AAWM Journal 10/2 (2022)

Rāgs of Western India and Sindh

Peter Manuel and Brian Bond^[1]

Hindustani music, North Indian music, rāg, gurbānī sangı̄t, Sindhi music, Langa, Mānganhār

It is generally taken for granted that the rāg repertoire of Hindustani music is a pan-regional corpus, and that whatever regional origins and associations some rāgs may have once had are long since lost and inoperant. Accompanying this axiom is an assumption that rāgs are entities essentially unique to classical Hindustani and Karnatak musics. These notions, in fact, are only partially true, being marred by a significant exception, in the form of rāgs which are exclusive to or closely associated with Western India, by which we refer here to pre-Partition Punjab, Rajasthan, and Sindh. Some of these rāgs are part of the classical repertoire, but are only or predominantly performed by gharānās presently or historically based in the West. Others are part of the repertoire of Sikh gūrbānī sangı̄t, or are constituents of Rajasthani Langā and Mānganiyār music and Sindhi kāfī and shāh jo rāg̈. In this essay we survey these alternate rāg traditions and make a set of broader observations about their implications for North Indian music culture, including the nature and extent of regionalism therein, and the importance and, indeed, richness of these vernacular genres in the subcontinent’s music cultures.

Peter Manuel is Professor Emeritus at the Graduate Center of the City University of New York.

Brian Bond is a Visiting Lecturer at the University of California, Berkeley.

Click to read the PDF version

Introduction

[1] North Indian classical music, like its South Indian counterpart, has traditionally—and not entirely without reason—been regarded as a pan-regional entity. Whether cultivated in Mumbai, Kolkata, or Delhi, Hindustani music features the same sub-genres, styles, performance formats, instruments, associated social practices, and repertoire of rāgs and tāls (roughly, modes and meters). As such, it represents a quintessential example of what Indianist scholars have called a “Great Tradition,” meaning a supra-regional cultural entity sustained by elite, literate connoisseurs and buttressed by a canonic body of written works, whether, in this case, musicological or pedagogical. Probably the most fundamental entity in this music system is rāg, meaning a mode distinguished by ascending and descending scales, pitch hierarchy, and characteristic phrases, to be used as a basis for improvisation and composition.

[2] While this pan-regional conception of Hindustani music has an obvious logic and general veracity, it is in fact marred by a significant exception, namely Western India. By “Western India,” we refer here to a pre-Partitioned cultural region that includes Punjab, Rajasthan, Sindh, and Kachchh.[2] In Western India, there exist unique repertoires of rāgs, distinct from what should properly be categorized as “mainstream” Hindustani music. Some of these rāgs are part of the classical repertoire, but are only, or are predominantly, performed by gharānās based in the West. Others are part of the repertoire of Sikh gurbānī sangı̄t, which should be understood as a parallel, though overlapping quasi-classical tradition. Finally, others are constituents of vernacular music traditions—especially Rajasthani Laṅgā and Māṅgaṇiyār music and Sindhi kāfī and shāh jo rāg̈—which have some features of a regional classical music, though lacking such explicitly elaborated and articulated music theory. Some of the rāgs appear to derive from folk origins; others appear to have derived from exposure—in some form in the past—to classical music, but have been either modified or else preserved as marginal survivals. Strikingly, in no other part of North India can one find counterparts to these regional modal repertoires.[3] The existence of these repertoires also obliges us to recognize that rāgs are entities which are not unique to Hindustani (and Karnatak) music, but can also be found in parallel and intermediate music genres.

[3] In this essay our primary goal is not to describe all of these rāgs, although we do present much information about their forms. Rather, we seek to make a set of broader observations about the implications these repertoires have for North Indian music culture. Their existence, among other things, calls for a more nuanced and holistic understanding of the nature and extent of regionalism in Indian music. For some musicologists—especially of Sikh music—, a recognition of their unique repertoire has socio-religious import, serving as a corrective to an unquestioning hegemony of mainstream Hindustani music theory that has “colonized” variant traditions. The study of these rāgs, their associated genres, and their different sorts of relations to Hindustani music also affords an enhanced understanding of the historical development of North Indian music in general. In particular, it sheds light on some of the dynamics—both socio-musical and purely formal—that shape the evolution of rāgs. It also highlights the importance and, indeed, the richness of these vernacular genres in the subcontinent’s music cultures.

[4] In an earlier essay (Manuel 2015a), one of the present authors described these vernacular genres as constituents of an “intermediate sphere” of Indian music, comprising various idioms that lie between “folk” and classical realms in terms of both formal and socio-musical features. As articulated by Harold Powers (1980), art musics—cross-culturally and in South Asia—can be characterized by their reliance on the patronage of elite connoisseurs, their grounding in a pan-regional “Great Tradition” of canonic texts and a body (whether written or oral) of explicit theory, and the ability of music performances to stand on their own as the centerpiece of a performance (as opposed, for example, to songs embedded in theater or religious ritual). The intermediate-sphere idioms discussed all have a certain degree of theory, which can be articulated, if often inconsistently and incompletely, by performers; this theory might be unrelated to that of Hindustani music, as in the case of Hathrasi rasiyā, whose theory pertains primarily to prosody, though with melodic associations (Manuel 2015b). In the case of the Rajasthani and Sindhi musics discussed in this article, the theory of relevance comprises the repertoire of rāgs, which, however, are not as standardized or as analytically conceptualized as are their counterparts in classical music. Further, the “elite” patrons are primarily the local, provincial landowning class rather than Mughal courtiers or the urban bourgeoisie.

[5] The prominence of these intermediate-sphere genres in Western India, and the existence of a set of rāgs in North Indian classical music associated with this region, derive in part from the area’s distinctive geographical status. Rajasthan, the Punjab, Sindh, and Kachchh—however distinct in their histories and cultures—have always constituted particular sorts of subcontinental frontier or buffer regions, whether in relation to the Sanskrit cosmopolis, the Mughal Empire, or the modern North Indian cultural sphere (Sindh and western Punjab, of course, being ceded to Pakistan in 1947). In pre-modern and early modern times, the Gangetic Valley, the core region of the Hindustani music tradition, was sharply circumscribed in the north by the Himalayas, while to the east of Bengal it dissipated in thick tropical forests. But to the northwest and west lay regions that were habitable—Punjab, Rajasthan, and Sindh—, and indeed inhabited, hosting their own modest cities and courts since ancient times. Although somewhat remote from historic metropolitan cities such as Kashi and Pataliputra, they were not separated from those epicenters by impassable geographical obstacles. Further, although they were border regions of sorts, they were not directly adjoining distinctive foreign cultural zones, such as those of Persia or Central Asia. With the decline of the Mughal Empire in the early 1700s, courts in these regions became effectively independent, while continuing to patronize musical establishments. This particular combination of isolation and relative accessibility contributed to the cultivation of regional rāgs, both in local performance traditions of Hindustani music and in intermediate-sphere genres.

Regionalism and Pan-Regionalism in North Indian Classical Music Culture, Past and Present

[6] The existence of a distinctive modal repertoire in Western India—both in intermediate-sphere genres as well as Hindustani music—should be seen in the broader historical context of regionalism and pan-regionalism in North Indian music culture as a whole. The basic geographical contours of Indian art music culture are familiar: sources—primarily Sanskrit treatises—suggest that in the first millennium and a half of the common era, Indian art music flourished as a relatively unified pan-regional entity. However, our knowledge of this cultural geography is woefully incomplete; for example, we do not even know where some of the most important texts—especially the Nāṭyaśāstra (2^nd c. BCE–2^nd c. CE?) and Brhaddeśī (6th–8th c. CE)—were written. Nevertheless, it is generally assumed that for most of this period, the epicenter of Indian court culture—including the evolution of rāgs—was the Gangetic plain, especially but not only during the Gupta Period (3^rd–6^th c. CE). It is clear that during this period, as later, there existed a distinction between a “Great Tradition” of pan-regional elite culture, canonized in Sanskrit texts, and regional vernacular “Little Traditions.” The Brhaddeśī suggested such a dichotomy in contrasting mārg music—implicitly eternal, inherited, and perhaps even supernatural in origin—with regional deśī music, including the rāgs that it described. By the time of the Sangı̄taratnākara (13th c.), all the rāgs in practice were effectively deśī. In an informative early study, O.C. Gangoly (1935, 41–45, 72–74) noted that several rāg names from this era denoted specific regions or, in some cases, tribes or ethnic/linguistic groups. Richard Widdess, in his article “The Geography of Rāgas in Ancient India,” building on Gangoly’s observations, pointed out the “preponderance of names relating to the north-west and western regions,” referring to such rāgs as Mārū, Sauvīra, Saindhva, Gandhāra, and Ṭakka, essentially connoting Rajasthan, Sindh, and the Punjab.

[7] By the fifteenth century, if not earlier, protracted Muslim rule in North India had led to a broad divergence of two pan-regional art music traditions: the Northern and Southern, or Hindustani and Karnatak, which were distinguished by rāg repertoire, favored instruments, song text languages, style, patronage patterns, and other features. It could be said that the Northern modal repertoire was a syncretic product of indigenous pan-regional Indian traditions, imports from Persia, and thirdly, a particular regional tradition that happened to encompass the imperial capital and acquire pan-regional status. Here we are speaking of a set of interrelated dhrupad traditions, especially the court music of Gwalior, the quasi-classical song traditions of Hindu temples in the Mathura area, and Mughal court music as cultivated primarily in Delhi and Agra, enriched by the presence of musicians from Gwalior. In the ‘Ain-i-Akbari (1593), Abul Fazl contrasted the mārg songs—“chanted by the gods and great rishis”—with the dhrupad of Gwalior, which, though deśī in origin, had “passed into universal favor” once incorporated into Mughal court culture.

[8] Two primary factors coincided in consolidating this composite musical idiom as pan-regional. One was, of course, the status and power of the Mughal court, whose artistic culture provided the model for provincial courts. A second factor was the celebration, among Vaishnavite Hindus, of the Braj region—especially Mathura and Brindavan, near Agra—as the ancestral homeland of the god Krishna. Concomitant with this identification was the adoption of the Hindi dialect of Braj Bhasha as a pan-regional idiom for Krishnaite poetry (and song texts) and subsequently lyric verse (rītī) in general. A related development was the further spread of the rāg repertoire of Braj temple music, as devotees from Bengal and other regions traveled to Krishna’s homeland, imbibed the local devotional music and poetry, and then brought it back to their own regions. However, as we shall note, the idiom that would become a canonic Hindustani music norm spread in an uneven fashion, and took root in Rajasthan and the Punjab considerably earlier than in Maharashtra and Bengal, which would become centers in the early twentieth century.

[9] A complex dialectic between regional and pan-regional traditions took shape and has persisted, in modified form, from the Mughal period to the present. On the one hand, it is assumed that certain regional vernacular modes were incorporated into the pan-regional canon. Raja Man Singh, in his sixteenth-century Mānkutūhala, surmised as such, attributing rāg Mālasiri to folk origin, and Bairārī to Tirhut (in Bihar). Notably, however, it is primarily genres and styles, rather than rāgs, that Faqirallah, in his Rāg Darpan supplement to that text, attributes to specific regional origins (1996 [1665], 39, 47, 101–19). In general, it could be argued that most rāgs have derived from innovations in the existing urban art music repertoire rather than regional vernacular music. This observation would apply especially to all the rāgs in Pūrvī, Mārvā, and Toḍī thāṭs, which are not at all common in North Indian folk music.[4]

[10] Meanwhile, during the Mughal period, as Sanskrit declined, regional vernacular languages such as Bengali, Marwari, Punjabi, and Avadhi were, in different forms, emerging as legitimate literary idioms, cultivated in poetry and song-texts alongside the pan-regional Braj Bhasha (as well as Persian, and, from the late eighteenth century, Urdu). In accordance with such developments, along with difficulty of travel and correspondence in the early modern period, it might be natural to expect regional potentates and nobility to take pride in their local culture and cultivate local music, perhaps in stylized form, in such realms as modal repertoire of court music. Some such cultivation certainly took place, but primarily in the realm of light entertainment rather than classical music. Mughal chronicles such as the ‘Ain-i-Akbari document how court entertainment consisted not only of art music, but also of all manner of local folk dancers, acrobats, and rustic thespians, as was also the case elsewhere, for example, in the nineteenth-century Lucknow court of Wajid Ali Shah.

[11] In the realm of art music, however, what was sought was not some celebration of provincial traditions but rather imitation of the cosmopolitan pan-regional forms of the imperial power. Courts and regional gentry sought legitimation, status, and aesthetic pleasure in their competitive patronage of Great Tradition art music, hosting prominent musicians and occasionally commissioning music treatises. Through this promulgation they perpetuated a well-established tradition that Sheldon Pollock described in reference to the Sanskrit cosmopolis in the first millennium CE. He wrote:

In many of these cases, qualifying as empire, whether imperial governance was actually exercised or not, seems to have required a language of cosmopolitan character and transethnic attraction, transcending or arresting any ethno-identity the ruling elites themselves might possess. It had to be a language capable of making the translocal claims—however imaginary these were—that defined the political imagination of this world. (Pollock 1998, 13)

[12] Such an attitude can be said to have persisted until the present in reference to North Indian music. Thus, for instance, if Bengalis might take pride in their contribution to Hindustani music culture, it would not be on the basis of having contributed some rāgs to the canonic repertoire, but rather in terms of the vibrancy (albeit now attenuated) of Kolkata as a center for performance, and the impressive number of Bengalis among the top rank of performers. Likewise, if a Bengali vocalist such as Chinmoy Lahiri (1920–84) composed rāgs such as Shyāmkos and Yogamāya that are performed primarily by his Kolkata-based disciples, these rāgs are not celebrated as regional “Bengali rāgs” per se, but rather as the creations of an artist who happens to be Bengali. It is for these reasons that the rāg repertoire of Hindustani music has constituted a relatively standardized pan-regional canon—with the partial but not insignificant exception of a small set of rāgs from the subcontinental West.

[13] The only scholarly work to explore the issue of regional variation in North Indian music is the 2005 dissertation of Jeffrey Grimes, “The Geography of Hindustani Music: The Influence of Region and Regionalism on the North Indian Classical Tradition.” Grimes correctly notes that Hindustani music is “primarily a national tradition with consistency, continuity, and coherence that cuts across regional or state boundaries” (37). He relates that his many informants deny that there are stylistic differences between regional renditions of Hindustani music (e.g., 211, 327); as one told him, “If you can make out that someone is a Maharashtrian when they are singing, they are doing something wrong” (27). Grimes does not discuss rāg repertoire at all, nor does he say anything about the area encompassing Rajasthan, Sindh, and the Punjab. Instead, he compares and contrasts Maharashtra and Bengal—two regions whose Hindustani music scenes did not commence until the mid-nineteenth century, and then did so in fairly distinct manners. In Maharashtra, rulers of several princely states patronized Brahmin vocalists who had learned from Gwalior founders, such that the region—and especially Bombay and Pune—came to abound in fine Gwalior-style singers, while supporting hardly any renowned instrumental soloists (that is, of sitar and sarod). By contrast, Bengal—meaning largely Kolkatta—has generated few renowned vocalists; instead, it has been a center for instrumentalists, whether local Bengalis such as sitarist Nikhil Banerjee, or ethnic Uttar Pradesh Muslims such as the families of sitarists Inayat Khan and Vilayat Khan.^[5]

[14] The historical development of classical music in the northwestern regions of Rajasthan and the Punjab differs in certain ways from that in both Maharashtra and Bengal. Firstly, evidence suggests that rāg-based music—whether for Hindu or Sikh devotion or for secular courtly entertainment—was cultivated considerably earlier in those former regions, especially given their relative proximity to Delhi and Gwalior. In the 1650s, much of the Pushti Marg community of Vaishnavite Hindus, persecuted by Aurangzeb, fled from the Braj region and settled in Mewar, Rajasthan, where their temples and practices flourished under the patronage of Rajput princes. These practices included the tradition of rāg-based dhrupad-style singing today called havelī sangı̄t. The initial flowering of this music in the Braj region was contemporary with and presumably linked to the emergence of dhrupad in the court of Man Singh of Gwalior, which, as mentioned, was later transplanted to the Mughal court; as Ho (2013) has discussed, evidence suggests that aspects of the Pushti Marg music tradition may well have predated the courtly dhrupad rather than being derivative of it. While Pushti Marg temples, with their associated havelī sangı̄t, remain active in Mathura and nearby Brindavan, since the seventeenth century the main strongholds of the tradition have been Rajasthan (especially Nathdwara, near Udaipur) and Gujarat. As we shall suggest, while the specifically Rajasthani character of Pushti Marg music should not be overemphasized, it is significant that some of the tradition’s rāg repertoire—such as its versions of Mārū and Dhanāsrī in Kāfi thāṭ—coheres with that of other distinctively western genres.

