The influence of the late Ustad Amir Khan on India's classical music ls spreading like rippks In a pond, such tht hardly a mustcbn In the Hlodustanl North can be unaware of wht this kAyal master did. He even stands as a muslclan's muslctan In the Karnatak South, an accolade that counts perhaps more than hts numerous national honors.
Those who have had the fortune of hearing Khansahab'sl music never forget It; and there are several recordings of performances which convey much of the experience. It {s easy to lox track of time while hearlug Amir Khan's kAy'I sinsing. In a reverie, a listener may fall to notice that Improvisattonal forays treating merely three or four notes had occupted twenty minutes, or mord
How does he achieve this7 His ux of maiestically kngthened tala-cycles and other elements of a grand design divert us from considerlog Amir Khan as miniaturist. But It {s precisely to the details that one must look in order to gain a better understanding of his festooned khyal-singing, and therewith to speculate on the notlons of Innovation and Improvisation from an Indlc vlewpolnt. For this, a rewarding avenue Is to Inqutre Into Amir Khnsahb's reconciltatlon of rapa (a raga's formal demands) with mirkhard (permutational patterns).2
The history, chief musical devices, and Gharana styles of kbyal are treated by Wade (1984) and Neuman (1980). Wade has also given good stylistic sketches of stngers, tncluding Amir Khan (Ibtd:266ff). But few {nstances can be found which probe the details of an Individual kAyal singer's style extensively. My two7ear discipleship with Amir Khansahab in Calcutta and Bombay ts a untque vantage from which to assay this. It would be revealing to examine his treatment of time, his sargam (tmprovised solfegel tans (rapid passages on vowelsl or compositional endeavors each of thex dexrves its owm Investigation. Here, however, I shall focus on hts {mprovisatlon in Ihe lower reglons of two ragas, Abhogi Kinai. and Kedar.
A two-hour khyal recital might consist of two mapr ragas, each rendered first in a slow or v/lambit tala cycle of 12 or 14 beats.3 During thex opening xcttons, the btstar (or viStar, "development"
Ross Amir Khan 99
i43565i-i765'454-21
A slightly expanded chalan might
be
1 4 3 5 -'4 5 6 b7 6 5 i - i 7 6 5'4 5 4 - 2 1
Ambivalence stemming from the use of both fourth degrees is helghtened in m7nri (slow slides) between the two notes. Very slow mind occurs around both 4s and 5. Phrases may dip below to 2 or upwards briefly to 6, but much time can be devoted to permutations of 4, '4, and 5 alone in the course of mind-laden phrases. Some will aver that '4 may not pass directly to 4, but must touch 5 first, although this is not evident in Amir Khan's performance of the raga.
Extensive sliding also occurs from 4 to 2, as long as one does not rest on 2 (sounding too much like ragas Megh or Chhaya) but resolves it as 4 - - 2 1 2 - - 1. In sliding from 4 to 2 (or in the phrase '4 5 4 - - 2), one should touch 4 with emphasts to avoid sounding like Chhaya, which shows ga, degree 3, a little more prominently during the same descending gliss. The phrase 4 2 5 - smacks of the Malhar riga family and should be avoided, as should 1 2 4 5, which evokes Ba rang. Komal (flatted) ni, degree 7, shows up in figurations such as I (7) 6 - b7 6 - 5 and 5 6 - b7 6 - 5. It is also possible to discern a hint of
komal ga (lower degree 3), in ornament but not in solfege, thus: 4 2 -(b3) 1.
