Chapter 2 The Social and Intellectual Contexts

In: The Books Sānk and Pātanğal
Noémie Verdon
Search for other papers by Noémie Verdon in
Current site
Google Scholar
Open Access

2.1 Building up Theoretical Knowledge on al-Hind

Our journey toward al-Bīrūnī’s encounter with al-Hind necessarily leads us to examine the ways in which became acquainted with its language, culture, philosophies and sciences. In his writings available to me, al-Bīrūnī does not use the term “Sanskrit.” He occasionally uses the expression “in the Indian language” (fī l-luġa al-hindiyya; ‮في الغة الهنديّة‬‎), and generally employs the word al-hind (‮الهند‬‎) as a collective term to designate “India” or to refer to “what is Indian.” He also employs the adjective derived from it, al-hindī (‮الهندي‬‎) meaning “Indian.” The accuracy of his Arabic transliterations in Al-āṯār al-bāqiya and Taḥqīq mā li-l-Hind, however, leave little doubt that he had knowledge of the classical Sanskrit lexicon related to astronomical science, literature, geography, philosophy and religion. In the Taḥqīq mā li-l-Hind, al-Bīrūnī also comments on the grammatical, semantic and phonological complexities of the literary Indian language, namely Sanskrit, the religious and scientific language at the time.1

In the year 1000, al-Bīrūnī dedicates Al-āṯār al-bāqiya (The Chronology of Ancient Nations) to Prince Qābūs of Ğūrğān.2 This treatise describes the astronomical calendars of different civilizations, explaining various methods by which their societies calculate days and nights, months and years, as well as longer eras. It also presents festivals linked to different calendars and covers historical events. The main civilizations considered in Al-āṯār al-bāqiya are those of Persians, Sogdians, Khwarizmians, Greeks, Jews, Syrians, Christians (Nestorians and Melkites), Zoroastrians (or Magians), Sabians, Arabs before Islam and Muslims. This work further includes information regarding India, especially terminology related to astronomy and the calendrical system.

An examination of the examples drawn from Al-āṯār al-bāqiya below reveals that al-Bīrūnī’s knowledge of Indian astronomical terminology was relatively accurate before he visited al-Hind. His proficiency in Sanskrit terminology is evident in how he provides transliterated Sanskrit names of the months, seven heavenly bodies and zodiacal signs in Arabic, as displayed in the following three tables:

Table 2

Names of the months in Sanskrit, as transliterated into Arabic by al-Bīrūnī in Al-āṯār al-bāqiya3

































mārgaśīrṣa, also mārga


















Table 3

Names of the seven planets in Sanskrit, as transliterated into Arabic in Al-āṯār al-bāqiya4


















(the Sun)












(the Moon)

Table 4

Names of the zodiacal signs in Sanskrit, as transliterated into Arabic by al-Bīrūnī in Al-āṯār al-bāqiya5



















































Al-Bīrūnī provides Arabic transliterations of the names of months, planets and zodiacal signs that in general correspond to their Sanskrit counterparts. Only two transcriptions significantly differ from their Sanskrit originals: mankis for mārgaśīrṣa and bākr for phālguna.6 Further, in two passages below, al-Bīrūnī discusses Indian astronomical methods to divide the celestial globe:

We say that the Indians divide the celestial globe by the number of lunar mansions (‮منازل‬‎), that is, twenty-seven for them. Accordingly, the [celestial globe] is divided by this number. Each mansion approximately amounts to thirteen degrees and a fourth. They draw [their] rules (‮احكام‬‎) from the stars’ entrances in their ribāṭāts (‮رباطات‬‎);7 this [process] is generally known as ğufūr (‮جفور‬‎)8 [and] it is applied [by them] to every single situation and need. Its report requires a long discourse foreign to [our] purpose and found in the books of sciences known by this [name] (i.e., ğufūr). […] The [Arabs] used the [notion of the lunar mansions] in a different manner than the Indians, as they intended to learn the conditions of the atmosphere at [different] times and weather phenomena according to the seasons of the year by [using] them. They were illiterate people, unable to [have] knowledge, except for visible [things].9

[…] This is a testimony of Abū Maʿšar showing that this method leads to correct results.10 Further, if examined by the Indian way of the ribāṭāt and the ğufūr, the matter would approach the correct [result].11

These passages show the scope of al-Bīrūnī’s knowledge on the topic and reflect his early interest in Indian astronomy. They also indicate the type of information which al-Bīrūnī had about Indian sciences before composing the Taḥqīq mā li-l-Hind. In the first extract, al-Bīrūnī explains—and shows appreciation for—the Indian lunar mansions and other related astronomical concepts. In the second quotation, he recognizes the use of the Indian concepts mentioned in the first quotation, that is, the ribāṭāt (stations) and the ğufūr, in order to obtain relatively accurate results on the times of the rising and setting of the lunar mansions. In the second half of the eighth century CE, at the Abbasid court, astronomical and medical treatises had been translated from Sanskrit into Arabic, as aforementioned in the introduction of the present book. Thus, Indian astronomy, which was known to Muslim thinkers for two centuries before al-Bīrūnī composed his Al-āṯār al-bāqiya, enjoyed prestige, and al-Bīrūnī was certainly indebted to this tradition.

Furthermore, several Indian siddhānta texts were among the works that had been translated into Arabic during the Abbasid caliphate.12 Some of these translations were known to al-Bīrūnī, as the following five excerpts show:

According to Ptolemy, the revolutions [of the sun] are equal, as he did not find that the apogee of the sun has any movement. For others than him, I mean the authors of the Sindhind and modern thinkers, they are unequal, because their observations led them to [make conclusions] about the existence of movement [of the sun]. However, whether equal or unequal, [the revolutions] encompass the four seasons and control their nature.13

These [cycles of the stars determined by Abū Maʿšar] differ from the cycles determined by the observations of the Indians, known as the Cycles of the Sindhind (‮السندهند‬‎ ‮ادوار‬‎), [and likewise] differ from the Days of Āryabhaṭa (‮الارجبهز‬‎ ‮ايّام‬‎) and the Days of al-Arkand (‮الاركند‬‎ ‮ايّام‬‎).14

The discrepancy in the cycles [of the stars], not the discrepancy in the observations, is a sufficient argument and stronger evidence in refuting what Abū Maʿšar perpetrated. Stupid [people] rely upon him, discrediting religion and making the Cycles of the Sindhind, or others, a means to abuse those who warn about the approach of the [ultimate] hour and who inform them on the gathering [of the dead] for the reward and punishment in the everlasting abode.15

As for the day of [the vernal] equinox, the Indians calculate it with their astronomical handbook (zīğ; ‮زيج‬‎), which they say, with ignorance, is eternally ancient and that other astronomical handbooks make use of it. Their Nowrūz (‮نوروز‬‎) (i.e., the Persian New Year celebration occurring at the vernal equinox) is a great festival for them. During the first hour of this [day], they worship the sun and pray [to the] spirits for happiness and bliss. At noon, they worship the [sun] and pray for the life to come and the beyond. At the end of the day, they worship the [sun] and pray for their bodies and health. During that [day], they worship every object of value and [every] living creature. They say that during that [day], the winds blow great auspicious spiritual beings. The people of heaven and hell look at each other with affection. Light and darkness are in equilibrium. During the hour of [the equinox], fire is burnt in sacred places.16

According to the astronomical handbook, the Sindhind, the second equinox is a great festival for the Indians, similar as Mihrğān (‮مهرجان‬‎) for the Persians. During that [day], they exchange all important goods and delicate jewels. They gather in the temples and places of worship at noon. Then they go out in their parks, they gather in their public places and bow to their [god of] Time and show obedience to Allah17—respected and exalted be He.18

The above passages reveal that al-Bīrūnī’s knowledge of India at the time of the Al-āṯār al-bāqiya’s composition, in the year 1000, was largely based on literary sources. He quotes and refers to writings on topics such as the astronomical revolution of the sun, cycles of the stars, the vernal and autumnal equinox or rituals performed on certain days. The Arabic term zīğ, translated here as “astronomical handbook,” was a generic appellation for a type of handbook which regrouped astronomical tables and explanatory material. The Zīğ al-Sindhind refers to al-Ḵwarizmī’s work on Indian astronomy, while the Days of Āryabhaṭa and of al-Arkand are Arabic works based on other Sanskrit astronomical treatises.19 These books were available to al-Bīrūnī who drew upon them when discussing Indian astronomy.

Arabic sources that were no translations nor interpretations of Sanskrit works also played a part in al-Bīrūnī’s account on India in Al-āṯār al-bāqiya, as the following three passages illustrate:

I have heard that the Indians use the appearance of the new moon for determining the months. They intercalate one lunar month to every 976 days. […] Abū Muḥammad al-Nāʾib al-Āmulī reported in the Kitāb al-ġurra, based on the work of Yaʿqūb Ibn Ṭāriq, that the Indians consider four types of periods. One of them is the revolution of the sun [starting] from a point that consists in a star in the constellations, [and returning] to its source. This is a solar year. The second is the rising of [the sun] 360 times. It is called the middle year, because it is longer than the lunar year and shorter than the solar year. The third corresponds to twelve revolutions of the moon [starting] from the two [stars, called] al-Šaraṭāni (‮الشرطان‬‎), which are both at the top of the Aries [constellation] (‮الحمل‬‎), and [returning] to the two [same points]. This is their lunar year. It approximately amounts to 327 days, seven hours and two thirds. The fourth is the appearance [of the moon] twelve times. It is the lunar year [commonly] in use.20

The author of the Kitāb maʾḵaḏ al-mawāqīt (i.e., Book on the Method for Determining Times) claimed that those [who] follow [the calendar] with the intercalated fourth [day], namely the Greeks21 and others, established the sun’s entrance into the constellation of Aries in the beginning of April, which is Naysān (‮نيسان‬‎) for the Syrians, as the beginning of their era. […] Further, he [said], speaking about the Greeks, that after they understood that the beginning of their year had changed its place, they had recourse to the years of the Indians and intercalated in their year the additional [day] between two years. […] He (i.e., the author of the above-mentioned book) assimilated the differences between the Greek year and the solar year in the manner the Indians [did].22

Al-Ğayhānnī reported that, at the Indian Ocean, roots of a tree spread along the seacoast in the sand, that [its] leaf rolls up before separating from its root, and that [the leaf] changes into a male bee and flies away.23

These three extracts point to some of al-Bīrūnī’s Arabic sources which informed him about India at an early date.24 First, al-Bīrūnī quotes Abū Muḥammad al-Nāʾib al-Āmulī (Kitāb al-ġurra) who refers to Yaʿqūb Ibn Ṭāriq in order to describe four different types of astronomical years in use among Indians.25 In the next passage, he refers to the Kitāb maʾḵaḏ al-mawāqīt26 without naming the author of this book, in order to point out different manners of calculating days and years. In the last excerpt provided above, al-Bīrūnī quotes al-Ğayhānnī27 to depict a tree found on the coast of Indian Ocean endowed with fantastic characteristics. The use of the expression “I have heard that the Indians […]” (‮سمعت أنّ الهند‬‎)28 in the first of these excerpts also suggests that information was transmitted orally to him.

