Dedication

Amir Khusrau (1253-1325), a celebrated court poet and musician of North India, is commonly credited with several innovations in Indian classical music. In a critical appraisal of those contributions, this paper shows that the claims have been exaggerated and remain largely unproven. To build his case, the author uses a wide range of primary and secondary Persian and Urdu sources. He has drawn heavily upon the writings of the late Pakistani musicologist Rashid Malik but has also made a number of original arguments. This paper was published in the journal Sangeet Natak ( Volume XLII, No. 4, 2008).

**********************************************************************

Introduction

The name of the poet Amir Khusrau (1253-1325) is associated with several innovations in Indian classical music dating to the medieval period. Though Khusrau’s fame outside India is largely based on his Persian poetry, in India he is also remembered for his many putative contributions to Indian music. The writer of an anthology of Persian poetry describes his musical talent as follows:

A superb musician in his own right and credited with the invention of several musical instruments and with having laid the theoretical basis for much of Indo-Muslim music, Khusraw imparted to his ghazals a lilt and melody that have assured their inclusion in musical programs in India to the present day.¹

Khusrau’s fame as a Persian poet is indeed richly deserved. Known throughout the Persian-speaking world as Amir Khusrau-e-Dihlawī (Amir Khusrau of Delhi), Khusrau was court-poet to several kings in Delhi, most notable of whom was Sultan ‘Alā-ud-dīn Khiljī. Contemporary (or even some later) historians, however, do not remember him as a musician, but only as a poet. For instance Firishta, who writes in detail of the court of ‘Alā-ud-dīn Khiljī, lists Khusrau’s name among the poets, not among the qawwāls or musicians.²

In this paper, we attempt the beginning of a critical study of Amir Khusrau’s contributions to Indian music. While this subject has been much discussed, it is the aim of this paper to show briefly that a revision of the existing history (at least in the English language) is long overdue. A very important book Ḥaẓrat Amīr Khusro kā ‘ilm-e-musīqī (The Music of Amir Khusrau), by Rashid Malik, exists in Urdu, which deals extensively with this subject. Unfortunately, it is still unavailable in English.

Musical innovations ascribed to Khusrau

Khusrau is said to be the inventor of the sitār³ and the tablā.⁴ But in addition to these musical instruments, he said to have been the originator of genres such as the tarāna⁵ and the qaul. He is also said to have composed numerous new rāgas. Muhammad Wahid Mirza, who is the author of an authoritative biography of Amir Khusrau, sums up Khusrau’s musical contributions:

According to an old Persian work⁶ on Indian music (which is supposed to a be a translation of an older work⁷ written in the time of Rajah Mānsingh of Gwaliyar), he invented the following new melodies: mujīr, sāzgārī, aiman, ‘ushshāq, muwāfiq, ghazan, zilaf, farghāna, sarparda, bākharz, firodast, mun‘am (?), qaul, tarāna, khayāl, nigār, basīṭ, shāhāna, and suhila.⁸

(In this quote, genres of music such as qaul, tarāna and khāyal have been conflated with names of rāgas such as ‘ushshāq and aiman. Such inappropriate conflations abound in the secondary literature on Khusrau.)

Difficulties in evaluating Khusrau’s musical legacy

The chief difficulty in studying Khusrau’s contributions to music is that his own writings mention none of the contributions now ascribed to him, but later works attribute several innovations to him. These later works include Abul Fazl’s ‘Āin-i Akbarī which was written in 1601. Five decades later came ‘Abd ul-Hamid Lahori’s chronicle of Shah Jahan’s reign, the Bâdshâh Nâmah. In the next decade, during the rule of the emperor, Aurangzeb appeared the Râg Darpan, which was followed in turn by the Tuḥfat-ul-Hind. The last-named work consisted of five volumes dealing with the arts and culture of India, one of which was dedicated to music. The Râg Darpan makes many more claims on behalf of Khusrau than does the Bâdshâh Nâmah.⁹The fact that most such claims arose over three hundred years after Khusrau’s death suggests that these claims may be spurious.

Khusrau’s biographer, Mirza is unable to confirm that Khusrau indeed invented the sitār and the tablā, claims which have been made so often that the matter is often assumed as proved.