[15] With the drastic decline of the Mughal court in the early 1700s, many musicians migrated to Rajasthani princedoms. By the mid-nineteenth century, with Delhi in further decline, many Rajasthani courts became local centers of art music, under the ardent patronage of their rajas (see, e.g., Deodhar 1993, ch. 7). Before Independence, Rajasthan hosted twenty-one semi-autonomous “princely states.” Of these, Komal Kothari wrote:

Each court had its own decorum, rituals, ceremonies, customs and practices. The performing arts played a very important role in the execution of all these aspects of the court. Music, of all these performing arts, had the most important role to play not only as an object of aesthetics and arts, but also as a symbolic paraphernalia of the state and the court. (Kothari 1995, 59)

[16] While Mewat, Jodhpur, Jaisalmer, Udaipur, Alwar, Jhajjar, Tonk, Bikaner, and other locales hosted courts, the most extensive was that of Jaipur, which was a particular center for court music, including rāg knowledge, in the seventeenth and eighteenth centuries (Bhatkhande 1951-57, III, 107). Maharaja Sawai Pratap Singh (r. 1778-1803) was especially renowned for his arts patronage, and his successors continued to employ many court musicians through the mid-twentieth century (Erdman 1978). A seminal role was played by dhrupad singer Bahram Khan, who, abandoning Delhi after the devastation inflicted by the British in 1857, resided and taught at the Rajasthani courts of Alwar and Jaipur until his death in 1880. Bahram Khan is seen as the progenitor of the lineage eventually known as the Dagar family (see Widdess and Sanyal 2004, 30, 107). Jaipur and Alwar also hosted a distinct tradition of bı̄n and sitar playing.

[17] Art music also took root early on in the urban Punjab, especially as the city of Lahore served as the Mughal capital—alternating with Delhi and Agra—for much of that dynasty’s sixteenth- to seventeenth-century heyday. Emperor Akbar, for example, based himself in Lahore for fourteen years. The Talwandi gharānā of dhrupad singers traces its origin to Mughal court singers who were awarded land grants in the Punjab in the sixteenth century (Widdess and Sanyal 2004, 30). After the decline of the Mughals, new Punjabi rulers supported court music in their manner, such as Ranjit Singh (r. 1799-39), who, though best known as a warrior-king, was also a patron of the arts and in the 1830s hosted the aforementioned Bahram Khan (see, e.g., Kapuria 2019). Another court musician, Irshad Ali Khan, founded the Kasur gharānā of Lahore-district dhrupad singers, and his descendants became propagators of the Patiala gharānā of khyāl and thumri singers. The Patiala court itself became a center of music patronage from the reign of Maharaja Narinder Singh (r. 1823–62) onward (van der Linden 2015, 149–50). Evidence suggests that there was a fair amount of interaction and travel between Rajasthani and Punjabi courts, with musicians such as the young Alladiya Khan (1855–1946) residing alternately in both regions.

[18] Another rāg-based music taking early root in the Punjab was the devotional music initially promoted by the first Sikh guru, Guru Nanak, from the early sixteenth century, and his successors and followers. As we shall discuss, this music, eventually known as gurbānī sangı̄t or gurbānī kirtan, constituted a parallel tradition to Mughal court dhrupad, while differing in its orientation toward devotion rather than secular entertainment. While the rāg repertoire of this music, along with other aspects, cohered in most respects with Mughal court music, it represented not a mere derivative tradition, and in some respects, like havelī sangı̄t it may have taken shape earlier than its courtly counterpart (Cassio 2015).

[19] Thus, the genre that would coalesce as Hindustani art music enjoyed a strong presence in Rajasthan and the Punjab from the mid-seventeenth century—some two centuries before it came to be extensively cultivated in Maharashtra and Bengal (the latter of which was characterized by Bernard Cohn, in a 1967 article on regionalism, as a “cul-de-sac”). When, in the 1800s, Hindustani music finally came to Maharashtra, it did so in a fairly uniform manner, with a uniform rāg repertoire—especially that of the Gwalior gharānā (with the exceptions being Rajasthani Jaipur-gharānā rāgs taken up by Maharashtrian disciples of that tradition). Moreover, both Rajasthan and the Punjab also hosted rāg-based “intermediate” genres—especially Mānḍ and gurbānī sangı̄t—that could serve as conduits between classical music and local vernacular genres. (Both regions declined drastically as centers for fine arts from the mid-twentieth century, as concert activity became concentrated in Calcutta, Delhi, and Bombay.) Such factors allowed for the perpetuation of various distinctive and unique rāgs, whether these represented archaisms, intermediate-genre idiosyncrasies, or local creations which for whatever reasons did not travel to other regions.

Figure 1. Map of India, indicating site of Figure 2.

Figure 2. Map of Western India and Sindh, showing relevant places.

Hindustani Rāgs of Western India

[20] Though the names of rāgs such as Mārū, Sauvīra, Saindhva, and Gandhāra invoke Western India, their inclusion in the rāg enumerations of major Sanskrit treatises suggests that they had long been incorporated into the mainstream courtly modal repertoire, in such a way that whatever regional origins they may have once had were no longer relevant or important. (Such, for example, is also the case with modern rāgs such as Pahāṛī, Sindh Bhairavī, and Brindāvanī Sārang.) However, the extent to which these rāgs retained regional associations is ultimately unverifiable, since the authors of such treatises preferred to present the music system they described as a universal canon rather than one sullied by geographical variations. In the late modern period, which we may roughly date from the nineteenth century, the Hindustani rāgs associated with Western India include a few from the Sanskritic era—especially Soraṭh and Mārū—but are otherwise a different group, comprising especially Āsā, Champak, Husenī Toḍī, Mānḍ, and perhaps Khokar. The cultivation of these rāgs can be further associated with a specific set of regional gharānās emerging in the nineteenth and early twentieth centuries from the socio-musical foundations we have just discussed.

[21] Some of these gharānās were based in the Punjab, which, as mentioned, hosted several regional courts, such as that of Patiala, fostering the emergence of the gharānā by that name in the mid-nineteenth century. Another hoary Punjabi lineage is the Talwandi gharānā, whose members, based in Ludhiana district, sang dhrupad and other genres for Sikh patrons and the Hindu court of Jammu (Widdess and Sanyal 2004, 31). A third Punjabi line was the Sham Chaurasi gharānā, whose most (indeed, only) renowned exponents were the brothers Nazakat and Salamat Ali Khan (1928–84, 1934–2001). Musicians in these lineages performed for local rajas and landlords, and for Sikh religious functions, whose related repertoire will be considered below.

[22] In Rajasthan the most important gharānā was the Jaipur khyāl gharānā, founded by the aforementioned Alladiya Khan, who was born, raised, and trained in Jaipur state.^[6] Alladiya Khan performed—and taught to his sons and Maharashtrian disciples—a prodigious number of rāgs, many of which are not performed by other gharānās. Some of these rāgs he seems to have invented, while others were obscure traditional rāgs that he resurrected from oblivion. Often it is unclear which is the case.^[7] However, some of the rāgs in question are also extant in other Rajasthani, Punjabi, or Sindhi traditions, including intermediate genres, and hence are clearly regional entities rather than his own idiosyncratic creations. Meanwhile, aside from his gharānā, another Rajasthani lineage is the Mewati khyāl gharānā, which, though previously obscure, came into prominence with the career of Pandit Jasraj (1930–2020).

[23] The art music patronized in these courts and performed by these gharānās was overwhelmingly that of the pan-regional Hindustani canon. Many of the seminal performers were refugees from Delhi, whose cultural life suffered repeated blows, from Nadir Shah’s invasion of 1739 to the massacres and pillage inflicted by the British in 1857. However, the early establishment of rāg-based musics in Rajasthan and the Punjab, coupled with the distance from Delhi, seems to have occasioned the cultivation of a set of distinctive rāgs. Whether these emerged from local vernacular musics, constituted marginal survivals, or were the creations of urban court musicians, they did not enter the Delhi and Gwalior canon that was exported to Maharashtra and Bengal from the mid-nineteenth century. These rāgs, though few, are not insignificant, and they also vary in their status and their historical relations to Hindustani music. Some of the rāgs of presumed western (especially Sindhi) origin—notably Sindhūra, Sindh Kāfi, and Sindh Bhairavi—have long since been absorbed into the pan-regional Hindustani canon, and in that sense are not uniquely “western rāgs” in terms of their performance practice and the criteria used in this essay.[8] Our focus here lies instead on rāgs that are performed primarily or even exclusively by musicians connected in training and/or residence with the west.

[24] Perhaps the single most prominent and venerable Western Indian rāg—both in Hindustani music and in the intermediate genres discussed below—is Āsā (or, less often, Āshā), a “major-scale” (Bilāval thāṭ) rāg that omits the third and seventh degrees in its ascending scale (hence, Sa Re Ma Pa Dha Sa), while including them, and occasionally flat Ni, in the heptatonic descent. Āsā is of considerable antiquity, being a prominent rāg in the Sikh Ādi Granth compiled in 1604, as discussed below; a verse in that volume attributed to Sufi poet Shaikh Farid (1173–1265) is designated to be sung in Āsā, supposedly indicating that Farid himself sang it in that rāg, though no associated melody survives. In modern times, Āsā (or Āsā Mānḍ) was sung by leading vocalists of the Patiala, Sham Chaurasi, and Jaipur gharānās, and by Pandit Jasraj of the Mewati gharānā.^[9] Vishnu Digambar Paluskar (1872–1931)—with V.N. Bhakhande, one of the two great modernizers of Hindustani music—also included two bhajans in Āsā in his pedagogical book Sangı̄t Bālprakāsh (1921, 12–14). Paluskar himself was trained in the Gwalior gharānā, whose exponents, like those of essentially all gharānās outside the Western region, generally do not sing Āsā; however, Paluskar lived in the Punjab for several years and founded his first music academy, the Gandharva Sangı̄t Mahavidyalaya, in Lahore in 1901. It is clear that Paluskar—with his double musical identity incorporating both Gwalior and the Punjab—learned Āsā in the Punjab and deemed it logical to include it in his songbook. Āsā is also included in a 1916 text, Sangı̄t Sudarshan, by Jaipur-based sitarist Amritsen, and it is described in the Rāg Kalpadruma (1842, pp. 16, 25), a massive anthology of song texts in Hindi, Rajasthani dialects, and other languages. Conversely, the rāg is conspicuously absent in treatises and pedagogical works representing the pan-regional mainstream, such as the publications of V.N. Bhatkhande (1860–1936), as well as Sanskrit texts preceding the modern era.^[10]

[25] Closely resembling Āsā is rāg Mānḍ, another Bilāval-thāṭ mode. Somewhat confusingly, “Mānḍ” denotes both this rāg and a genre, specifically, a Rajasthani court and salon counterpart to thumri and dādra, flourishing in the nineteenth and early twentieth centuries. This “rajvāṛi” (“courtly”) Mānḍ did not differ stylistically from thumri and dādra, but featured Marwari texts, an introductory dohā couplet, and occasional use of distinctive tāls. It was most characteristically sung by pātar courtesans, in a quintessentially bāijī style (see Manuel 1989, 105-144). As in eastern Uttar Pradesh, the courtesan milieu of specialists and connoisseurs of light-classical music declined in the mid-twentieth century, and the best exemplars of Mānḍ are recordings of deceased courtesans such as Allah Jilai Bai (Allajilai Bai, 1902-1992). As a genre, Mānḍ, like thumri and dādra, could be sung in various light rāgs, such as Khamāj, Pı̄lū, Bhairavī, and also the eponymous rāg Mānḍ, which is of relevance here. Rāg Mānḍ is somewhat distinctive among the modes considered here in the sense that it is fairly well-known (though not necessarily widely performed) outside Rajasthan, while at the same time being universally recognized as originating in that state (or perhaps, more specifically, from Malwa, in the west). Mānḍ was a popular rāg used in Maharashtrian theater songs (Subba Rao 1993, III, 124); it is also routinely included in rāg encyclopedias and song anthologies such as Bhatkhande’s six-volume Kramik Pustak Mālika anthology (henceforth, KPM, 1954-59, V, 239) and Onkarnath Thakur’s Sangı̄tājjali (1957, 179).^[11] However, Mānḍ is most widely performed (both as a rāg and a genre) in Rajasthan and Punjab, or among singers trained in these traditions.

[26] As a “small,” light rāg, Mānḍ’s particular form is variable and flexible, and some non-Rajasthani performers may base their interpretations on little more than a familiarity with the iconic Marwari song “Kesariya Bālam,” the first lines of which are shown in Example 1 (see Bhatt 2014, 72). Meanwhile, as will be discussed below, varieties of Mānḍ are sung in various “intermediate” genres, especially the music of Laṅgās and Māṅgaṇiyārs of western Rajasthan.^[12]

Example 1. “Kesariya Bālam.”

[27] In a somewhat different category is rāg Soraṭh. Like Āsā, Soraṭh (or Soraṭhī) is a venerable rāg, but unlike Āsā, it is documented in several Sanskrit treatises, both Southern and Northern, including Lochana Kavi’s Rāga-Tarangini (15th c.?), Ksemakarna’s Rāgamālā (16th c.), and the Rādhagovind Sangı̄tsār written in Jaipur around 1803. It is also typically included in twentieth-century rāg surveys, such as that of Subba Rao (1993), and is discussed in V.N. Bhatkhande’s monumental history of rāgs—an important resource in this article—, the Hindustāni Sangı̄t Shāstra (1954–57, I, 236–39). Further, it is a rāg in the dhrupad-related Pushti Marg havelī sangı̄t repertoire. While these factors would seem to suggest inclusion in the pan-regional canon, other considerations link the rāg to Western India. To begin with, Soraṭh is the colloquial and archaic name for the Saurashtra/Kathiawar peninsula, on the southernmost part of Gujarat, itself to the south of Rajasthan and southeast of Sindh. Bhatkhande points out that Soraṭh song texts in treatises are often in Gujarati or Marwari, the dialect of western Rajasthan and the primary traditional poetic idiom of Rajasthan in general (1954–1957, I, 238). Several of the Soraṭh song-texts included in the Rāg Kalpadruma are in Marwari, as are some of those presented in Bhatkhande’s KPM (1954–1959, V, 316–19); one of the latter is addressed to Mārū, a folk-ballad hero of Rajasthan and Sindh. Further, Soraṭh is the name of a heroine in a romantic epic ballad popular in Rajasthan, and in Sindh, in its retelling by Shah Abdul Latif, to be sung in the eponymous sur or mode (as discussed below). Most significant is that however hoary the rāg may be, it is rare in the modern Hindustani repertoire, being occasionally performed only by Patiala and Jaipur gharānā exponents. Meanwhile, as we shall note, it is a basic rāg in the repertoires of Sikh gurbānī sangı̄t, Laṅgā and Māṅgaṇiyār music, and Sindhi kāfī and shāh jo rāg̈. The distinctive current prominence of Soraṭh in western regions strongly suggests that in spite of its customary inclusion in treatises, both Sanskritic and modern, it may have always retained the regional character suggested by its name. As we shall hypothesize, its lack of prominence in the mainstream canon may have to do with its close resemblance, depending on its form, to either Brindāvani Sārang or else, if Ga is included in descent, Desh.