In conclusion, all these examples show that al-Bīrūnī had material regarding Indian astronomy at his disposal before the year 1000 CE and illustrate his respect for it.29 Based on the above, I argue that al-Bīrūnī accessed oral and written sources for his account of India in Al-āṯār al-bāqiya¸ while he chiefly based his report on written documents. These writings were available to him in Khwarezm, Ray or Ğūrğān, that is, before he travelled eastward, came in contact with Indian thinkers, and veritably entered the territory of al-Hind.

Furthermore, as I aim to show below, by the time he had written the Taḥqīq mā li-l-Hind, al-Bīrūnī’s knowledge of Sanskrit had increased considerably.30 According to Sachau, al-Bīrūnī had gained good proficiency in the lexicon and grammar of Sanskrit when he composed this book. Sachau also considers that al-Bīrūnī’s work is the result of both his endeavours to understand Sanskrit and his collaboration with Indian thinkers.31 David Pingree, however, contends that al-Bīrūnī was not highly skilled in Sanskrit and that his translation of the Sanskrit astronomical treatise Brāhmasphuṭasiddhānta relied, for the most part, upon the Indian pandits whom he met and who misled him in his interpretation.32

Nevertheless, al-Bīrūnī’s faithful transfer of Sanskrit terms into Arabic in his works provides evidence for his good command of the Sanskrit lexicon related to Indian astronomy, literature, philosophy, geography and religion. His transliterations in Al-āṯār al-bāqiya and the Taḥqīq mā li-l-Hind are mostly true to the corresponding Sanskrit words. The above table displays a random sample of Sanskrit terms transliterated into Arabic drawn from the table of contents of the Taḥqīq mā li-l-Hind as printed in the Hyderabad edition.33

Table 5

Transliterations from Sanskrit to Arabic by al-Bīrūnī in the Taḥqīq mā li-l-Hind34





bīḏa (‮بيذ‬‎)


nārāyan (‮ناراين‬‎)


purānāt (‮پرانات‬‎)

purāṇāḥ (pl.)

bāsudīwa (‮باسديو‬‎)


mīrū (‮ميرو‬‎)


bhārata (‮بهارت‬‎),

bhārāṯa (‮بهاراث‬‎)


dībāt (‮ديبات‬‎)

dvīpāḥ (pl.)

akšauhinī (‮اكشوهنى‬‎)


lanka (‮لنك‬‎)

laṅkā (f.)

adimāsah (‮ادماسه‬‎)


māna (‮مان‬‎)


ūnarātra (‮اونراتر‬‎)


brahmānda (‮برهماند‬‎)


ahargana (‮أهرگن‬‎)


kalpa (‮كلپ‬‎)


parba (‮پرب‬‎)


catur jūga (‮چترجوك‬‎)


sanbajjara (‮سنبجر‬‎)


In this table, most of the long vowels in Sanskrit are rendered with long vowels in Arabic. The nasal (guttural) and (retroflex) are generally reproduced by the same letter nun in Arabic, as it is the sole (dental) nasal of the Arabic alphabet. The Arabic letters bā, fā and wāw are each employed at different times to transliterate the Sanskrit sound v, which does not exist in Arabic. In other cases where the Arabic language did not have sounds specific to Sanskrit, al-Bīrūnī inserted Persian characters, such as ch, g and p, into the Arabic script. He generally rendered the Indic phoneme e into ī when writing in Arabic.

Based on the renderings of Sanskrit terms into the Semitic alphabet in the Taḥqīq mā li-l-Hind, Sachau suggests that al-Bīrūnī knew several vernacular languages alongside Sanskrit and that his transliterations of Indic terms reflect this plurality.35 A recent study of al-Bīrūnī’s rendering of terms related to Indian religious festivals by the present author showed a relation between his various ways of transliterating Sanskrit terms into Arabic and the two types of sources, textual and oral, that he used to collect information on India. Al-Bīrūnī tended to transfer words that he had heard in an abbreviated form close to vernacular languages such as Panjabi and Sindhi, whereas he rendered words that he knew from literary sources in a way that is more faithful to the classical Sanskrit terms.36

Lastly, a comparison between al-Bīrūnī’s use of Sanskrit terminology in Al-āṯār al-bāqiya and in the Taḥqīq mā li-l-Hind points to the evolution of his knowledge of this language. In the former, al-Bīrūnī’s knowledge of Sanskrit terminology was largely confined to the astronomical field and based on literature. In the latter, his field of expertise, based then on literature and on oral transmission, expands to other areas of knowledge. Al-Bīrūnī indeed quotes Sanskrit works belonging to various literary genres and scientific domains, such as the Purāṇas, the Kitāb Gītā, two texts related to Sāṅkhya and Yoga, and to a considerable amount of astronomical literature.37 Al-Bīrūnī’s degree of proficiency in Sanskrit can also be appreciated by the two texts he enumerates in his auto-bibliography that are interpretations from Arabic into Sanskrit of Euclid’s Elements, of Ptolemy’s Almagest and of a book on astrolabes.38

Thus, I highlighted developments in al-Bīrūnī’s knowledge of Sanskrit and of Indian sciences that occurred from the time of the composition of Al-āṯār al-bāqiya up to that of the Taḥqīq mā li-l-Hind, an evolution that necessitated a long-term collaborative work with Indian thinkers. Therefore, I next explore the role of intercultural and intellectual exchanges taking place in the Ghaznavid royal court to which al-Bīrūnī belonged.

2.2 Collaborations and Multiculturalism at Royal Courts

Socio-historical elements help retrace the manner in which al-Bīrūnī learned Sanskrit so that he could eventually acquire the ability to translate two works related to Sāṅkhya and Yoga philosophy into Arabic. Although al-Bīrūnī remains an isolated historical figure of his period known to have conducted in-depth research on al-Hind, the context of royal courts of the time, and that of the Ghaznavids in particular, played a crucial role in his work. The existence of various instances of intellectuals working together on translations demonstrates that this was common practice from an early date. The Bayt al-Ḥikma (House of Wisdom) of Baghdad, an academy where philosophers, translators, secretaries, clergymen, copyists, librarians and astronomers worked together, is just one example of these early collaborations.39 Another instance centers around the work of Ḥunayn Ibn Isḥāq (b. 808) who explains how after he had translated Galen’s De motu muscolorum into Syriac, another scholar translated it into Arabic.40 As Travis Zadeh notes, Ḥunayn’s explanation illustrates a “professional process of translation,”41 as well as the necessity of teamwork in such processes.

The Marvels of India (‮عجايب الهند‬‎) is an example of a literary work that illustrates early intercultural exchanges and points to the existence of polyglotism. Authored by Buzurg Ibn Šahriyār in the mid-tenth century CE, it gathers 134 whimsical travellers’ tales.42 This book not only exemplifies how information was propagated throughout different regions connected by the Arabic Sea, the Indian Ocean and the Mediterranean Sea, but also demonstrates how sailors and other travellers used multiple languages to communicate. One report narrates how an Indian king in a region of Kashmir wanted to have the laws of Islam translated and requested a person from Iraq, who lived in India and knew several of its languages, to come to his court.

Incidentally, the same Indian king asked Buzurg Ibn Šahriyār to translate the Quran into his own Indian language.43 Another story tells of a person from Siraf, in present-day Southern Iran, who travelled with an Indian guide. According to this story, the two travellers were able to converse, although no information regarding the language they used is provided.44 While the report in The Marvels of India may not be completely historically reliable, it at least reveals the existence of polyglot persons able to speak different languages, most probably Arabic, Persian and some Indic languages, in the mid-tenth century.

In a context closer to al-Bīrūnī’s, the Persian translation of al-ʿUtbī’s Al-Yamīnī describes Maḥmūd’s army as composed of many ethnic groups, including Indians, who may have spoken various languages.45 In this period, it was common for the Muslim nobility to have foreign servants, referred to as ġulāms (slaves, servants, young man).46 Indian ġulāms were regularly seized during military campaigns and some, after spending time serving their new masters, appear to have been assigned to relatively high-ranking positions at the Ghaznavid court. One case is that of Tilak, an Indian ġulām and son of a coppersmith, who became a military leader after having been an official interpreter for Maḥmūd’s son in his administration. Bayhaqī reports that Tilak was known as having “a good hand for writing both the Indian (hendavi, hendui) and Persian languages.”47

This example illustrates that the Ghaznavids included different ethnic and linguistic groups in their army and administration, that they appointed people of Indian origin to higher positions, and that they needed interpreters for Indic languages and Persian. Thus, as part of cultural and ethnic encounters, languages were exchanged at the Ghaznavid royal court. In addition, the practice of promoting former ġulāms to higher positions provides a valuable clue for the need of the Ghaznavids to employ Indian interpreters, some of whom may have been Brahmins proficient in Sanskrit, to help them govern their newly conquered territories and communicate with the population.48

Regarding bilingualism and polyglotism, further evidence points to the use of different languages in Islamic royal courts of early medieval times. Rehman calls attention to an epigraph dated to 1011 and inscribed on a foundation stone of a tomb found in Zalamkot in the lower Swāt. This epigraph, which dates to the reign of Maḥmūd, bears a bilingual inscription in Persian (seven lines) and Sanskrit (three lines in Śāradā script), indicating an early interest in writing records in these two languages.49 In this official document, Persian was used next to Sanskrit, in contrast to the Arabic legend of the bilingual coins minted in the region of Lahore.50

Further, intellectuals, alongside ideas and languages, travelled beyond cultural borders amid military and commercial interests of the rulers, thereby enabling early intercultural exchanges. As illustrations of intellectuals’ moving through cultural and linguistic boundaries during this period, I provide a few additional examples. Al-Bīrūnī refers to an Indian physician who travelled in the region of Gardez, between Ghazna and the area of Panjab now located in Pakistan, in his introduction to the Pharmacology (Kitāb al-ṣaydana fī l-ṭibb) composed at the end of his life.51 Not many years after al-Bīrūnī had composed the Taḥqīq mā li-l-Hind, the Persian Sufi Ali Huğwīri, born in Ghazna, spent the latter part of his life in Lahore and died there in 1071/1072.52 The poet Masʿūd-I Saʿd-I Salmān (1046/1049–1121/1122), who was of Persian origin and lived in Lahore, is said to have composed poems in Persian, Arabic and Indic languages. Only his Persian poems are however extant.53

Thus, the context of the Ghaznavid court was rich in intercultural and intellectual exchanges, while polyglotism was probably more common at the time than generally assumed. These above examples also support Finbarr Barry Flood’s statement that “[b]ilingualism and/or polyglossia may in fact have been relatively common phenomena of the South Asian borderlands.”54

Thus, this context favoured al-Bīrūnī’s constant development of knowledge and, in this case, his learning of Sanskrit and of Indian sciences thanks to collaborations with Indian thinkers. The existence of intermediary languages known by the involved parties, including Sindhi and Panjabi, as Carl Edward Sachau and Suniti Kumar Chatterji have noted, but also Persian made these interactions possible.55 As the available evidence suggests, al-Bīrūnī met people from Multan and surely spent some time in Panjab56 which enabled him to become acquainted with some vernacular languages, in all likelihood a form of Sindhi and of Panjabi. Persian, which belongs to the Indo-Iranian linguistic family, became the official language of Islamic royal courts at the time and would also serve as an intermediary language in these intercultural exchanges.