The lore of Kedir's riga-rapa (and the 11st could go on!) is thus considerably more extensive than that of Abhogl, and its improvisatory scope is accordingly more regulated. But as in any art, mastery is required to discover the freedoms within these confines. In the following passage of Kedar bistar, Amir Khan spends some four minutes shifting a balance from degree 2 to 4 (re to ma) 7
Astan Music, Spring/Summer
1993
akin to the free-rhythm alap) showcases the sinEer's learning and Improvisatory mettle. One can also say that It explores both a riga's highilghts and hidden corners durlng a purney through {ts gamut. Amir Khan's bistar plumbed untold depths In ragas through an especially penetrating concentration on their emphasized notes and characlerlsttc phrases. Passages In a raga's unfolding which, in the hands of another muslclan, would be cursorily alluded to were given a new scrutiny: permuting thelr xquentlal possibilltles. focusing on one note for some time while neglecting a nelghbor, and so on. Thus the bistar of a "large" raga like Darban could easily exceed an hour's time--no easy matter, tf one wlshes to hold an audlence's attentlon wlth such slow-moving exposition. A commentator notes some traits of Khansahab's muslc:
Amir Khan's vocalism was marked by classlcal purity. It revealed a leisurely pace and a meticulous unfolding of the muslcal mode. He never displayed virtuosity for Its own sake but concentrated on the spirit of his raga. The qulet, unhurrled progresslon ... showed his fine sense of the subtle and the beautiful. What finally stood out were the architectonics of a master bulider.4
By 'architectonics" one first understands the structure of A bistar improvisatlon taken as a whole. But this consists In turn of nested, smaller-scale structures statements of bistar occupying one or more tala-cycles; wlthin each statement, "gestures" (like paragraphs) bullt from a number of phrases; and note-patterns wlthin each phrax. It is to this last microcosmic structure-level that we shall 'zoom In" for a closer look.
Perhaps the prlme generator of such note-patterns is a raga's r~pa, its "shape' or canon of formal characterlstics. The terxst way to evoke the rtipa Is by the ascent and descent, the aroha-avaroha Thus Xbhog'Kinada, a five-note raga, writ small ts
A silghtly expanded chalan ('way of moving=) might be
1 b3 2 b3 4 6 1 - 1 6 4 b3 2 - b3 4 - 2 -1
Re, the second degree, can be omittett In the ascent: I b3 4 6 1. Along wlth Ihese Introductory ".uidellnes for Abhogl, a teacher will hasten to underline the Importance given to ma and sa, degrees 4 and I (ma Is stresxd further by tuning the tJmpurJ drone 4 - I - I -1, Instead of the Usual 5 - I - I - 1). The third degree of Abhogl, p, sometimes has andolan, a slowly undubting ornament. This oscilblion on ga, of len appearlng in the catch-phrase b,ga ma - re - sa (b3 4 -2 - 1), places {t in the K.nada raga group hence the name Abho.it Kanada.
The latter are the primary givens which honor the raea-repa of Abhog1 when consclentlously carrled out. One notes ihem In a passage performed by the brothers Salamat and Nazakat All Khan.s
The range covered tn this excerpt, lasting a minute and a ha is large: an octave from low ma (degree 4) to middle ma. Nazakat Kha expands in both directions from the central sa tonic, in note-value! ranging from rather short to extremely quick. Above sa, degree receives some attention (6 2 - 1), and one hears b3 in the Kanide phrase b3 4 2 - I (statement Cl).
Nazakat Khan tends to sing short sustained-note phrase: followed by quick-value longer phrases, covering a large ambitus. Th brothers' approach to presenting the rsga is to paint a broad pictur of It (here, the octave from low to middle ma), and then to go over th, same area several times in detail. One senses that the statement divided by mukhras (a fragment of the composition's beginning used a a point of return), could be offered in any order without ill effect. Th focus shifts rapidly from dha, to ma, to sa, to ga and re, yielding presentation of all the raga's facets which gradually drifts upward.