But unfortunately, I have been unable to trace the name ‘sitār’ anywhere in Khusrau’s writings, although there are pages full of the description of the various instruments used in his time. Nor does any of his contemporary, or even later writers mention the name.¹⁰

The Tablā

There is a widely quoted story that Khusrau invented the tablā by cutting up an existing instrument into two halves. But this story can be dismissed very easily, even with a very simplistic explanation:

The story current among musicians that Khusrau cut the mridang into two halves and it thus became the tabla has no basis, for a mridang cut into two will not acquire the shape of a tabla.¹¹

The two drums that make up the pair of drums called the tablā fulfil different functions. The left drum bāyān is used to generate deeper sounds and support the melody which is chiefly carried by the right drum. The two drums are different in size and shape and have different membranes.

The legend may also have a simple explanation based on folk etymology, though it does not explain the association of the name of Khusrau with the instrument. Classical Sanskrit sources and also Indo-Persian sources on music mention an instrument called the āvaj which consisted of two drums. A single drum was called ardhāvaj (half an āvaj).¹² This etymology could well be at the root of the persistent claim that the mŗdaṅga (or in some renderings, the ḍhol) was divided into two parts, giving rise to the tablā. The tablā is also regarded by some as being of Middle Eastern origin, based on its name, derived from the Arabic word t̤abl. However, the drums which go by this name in the Middle East are single-membrane drums, and lack the additional black central circular membrane (gābh/syāhī) which is a feature of several Indian drums such as the pakhāvaj and the mŗdaṅga. The Indian tablā may be a drum indigenous to India, which may have been renamed in Islamic times. This view can be bolstered by the observation that the tablā has indigenous regional names such as the dukaṛ in Punjab¹³ and dokra in Kashmir.¹⁴ Also, according to the influential Persian encyclopedia Loghatnâmeh-ye Dehkhodâ, the word t̤abl in Indian Persian contexts refers to the Persian drum tombak, which is a favourite instrument among the mystics (Sufis) of India (‘irfân-e Hind).¹⁵ That is, the word t̤abl does not refer to the instrument now known as the tablā at all. This would again support the contention that the Perso-Arabic word t̤abl, modified as tablā has been transferred to an indigenous drum. At any rate, Khusrau himself refers to a drum, tabīra-i Hindī, “the Indian drum”, in which he explicitly acknowledges its Indian origin, without claiming to be its inventor.

The Tarānā

Another of Khusrau’s contentious contributions is the genre tarānā, described by Wade as a

rhythm-oriented vocal genre featuring vocables and sometimes poetry, sargam, or drum syllables as text, frequently performed after khyāl in medium or fast speed but occasionally sung slow speed; counterpart of Karnatak tillānā.¹⁶

Wade adds that “the Karnatak version of tarānā, which is called tillānā, is very similar and is said to have developed at about the same time.”¹⁷ In saying this, Wade confirms the accepted idea that Khusrau’s invention, the tarānā, led to the development of the tillānā. The textual content of the tarānā is not verse, but vocables, such as in the phrase

“Ta re da ni ta da ni”, (excerpted by Wade from a tarānā sung by Salamat Ali and Nazakat Ali Khan). Occasionally a tarānā may include a Persian couplet, but this couplet does not function as a poetic verse might in a song genre such as the khyāl. Similarly, in a tillānā

…drum syllables, solfège, and brief passages of poetry provide the text. In dance tillānās, the rhythmic passages are composed so as to correspond with footwork, and drum syllables provide the only text.¹⁸

We will address the question whether Khusrau was indeed the originator of all these modes and genres mentioned above, by first looking at some writings which are widely quoted. We begin with S.Q. Fatimi, whose book contains much information about Khusrau’s music. Fatimi discusses some of Khusrau’s contributions with translations of excerpts from the Bâdshâh Nâmah:

Half a century after the ‘Ain-i-Akbari came ‘Abd ul-Hamid Lahauri’s official chronicle of the first twenty years of Shah Jahan’s reign, named Badshah Namah. He wrote that before Amir Khusrau’s times git, chhand, dhurpad, and astiti used to be sung in Hindi, but the Amir introduced the following:

Avval, qoul, keh be-qânûn-e gît mushtamal ast bar ‘arabî-o-farsî be-nazm yâ be-nasr va binâ-ye ân bar yek tâl ast yâ do yâ seh yâ chahâr

(First, qaul, i.e., Arabic and Persian, poetic or prose, passages sung according to the rules of git, based on a single, or duple [sic], or triple, or quadruple tal (measure of time).)