According to Houari Touati, Muslim travellers did not leave their native countries solely out of curiosity, but their journeys were often first motivated by governmental interests (embassies, conveyance of messages, administrative organisation in all regions of the empire).57 The testimony of al-Balāḏurī (d. 892 CE) furnishes an example of how a ruler appointed somebody to visit India in order to gather information about this foreign land.58 This report indicates that Muslim rulers showed interest in Indian countries at an early date for commercial, administrative or military reasons. Similarly, Maḥmūd benefited from al-Bīrūnī’s skills in Indic languages in his conquests in the East. Further, al-Bīrūnī’s wish to access Indian science, an interest that was part of an existing tradition among his peers is evident from a passage found in the preface to the Taḥqīq mā li-l-Hind, in which he explains that the composition of the work was requested by another learned man.59

Al-Bīrūnī’s position at Maḥmūd’s court was thus conducive for him to learn Sanskrit and study Indian sciences, religion and philosophies. He came in contact with Indians thanks to Maḥmūd’s military and commercial interests. Maḥmūd is also known to have sought to gather scientific writings in Ghazna, for instance, from the cities of Ray and Isfahan in Iran,60 and to have requested a considerable number of scholars and poets to come to his court.61 As I suggested above in Section 1.2.3, numerous people accompanied the sultan during his campaigns: soldiers, workers, officials, poets, secretaries, interpreters, etc. Al-Bīrūnī records in his book on gemmology, Al-ğamāhir fī l-ğawāhir (The Collection of Gemstones), that the encounter with the ambassadors of the Kʾitan dynasty provided him with information on the Far East.62 Farruḵī’s poems also offers information about the life of the sultan, who received delegates and military leaders from foreign states.63 It is likely that there were members of the Indian elite among these delegates, such as royal advisors, astronomers or officials, who were educated Brahmins. Maḥmūd’s court thus helped thinkers access various resources, whether written documents or scholarly interactions. This context enabled al-Bīrūnī to complement his—rather theoretical—knowledge of India preceding his actual travels in al-Hind, with a practical approach of Indian languages and sciences.

2.3 Al-Bīrūnī and Indian Scholars

After having highlighted how important were the socio-cultural context and the collaboration work in al-Bīrūnī’s intellectual project, I examine his interactions with Indian thinkers with the aim of specifying the identity of his interlocutors. First, the Taḥqīq mā li-l-Hind reveals that Brahmins were essential informants to his project. The two following passages drawn from this work are direct evidence of al-Bīrūnī’s interactions with Brahmins:

I saw [some] Brahmins who allowed to their table companions, [who are also] relatives, [to eat] from one plate [with them], but the rest of them denied this.64

I [repeatedly] heard that when Indian slaves escape [from another land] and return to their country and religion, they are forced to expiate by fasting, then they are soaked in cow’s dung, urine and milk for a certain number of days, until they become mature there. Then, they go out of the dirt, and they eat additional similar things. I asked the Brahmins about this, but they denied it, pretended that there is no expiation for these [Indians], and that they are not allowed to return to their previous situation.65

As aforementioned in Section 1.2.4, al-Bīrūnī devotes an entire chapter to the life and practices of Brahmins, whereas he portrays the other classes of the society in one chapter.66 Further, he describes the four varṇas (lit. colours) and the classes that are outside the caste system in a chapter entitled “On the classes called colours and on the [classes] which are lower [than them].”67 In general, the Taḥqīq mā li-l-Hind stands as evidence of a Hindu society organized according to the so-called caste system and following a Brahminical model.68 Another passage of the Taḥqīq mā li-l-Hind is instructive on the type of society al-Bīrūnī describes:

Their [religion] chiefly [revolves] around the Brahmins, as the [latter] are trained to preserve and maintain it.69

This passage is found in a chapter entitled “On their belief in the existents, both intelligible and sensible.”70 It presents Brahmins as the custodians of the religion of the society encountered by al-Bīrūnī. Al-Bīrūnī’s informants conveyed to him the picture of a society in which not only the caste system prevailed, but also the authority of the Brahmins. Further, the latter constituted the literate class of the population, as advisers, interpreters or astronomers/astrologers at the kings’ courts. The Indian rulers, accompanied by their courts, formed the sector of the society that was the most likely to enter into contact with their Islamic counterparts. Therefore, it is no wonder that al-Bīrūnī, who held a position at the Ghaznavid court, collaborated with Brahmins, rather than with people at any other level of the society, such as soldiers or peasants.71

The question arises then of the field of expertise of these Brahmins: where they astronomers, priests, philosophers, or these several things at the same time? First elements of answer regarding the exact nature of al-Bīrūnī’s intellectual encounters with Indians lie in his writings:

[At first,] I stood among their astronomers in the position of a student with [his] master, because of my difficulty in speech (‮لعجمتى‬‎) [ignorant of their language] among them and my shortcomings about what they were [involved] in, such as their conventions.72 When I had made some progress in these [matters], I began instructing them on the defects [of their conventions],73 pointing out elements of demonstrations and correct methods of arithmetic. They swarmed around me, from all parts, being astonished [and eager] to learn [from me, and] while crowding in, they asked: “With whom among the Indians did you stay to acquire [this knowledge]?”74

Some of al-Bīrūnī’s informants were thus astronomers. His particular interest in astronomical science and the need of the Indian rulers to be advised by astronomers and astrologers may account for this fact. As for the connection between religion and the science of the stars, the extract below is informative:

The science of the stars is most well-known among them (i.e., the Indians), because matters of religion are dependent on it. Those among them who do not know how to make judgments [based on the stars] (i.e., astrology) see the [fundamental] characteristic of astrology as nothing but arithmetic.75

This connection is particularly relevant to the present discussion, as it opens the possibility that religious officials were astrologers/astronomers and suggests that al-Bīrūnī met such Indians expert in several domains.76 In addition, al-Bīrūnī also met Indian philosophers who assisted him in his interpretations of the Sāṅkhya and Yoga texts, as shown in Chapters 4, 5 and 6. Interestingly, Gardīzī, a writer contemporaneous to al-Bīrūnī also attached to the Ghaznavid court, enumerates Indian communities in an account based on al-Ğayhānnī’s report. He describes one group of persons as including thinkers who combined the astronomical and medical sciences in their work and designates them as “philosophers” (Persian: khudāvandān-i andīsha; “masters of thought”).77

Although the semantic field of this term is vast, Gardīzī’s description suggests that Indian thinkers specialised in astronomy and medicine also cultivated the science of thought, in all likelihood including reasoning, logic and metaphysics. Gardīzī further describes these thinkers as eating “dates, plants and herbs so that it should be light for their senses” (Minorsky 1948: 633). This diet and the reason for adopting it reminds one of the lifestyle promoted by ascetics (e.g., yogis) and orthodox Brahmins. Gardīzī’s report may thus support the hypothesis that the Brahmins who were al-Bīrūnī’s informants were versed in astronomy and philosophy.

Moreover, in the Taḥqīq mā li-l-Hind, al-Bīrūnī constantly distinguishes between the views of educated and uneducated people, referred by him as the elites (‮الجاصّ‬‎) and the masses (‮العامّ‬‎) respectively. He generally approves of the intellectual and religious attitudes of the elites, who, in his view, are able to consider abstract notions and whose conceptualization of the divine can be compared to the monotheism of Islam. He described the masses, on the contrary, as idolatrous people. Al-Bīrūnī certainly interacted with members of the elite, as his comments in the preface to the Kitāb Pātanğal confirm:

I had not ceased to translate books by arithmeticians and astronomers from the Indian [language into Arabic], until I turned to books on wisdom (‮في الحكمة‬‎) preserved by their elite (‮خواصّهم‬‎), and about which the ascetics compete [with each other] concerning the paths [leading] to devotion. When they were read to me letter by letter, and when I grasped their content, my conscience could not overlook [the occasion] to share [my knowledge] with those who wish to study these [books].78

In another excerpt drawn from the Taḥqīq mā li-l-Hind which presents the views of several Purāṇas concerning the names of the planets, al-Bīrūnī comments on those who assisted him in understanding these texts:

Those who explained to me [these texts] by way of translation were well versed in the language and were not known to be treacherous.79

The above two passages illustrate how educated people, whom al-Bīrūnī considered reliable, helped him study philosophical and purāṇic Sanskrit literature. The elite and the people “well versed in the language”, whom al-Bīrūnī refers to, likely belonged to the class of literate Brahmins. These two above passages also confirm two observations made earlier in this book: 1) al-Bīrūnī was originally interested in mathematics and astronomy, and then turned to other sciences; and 2) he did not translate his Sanskrit sources into Arabic all by himself, but in collaboration with Indians who orally conveyed to him their readings of these sources, probably using an intermediary language.

The prevalence of certain types of Indian literature, as opposed to others, reflects the intellectual and religious inclinations of al-Bīrūnī’s informants. As seen earlier in Section 1.2.4, the Sanskrit works which al-Bīrūnī mentions or quotes in his book on India mostly belong to a Brahminical scientific and religious literature.80

The material presented above, in my view, strongly suggests that among the Brahmins al-Bīrūnī met, there were astronomers/astrologers, religious officers and philosophers. If so, Brahmins and advisers worked in several capacities for the rulers of royal courts of north-western early medieval India, including that of the Late Shahi kings.

Further, as I show in Chapters 4, 5 and 6, al-Bīrūnī reached an excellent proficiency of the Sāṅkhya and Yoga philosophies. In addition, I have already suggested in-passing that al-Bīrūnī had to engage in relatively long-term collaborations with Indian scholars in order to achieve such high level of knowledge about Sanskrit and Indian sciences, and to produce his numerous translations from Sanskrit to Arabic. I thus conclude this section with a discussion on the types of exchanges between the Ghaznavid court and that of local Indian rulers that occurred then, and on the possible geographical sites where a long-term collaboration would have taken place. In addition, the examination of these two questions offers an overview into some specifics of intercultural interactions of the time.

First, it is hard to fathom whether from the Indian side this collaboration occurred willingly or due to force. Abdur Rehman discusses how the relationship of the Late Shahi king Ānandapāla and his son Trilocanapāla with the Ghaznavid rulers fluctuated between being uncivil and amicable.81 Other kings, such as the ruler of Nārāyaṇapura, surrendered and spontaneously offered to pay tribute to the Ghaznavids.82 The existence of such peaceful tributes—discontinuous, however, in the case of the Late Shahis—point to the possibility of a dialogue between the two political spheres, which would have facilitated a dialogue between intellectual spheres.

Regarding the possibles places where al-Bīrūnī actually collaborated with Indians, evidence is scarce. However, both Kābul and Ghazna, located at the crossroad of the Islamic and Hindu worlds hosted books and prisoners brought from India. They may have thus served as centers of intellectual exchanges. Fort Nagarkot, situated in present-day north-western India, appears to have housed silk manuscripts and to have been a place where knowledge was preserved. Carl Ernst recalls that Muslim armies led by Sultan Firuz Ibn Tuqhluq looted some temples located near Fort Nagarkot in the mid-fourteenth century and that they collected a large quantity of books there.83 In a similar way, Maḥmūd could have gathered books from his raids in temples and fortresses of al-Hind.