The slow oscillation on the third degree, the emphasis on 4 an' 1, the b3 4 2 I Kanada catch-phrase all these are idiosyncrasies tha give Abhogi Kinada its irrational charm. Each riga has a discret explicit canon or raga-rJpa of this sort, and some are simple, whil others are complex. But outside of any raga's canon lie practice techniques which are relentlessly rational, and therefore are bound e times to be at odds with the demands of the r(ipa. In the styl advanced by Amir Khan, perhaps the most stressed technique I m7rkhar~d. As a pedagogicai exercise, it exhausts, in systemat fashion, all sequential combinations of a number of contiguous note Three-note and four-note arrays are deemed, because of thei. relative brevity, the most practical. The procedure for three note has, as its first set (a), degrees 1, 2, and 3 (sa, re, ga):
The phrases are in pairs which end in successwely lower notes. The exercise then applies the same pattern to higher arrays of the seven possible notes:
(b) 234 (c) 345 (d) 456 (e) 567 (f) 671 (B) 7 i 2
324 435 546 657 761 172
243 354 465 576 6i7 721
423 534 645 756 i67 271
342 453 564 675 716 127
432 543 654 765 176 2l7
With four notes, a similar pattern of successwely lower end-notes is seen (and it can be heardl as in the array using degrees 1,2,3, and 4 (sa, re, ga, ma):
(a) 1234
(b) 1243 (c) 1342 (d) 2341
2143 3142 3241
1423 1432 2431
4123 4132 4231
1432
4132
3412
4312
3421
4321
The exercise then applies the same permutations up one scale degree
(a) 2345 (b) 2354 (c) 2453 (d) 3452
3245 3254 4253
2435 2534 2543
4235 5234 5243
3425 3524
4325 5324
4523
4523
4352
3542
5342
4532
5432
One would then put the degrees 3 4 5 6 through the same procedure, reminiscent of English change-ringing or bell-tone permutation. All possible combinations of four contiguous notes from seven scale degrees generate 168 (7 x 24) patterns. Adding a note to the array increases the combinations factorlally: although generally considered tmpractical, the same process produces 840 combinations of five notes (the renowned sitarist Nikhil 8aneriee, himself an ardent admirer of Amir Khansahab, was said to be a master of the five-note set).
Besides having its own elegant rationality, the organization of the patterns stresses ways by which to arrive at notes, and this can be especially useful if the note is a strong one In the rsga. A student is encouraged to practice mirkhand, in slow phrases bearing in mind the raga at hand, as well as in quick, percussive renditions. Mastery of mirkhand promotes agility and precision; it lends the musician control over the most convoluted of sargam (solfege) passages, and is useful for tan melismas if practiced using a vowel instead of solfege. Especially if they aren't too laden with rules, major seven-, six-, or five-tone ragas lend themselves easily to mirkhand for daily practice.
However, the student receives the caveat6 that mirkhand are to be cultivated at home but never implemented literally in public performance. One should learn mirkhand and then "forget" them. An obvious reason for this is that too much order, unmodulated, yields a machine-like boredom. Moreover, the process discovers some notesequences which may sit uneasily with a given raga's rules. Mirkhand applied to the first four notes of Abhogl would produce
1 2b3 4
2 1 b3 4
1 b3 2 4
b3 1 24
2b3 1 4
b3 2 1 4,
of which 2 I b3 4 and b3 2 1 4 are viable phrases in Abhogi. Phrases I b3 2 4 and b3 1 2 4 depart from the rcipa by omitting degree 3, ga in ascent One might counterbalance the unidiomatjc b3 - I of 2 b3 1 4 by surroundjng U with unequivocal Abhogi phrases. Thus, the b3 _ 4
2 - I Ksnads phrase, and the resemblance to other ragas jf notes such as ga are skipped, makes choices suggested by mirkhar~d somewhat limited jn this four-note region of Abhogi Kansda.
If the procedure is applied to a less cjrcumscribed array of Abhogi-notes, more phrases devolve which are legitimate
b3 4 6 1
+ b3 6 1
b364 1
6 b3 + 1
4 6 b3 1
6 4 b3 1
b34 1 6
4b3 1 6...
These combinauons, and the ones that would follow, are wNhin a freewheeling area of Abhogi (b3 4 6 1). If some of them are chosen, taste will decide how they should be coUched. Ma dha bga sa (4 6 b3 1), for example, js extremely angular and might be best offset if prepared and followed by less adventuresome phrases.
Thus mirkhapd can seldom be directly quoted: the patterns must be "forgotten," even while taking advantage of some of the notesequences they suggest. To avoid sounding perfunctory or simply wrong (jn raga), Amir Khansahab applies markhar!d by the spirjt rather than the letter, and thts ts one of the keys to hjs art. The jnfluence of this technique may be seen jn passages which manage to reconcile a tjghtly ordered permutational system with a raga's grab-bag of strictures and freedoms, as in this Abhogi excerpt:
On the "FOCUS" staff of the transcription, showing the structural tones of each passage, notice that in each of the statements A through D, the sixth degree, dha, is prominent, with subsidiary notes in parentheses alongside it. Amir Khan devises phrases in the raga which center on a particular note as a theme in this passage, the note dha -- and each phrase brings out a previously hidden shade in this theme note.