[28] Somewhat similar in this sense are rāgs Bihāgṛā and Barārī (Bairārī, Varātī), which, though included in most early modern and modern rāg surveys, are sung almost exclusively by Jaipur gharānā vocalists, and by singers of Sikh gurbānī sangı̄t, as discussed below. Ahobal’s seventeenth-century Sangı̄t Parijata lists nine forms of Varātī, and at least three versions are present in the Jaipur gharānā tradition^[13]; of these, one is in Pūrvī thāṭ (i.e., with komal Dha), one is in Mārva thāṭ (with raised Dha), and the third uses both raised and lowered forms of Dha.

[29] A more unambiguously western rāg is Champak. Champak appears in at least one early modern treatise—the Rāg Darpan (Sarmadee 1996, 28–29)—and, significantly, in the Rādhagovind Sangı̄tsār from Jaipur. In modern times, Champak has been a fairly well established rāg among Punjabi and Rajasthani classical singers, as well as disciples of Paluskar.^[14] In most versions it resembles rāg Jhinjhoṭī, but certain characteristic phrases—especially the leap Re-Pa—lend it a distinct and quite attractive contour (chalan), rather than being a redundant cognate of another more familiar rāg.[15]

[30] Another distinctively western rāg is Husenī Toḍī (Husainī Toḍī), which may be of Middle Eastern origin. Amir Khusrau (1253–1325) is said to have introduced it from Persian music, and Husenī remains a familiar maqām in Iraqi and Turkish systems, with a neutral second degree that would not be admissible in Indian art music. A similar rāg Husenī—though with a natural second degree—became well established in South Indian music, being outlined in various treatises (including Somnath’s 17th-c. Rāg Vibodh and Tulaji’s 18th-c. Sangı̄t Sāramrta), and remaining familiar in the modern repertoire. In North India it seems to have survived—especially as “Husenī Toḍī”—only in Rajasthani traditions. Husenī is presented in the seventeenth-century treatises of Bhavbhatt (Bhava Bhatta), commissioned by Maharaja Anup Singh of Bikaner, and Husenī Toḍī is mentioned in the Nādodadhi, written in nineteenth-century Jaipur by Puran (Purna) Kavi (in Bhatkhande 1951-1957, II, 552). The rāg was performed by early-twentieth-century Jaipur singers Ashiq Ali Khan and his father Muhammad Ali Khan, who taught it to Vilayat Hussein Khan of the Agra gharānā, and to Bhatkhande (Yodh 1978, 14, Ratanjankar 1967, 14-15). In modern times, Husenī Toḍī is sung primarily by exponents of the Jaipur gharānā.^[16] While sharing some phrases with Jaunpūri (“Jaunpūrī Toḍī”), it more closely resembles rāg Desī, indeed, to such an extent that its presence in the mainstream Hindustani repertoire might be redundant. As we discuss below, Husenī Toḍī, as a thoroughly Rajasthani entity, is also sung by some Māṅgaṇiyārs and Laṅgās, some of whom, interviewed in the 1970s by Nazir Jairazbhoy, claimed familiarity with Ashiq Ali Khan.

[31] Meriting passing mention here is Khokar. Like Husenī and Champak, this rāg is cited in at least one medieval treatise, the Rāg Darpan, and is also described in the aforementioned Rādhagovind Sangı̄tsar, and though mentioned in such rāg surveys as those of Subba Rao (1993) and Kaufmann (1968, 585)—who describes it as “rare”—, it is standard only in the repertoire of Jaipur gharānā singers (several of whose renditions are available on YouTube). Like Husenī Toḍī and Champak, it closely resembles another more familiar rāg—in this case, Bihāgṛā—, indeed, to the extent that the two rāgs and their names may be essentially interchangeable.^[17] This redundance may explain its absence in the mainstream canon, and its presence only among Jaipur singers who have perpetuated Alladiya Khan’s cultivation of a wide rāg repertoire encompassing many otherwise obscure modes. Note, however, that we are characterizing as “western” only rāgs that are performed in more than one gharānā or genre; thus, if Hinḍol-Bahār may be characterized as a mere “Jaipur-gharānā rāg,” Huseni Toḍī has a broader western character manifest in its presence in Rajasthani treatises and Māṅgaṇiyār/Laṅgā repertoires as well as Jaipur-gharānā music.

[32] Passing mention should also be made of a set of now extinct rāgs that seem to have been current in nineteenth-century Rajasthan, especially in the realm of light-classical bhajan singing. The Brahmanand Bhajan Māla, an anthology of song lyrics published in Ajmer around 1900, prefaces many of its verses with headings specifying the rāg in which they are to be sung; these headings include such local archaisms as Banjāra, Kasūrī, Rāsra, Punjabi Kāfī, and Mangal, along with the more familiar Rajasthani rāgs Mānḍ and Janglā.

[33] A number of observations could be made or reiterated about the existence of this distinct set of Western Indian rāgs. First, as we have noted, no other region can claim such a presence in Hindustani music; that is, for example, there are no “Bengali rāgs” or “Maharashtrian rāgs” per se. In this regard, it is also significant that the only language other than Braj Bhasha that has been used in khyāl and dhrupad is Marwari (which, in such contexts, is roughly intelligible to Hindi speakers). Thus, for example, Bhatkhande’s massive KPM contains several khyāls and dhrupads in Marwari, but none in any other languages; in Hindustani music per se, with a very few idiosyncratic exceptions, there is no tradition of composing or singing classical khyāls or dhrupads in Bengali, Marathi, Gujarati, or Punjabi (aside from devotional gurbānī sangı̄t), but the Marwari tradition is well enough entrenched as to persist, especially in the context of its special rāg repertoire.^[18]

[34] Secondly, as we have suggested, the “Western” rāgs discussed are not notably distinctive in form, and indeed, they all resemble other more common Hindustani rāgs. It is as if the “space” that the Hindustani system allowed, or even desired, for certain modal entities came to be occupied by canonic rāgs in the mainstream, but alternately or additionally by a distinctive set of rāgs in the west. Their existence also suggests that such a regional dimension may have existed for centuries. Thus, while the Sanskrit treatises opted to present art music as a canon unsullied by regional irregularities, it is quite possible, for instance, that Soraṭh was always more popular in the west than elsewhere.

Gurbānī Sangı̄t

[35] As we have noted, Mughal court dhrupad evolved as a transplanted elaboration of the dhrupad cultivated in the Gwalior court of Man Singh Tomar (r. 1486-1516). This Gwalior tradition itself evolved in some sort of interaction with the similar music cultivated in the wealthy Vaishnavite temples in Mathura and nearby pilgrimage sites (Thielemann 2001). This latter music—especially the Pushti Marg tradition of kı̄rtan-singing or havelī sangı̄t—developed as a form of dhrupad, featuring four-part compositions in classical rāgs, set to “large” tāls, with predominantly Krishnaite lyrics in Braj Bhasha, anthologized in editions organized according to rāg. It differed from court dhrupad primarily in emphasizing devotional bhakti, as conveyed in the lyrics, rather than abstract elaboration of rāg and tāl. In the Punjab, a similar and closely related form of temple dhrupad singing was adopted by the founder of Sikhism, Guru Nanak. In collaboration with classical vocalists, Nanak set almost a thousand devotional hymns (shabads, shabds) to nineteen rāgs, creating a repertoire that came to be known as gurbānī kı̄rtan (or gurbānī sangı̄t). His equally prolific successors—especially the fifth Guru, Arjan (1563-1606)—amplified this repertoire, adding their own compositions and settings of lyrics, in more than twenty languages, attributed to other poet-saints, including Kabir, Mira Bai, and Shaikh Farid. Guru Arjan also established norms for daily singing of these compositions—anthologized in his Ādi Granth compilation—at the Sikh shrine. A subsequent and definitive version of the repertoire was formalized in the early 1700s in the Srī Gurū Granth Sāhib (henceforth, GGS) by the tenth and last Guru, Guru Gobind Singh (1666-1708). This anthology suggested an appropriate rāg and tāl for each hymn, using a repertoire of thirty-one basic rāgs and another thirty-one derivative compound rāgs. However, the Gurus also countenanced the setting of hymns to contemporary popular folk tunes. As we shall discuss, most of the rāgs are canonic in Hindustani music, though a significant minority are unique to gurbānī sangı̄t, or, while not entirely unknown in Hindustani sangı̄t, are rare in that idiom.

[36] Meanwhile, over the centuries, many temple singers—whether Sikh rāgīs or Muslim dhāṛı̄s—perpetuated their own orally transmitted “paramparik” (“family tradition”) repertoire, which included both the canonic GGS settings as well as their own renditions, and which employed around forty rāgs (including Mānḍ and Champak) not included in the GGS guidelines. In 1979, two eleventh-generation custodians of this repertoire, the brothers Avtar Singh (1926-2006) and Gurcharan Singh (1915-2017), sons of Jwala Singh, recorded around five hundred of their songs and notated them in a published anthology, the Gurbānī Sangı̄t: Prachīn Rīt Ratnāvalī. Several of their recordings are accessible on YouTube, along with others made by their disciple and grand-nephew, Bhai Baldeep Singh, an energetic vocalist, scholar, and pedagogue. Other song anthologies include those of Principal Dayal Singh and Master Tara Singh (Inderjit Kaur, p.c.).

[37] Gurbānī kı̄rtan, like havelī sangı̄t, is a text-driven (shabd-pradhān) genre, with primary emphasis on the sentiment of the devotional lyrics rather than abstract or virtuosic elaboration of rāg and tāl. In this respect, and in that it is traditionally performed in religious contexts rather than as stand-alone presentations, it has the character of an “intermediate sphere” genre, though its grounding in rāg and tāl is a distinctively classical feature (Manuel 2015a).

[38] As with havelī sangı̄t, the precise historical relationship between gurbānī kı̄rtan (including its rāg repertoire) and the Mughal court idiom which became mainstream Hindustani music has been in many respects unclear, and indeed has been a subject of lively disputation in some circles. On one side are those—especially Bob van der Linden (2015)—who have argued that gurbānī kı̄rtan, in its formal aspects, developed essentially as a regional efflorescence of Mughal court music and is best regarded as Hindustani music rather than as any sort of distinct tradition. Linden asserts that despite the frequent conflicts between the Sikhs and the Mughals, “the Sikh gurus and later Sikh maharajas increasingly took the Mughal court (darbār) as a point of reference for their own culture” (2015, 143). However, Bhai Baldeep Singh takes umbrage at this viewpoint, arguing instead that gurbānī kı̄rtan evolved in a parallel rather than derivative relationship to Mughal art music. Singh points to the unique rāgs and tāls in the Sikh repertoire, and to the refusal of many gurbānī musicians to accept Mughal patronage (2011, 270, 272-73). Similarly, Francesca Cassio argues that aspects of the gurbānī dhrupad tradition may have actually emerged before both its Mughal counterpart as well as havelī sangı̄t; she points out, for example, that Guru Nanak himself was born thirty-seven years before Tansen, the celebrated vocalist of Akbar’s court (2015, 10-11).

[39] For purposes of the present study, we need not dwell extensively on this debate, much less attempt to resolve it. Obviously court music of the sixteenth to nineteenth centuries was a shared idiom in the Punjabi as well as Mughal courts, and some of the rāg repertoire may have originated in the Punjab. However, with the liberation of Punjabi courts from Mughal dominance, the Punjabi repertoire may have developed or preserved distinctive elements, especially as cultivated in Sikh contexts, which were distinct from the secular entertainment ones. What is of particular interest here is that for someone such as Baldeep Singh, the unique rāg repertoire of gurbānī kı̄rtan is a matter of pride, and a cherished tradition that should be protected against the “colonizing” hegemony of the mainstream canon (p.c.).

[40] In modern times, the GGS hymns have come to be sung in a wide variety of styles—collectively called gurmat sangı̄t—ranging from settings using film tunes to khyāl-style renderings of the specified rāgs, with much showy display of virtuoso technique. Since the years around 1900, the Singh Sabha, an influential Sikh reformist organization, has attempted to standardize and formalize many aspects of Sikh devotional practice, including the music performed in gurdwāra shrines. In 1991 a group of concerned scholars and musicians met at Jawaddi Taksal, in Ludhiana (Punjab), forming the “Rāg Nirnayak Committee” (RNC), which sought to re-establish tradition and impose order on gurmat sangı̄t by insisting that the shabads be taught and performed in the rāgs specified in the GGS. They also promoted standardized versions of these rāgs which, however, either cohered with modern Bhatkhande-style Hindustani music or, in a few cases, were modern inventions, in contrast to the paramparik traditions. (These “reformed” rāg versions are presented at jawadditaksal.org.) The result has been a sort of “hermeneutic chaos” that, as has been noted, has exhibited some of the contradictions and distortions as are found in the European early music movement (see Kaur 2014; van der Linden 2013, ch. 5, Cassio 2015; Kalra 2014, ch. 4). Bhai Baldeep Singh has been particularly critical of the “colonizing” reform movement, which has devalued and dismissed the distinctive rāg versions perpetuated and laboriously documented by the Jwala Singh lineage. In yet another debate, van der Linden (2008, 12; 2012; 2013, 148) has argued that this orally transmitted repertoire cannot be assumed to faithfully represent the music of the sixteenth- to seventeenth-century gurus, to which Baldeep Singh replied that the consistency of song versions of distinct paramparik performers illustrates the fidelity of their transmission (Singh 2011, 268).

[41] The thirty-one basic rāgs in the GGS are: Srı̄ rāga, Mājh (Mānjh), Gaurı̄, Āsā, Gujrı̄, Devgandhārī, Bihāgṛā, Baṛhaṅs, Soraṭh, Dhanāshrī, Jaiṭsrī, Toḍı̄, Bairārī, Tilang, Sūhī, Bilāval, Gonḍ, Rāmkalī, Naṭ Narāyan, Mālī Gaura, Mārū, Tukhārī, Kedār, Bhairo, Basant, Sārang, Malhār, Kānṛa, Kalyān, Prabhāti, and Jaijaivantī. Most of these are mainstream Hindustani rāgs, but around a dozen are distinct either to Sikh tradition or to Western India in general, and are thus of interest in the present study. (Most of the other thirty-one rāgs are relatively obscure chhāya-lag compounds of the basic set, such as the eleven Gaurī mixed rāgs, and will not be discussed here.^[19]) The GGS concludes with an extended lyric, the “rāgmāla,” which enumerates these and other rāgs, including “Suhav” and the aforementioned Champak.