Dovvom, fârsî, ‘ash‘âr-i fârsî bâ tarâneh mubnî bar yek tâl farâham âvardeh

(Second, Farsi, i.e. Persian couplets sung in the tarana (form of music) basing on a single tal (measure of time).)

Sevvom, tarâneh keh bî ‘ash‘âr asâs-e ân bar yek tâl gozâshteh

(Third, tarana, i.e., the singing of tarana without (words of) couplets based on a single tal.)

Chahârom, tasnîfî, keh be-hindustânî zabân bar gozârad va ânrâ khayâl nâmîd va khayâl bish az bar yekjandî (?) bar sarâ’îdeh and.

(Fourth, tasnifi, (lit. related to authorship, i.e. original) which he composed in the Hindustani language and called it khayal ….)¹⁹

We can see that Fatimi translates the word “tarâneh” as “tarana (form of music)” in both the second and third items listed in the Bâdshâh Nâmah. But this is incorrect. “Tarâneh” in Persian simply means “song”, “singing” or “poem”.²⁰ In the second item, the word “tarâneh” simply means “song” or “singing”. That is, we are dealing with couplets in Persian being sung (as opposed to being recited). In the third item, we are told about a song or melody consisting of words without “poetry”, sung to a single beat. Fatimi translates the word “bî ‘ash‘âr” as “without couplets” but it is more appropriate to render it as “without poetry” because the word “‘ash‘âr” (‘ash‘âr being plural of the word shi‘r) more generally means “poetry”/“verses” and not necessarily “couplets”.

Fatimi’s translation of the word tarâneh assumes that the word tarāna in medieval Persian already meant what is now understood by tarāna in Indian music. That is, he assumes the present meaning of the tarāna in order to prove that it meant the same thing in Khusrau’s time. In fact, the correct translations of the second and third items should read:

Dovvom, fârsî, ‘ash‘âr-i fârsî bâ tarâneh mubnî bar yek tâl farâham âvardeh

Second, Fârsî, Persian verses sung (literally, with song), based on a single tāla (or based on the beat ektāla).²¹

Sevvom, tarâneh keh bî ‘ash‘âr asâs-e ân bar yek tâl gozâshteh

Third, a song without verses, based on single tāla (or based on the beat ektāla).

However, for the very first item we see that Khusrau’s contribution was that of a poet, and not that of a musician. Khusrau’s qaul clearly consisted of Arabic and Persian passages, sung according to the (existing) rules of gīt.²² Even in the case of the second and third items, there is no evidence that Khusrau introduced a musical innovation, there being only a mention of the words yek tâl, and the use of Persian text. The question still remains whether Khusrau really invented what is now called the tarāna. A modern commentator on Indian music, Thakur Jai Deva Singh, answers in the affirmative. He writes:

This was entirely an invention of Khusrau. Tarana is a Persian word meaning a song. Tillana is a corrupt form of this word. True, Khusrau had before him the example of Nirgit songs using śuṣk-akṣaras (meaningless words) and pāṭ-akṣaras (Mnemonic syllables of the mridang). Such songs were in vogue at least from the time of Bharat. But generally speaking, the Nirgit used hard consonants. Khusrau introduced two innovations in this form of vocal music. Firstly, he introduced mostly Persian words with soft consonants. Secondly, he so arranged these words that they bore some sense. He also introduced a few Hindi words to complete the sense….

It was only Khusrau’s genius that could arrange these words in such a way to yield some meaning. Composers after him could not succeed in doing so, and the tarana became as meaningless as the ancient Nirgit.

While Jai Deva Singh clearly admits the existence since ancient times of songs using words without semantic meaning, and drum syllables, he regards Khusrau as having invented the tarāna genre for having introduced Persian words and for rearranging them to make some sense. The sense also needed to be complete only with the addition of Hindi words. Jai Dev Singh gives some examples of these words, but we prefer here to quote Ustad Amir Khan who seems to have been the first person in modern times to have expressed this view:²³

Tanan Dar Aa — Enter my body;

O Dani — He knows;

Tu Dani — You know;

Na Dir Dani — You are the complete wisdom;

Tom — I am yours, I belong to you;

Yala — Ya Allah;

and, Yali — Ya ‘Ali.

These translations are only partially correct. Tanan does not mean “my body” (but tanam would have meant it.) While tû dânî correctly means “you know”, û dânî is ungrammatical. Nâdir means “rarity”, and has meaning only as a single word — i.e., dir has no meaning at all. The translation “You are the complete wisdom” is simply incorrect; so is the translation “I am yours, I belong to you”. Such difficulties can be illustrated with the help of the following verse attributed to Khusrau by Jai Deva Singh:²⁴

Hayya ya dir tala laye — Hasan-o-Nizamuddin Auliya, dem dem dir dir tan tan tale ta — na na, na na, na na.