Subtle qualities such as the sonority of Khansahab's voice and the timings of the phrases aid in ways that are perhaps beyond a written portrayal. But it is possible, with such a reductive transcription, to discuss the shifting balance of pitches surrounding the theme note, which usually appears at the ends of phrases. Bor example, all four phrases of statement A center on dha. In Al, it is approached from below by ga and ma (degrees b3 and 4L b3 ~ 9 6 ~ ~ ~4
A2 devotes a little more attention to ga and ma, but contains a long dha toward its beginning; it also ends on dha 4 6 - - b3 4 b3 4
6
A3 shifts a little toward ga, but again ends with dha: b3 4
b3 6.
And A4 steers away from ga, touching dha repeatedly, while introducing at its end a new note, the tonic sa. Looking more closely at this last phrase, one notices that it consists of three phrases, each casting a different light on dha: (6) b3: 4 6; 4 6 - -; (1) 4 6 1 - 6
Many phrases from the bistar passage appear to have their origins in b3-4-6 and 4-6-1 mirkhand patterns, as tabulated below:
mirkhand pattern bistar phrase
b3 4 6
4 b3 6
b3 6 4
6 b3 4
4 6 b3
6 4 b3
4 6 1
6 4 1
4 1 6
1 4 6
6 1 4
1 6 4
Al, A2, A4, Bl, C3, C4
A3
A2
Dl, El
Dl
D2
D3
C2, C3, Dl, D2
A few phrases such as 1 6 b3 4 (D3) may plausibly derive from fournote arrays. Some patterns are especially prominent because their last note ends a phrase, while others are imbedded in a larger gesture. Here, Amir Khan's ability to keep dha fresh under scrutiny (and to suspend the listener's perception of time) seems beholden to the shell-game dexterity that m1rkhan4 imparts.
In untrammeled arrays such as the b3-4-5-1 region of Abhogl, mirkhan4's note-for-note influence is demonstrable in Amir Khan's music. But what about the thornier, more circuitous ragas, such as Kedar7 This melodic type shares with Abhogl the use of ma and sa as the stressed and end-note pitches. Beyond this, its nature lies in contrast. It is quite vakr'(angular) in melodic habit, with an ascent and descent as follows:
The rather straightforward canon for Abhogi, as we have seen, admits some mirkhand-inspired passages. But the angularity and catch-phrases of Kedar's melodic type would make a search for note for-note mirkhand thinking fruitiess. However, the principle of this permutational technique can be uncovered here if a few basic Kedar phrases (and their paraphrases) are substituted for Individual notes in the patterning. Our excerpt can be boiled down to the following gestures:
I (4) 2 - I (ma) re - sa
11 5 ~ I pa - sa
111 1-5 sa-pa
IV 5 - 4 pa - ma
V I (4) - 5 sa (ma) - pa
When each gesture is given a Roman numeral (as seen in the transcription), and the numbers grouped according to breath-marks and other dividers, sequences emerge which may owe much, by extrapolation, to the practice of mirkhand.
A
Statement Phrase Cesture-Sequence
1 11 1 111
2 111 1 11
3 111 11 1
4 V 1 11 IV
5 V 11 1
6 Vl I 11 1
7 111 11 1 1 {mukhra)
B
1 V IV I
2 IV I V
3 IV I IV
4 11 1
5 IV V IV I IV
6 I III II . . . I [mukhra)
C 1 V IV I
2 IV I V
3 V IV 11 IV
4 1 111 11 1
5 V IV I
6 11 IV 1
7 V IV I [mukhra)
D 1 I V IV
2 V IV I
3 V IV V IV I 11
4 11 1 IV
5 IV 111 11 1 {mukhral
Here mirkhand's putative influence can be discerned as a permutor of gesture-sequences rather than note-sequences. Gesture-sequences such as (1) 11 1 111 - (1) 111 1 11 - (1) 111 11 1 in bistar statement A; V IV I - IV V I in B; and V IV I - IV I V in C demonstrate the permutational principle.