[42] The gurbānī rāgs are diverse in origin as well as form. Previously discussed are Āsā and Champak, which have the status of uniquely Punjabi and Rajasthani rāgs, also performed by Hindustani vocalists of that region. In a related category, as mentioned earlier, are Bihāgṛa, Barārī (Bairārī, Varātī), and Soraṭh, which, though recognized as legitimate Hindustani rāgs, are performed almost exclusively by musicians of western origin or training, especially those of Patiala or Jaipur gharānās, aside from being basic gurbānī kı̄rtan rāgs; significantly, however, they are also sung in Pushti Marg havelī sangı̄t, suggesting a historical connection between these two devotional dhrupad idioms. Also sung in both devotional genres are Kānrā (Kānṛā) and Mārū, two rāgs cited in various Sanskrit treatises (such as the Sangı̄t Parijāta), though being effectively extinct in modern Hindustani music. Kānrā, in its most common form, features the vakra “crooked” Ga♭-Ma-Re-Sa phrase iconic in modern Kānrā variants (such as Dārbārī), and most closely resembles the modern (though obscure) rāg Sūghrāi in its passing use of shuddh Dha.

[43] For its part, the gurbānī rāg Mārū, in its distinct versions, illustrates the “hermeneutic chaos” mentioned earlier. The most prevalent form (several renditions of which can be heard on YouTube) is particularly distinctive, using both raised and lowered forms of Ma, Dha, and Ni, along with shuddh Re. One might be inclined to posit intriguing parallels with the somewhat similar Pushti Marg version of Mārū, which, however, has komal Re, in accordance with the Bhairav-thāṭ (Gauri melā or scale) version presented in the seventeenth century by Pundarika Vitthala. However, this gurbānī version (according to Bhai Baldeep Singh, p.c.) appears to be a thoroughly modern concoction. By contrast, the paramparik Mārū versions are in either Kalyān or Khamāj ang (style, form), with no particular resemblance to either the haveli sangı̄t or Pundarika forms.

[44] A more probable anachronism is represented by a version of Srī rāg in Kāfi thāṭ, which is performed in the paramparik tradition alongside the Pūrvī-thāṭ Sri rāg standard in both the GGS canon and modern Hindustani music. The historical relationship between these two entirely different rāgs of the same name is enigmatic (as discussed by Bhatkhande 1951-1957, III, 41-64). Sanskrit texts describe only the Kāfī-thāṭ version, which is maintained in modern Karnatak music (with the corresponding melābeing renamed Kharaharapriya). At some point—presumably the nineteenth century—the Pūrvī-thāṭ Srī rāg coalesced in the North,^[20] with the Kāfi-thāṭ version surviving only in the Sikh tradition, together with several GGS compound rāgs whose names—such as Bilāval Dakhnī (“southern Bilāval”)—also suggest some sort of Karnatak connection.

[45] In a special category are rāgs Tukhārī and Sūhī, which exist only in gurbānī music. Tukhārī, like Mārū, is another rāg whose form has been confounded by modern revisionism. It is often rendered in a manner nearly identical to Hindustani Madhūvantī, but the paramparik version introduces komal Ni and shuddh Ma in quite a distinctive manner, as shown in Example 2.^[21] Though Tukhārī is certainly obscure in comparison to Madhuvanti, it actually features much more of the tetrachordal symmetry Nazir Jairazbhoy (1971) astutely showed to characterize most Hindustani rāgs. While Madhuvanti’s lower and upper tetrachords are completely unbalanced, those of Tukhārī achieve clear symmetry with the phrases Sa-Ma and Ga♭-Re-Sa being intervalically duplicated in the upper tetrachord by Pa-Sa and Ni♭-Dha-Pa.

Example 2. Rāg Tukhārī, sthāi and antarā of gurbānī shabad “Gol Ghumāi,” in chārtāl (12 beats).

[46] For its part, Sūhī (which bears no relation to Hindustani Sūhā)^[22] features the distinctive phrase Ni♭-Dha-Ni♮-Sa (as in rāg Miāṅ ki Malhār, though Sūhī features shuddh rather than komal Ga). The GGS contains a poem attributed to Shaikh Farid, to be sung in Sūhī, whose specific melody survives in the paramparik tradition and is thus believed to be that performed by Farid himself (Bhai Baldeep Singh, p.c.), though this alleged perpetuation of a thirteenth-century tune certainly cannot be verified. (GGS verses composed by the Sikh gurus were set by them and their associates to original tunes, but, according to oral tradition, verses incorporated from other saints and poets, such as Farid, retained their extant melodies and rāg settings.)

[47] The gurbāni rāg Naṭ Narāyan might constitute another distinctively “western” entity in that at present, as a Hindustani rāg it seems to survive almost exclusively in the repertoires of Jaipur, Mewati, and Patiala gharāna musicians, despite appearing in several Sanskrit texts. However, the gurbāni version of this, somewhat resembling Shām Kalyān, differs from that Hindustani namesake (sometimes called Naṭ Narāyani).

[48] In general, the distinctive gurbānī rāgs, like the Hindustani rāgs discussed in the previous section, are of diverse origins. Some, such as Champak and Āsā, are also sung by classical vocalists of Western gharānās. Others, especially Bihāgṛā, Barārī, and Soraṭh, are relatively canonic Hindustani rāgs which, however, are rarely performed by non-western musicians. Still others—especially Kāfī-thāṭ Srī rāg and Kānṛā—appear to be atavisms, the latter of which could be “western” in the sense that it is also perpetuated in the Pushti Marg tradition, with its epicenter in Nathdwara, Rajasthan.

[49] Most renditions of gurbānī and gurmat sangı̄t, as text-driven shabd-pradhān genres, consist primarily of reiterations of the composition rather than extended elaboration of the rāg in question. Hence some people might be inclined to dismiss the distinctive gurbānī rāgs as simple modal categorizations of songs rather than the fully fleshed-out rāgs of classical music. However, an activist such as Bhai Baldeep Singh would counter that some renditions, whether in dhrupad or khyāl style, do in fact develop the rāg in depth and illustrate how the gurbānī rāgs are just as legitimate as are Hindustani rāgs.

Laṅgā and Māṅgaṇiyār Music of Western Rajasthan

[50] The rāgs discussed thus far have included some associated with traditions originating or effectively based in Rajasthan, including the Jaipur and Mewati gharānās, courtly Mānḍ singing, and Pushti Marg havelī sangı̄t. Meanwhile, western Rajasthan is host to a particularly remarkable music genre—with a particularly distinctive rāg repertoire—, namely, that associated with the Laṅgā (Laṅghā) and Māṅgaṇiyār (Māṅgaṇhār) musician castes. Hereditary Muslim performers of the arid Jaisalmer, Barmer, and western Jodhpur districts, Laṅgās and Māṅgaṇiyārs have traditionally provided ritual, ceremonial, and entertainment music for their more affluent patrons in towns and villages. Since the 1970s, while many such patrons have ceased employing traditional performers, quite a few Laṅgās and Māṅgaṇiyārs have found work performing for tourists and at urban festivals (e.g., of “Sufi Music” and even “Gypsy music”), and some have even toured the West, where they are appreciated for their virtuoso singing, catchy tunes, animated performance style, and the dulcet timbres of the sārangī and kamāichā fiddles played by the Laṅgās and Māṅgaṇiyārs, respectively. (The two castes are endogamous, serve different patrons, dress differently, and have different song repertoires, but their performance styles are similar and their rāg repertoire is more or less shared.[23])

[51] Laṅgā and Māṅgaṇiyār music (insofar as it can be regarded as a unified tradition) is a quintessential intermediate-sphere idiom in the way it combines features of classical music with others normally associated with folk music (see Manuel 2015a). Much of the repertoire consists of traditional songs, in many cases stylized versions of women’s songs, which Indian scholars would categorize as lok gı̄t, that is “folksong.” However, musicians identify many of their more elaborate songs (especially those categorized as jāngdā, moṭā gı̄t or baṛā gı̄t—“big song”) with specific rāgs, some of which correspond to Hindustani equivalents, but others of which are unique. Many vocalists cultivate formidable vocal technique enabling them to adorn their songs with impressive khyāl-style tāns. Another affinity with art music is the traditional dependence on elite patronage—in this case, the local, often rural gentry, rather than the urban upper class. At the same time, musicians do not articulate the sort of explicit theory that is characteristic of a classical music. When queried by researchers, they are not able to enumerate beats of meters nor specify names of notes, nor are they able to describe in formal terms the distinctions between rāgs—which themselves are often unclear in performance; indeed, their versions of rāgs (and their rāg repertoires) may vary from one village or even family to another. Many performers are illiterate, and hardly any have had any direct exposure to or familiarity with Hindustani music per se. Morgan Davies, in his insightful dissertation on rāg in Laṅgā/Māṅgaṇiyār music, notes that many performers do not conceive of rāg as an important element in their music, but rather as a mere descriptive aspect of their extant song repertoire.^[24] The knowledgeable elderly kamāicha players he worked with had more extensive knowledge of rāg than many performers, but even their solo instrumental renditions of rāgs tended to consist of strings of established, fixed phrases interspersed with other stock phrases that were not rāg-specific (Davies 2017,135, 212).

[52] As Davies and others have noted, it is likely that the Laṅgā/Māṅgaṇiyār rāg repertoire derives less from inventions by their own communities than from contact, however incomplete and sporadic, with more elaborate and formalized genres—especially Hindustani music, but also Sindhi surs, gurbānī sangı̄t, and possibly Pushti Marg havelī sangı̄t. This contact may well have occurred over the course of several centuries, such that some of the borrowed rāgs have long since acquired (or lost) their own distinctive features. It is evident that in the past at least a few members of these communities had some sort of engagement with classical musicians in local courts such as that of Jaisalmer and nineteenth-century Jaipur, where both folk and classical performers were patronized (see, e.g., Jairazbhoy 1980, Erdman 1985). Through such interactions they acquired some knowledge of rāgs—or at least rāg names—, which they conveyed to their rural kinsmen. The performers incorporated these materials, albeit in an inconsistent and idiosyncratic manner, into their art, lending it a prestige and sophistication uncharacteristic of other village song idioms. These features qualified them to provide music for rural landlords and other gentry, who, in accordance with their own status, for their weddings and other occasions sought music that was more elaborate and prestigious than, say, the lively but unpretentious songs of the Kalbelia “gypsies.”

[53] Laṅgās and Māṅgaṇiyārs do not perform rāgs per se, but if asked to do so, will typically render songs, or a prefatory dohā (lit., couplet, but also the prosodic meter typically used therein), that they regard as being in certain rāgs. The dohā verses either cohere in subject matter with the subsequent song, or, like the contemplative rāg-dhyān verses appended to rāgmālā (rāgamālā) paintings, they praise or personify the rāg in question, as in these examples (from Bhatt 2014, 44):

Nadiyā meiṅ Gangā barī, tirath baro Kedār

Runkhā meiṅ chandana baro, rāga Gūnḍ Malhār.

Greatest of rivers is Ganga, of pilgrimage sites Kedarnath

Greatest of trees is the sandalwood, the greatest rāga Gūnḍ Malhār.

Āsā mhārī lādlī, jhilan gayī tālāb

Mailā to sab dho liyā, viraha na dhoyo jāi

Lovely Āsā went to the pond for a dip, all her clothes and herself were washed

except she could not wash away the feeling of separation from her beloved.

[54] In several North Indian vernacular genres—such as Hathrasi rasiyā (see Manuel 2015b)—it is customary to preface a metered song with a free-rhythmic dohā, which has something of the character of a brief introductory ālāp. In such genres, the dohās typically pertain to the subject matter of the song; by contrast, the Laṅgā/Māṅgaṇiyār practice of singing dohās regarding rāgs is unique to their music and presumably derives from distinctive features of early modern Rajasthani court culture. The tradition of penning metered verses about rāgs is itself ancient; Sanskrit treatises about music were written entirely in verse (for ease of memorization), and they typically described the rāgs both in formal features (especially scale) as well as with fanciful poetic imagery. After the sixteenth century the art of penning such iconographic descriptions (e.g., portraying Toḍī as a damsel in the woods) came to be greatly elaborated by poets of diverse Hindi dialects—including Marwari—, both as a literary idiom and for insertion in rāgmāla paintings. From the seventeenth to early nineteenth centuries, Rajasthan was the primary center for rāgmālā painting and the allied literary art of rāg-dhyān verses. In Rajasthani courts and homes of nobles, such paintings and poetry manuscripts would be passed around, relished, and discussed. On the whole, these arts were conventional pictorial and literary genres whose composition and consumption might have little or nothing to do with music per se.^[25] It is clear, however, that as dynamic and valued parts of western Rajasthani court culture, rāg-dhyān verses were picked up by such literate Laṅgās and Māṅgaṇiyārs who were connected, in whatever capacity, to court culture; transmitted to others in the community, the dohās were adopted as markers of prestige, along with rāg time-theory and, in idiosyncratic forms, the rāgs themselves. As Davies notes, some of the dohās reference Shah Abdul Latif sūrs such as “Sūr Rāno,” or they invoke certain Sindhi ballad personages such as the tragic heroine Soraṭh, suggesting that the rāg by that name may have developed as a stylization of a mode traditionally used in singing the eponymous tale (Davies 2017, 69–71, 344–46, 351–59).

[55] Several of the rāgs, as abstracted from songs and dohās, lack consistently performed distinctive modal features. Quite possibly, musicians of earlier generations encountered these Hindustani rāgs or rāg-names in some context, and then applied these names to items in their extant song repertoire. Thus, for example, though the catchy and popular song “Mūmal” is said by some Laṅgās to be in rāg Mānḍ (Rajhans n.d., 13), it does not exhibit any particular distinctive features of that rāg, and its attribution might be akin to saying that the song “Amazing Grace” is in rāg Durgā.^[26] However, other rāgs in their repertoire—especially those in Bhairav and Pūrvī thāṭs—are quite distinctive, and in their modes they are quite uncharacteristic of any sort of North Indian folk or vernacular music.

