1. CONCEPTS, CONVENTIONS,
HISTORY AND CULTURE

1.1 Elements of Indian Classical Music

Survey the scene when a Hindustani classical music performance is in full flow, and you will see and hear a number of things going on simultaneously. Centre stage is the main artist (or artists). He or she may be an instrumentalist or vocalist; and if a vocalist they may be supported by one or more accompanists on melody instruments, such as harmonium or sāraṅgī (a bowed instrument with many sympathetic resonating strings). Then there will be one or more percussionists, usually playing tabla (a pair of hand drums) or possibly pakhāvaj (a barrel drum used in the dhrupad style). And in the background to all this is the constant buzzing drone of one or more tānpurās , long-necked lute-like instruments—these days increasingly supplemented, sometimes even supplanted, by their digital counterparts.

Much of the joy and intensity of Indian classical music comes from the fact that it is substantially improvised ; unlike its western counterpart it is not performed from notation, but is extemporised in the moment. Even so, the performers do not produce their ideas from a void. Rather, they generate them from well-recognised conventions and structures that help them mobilise a stock of musical materials and formulas acquired over many years of practice into ever new variants . (For more detailed discussions, see Sanyal and Widdess 2004: 130, 143; Slawek 1998; Zadeh 2012.)

The rich inventiveness of a live Indian classical music performance draws ultimately on a small number of fundamental musical elements and their organising concepts. These are: (i) melody, organised by the modal principles of rāg; (ii) rhythm, organised by the cyclic principles of tāl; and (iii) drone, which fixes the tonic note of the rāg as a continuous sounding presence—in Indian classical music there is no concept of key change, and no functional harmony as such. The resonance and manifold overtones of the drone -sustaining tānpurā embody a further fundamental concept in Indian music that also permeates melody: svar. This term could be superficially translated as ‘note’, but it signifies something richer and more aesthetically resonant.

These concepts are explored in the following sections. Prefacing these discussions, we also consider sargam notation —part of the common tongue of musicians, and, as a solmisation system, essential to the discussion of scales, in turn an important component—though arguably not the substance—of rāgs. In the later sections of Part 1, I provide perspectives on the khayāl vocal style and also explore aspects of history and culture that inform the lifeworld of Hindustani musicians.

1.2 Sargam Notation

Although Indian classical music is principally an oral tradition , notation is no stranger to it and is in practice embedded in the oral transmission of musical ideas. The identification of tabla strokes with bols (words) is one example of such oral notation (see Section 1.4). So too is the naming of notes (svar) under the system known as sargam—a term derived from the first four scale steps, Sā, Re, Ga, Ma. Sargam notation is a solmisation system: like western sol-fa, it indicates scale degrees, not absolute pitches. The system tonic, Sā, can be placed at whatever pitch suits the performer or the circumstances; all the other scale degrees are placed relative to it.

The terms for the full heptatonic gamut are: Sā, Re, Ga, Ma, Pa, Dha, Ni. For convenience, these are often abbreviated to their first letter, but they are in fact the short forms of longer Sanskrit names—ṣaḍj, ṛiṣabh, gāndhār, madhyam, pañcam, dhaivat and niṣād. While musicians usually use the shortened note names (‘sing Pa’) they not infrequently use the longer ones too (‘sustain pañcam’). These details are summarised in Figure 1.2.1 (of course, in their original Sanskrit form, these syllables and words would be represented in Devanāgarī characters, not the Roman ones used here and in western transliteration generally).

A table that labels the degrees of the scale with their Sanskrit note names. The table has four columns. The first column lists the scale degrees using numbers from 1 to 7. The fourth column labels these with their Sanskrit note names in full (e.g. ṣaḍj, ṛiṣabh, gāndhār). In between, the third column gives the shortened note names, as used in everyday musical practice (e.g. Sā, Re, Ga); while the second column shows these as single letter abbreviations (e.g. S, R, G).

Fig. 1.2.1 Scale degrees/note names in sargam notation. Created by author (2024), CC BY-NC-SA.

Indian classical music normally operates within a three-octave compass, and the register in which a note is to be sung or played can also be included in the notation. This is the case in the notation system of Vishnu Narayan Bhatkhande (1860–1936), which is used by many Hindustani musicians, and which we have adapted in notating the song compositions in Rāgs Around the Clock. Under these conventions, notes in the lower octave are notated with a dot below the note name; notes in the upper octave with a dot above; and notes in the middle octave with the note name only. For example: Ṣ = lower tonic (mandrā Sā); S = middle tonic (madhya Sā); and Ṡ = upper tonic (tār Sā).

Pitches can be inflected upwards or downwards. Letters without modifications indicate the natural (śuddh) form of a note. A line below a letter indicates its flat (komal) form—applicable to Re, Ga, Dha and Ni; hence R = komal Re, flattened 2nd. In Hindustani classical music only the fourth scale degree can be sharpened; this is known as tivra Ma, and is indicated with a wedge above the note name, thus: Ḿ. These five inflections added to the seven natural notes theoretically make available a full twelve-note chromatic gamut, though not all notes are available in any given rāg, and these are not arranged in equal temperament. Rather, svars may be subject to microtonal inflection (śruti) according to the rāg.

Already this hints at a much more extensive and complex theoretical background. The discussion of microtones, of different possible divisions of the octave (for example into twenty-two śrutis ), is just one aspect of a large body of theoretical treatises (śāstras) on Indian music and related arts. Many of these have been lost, but the oldest known is the Nāṭyaśāstra of Bharata Muni (in fact a treatment of theatre from the early first millennium CE); while the earliest known example of sargam notation is found on the Kuḍumiyāmalai Inscription dating from seventh- or eighth-century Tamil Nadu (see Widdess 1979; 1995: 104–24; 1996). The principal language of these works is Sanskrit, but scholars such as Françoise ‘Nalini’ Delvoye and Katherine Butler Schofield (née Brown) shed light on a further corpus of Indo-Persian texts written between the thirteenth and late-nineteenth centuries that highlight the significance of Muslim scholars and artists in the Hindustani tradition (see Nijenhuis and Delvoye 2010; Brown 2003, 2010). Further, the bigger context for Hindustani music also includes the contribution of Sikh musicians, gurus and scholars (see, for example, recent work by Gurminder Kaur Bhogal (2017, 2022), Harjinder Singh Lallie (2016), and Kirit Singh (2023)).

It is useful for musicians to be aware of the theoretical and historical hinterland behind their practice; for, in truth, Indian classical music has constantly evolved out of a complex and elliptical relationship between oral discipular traditions (sampradāya) and canonical works of theory (śāstra). The quotidian use of sargam notation by practitioners—as a way of talking about music, of communicating knowledge and ideas about it, of making things happen with it (in the context, say, of a class or a rehearsal)—is one example of the mediation of practice by theory . But the implications of this confluence extend beyond pragmatics. The way notation transmutes sounds into concepts (like Sā, Re, Ga ...), the way it structures them into a systemic relationship, the way it brings a pattern and an order to our musical thinking—all these features condition our experience of the music, and pre-empt any naïve dislocation between orality and literacy , and between improvisation and composition .

Critical twists and turns in this argument are possible. For example, Dard Neuman (2012) interrogates notation and classificatory knowledge as ideological aspects of the modernisation of Indian music ushered in by the likes of Bhatkhande in the earlier twentieth century. Neuman cites accounts of how hereditary musicians (of the kind Bhatkhande tended to disparage) would traditionally withhold information about a rāg and inhibit the use of sargam notation until the student had memorised and absorbed the material in an embodied way—though, even then, information was withheld only temporarily, not suppressed permanently.

At subsequent points in this book—notably in Section 4.3—we will again have occasion to consider the creative tensions between notation and practice. For now, it is sufficient to underline notation’s mediating significance in our contemporary musical world, and to note its historical presence in arguments (traced in Powers 1992) for what makes Indian classical music classical—śāstrīya saṅgīta.

1.3 Rāg

Rāg is the source from which all melodic invention in Indian classical music flows. If its concept cannot be finally captured in language this is because rāg has an aesthetic as well as a technical dimension; it is a world of feelings and a mode of tonal organisation.

Because Indian classical music organises tones under melodic rather than harmonic principles, rāg can in certain respects be thought of as a modal system. It is indeed considered as such in Grove Music’s capacious article on Mode (Powers et al. 2001), which notably locates rāg in a pan-Asian and wider global context. One of the modal features of rāg is that it is built on scale forms. A rāg must deploy no fewer than five and may have up to seven scale degrees, with varying inflections. In other words, its scales may take pentatonic, hexatonic or heptatonic form (known as auḍav, ṣāḍav and sampūrṇ, respectively). However, ascending and descending scale forms (āroh and avroh) may differ. For example, the ascending form of Rāg Bihāg (heard on Track 11 of Rāg samay cakra) is based on a pentatonic scale with a natural fourth degree, while its descending counterpart deploys a heptatonic scale with a sometimes sharpened fourth. Sometimes but not always: when and how to execute the sharpened scale degree, so as to enhance rather than disturb the feel of the rāg, is just one of the subtleties a student must learn from their teacher—and just one instance of how the technical and aesthetic blur into one another.

From this we begin to see how a rāg amounts to something much more than its raw scale form. In one rāg, certain notes may be particularly prominent (these may be termed vādī and saṃvādī) while others are only fleetingly touched upon or used to pass between adjacent pitches. In a different rāg using the same scale, degrees of relative prominence may differ, as may the way one note moves to another. Compare, for example Rāg Toḍī and Rāg Multānī (Tracks 2 and 6 respectively of Rāg samay cakra), both of which have flattened second, flattened third, sharpened fourth and flattened sixth degrees, but each of which deploys these notes with different emphases, grammars and expressive palates. Perhaps their most explicit difference lies in their respective vādī and saṃvādī pitches: in Todī, Dha and Gha; in Multānī, Pa and Sā. Furthermore, every rāg has its distinctive melodic turns of phrase—a characteristic that bears out Harold Powers’ claim that ‘a rāga is not a tune, nor is it a “modal ” scale, but rather a continuum with scale and tune as its extremes’ (Qureshi et al. 2020: §III.2.i.a). And certain notes in a rāg may receive particular ornamentation or microtonal inflections (śruti), or both—listen, for example, to the distinctive oscillation (āndolan) around the flattened second and sixth degrees of Rāg Bhairav (Track 1 of Rāg samay cakra; Tracks 1–3 of Twilight Rāgs).