The reader who tries to make sense of this verse will probably agree that the “tarana became as meaningless as the ancient Nirgit” even in Khusrau’s time! The syllables Dem, Dir, Tale have no particular lexical meaning whether in Persian or in an Indian language.

In spite of great efforts to read “meaning” into the tarāna, we find it makes sense only as described by a distinguished artiste:

For Bharata Natyam, Tillana is basically a structure which follows a particular sequence of phrasing and evolves in a certain way. It is performed, traditionally, at the end of a recital — usually fast-paced, rhythmic and exciting. There are a set of syllables, or sollukottus, that are typical to a Tillana. They have no meaning — they are not meant to have any meaning. Usually, the syllables are something like this: tom till ana udanata deem deem tana na dari tat da, etc. They are composed purely based on how beautiful they sound together. Traditionally, there is a short two line prayer within the Tillana towards the end.

On the other hand, in the pushpanjali (flower offering usually done at the beginning of a recital), the Natya Shastra actually lays down certain syllables called “nandi shabda” which are said to have emanated from Shiva’s drum. Subsequently, the sounds became words and thus the creation was born. These “nandi shabda’s” are said to have auspicious vibrations that bless the rest of the performance. Again, they have no meaning but have been specifically prescribed by the Natya Shastra.²⁵

The use of onomatopoeic syllables to mimic or notate music and dance is very ancient and traditional.²⁶ There are many such schemes which cannot be attributed to Khusrau. However, a practice of attributing mystical significance to some syllables used in music did exist, but in Indo-Persian writings on music, it has been traced only as far back as the Shams al-As̤wāt of Ras Baras Khan Kalawant,²⁷ which is dated 1698. This practice, which is attested in the practice of dhrupad ālāp, is without precedent, at least in the extant literature. Thus this tradition also cannot be ascribed to Khusrau, on the basis of the evidence at hand.

Rāgas attributed to Khusrau

So far we have only discussed the linguistic contributions made by Khusrau (namely, the introduction of Persian and Arabic poetry and or terminology), or the instruments he is said to have invented. Now let us look at the more specific claims that attribute new rāgas to Khusrau. For this, we first look at the text which started the trend, namely, the Râg Darpan²⁸ by Faqirullah Saif Khan, a work begun in 1662/1663 and completed in 1666. (In what follows, the Persian text has been taken from Malik’s edition, pp. 98-99.²⁹)

Amîr ‘aleiḥ raḥmatullah az jomleh-ye râghâ davâzdah râg râ gozîn namûdah ânrâ nâmhâ nehâdah badîn tartîb:

The Amir, God’s Mercy be upon him, from among the ragas, chose twelve, and named them in this manner:

Dar berârî va mâlasrî dogâh ḥoseinî ẓamm namûdah mowâfiq nâm kardah vîvâlî nîz gûyand.

In Bairârî and Mâlasrî, he mixed Dogâh Ḥoseinî, and named it Movâfiq — it is also called Vîvâlî.

Dar toḍî panjgâh va muḥayyir keh gosheh-ye ûst yekjâ kardah muḥayyir nâm kardah.

In Toḍî he put Panjgâh and Muḥayyir (which is a gosheh of Panjgâh) together, and named it Muḥayyir.

Pûrbî râ ghanam guyad va az maqâmât-e fârsî shahnâz dâkhil kardah.

He called Pûrbî Ghanam, and of the Persian maqâms he introduced Shahnâz.

Khaṭ-râg-râ zîlaf nâm gozâshteh.

He named Khaṭrâg as Zîlaf.

Dar fârsî Khaṭ-râg -râg-râ ghazâl gûyand — dar pârsî va mârag va desî Khaṭ-râg yek ast. Dar ân tafâvut nîst. Gheir az Khaṭ-râg hîch râg nîst keh dar fârsî va hendî yekî bâshad. Âre ba‘azî râghâ hastand keh dar desî va mârag meyân-e ânhâ tafâvut nîst. Avval shash râg, dîgar kalyân, va deshkâr, va desâkh, gûjrî, gonḍ, soraṭhî, sindhû, saindhavî, madhmât, sâvant, tarûn, bholâ, jaijâvantî, mangal bhairavî, marû, bangâl — shâyad chandî dîgar bâshad.