[56] A few of the more common Laṅgā/Māṅgaṇiyār rāgs, such as (Sindhi) Bhairavī and Khamāichī (Cf. Khamāj), may roughly cohere with their namesakes in Hindustani music, and versions of Gūnḍ Malhār, Birbhās/Burwās, Sālang, and Jangla may loosely resemble Gaur Malhār, Vibhās, Brindāvanī Sārang, and Hindustani Janglā, respectively (the latter being in any case an obscure archaism in North Indian music). However, other rāgs differ conspicuously from their classical namesakes. Kalyān (Cf. Yaman Kalyān, Karnatak “Kalyānī”) lacks the distinctive raised fourth degree of its classical counterparts, Kāfī is in the “major-scale” Bilāval thāṭ rather than the “Dorian” eponymous thāṭ, Tilang includes the notes Re and Dha (absent in khyāl-style Tilang, though present in old thumri renditions), and Jog resembles not Hindustani Jog—a twentieth-century invention—but the older rāg Jogiya (see Davies 2017, 183, 252)

[57] More relevant to the present article are the Laṅgā/Māṅgaṇiyār rāgs that correspond to the western rāgs we have discussed. Rāg Mānḍ, being more associated with eastern Rajasthan court culture, is not a specialty of Laṅgās/Māṅgaṇiyārs, though eclectic as they are, they can certainly sing “Kesariya Bālam” and a few other Mānḍ songs on request. Also in their repertoire are songs performers identify as exemplars of the distinctively western rāgs Āsā, Soraṭh, Husenī Toḍī, and Mārū, though their versions of the latter tend to resemble the Bhairav-thāṭ havelī sangı̄t rāg more than its gurbānī sangı̄t namesake. While Sindh(i) Bhairavi might be a familiar, if uncommon, rāg throughout the Hindustani music ecumene, its name certainly suggests a western provenance, and it is basic in the Laṅgā/Māṅgaṇiyār repertoire, as in the popular song “Hı̄chkı̄” (Hı̄ṅchkı̄, “hiccup,” also identified as rāg Janglā by some). Other rāgs are idiosyncratic. Gūnḍ Malhār, Shām Kalyān, Kāfi, and some versions of Toḍī are what some might call “nondescript” Bilāval/Khamāj-thāṭ entities, though Davies endeavors, through detailed analysis, to abstract distinctive phrases from the renditions of such rāgs by senior kamāicha players. Lūr Sārang has some character of Hindustani Sārang, though with some use of komal Ga. Sūb, like Mārū, is mostly in Bhairav thāṭ but, in some versions, tīvra Ma is judiciously introduced, sometimes in ways that suggest shuddh Ma as a secondary tonic, as in rāg Lalit. Shuddh Dha can also appear, lending the upper tetrachord some similarity with Mānḍ, such that some Laṅgās refer to “Sūb Mānḍ,” along with “Āsā Mānḍ,” “Sāmerī Mānḍ” and the like insofar as they bear similar affinities (see, e.g., Rajhans n.d., 42–44). Given the inconsistency of versions, renderings of Sūb, Mārū, and Jog may be essentially indistinguishable (see, e.g., Davies 2017, 206). A few Laṅgā/Māṅgaṇiyār rāgs—especially Kohiyārī/Sasu’ī, Kachchhi Kāfi, Rāṇo, and possibly Sāmunḍī—are shared with and presumably derived from Sindhi music, as many hereditary musicians of western Rajasthan historically had ties with Sindhi-speaking patron communities or lived and performed in Sindh itself. Finally, Jairazbhoy encountered Māṅgaṇiyārs whose especially large rāg repertoire included versions of Toḍī in Āsāvarī thāṭ, Sāverī in Bhairav thāṭ, and Paraj in Khamāj thāṭ, all of which might constitute either idiosyncrasies or else remarkable survivals of medieval forms of these rāgs.^[27]

Sindh and Kachchh

[58] Sindh is home to multiple performance traditions that, like gurbānī sangīt and the repertoire of Laṅgā and Māṅgaṇiyār musicians in western Rajasthan, evince connections with the Hindustani music tradition through shared rāg names but are distinguished from it by the melodic content and stylistic delivery of these rāgs. We focus in this section on the performance of rāgs—or surs, as they are calledin Sindhi—in the kāfī and shāh jo

rāg̈^{^[28]} genres. These genres are distinct from one another but have substantial overlap in poetic content and melodic repertoire. Like the musical practices discussed in previous sections, kāfī and shāh jo rāg̈ may be considered intermediate-sphere genres insofar as they are clearly linked to the Hindustani tradition but emphasize poetic texts over rāg elaboration and other forms of musical improvisation. It is worth noting, though, that some stylistic variants of the kāfī genre—namely those performed by artists with training in the classical tradition—could be described as “light classical” to the extent that they feature virtuosic and improvisatory vocal and drumming practices.

[59] Due to historical migration patterns as well as mediatization in the post-Partition era, Sindhi music traditions are found not only in Sindh but also among Muslim communities in Kachchh and western Rajasthan. Kachchh, now a border district of Gujarat, was historically a kingdom—and later a princely state under the British—that had close cultural and economic connections with Sindh. Almost all of Kachchh’s myriad endogamous Muslim communities trace their ancestry to Sindh and speak Kachchhi—a southern dialect of Sindhi—or other variants of Sindhi. Since Partition, many Muslims in Kachchh have sustained their Sindhi musical and poetic heritage by listening to recordings of Sindhi artists from Pakistan on cassettes and, until the late 1990s, on cross-border Pakistani radio. Although Sindhi language literacy has declined in Kachchh since Partition, there are countless enthusiasts of Sindhi music and poetry in the region. Performers from the Laṅgā community of Muslim hereditary musicians, who mostly perform as drummers and soṇā’iṅ (shehnā’ī) players, are a particularly important source of musical knowledge in Kachchh. Although Laṅgās typically do not sing in public, local kāfī singers—who are almost all from agriculturalist and pastoralist castes—often look to them for their knowledge of both Sindhi surs and Hindustani rāgs.

[60] In western Rajasthan, communities known collectively as Sindhi Sipahi, many of whom are Marwari-speaking, have also maintained an interest in Sindhi language, poetry, and music. The Marwari-speaking hereditary musicians who provide musical services for these communities continue to perform Sindhi songs for their patrons. In the past, members of these communities often migrated to Sindh for musical work. On a research trip to Jaisalmer district in early 2018, one of the present authors (BB) met two Māṅgaṇiyār musicians who had lived in Pakistan up until their early teens before returning to India in the 1970s, when the border was more porous than it is today. As mentioned in the previous section, Laṅgā/Māṅgaṇiyār musicians in Jaisalmer and Barmer districts perform numerous Sindhi melody types, some of which they explicitly identify with the suffix “Sindhi”—e.g., Sindhi Soraṭh, which is distinct from the local variant that hereditary musicians there call Soraṭh.

[61] The Sindhi music genres we discuss here rest on the musico-poetic foundation established by the Sufi poet-saint Shah Abdul Latif Bhitai (Shāh ‘Abdul Latīf Bhiṭā’ī, 1689–1752). The shāh jo rāg̈ genre in particular revolves almost entirely around Shah Bhitai’s poetry.^[29] And while Sindhi kāfī repertoire includes countless poetic texts by later poets, Shah Bhitai is the primary touchstone in that genre too; Sindhi kāfī singers typically sing one or more of Shah Bhitai’s baits (verses) as a preface to a song composition (kalām), whether by Shah Bhitai or another poet. Shah Bhitai’s poetic compendium, Shāh Jo Risālo—usually translated as “Shah’s Message”(also known as Ganj, “treasure”)—is organized into discrete groupings of thematically related poems (bait) and strophic song texts (vā’ī). Each of these groupings is referred to as a sur. Before continuing, it is worth noting that sur in Sindhi has a meaning equivalent to rāg as it is used elsewhere in South Asia—i.e., “melodic type” (Powers 1980)—, while the term rāg̈^u in Sindhi usually denotes “music” in a general sense. Sur in this usage appears to be related to the Persian term surūd (melody, song, singing), which was used to denote these groupings in early manuscripts of Shah’s poetry. It is not clear precisely why or when the transition from surūd to sur occurred, except that Ernest Trumpp used the term sur in his influential 1866 edition of Shah Jo Risālo.^[30]

[62] Editions of Shah’s Risālo typically contain between twenty-nine and thirty-six surs, as some editors have excised certain surs on the grounds of inauthenticity. Most of Shah’s surs are based on regional tales (Sindhi qiṣṣo). While surs are often described as “chapters” of the Risālo, it is important to stress that Shah’s surs are firstly performance entities. A sur in this sense is a hybrid entity that is at once a melodic type and a set of linked texts. (The term sur is also used to refer to melodic types not included in or associated with Shah’s Risālo.) Shah’s surs are made up of varying numbers of baits (verses) and vā’īs (song texts), which are intended to be sung in their associated melody type. While the shāh jo rāg̈ tradition has preserved the practice of singing each of Shāh Bhiṭā’ī’s sur-texts in its associated sur-melody, the kāfī genre is more flexible in this regard, with kāfī singers taking creative license to perform some poetic texts/topics in a variety of melodic types.

[63] Like Guru Nanak, Shah Bhitai is said to have received assistance in the process of musically organizing his poetry. According to Agha (1985, 2), Shah Bhitai was close with “two noted musicians from Delhi” named Atal and Chanchal, who stayed with him and his disciples and would sing Shah’s verses for them. Agha writes:

two Indian musicians […] who visited the court of Miyan Noor Muhammad [Kalhoṛo] and later Shah, had considerably helped [Shāh] in the selection of the Indian melodies for his Risalo. In fact they split up the Ganj [manuscript of Shah’s poetry] in appropriate melodies and the verses were sung accordingly. That arrangement has not been disturbed so far. (1985, 53)^[31]

[64] In an essay on the connections between Shah Bhitai’s musico-poetic repertoire and Sindhi musical drama, Tirathdas Hotchand has posited that Atal and Chanchal were members of a bhagat—a traveling band of Muslim and Hindu male performers who wore ankle bells and gowns to “indicate that they were females” ([1960] 1988, 89).^[32] According to Hotchand, these two Hindu musicians traveled with Shah Bhitai everywhere he went. Shah Bhitai’s interest in the Sindhi tradition of musical drama is evident in his choice to allegorize the major tales of the region. Reflecting on Shah Bhitai’s creative repurposing of regional narrative performance traditions, in which stories were associated with specific melodies, the Sindhi scholar Aziz Baloch accurately described Sindhi music as a “thematic music” (1988, 24). Table 1 lists the sur names included in Shāh Jo Risālo and the topical content of their respective dominant poetic themes, about half of which are allegorical poetic reflections on regional stories. This list includes all thirty-six surs traditionally credited to Shah Bhitai, some of which have been excised from certain published editions of the Risālo on the basis of inauthenticity. From this list, twenty-one surs share names with, or include the names of, Hindustani rāgs, as shown in Table 1; however, these correspondences do not imply that the rāgs are identical.

[65] Although Shah Bhitai’s sur repertoire is certainly the best known and most imitated example of the “thematic music” of greater Sindh, he was not the first to set his poetry to specific melodic types. Shah Bhitai’s older contemporary Shah ‘Inayat (or ‘Inat) Rizvi of Nasarpur (1622–1712) composed poetry along thematic lines that was to be performed in particular melodic types (surūd), and his repertoire was a model for Shah Bhitai (Sayed [1988] 2000, 9; N. Baloch 1978, 134).Later poets of Sindh and Kachchh wrote poetry modeled on the thematic basis that Shah Bhitai so firmly established. The Sindhi verses of Sachal Sarmast (1739–1827), for instance, are organized according to melody and theme. In Kachchh, poets such as Umar Luhar of Mathal village (early twentieth century) followed Shah’s model, albeit in the Kachchhi dialect. Luhar’s verses were performed in a now virtually extinct local musical style known as kacchī rāg (“Kachchhi music”), a musical tradition that operated on the same thematic, melodic, and temporal organizational principles as Sindhi music, with singers performing poetic topics in their accompanying melodic types at specific times of night.

[66] One of the present authors conducted field research with kāfīand shāh jo rāg̈ performers in Sindh in summer 2022 and in Kachchh from 2014–2018 (see Bond 2020a), and with Māṅgaṇiyār musicians in western Rajasthan with knowledge of Sindhi kāfīrepertoire. Of the two genres, shāh jo rāg̈ is the most closely associated with the legacy of Shah Bhitai. While in Pakistan there are around 150 shāh jo rāg̈ī faqīrs—as singers of the genre are respectfully called—, there are only four faqīrs left on the Indian side of the border, all living in a single remote village.^[33] In a typical performance of a sur, faqīrs take turns singing baits solo and then perform a vā’ī (strophic song) as an ensemble. Faqīrs accompany themselves on the danbūro, a five-stringed unfretted instrument that provides a tonic-fifth drone

Sur	Dominant themes, associations	Corresponding Hindustani rāg name
Kalyāṇ	Oneness of God; devotion to God & Prophet	Kalyāṇ (Yaman)
Yaman Kalyāṇ	Physician and patient; wine; sacrifice	Yaman Kalyāṇ
Khambhāt	Moon (confidant); camel (ego); the name also references the Gulf of Khambhāt	Khamāj
Khambhāt		Khamāj
Surīrāg	Dangers of seafaring	Srīrāg
Sāmunḍī	Woman longing for her seafarer husband
Suhṇī	Suhṇī-Mehār qiṣṣo	Sohṇī
Sāraṅg (“rainy season”)	Longing for rain; praise of the Prophet	Sāraṅg
Ked̤āro	Battle of Karbala	Kedār
Ābrī (“weak”)	Sasu’ī-Punhūṅ qiṣṣō
Mā’zurī (“helpless”)	Sasu’ī-Punhūṅ qiṣṣō
Desī (“native”)	Sasu’ī-Punhūṅ qiṣṣō	Desī
Kohīyārī (“mountaineer”)	Sasu’ī-Punhūṅ qiṣṣō
Ḥusainī	Sasu’ī-Punhūṅ qiṣṣō	Ḥusenī
Soraṭh	Qiṣṣo of Rāi Diyāch, who ruled in Saurashṭra	Soraṭh
Barvo Hindī	Story of renunciant Barvo faqīr	Barva
Barvo Sindhī	Story of renunciant Barvo faqīr	Barva
Mūmal-Rāṇo	Mumal-Rāṇo qiṣṣo
Ḍhol-Mārū*	Ḍhol-Mārū qiṣṣo
Khāhoṛī	Yogis
Rāmkalī	Yogis
Rip (“pain”)	Lonely wife longing for husband	Rāmkalī
Līlāṅ-Canesar	Līlā-Canesar qiṣṣo
Bilāval	Characters from the end of the Dodo-Canesar qiṣṣo; praise of the Prophet	Bilāval
D̤ahar (“valley”)	Multi-themed: references to bandits of Kachchh; dried-up valley; the Prophet
Kapā’itī	Women’s thread spinning
Pirbhātī (“morning”)	Traveling musician; name refers to early morning	Prabhātī
Ghātū	Morīro (a fisherman) qiṣṣo; dangers of sea
Shīṅh Ked̤āro*	Assorted animal imagery (lion, dog, birds)	Kedār
Āsā (“hope”)	Multi-themed: mystical knowledge, hypocrisy, etc.	Āsā
‘Umar-Māru’ī	‘Umar-Māru’ī qiṣṣo
Dhanāsirī	Praise of spiritual teacher (murshid), specifically Abdul Qādir Jilānī (1078–1166) and Bahā’uddīn Zakariyā (1170–1262)	Dhanāsrī
Pūrab (“east”)	Crow as messenger bird for longing woman; yogis	Pūrab
Kāmod̤ (“love”)	Nūrī-J̈ām Tamācī qiṣṣo	Kāmoḍ
Karāyal	Swan as symbol of spiritual purity
Basant Bahār* (“spring”)	Arrival of spring (coming of the Prophet)	Basant Bahār

Table 1. Thematic Associations and Transregional Musical Connections of the 36 Surs.
* = not included in many editions of the Risālo

spanning two octaves. In Pakistan, the singing portion of a performance is often preceded by taṅd (“string”), an introductory non-metrical instrumental exposition of the melody type played on the zabān string of the danbūro. The rāg̈ī faqīrs in Kachchh very rarely perform taṅd.^[34]

[67] Kāfī is the major Sufi music genre of the Indus Valley region and refers to the performance of song texts composed in the kāfī poetic form, in which strophes alternate with a refrain. Kāfī is performed in the Punjabi, Siraiki, Sindhi, and Kachchhi languages, and Sindhi kāfī is by far the most popular Sufi musical genre in Kachchh. Sindhi-language kāfī is performed in a variety of styles with varying degrees of proximity to Hindustani musical practices. Numerous Sindhi kāfī artists of the twentieth century performed Sindhi-language Sufi texts in a “classicized” form of kāfī, notably Ustad Manzoor Ali Khan, Ustad Muhammad Juman, Ustad Waheed Ali, and Ustad Muhammad Yusuf. Across the stylistic spectrum, kāfī performances typically begin with a brief exposition of the melodic type (ālāp), after which the singer performs a bait, followed by the main song text (kalām). The form of kāfī most popular in Kachchh is based on recorded performances of artists based in southern Sindh, most notably Ustad Mithoo Kachi and his brothers Haji Usman Kachi and Ustad Hashim Kachi. The Kachi brothers were Laṅgā hereditary musicians originally from northwestern Kachchh who incorporated recited storytelling and/or the explication of the metaphorical Islamic meanings of baits in their performances (see Bond 2020b).