These, then, are some of the ways in which rāg combines and colours tones to generate subtleties of mood and emotion—or ras. While there is no history of formal correlation between the terms, musicians sometimes invoke ras as a way of indicating the appropriate affect of a rāg—as we sometimes do in our commentaries in Section 2.5 of this book. The Sanskrit word rasa literally means ‘juice’ or ‘essence’ or ‘flavour’, and came to denote a theory of emotion in Hindu aesthetics that goes back to Bharata Muni’s Nāṭyaśāstra (2006/1989: 70–85), compiled in the early centuries of the first millennium CE. Bharata outlined eight rasas, of which the most relevant to khayāl are probably śr̥ṅgāra (romance), karuṇa (compassion or pathos), and vīra (heroism); so too is a ninth rasa, śānta (peace), which was adopted by later theorists.

Another way in which a rāg acquires its identity is through its differences from similar rāgs. Hence, part of the work of learning a rāg is also to acquire familiarity with its relatives. In our commentaries we have taken care to indicate, where relevant, some of the salient similarities and differences between a rāg and its neighbours. This points to the fact that rāgs are organised into families—although exactly how one construes the interrelationships of the several hundred rāgs in an ever-evolving repertory has historically been a moot point. Rāgs both invite and resist totalising classification systems, and over the centuries rāgs themselves and their systems of organisation and taxonomy have mutated or been supplanted by new ones. For example, the rāga–rāgini system , in which a series of principal (male) rāgs were construed as governing their own family of ‘wives’, held currency in various versions between the fourteenth and early-nineteenth centuries, but eventually became obsolete when it no longer seemed to conform to actual usage. Joep Bor et al. (1999: 2–4) provide a historical sketch of this and other classification systems; see also Harold Powers and Richard Widdess’s more extended account (in Qureshi et al. 2020: §III.1.ii).

A key reformer in modern rāg taxonomy was V. N. Bhatkhande (a figure discussed at greater length in Section 1.10). Bhatkhande organised rāgs into ten groups, each identified by a parent scale, which he termed ṭhāṭ—as shown in Figure 1.3.1.

A table listing Bhatkhande's ten ṭhaṭs. There are two columns. The first column lists the ṭhaṭ names. The second column indicates the corresponding scale for each ṭhaṭ, using sargam single-letter abbreviations (see column 2 of Figure 1.2.1).

Fig. 1.3.1 Bhatkhande’s ṭhāṭs and their scale types (after Powers 1992: 13). Created by author (2024), CC BY-NC-SA.

It needs to be stressed that the ten ṭhāṭs and their associated scale types remain abstract, theoretical constructs. One would never imagine ‘performing’ the Kalyāṇ ṭhāṭ as such, but one might be aware when performing certain rāgs, such as Yaman , Bhūpālī , Kedār and Kalyāṇ itself, that they are related through being members of this ṭhāṭ—at least according to Bhatkande’s construction. But then, like so much else in Bhatkhande’s life and work, this system remains contentious—among other reasons because it contains many inconsistencies and anomalies (a number of rāgs do not fit readily into it), and because scales are not the only, nor even necessarily the main, principle through which rāgs can be defined and related. These were arguments made by, among others, Omkarnath Thakur (1897–1967), ‘the most articulate and persuasive of Bhatkhande’s detractors’ (Powers 1992: 18). Yet, for all this, Bhatkhande remains a continuing, if qualified, point of reference, both for performers and theorists (see, for example, Nazir Jairazbhoy’s (1971) extended application of Bhatkhande’s ṭhāṭ system).

While we cannot adequately recreate the sound worlds of the now obsolete grāma-jāti, grāmarāga and deśī-rāgā systems, which range from the early first to the early second millennia, scholars such as Widdess have nonetheless adduced ‘evidence for continuity and change in musical concepts, structures and performance’ across these antecedents of our present-day rāg system (1995: 371). So, if the evolutionary timescale of this process and the discontinuities within it warn us not to construe rāg as something unchanging or timeless, this evidence nonetheless points to its sheer historical depth.

1.4 Tāl

Tāl is the term used for the cyclic organisation of rhythm in Indian classical music. In the Hindustani tradition, the job of projecting and sustaining the tāl falls chiefly to the tabla player—or pakhāvaj player in a dhrupad performance. Common tāls include tīntāl (based on a sixteen-beat cycle), ektāl (based on twelve beats), jhaptāl (ten beats), rūpak tāl (seven beats), keharvā tāl (eight beats; used for light classical music , including devotional bhajans ) and dādrā tāl (six beats; also used for light classical music, including ṭhumrī ).

A tāl’s feeling of endless recurrence seems of a piece with the unchanging background drone on Sā; but what gives a tāl its cyclic quality? This is a more complex matter than might at first appear (a point explored in Clayton 2000: chapters 4–5). Some obvious features include the unchanging length of each tāl cycle (āvartan), measured as a fixed number of beats (mātrās ); and the unbroken succession of one cycle by the next. Moreover, each tāl organises its cycle in a characteristic way that gives it a unique shape and flow . This is related to the fact that tāl is not only cyclic but also metrical: the mātrās of a tāl cycle are organised into subdivisions known as vibhāgs (broadly similar to western music’s grouping of beats into bars). This gives a tāl its distinctive pattern. Consider Figure 1.4.1, which shows the metrical profile of tīntāl, one of the most common tāls in Hindustani classical music. Looking along the top row of this figure , you can see how one āvartan (cycle) comprises sixteen mātrās (beats) organised into four vibhāgs (subdivisions) of four mātrās each:

A tabulated representation of tīntāl. The sixteen beats of the tāl cycle are numbered along the top row (1, 2, 3, 4 etc.). This row of figures (and the information below it) is divided into four four-beat subgroups (vibhāgs) using vertical lines. On the next row down, each beat is labelled with its respective tabla stroke (e.g. dhā, dhin, dhin, dhā for beats 1-4). The row below this gives the clap pattern for the tāl, indicating whether to clap or wave at the beginning of each four-beat subgroup. The bottom row names the claps or waves as tālī or khālī respectively, except for the first beat which is labelled sam.

Fig. 1.4.1 Tīntāl: metrical structure and clap pattern. Created by author (2024), CC BY-NC-SA.

One way in which this metrical structure can be communicated is through its clap pattern. This is partly a didactic device, but it is not absent from performance either. In a live concert, audience members can often be seen discreetly tapping out the tāl’s clap pattern with their fingers or on their knee; and in the Karnatak tradition, vocalists themselves may overtly execute the clap pattern as they sing. The details of the clap pattern for tīntāl along with its accompanying notation are shown in the bottom three rows of Figure 1.4.1, and can be explained as follows:

The beginning of each vibhāg is signalled with a clap or a wave depending on its place in the cycle (see third row up).

Beat 1 of the entire cycle is known as sam (see bottom row). In tīntāl and almost all other tāls, this is indicated with a clap, which is notated ‘x’ (a notable exception is rūpak tāl, which begins with a wave). Because tāls are cyclic, sam simultaneously marks both the beginning of each āvartan and the end of the preceding one. It is structurally the most significant feature of a tāl, and, after an accumulation of several āvartans, sam can mark a climactic point of release.

By contrast, the complementary point of a tāl (and its associated vibhāg) is termed khālī, meaning ‘empty’; this point is signalled with a wave, and is notated ‘o’. In tīntāl, khālī falls on beat 9, at the opposite pole to sam, giving this tāl a distinctive symmetrical aspect. Some tāls have more than one khālī—for example, ektāl, discussed below.

Stressed beats other than sam are termed tālī, and, like sam, are clapped. In tīntāl these are found on beats 5 and 13, and are notated with the numbers 2 and 3, denoting the second and third claps of the cycle (see second row up in Figure 1.4.1).

This technical description perhaps risks making the structure seem more complex than it actually is. The best way to grasp these points in the first instance is simply to count the beats out loud and clap or wave in the correct place, repeating this until the pattern is ingrained in the body. A further important point is made by Neil Sorrell , who reminds us that sam (or tālī) and khālī do not simplistically correspond to stressed and unstressed beats as they might in western metre (Sorrell and Narayan 1980: 117). Sam might be a point of focus, a locus of organisation, but it is not necessarily articulated with a major accent (rūpak tāl is a case in point). Similarly, khālī may be associated with lighter strokes, but in symmetrical tāls such as tīntāl it is nonetheless a complementary point of focus that is in its own way structural—as saṃvādī is to vādī in a rāg, we might conjecture.

Just as a rāg is more than a scale, so a tāl is defined by more than the number of its beats or its combination of claps and waves. In the Hindustani tradition, another way in which a tāl is expressed is through its ṭhekā—a pattern of drum strokes unique to the tāl and complementing the clap pattern . Drum strokes in Indian classical music are taught and identified by their bols (meaning ‘words’), such as ‘nā’, ‘tin’, ‘ghe’, ‘ke’; and the ṭhekā for any given tāl combines such strokes into a unique sequence. For example, the ṭhekā for tīntāl is as shown in the second row of Figure 1.4.1, beginning with the bols ‘dhā, dhin, dhin, dhā’. These particular bols onomatopoeically convey the simultaneous combination of a percussive stroke on the smaller, tuned tabla drum (dāyā̃) and a resonant stroke on the larger one (bāyā̃). Conversely, the strokes that follow khālī, on beats 10–13 (... tin tin tā | tā...), do not involve the larger drum, and hence the feel is lighter at this point. All musicians should know the ṭhekā of any tāl they are using. So, students should practice speaking it out loud until it becomes second nature, and then combine it with the relevant clap pattern . Again, this is a way to know the shape and flow of a tāl experientially.

The tabla player may decorate the ṭhekā, but usually ensures it remains discernible. This helps the vocal or instrumental soloist remain oriented within the tāl, so that after improvisatory passages they are able to resume the composition at the right point. Hence, all the performers have a stake in tāl and need constantly to hold it in their awareness. The same goes for the audience : because the tāl forms the organising metrical framework against which the musicians improvise, listeners who are able to follow it are likely to gain greater insight into the performers’ invention—hence the subtle participation of audience members in keeping time, a gesture of identification with the musicians.

While almost all the compositions on the Rāg samay cakra album are in tīntāl, the bandiś for Rāg Basant deploys ektāl, as do several compositions on Twilight Rāgs from North India. Details of this twelve-beat tāl are given in Figure 1.4.2. Note here how ektāl has two khālī vibhāgs , yet in the first of these, on beat 3, khālī is simultaneously associated with a wave within the clap pattern and a heavy tabla bol within the ṭhekā. The converse is also true on mātrā 5, where a tālī in the clap pattern corresponds with a khālī bol (‘tū’) in the ṭhekā. These seeming contradictions between ṭhekā and clap pattern —discussed at greater length by Martin Clayton (2000: 65–6)—reinforce Sorrell’s point that sam and khālī do not necessarily correspond to phenomenal stress or accentuation.