In Persian, Khaṭ-râg is called Ghazâl. Khaṭ-râg is the same in the Persian system and in the Marga and the Desi systems. There is no difference between them. Other than Khaṭ-râg, there is no rāga which is identical in the Persian and Indian systems. Of course, there are many rāgas which are the same in the Desi and Marga systems. The first six ragas, then Kalyân, Deshkâr, Desâkh, Gûjrî, Gonḍ, Soraṭhî, Sindhû, Saindhavî, Madhmât, Sâvant, Tarûn, Bholâ, Jaijâvantî, Mangal Bhairavî, Marû, Bangâl — there may be some more.

Gaurah râ farghânah nâm kardah, chûn farghânah az maqâmât-e fars dâkhil kardah.

He named Gaurah as Farghânah because Farghânah is one of the Persian maqâms.

Va dar sârang navâ va basant ẓamm namûdah ‘ushshâq laqab gozâshteh.

And adding Navâ and Basant to Sârang, he named it (i.e., the result) the ‘Ushshâq.

Dar gonḍ, bilâval va gaur sârang va az maqâmât-e fars râst râ mulhaq sâkhteh, sarpardah nâm nehâdeh,

In Gonḍ, he added Bilâval and Gaur Sârang, and the Persian maqâm Râst, and named it Sarpardah.

Dar kânhrah chand râg bâham makhlût kardah chonâncheh bâlâ taḥrîr yâft az aṣl nuskhah, va ân nîz âhangî ẓamm namûdah firodast esm gozâshteh.

In Kânhrah, he blended a few rāgas, as written above in the original manuscript, and further adding an âhang, named it Firodast.

Dar aiman neirez ẓamm namûdah ânrâ aimanî gûyad.

He added Neirez to Aiman, and named it Aimanî.

Pûrbî, Bibhâs,Gaurah, Gunkalî, va az maqâmât-e fars ‘irâq dar ân dâkhil kardah sâzgîrî nâm kardah.

He blended Pûrbî, Bibhâs, Gaurah, Gunkalî, and a Persian maqâm ‘Irâq, and named the result Sâzgîrî.

Va dar deshkâr bâkharz kaz maqâmât-e fars ast ẓamm namûdah ânrâ bâkharz laqab gozâshteh.

And adding the Persian maqâm Bâkharz to Deshkâr, he named it Bâkharz.

Va dar kalyân nei-rez mulhaq sâkhteh ghanam laqab kardah.

And blending Neirez in Kalyân, he called it Ghanam.

Muḥtajib namânad keh dar sâzgirî bâkharz va ‘ushshâq va mowâfiq dar în chahâr râg kheilî kâr kardah tâ dîgar râg-o-maqâm be-t̤arîq-e onîq ẓamm namûdah. Dar dîgar râghâ chandân kâr nakardah be-joz ân-keh maqâmî makhlût̤ namûdah va nâmî gozâshteh.

Dîgar az jomleh-ye râghâ-ye amîr aiman basant ast keh aiman va basant râ yekjâ kardah.

Let it be known that he did a lot of work on the four rāgas Sâzgirî, Bâkharz, ‘Ushshâq and Mowâfiq, and only then beautified the other rāgas and maqâms. On the other rāgas he did not do much work other than blending a certain maqâm or giving them a new name. Another of Amir’s rāgas is Aiman Basant which is Aiman and Basant brought together.

It is difficult to determine what exactly Khusrau’s innovations were, because the words “ẓamm kardan/namûdan” would literally mean “to add, annex, append”. For instance, in the absence of independent complementary information from other sources, it is difficult to understand what is meant by “adding the Persian maqâm Bâkharz to Deshkâr, and naming it Bâkharz”.³⁰ But in any case, we see that at least some of the rāgas Khusrau is said to have invented are no more than previously existing rāgas renamed(such as the ràga Pûrbî which he called Ghanam, or the ràga Gaurah which he renamed Farghânah).

The later work Tuḥfat-ul-Hind by Mirzā Muḥammad Ibn Fakhr-ud-dīn has its fifth volume devoted to music. Its eighth chapter has a discussion of rāgas composed by Amir Khusrau. It lists as Khusrau’s contributions the following rāgas:

First, Muḥayyir: It is said to be a composite of Ghârâ and a Persian maqâm. Some people say that is a composite of Toḍî and ‘Irâq.