[68] Space does not permit a detailed discussion of the differences between surs as performed in shāh jo rāg̈ and kāfī, but one can make a few generalizations. Most importantly, in comparison to shāh jo rāg̈, surs performed in contemporary kāfī usually sound more similar to Hindustani rāgs of the same name. For instance, Sur Kalyāṇ in shāh jo rāg̈ draws from notes in Bilāval and Khamāj thāṭs.^[35] Sur Kalyāṇ as performed in contemporary kāfī, however, sounds substantially close or identical to rāg Yaman, depending on the performer and their training. (Kalyāṇ is an older name for rāg Yaman.) In general, versions of surs as performed in contemporary shāh jo rāg̈ are less diverse in terms of pitch class set, with many surs drawing from the notes of Khamāj and Bilāval thāṭs, and sometimes also incorporating flat Ga.^[36] By comparison, surs as performed in the kāfī genre are more distinct from one another in terms of pitch class set and chalan.

[69] The respective differences in the melodic execution of surs in shāh jo rāg̈ and kāfī make it difficult, in turn, to generalize about the relationship between Hindustani rāgs and the Sindhi sur repertoire. In some cases, a sur may sound quite different from a rāg of the same or similar name, but can nonetheless be recognizably related. This is the case with Sur Kalyāṇ in shāh jorāg̈, which contains signature phrases also found in rāg Shuddh Kalyān (see Huang 2021, 183–191). In other cases, a surwill have a completely different pitch class set from the Hindustani rāg of the same name. The version of Tilaṅg performed by kāfī singers, for instance, draws from the notes of Kāfī thāṭ, while the Hindustani version of Tilaṅg is set in Khamāj thāṭ. (Tilaṅg is not included in the Risālo and is thus not performed in shāh jo rāg̈.)

[70] Addressing these sorts of discrepancies, the renowned Sindhi researcher and writer Nabi Bakhsh Khan Baloch (also known as N.A. Baloch) posited that Shah Bhitai “retained Kalyan, Bilawal, and Khambhat [i.e., Khamāj] in their classical (shuddh) form” and that the other fourteen surs that share names with the Hindustani tradition were “retained in the form in which they were being sung by the people” (N. Baloch 1988 [1973], 65). Baloch concluded that “the functional composition of each of these melodies under Shah’s rāga does not necessarily conform exactly to its classical composition” (ibid.). The names of the surs, therefore, should not be read as transparently denoting the same melodic types as the rāgs with which they share names.

[71] The complex and often murky relationship between Sindhi surs and Hindustani rāgs is particularly apparent when reviewing Sindhi musicological texts that attempt to pin down the melodic content of the surs. A comparison of the essay “Sindhī Sangı̄t Jo Saṅvārīṅdaṛ Bhiṭā’ī” (“Sindhi Music’s Arranger, Bhitai”) by Abdul Aziz Shaikh (1992) with the monograph Sur, Shāh, Samuṅḍ (“Sur, Shah, Ocean”) by Ustad Ameer Ali Khan (2007) reveals that, out of twenty-four surs notated by them, the two authors disagree on the pitch class sets of fourteen surs. (It is possible that the differences between kāfīand shāh jo rāg̈ performance may have contributed to this discrepancy.) Khan seems to base his account of the surs’ melodic content on Hindustani musical practice. For example, whereas Shaikh observes how, “in classical music Pirbhātī is classified as belonging to Bhairav ṭhāṭh, it falls in Bilāwal ṭhāṭh according to the notes of the Sindhi version” (Shaikh 1992, 129), Khan asserts that Pirbhātī is in Bhairav thāṭ.^[37] Similarly, the pitch-class set that Khan provides for sur Pūrab (Sa Re♭ Ga Ma# Pa Dha♭ Ni) is derived from Hindustani Purvī (Purbī) thāṭ, which is rarely if ever heard in Sindhi music (though, as noted, it occurs in Laṅgā/Māṅgaṇiyār music). For Surīrāg, Khan again follows Hindustani practice and writes that the Sindhi version is in Purvī thāṭ (Khan 2007, 147), while Shaikh writes that it is in Khamāj thāṭ (Shaikh 1992, 134–35).^[38] For these reasons, it seems fair to conclude that Khan’s account of the surs is less reflective of contemporary Sindhi musical practice than the accounts of Shaikh and Baloch.

[72] A survey of performances and musicological descriptions of sur Āsā illustrates the difficulty in determining a definitive version of a sur. Shaikh (1992, 130), who wrote that “this rāgiṇī is a special creation (khās pedāvar) of Sindh,” described the Sindhi version of Āsā as largely in Bilāval thāṭ, but with occasional use of the komal Ni from Khamāj thāṭ. Khan (2007, 267), meanwhile, asserts that there are two versions of Āsā in circulation in Sindh: one in Bilāval thāṭ and another in Āsāvarī thāṭ. The version of Āsā performed by the masterful kāfī singer Abida Parveen, is set in Bilāval thāṭ but diverges from other western examples of this melody type with its use of Ga in ascent.^[39] A shāh jo rāg̈ version of Āsā diverges greatly from Parveen’s performance, with prominent use of the flat Ni as well as the flat Ga.^[40] For its part, the Āsāvarī thāṭ version might suggest that certain performers came to identify the name “Āsā” with the similar (though probably historically unrelated) Hindustani name “Āsāvarī” and revised their performance tradition accordingly. In Rajasthani Laṅgā/Māṅgaṇiyār music, similar phonetic linking may have been involved in the case of the obscure rāg Sāmerī, resembling the local Sāverī—an old Hindustani alternate name of Āsāvarī (Manuel 1981, 17–20). Lest such jumbling of phonetic and musical similarities seem indicative of the confusion of poorly educated provincials, it is entirely likely that such transformations and adaptations have been central to the evolution of Indian classical music from its very inception.

[73] Some Sindhi surs are distinctly regional and have no equivalent in Hindustani music, most notably Kohiyārī and Rāṇo. These surs are popular in Sindh and Kachchh and are also performed by Laṅgās and Mānganiyārs in western Rajasthan, where Kohiyārī is called “Sasu’ī” because of its indelible association with the Sasu’ī-Punhūṅ tale. Referring to the version of Kohiyārī performed by kāfī singers and instrumentalists such as alghozo (double flute) players, Shaikh (1992, 122) reflected, “This rāgiṇī is said to be Sindhi music’s representative (numā’indā) rāgiṇī, and this is not incorrect.” A primarily Bilāval thāṭ melodic type, Kohiyārī ascends with a natural Ni and descends with a flat Ni. The signature melody of Kohiyārī is very distinct, and listeners in Sindh and Kachchh instantly recognize it when kāfī singers perform it as a brief introductory ālāpor use its descending phrases as a melodic vehicle for singing baits (Example 3).

Example 3. Signature Melody of sur Kohiyārī.

[74] Rāṇo is another sur that has no equivalent in Hindustani music. This Kāfī-thāṭ melody type is associated with the story of Mūmal and Rāṇo (also known as Mendhro), which takes place in the Jaisalmer area of Rajasthan in the fourteenth century. The version of Rāṇo performed in the kāfīgenre shares a pitch class set with, and is clearly related to, the version of Rāṇo performed in shāh jo rāg̈, but the two versions diverge slightly in their melodic contours. In Kachchh and Sindh, Rāṇo and Kohiyārī are exclusively reserved for singing the Mūmal-Rāṇo and Sasu’ī-Punhūṅ stories, respectively, while singers in western Rajasthan also use these melody types for singing other poetic topics.

[75] In addition to the surs discussed above, there are a few notable regional surs not mentioned in the modern version of the Risālo. Māṅjh, a rāg name also found in Sikh gurbānī sangīt, is a popular Bilāval-thāṭ melody type.^[41] Ilyas Ishqie ([1973] 1988, 58) observed a resemblance between Mānjh and rāg Māṅd as sung by musicians of Rajasthan. The version of Māṅjh performed in contemporary Sindhi kāfī includes an occasional andolan on Pa that touches Ma#. A classic example of this melody type is Noor Banu’s famous rendition of the kāfī composition “Kaḍh Koṭana Māṅ.”^[42] Loṛā’o is another popular Bilāval-thāṭ rāg popular in Sindh and Kachchh. According to Aziz Baloch (1988, 24–25), Loṛā’o is one of the oldest melody types of Sindh and its name is derived from the nomadic Loṛā/Loṛī community (of modern-day Iran and western Pakistan), which historically included musicians. Loṛā’o bears similarities to Māṅjh but is sung in a higher register and has a smaller melodic range.

[76] Other Sindhi surs share names with some of the distinctively western rāgs mentioned in sections above. The Sindhi version of Pirbhātī is set in Bilāval thāṭ and focuses on the figure of the traveling musician in Shāh Bhiṭā’ī’s Risālo. Dhanāsirī—the text of which focuses on praise of the spiritual master in Shāh’s Risālo—is harder to pin down: both Shaikh (1992) and Khan (2007) state that this sur is in Kāfī thāṭ (like the gurbānī sangı̄t namesake), but renditions by shāh jo rāg̈ singers suggest that the version in that genre draws from Bilāval and Khamāj thāṭs, albeit with touches of Re and Dha natural. Shah Bhitai’s Sur Soraṭh is an allegorical retelling of a story concerning the hereditary musician B̤ījal and King D̤iyāc, which takes place in and around Junagadh, on the Saurashtra peninsula. In Sindh and Kachchh, the Soraṭh melody is exclusively reserved for singing poetry based on this story. The Sindhi version of Soraṭh performed by kāfī singers such as Ustad Manzoor Ali Khan and Abida Parveen is essentially equivalent to rāg Desh of the Hindustani tradition.^[43] The version of Soraṭh performed by shāh jo rāg̈īfaqīrs shares some melodic movements with the kāfī genre version of this sur, such as the ascending line Re-Ma-Pa-Ni♭-Dha-Pa, but includes touches of flat Ga, a melodic tendency that is common to many surs in the shāh jo rāg̈ genre. In addition to these surs, kāfīsingers also perform a number of additional melody types not included among the surs of Shah’s Risālo, such as Jilo (Zilā), Durgā, Kalingṛo, and Bhairavī.

[77] For all its connections and parallels to other practices discussed in this article, the Sindhi sur repertoire constitutes a thoroughly regional tradition with much internal variation and diversity. Certainly, it has some affinity with the Sikh gurbānī sangīt repertoire in that both revolve around a compendium of texts that are intended to be sung in particular melody types. And like some Hindustani rāgs, most Sindhi surs have temporal and sometimes seasonal associations, such as Sur Sārang, which is sung in the rainy season. But the Sindhi sur repertoire is unique among the musics discussed here insofar as most Sindhi surs have deep extramusical associations with regional folktales and other themes. While some sur melodies—particularly those not included in the Risālo, such as Loṛā’o and Bhairavī—are used as melodic vehicles for a range of topics in kāfī performance, others such as Rāṇo, Kohiyārī, and Soraṭh have such strong thematic associations that singers consider it improper to use them for singing any other topic.

Conclusions

[78] When one of the present authors attempted, via email, to engage a respected Indian Hindustani music savant on the subject of regional rāgs, the latter curtly dismissed the topic with a categorical, “There are no ‘regional rāgs’.” The authors hope that in this essay they have shown that there is in fact a small but significant category of Western Indian and Sindhi rāgs, involving both Hindustani music as well as regional intermediate-sphere genres which have their own sort of legitimacy. The existence of this category reveals a hitherto ignored regional dimension of Hindustani music culture, and of the musical geography of North India in general.

[79] As we have mentioned, some of these rāgs, such as Champak, may constitute archaic survivals of entities documented in Sanskrit texts, or in regional Hindi treatises and song anthologies of nineteenth-century Rajasthan. Some, such as Soraṭh and Mārū, may always have had a regional character, which Sanskrit writers chose not to acknowledge. Still other rāgs, such as Tukhārī, constitute unique creations of regional communities—in this case, the early Sikhs—which never spread to other regions. In a related category are the repertoires of the intermediate-sphere genres of Sindh and western Rajasthan, which appear to have evolved less through relative proximity to foreign realms—e.g., Persia—than through the specific historical circumstances, including geographical factors and the early presence of art music. Thus, there is nothing particularly “Persian” about Sūb or Sasu’ī, which emerge rather as products of a particular combination of regional isolation and connections.

[80] The rāgs discussed here vary not only in terms of form and origin, but also in their relative status and importance as structural entities in performance. It is clear that for many musicians in the intermediate-sphere genres, the notion of “rāg” functions primarily as a descriptive designation applied to extant songs, rather than a conceptual entity that actively inspires and regulates improvisation and composition. Relevant here is John Baily’s (1981) distinction between “representational” and “operational” models, respectively (see also Davies 2017, 57). Hence, in an intermediate genre such as Laṅgā/Māṅgaṇiyār music, performers may be unlikely to describe rāgs in formal terms, and their conception of it may be intuitive rather than explicit; however, in a genre such as gurbānī kı̄rtan, the original songs might certainly have been composed by musicians with clear understanding of rāg, in the classical sense.

[81] As noted, the Hindustani rāgs in question are specialties of Punjabi vocal gharānās and of the Jaipur gharānā, whose regional origins are somewhat obscured by the fact that most of its current exponents are Maharashtrian. These rāgs are performed primarily by vocalists, not instrumentalists. Of particular regional character, and of special interest in this essay, are rāgs that are performed in more than one western genre, rather than being idiosyncratic items in the repertoire of a single gharānā or community. Thus, some of the rāgs in question are also components of gurbānī sangı̄t, most of whose versions differ from Hindustani music in their emphasis on devotional texts, but whose venerability and, in some cases, performance style, bestow ample status on its rāg repertoire. Hence, performer-scholar Baldeep Singh would not be the only Sikh tradition-bearer to criticize histories of Hindustani music that “ignore everything west of Gwalior.” The occasional rendering of rāgs such as Tukhārī and Mārū in dhrupad and khyāl styles certainly effectuates their legitimacy, and a logical next step would be for instrumentalists and non-Sikh performers to take them up and explore their melodic potential. Indeed, in recent years a few Hindustani vocalists have taken to performing items from the havelī sangı̄t repertoire (Ho 2013), and sitarist Krishna Bhatt has included in his book on Mānḍ a CD in which he performs Sūb Mānḍ and other Rajasthani specialties.

[82] The relationships between regional rāgs and the Hindustani mainstream canon may vary. Hindustani music is in general a sufficiently open and flexible system that it can accommodate new and diverse rāgs. Thus, for example, despite the standardizing effects of Bhatkhande’s works, several new rāgs have entered the mainstream since his time, and some of the rāgs he described have changed in form. Hence, rāgs such as Champak and Husenī Toḍī could certainly continue to be cultivated. However, as has been shown by the attempts of Sikh reformists to align the gurmat sangı̄t rāg repertoire with the North Indian mainstream, relations with Hindustani music can be problematic and controversial. On the whole, it could be said that gurmat sangı̄t is sufficiently vital and popular among Sikhs that it can flourish in several forms, including its distinctive rāg repertoire. It is not inconceivable that organizations might emerge among Laṅgās, Māṅgaṇiyārs, and Sindhi singers seeking to standardize their repertoire and perhaps “legitimize” it by shoehorning their regional rāg versions into Hindustani versions. Davies notes that some of the events organized by Rajasthani folklorist Komal Kothari for Laṅgās and Māṅgaṇiyārs may have inadvertently promoted various classical-type approaches and conceptions, despite his enjoinders to the musicians that they preserve their own rāg forms (2017, 143). One of the present authors has written of a similar sort of tension operant in Trinidad involving the relation between Hindustani music and that island’s “local-classical music,” which has its own prodigious charm and beauty (Manuel 2000). One can only hope that such revitalizing movements can enrich distinctive traditions without erasing or devaluing their unique features.