A tabulated representation of ektāl. The twelve beats of the tāl cycle are numbered along the top row (1, 2, 3, 4 etc.). This row of figures (and the information below it) is divided into six two-beat subgroups (vibhāgs) using vertical lines. On the next row down, each beat is labelled with its respective tabla stroke (e.g. dhin, dhin, dhāge, tirakaṭa for beats 1-4). The row below this gives the clap pattern for the tāl, indicating whether to clap or wave at the beginning of each two-beat subgroup. The bottom row names the claps or waves as tālī or khālī respectively, except for the first beat which is labelled sam.

Fig. 1.4.2 Ektāl: metrical structure and clap pattern. Created by author (2024), CC BY-NC-SA.

Tāls can be performed across an entire gamut of speeds, but Hindustani classical music nominally uses three categories of lay (tempo): vilambit (slow); madhya lay (medium tempo); and drut (fast). While these represent quite broad tempo bandwidths as measured by the metronome, each has its own personality. Madhya lay, for example, does not just occupy a mid-point between vilambit and drut lays, but has an unhurried, relaxed character of its own. It is also possible to perform in ati vilambit (very slow) and ati drut (very fast) lays—the former usually the hallmark of a baṛā (large, grand) khayāl (as discussed in Section 4.3); the latter often coming into play at the culmination of many an extended instrumental or vocal performance (I make a detailed analysis of just such a drut khayāl in ektāl by Vijay Rajput (henceforth VR) in Section 4.2).

1.5 Tānpurā Drone, Svar

The tānpurā , a long-necked fretless lute, is responsible for providing the most elemental feature of Indian classical music: its drone . This instrument is ostensibly easy to play, requiring only that the player pluck the open strings at an unchanging tempo—different from that of the other artists—remaining outside the accumulation of speed and intensity characteristic of Indian classical music performances (see Clayton 2007 for an empirical investigation of these conditions). Arguably, tānpurā players do not overtly perform, but occupy a liminal place both inside and outside the performance.

Nowadays the tānpurā drone is often generated digitally, either by electronic śruti boxes or—increasingly commonly—by smartphone apps. Although ostensibly for practise purposes, such devices are commonly plugged into the sound system of concert performances, boosting the sound of the live tānpurās. Occasionally, in an instrumental performance, the live tānpurā might be entirely obviated by its electronic counterpart softly unfurling its drone in the background.

The tānpurā player’s relatively low status in the performer hierarchy is mitigated by the fact that he or she may be a senior student of the principal artist; hence this accompanying role can carry a degree of kudos. If the soloist is a singer, the tānpurā-playing student may be solicited to provide short periods of vocal support during a performance. And for all its modest technical demands, the tānpurā itself is held in high esteem—philosophically because it embodies svar (see below), and practically because it provides the essential reference point for the other performers’ tuning.

Most tānpurās have four strings that are plucked in a steady repeating sequence. The middle two strings are tuned to the tonic in the middle register (madhya Sā); the lowest to the lower-octave tonic (mandrā Sā); and the second lowest usually to the lower fifth degree (Pa) or, if that note is not present in the rāg being performed, to some other appropriate scale degree—most usually the fourth (Ma) or seventh (Ni).

What gives the instrument its characteristic buzz is the sandwiching of a cotton thread between each string and the flat bridge at a key nodal point. This generates a cascade of harmonics—an effect known as javārī, meaning ‘life-giving’ (see Datta et al. 2019). It is to this acoustic panoply that the soloist attunes, listening to the vibrations and seeking to match and intensify them. Musicians sometimes say that all the notes of a rāg are already there in the sound of the tānpurā ; the performer’s job is merely to tap into them and release them.

This is the experience of svar—not merely ‘note’ in the prosaic sense of a specific pitch with a specific amplitude and duration, but also a phenomenal convergence of sound and self, of tone and feeling. Svar can be coloured and enhanced by the śruti (microtones) that lie within it; and as one svar melts into the next, so this concept blends into that of rāg. Some musicians also invoke a relationship between svar and nād—a term for musicalised sound that has metaphysical resonances going back to the Vedic era (ca. 1500–500 BCE) (see Beck 1995; Rowell 1998). Related to this are philosophical ideas such as nāda yoga—the oneness of mind and body achieved through sound—and nāda brahma—the convergence of sound and consciousness .

As evocative as these ideas are, we have to be careful about regarding them as essential or universal (see Clayton 2000: 6–7, 10–13). For one thing, the omnipresence of a tonic drone in Indian classical music is probably a relatively recent development. Research by Chaitanya Deva (1980b: 47–75), Lewis Rowell (1998: 293), Bonnie Wade (1998: 195–8) and Widdess (1995: 7)—based variously on the historical evidence of iconography, instrument construction and primary textual sources—suggests that this performing practice may go back to no earlier than the fifteenth century. In the present day, notions such as nāda yoga and nāda brahma hold greater currency for some genres, gharānās and individual musicians than others. They feature, for example, in the discourse of certain lineages of the dhrupad style, such as the Ḍāgar bānī (notably the Gundecha brothers ), and among some exponents of the Kirānā gharānā, including VR.

1.6 Rāg and Time: Samay Cakra

The title of our book, Rāgs Around the Clock, references a convention fundamental to Hindustani classical music: that a rāg should be performed at its proper time. The Sanskrit term for this principle is samay cakra—where samay means ‘time’, or, more specifically, ‘at the appointed time or right moment’ (Monier-Williams 1899), and cakra denotes ‘wheel’ or ‘circle’.

There is no single agreed representation of this time cycle (see Wade 2004/1999: 77–8). One example, from a now superseded incarnation of the ITC Sangeet Research Academy website, divides the diurnal cycle into twelve segments—as shown in Figure 1.6.1. Under another convention—the one we adopt in Rāgs Around the Clock—day and night are each divided into four quarters, or prahars . The Sanskrit prahara literally means ‘watch’ as in the watch of a guardsman; a similar word, prahāra, means a striking or hitting, as in the sounding of the hours on a gong. This eightfold schema is represented, in conjunction with the rāgs of our albums, in Figure 1.6.2.

A visual representation of the performing times of rāgs within the daily cycle. Names of rāgs are shown within a circle divided into twelve segments - six segments for each phase of the day and night. The twelve phases are labelled around the outside of the circle, e.g. 'Morning' for the period between 8 am and 10 am. Rāgs are allocated to the segment appropriate to their performing time.

Fig. 1.6.1 Samay Rāga—based on twelve time periods (after website of ITC Sangeet Research Academy). Created by author (2024), CC BY-NC-SA.

An alternative visual representation of a rāg cycle (cf. Figure 1.6.1). Names of rāgs are indicated around the outside of a circle divided into eight segments - four segments for each phase of the day and night. Performing times are indicated inside each segment, e.g. '1st quarter of the day'. The rāgs shown here are those performed on Rāg samay cakra, the first album accompanying this book.

Fig. 1.6.2 Rāg samay cakra—based on eight time periods. Created by author (2024), CC BY-NC-SA.

Beginning with the dawn rāg, Bhairav, our first album, Rāg samay cakra takes us through to Rāg Mālkauns , a midnight rāg. Some rāgs have associations with the annual cycle of the seasons, so we have also included performances of Rāg Basant , which celebrates springtime and at other times of the year can be performed in the final quarter of the night, and Rāg Megh , a monsoon rāg.

Just how essential—and how old—is the connection between rāgs and their sanctioned performing times? Although Karnatak (South Indian) music has largely dropped the association, in the Hindustani tradition it still has currency. Musicians may privately practise a rāg at any time, but to publicly perform a rāg at the incorrect time would be to invite disapproval. Some leeway is allowed, however—for example, in the licence to perform a late-night rāg such as Mālkauns in the latter part of a concert, provided that it does not precede a rāg with an earlier performing time.

This culturally regulated practice may have some psychological basis. But, as with so many cultural factors, it also carries the weight and sanction of history. The samay principle has a long pedigree, although not as long as that of rāg itself. Antecedents of the latter are discussed in written treatises going as far back as Bharata’s Nāṭyaśāstra, dated between approximately 500 BCE and 500 CE. On the other hand, the earliest sources in which we find connections between rāg and performing time come from the medieval period. Possibly the earliest documented account of rāg and time is found in the Saṅgīta-makaranda of Nārada, where the author states:

One who sings knowing the proper time remains happy. By singing raga-s at the wrong time one ill-treats them. Listening to them, one becomes impoverished and sees the length of one’s life reduced. (Part 1, slokas 23–4; as cited in Daniélou 2003: 95.)

Typically of such treatises, Nārada’s work is difficult to date with precision. In his introduction to the 1920 edition of the text, Mangesh Rāmakrishana Telang locates it between the seventh and eleventh centuries CE (Nārada 1920: viii–x), a view consistent with M. Vijay Lakshmi’s overview of the extant scholarship (1996: 40–1, 65). On these views, the treatise would predate Śārṅgadeva’s early thirteenth-century Saṅgīta-ratnākara (1978, 2023/1993), one of the most important medieval treatises on Indian music, and another early source that links rāg and performing time (Śārṅgadeva 2023/1993: chapter II). However, Widdess (personal communication) queries whether the Saṅgīta-makaranda is the earlier text, since it anticipates the later rāga-rāgiṇi system; Widdess nonetheless acknowledges that it may be the earliest rationalisation of rāg–time association (as opposed to the mere assertion of the principle in treatises such as the Saṅgīta-ratnākara and the Bharata-bhāṣya of Nānyadeva). By contrast, Shripada Bandyopadhyaya implicitly endorses the view of Nārada’s treatise predating Śārṅgadeva’s; in his chronological exposition, Bandyopadhyaya places the Saṅgīta-makarand between the Br̥had-deśī of Mataṅga (ca. sixth to eighth century) and the Saṅgīta-ratnākara (thirteenth century), and also comments that ‘Saranga Deva … followed in the footsteps of his predecessors, denying only the principles of masculine and feminine Rāgas expounded by Nārada Muni in his work “Sangeet Makaranda”’ (1977:17).