As for other sites located in the territory of al-Hind, the lack of available data concerning the sites of Laghmān, Peshawar, Fort Rājagirī and Fort Lahūr, locales visited by al-Bīrūnī in al-Hind, prevents us from determining their significance in his collaboration with Indians. Wayhind/Udabhāṇḍa was an important Late Shahi site located on the way of the Ghaznavid military conquests. Very little is known about this town, because much of the ancient site is at present submerged in the Indus River, or under the present-day town of Hund.84 According to primary sources, Maḥmūd stayed in the region of Peshawar from September 1001 to April 1002.85

As for Lahore, it became the capital city of the Late Shahi kingdom, after Udabhāṇḍa and Nandana, and before they took shelter in Kashmir.86 Indian scholars likely dwelt in Udabhāṇḍa and Lahore. The two cities were located on a road often used by Maḥmūd for his military excursions in al-Hind, which al-Bīrūnī describes, and which opened new territory to the Ghaznavids. However, al-Bīrūnī does not mention any observations or calculations he might have made in these two cities. It is thus possible, although not ascertainable, that al-Bīrūnī collaborated with Indians in both Udabhāṇḍa and Lahore.

Multan was an equally important city in al-Hind at the time, as al-Bīrūnī’s many references to it suggest. He explains that different appellations were given to this city, describing it as a place of pilgrimage on account of its pond and Sun idol.87 Al-Bīrūnī communicated with people from Multan and consulted books by authors from this city. Already in the mid-eighth century, scholars from Sind played a role in the translation of Sanskrit astronomical works into Arabic.88 It is likely that 250 years later, when al-Bīrūnī wrote about India, the region still hosted Indian scholars who may have interacted with him. In addition, al-Bīrūnī says that he himself calculated the latitude of Multan.89

However, in contrast with the five locales discussed in Section 1.3.1, al-Bīrūnī does not explicitly express his presence at Multan. In addition, most of the other pieces of evidence point to al-Bīrūnī visiting some parts of Gandhāra and Panjab, not upper Sind. For instance, as I argued above, Multan was not located on the northern road that Maḥmūd chiefly utilized for his campaigns to the East and through which al-Bīrūnī travelled the most. Therefore, even if he visited the region, it is not certain whether he spent much time in this city. In my view, al-Bīrūnī did not visit Multan, or very briefly, despite him mentioning his calculation of its latitude. Lastly, Multan has seen Muslim Arabic speakers and local non-Arabic speakers coexisting since at least the eighth century CE. Maḥmūd was perhaps benefiting more from having an interpreter in newly conquered regions than in areas where bilingualism, or polyglotism, was already rooted.

With regard to Nandana in the Salt Range, the ruins of two important temples are found there. These temples most probably housed Indian Brahmins, along with Sanskrit texts. Traditional education and the opportunity to study may have been provided by the learned temple attendants.90 After having plundered the temples of Nandana in 1014, the Ghaznavids appointed governors at the place, which, as argued above, implies a will to strengthen their local political authority.91 In addition, al-Bīrūnī spent sufficient time in this fort to experiment with his method of calculating the circumference of the earth.92 It is thus likely that Maḥmūd appointed the scholar to stay there for some time between the years 1017 and 1030. On this occasion, priests of the temples, who may have had proficiency in astronomy and/or philosophy, would have assisted al-Bīrūnī in learning Sanskrit and studying Indian culture.93

Lastly, Nagarkot, the modern Kangra in Himachal Pradesh, which is located farther east, may have been a significant site for al-Bīrūnī’s study of Indian sciences. Maḥmūd attacked Nagarkot in Winter 1008. In the same way as in Fort Nandana and Lahore, he appointed a governor which indicates a strong desire to establish governmental authority in the place. As shown above, this fort was also a place where the chronicles of the Early and Late Shahis were stored, and thus where other texts might also have been found.

Although the above short outline may not suffice to ascertain where the collaborative work between Indians and al-Bīrūnī took place, the region of Ghazna and Kābul outside al-Hind and Nandana Fort and Nagarkot within its territory are particularly good candidates. Other sites, such as Wayhind/Udabhāṇḍa, Lahore and Multan present, in my view, less indication for a possible long-term interaction between al-Bīrūnī and Indians.

2.4 The Transmission of Living Traditions

In the preceding sections, I argued that al-Bīrūnī described a society that adopted Brahminical precepts, literature and sciences, as well as Hindu cults. I have also highlighted that he relied on his interactions with Indian thinkers in order to study Indian literature, sciences and religion. In addition, elements coming from his direct observations pertains to a specific region of al-Hind, that is, Gandhāra and Panjab, which belonged to the kingdom of the Late Shahis not long before al-Bīrūnī’s time. These rulers adhered to Brahminism and to a certain form of Hinduism. Thus, from this rather broad point of view, al-Bīrūnī’s descriptions in the Taḥqīq mā li-l-Hind match the Late Shahi society. However, Brahminism and Hindu cult can be associated with almost all ruling dynasties of early medieval India. Therefore, I now explore the possibility that for more specific cultural features, notably the philosophy, al-Bīrūnī’s descriptions also fit with the society that was based in Gandhāra and Panjab. In other words: were the teachings of Sāṅkhya-Yoga popular among the Brahmins of Gandhāra and Panjab in the early eleventh century CE?

This part of the investigation has two aims: 1) to consider the question of a geographical foyer of the Sāṅkhya-Yoga philosophies in the north-western subcontinent in the early eleventh century CE, an information which is often lacking when Indian philosophy is concerned; 2) to demonstrate that portions of the Taḥqīq mā li-l-Hind indeed constitute accurate testimony to the religious, literary and intellectual traditions of the Late Shahis, provided the information is well contextualized and analysed.

Addressing these questions presupposes that local traditions, for instance sciences and religion, continued to be practiced after the Ghaznavids entered the region. I have already observed that people continued to adhere to local customs, such as the timekeeping ritual and traditional calendrical system, in Laghmān and Peshawar areas, some years after Muslim conquests there.94 Another example of such phenomenon is that of Somnāth. Richard M. Eaton notes that, after the expedition which Maḥmūd led there, extant local Sanskrit inscriptions did not refer to the event, thus suggesting a relatively low impact on local life.95 It is likely that in the domain of philosophy, too, locals continued to read their texts and to practice their teachings after the Ghaznavids’ arrival in Gandhāra and Panjab. Al-Bīrūnī then most probably encountered such living traditions there.

Furthermore, if the philosophers who helped al-Bīrūnī understand the two philosophies were based in Gandhāra and Panjab, the intellectual context of the Late Shahis was thus conducive to the development of these two philosophies in particular. In other words, I posit that the Late Shahi rulers supported philosophers versed in these two philosophies, rather than any other one. Thus, the present section considers the question of how al-Bīrūnī accessed the Sanskrit Sāṅkhya and Yoga manuscripts. It explores the possibility that they were read among the Brahmins he met in Gandhāra and Panjab, and finally examines the reasons why he translated these two philosophical works in particular into Arabic.

There is however very little indication of the possible geographical provenance of the books related to Sāṅkhya-Yoga that al-Bīrūnī consulted for his translations. Therefore, I first discuss places that could most probably not constitute these geographical foyers for al-Bīrūnī, despite their importance as centre knowledge of the time. They are Kanauj, the Valley of Kashmir, Somnāth, Multan and Varanasi.

One passage of the Kitāb Pātanğal appears to rule out Kanauj as a place where al-Bīrūnī could have found his Sanskrit source for this translation. In the passage on the different means of knowledge, a simile is offered regarding āgama, that is, authoritative tradition:

Just as our knowledge that the city of Kanauj (‮كنوج‬‎) is on the bank of the Gaṅgā (‮كنك‬‎) River, this [knowledge] results from [oral] report but stands for its (i.e., the knowledge’s) apprehension by eyesight.96

This example is absent from any extant Yoga text predating al-Bīrūnī’s time which could have thus inspired him, that is, the bhāṣya-part of the Pātañjalayogaśāstra, the Vivaraṇa, the Tattvavaiśāradī and the Rājamārtaṇḍa. It appears to be his own creation, or that of his informants. If it is so, the use of this illustration suggests either that al-Bīrūnī never went to Kanauj—which appears to be the case—or that his interlocutors, who read the Kitāb Pātanğal with him, were not from Kanauj.

As for the Valley of Kashmir, as I have argued in Section 1.2.4, al-Bīrūnī was very well informed about this region, although he never visited it at least by the time of the composition of the Taḥqīq mā li-l-Hind.97 At the time, important intellectual exchanges were indeed taking place between inhabitants of Gandhāra and Panjab on the one hand, and those based in the Kashmir Valley on the other. An extract drawn from the Taḥqīq mā li-l-Hind explicitly mentions such exchanges:

I have been told that this man (i.e., Ugrabhūti, the grammarian) was the educator and teacher of the Šāhi Ānandapāla, son of Jayapāla (‮انندپال بن جيپال‬‎), [who ruled] in our time. After [Ugrabhūti] completed [his] book, he sent it to Kashmir (‮كشمير‬‎). People there did not accept it because of their arrogance in these [things] […]. [Ānandapāla] ordered the sending of 200,000 dirhams and similar presents to Kashmir, in order to distribute [these gifts] among those who were occupied with the book of his master.98

This anecdote illustrates the vigour of intellectual exchanges between the two regions at the time. Further, as aforementioned, al-Bīrūnī’s own bibliography also provides evidence suggesting that the scholar corresponded with Kashmiris, as one of his works is entitled Answers to the Ten Kashmiri Questions (‮الجوابات عن المسائل العشر الكشميرية‬‎).99

Interactions and relations between the kingdom of the Shahi rulers and the Valley of Kashmir were part of an earlier tradition. In the Rājataraṅgiṇī, composed in the mid-twelfth century, Kalhaṇa states that Lohara (Fort Lahūr) was dependent on the Kashmiri kings, at the time of Lalitāditya Muktāpīḍa (r. ca. 724–760).100 According to the same author, the ruler of Lohara in the tenth century, Siṃharāja, was the son-in-law of Bhīma the Shahi (śrībhīmaśāhi), namely the king who precedes Jayapāla in the list of the kings provided by al-Bīrūnī. Siṃharāja had his daughter Diddā married to the Kashmiri king Kṣemagupta (r. ca. 950–958).101

Incidentally, Kashmir was flourishing at the turn of the first millennium, when Queen Diddā had a college (maṭha) built, which hosted young Brahmins from Madhyadeśa (Madhya Pradesh), Hāṭa (or Karahāṭa, in Uttar Pradesh) and Saurāṣṭra (Surat, Gujarat).102 The situation of Bilhaṇa, a Kashmiri minister and poet who lived in the eleventh century, also demonstrates this dynamism and mobility, as he travelled from Kashmir to Mathura, Kanauj, Prayāga, Anahilwada and Somnāth.103 Al-Bīrūnī who describes Kashmir as a shelter for the Indian sciences and who had access to some of these sciences from Kashmir, may have thus received Sāṅkhya-Yoga manuscripts from Kashmiri travellers.