Second, Sâzgîrî: It is a composite of Pûrbî, Gaurâ and Gunkalî, and is one of the Iranian maqâms. Some people mention Bibhās instead of Pûrbî.

Third, Yaman: It is a composite of Hinḍol and a Persian maqâm. Some people regard it as a composite of Aiman and a Persian maqâm.

Fourth, ‘Ushshâq: It is a composite of Sârang, Basant and a Persian maqâm.

Fifth, Movâfiq: It is a composite of Toḍî, Mâlasrî, Dogâh and Ḥoseinî, and it is also called Dîvalî.

Sixth, Ghanam: It is derived by making small variations in Pûrbî.

Seventh, Zîlaf: It is derived by making small variations in Khaṭ-râg.

Eighth, Farghanâ: It is a composite of Gunkalî and Gaurâ.

Ninth, Sarpardah: It is a composite of Gauḍ Sârang and a Persian maqâm. Some people regard it as a composite of Gonḍ Bilâval, Pûriyâ and a Persian maqâm.

Tenth, Bâkhrez: It is a composite of Deskâr and a Persian maqâm.

Eleventh, Firodast: It is a composite of Kânhrâ, Gaurî, Pûrbî Syâm and a Persian maqâm.

Twelfth, Ghanam: It is a composite of Kalyân and a Persian maqâm. Some people call it Neirez instead, and regard it as a composite of Paṭmanjarî and a Persian maqâm.³¹

There are some contradictions in the claims of the Râg Darpan and those of the Tuḥfat-ul-Hind. There are some minor differences of spelling, such as in the case of Bâkharz and Bâkhrez. But there are more serious differences between the two sources. For example, the Râg Darpan unambiguously asserts that Khusrau did not compose all the twelve rāgas associated with him — that he composed only a few of them, and renamed some (Ghanam, Zîlaf, Farghânah). The Tuḥfat-ul-Hind, however, claims that he composed all the twelve rāgas associated with him, even if some of them involved only small variations in existing rāgas.

Perhaps the most significant difference is the fact the later work, the Tuḥfat-ul-Hind, is less detailed in its information. It mentions only two of the Persian maqâms which Khusrau is supposed to have used in his creations, namely, ‘Irâq and Neirez. However, while being less detailed, it is more emphatic in his conclusion that Khusrau actually composed the rāgas associated with him. The process of myth-making involving Khusrau seems to have already been underway by the time the Tuḥfat-ul-Hind was composed.

While the Râg Darpan attributes to Khusrau the creation of a rāga Aimanî, whereas the Tuḥfat-ul-Hind attributes to him the creation of a rāga Yaman. The rāga Yaman (often called Yaman Kalyān, sometimes Aiman or Aiman Kalyān) is regarded by some to be a Middle Eastern rāga borrowed into Indian music. It is sometimes credited to Khusrau, but sometimes more involved explanations are offered, such as this one by Sarmadee:

Aiman is undoubtedly Yamana Indianized. Yamana (Southern Arabia) and Kalyana (near Bombay coast) have been trade-links and culturally congenial places of early medieval days. Hence the two ragas Yamana and Kalyana have always fraternized the way they have.³²

This argument seems far-fetched in view of the fact that early Sanskrit texts do not mention any rāga named Yamana. It is indeed true that there were far-reaching trade links between India and Arabia, dating to pre-Islamic times. For example, the island of Socotra (off the coast of Yemen) had a large Indian merchant population and even may have taken its name Socotra from the Sanskrit word Sukhātara-dvīpa.³³ However, a rāga that entered Indian musical culture from Yemen, ostensibly along the west coast, would surely not have needed a thirteenth-century poet from Delhi to “invent” it, so that its ascription to Khusrau is suspect. (In fact, it is only the Arabic name that suggests a Middle Eastern origin for this rāga; there is no other internal evidence that it is otherwise an innovation in Indian music.)

The explanation is probably much simpler than the speculations offered by Sarmadee. It has been long noted that the Arabic word Aiman and the Sanskrit word Kalyāṇa have the same meaning.³⁴ After the conquest of the kingdom of Devgiri by ‘Alā-ud-din Khiljī, its capital Kalyāṇa was renamed Aimanābād. The use of the compound name for this rāga, consisting of the juxtaposed words Aiman and Kalyāṇa probably dates from this time.