Rāg	Hindustani	Gurbānī	W. Raj.	SJR	Kāfī	Pushtī
Āsā	X	X	X	X	X
Bairārī/Varātī		X				X
Bihāgṛā/Vihāgṛo/ Bihāg	X	X		X	X	X
Birbhās/Burwās			X
Barvo Hindī/Sindhī				X
Champak	X	X
Dahar				X
Devgandhāri		X
Dhanāsri (Kāfi thāṭ)		X		X	X	X
Durgā	X				X
Gonḍ		X
Gūnḍ Malhār			X
Hı̄r				X
Husenī Toḍī/Husenī	X		X	X
Jangla/Jhanglo	X		X		X	X
Jog			X		X
Kalingaṛo/Kalingaṛā	X				X
Kalyāṇ	X			X	X
Kāmod/Kāmod̤	X			X	X
Kānra		X				X
Kārāyāl				X	X
Kachchhi Kāfi			X
Kedār / Ked̤āro	X			X
Khahoṛī				X
Khambhāt				X	X
Khokar	X
Kohiyāri / Sasu’ī			X	X	X
Lilāṅ			X	X	X
Lūr Sārang			X
Mājh (Māṅjh)		X			X
Mānḍ	X	X	X
Mārū		X	X			X
Māru’ī				X	X
Ḍholā-Mārū				X
Ma‘zūrī				X
Pahāṛī	X				X
Prabhātī		X	X	X	X
Pūrab				X
Rāmkalī	X	X		X
Rāṇo			X	X	X
Rı̄p				X
Sālang			X
Sāmeri			X			X
Sāmunḍī			X	X
Sindh(i) Bhairavi	X		X		X
Soraṭh	X	X	X	X	X	X
Srı̄ rāga (Kāfi thāṭ)		X		X
Sūb			X
Sūhi		X
Suhṇī				X
Tilaṅg		X	X		X
Toḍı̄ (Bilāval thāṭ)			X
Tukhāri		X
Zilo / Jilo (Kāfī)	X				X

Table 2. Distinctive Western Rāgs. “Hind.”=Hindustani music; “Gurbānī”=gurbānī sangı̄t; “W. Raj.”=Langhā/Māṅgaṇiyār music; “SJR”=Shāh jo rāg̈; “Kāfī”=Sindhi Kāfi; “Pushti”= Pushti Mārg. This table includes Durgā, Kalyāṇ, Kāmod, Kedār, Rāmkalī, Tilang, and Srı̄ rāga (Kāfi thāṭ) because their western versions are clearly distinct from their Hindustani namesakes.

Acknowledgments

Peter Manuel wishes to thank Francesca Cassio and especially Bhai Baldeep Singh for guiding him in seeking to understand matters pertaining to gurbānī sangı̄t, and Krishna Bhatt for invaluable information regarding Rajasthani music. Srijan Deshpande also provided insightful comments on this essay. Brian Bond would like to thank Gani Langa, Manthar faqīr Junejo Arab Jat, Abdullah Jat, Mazharuddin Mutwa, Sumar faqīr Jat, and Mitha Khan faqīr Jat for patiently teaching him about the sur repertoire of Sindhi music, Pei-ling Huang for comparing findings on surs in shāh jo rāg̈, and Khete Khan for discussing the music of western Rajasthan. Brian’s field research was funded by the Social Science Research Council and the American Institute of Pakistan Studies.

REFERENCES

Agha, Muhammad Yakoob. 1985. Shah Jo Risalo Alias Ganje Latif. Hyderabad: Shah Abdul Latif Bhitshah Cultural Centre Committee.

Ayyagari, Shalini. 2012. “Spaces Betwixt and Between: Musical Borderlands and the Manganiyar Musicians of Rajasthan.” Asian Music 43(1): 3–33. https://doi.org/10.1353/amu.2012.0005

Baloch, Aziz. 1988. “Sindhi Melodies and Cante Jondo.” In Rhythms of the Lower Indus: Perspectives on the Music of Sindh, edited by Zohra Yusuf, 35-40. Karachi: Dept. of Cultural and Tourism, Govt. of Sindh. [Reproduced from “Spanish Cante Jondo and its origin in Sindhi Music,” Mehran Arts Council, Hyderabad, 1968].

Baloch, Nabi Bakhsh Khan. 1978. Sindhī mūsīqī jī mukhtaṣir tārīkh. Bhit Shah & Hyderabad, Sindh: Shah Abdul Latif Bhit Shah Siqafati Markaz.

_____. 1988. “Shah Abdul Latif – The Pioneer of a New Era.” In Rhythms of the Lower Indus: Perspectives on the Music of Sindh, edited by Zohra Yusuf, 62–66. Karachi: Dept. of Cultural and Tourism, Govt. of Sindh. [Reproduced from Development of Music in Sind, Sind University Press Hyderabad, 1973].

Beach, Milo Cleveland. 1992. The New Cambridge History of India: Mughal and Rajput Painting. Cambridge, UK: Cambridge University Press.

Bharucha, Rustom, and Komal Kothari. 2003. Rajasthan, an Oral History: Conversations with Komal Kothari. London: Penguin Global.

Bhatkhande, Vishnu Narayan. 1951–1957. Sangeet Shastra: Hindustani Sangeet Paddhati (four volumes, in Hindi). Hathras: Sangeet Karyalaya.

_____. 1954–1959. Kramik Pustak Malika (six volumes, in Hindi). Hathras: Sangeet Karyalaya.

Bhatt, Krishna. 2014. Mānd: Court Singing of Rajasthan. Jaipur: Gurukul Music India Publications.

Bhutta, Saeed. 2008. “Kafi: A Genre of Punjabi Poetry.” South Asian Studies 23(2): 223–229.

Bond, Brian. 2020a. “A Heavy Rain Has Fallen upon My People: Sindhi Sufi Poetry Performance, Emotion, and Islamic Knowledge in Kachchh, Gujarat.” PhD dissertation, the Graduate Center of the City University of New York.

_____. 2020b. “Teaching Islam in Song: Storytelling and Islamic Meaning in Sindhi Sufi Poetry Performance.” Asian Music 51(2): 39–71. https://doi.org/10.1353/amu.2020.0019

Cassio, Francesca. 2015. “Gurbānī Sangı̄t: Authenticity and Influences.” Sikh Formations: Religion, Culture, Theory 11(1–2): 23–60. https://doi.org/10.1080/17448727.2015.1023105

Cohn, Bernard S. 1967. “Regions Subjective and Objective: Their Relation to the Study of Modern Indian History and Society.” In Regions and Regionalism in South Asian Studies: An Exploratory Study, edited by Robert I. Crane, 5–37. Durham, North Carolina: Duke University Program in Comparative Studies on Southern Asia.

Davies, Morgan. 2017. “Music Knowledge of the Laṅgās and Māṅgaṇiyārs.” PhD dissertation, SOAS, University of London.

Dennon, Frank. 2010. “The Third Stream: Oḍiśī Music, Regional Nationalism, and the Concept of ‘Classical’.” Asian Music 41(2): 149–179. https://www.jstor.org/stable/40930325

Deodhar, B.R. 1993. Pillars of Hindustani Music. Bombay: Popular Prakashan.

Erdman, Joan. 1985. Patrons and Performers in Rajasthan: The Subtle Tradition. Delhi: Chanakya Publications.

Faqirallah. 1996 [1665]. Tarjuma-i-Mānakutuhala & Risāla-i-Rāgadarpana. Edited and annotated by Shahab Sarmadee. New Delhi: Indira Gandhi National Centre for the Arts.

Gangoly, O. C. (Ordhendra Coomar). 1935. Rāgas & Rāginīs, a Pictorial & Iconographic Study of Indian Musical Modes Based on Original Sources. Bombay: Nalanda Publications.

Grimes, Jeffrey. 2008. “The Geography of Hindustani Music: The Influence of Region and Regionalism on the North Indian Classical Tradition.” PhD dissertation, University of Texas at Austin.

Hawley, John Stratton. 2012 (2005). Three Bhakti Voices: Mirabai, Surdas, and Kabir in Their Times and Ours. New Delhi: Oxford University Press.

Ho, Meilu. 2013. “Connecting Histories: Liturgical Songs as Classical Compositions in Hindustānī Music.” Ethnomusicology 57 (2): 207–235. https://doi.org/10.5406/ethnomusicology.57.2.0207

Hotchand, Tirathdas. 1988. “Sindhi Drama and the Work of Shah Abdul Latif.” In Rhythms of the Lower Indus: Perspectives on the Music of Sindh, edited by Zohra Yusuf, 87–92. Karachi: Department of Culture and Tourism, Government of Sindh, Pakistan. [Originally published by the Bhitshah Cultural Centre, 1960.]

Huang, Pei-ling. 2021. “Ways of Love: Self-making and Repertoire Formation through the Musical Legacy of Shah Abdul Latif in Sindh, Pakistan.” PhD dissertation, Harvard University.

Ishqie, Ilyas. 1988. “Surs in Sindhi Music.” In Rhythms of the Lower Indus: Perspectives on the Music of Sindh, edited by Zohra Yusuf, 56–61. Karachi: Dept. of Cultural and Tourism, Govt. of Sindh. [Reproduced from Development of Music in Sind, Sind University Press Hyderabad, 1973]

Jairazbhoy, Nazir Ali. 1971. The Rags of North Indian Music: Their Structure and Evolution. Middletown, CT: Wesleyan University Press.

_____. 1980. “Embryo of a Classical Music Tradition in Western Rajasthan.” In The Communication of Ideas, 10^th Annual ICAES Series No. 3, 99–109. New Delhi: Concept Publishing.

Kalra, Virinder. 2014. Sacred and Secular Music: A Postcolonial Approach. London: Bloomsbury.

Kapuria, Radha. 2019. “Of Music and the Maharaja: Gender, Affect and Power in Ranjit Singh’s Lahore.” Modern Asian Studies 54(2): 654–690. https://doi.org/10.1017/S0026749X18000446

Kaufmann, Robert. 1968. The Rāgas of North India. Bloomington: Indiana University Press.

Kaur, Inderjit N. 2014. “The Renaissance of Sikh Devotional Music: Memory, Identity, Orthopraxy.” PhD dissertation, University of Michigan.

Khalsa, Nirinjan K. 2014. “The Renaissance of Sikh Devotional Music: Memory, Identity, Orthopraxy.” PhD dissertation, University of Michigan.

Latif, Shāh ‘Abdul. 2012. Shāh Jo Gaṅj, edited by Mohmmad Qasim Rahemoon. Hyderabad, Sindh: Roshni Publications.

Mansukhani, Gobind Singh. 1982. Indian Classical Music and Sikh Kirtan. New Delhi: Oxford & IBH.

Manuel, Peter. 1981. “The Evolution of Todi Raag-s in Indian Music.” Journal of the Indian Musicological Society 12(3): 14–30.

_____. 1989. Thumri in Historical and Stylistic Perspectives. Delhi: Motilal Banarsidass.

_____. 2000. East Indian Music in the West Indies: Tan-singing, Chutney, and the Making of Indo-Caribbean Culture. Philadelphia: Temple University Press.

_____. 2015a. “The Intermediate Sphere in North Indian Music Culture: Between and Beyond ‘Folk’ and ‘Classical’.” Ethnomusicology 59(1): 82–115. https://doi.org/10.5406/ethnomusicology.59.1.0082

_____. 2015b. “Hathrasi Rasiya: An Intermediate Song Genre of North India.” Asian Music

46(2): 3–24. https://www.jstor.org/stable/24913557 Miner, Allyn. 2015. “Raga in the Early Sixteenth Century.” In Tellings and Texts: Music, Literature and Performance in North India, edited by Francesa Orsini and Katherine Butler Schofield, 385–406. Open Book Publishers.

Mir, Farina. 2006. “Genre and Devotion in Punjabi Popular Narratives: Rethinking Cultural and Religious Syncretism.” Comparative Studies in Society and History 48(3): 727–758.

Nag, Ravi Prakash. 1987. Rājasthāni Gı̄ton ro Gajro. Jaipur: Sahityanagar.

Neuman, Daniel, and Shubha Choudhury, with Komal Kothari. 2006. Bards, Ballads and Boundaries: An Ethnographic Atlas of Music Traditions in West Rajasthan. Calcutta: Seagull Books.

Pacholczyk, Jozef. 1978. “Sufyana Kalam, the Classical Music of Kashmir.” Asian Music 10(1): 1–16.

Pollock, Sheldon. 1998. “The Cosmopolitan Vernacular.” The Journal of Asian Studies 57(1): 6–37. https://doi.org/10.2307/2659022

Powers, Harold S. 1980. “Classical Music, Cultural Roots, and Colonial Rule: An Indic Musicologist Looks at the Muslim World.” Asian Music 12(1): 5–39. https://doi.org/10.2307/833796

Qureshi, Regula. 1988. “An Ethnomusicological Perspective.” In Rhythms of the Lower Indus: Perspectives on the Music of Sindh, edited by Zohra Yusuf, 47–55. Karachi: Dept. of Cultural and Tourism, Govt. of Sindh. [Reproduced from Sind Through the Centuries: Proceedings of an International Seminar Held in Karachi, Spring 1975, edited by Hamida Khuhro. Karachi: Oxford University Press, 1981].

Rajhans, Sudha. n.d. Chirmi: Laṅghā Jāti Gāye Dwāra Jāme Wāle Pachchı̄s Gı̄ton ko Svarlipi. [Chirmi: Notations of Twenty-five Songs of the Langhā Caste]. Jodhpur: Rajasthan Sangı̄t Natak Akademi.

Saleem, Agha Khalid. 1984. Sindh meṅ mūsīqī. Islamabad: Pictorial Printers Limited.

Sayed, Durreshahwar. 2000 (1988). The Poetry of Shāh Abdal-Latif. Jamshoro, Sindh: Sindhi Adabi Board.

Shaikh, Abdul Aziz. 1992. “Sindhī sangı̄t jo saṅvārīṅdaṛ bhiṭāī.” In Shāh Latīf jī mūsīqī, edited by Dr. Dure Shawar Sayyad, 119–143. Karachi: Shah Abdul Latif Chair, University of Karachi.

Singh, Bhai Avtar, and Gurcharan Singh. 1988/1979. Gurbānī Sangı̄t: Prācīn Rīt Ratnāvalī, Volume 1 (1988), Volume 2 (1979). Patiala: Panjābī University Publications Bureau.

Singh, Bhai Baldeep. 2011. “What is Kirtan? Observations, Interventions, and Personal Reflections.” Sikh Formations: Religion, Culture, Theory 7(3): 245–295. https://doi.org/10.1080/17448727.2011.637382

Subba Rao, B. 1993. Rāganidhi: A Comparative Study of Hindustani and Karnatak Rāgas. Madras: The Music Academy.

Thakur, Onkarnath. 1957. Sangı̄tājjali, v. 4. Varanasi: Kashi Vishvavidhyalaya.

Thielemann, Selina. 2001. Musical Traditions of Vaisnava Temples in Vraja: A Comparative Study of Samaja and the Dhrupada Tradition of North Indian Classical music: Volume I and II. New Delhi: Sagar Printers and Publishers.