Moreover, Abhinavagupta, who lived in Kashmir during the second half of the tenth century, extensively elaborates upon the ideas referred to collectively as Kashmiri Śaivism. Both Kashmiri Śaivism and Śaiva Tantra make use of Sāṅkhya-Yoga concepts in their own philosophical constructions.104 Thus, one may wonder whether the Sāṅkhya-Yoga ideas which are found in the Taḥqīq mā li-l-Hind were actually drawn from Kashmiri Śaiva theories or not. If so, then the Kitāb Sānk and the Kitāb Pātanğal may be interpretations of Sanskrit works originally found in the Valley of Kashmir.

The content of the two books, that is, the Kitāb Sānk as it is transmitted to us in the Taḥqīq mā li-l-Hind and the Kitāb Pātanğal, however, makes it clear that the ideas they developed relate to the traditions of the Sāṅkhyakārikā and the Pātañjalayogaśāstra respectively. This will become clear in the subsequent chapters of the present book. In addition, al-Bīrūnī’s Taḥqīq mā li-l-Hind does not disclose explicit signs of influences of Kashmiri Śaivism. Thus, despite the favourable situation for intellectual exchanges with Kashmir, there is no evidence of such exchanges in the domain of philosophy between al-Bīrūnī and Kashmiri thinkers, nor is there any indication that his interpretations of Sāṅkhya and Yoga texts were based on manuscripts originating from this region.

Further, only in few cases, al-Bīrūnī provides the geographical provenance of the authors of the works he used: the Brāhmasphuṭasiddhānta is by Brahmagupta from Bhillamāla, a book by Durlabha from Multan, the Srūdhava by Utpala from Kashmir, and the Karaṇatilaka by Vijayanandin from Varanasi. In the same way, he very rarely specifies the geographic provenance of his informants. He only does so in the case of Kanauj, Multan and Somnāth. In addition, he never states in the Taḥqīq mā li-l-Hind if an information comes from the region he visited. I thus suggest that al-Bīrūnī may have only considered it important to inform his readers about the provenance of his information—oral and written—when it came from places he did not visit himself. In contrast, al-Bīrūnī may have not deemed it necessary to explicitly state the geographical provenance of books, sciences and customs when they were locally studied and practiced. If this is accepted and in view of the above discussion, the Sanskrit sources of the Kitāb Sānk and Kitāb Pātanğal are thus unlikely to have come from Kanauj, Multan, Somnāth or Varanasi. Lastly, some texts, like the Vedas, some great Purāṇas (Mahāpurāṇas), such as the Viṣṇupurāṇa, the Ādityapurāṇa, the Matsyapurāṇa and the Vāyupurāṇa, the Bhagavadgītā and the Mahābhārata, certainly belonged to a category of literature widespread among Brahmins of early medieval India. One may thus wonder whether the sources of the Kitāb Sānk and the Kitāb Pātanğal were also to be counted among this category of literature. This large diffusion would account for al-Bīrūnī’s failure to specify their geographic provenance. A cross-examination of numismatic and textual data, however, suggests that the Sāṅkhyakārikā and the Pātañjalayogaśāstra traditions were popular teachings among people living in the north-western subcontinent at the beginning of the eleventh century.

First, the use of technical Sāṅkhya term avyakta in the legend of the bilingual coins of Lahore described in Section 1.2.3 of the present book may suggest that classical Sāṅkhya philosophy had adepts in Panjab.105 The concept of avyakta, namely the original cause (prakṛti), is employed to refer to God (Allah) in the Arabic šahāda. This fits well with the fact that in classical Sāṅkhya there is no notion of a creator God: the original cause is the active origin of the phenomenal world.106

Thus, the notion of avyakta is the Sāṅkhya concept that best renders the concept of the Islamic creator God. The Sanskrit legend of Maḥmūd’s bilingual coins likely suggests that the principles of Sāṅkhya metaphysics influenced the legend’s composition, thereby serving as evidence that the literate population of the western Panjab, a part of the Late Shahi kingdom, most probably was familiar with classical Sāṅkhya.

Furthermore, there is a series of instances in the Taḥqīq mā li-l-Hind suggesting that al-Bīrūnī’s informants considered the topics of the Kitāb Sānk and the Kitāb Pātanğal as essential teachings. In these instances, al-Bīrūnī does not attribute a specific geographical provenance to his statements, implying thus that their principles were well established practices among al-Bīrūnī’s local informants. Al-Bīrūnī mentions the Kitāb Sānk and the Kitāb Pātanğal in the preface to the Taḥqīq mā li-l-Hind. These books are described as containing “most principles around which their (i.e., the Indians) faith revolves, without the subdivisions of their religious laws” (‮فيهما أكثر الاصول التى عليها مدار اعتقادهم دون فروع شرائعهم‬‎).107 In the Taḥqīq mā li-l-Hind, al-Bīrūnī also qualifies the Kitāb Pātanğal as a “famous book” (‮الكتاب المشهور‬‎).108 The Kitāb Pātanğal and Kitāb Sānk—together with the Kitāb Gītā and the Purāṇas—are quoted throughout the Taḥqīq mā li-l-Hind when the latter does not deal with astronomy. The descriptions made by al-Bīrūnī of these two books, their mention at the very beginning of the Taḥqīq mā li-l-Hind, and the frequent references to them throughout this work indicate their central importance to the Indian thinkers he encountered.109

Another clue to the popularity of Sāṅkhya among the Indians with whom al-Bīrūnī interacted lies in the way he sometimes describes concepts related to this philosophy in the Taḥqīq mā li-l-Hind, as though they were common beliefs of these Indians and part of an oral local tradition. In a chapter entitled “On their belief in the existents, both intelligible and sensible” (‮فى ذكر اعتقادهم فى الموجودات العقليّة الحسّيّة‬‎), al-Bīrūnī presents the opinion of “those [among the Indians] who deviate from allusions [but direct themselves] to investigation” (‮الّذين يعدلون عن الرموز إلى التحقيق‬‎)110 and enumerates twenty-five constitutive principles (tattva). For the most part, this exposition matches the definitions found in the Sāṅkhyakārikā and its commentaries.111

Al-Bīrūnī’s enumeration begins with pūruša (‮پورش‬‎), the Arabic transliteration of the Sanskrit puruṣa, an essential Sāṅkhya principle that every living being possesses and which I translate here as “conscious self.” Al-Bīrūnī defines it as the human soul or nafs (‮نفس‬‎). According to al-Bīrūnī’s report, puruṣa is only characterized by life, and presents a succession of knowledge and ignorance, as it is ignorant in actuality and intelligent in potentiality, the cause of action being its ignorance.112 This description of puruṣa reflects to some extent that of the Sāṅkhyakārikā. Indeed, according to this tradition, the conscious self is inactive, indifferent and defined as the knower (jña).113

The next element enumerated by al-Bīrūnī is abyakta (‮ابيكت‬‎), an Arabic transliteration of the Sanskrit avyakta, meaning “unmanifest,” which al-Bīrūnī defines as absolute matter (‮المادّة المطلقة‬‎) or pure primary matter (‮الهيولى المخرّدة‬‎), a philosophical term drawn from Aristotle’s works and known to his readership. It is inanimate and possesses the three forces (‮القوى الثلاث‬‎) in potentiality but not in actuality.

The three forces are satta, raja and tama (‮ست‬‎; ‮رج‬‎; ‮تم‬‎) and correspond to the three constituents (guṇa), sattva, rajas and tamas. Al-Bīrūnī describes them as: 1) quietude and goodness, from which existence and growth originate and which are ascribed to angels (‮الملائكة‬‎), namely the deities (deva), 2) exertion and labour, from which constancy and continuation originate and which are ascribed to man and 3) languor and indecisiveness, from which decay and annihilation originate and which are ascribed to animals.114

In his enumeration, al-Bīrūnī also describes the byakta (‮بيكت‬‎), a transliteration of the Sanskrit vyakta, meaning “manifest,” and qualifies it as the shaped (‮متصوّرة‬‎) matter that possesses the three forces and moves outward in actuality (‮المادّة خارجة إلى الفعل‬‎). He reports that the term prakriti (‮پركرت‬‎) designates the whole of pure primary matter and of shaped matter. He then turns to āhangāra (‮آهنگار‬‎), which he identifies with the [innate] temperament (‮الطبيعة‬‎). The mahābhūtas (‮مهابوت‬‎), the five gross elements, are then described as constituting all existents of this world. He refers to them employing the Arabic expression commonly used to designate the four elements in Islamic tradition, namely the great natures (‮كبار الطبائع‬‎).

At this point in the passage, al-Bīrūnī quotes from the Vāyupurāṇa. After this quotation, he discusses the panğ mātar (‮پنج ماتر‬‎), a transliteration of Sanskrit pañca tanmātra which refers to the five subtle elements and interprets the expression as signifying “five mothers” (‮أمّهات خمسة‬‎) and “simple elements” (‮بسائط‬‎). According to him, these subtle elements precede the gross elements. In parallel with the Sāṅkhyakārikā, al-Bīrūnī connects each of the five tanmātras to one of the mahābhūtas: ether is sound, šabda (‮شبد‬‎); wind is what is touched, sayiras (‮سيرس‬‎);115 fire is form, rūpa (‮روپ‬‎); water is what is tasted, rasa (‮رس‬‎); and earth is what is smelled, ganda (‮گند‬‎).116 Attempting to explain the seemingly strange connection between sound and ether, he invokes quotations from Homer, Porphyrus, Diogenes and Pythagoras.117

Al-Bīrūnī also describes the five sense organs, indriyān (‮اندريان‬‎), corresponding to the buddhīndriyas of classical Sāṅkhya, which he defines as “hearing by the ear, sight by the eye, smelling by the nose, tasting by the tongue and touching by the skin.” He further explains mana (‮من‬‎), namely the mind (manas), as “the will (‮إرادة‬‎) [that] directs [the senses] to [their] various locations (of action)” and “as [having] its residence in a [person] (‮منه‬‎) [… in] the heart (‮قلب‬‎).”118 Al-Bīrūnī further explains the five senses by action (‮الحواسّ بلفعل‬‎), which he calls the karma indriyān (‮كرم اندريان‬‎)119 and the five necessities (‮ضروريّات‬‎). He describes these principles as follows: production of a sound for [different] kinds of needs and wishes; strength by the hands for fetching and putting away; walking with the feet so as to seek [something] or to flee [from it]; and shaking off the excess of food through each of the two holes destined for it.

At the end of this explanation, al-Bīrūnī provides a summary in which he again lists all principles along with their generic designation as tatwa (‮تتو‬‎), that is, the Arabic transliteration of the Sanskrit term tattva.

Al-Bīrūnī appears to interpret and explain some of the above Indian concepts on the basis of his intellectual background, for instance, by using Aristotelian and Neo-Platonic terminology and concepts, such as “potentiality” and “actuality.”120 As I show in Chapters 4, 5 and 6 of this book, al-Bīrūnī often used a terminology that his Muslim readership knew so as to make his description of Indian thought and religion more palatable to them. This strategy of translation, however, results in transformed and unliteral translations when compared to their possible sources. As aforementioned, this strategy resembles that which was adopted to translate the Arabic šahāda into Sanskrit on the bilingual coins found in Lahore.