The following depicts how the paraphrases reduce to gestures that implement the principle:
I (4) 2 - I (ma) re - sa
11 5 I pa- sa
111 1 - 5 sa - pa
IV 5 - 4 pa - ma
V I (4)- 5 sa (maL pa
(All the staff notations missing due to lack of pictorial clarity. JC)
The sequences Al through A3 all begin with 1, but then permute gestures 1, 11, and 111. Sequences Bl-2 and Cl-2 permute 1,
IV, and V, serving incidentally as a good illustration of two different paraphrases of the sequence V IV I - IV I V. However, sequences such as V IV - V IV I - V IV V IV I 11 in bistar D (1-3) show a cumulative gesture-sequence, unlike mirkhand patterns.
As far as I know, seeing mirkhap4's influence in the arrangement of gesture-sequences is unrecognized in both written and oral commentary about Indian music: and yet if mirkhand is rightly taken as a principle of permutation rather than solely as a specific series of note-sequences, its presence may well be discernible in certain parts of Khansahib's bistar in Kedar.
Focusing on these details reveals the deftness of Amir Khan's minutiae: to vary the approaches to and departures from important notes is continually to refine their evocative power. A raga such as the relatively straight-up-and-down Abhogi admits an application of m1rkhand in note-sequences, while m1rkhand 's influence may be deduced in the permutation of gesture-paraphrases in the angular Kedsr. Whether on the level of note or phrase, the singer must control details even while keeping track of a larger design, and this is a clue to Khansahab's artistry. How the sequencing of a raga's phrases fits into the yet larger time-spans of the ultra slow 14-beat Jhumra tala favored by Amir Khan, and how these tala cycles in turn may be seen to be counterbalanced in an overarching awareness of architectonic design, await future analyses. Such studies may once again reveal the permutational principle: indeed, it is from this vantage that Jairazbhoy has noted that "the whole concept of the development of the melodic line in . . . khayal [sicl seems to have a marked similarity to the development of the sequences of [m1rkhand]" (Jairazbhoy 1961:313).
It appears that Amir Khan employed m1rkhand obliquely to meet the rupa or formal demands of each raga.8 The way in which these sometimes inimical forces interact in Khansahab's music is, I hope, clearer through the examples presented here. "Khyal" means "idea" or "imagination." Mirkhand serves, then, as a lever for the imagination, too mechanical and thoroughgoing to be quoted directly. But in a master's hand, a mundane tool can produce great art.
Descrying m1rkhand's variational role helps to show how Amir Khansahab is able to hold just a few notes under a powerful lens, revealing hitherto unheard wonders; and it is especially this that has attracted and inspired generations of musiclans and musiclovers from both North and South India. He has been called innovative. But this plays down the fact that Amir Khan's style is illustrious equally because it honors traditional precepts of good singing (hence the "classical purity" alluded to by our earlier commentator): the treatment of tala is both assured and ingenious, and the approach to raga, one component of which I have examined here, manages to honor and illuminate the proper rupa even while coaxing out the unexpected. Innovation here is in the manner of old wines in new bottles, rather like the work of Miles Davis in jazz.9
This view of mirkhand's role also sheds light on that sometimes befogged term "improvisation" (see Nettl 1974; Powers 1980) it affords instances wherein a performance is neither purely out-of-the-blue extemporizing nor rote recitation. At least seen from the perspective of Indian music, one can say (with a nod to T. S. Eliot) that a good improvisation mixes memory with desire: the concordance of familiar raga-phrases meets the striving of bistar's invention an invention aided, as we have seen, by m1rkhand.
Union College
Notes
N.B.: In addition to the cited bibliography, much of the raga-rupa descriptions are from my direct discipleship with Ustad Amir Khan and tutelage from his senior disciples, Srikant Bakre and Kamal Banerjee. I am also indebted to A. Kanan and Malabika Kanan for additional information.