Another account of time theories in the early sources is given by Mukund Lath (1987), who remains sceptical about any ‘psycho-physiological’ basis (see also Wade 2004/1999: 78–9). Lath reminds us that another important figure in the promotion of time theor y was the twentieth-century musicologist Bhatkhande (discussed in Section 1.10, below). Bhatkhande integrated time theory with his ṭhāṭ system, which groups rāgs into families based on their scale form (as recounted in Section 1.3). He attempted to show how rāgs associated with the same performing time often also have commonalties in their scale form. For example, rāgs using flattened second and natural third and seventh degrees are commonly associated with dawn or dusk (sandhi prakāś rāgs). Bhatkhande also argued that rāgs sung in the hours of darkness tend to prioritise the lower tetrachord of the scale (pūrvaṅg), while those sung in the daylight hours favour the upper tetrachord (uttaraṅg). These correlations and others are also raised by Chaitanya Deva (1980a: 19–20) in an account not dissimilar to Powers’ paraphrase of Bhatkhande (cf. Powers 1992: 15–16); subsequent engagements with Bhatkhande’s theories can be found in work by N. A. Jairazbhoy (1971: 61–4). Powers tells us that Bhatkhande looked to treatises written between the fifteenth and eighteenth centuries—rather than to earlier texts such as the Saṅgīta-ratnākara—for theoretical evidence of the living practice he knew, and on which to develop his own time theory and ṭhāṭ system (1992: 11). But Powers also relates how practitioners such as Thakur disparaged Bhatkhande’s system and looked back to the earlier treatises for his own theories of rāg organised on principles other than scalic ones.

Where does this leave us? On the one hand, critiques such as Thakur’s point to the often self-confessed inconsistencies in Bhatkhande’s system, and to the lack of any conclusive empirical evidence for the association between rāg and samay. On the other hand, in the absence of any subsequent alternative systematic theory, Bhatkhande’s word continues to hold considerable authority—a point conceded by Bor et al. (1999: 4). There may be no objective resolution to the issue. The connection between rāg and time is arguably not an essential one, but is made ‘natural’ or ‘real’ as part of a historically mediated cultural practice that also has regard for the ambience of each phase of the diurnal cycle. Rāg Yaman , for example, can be felt to tap into the stillness of the early evening, the sun just set, the day’s work completed; complementing this, the rāg’s tone colours—Ni and Ga in reposeful prominence, Re their congenial mediator, Ma gently yearning, all resonating peaceably against the prevailing tonic—feed back into those sensibilities and reinforce them. Deva (1980a: 20) makes a similar point about sandhi prakāś (twilight) rāgs:

Sandhi means a junction: the passing of night into day and day into night. There seems to be a psychological significance in this. For, it is the time of mental twilight between the conscious and the non-conscious : a time when one sits for prayer and meditation. The dissonances engendered by [komal] ri, [komal] dha and [suddh] Ni go well with the ‘dreamy’ state of mind during these hours.

Deva’s subsequent ruminations (ibid.) include the observation that some twilight rāgs have both tivra and suddh Ma, presumably resonating with their liminal place in the time cycle.

The global diaspora of South Asian musicians and their music adds a further twist to these experiences. For example, the subtropical daylight hours and climate of northern India do not map comprehensively onto more remote northern or southern latitudes. At VR and David Clarke’s (henceforth DC) own geographic location at 55 degrees North, where the length of daylight varies significantly between Summer and Winter solstices, there may be no right time or ambience for certain rāgs at certain points of the year. Consider Rāg Yaman once again, which in India would be sung after sunset in the first quarter of the night, between around 6 and 8 pm. At midsummer in the North of England, however, the sky may not be fully dark even at midnight, the time of Rāg Mālkauns . As nature and culture slip out of sync in these different geographic contexts, it requires a work of the imagination (the meaning of khayāl), to reunite India and the foreign land (an idea so often invoked in khayāl songs) in our minds.

1.7 Khayāl: Stylistic and Performance Conventions

All the rāgs in Rāgs Around the Clock are performed in the khayāl style—a vocal idiom that rose to pre-eminence in the eighteenth century and has remained centre stage among Hindustani classical vocal genres ever since. Aesthetically, khayāl occupies a middle ground between the older, more sober dhrupad style and the romantic, light-classical genre of ṭhumrī . In Section 1.9 we will look at some recent perspectives on the historical origins of khayāl, but first it will be useful to examine some of its present-day conventions. What is it that performers expect to do, and that listeners expect to hear, in a khayāl performance?

Khayāl is an Indo-Persian /Arabic term meaning ‘imagination’, a quality reflected in the music’s largely improvised nature. The vocalist must extemporise within the constraints of rāg and tāl, also drawing inspiration from one or more short song compositions (bandiśes) based on just a few lines of poetry. The texts are usually devotional or romantic, or maybe both, since divine and human love are not opposed in this imaginative world (a topic discussed further in Section 2.2).

A khayāl performance may be expansive, or concise, or somewhere in between. A fully-fledged presentation—which normally begins a recital and in present-day practice might last 30–60 minutes—will comprise all of the components listed in the outline below: ālāp, baṛā khayāl and choṭā khayāl. For shorter performances, it is possible to perform just an ālāp and a single choṭā (small) khayāl. It is this latter, simpler framework that students first aspire to master, and that forms the basis for all fourteen tracks of our first album, Rāg samay cakra. By contrast, the second album, Twilight Rāgs from North India, presents two extended rāg renditions, both of which include a baṛā (large) khayāl; this provides a model for more advanced students. The three stages of the schema below are cross-referenced to further, in-depth explorations undertaken in Parts 3 and 4 of this book, which are also illustrated by examples from VR’s performances.

Ālāp: an unmetred, meditative exploration of the chosen rāg, improvised by the soloist without tabla accompaniment (cf. Section 3.2). If followed by a baṛā khayāl, this may be very short, as ālāp principles are in any case impregnated within the first phase of that section.

Baṛā (large) khayāl: the most substantial section of an extended khayāl performance (cf. Section 4.3). The inception of a baṛā khayāl is marked by the entry of the tabla, which joins the voice and establishes a slow, or very slow, tāl—hence, this stage is sometimes known as a vilambit (slow) khayāl. A baṛā khayāl is based around a rhythmically fluid composition, and comprises two principal phases:
1. The soloist sings the first part (sthāī) of the composition (bandiś) and then embarks upon an extended series of improvisations that retain an ālāp-like feel, despite the presence of the tāl, and create a sense of staged development (baṛhat). Each improvisation ends with the opening motif (mukhṛā) of the composition, which articulates sam (the beginning/end point of each cycle of the slow tāl). The music gradually intensifies , rising in register until the soloist achieves and sustains the upper tonic (tār Sā). This triggers the second part (antarā) of the composition, which is followed by the second phase of the baṛā khayāl.
2. The soloist returns to the mukhṛā of the first part of the composition, and the feel now becomes more dynamic (the tempo may also move up a gear) as the soloist improvises using a variety of devices, such as:

The intensity continues to build, leading to the next stage of the performance.

Choṭā (small) khayāl: a faster or up-tempo khayāl—hence also known as drut (fast) khayāl—based around a further bandiś in the chosen rāg, with improvised elaborations (cf. Section 3.3). Again, the bandiś is in two parts: sthāī and antarā. The soloist is at liberty either to sing both parts at the outset or (quite commonly) to delay the introduction of the second part until completing a significant period of improvisation around the first. The first line of each part (especially that of the sthāī) usually receives the most emphasis, and is often treated as a refrain (mukhṛā). Between statements, the soloist improvises using a similar repertory of devices to that listed above for the baṛā khayāl.

The performance may continue with one or more further compositions, often with a corresponding increase in lay. It sometimes concludes with a tarānā, deploying a composition based on non-semantic syllables such as ta, na, de, re, nūm. This could be considered as an optional, fourth section of the performance.

1.8 Khayāl: Ornamentation

Ornamentation is a vital aspect of rāg music in its many guises. Indeed, there is probably no type of Indian music in which ornamentation does not in some way play a part. Degrees and styles of ornamentation vary according to genre, gharānā (stylistic school), performer, mood and circumstance: relatively sparingly in dhrupad , fulsomely in ṭhumrī , abundantly in some light music and film music idioms. Midway along this continuum is khayāl, which has its own repertory of ornaments, some distinctive to itself, others held in common with related genres. Pick a few seconds of any track on the albums accompanying this book, and you will hear that many of the notes VR sings are either ornamented by or ornamenting some other note—sometimes subliminally, discreetly, delicately; at other times overtly, effusively, exuberantly.

Terminology with respect to ornamentation is beset by ambiguity. The word alaṅkār, often used in Hindustani music to mean adornment or decoration, may also (or instead) be used by musicians to mean sequential practise exercises. Meanwhile, in Karnatak music, the general term for ornament is gamak, which, confusingly, signifies a particular kind of ornament in Hindustani music. Moreover, different musicians or different stylistic schools may mean slightly or appreciably different things by individual names of ornaments.

My aim here is not to definitively resolve such ambiguities, but rather to offer my own window onto their complexities. To my knowledge, the most perspicacious account in English of ornamentation in khayāl is by Nicolas Magriel (Magriel and du Perron 2013, I: Chapter 6), who does not shy away from ambiguities of terminology. Here I offer a more concise treatment, which is based on VR’s gāyakī, and adds further to this not entirely reconcilable mix of viewpoints. In what follows, I will consider the main classes of ornament in khayāl, providing short audio illustrations from Rāg samay cakra, with accompanying notations.

1.9 Khayāl: Origins

For all that many of its songs invoke Hindu deities and legends, khayāl developed and flourished in and around the Muslim courts of the Mughal era . One problem is that we do not know exactly what the music would have sounded like in the earlier centuries of its evolution. Like the music itself, accounts of khayāl’s history have tended to be transmitted orally, accruing a variety of myths in the process. For a factually grounded account of the genre’s origins it is important to look to reliable scholarship based on consultation of the available manuscript sources.

Significant among such historical treatments is Katherine Butler Schofield’s (née Brown ) revisionist essay ‘The Origins and Early Development of Khayal’ (2010). Brown painstakingly pieces together an account whose authority is based on the evidence of Indo-Persian sources. Interestingly, and perhaps provocatively, it differs in key respects from received wisdom. Taking the earliest known uses of the term khayāl in the manuscript sources as a starting point, Brown surmises that a form of the genre may have first appeared at the Mughal court between 1593 and 1637—significantly later than the life of Amir Khusrau (1253–1325) to whom the invention of khayāl is commonly attributed.