Thompson, Gordon. 1995. “What’s in a Dhāl? Evidence of Rāga-like Approaches in a Gujarati Musical Tradition.” Ethnomusicology 39(3): 417–432. https://doi.org/10.2307/924629

Van der Linden, Bob. 2008. “Sikh Music and Empire: The Moral Representation of Self in Music.” Sikh Formations: Religion, Culture, Theory 4(1): 1–15. https://doi.org/10.1080/17448720802075389

_____. 2012. “History Versus Tradition Again? A Response to Bhai Baldeep Singh.” Sikh Formations: Religion, Culture, Theory 8(2): 247–251. https://doi.org/10.1080/17448727.2012.720476

_____. 2013. Music and Empire in Britain and in India. Identity, Internationalism, and Cross-Cultural Communication. New York: Palgrave Macmillan.

_____. 2015. “Pre-Twentieth-Century Sikh Sacred Music: The Mughals, Courtly Patronage and Canonisation.” South Asia: Journal of South Asian Studies 38(2): 141–155. https://doi.org/10.1080/00856401.2015.1014128

Verma, Vijay. 1987. The Living Music of Rajasthan. New Delhi: Office of the Registrar General.

Widdess, Richard. 1993. “The Geography of Rāga in Ancient India.” The World of Music: Journal of the International Institute for Traditional Music 35(3): 35–50. https://www.jstor.org/stable/43616472

Widdess, Richard and Ritwik Sanyal. 2004. Dhrupad: Tradition and Performance in Indian Music. Burlington, Vermont: Ashgate.

Yodh, S.N. 1978. “Vilayat Hussein Khan of the Agra gharānā.” Journal of the Indian Musicological Society 9(3): 17.

1. Most portions of this article are written Peter Manuel; the section on Sindhi music is written by Brian Bond.

[2]. Although Maharashtra, Goa, and Gujarat are typically included in “Western India,” we are not including these in our discussion (aside from Kachchh, which is culturally distinct from eastern Gujarat and was only incorporated into that state in 1960).

[3]. The closest counterpart would be the Kashmiri Sufyāna kalām, which, as documented by Pacholczyk (1978), is based on a repertoire of songs, in Persian or Kashmiri, set to one of around twenty “maqāms” (not “rāgs”) whose names—such as “Bayāti” and “Jinjoti”—reflect the mixture of Persian and Indian influences the region has absorbed. Orissa, though not generally regarded as part of North India, also hosts a quasi-classical music with some distinctive local rāgs (Dennon 2010). In Gujarat (aside from Kachchh), the song genre “Prabhāti” has some aspects of a mode (Thompson 1995), but if that state’s vernacular music comprises any distinctive modal system, it has yet to be documented.

[4]. This article assumes some basic familiarity with Indian classical music elements. Hence, we shall freely use basic Hindustani nomenclature and sargam, e.g., komal=flat, shuddh=natural, tı̄vra=sharp. The seven scale degrees are: Sa Re Ga Ma Pa Dha Ni, corresponding to (from C): C D E F G A B; the chromatic scale would be: Sa Re♭ Re Ga♭ Ga Ma Ma♯ Pa Dha♭ Dha Ni♭ Ni. The ten thāṭs, though less than satisfactory as frameworks for classifying the entire rāg repertoire, are convenient descriptive concepts, often referred to by musicians themselves. They are as follows:

Bilāval: Sa Re Ga Ma Pa Dha Ni (Cf. “major scale”)

Khamāj: Sa Re Ga Ma Pa Dha Ni♭ (Cf. “Mixolydian”)

Kāfi: Sa Re Ga♭ Ma Pa Dha Ni♭ (Cf. “Dorian”)

Āsāvarī: Sa Re Ga♭ Ma Pa Dha♭ Ni♭

Bhairavi: Sa Re♭ Ga♭ Ma Pa Dha♭ Ni♭

Toḍī: Sa Re♭ Ga♭ Ma# Pa Dha♭ Ni

Bhairav: Sa Re♭ Ga Ma Pa Dha♭ Ni

Pūrvī: Sa Re♭ Ga Ma# Pa Dha♭ Ni

Mārva: Sa Re♭ Ga Ma# Pa Dha Ni

Kalyān: Sa Re Ga Ma# Pa Dha Ni (Cf. “Lydian”)

[5]. Grimes goes on to make quite fanciful arguments about alleged differences in style and their supposed roots in topography, e.g., a supposed Maharashtrian preference for angular tunes and ornaments as reflecting the iconic Sahyadri mountains, as opposed to a Bengali fondness for “flowing” mīnḍ (glissando) expressing that state’s “riverine” character.

[6]. The gharānā is often referred to as the Alladiya Khan gharānā, or the Jaipur-Atrauli gharānā. Alladiya’s ancestors in the two generations preceding him lived in Jaipur state, though the ancestral home was Atrauli, in Uttar Pradesh (Grimes 2008,142). Alladiya is also said to have learned some material from the Pushti Marg kı̄rtan/dhrupad tradition of Nathdwara, Rajasthan.

[7]. A similar ambiguity may pertain to Alladiya’s practice of rendering the leisurely rāg development (barhat, vistār) in medium-tempo vilambit tāl, rather than the standard ati-vilambit (“extremely slow”) tempo. Deodhar seems to suggest that Alladiya adopted this practice because he had ruined his voice and was incapable of singing extended slow development (1993, 30). However, it is generally believed that the ati-vilambit tempo only came into prominence in the mid-twentieth century, in which case Alladiya’s use of medium tempo would constitute an archaism rather than an innovation.

[8]. Accordingly, while Sindh Bhairavī is commonly performed by Rajasthani Laṅgās and Māṅgaṇiyārs, rāg Multānī—whose name suggests an association with western Punjab—has no particular presence in the region. Yet another contrast would pertain to the light rāg Pahāṛī, which, though a pan-regional stylization of hill-region melodies, is not performed as a rāg per se in those regions.

[9]. Renditions of Āsā Mānḍ by Munawwar Ali Khan (Patiala), Nazakat and Salamat Ali, and Pandit Jasraj can be found on YouTube.

[10]. Subba Rao, in his thorough rāg encyclopedia Rāganidhi, mentions it in passing, saying, “Information regarding Āsā is scanty” (1993: I, 35).

[11]. Surprisingly, the Rādhagovind Sangı̄tsār, though written in Jaipur around 1803,makes no mention of Mānḍ, presumably because its authors, though writing in Hindi prose, wished to align it more with Sanskritic textual tradition, especially the seventeenth-century Sangı̄t Parijata, than with local vernacular music.

[12]. Meriting passing mention is the Bilāval-thāṭ rāg Mewāṛā which, as its name suggests, is, like Mānḍ, attributed to Rajasthani origin (specifically, Mewar/Udaipur). Though included in rāg anthologies such as Bhatkhande’s KPM (V, 243), it is effectively extinct in current Hindustani practice.

[13]. Jaipur vocalist Manjari Asanare Kelkar demonstrates these at: https://www.youtube.com/watch?v=bCND4p5OM80

[14]. Renditions of Champak by Punjabi and Rajasthani (Jaipur gharānā) vocalists currently accessible on YouTube include those of Fateh Ali Khan and Umeed Ali Khan (both Punjabi, though of Gwalior gharānā), Roshan Abbas Khan, Ram Krishna Verma, and Khadim Hussein Khan, as well as Paluskar disciples Onkarnath Thakur and Narayanrao Vyas. Umeed Ali also sang a rāg “Tirvan,” about which we have not been able to gather information.

[15]. Typical Champak phrases are (with komal ni): sa re pa, pa dha ni dha pa, dha ma pa ga, re sa re ma.

[16]. YouTube currently offers renditions of this rāg by Kishori Amonkar, Manjiri Asanare Kelkar, and Arun Dravid, all of the Jaipur gharānā. For further discussion of Huseni and Huseni Toḍī, see Manuel 1981, 22-24.

[17]. See, e.g., Deepak Raja’s comments in his blog: http://swaratala.blogspot.com/2011/03/bihagda-and-khokar-whats-difference.html

[18] An idiosyncratic composition is the Punjabi choṭā khyāl “Sāḍe nāl ve miyāṅ.”

[19]. The compound rāgs are: Gaurī Guārārī, Gaurī Dakhnī, Gaurī Chaitī, Gaurī Birāgan, Gaurī Dı̄pakī, Gaurī Purbī Dı̄pakī, Gaurī Pūrbī, Gaurī Māhī, Gaurī Mālwa, Gaurī Mālā, Gaurī Soraṭhī, Āsā Kāfī, Āsāvarī, Āsāvarī Sudhang, Devgandhār, Baṛhaṅs Dakhni, Tilang Kāfi, Sūhī Kāfī, Sūhī Lalit, Bilāval Dakhnī, Bilāval Mangal, Bilāval Gonḍ, Rāmkalī Dakhnī, Naṭ, Mārū Kāfī, Mārū Dakhnī, Basant Hinḍol, Kalyān Bhūpālī, Prabhātī Bibhās, Bibhās Prabhātī, and Prabhātī Dakhnī.

[20]. In the Sarmāya-i Ishrat ma’aruf Qānun-i Mūsiqi (Delhi, 1869), Sri has the scale of Bhairav thāṭ, differing only from Pūrvi-thāṭ in its lowered fourth degree. See presentation by Allyn Miner, at: https://archive.org/details/calauem_200201_omvf0000220_ac

[21]. Sung by Bhai Avtar Singh and Gurcharan Singh, at https://www.youtube.com/watch?v=l2hlhsMrecE

[22]. It scarcely needs to be pointed out that similar-looking words may or may not be cognates. Thus, as Bhatkhande notes (1951–57, IV, 167–69), historical documents verify that “Bairārī,” “Barārī,” “Barātī,” Varātīʼ,” and “Varālī” have all denoted versions of the same basic rāg, but there does not appear to be any etymological or musical relationship between gurbānī “Sūhī,” Rajasthani “Sūb,” and the “Sūhvī” and “Sūhā” of Sanskrit texts and Hindustani music, respectively.

[23]. Davies (2017, Ch. 4) makes a solid attempt to address the extent to which Laṅgā and Māṅgaṇiyār rāg conceptions and repertoires are shared.

[24]. Aside from Davies’ dissertation, some basic information and observations about rāg usage in Laṅgā/Māṅgaṇiyār music are provided by Verma (1987), Rajhans (n.d.), Kothari (1995), Neuman and Chaudhury (2006, 94–111), and Bhatt (2014). Nazir Jairazbhoy gathered substantial data on rāg repertoire in the 1970s, but did not publish his findings, though one of the present authors worked with him and published an early article on Toḍī rāgs (Manuel 1981) drawing from this research; for his part, Bond conducted productive interviews with Māṅgaṇiyārs in western Rajasthan and Laṅgās in Kachchh (who are distinct from the Laṅgās of western Rajasthan).

[25]. Thus, for example, an artist painting “rāg Kāmod” would base his picture not on pensively listening to that rāg but on a tracing—often complete with paint-by-number color indications—that would circulate and be used as a template by many artists (see Beach 1992, 160–62, Gangoly 1935, 96–104, Miner 2015). Similarly, a Hindi poet could even compose a treatise on music—as a conventional literary topic—without knowing anything of that art, rewriting in his own style and language material cribbed from one or more Sanskrit texts. Such an author, like the rāgmālā painter, the poet writing rāg-dhyāns, and the patron who enjoyed such projects, could all easily be deaf, or at any rate ignorant of and indifferent to rāg-based music. Hence the unique nature of the Laṅgā/Māṅgaṇiyār practice of singing rāg-dhyān lyrics which might otherwise be independent of music.

[26]. Similarly, some Indian folklorists (e.g., Nag 1987, 23) have stated that some Rajasthani songs are “based on” Hindustani rāgs such as Pilu, Tilak Kāmod, and Durgā. In this article, by contrast, we are discussing rāgs as identified and named by the performers themselves.

[27]. See Manuel (1981). Lochana’s Rāga-Tarangini places Paraj in Karnāta melā (modern Khamāj thāṭ); modern Paraj is in Purvi thāṭ. Most sixteenth-century treatises place Āsāvari/Sāveri in Bhairav thāṭ rather than its current eponymous thāṭ.

[28]. Rāg̈^u in Sindhi is pronounced with an implosive /g̈/. One also encounters rāg̈iṇī in discussions of melodic types in Sindhi musicological literature, in which it is typically used as a synonym of sur.

[29]. Shah’s poetic compendium has absorbed some verses by other poets, such as Kabir (1398/1440–1448/1518) and Shah Husain (1538–1599).

[30]. Akhtar Dargahi pointed this out in an April 2018 Sindhi-language lecture entitled “Surs of Shah Latif and their Raagas.” (https://www.facebook.com/ElectronicDiary/videos/415454445923910/)

[31]. Pei-ling Huang discussed these two musicians with Ustad Ameer Ali Khan of Hyderabad, a Māṅgaṇhār hereditary musician whose family is originally from the Jaisalmer region of western Rajasthan. Khan claimed that the two musicians’ names were Aṅtarīo and Caṅcarīo, and that they were Māṅgaṇhār ancestors of his from Jaisalmer, not Delhi (Huang, pers. comm.). Bond asked hereditary musicians about this when visiting Jaisalmer in January 2018, but none could confirm the account.

[32]. Members of a bhagat were also known as bhagats, a title still used for many Sindhi Hindu singers.

[33]. Two other shāh jo rāg̈ singers have passed away since 2018.

[34]. For a more in-depth discussion of shāh jo rāg̈ performance form, see Bond (2020a) and Huang (2021).

[35]. For a detailed discussion of melodic performance in shāh jo rāg̈, see Huang (2021).

[36]. Aside from melodic content, surs in shāh jo rāg̈ are distinguished by verbal cues particular to each sur, which are sung at the beginning and end of baits. The verbal cue sung prior to a bait is called sad̤ and the cue sung after a bait is called hungāro.

[37]. South Asian authors have used different spellings of thāṭ / ṭhāṭh. Ṭhāṭh is the Sindhi spelling.

[38]. Shāh jo rāg̈ example of Surīrāg: https://www.youtube.com/watch?v=rca0KgIBOoc .

[39]. https://www.youtube.com/watch?v=3SpWxDlXixM

[40]. https://www.youtube.com/watch?v=WlHbY2QAutY

[41]. Surūd Māṅjh is included in the 1793 manuscript of Shāh’s Gaṅj (Huang 2021, 135). This was the older name for Sur Hīr-Rānjho.

[42]. “Kadh Kotan Man Umar Noor Banu” https://www.youtube.com/watch?v=1yqI3vISjzQ

[43]. “Ustad Manzoor Ali Khan, Dino Rai Diyach (Surr Soorath).” https://www.youtube.com/watch?v=a9swG5ybjBs

Analytical Approaches to World Musics

Analytical Approaches
to World Musics

AAWM Journal 10/2 (2022)

Rāgs of Western India and Sindh

Peter Manuel and Brian Bond^[1]

Introduction

Regionalism and Pan-Regionalism in North Indian Classical Music Culture, Past and Present

Hindustani Rāgs of Western India

Gurbānī Sangı̄t

Laṅgā and Māṅgaṇiyār Music of Western Rajasthan

Sindh and Kachchh

Conclusions

Acknowledgments

REFERENCES

Analytical Approaches to World Musics

Analytical Approaches to World Musics

Manuel_Bond

AAWM Journal 10/2 (2022)

Rāgs of Western India and Sindh

Peter Manuel and Brian Bond[1]

Introduction

Regionalism and Pan-Regionalism in North Indian Classical Music Culture, Past and Present

Hindustani Rāgs of Western India

Gurbānī Sangı̄t

Laṅgā and Māṅgaṇiyār Music of Western Rajasthan

Sindh and Kachchh

Conclusions

Acknowledgments

REFERENCES

Analytical Approaches
to World Musics

Peter Manuel and Brian Bond^[1]