A Muslim master (Ustad) such as Amir Khan may be referred to as "Khansahab," "Amir Khansahab," or with a reduplicative honorific: "Khansahab Amir Khan." In addition to providing the variety good prose enjoys, these usages (with all the familiarity wielded by a devotee or student) seem to me to be as appropriate as referring, on occasion, to Lester Young as "Pres," to Davis as "Miles," or indeed to desPres as "Josquln" in academic writing.
2 From Sanskrit khanda-meru, "mountain of squares," referring to the pyramidal form obtained by writing out the successively lengthening patterns. Today's parlance may use mirkhand interchangeably with prastara, "permutation." See Jairazbhoy 1961.
3 If our ragas were selected for concert performance, a proper timezone and sequence for them might be: Kedar, 7-8 pm; Abhogi, 8-9 pm.
4 Anonymous liner-notes to Amir Khan: Great Master, Great Music, EMI EALP 2765, 1976.
5 Excerpt, Music from Pakistan Nazakat and Salamat Ali Khan, EMI EALP 308, 1964. A line of vocal melody is represented on two staffs, comprising one system. The upper of these shows the rhythmic cycle in large divisions: in this case, the tala is a medium slow Jhumra of 14 beats, taking up two systems wlth each beat marked by a half-bar. The corresponding passage in Abhogi by Amir Khan is in the 10-beat cycle Jhaptal of medium tempo, all on one system with a dotted baritone showing the 5 plus 5 subdivision. In the Kedar transcription, which is in Amir Khansahab's famously slow Jhumra, the systems are in groups which alternate between 3 and 4 beats. Passages of bistar are separated in all examples by returns to the mukhra, which consists of the first few notes of the composed khyal. I omit these returns in the transcription, since they do not figure In the
phenomenon under examination; for the same reason, the singing-text does not appear. Since a vocal performance is expected to float over the slow tala beats given by the tabla drums, rhythmic values for the singing are self-relating approximations placed with enough precision to show how many cycles are taken for each improvisatory statement before a return to the composed mukhra. Notes without flags are relatively qulck, and single notes in parentheses and tied sixteenths are respectively quicker and quickest, serving as ornaments to slower values. On the lower staff are written the prlme Indlan solfege pitches, or area of FOCUS, upon which each phrase concentrates. Solfege pitches of secondary importance are in parentheses.
6 Personal communication, Kamal Banerjee (a senior disciple of Amir Khan), 1967.
7 Recording courtesy Vijay Kichlu, Sangeet Research Academy, Calcutta.
8 I first heard this asserted by- the singer Malabika Kanan (personal communication: Calcutta, 1986).
9 In the chapter "Men of Few Notes: Miles Davis and Amir Khan," from A Tune Beyond Us, Yet Ourselves On Transcultural Hearing (Ph.D. Thesis, Wesleyan University: 1985), I pursue this parallel in aesthetic viewpoint.
References Cited
8hatkhande, V.N.
1956 Hindustanl Sangit-Paddhati (Hindi translation), I-IV. (Marathi originally publ. 1910) Bombay: Hathras.
Jairazbhoy, Nazir "Swaraprastara in North India Classical Music", Bulletin of the School of Oriental and African Studies. Vo. 24 Pt 2, University of London. 1961
Thakur,Omkarnath
1954-62 Sangitanjali (I-IV), Banaras (Vol. 11, publ.1938).
Nettl, Bruno
1974 "Thoughts On Improvisatlon: A Comparative
Approach," The Musical Quarterly 60:1.
Neuman, Daniel
1980 The Life of Music in North India: The Organization
of an Artistic Tradition. Detroit: Wayne State
University Press.
Powers, Harold S.
1976 "The Structure of Musical Meaning: A View from
Banaras," Perspectives of New Music 14:2/lS:I.
1980 Language Models and Musical Analysis,"
Ethnomusicology 24:1.
1983 "North Indian R.gas." Graduate seminar, Wesleva~
Untversity.
Wade, B.C.
1984 Khyal: Creativity within North India's Classlcal
Music Tradition. Cambridge: Cambridge University
Press.