Alternative protagonists in Brown’s account include the Qavvāls of Delhi —Sufi musicians who were regarded as the primary exponents of khayāl throughout the seventeenth century (Brown 2010: 168ff.). They cultivated the raviś (style) of Amir Khusrau, which included the genres of qaul , tarānā and—in some accounts—khayāl. However, Brown argues that khayāl was not a direct inheritance from Amir Khusrau, but was rather part of a legacy in which the Qavvāls synthesised his raviś with other influences. Important among these was a genre known as cutkulā , which was a legacy of Sultan Husain Shah Sharqi of Jaunpur (r. 1458–83).

Significant in the nexus between Delhi and Jaunpur was the figure of Shaikh Bahauddin Barnawi (d. 1628/9), associated with the village of Barnawa, East of Delhi (ibid.: 174–8). A renowned musician, he is known to have composed khayāl. His great-grandfather, Shaikh Pir Buddhan (d. 1498) was the pīr (spiritual master) and fellow connoisseur of Husain Shah Sharqi of Jaunpur (ibid.: 177), and hence may have been a conduit for the transmission of cutkulā —and its subsequent evolution as khayāl—into the lineage of the Barnawa shaikhs. Indeed, Brown conjectures that the circumstances were exactly right for Shaikh Bahauddin Barnawi himself to have been a creator of khayāl (ibid.: 178). And among the Shaikhs’ retinue were qavvāls—a possible line of transmission to the Qavvāls of Delhi .

Crucial to these many interconnections are the religion and culture of Sufism , especially its Chishti Order. This is the common denominator that links many of the protagonists and places in Brown’s account: Amir Khusrau , Husain Shah Sharqi , the Qavvāls, the Barnawa Shaikhs, other khayāl exponents such as Shaikh Sher Muhammed (seventeenth century), and the centres of Delhi and Jaunpur . A possible channel of transmission could have been Sufi samā‘ gatherings—ecstatic devotional assemblies in which khayāl, or generic variants of it, may have been sung (ibid.: 180–2). Importantly, such gatherings were sympathetic to Hindu devotional imagery —a fact which remained salient for the evolution of khayāl in later centuries. There seemed to be no contradiction in Muslims singing songs informed by the Hindu bhakti tradition , in which the pain of separation between lovers (most emblematically Rādhā and Kr̥ṣṇa ) might be implicitly equated with longing for union with the divine (ibid. 186–7).

Brown’s already complex narrative forms part of an even more complex wider picture. Although any extended exploration of the subsequent history of khayāl is beyond the scope of this account, several points are worth briefly highlighting.

First, khayāl, alongside various other forms of Hindustani music, eventually established a place in the secular court mehfil of the Mughal empire in the seventeenth century. While dhrupad was the pre-eminent classical genre (with Tānsen (ca. 1500–89) its most prominent exponent) in the court of Akbar the Great (r. 1556–1605), we know from Faqīrullāh’s Rāg Darpan (Mirror of Music, compiled mid-seventeenth century), that two khayāl singers were listed in the retinue of the Emperor Shah Jahan (r. 1627–58) (Wade 1997 : 1–2), albeit that dhrupad was still in the ascendant at this point. It was by the beginning of the reign of Aurangzeb (r. 1658–1701), that khayāl probably began to rival dhrupad, in popularity if not prestige (Brown 2010: 182–4).

Secondly, while khayāl had acquired the characteristics of a classical form by this time, it had arisen and would continue to develop through cross-fertilisation with other styles and genres, and through a confluence of classical (mārga) and regional (deśī) features. Madhu Trivedi tells of a flourishing of the musical arts in the first half of the eighteenth century under Emperor Muhammad Shah (r. 1719–48), which also included a diversification of patronage. Mehfils were ‘arranged by nobles, affluent people and eminent musicians [and] attracted large audiences. As a result the number of professional artists increased greatly’ (Trivedi 2010: 83). Paradoxically, the crisis of the Mughal empire at this time (which marked the beginning of its disintegration) further fostered this diversification. After the invasion and sacking of Delhi by the Safavid Persian Emperor Nādir Shah (1688–1747) in 1739, court artists were forced to find employment from a wider sphere of patrons, and this was complemented by a cross-pollination of classical and folk (dhun)-based forms (ibid.: 84).

Thirdly, under this changing political and social background—increasingly marked by the British presence in India—khayāl flourished during the eighteenth, nineteenth and early twentieth centuries across various Northern Indian centres of patronage, including Delhi , Gwalior, Lucknow, Benares, Baroda, Rampur , Calcutta and Bombay (Bor and Miner 2010; Wade 1997: 5–10).

Fourthly, there is the question of Ni‘mat Khan , also known by his pen name Sadāraṅg (1670–1748). Ni‘mat Khan was active in the early eighteenth century, and is canonical in histories and orally transmitted stories of khayāl. He was a celebrated player of the bīn (a form of lute), a famed exponent of dhrupad and khayāl, and a renowned teacher. Sulochana Brahaspati (2010: 271–5) describes how he was not only favoured by the Emperor Muhammad Shah , but also established an important musical lineage associated with Rampur , which included his nephew Firoz Khan , known under the pen name Adāraṅg. Trivedi credits Ni‘mat Khan with the ‘renovation’ of khayāl through features still recognisable in today’s versions of the genre (2010: 84). On the other hand, Brown refutes claims that he uniquely popularised or classicised khayāl—though she concedes him a place within the broader landscape of the genre (2010: 189–91). She argues that Ni‘mat Khan’s espousal of khayāl under the patronage of Muhammad Shah may have been related to his need to distance himself from his association with dhrupad , which he practiced under the earlier patronage of the eventually discredited Emperor Jahandar Shah (r. 1712–13) and his concubine Lal Kunwar . If this was the case, the strategy worked: to this day, songs referencing Sadāraṅg (by no means all of which are authenticable) continue to circulate in the khayāl repertory.

1.10 V. N. Bhatkhande

Any student or enthusiast of Hindustani classical music will sooner or later encounter the figure of Vishnu Narayan Bhatkhande, whose name has indeed already surfaced several times in this book. A seminal scholar-musician, Bhatkhande was instrumental in ushering Indian music into the modern era. He organised four All-India Music Conferences held between 1916 and 1925 (see Trasoff 2010). His lifelong efforts to modernise the way Indian music was taught, understood and practiced was of a piece with the momentous social, cultural and historical changes of India’s struggle for independence from the British . He published a substantial body of historical and theoretical writings on Indian music, some in Sanskrit, some in his mother tongue, Marathi (some of which were translated into Hindi ). To this day, his legacy remains pertinent, if contentious, to practitioners and scholars of South Asian music.

Three traces of Bhatkhande’s influence can be discerned in Rāgs Around the Clock. First, his time theory of rāg is of obvious relevance (cf. Section 1.6). Secondly, in our descriptions and specifications of individual rāgs (in Section 2.5, below), we have drawn on his Kramik pustak mālikā (1937)—a six-volume compendium of rāg commentaries and bandiś transcriptions compiled from hereditary musicians of his day (a copy sits on VR’s bookshelf, and is a not-infrequent reference point in our lessons). Thirdly, in our transcription of bandiśes we have adapted Bhatkhande ’s notation method, since this is almost universally recognised by present-day Hindustani musicians.

Typically of many modernisers , Bhatkhande achieved historical significance and influence through a process of othering—strategically differentiating himself from rival individuals, peoples, ideas and tendencies. He asserted a future Indian music against a colonial western present—a future underpinned by an institutionalised learning culture distinct from the hereditary musical lineages of India’s own past. To this end, he and Rai Umanath Bali established an academy for the study of Hindustani music in Lucknow in 1926, originally known as Marris College (see Katz 2017: 109–16). Bhatkhande’s vision of Indian music as a culturally unifying force in a new, independent nation was an ostensibly secular one. In this he differentiated himself from his rival reformer Vishnu Digambar Paluskar (1872–1931), whose own urban music academies—which bear the name Gandharva Mahavidyalaya and remain operative today—were intimately connected to Hindu nationalism and the notion of a return to the imagined Vedic roots of a pre-Muslim era.

Janaki Bakhle discusses Bhatkhande and Paluskar against the wider political background of their projects in her 2005 monograph Two Men and Music. Bakhle ’s account of Bhatkhande is especially polemical, highlighting his antagonism toward the very hereditary Muslim musicians whose musical knowledge he solicited. She underlines how he characterised them as illiterate ; how he held them responsible for a perceived decline of Indian music into degeneracy; and how he regarded them as imperilling their musical legacy by relying solely on oral transmission within their own khāndāns (family lineages). Bakhle ’s evidence includes Bhatkhande’s own journal, which documents his frustration at what he perceived as the shortcomings of musicians such as the sarod player Karamutallah Khan (Bakhle 2005: 109–13). Max Katz , on the other hand, evidences an altogether more cordial and constructive relationship between Bhatkhande and the sarod player Sakhawat Hussain Khan (1877–1955), who was an important figure in the establishment of Marris College (Katz 2017: 116–22). Yet Katz also argues that this relationship was played out against the wider backdrop of a Hindu–Muslim communalist struggle in which the former group was to achieve ascendancy—both within the College and within India as a whole (2017: 100–28).

Bhatkhande’s ultimate goal was to systematise and unify Indian music—to present it as a classical tradition of equal stature to that of the West. His invention of a musical notation system was part of this; and so was his search for a theoretical and historical basis for Indian music that would demonstrate systemic linkages between its various forms (one goal, eventually abandoned, was to unify the Hindustani and Karnatak traditions). Yet, he was unable to find conclusive evidence for such a unified picture. For all his scholasticism, his research eventually led him to believe that Indian music as actually practiced in his own day had only a relatively short history of some two to three hundred years, and that ancient canonical treatises such as the Nāṭyaśāstra of Bharata or the thirteenth-century Saṅgīta-ratnākara of Śārṅgadeva were of limited relevance (Bakhle 2005: 105–6, 114–16). In his attempt to write a systematic theory of his own, his Hindustānī-saṅgīta-paddhati, he placed more value on treatises written between the fifteenth and eighteenth centuries (see Powers 1992: 11), which he believed had a more discernible relationship to modern-day musical practice.