Conclusion

Rashid Malik, the author of the definitive work on Khusrau’s music alluded to above, points out that unlike other composers such as Tānsen, Mirābāī, Sūrdās or Rāmdās, whose names are commemorated both in the texts and in the living traditions of the musicians themselves in form of rāga-names such as Mīyān kī Malhār, and Rāmdāsī Malhār, Khusrau’s name does not explicitly figure in any such rāga-name.³⁵ It is possible that Khusrau has been credited with the contributions made by a long list of musicians, whose names are now lost to us. Even the rāgas now associated with Khusrau (by historians) have turned out to be largely ephemeral, and hardly figure in the repertoires of Indian and Pakistani musicians, whether Muslim or Hindu.³⁶ In spite of the claims such as the one made by Thackston (see footnote 1), Khusrau’s impress on Indian music is simply not as great as his enormous fame could lead one to believe. As one of the brightest stars in the firmament of Indo-Persian poetry, he will certainly continue to be remembered by posterity with respect. But a careful reappraisal of his musical legacy may rehabilitate the work of many great but as yet unknown musicians, whose innovations have been attributed to him.

Transliteration Scheme – Long vowels in Indic words have been transliterated with the help of a macron, but long vowels in Persian words are indicated with the use of the circumflex. In the case of Indic words appearing in a Persian text, the Persian transliteration conventions have been followed. Thus, for instance, the words Raga is spelled as Rāga if it occurs in a Sanskrit work, but râg if it occurs in a Persian text. Arabic velar consonants have been transliterated with the help of two dots below the corresponding letter; Sanskrit retroflex consonants have been represented with one dot below the letter. Thus, t̤ and s̤ are Arabic velars, and ṭ is an Indic retroflex.

Acknowledgements – I am grateful to Dr Pierre-Sylvain Filliozat and Dr Vasundhara Filliozat for their initial willingness to share information in response to a stranger’s email, and their subsequent encouragement over the years. Many of the ideas expressed here have evolved through email discussions especially with Dr Vasundhara Filliozat. Anuradha Naimpally, an artiste well-known to dance-lovers in Austin, Texas, shared with me her knowledge of Indian dance. Dr Katherine Butler Brown, a musicologist now at King’s College (London) sent me her PhD dissertation, and it has proved to be an invaluable resource in my study. Dr Yvette-Claire Rosser, also of Austin, Texas, kindly brought me books and manuscripts relating to music, from her study-trips to Pakistan and Bangladesh.
References / Footnotes

1. A Millennium of Classical Persian Poetry, Wheeler M. Thackston, Bethesda, Maryland, 1994, Page 50.

2. Ḥaẓrat Amīr Khusro kā ‘ilm-e-musīqī, Rashid Malik, Lahore, 1975, pp. 198-199.

3. Sitar technique in Nibaddh forms, Stephen Slawek, Page 6. (Slawek rejects the claim, for it is lacking in substantiation, but notes that the legend is persistent.)

4. Music In India: The Classical Traditions, Bonnie C. Wade, New Jersey, pp. 135-136.

5. The Tillānā music of Bharat Natyam is regarded as a genre derived from the tarāna. The word “Tillānā” is said to be a derivative of tarāna.

6. Mirza here refers to the Rāg Darpan by Faqirullah Saif Khan, a work begun in 1662/1663 and completed in 1666.

7. The work referred to here is the Mān-Kutūhal, a work dedicated to Raja Man Singh of Gwalior (r. 1486-1517).

8. The Life and Works of Amir Khusrau, Mohammad Wahid Mirza, Idarah-i Adabiyat-i Delli, Delhi 1974 (1935), p. 238. (The words “mujīr”, “sāzgārī” and “mun‘am” are misspelled — they should have been “Muḥayyir”, “Sāzgīrī” and “Ghanam” respectively.)

9. Amir Khusrau’s Contribution to the Indus-Muslim Music, S. Qudratullah Fatimi, Pakistan National Council of the Arts, Islamabad, 1975, p. 21.

10. Ibid., p. 239.

11. Ibid.

12. The ‘Āin-i Akbarī says that the “Ardhāvaj is half of an āvaj”. (Page 166, Excerpt related to music from the ‘Āin-i Akbarī, reproduced in Barr-i S̤aghīr meñ mūsīqī ke Fārsī ma‘ākhiz, editor, Rashīd Malik; translated and annotated by Khvājah Hamīd Yazdānī, Lahore, Idārah-i Taḥqiqāt-i Pākistān, 1983.)

