure B-1. Selection of seals unearthed at Mohenjo-daro, Sindh, with undeciphered symbols
Supplementary figures Figure B-1. Selection of seals unearthed at Mohenjo-daro, Sindh, with undeciphered symbols
Related topics:
Phonetics and PhonologySociolinguistics of Writing
580 California St., Suite 400
San Francisco, CA, 94104
© 2025 Academia. All rights reserved