Bhatkhande remains an ambiguous figure—canonised and criticised in equal measure. Authors such as Sobhana Nayar (1989) and Shripada Bandyopadhyaya (1977) are unstinting in their praise of him and unequivocal about his significance. Others, as we have seen, can be searing in their critique. Yet even Bakhle is careful to consider the nuances, ambivalences and contradictions in Bhatkhande’s thinking. His musical theories have been criticised by scholars and musicians on theoretical and practical grounds, yet he remains a key point of reference for many (for example, Jairazbhoy 1971). In a sympathetic appreciation, Sulochana Brahaspati (2010: 278–9) reminds us that in his youth Bhatkhande learned the vīṇā as ‘a disciple of the [Muslim ] maestros of Rampur’, and that the historical texts he found valuable included sources in Urdu and Persian . As the musical and geopolitical debates into which he pitched continue to play out in the present day, we can only continue to hold his formidable achievement and the contradictions in his outlook in tense juxtaposition.

1.11 The Guru-Śiṣyā Paramparā

Introduction

The content of Indian classical music is organically connected to the way it is taught. The subtleties of rāg, svar and śruti; the spontaneity of improvised performance; the oftentimes elliptical relationship between melody and tāl; the emotional flavour (ras) that can be savoured in every phrase and ornament: these elements of the music’s lifeblood flow from its distinctive oral pedagogy founded on the immediacy of face-to-face communication and the bond between student and teacher. This historically longstanding tradition, which has the figure of the guru at its heart, is known as the guru-śiṣyā paramparā.

Let us consider these words in turn. ‘Guru’ means more than simply ‘teacher’: it carries connotations of ‘master’ or ‘preceptor’, and has a history going back to the Vedas, the foundational texts of Hinduism. As Joel Mlecko explains (1982: 34): ‘Gu means “ignorance” and ru, “dispeller.” The guru is a dispeller of ignorance, all kinds of ignorance’. Similarly, śiṣyā means more than merely ‘student’: it implies discipleship and devotion. Finally, paramparā, means ‘lineage’ or ‘succession’: it points to a would-be unbroken transmission of knowledge as śiṣyās themselves eventually become gurus and pass on their skills and wisdom to the next generation.

These Sanskrit terms have their Urdu equivalents in Muslim traditions of learning—a significant point given the prominence of Muslim hereditary lineages in Hindustani classical music. The Muslim counterparts of guru and śiṣyā are ustād and śāgird; and the equivalent of paramparā is silsilā. As James Kippen puts it (2008: 127): ‘what both [sets of terms] have in common, in an ideal sense, is a system where the master becomes the complete role model for the disciple not only in terms of the transmission of musical understanding and the technical means to perform it but also in terms of moral and ethical integrity, self-realization, vision, and personal depth’. What both versions of the tradition also have in common is the place of the student as a member of the master’s household, with its intimate and immersive learning environment. This last feature continues to be regarded as the ideal milieu for the passing-on of musical knowledge and its attendant values, even though, since the twentieth century, it has been challenged by other learning systems that have emerged from it or have sought to displace it.

Which brings us to a further crucial aspect of the guru-śiṣyā paramparā: the impact of modernisation . The tradition’s passage into modernity goes hand in hand with the rise of a Hindu middle class that began to cohere in the nineteenth century (Van der Meer 1980: 122–6), came to prominence contemporaneously with the Indian independence movement (Neuman 1990: 18–21), and achieved dominance in India after independence and Partition in 1947 (ibid.: 142). Voices in recent anglophone musicology (for example, Bakhle 2005, Katz 2017) have critically examined how this also involved an emerging Hindu hegemony over hereditary Muslim musicians. However, complementing this, Justin Scarimbolo (2014) argues for softening ‘polarized understandings’, based on his research into the agency of Brahman musicians ‘beyond nationalism’ during the colonial era.

What remains clear is that, in the earlier twentieth century, the Hindu middle-class sphere became fertile ground for the learning of Indian classical music (and largely remains its habitus in India today). Since that time, musical pedagogy has taken institutional form within specialist music academies or music departments of universities (the earliest modern Indian universities were founded on the colonial model of London University in 1857). Alongside such establishments, gurus and ustāds continue to teach privately, but often along lines closer to western musical tuition—what Regula Burkhardt Qureshi terms the ‘Indian bourgeois version of music lessons’ (2009: 168).

To summarise and elaborate a little: we have so far identified three particularly important moments in the long duration of the guru-śiṣyā paramparā. First, its historical antecedents in the figure of the guru in the Vedic era (dating back to probably the second millennium BCE); second, the significance of hereditary musicians as tradition bearers, who were already a presence in the court culture of the Delhi Sultanate (which arose in the thirteenth century) and further flourished under the Mughal dynasty (from the sixteenth to the nineteenth centuries); third the impact of modernity , with its attendant shift in the nineteenth and twentieth centuries to a middle-class culture of musical pedagogy and consumption. To these three moments we might add a fourth: a problematic side to the paramparā, arguably rooted in its patriarchal underpinnings, that has begun to loom especially large in the twenty-first century.

It is important to stress that these moments do not resolve into a unified, seamless historical chronology or narrative. As Bakhle reminds us (2005: 257–8), ‘no single historical trajectory is adequate to the task of telling the robust history of music’. Nonetheless, these themes, with their overlaps and interpenetrations, will form important elements in the following discussion of the paramparā. My starting point, however, will be what is probably the tradition’s most crucial aspect (also mentioned above): the fact that the student has historically acquired their musical training through living in (or near) their teacher’s household, in an environment where learning becomes inseparable from daily living. If this account takes more extended form than the other essays so far in this book, this is largely because of the sheer significance of the paramparā to Indian classical music, and because its social, historical and cultural features throw considerable light on the practice as a whole.

The Tropes of the Tradition

What are the sources of our knowledge of the guru-śiṣyā paramparā? Like so much else in Indian culture, much of what we know about the tradition has been orally transmitted within it, through stories and anecdotes. These have assumed a status close to myth or lore. Such narrative themes are often termed tropes; and if they project a highly idealised picture of the relationship between master and disciple that is not always realised in practice, this seems essential to sustaining and reproducing the ethos and ethic of the learning culture (for does not every musician speak devotedly of their ustād or guruji?).

Another source of our knowledge comes from scholarly accounts by ethnomusicologists who themselves have experienced musical discipleship as part of their fieldwork . These accounts include writings by Daniel Neuman (1990: 43–58) who learnt sāraṅgī with Ustād Sabri Khan (1927–2015); Qureshi (2009) who, some years later, learnt with the same ustād; Kippen (1988; 2008), who studied tabla with Ustād Afaq Hussain (1930–90); Magriel (2001), also a disciple of Ustād Sabri Khan; and Wim van der Meer (1980: 138–50), who studied voice with Pandit Dilip Chandra Vedi (1901–92). Our knowledge of the guru-śiṣyā paramparā is also informed by historians’ investigations of the guru concept in ancient Vedic sources. Further insight is afforded by musicological genealogies of hereditary musicians, and by related accounts of the gharānās (stylistic schools) of the later nineteenth and earlier twentieth centuries (see for example, Das Sharma [Dasasarma] 1993; Neuman 1990; Wade 1997).

Salient within these sources is the idea that the traditional locus of musical instruction is the guru’s own home. This point is underlined by, among others, Qureshi (2009), who describes first-hand how pedagogical transmission is predominantly oral, and how the distinctive feature of discipleship under a hereditary musician is that the śagīrd or śiṣyā live with or near their ustād or guru as an actual or honorary family member. Qureshi recounts how the Urdu phrase ‘sīna ba sīna’, which loosely means ‘from heart to heart’, was interpreted by her ustād, Sabri Khan , as meaning in effect ‘from father to son’—thus stressing the intimate, familial nature of this culture (2009: 167–71). Familial might also mean quasi-familial—in that non-family members can also be admitted to the teacher’s circle and treated as a family member (as in Qureshi’s own case)—or extended familial—in that teaching may also be from grandparent to grandchild, uncle to nephew etc. This domain extends out into a wider kinship network or community, known as birādarī or ‘brotherhood’ (ibid.: 170).

In this familial setting—often known as a gurukul (meaning, roughly, domain or clan of the guru)—teaching, learning, practice (riyāz), talk about music, and the goings-on of everyday life blend seamlessly (Neuman 1990: 54–5). This is a particularly congenial environment for younger family members to learn in. Magriel (2001: 104–40) observed how the young sons and nephews of several hereditary ustāds were inducted into learning music almost by osmosis, since music making was going on around them all the time as extended family and community flowed in and out of the house. Learning for these youngsters was in the first instance informal, even playful, and gradually became more formalised and focused as they got older and acquired competence. The ustāds’ aspiration was for this younger generation to become professional artists themselves, and so pass on the tradition and continue the family lineage (assuming the potential was there and the opportunities for a career looked favourable). Such hereditary musicians are known as khāndānī (Neuman 1990: 58).

The admission of non-family members into discipleship is one way that Hindu musicians have learnt from Muslim teachers and vice-versa; it has also enabled the diversification of traditions, and, in the more recent past, made it possible for western musicians to study with South Asian masters. Nonetheless, entry by non-family members into a guru’s or ustād’s domain has historically been difficult to negotiate. Another trope among musicians’ stories is of having to demonstrate the seriousness of their intent to a potential guru or ustād, who might in the first instance appear indifferent or even discouraging. Van der Meer provides examples of such stories, some of them legendary (1980: 144–8); Kippen’s account of his own experience of gaining admission as a disciple of Ustād Afaq Hussain is also paradigmatic (2008: 125–6, 128–9); and VR himself recounts how, after initially approaching Pandit Bhimsen Joshi in person to take him on as his śiṣyā, it took another two years before the master finally confirmed he would teach him, during which time VR would repeatedly turn up to Joshi’s concerts in Delhi , sometimes playing tānpurā for him, understanding that the whole process was a test of his patience and commitment (Rajput 2012).

Once admitted into the guru’s household, a student’s devotion might continue to be tested before they are formally accepted and begin tālīm (tuition). A further trope among musicians’ stories is of being expected to perform menial chores for their guru, such as doing his shopping, running errands, and generally making his life more comfortable—all of which could be seen as a demonstration of their submission (Kippen 2008: 129; Neuman 1990: §55). During this probationary phase, the student typically sits in on other students’ lessons, learning by immersion, rather than being directly taught.