13. Pakhawaj & Tabla: History, Schools and Traditions, Aban E. Mistry, Mumbai, page 157.

14. Ṣūfyānā musīqī: the classical music of Kashmir, Józef M. Pacholczyk, 1996, Berlin, p. 34. Pacholczyk points out that the Kashmiri dokra is simply the Hindustani tablā brought in from Punjab.

15. Loghatnâmeh, Ali Akbar Dehkhoda, edited by Mohammad Moin and Dj. Shahidy, University of Tehran, NS 133, Lettre T, Fascicule 10, Oct 1967 (Mehr 1346), p. 184.

16. Music In India: The Classical Traditions, Bonnie C. Wade, 1979, p. 241.

17. Ibid., p. 177

18. Ibid., p. 204.

19. Amir Khusrau’s Contribution to the Indus-Muslim Music, S. Qudratullah Fatimi, Pakistan National Council of the Arts, Islamabad, 1975, pp. 15-16

20. In this sense, the word has entered Urdu as in the expression “Qaumī Tarāna” (national anthem).

21. From the text, it is not possible to decide whether the Persian expression “yek tâl” means “one tāla” or is a translation for the word “ektāla”.

22. This fact has long been known. For example, footnote 2 of page 45, Essays of History of Indo-Pak Music, Abdul Halim, Dacca, 1962.

23. In a paper that he read at the Conference on the Music of East and West held at New Delhi in February, 1964. (Amir Khusrau’s Contribution to the Indus-Muslim Music, S. Qudratullah Fatimi, Pakistan National Council of the Arts, Islamabad, 1975, p. 17.)

24. Singh, op. cit., p. 274.

25. Anuradha Naimpally, personal communication, 8 October 2001.

26. For instance, both pre-Islamic Sanskrit texts like Bharata’s Nāṭya Śāstra and the Viṣṇudharmottara purāṇa, and post-Islamic texts like the Sangīta-Ratnākara mention the 16-syllable akṣara scheme for notating drum syllables. (Textes des Purāṇas sur la théorie musicale, Vol. I, Alain Daniélou and  N.R. Bhatt, Institut Français d’Indologie, Pondicherry, 1959, p. 157.).

27. Hindustani Music in the time of Aurangzeb, Katherine Butler Brown, Unpublished Ph.D. thesis, SOAS, 2003, p. 77.

28.  Edited and translated into Urdu by Rashid Malik, Majlis-e Taraqqi-e Adab, Lahore, 1997.

29. Malik’s Urdu translation has been rendered into English by the author of this article.

30. It is also not clear what is meant by the word maqâm. Later Indo-Persian texts such as the Kitab-i-Nauras treat the word maqâm simply as an equivalent of the word rāga/rāginī. (Kitab-i-Nauras, (ed. Nazir Ahmad), Bharat Kala Kendra, New Delhi, 1956, p. 68.)

31. Tuḥfat-ul-Hind, by Mirzā Khan Ibn Fakhr-ud-dīn Muḥammad, edited by Dr. Nurul Hasan Ansari, Intesharat-e Farhang-e Iran, Khordad 1353 (May-June 1974), Volume 5, Chapter 8, pp. 421-423. (This excerpt was translated from Persian by the author of this article.)

32. Tarjuma-i-Mānakutūhala And Risāla-i-Rāgadarpaṇa, Ed. Shahab Sarmadee, New Delhi, 1996, p. 270, footnote 94.

33.  The Wonder that was India; a survey of the history and culture of the Indian sub-continent before the coming of the Muslims, A. L. Basham, New York, 1968, 3rd rev. ed., page 230.

34. Tagore and the Music of Iran, Rajyeshwar Mitra, Indo-Iranica, June 1961, page 62. Khusrau’s Musical Compositions by Thakur Jai Deva Singh, in Life, times & works of Amir Khusrau Dehlavi (ed. Z. Ansari), p. 276, New Delhi, 1975.

35. Malik, Ḥaẓrat Amīr Khusro kā ‘ilm-e-musīqī, p. 102, pp. 236-237.

36. The Urdu sitar manual Qānūn-i sitār, (Sayyid Safdar Husain Khan, Munshi Naval Kishor Press, Lucknow, 1873), for instance, mentions only the rāgas Sāzgiri and Sarparda. The most influential modern Urdu work on Indian music Ma‘ārifunnaghmāt (Thakur Nawab Ali Khan, 1873) mentions only Sāzgiri and Aiman.