Al-Bīrūnī’s account also diverges from the Sāṅkhyakārikā tradition on a few other points. Instead of using the terms avyakta and vyakta as generic designations for some of the twenty-five elements as the Sāṅkhyakārikā does, al-Bīrūnī, it seems, understands that the avyakta and vyakta are principles themselves. He makes no mention of mahat or buddhi, which is the principle that originates from prakṛti, or avyakta; but in his scheme, the constitutive principle emerging from avyakta is vyakta, and thus there are still twenty-five tattvas. This confusion, whether due to a misunderstanding on the part of al-Bīrūnī or on the part of his informants, points to an oral transmission of this account.

On the whole, however, al-Bīrūnī’s description of the Sāṅkhya tattvas (principles) globally matches that of the Sāṅkhyakārikā and its commentaries. In contrast, it does not correspond to Sāṅkhya ideas which are found in the Buddhacarita or the Mahābhārata, nor those of the thirty-six tattvas of Kashmiri Śaivism, from which al-Bīrūnī could have hypothetically drawn his information. Therefore, the account in the Taḥqīq mā li-l-Hind is most likely either a summary of what al-Bīrūnī had heard from his informants about the Sāṅkhya system of thought or of a passage of the Kitāb Sānk. Al-Bīrūnī’s confusions in this passage, however, tends to suggest that the passage was mostly orally transmitted to him, and thus a popular teaching among the people he met. Thus, these few examples tend to show that Sāṅkhya and Yoga philosophies were popular among al-Bīrūnī’s informants and in the regions he visited in al-Hind.

In order to supplement the discussion of the geographical provenance of al-Bīrūnī’s description, I propose to consider the question as to how al-Bīrūnī’s personal inclinations may have influenced his choice of topics he covers in the Taḥqīq mā li-l-Hind. In the case of astronomy and mathematics, al-Bīrūnī’s interests played a crucial role for the information he transmitted in the Taḥqīq mā li-l-Hind. Regarding Indian philosophies, it is possible that al-Bīrūnī considered the teachings found in the Kitāb Sānk and Pātanğal to have affinities with Falsafa, or Islamic philosophy, and Sufism. He sometimes used a terminology of these systems of thought to render concepts of the metaphysics of Sāṅkhya and of the ethics of Yoga.121 The parallel that al-Bīrūnī drew between Indian and Islamic thought may have led him to transmit Sāṅkhya and Yoga, rather than any other Indian philosophies, to his Arabic readership.

Evidence, however, shows that he was much dependent on his Indian informants and on observation of the places he visited in al-Hind for his description of Indian culture. He appears then not to have solely let his personal interests guide his work.

For instance, the Taḥqīq mā li-l-Hind hardly ever deals with Buddhism, despite al-Bīrūnī’s knowledge of it and possible interest in it.122 Al-Bīrūnī himself states the reasons why he did not take into account Buddhist communities:

As I never found a Buddhist (‮الشمنيّة‬‎) book and none of the [Indians could] clarify [to me] their [theories] on this subject, my account of them (i.e., the Buddhists) is based on al-Īrānšahrī (‮الايرانشهرىّ‬‎).123

This passage reveals that the absence of information on Buddhism in the Taḥqīq mā li-l-Hind is due to al-Bīrūnī’s not having access to books related to Buddhism, and not due to his lack of interest in the subject. It is however clear that al-Bīrūnī made a distinction between Buddhists and Brahmins, as the following passage illustrates:

[Another circumstance] which intensified [the Indian] aversion and hostility [to foreigners]124 is [related to] the group known as Šamaniyya (i.e., Buddhists), [who] despite their intense hate toward the Brahmins (‮البراهمة‬‎) are closer to the Indians (‮الهند‬‎), than any other [men]. Formerly, the Khurasan (‮خراسان‬‎), Persia (‮فارس‬‎), Iraq (‮العراق‬‎) and Mosul (‮موسل‬‎)125 up to the frontier of Syria (‮الشام‬‎) [belonged to] their religion until Zarathustra (‮زردشت‬‎) arrived from Azerbaijan (‮اذربيجان‬‎) and promoted Mazdeism (‮المجوسيّة‬‎) in Balkh (‮بلخ‬‎). His promotion was successful with the [king] Kuštāsb (‮كشتاسب‬‎), and his son Isfandiyār (‮إسفنديار‬‎) continued to promote it in the eastern and in the western lands, forcibly and peacefully. He erected fire temples from China (‮الصين‬‎) to the Greek (‮الروم‬‎) [empire]. The kings after him made way for their religion (i.e., Mazdeism) in Persia and Iraq. Thus, the Šamaniyyas moved from these [lands] to the east of Balkh.126

The term al-šamaniyya (‮الشمنيّة‬‎) is the Arabic for naming the Buddhists. Al-Bīrūnī did make use of this term, and in the Taḥqīq mā li-l-Hind he differentiates al-šamaniyyas from the Brahmins, namely al-barāhimas (‮البراهمة‬‎).127 Here, he attempts to explain the decline of Buddhism in Central Asia as due to the advent of Zoroastrianism.128 This question is a much debated one. The types of interaction between the several religious communities established in Central Asia and in the north-western parts of South Asia, such as Buddhist, Muslim, Hindu and Zoroastrian, indeed, varied regionally,129 and it is not the place here to discuss this issue.

Nevertheless, in addition to his statement about the inaccessibility of Buddhist books for him, al-Bīrūnī did not describe any well-known Buddhist site, in the way he described Hindu temples and idols, for instance in Taneshwar, Multan and Somnāth. It is thus likely that he indeed did not directly interact with a Buddhist who could have informed him about important Buddhist sites at the time. His testimony perhaps also indicates that the significance of Buddhist sites as intellectual or cultural centres had waned in Gandhāra and Panjab and that Buddhist communities were no longer supported by the ruling dynasties at the time. In addition, as I showed above, most information that al-Bīrūnī gathered in the Taḥqīq mā li-l-Hind relates to a socio-religious context different from that of a Buddhist population.

Furthermore, while al-Bīrūnī generously quotes from texts linked to Sāṅkhya and Yoga philosophies in the Taḥqīq mā li-l-Hind, he remains silent with regard to other Indian systems of thought. For instance, he did not engage with Nyāya-Vaiśeṣika tradition, nor with the Mīmāṃsā or Vedānta, generally considered predominant in India at the time. Why the Taḥqīq mā li-l-Hind is silent about Advaita-Vedānta philosophy, for example, is another relevant question to consider.130 One may argue that this absence could either be due to al-Bīrūnī’s particular preferences or to the fact that he did not access books related to these philosophical systems in Gandhāra and Panjab. In view of the above, it is likely that the people al-Bīrūnī met in Gandhāra and Panjab were no philosophers of Mīmāṃsā and Vedānta, nor of Nyāya-Vaiśeṣika.

The passage below contains the only explicit reference to philosophies other than Sāṅkhya and Yoga in the Taḥqīq mā li-l-Hind. In contrast with his comments on Buddhism, al-Bīrūnī does not explain his lack of information on these philosophies. The passage reads as follows:

They (i.e., the Indians) have books on the jurisprudence of their religion, on theology, on asceticism, on deification and on the quest for liberation from the world,131 such as the eponymous book composed by Gauḍa (‮گور‬‎) the ascetic; [the Kitāb] Sānk (‮سانگ‬‎), composed by Kapila (‮كپل‬‎), on divine subjects; [the Kitāb] Pātanğal (‮پاتنجل‬‎), on the quest for liberation and for union between the soul and its object of apprehension; [the Kitāb] Nyāyabhāṣya,132 composed by Kapila, on the Vedas (‮بيذ‬‎) with their interpretation, [on the position] that they (i.e., the Vedas) have been created, and on the distinction in these [works] between divine precepts and customs; [the Kitāb] Mīmāṃsā (‮ميمانس‬‎), composed by Jaimini,133 on the same subject; [the Kitāb] Lokāyata (‮لوكايت‬‎), composed by Jupiter (‮المشترى‬‎),134 which accepts sensory perception as the only [means to know] about an object of investigation; [the Kitāb] Agastimata (?) (‮آگست مت‬‎),135 composed by Canopus (‮سهيل‬‎),136 on the use of the senses, alongside with the oral tradition, [as means to know] about the [object of investigation]; and the book Viṣṇudharma (‮بشن دهرم‬‎). The [general] interpretation of [the term] dharma [takes it to refer to] reward, but [here] it means “religion,” as if the book were [titled] The Religion of Allah, related to Nārāyaṇa (‮ناراين‬‎).137

Thus, al-Bīrūnī is aware of the existence of a certain number of works connected to Indian philosophy and religion. Some problems and questions however arise with regard to his account. For instance, al-Bīrūnī enumerates a work which he entitles Nyāyabhāṣya and which he attributes to Kapila. If the work mentioned here by al-Bīrūnī refers to the extant Vātsyāyana’s Nyāyabhāṣya, he was ill-informed when he associates its authorship with Kapila. In addition, al-Bīrūnī’s description of the subject dealt with in this book does not reflect the actual main topic of the Sanskrit Nyāyabhāṣya that pertains to debate, logic, epistemology and metaphysics. This confusion contrasts with his relatively accurate description of the content of the books related to Sāṅkhya and Yoga. Also of note is that titles of works related to the Vaiśeṣika or Vedānta systems are not referred to in the above enumeration. This absence suggests that al-Bīrūnī did not access accurate information about these two traditions. As in the case of literature relating to Buddhism, al-Bīrūnī probably did not find books related to them; at the same time, this account rather reflects the scope of his informants’ philosophical knowledge and training.

Two additional elements indicate that the selection of literature transmitted by al-Bīrūnī does not necessarily reflect his personal preferences: his criticism of the content of some texts, despite him quoting them in the Taḥqīq mā li-l-Hind, and the expression of his constant quest for knowledge and books.