Eventually, if convinced by the student’s commitment and general good character, the master may formally take them on as a śiṣyā or śāgird. This rite of passage is traditionally marked by a ceremony known as gaṇḍā bandhan, at which a thread is tied around the wrists of master and disciple, symbolising the bond between them. Kippen describes this ceremony as a mixture of Muslim and Hindu ritual (2008: 130); Qureshi (2009: 171–4) references the ceremony by its Muslim name—śāgirdī—in an account of sāraṅgī player Nasir Khan’s admission into discipleship under his uncle, Ustād Sabri Khan . Such ceremonies are undergone by hereditary and non-hereditary musicians alike; indeed, the distinction between the categories would seem to be erased from this point. The śiṣyā commits to a life devoted to musical learning and its attendant values, and the guru commits to the śiṣyā as if their own offspring, and to passing the tradition on to them. If not already a family member, the śiṣyā is henceforward treated as such, and refers to fellow students as their guru-bhāī or guru-bahan (‘guru brother’ or ‘guru sister’). There is a strongly affective dimension in all these relationships: another trope has it that the guru should love his student like (or more than) he loves his own son, and the śiṣyā love his guru like (or more than) he loves his own father (Neuman 1990: 45–50).

At the gaṇḍā bandhan ceremony , the śiṣyā customarily makes a financial offering known as a guru dakśinā; and although further offerings may also be subsequently made (even expected), notionally these would not be regarded as direct payment for tālīm which in principle is given freely (Neuman 1990: 51–2). ‘Guruji never asked for payment’, confirms VR; in other accounts of the guru dakśinā, the śiṣyā simply pays what they can afford. What the guru has to offer is regarded as being beyond price (as it were, outside of any commodity exchange); in return the śiṣyā is expected to adapt his own behaviour, showing respect and obedience. Even after initiation, a śiṣyā continues to undertake tasks and favours (such as carrying the guru’s instrument, bringing his food to the table) as acts of respect and gratitude (Neuman 1990: 46, 51). Reverence is also conventionally shown in the ritual greeting (praṇām) of touching a guru’s feet.

Perspectives from History

The historical depth of the paramparā—and of lineages within it—matters. There is status for a musician in hailing from a long and distinguished pedigree. Lines of descent from father to son (patrilinear) or from blood relative to blood relative (consanguineal) are known as khāndān, and its members as khāndānī; wider family groups connected by actual or potential marriage ties are known as birādarī (Neuman 1990: 95–9; Qureshi 2009: 170–1). Scholars seeking to compile genealogies must contend with sometimes misremembered or conflicting accounts (where brothers may have been confused with father and son or vice-versa (Neuman 1990: 166)) and sometimes tenuous claims by musicians to have belonged to the khāndān or birādarī of a particularly eminent ustād in order to enhance their own prestige. Even if some might be stretching a point to claim themselves khāndanī of the legendary sixteenth-century dhrupad singer Tānsen , present-day continuities with the four singing styles (bānīs) dating back to Akbar’s court, and with instrumental lineages putatively originated by the sons and daughters of Tānsen, mean that this is at least imaginable (ibid.: 147–8, 164–5). Neuman asserts that the likely limit point of any hereditary musician’s plausible ancestry would be the Sufi musician Amir Khusrau who was a famed singer in the era of the Delhi Sultanate. Before and alongside this, one would need to look to pedagogical lineages of Hindu temple musicians, which were not necessarily hereditary (ibid.: 85, 104, 105). For that matter, not all lineages associated with courts were Muslim —for example, Tānsen and his teacher, the equally legendary Swami Haridas (1480–1573), were both Brahmans (though the former is said to have become a Muslim when he entered Akbar’s court).

A further aspect of lineage is the institution of the gharānā, which was ‘conceived in the mid-nineteenth century and born in the twentieth’, as Neuman puts it (1990: 146). In seeking to distinguish a gharānā from a khāndān (beyond the fact that the former concept has a more recent historical provenance), we might say that gharānā more strongly implies the idea of a stylistic school. Gharānās are often named after their place of origin (for example, Gwalior, Agra, Delhi , Kirānā) or their founder or most eminent pioneer (for example, sitarist Imdad Khan). Gharānās may be predominantly vocal or instrumental, or reputed for both. Each gharānā is distinguished from the others by particular forms of performance style which have evolved across successive generations (Deshpande 1987; Wade 1997). For example, while the vocal strand of the Kirānā gharānā, with which VR is affiliated, is known for its purity of svar and for giving less priority to lay (as can be heard from historic recordings of its founder, Abdul Karim Khan ), the reverse is true for the Agra gharānā (Deshpande 1987: 41–5).

Although lines of transmission between generations of a gharānā may be hereditary , this is not universally the case. For example, tracing back a line of pedagogy back from VR within the Kirānā gharānā, we have Pandit Bhimsen Joshi , Pandit Sawai Gandharva (1886–1952) and Ustād Abdul Karim Khan (1872–1937), none of whom are consanguineal. For a gharānā to be recognised as such, it must customarily have at least three generations; it is not enough for a teacher and his students to declare themselves a gharānā. Despite their often geographic titles, gharānās do not necessarily designate physical or geographic communities. Membership is generally consolidated through identification with a performance style, and some artists may identify with more than one gharānā (for example, Bhimsen Joshi’s singing style also drew influence from musicians of other gharānās (Wade 1997: 194)). Conversely, although the gharānā concept did not arise until the nineteenth century, once established, gharānās tended retrospectively to reconstruct their roots back to earlier generations. At stake here was what Neuman terms ‘the politics of pedigree’—the need for agency and influence, which became particularly acute at a time when the princely courts that supported professional hereditary musicians were in decline (Neuman 1990: 146–7; 145–67).

As might be surmised, paternalism and patriarchy are notable features of the guru-śiṣyā paramparā; in recounting its social structure, it is difficult not to reproduce the implicit tendency to represent the actors as male. So it is important to stress that women have also had a longstanding place in Indian classical music (Post 2000), albeit not outside the relations of patriarchy . As Neuman explains and documents (1990: 97–9; 248–53), wives and mothers had a role in preserving and expanding the hermetic lineages of khāndans. Women also historically constituted their own class of hereditary musicians (ibid.: 100–2), as temple dancers and court musicians (devadāsīs , tawaīfs )—professional functions that overlapped with that of courtesan. In the temple and courtly milieux of their day, such women were highly regarded for their cultural knowledge and artistic prowess. However, they became vilified under colonial Victorian moralising attitudes in the nineteenth century, and under related social reforms in the early twentieth that included the banning of temple dancers (Post 2000). Women subsequently regained respectability within Indian classical music largely through middle-class (and predominantly Hindu) teaching academies such as Paluskar’s Gandharva Mahavidyalaya (Bakhle 2005), which, significantly, allowed non-hereditary aspiring musicians access to tuition. Hereditary female performers nonetheless maintained a place as professional musicians, though needed to take steps to sanitise their former courtesan associations. Learning with respected gurus and ustāds was one way to do this, which gave them affiliation with gharānās (Neuman 1990: 100–1; 207–8). The suffix bai—which can be intended honorifically—is not uncommonly added to the first names of professional female performers (ibid.: 100).

Qureshi (2002) sees the endogenous culture of hereditary musicians as reproducing the feudal relationships in which they were employed. When their princely or landowner patrons could dismiss them on a whim, it was understandable that musicians would keep the only commodity they had—their artistic skills and knowledge—within the family. Such a feudal mentality partly explains the patriarchal authority one can still find operating in the ustād-śāgird/guru-śiṣyā culture of today—where the guru or ustād tends to be regarded as a figure of absolute authority.

An even longer historical view reveals the guru as a venerated, near-deified religious figure within the deep past of Hinduism. According to Mlecko (1982: 34), ‘[i]n his earliest role the guru was a teacher of the Vedas and the various skills needed for their study, such as grammar , metrics, etymology and mnemonics’. Mlecko stresses the importance of orality in the transmission of this knowledge in which ‘proper accent and pronunciation’ were paramount (ibid.); and it was the milieu of the gurukul that afforded the close personal contact necessary to this learning process. In other words, the institutions of both guru and gurukul, as we have continued to see them in the pedagogy of Indian classical music, were already operative at a very early historical stage—in what Mlecko loosely terms ‘ancient India’, probably meaning the first millennium BCE or earlier. Indeed Mlecko evidences the importance attributed to the guru within the scriptures themselves—for example in the R̥g Veda, Yajur Veda and several of the Upaniṣads, dating variously from the second millennium BCE to the early centuries BCE. The writers of these texts often impute divinity to the guru, some depicting him as a figure from whom even the gods learn. Jan Gonda’s account (1965: 229–83) similarly evidences the exalted status of the guru as well as the antiquity of the guru concept, which he argues as being co-extensive with Hinduism . Mlecko likewise illustrates the continuity of these ideas through the Hindu epics and tantrism . He goes on to demonstrate their later historical evolution in the medieval bhakti tradition and beyond (1982: 47–52), in which the guru becomes emancipated from his priestly or Brahminical origins and becomes a figure within popular devotional movements (see also Sooklal 2010).

The Impact of Modernity

The guru-śiṣyā paramparā has continued to evolve under late modernity. Since the twentieth century, the paramparā has had to find a way to survive without the patronage of princely courts and wealthy zamīndārs (hereditary landowners)—with the concert hall largely taking the place of the mehfil (courtly gathering), the academy functioning as an alternative site of pedagogy to the gurukul, and liberal-democratic social relationships beginning to impinge on the feudally -sanctioned authoritarianism of the guru.

In modern times many gurus or ustāds still teach from their own home, but students are more likely to be visitors than live-in family members, taking their lesson perhaps weekly rather than daily, and commonly availing themselves of smartphone technology to record their class as a way of maintaining their teacher’s virtual presence between lessons. Alternatively, or additionally, a guru might be employed by one or more educational institutions. Taking the residency of bansurī maestro Pandit Hariprasad Chaurasia at the Rotterdam Conservatory as a case study, Huib Schippers (2007) analyses how such artists manage traditional guru-śiṣyā teaching styles against the regulative processes of qualification-awarding institutions, as well as balancing their students’ needs against the itinerant lifestyle of a performer on the international stage. The profile of many jobbing modern gurus takes the form of a freelance portfolio career in which they operate as self-employed individuals juggling performing and private and institutional teaching commitments across various locations.

The physical gurukul has also taken new forms. Some teachers rent alternative premises—in effect an annexe to their home—from which to teach students locally. Others have developed purpose-built gurukuls complete with residential accommodation for national and international students. One example would be the Gundecha brothers’ Gurukul Dhrupad Sansthan in Bhopal, where, even though the scale is large, the core pedagogical principles remain daily face-to-face contact with the gurus and the sustaining of a community of śiṣyās who predominantly live together (Sankaran 2020). The Kolkata-based ITC Sangeet Research Academy is another example, with its more formal institutional infrastructure, research library and associated scholars; notwithstanding all this, the Academy’s key objective remains to cultivate performers in the guru-śiṣyā paramparā (Kashalkar 2013).

Such modern gurukuls , which resist the idea of a formal curriculum or syllabus, remain distinct from Indian university music departments and music schools which offer tuition for a formal award such as a degree. As Andrew Alter summarises it (paraphrasing Banerjee 1986), tuition in these latter institutions is likely to be syllabus-based, involve a variety of teachers rather than a single guru, be delivered to groups of students rather than individuals, and emphasise musical literacy and academic training alongside practical tuition (Alter 2000: 447–8). Yet, even in the face of more institutionalised and externally regulated models, such as the western-influenced, curriculum-based, academy, the guru-śiṣyā paramparā continues to be acclaimed by performers as the essential pathway to learning Indian classical music, with the guru or ustād as its lynchpin.

Problematics of the Paramparā

The image of the guru conveyed in most narratives about the paramparā is an idealised one. Sanyukta Kashalkar (2013: 83) puts it thus (in a statement mirrored on a now-superseded version of the ITC Sangeet Research Academy website): ‘To the Shishya, the Guru symbolises the art itself, while for the Guru, the Shishya signifies the continuity of the art. The Guru shares the sacred knowledge of the art only with kindred souls, sincere in their quest’. Given human nature, however, any individual guru or ustād may map more or less perfectly onto the quasi-divine ideal they inherit; and a properly critical assessment of the guru-śiṣyā paramparā needs also to address its shadow side.

Problems, perhaps as longstanding as the tradition itself, have not gone unacknowledged. Mlecko’s historical long view includes the statement that ‘[g]urus can be completely selfless, desiring nothing for themselves or they can be avaricious, seeking only an easy livelihood off the naive or guilt-ridden—they use the śiṣyās’ (1982: 55). Kippen too gives an account of ‘the problems of exploitation and manipulation’ by musical ustāds (2008: 134–7), including a commonly referenced complaint: the guru who teaches ‘with a closed fist’—who, in other words, is parsimonious with the knowledge he passes on to his students (see also Slawek 2000: 462–3). The field also has its stories of gurus no less prone than other human beings to addiction and desire; stories of sexual misconduct, harassment and abuse have also come to light.

One aspect of the tradition’s grating up against modernity is the way these behaviours are, in the earlier twenty-first century, being publicly called out, especially via social media; the guru-śiṣyā paramparā is now also experiencing its #MeToo moment. At the time of writing, this tendency has reached the point where some gurus are themselves calling for reform. Karnatak vocalist and writer T. M. Krishna has been at the spearhead: ‘Let me say it as it is’, he writes, ‘[t]he parampara is … structurally flawed’ (Krishna 2020: n.p.). He continues:

we need to reimagine our structures of learning. The system must begin with respect for students, and recognition of their independence and rights as individuals. This is vital because the power structure is naturally tilted in favour of the guru. But for this to happen, we need to first ‘humanise’ gurus. The parampara that demands obedience and unquestioning deference, only because someone is a guru, needs to be demolished. Simply put, gurus must be respected for being domain experts—nothing more.

Krishna holds fire—at least in this article—on what this reimagining might be, while insisting he is not arguing for the institutionalisation of teaching. The issue is perhaps whether what is distinctive and valuable about the guru-śiṣyā paramparā can be transmuted into something compatible with liberal-democratic values . For the personal, affective and relational aspects of the guru-śiṣyā paramparā at its best still represent something valuable and humanising. Indeed, against the increasing regulation and corporatisation of learning institutions in countries such as the UK (and their growing culture of transactionalism between students and teachers) the values of the paramparā might, paradoxically, suggest a more salutary critical alternative.

Krishna is surely right to argue for re-constructing gurus as nothing more (or less) than human beings, and for unburdening them of the status of gods—a status which risks the corresponding infantilisation of the śiṣyā. Yet in order not to lose what is heart-centred in the tradition, perhaps we might still aspire for gurus and ustāds ethically to function as something beyond domain experts. Kashalkar chooses her words well in the above quotation when she says the guru symbolises the art itself and that its continuity is signified by the śiṣyā. For this potentially defends both parties from the mistake of attributing these qualities to them a priori—from conflating the individuals with the art they hold in trust. In reality both (human) parties might fall short of the ideal, but arguably they are less likely to do so if they recognise that neither is bigger than the tradition they uphold.

1.12 Riyāz

Like musical practitioners everywhere, Indian classical musicians practise for a purpose—indeed several purposes. They practise to improve: to increase their technical mastery of their voice or instrument. They practise to acquire knowledge: to internalise a repertoire of rāgs, tāls, compositions and more. They practise to learn how to perform: how to create and extemporise musical materials, and how to form them according to culturally recognised conventions of style and syntax . Above all, they practise in order ultimately to transcend these technical and material things: to reach for, and occasionally touch, moments of the intangible. This last purpose also relates to a non-purposive aspect of practise which is captured in the Urdu /Persian word by which Hindustani musicians know it: riyāz.

Riyāz construes practising as a devotional or spiritual act. It is often associated with another, Sanskrit term (familiar to yoga practitioners): sādhana. This refers to any daily practice aimed at attaining freedom from the ego. Despite this intent, the paradox is that sādhana ‘should be undertaken without any specific goal in mind. [It] should be practiced for the sake of maintaining the practice, and as a means of cultivating discipline’ (Yogapedia 2023). In the same vein, a musician’s riyāz is also often undertaken as an end in itself rather than a means to an end—such as becoming a professional musician. Riyāz forms the very heart of musicians’ musical lives and identities. They might routinely enquire of one another not when their last performance was, but how their riyāz is going.

In this sense, riyāz occupies the same idealised value-sphere as the relationship between guru and śiṣyā. Just as the guru ideally acts as spiritual preceptor, which makes him more than an everyday teacher; and just as the śiṣyā ideally approaches their learning with an attitude of discipleship, which makes them more than an everyday student; so riyāz is ideally undertaken—by both guru and śiṣyā—with mindful devotion, which makes it more than everyday practising.

And just as musicians’ received wisdom about the guru-śiṣyā paramparā is transmitted through various narrative tropes, so too their understanding of riyāz is passed on through tales that have assumed the status of myth or lore. These principally concern (i) prodigious feats of riyāz (often told in respect of renowned pandits or ustāds ); (ii) the demonstration of bodily signs of riyāz; and (iii) the importance of formidable levels of repetition in doing one’s riyaz. The following tale by eminent khayāl singer Ajoy Chakrabarty (2002: 32) is entirely typical:

My first guru, my father Sri Ajit Kumar Chakrabarty , used to have me practice one song or a single taan for up to eight hours at a stretch for several days. … [He] would never let me stop until my practice was complete. If mother intervened at this time, he used to become angry and say, ‘It is better to be childless than to have a worthless son’.

Daniel Neuman (1990: 31–43) recounts similar tales of musicians practising between eight and sixteen hours a day during the formative years of their training. He also tells of fellow sāraṅgī players inspecting the callouses on his cuticles as physical evidence of his riyāz: ‘The practiced eye could gauge very accurately how much practice had been accomplished and from it … the degree of dedication’ (2002: 32). Comparable stories in respect of the sitar and sarod are told by Gerry Farrell (1986: 271).

While not every musician is in a position (or has the disposition) to practise the whole day or night long, the most significant principle of riyāz, regardless of the actual hours applied, is repetition . Commonly, Indian classical musicians spend a significant portion of their riyāz practising scales, alaṅkārs (ascending and descending sequential patterns), palṭās (similar, more developed patterns), merukhaṇḍ (permutation of a set of notes), pakaḍs (catch phrases that capture the gist of a rāg) and tāns (running figurations). Sorrell provides a more detailed account of such practise techniques and how these eventually translate into performance (Sorrell and Narayan 1980: 67–91). Farrell notes that such routine exercises are practised at all stages of learning, by beginner and maestro alike, the only real difference being the speed of execution and level of sophistication (1986: 269).

A vital point is that in any riyāz session, one should not practise many items a few times, but one of them very many times. Again, this principle is enshrined in musicians’ anecdotes. Chakrabarty writes: ‘As I grew up, I became habituated to practising the same taans 500–600 times throughout the night, understanding my father’s and my gurus’ demands and expectations’ (2002: 32). Farrell discusses a palṭā ‘which one of my teachers told me to play one thousand times (and he wasn’t speaking figuratively!)’ (1986: 268). As well as being the principal means of absorbing musical material, repetition (done mindfully) is regarded as an essential route to ironing out one’s defects—if necessary to the point of exhaustion: ‘you can take a small break if you start dying’, vocalist Veena Sahasrabuddhe (1948–2016) is reported to have said (Phansalkar 2017: 48). Some gurus commend the use of a mālā (prayer beads) to count off repetitions (an aid more practicable for vocalists, who have their hands free, than for instrumentalists): a mālā usually has 109 beads, and is conventionally used when chanting mantras—significantly, a spiritual activity based on repetition.

Insistence on apparently pathological levels of repetition may seem like a means for a guru or ustād to assert their authority over a student, but essentially it is a way to enculturate a foundational pedagogical principle: a discipline; an almost meditative approach to internalising material at a beyond-conscious level. This is important because in improvised performance a musician needs automatically to reproduce and re-permutate musical ideas into new transformations without having to think (for there is no time for conscious thought when one is on the spot). Dard Neuman coins the term ‘automaticity’ for this process (2012: 438, 447 n. 3, 448 n. 16). He recounts how hereditary ustāds would seek to impede their students from thinking in their practising by withholding any technical or theoretical information about a rāg, pakaḍ, palṭā etc., even declining to use sargam note names, until the student had mastered the material in question (ibid.: 432–6). The aim was to achieve an essentially embodied practice—a ‘body-instrument’ in which the throat (in the case of vocalists) or hands (in the case of instrumentalists) ‘think’ for themselves (ibid.: 438–42, 445–6).

While not all Indian classical musicians aspire to become professionals, and while many undertake their practice and practising alongside other jobs or activities as part of busy, complex lives, a commitment to riyāz—and to repetition as its key technique—remains essential. The hard-won truth of the matter is that any musician is only as good as the quality and rigour of their riyāz. Only by this route do they ultimately attain the creative freedom in performance that can touch an audience ’s heart.

Kaṇ

Mīṇḍ

Āndolan

Gamak

Kaṭhkā and Murkī

Cultural and Aesthetic Significance

Introduction

The Tropes of the Tradition

Perspectives from History

The Impact of Modernity

Problematics of the Paramparā