Although al-Bīrūnī heavily quotes from purāṇic literature, specifically the Viṣṇudharma,138 the Viṣṇupurāṇa, the Matsyapurāṇa, the Vāyupurāṇa and the Ādityapurāṇa, he also criticizes their content. For instance, referring to a passage he draws from the Viṣṇudharma, he writes:

Further, the [Viṣṇudharma] says: “A [man] who reads this [statement regarding the celestial pole] and knows it accurately, him [indeed] Allah forgives [his] sins of the day and adds fourteen years to his predetermined age.” How simple are [these] people! And among us, some [scholars] know between 1020 to 1030 stars. He (i.e., Allah) does not take their breaths, nor deduct from their age only because of this.139

Further, after having quoted the Viṣṇupurāṇa, the Matsyapurāṇa, the Vāyupurāṇa, the Ādityapurāṇa and the Kitāb Pātanğal regarding the size of Mount Meru, al-Bīrūnī states:

The excessive dimensions of Mount [Meru] only make sense because of the excessive dimensions of the earth which these [works] report. If conjecture has no limit, then there is open space for [further] lying based on what was presumed.140

More generally, al-Bīrūnī notes the following:

This sum is more than thrice the one we mentioned according to the commentator Pātanğal. This is a habit of the copyists in all languages. The authors of the Purāṇas (‮پرانات‬‎) are not free from it, as they are not adherents of scientific studying.141

As for the authors of the Purāṇas, they represent heaven as a still dome above the earth and the stars wandering from east to west. How [then] would they have knowledge of the second motion? And if they would, how would [their] opponent let them [believe] that a unique thing (i.e., star) [can] move in two different directions by itself? We mention what has reached us from them, not because it is useful, as there is no usefulness in it.142

Despite his complaints regarding some ideas found in the purāṇic literature, al-Bīrūnī still mentions them. The transmission of such theories, which he considered worthy of critique, in his writing thus is not due to his personal inclination. Similarly, although al-Bīrūnī translated the Kitāb Pātanğal, he disliked the presentation of cosmography by the author of this book:

We have already found annoying the mention of the seven oceans, together with the [seven] earths, and even so this man [considers] making our burden lighter by adding earths below.143

Furthermore, two passages indicate that al-Bīrūnī actively searched for different kinds of sources in order to inform himself. The first passage where he describes his search for source materials occurs in his aforementioned historical account of the Late Shahis:

[… The history of] such a lineage, [written] on a piece of silk, is found in the fortress of Nagarkot (‮نغركوت‬‎). I desired to find it but was prevented from doing so for [different] reasons.144

His constant search for written documents is also evident in the following extract:

I have found it very hard to work my way into this [subject] (i.e., al-Hind), despite my desire which I alone possess in my time, and although I sacrificed myself generously and as much as possible in collecting their books from places where they were likely to be found and in gathering those who were on the right way to [find] them from places where they were hidden.145

Al-Bīrūnī’s intellectual curiosity was thus not limited to the works he may have been sympathetic to. It is likely that had he discovered books related to Buddhism or to other traditions of Indian thought, he would have turned his attention to them and reported on them. This assumption concurs with the following statement by al-Bīrūnī in the preface to the Taḥqīq mā li-l-Hind: “[T]his book is a report [of the facts]. I convey the words of the Indians as they are.”146 All of the above indeed strongly suggests that al-Bīrūnī mostly described aspects of the Indian society as it was presented to him through personal collaborations taking place in Gandhāra and Panjab, including Brahmins belonging to the Late Shahi court.

2.5 Concluding Remarks

This chapter first surveyed the intellectual framework in which al-Bīrūnī encountered Indian society, science and literature. It was possible to show that al-Bīrūnī gradually familiarized himself with Indian culture throughout his life. Having been originally trained as an astronomer and mathematician, he later on expanded his knowledge to the history of civilizations, Indian culture, gemmology and pharmacology. While his pre-existing knowledge of Indian science and culture was chiefly based on textual materials, the information found in the Taḥqīq mā li-l-Hind relies on his knowledge of Sanskrit literature, on his interactions with Indians and on his observations made during his visits to al-Hind.

Further, the context of Maḥmūd’s court was favourable for al-Bīrūnī to learn Sanskrit and to act as an interpreter. In this context, al-Bīrūnī not only met traders or travellers who, for instance, informed him about the geography of different provinces of India. He also actively interacted with Brahmins, some of whom were well versed in religious matters, astronomy, purāṇic literature and philosophy, and who guided him in understanding such areas of study. The elite with which al-Bīrūnī interacted was certainly part of the royal Indian courts which were directly challenged by Maḥmūd’s military campaigns, including that of the Late Shahis.

Two circles of literate persons—astronomers, priests, philosophers, interpreters, etc.—belonging to two distinct courts, namely that of Maḥmūd and of Indian kings, met thanks to the political context of the time. Further, in the case of the Ghaznavid court, the interacting group included Muslim thinkers and Brahmins who conducted various empirical studies and shared an interest in scientific problems.147 The case of al-Bīrūnī and the assumed circumstances in which he interacted with Indian thinkers also illustrate how political power, commercial interests and religious context strongly influenced access to one’s object of research at the time. I highlighted in Chapter 1, that a war context, also motivated by commercial concerns, contributed to the opening and the protection of roads network. The control over these roads encouraged knowledge to circulate among scholars. Chapter 2 of the present book confirmed these observations and highlighted the dynamism of intercultural and intellectual exchanges at Muslim royal courts in early medieval times.

In addition, when al-Bīrūnī studied Sanskrit and Indian sciences he had to collaborate with thinkers versed in Sanskrit, but also acquainted with Persian or Arabic. This group of scholars may also have worked with him through the intermediary of a vernacular language. Thanks to these collaborations and to the many texts available to him, al-Bīrūnī was well informed about Sanskrit phonology and terminology.

Places where al-Bīrūnī’s long-term interactions with this elite occurred may have been in the region of Ghazna and Kābul—to where Indians might have been taken as prisoners or where they were employed as advisors and interpreters. In the territory of al-Hind, Fort Nandana, Nagarkot, and possibly Wayhind/Udabhāṇḍa, Peshawar, Lahore and Multan, would have hosted such collaborations.

Discrepancies between the Indian textual tradition and al-Bīrūnī’s testimony addressed in this chapter may reflect his informants’ confusion, thereby foregrounding the importance of orality in al-Bīrūnī’s reception of the two Indian philosophies. Oral tradition indeed played an important role in al-Bīrūnī’s acquisition of knowledge about India and in his translations.

Lastly, data drawn from al-Bīrūnī’s writings and from numismatic evidence, if considered from a circumstantial perspective, strongly suggests that the sources of the Kitāb Pātanğal and Kitāb Sānk were popular readings and textbooks among the Indian scholars whom al-Bīrūnī met, possibly in Gandhāra and Panjab. If these thinkers belonged to the Late Shahi court, one may infer which sciences, religions and philosophies these kings were supporting and promoting. In this way, the study and teaching of Sāṅkhya and Yoga philosophies may have been financially supported by the Late Shahi kings.


Sachau 1910: I/18. See also Strohmaier 1991: 153.


Al-āṯār (1878); Al-āṯār (2001); Sachau 1879.


Al-āṯār (1878), p. 71; Al-āṯār (2001), p. 80; Sachau 1879: 83. The following comments concern variant readings of some of the names of the months given in this table: bhadrabad (‮بهدربد‬‎) is an emendation by Sachau; the manuscripts read bharūnda (‮بهروند‬‎). Azkaei’s edition reproduces the manuscripts’ readings as follows: bawš (‮بوش‬‎) reads bawšn (‮بوشن‬‎) in Al-āṯār (2001), and bākr (‮باكر‬‎) reads yākn (‮ياكن‬‎).


Al-āṯār (1878), p. 192; Al-āṯār (2001), p. 221; Sachau 1879: 172. The sun and the moon were included in the concept of the planets (graha) during a specific phase in the history of this Indian concept (Yano 2003, 2004: 331–332 and 335–337). My comments on the variant readings are as follows: brhasbatī (‮برهسبتى‬‎) is an emendation by Sachau, as the manuscripts have various readings; ādīda (‮اديد‬‎) reads adiṯah (‮ادثه‬‎) in Al-āṯār (2001), and šurk (‮شرك‬‎) reads šūk (‮شوك‬‎).


Al-āṯār (1878), p. 193; Al-āṯār (2001), p. 222; Sachau 1879: 173; Yano 2003: 384–385; brša (‮برش‬‎) reads bršā (‮برشى‬‎) in Al-āṯār (2001), and makar (‮مكر‬‎) reads makad (‮مكد‬‎).


In the Taḥqīq mā li-l-Hind, al-Bīrūnī equally enumerates the names of the Indian months transliterated from Sanskrit. The spellings there slightly differ from the same list found in Al-āṯār al-bāqiya (Taḥqīq [1958], p. 302; Sachau 1910: I/358).


The Arabic term ribāṭāt is understood as an astronomical technical term referring to stations.


The expression ʿilm al-ğafr can be translated as “divination,” but the exact meaning of the plural term ğufūr in the context of the above quotation is unclear.


Al-āṯār (1878), p. 336.12–22; Al-āṯār (2001), p. 432.1–11; Sachau 1879: 335–336.


A quotation drawn from Abū Maʿšar precedes this passage. It exposes a method to calculate the influences of the rising and setting of a lunar mansion (Sachau 1879: 341–342). The astrologer Abū Maʿšar was a native of Balkh living in the eighth or ninth century CE. He played an important role in the transmission of Indo-Iranian astrology to the Islamic world (Sachau 1879: 375; Pingree 1963: 243–245).


Al-āṯār (1878), p. 341.6–7; Al-āṯār (2001), p. 437.13–14; Sachau 1879: 342.


These translations include the Zīğ al-arkand and the Zīğ kandakātik, both based on Brahmagupta’s Khaṇḍakhādyaka, as well as the Zīğ karanatilaka (Vijayanandin), Zīğ karanasara (Vaṭeśvara) and the Kitāb al-adwār wa l-qirānāt (Ahmed 2001: 161–165). See also Pingree 1963, Baloch 1973: 24–33, Said & Khan 1981: 45 and Sarma 2009: 214–215.


Al-āṯār (1878), p. 9.15–18; Al-āṯār (2001), p. 13.6–9; Sachau 1879: 11.


Al-āṯār (1878), p. 25.12–13; Al-āṯār (2001), p. 31.11–12; Sachau 1879: 29.


Al-āṯār (1878), p. 26.18–20; Al-āṯār (2001), p. 32.15–18; Sachau 1879: 31.


Al-āṯār (1878), p. 259.2–8; Al-āṯār (2001), p. 323.1–7; Sachau 1879: 249–250.


The original term allāh is kept here, as it is difficult to determine which Indian deity al-Bīrūnī is referring to.


Al-āṯār (1878), p. 274.13–16; Al-āṯār (2001), p. 339.15–18; Sachau 1879: 266.


Pingree 2012b. The mathematician and astronomer al-Fazārī also composed the Zīğ al-Sindhind al-kabīr which is either based on the Brāhmasphuṭasiddhānta or on a hypothetical Mahāsiddhānta drawn on the former. See Pingree 2012a and fn. of the introduction to the present book.


Al-āṯār (1878), pp. 12.19–13.10; Al-āṯār (2001), pp. 16.21–17.11; Sachau 1879: 15.


The Arabic term al-rūm (‮الروم‬‎) is employed to refer to the people of the Eastern Roman Empire, including the Greeks, in contrast to al-yūnānī, (‮اليونانى‬‎) which refers to the ancient Greeks.


Al-āṯār (1878), pp. 51.1–2, 5–6 and 9–10; Al-āṯār (2001), pp. 59.5–7, 9–11and 13–14; Sachau 1879: 60.


Al-āṯār (1878), p. 228.2–3; Al-āṯār (2001), p. 283.9–11; Sachau 1879: 214.


In the Taḥqīq mā li-l-Hind, al-Bīrūnī also makes mention of Arabic writers acquainted with India (Taḥqīq [1958], p. 351.3; Sachau 1910: II/18).


On Yaʿqūb Ibn Ṭāriq see Pingree (1968).


This work is unknown to me.


Al-Ğayhānnī was probably a vizier of the Samanid dynasty (ca. 10th c.); see Sachau 1879: 424 and Pellat 2012. Al-Bīrūnī also made reference to him in the Taḥdīd al-amākin, when he writes: “Once, I had the intention to glean the information provided by the method of Ptolemy, in his book, the Geography, and by the method of al-Jaihānī and others, in their books on al-Masālik, for the following purposes: the collection of data, the clarification of obscurities, and the perfection of the art” (translation Ali 1967: 14).


Al-āṯār (1878), p. 12.19; Al-āṯār (2001), p. 16.21.


It is worthy of note that more than the half of the Taḥqīq mā li-l-Hind deals with Indian astronomy.


In a passage of the Taḥqīq mā li-l-Hind, al-Bīrūnī explains how his knowledge of the language grew while staying with Indian astronomers. See below pp. 74–75.


Sachau 1887: xv–xix. Sachau (1887: xiv) also assumes that al-Bīrūnī used a grammar book and a dictionary. See also Chatterji 1951: 86–87 and 95 on al-Bīrūnī’s knowledge of Sanskrit.


Pingree 1983: 353. Al-Bīrūnī titled his Arabic translation of the treatise Brāhmasiddhānta (see Verdon & Yano 2020: 60–62 and 68–71). See p. , below, on Pingree’s assessment of al-Bīrūnī’s knowledge of Sanskrit.


Taḥqīq (1958), pp. 7–12. With regard to the Latin transliteration of short vowels when the Arabic script does not specify them, I attribute the length of the vowels of the corresponding Sanskrit term to them. Similarly, diphthongs have been inferred from the spelling of the original Sanskrit terms.


Al-Bīrūnī appears to have transliterated the nominative plural masculine of the word purāṇa, even though the Sanskrit original, used with reference to the thus-designated works, should have read purāṇāni, i.e., the nominative plural neuter.


Sachau 1887: xxii–xxvii and 1888.


Verdon 2019a: 71–75.


See above p. .


Boilot 1955: 238–239, nos. 175–177. These translations are not extant, but the Elements and Almagest are mentioned in the Taḥqīq mā li-l-Hind (Taḥqīq (1958), p. 102.5.7; Sachau 1910: I/137).


Balty-Guesdon 1992: 141–146.


As quoted by Travis Zadeh (2011: 60).


Zadeh 2011: 60.


Fück 2012.


Devic 1878: 2–3.


Devic 1878: 90–91.


Reynolds 1858: 335–336.


Sourdel et al. 2012.


Bosworth 2011e: 57. See also Bosworth 1963: 101 and 2011: 57–59; Flood 2009: 4 and 78. See further Cappelletti 2015: 110. On slavery under the Ghaznavid see Bosworth 1963: 99–106. Richard M. Eaton also highlights the significance of the inclusion of Indians in the Ghaznavid army by stating “despite the dynasty’s rhetoric about defending Sunni Islam, religion posed no bar to military recruitment, as Indians had always been prominent in Ghaznavid armies” (2020: 34–35).


Said & Khan 1981: 89.


Rehman 1998, Rodziadi Khaw 2016: 142–144 and Shavarebi 2022.


See above pp. –.


Said 1973: 6.


Böwering 2012.


Grover 2006: 61; Clinton 2012.


Flood: 2009: 42.


Sachau 1887: xxiv and 1888: 37; Chatterji 1951: 93–94. Chatterji also observes various spellings in al-Bīrūnī’s Arabic transliterations of Sanskrit words that do not reflect the pronunciation of northern Panjab or the Gangetic Plain. This linguistic observation leads Chatterji to think that al-Bīrūnī interacted with people from regions of India other than those two (Chatterji 1951: 89). See also Sachau 1888: 5–6 and 10–41. Fabrizio Speziale (2010: 419–420) observed that later Muslim authors studied Indian sciences under similar conditions.


See chapter 1, section 2 of the present book.


Touati 2000: 12.


Elliot & Dowson 1867: 116.


In the Taḥqīq mā li-l-Hind, al-Bīrūnī writes: “[H]e incited me to write down what I know about the Hindus” (Sachau 1910: I/7). He also states in Al-āṯār al-bāqiya: “[A] learned man once asked me regarding the eras used by different nations” (Sachau 1879: 2). See also Kozah 2016: 15 and 34.


Nazim 1931: 158.


Bosworth 1963: 132.


See above p. . This episode is referred to by Minorsky (1951: 233–234), Shamsi (1979: 271) and Said & Khan (1981: 80, 82 and 222, n. 178). For a complete English translation of Al-ğamāhir fī l-ğawāhir, see Said 2001. See also Boilot 1955: 230, no. 156.


Bosworth 1991: 47.


Taḥqīq (1958), p. 456.12–13; Sachau 1910: II/134.


Taḥqīq (1958), p. 475.11–16; Sachau 1910: II/163. This passage shows that notion of impurity and pollution resulting from the contact with other castes or with foreigners (mleccha) was acknowledged by these Brahmins in the early eleventh century.


Taḥqīq (1958), pp. 452.5–459.5; Sachau 1910: II/130–139.


Taḥqīq (1958), pp. 75.11–80.1; Sachau 1910: I/99–104.


Mishra 1983: 103.


Taḥqīq (1958), pp. 29.19–30.1; Sachau 1910: I/39.


Taḥqīq (1958), pp. 24.4–34.4; Sachau 1910: I/33–45.


This observation parallels Thapar’s remark that early European thinkers chiefly based their research on India on their interaction with Brahmins and thus depended on the latter’s view of the Indian society (Thapar 2002: 9).


Sachau paraphrases this passage, as follows: “being a stranger among them and not acquainted with their peculiar national and traditional methods of science” (Sachau 1910: I/23).


I translate ʿilal (‮علل‬‎) as “defects,” whereas Sachau renders it by “the elements on which this science rests” (Sachau 1910: I/23).


Taḥqīq (1958), pp. 17.16–18.2; Sachau 1910: I/23.


Taḥqīq (1958), p. 118.7–9; Sachau 1910: I/152–153.


See below Section 2.3.


Minorsky 1948: 633.


KP, p. 167.8–11; Pines & Gelblum 1966: 309.


Taḥqīq (1958), p. 186.11–12; Sachau 1910: I/229.


See above p. .


Rehman 1979: 147–158.


Nazim 1931: 102. See also Cappelletti 2015: 109–110. Anooshahr (2021) discusses textual sources on the question of the political relationship between the Ghaznavids and local Indian rulers.


Ernst 2003: 175.


Verdon & Lončar 2016. Excavation reports are unpublished or inaccessible to me. On Wayhind, see Kimmet 2020 and Verdon 2021.


Rehman 1998: 472.


See Dar 1994 and 2001: 53–60.


Sachau 1910: I/116, 298, II/145 and 148.


Baloch 1973: 19–33.


See above pp. 28–29.


Scharfe 2002: 169.


Maḥmūd’s political interest is discussed in Section .


Ali 1967: 188–189. Said & Khan 1981: 84.


On al-Bīrūnī’s stay in Nandana, see Said & Khan 1981: 77–78. Baloch (1987) also wrote about al-Bīrūnī’s sojourn at Nandana and his calculations there.


See above Section 1.3.1.


Eaton 2020: 22. On the historiography of Maḥmūd’s raid on Somnāth, see Thapar 2004.


KP 5, p. 171.4–5; Pines & Gelblum 1966: 315.


See above pp. –.


Taḥqīq (1958), p. 105.1–6; Sachau 1910: I/135–136.


Boilot 1955: 200, no. 72.


RT IV.177, p. 50; Stein 1900: I/138.


RT VI.176–178, p. 97; Stein 1900: I/249; Majumdar 1957: 65; Pandey 1973: 94.


RT VI.300, p. 102; Stein 1900: I/260; Gopal 1989: 91.


Gopal 1989: 92.


Torella 1999: 555–557.


See above pp. –.


See below Section 6.3.3.


Taḥqīq (1958), p. 6.3–4; Sachau 1910: I/8.


Taḥqīq (1958), p. 21.16–17; Sachau 1910: I/29.


On the relationship between the Kitāb Sānk and the Kitāb Pātanğal according to al-Bīrūnī, see below Section 3.4.2.


Taḥqīq (1958), p. 30.10; Sachau 1910: I/40. See passage II in the Appendix (Taḥqīq [1958], pp. 30.10–34.4; Sachau 1910: I/40–44).


See chapter 3, section 2 of the present book for an overview of Sāṅkhya philosophy.


This definition echoes al-Bīrūnī’s definition of the knower (‮العالم‬‎) in KP 36 and 37 of the Kitāb Pātanğal (KP, p. 181.9–17; Pines & Gelblum 1977: 525). A similar description of the soul (‮النفس‬‎) is found in the following chapter of the Taḥqīq mā li-l-Hind (Taḥqīq [1958], p. 34.7–9; Sachau 1910: I/45).


GPBh, p. 4.4–5 on kā. 2.


Sachau 1910: I/40–41.


Here the reading should probably be sapiras (‮سپرس‬‎), as the corresponding Sanskrit word is sparśa, referring to the quality of tangibility.


For the related account in classical Sāṅkhya, see the commentaries on . 10.


Concerning the five gross elements (mahābhūta) and the five subtle elements (tanmātra), Sachau’s translation is as follows: “As these elements are compound, they presuppose simple ones which are called pañca mâtáras, i.e. five mothers” (Sachau 1910: I/42). This translation leaves space for some confusion about how al-Bīrūnī’s understood the exact sequence of the subtle and gross elements. A literal translation of the above runs as follows: “These elements (i.e., the gross elements) are composite. Thus, they have simple [ones], which precede them and are called pañca [tan]mātra (‮پنج ماتر‬‎), meaning ‘five mothers’ (‮أمّهات خمسة‬‎)” (‮هذه العباصر مركّبة فلها بسائط تتقدّمها تسمّىپنج ماترأى أمّهات خمسة‬‎) (Taḥqīq [1958], p. 32.3–4). This alternative translation renders the meaning of the fifth form of the verbal root q-d-m (‮تتقدّمها‬‎) as signifying “they precede them,” an important specification in order to precisely understand al-Bīrūnī’s description.


Taḥqīq (1958), p. 33.11–12; Sachau 1910: I/43–44.


Interestingly, al-Bīrūnī here provides an Arabic transliteration of the inflected form indriyāṇi (nominative plural) of the neuter stem indriya.


See Verdon 2019b.


See for instance below pp. 143–144 and 148–149.


Sachau 1910: I/xlv. As for Jainism, Sachau notices that a few features of al-Bīrūnī’s work may go back to an encounter with Jainism (Id. I/xl). For instance, al-Bīrūnī uses the word Jina to refer to the Buddha when he quotes the works of Varāhamihira (Id. I/119) and Brahmagupta (Id. I/243). See also Bhattacharyya 1964: 54.


Taḥqīq (1958), p. 206.4–5; Sachau 1910: I/249. Īrānšahrī was a Persian scholar from Nišāpur who lived in the second half of the ninth century. He inspired al-Bīrūnī’s works, but also those of the physician and philosopher Moḥammad b. Zakariyyāʾ Rāzi (b. 854).


Prior to this passage, al-Bīrūnī discusses the general antipathy of Indians toward foreigners, i.e., mleccha (‮مليج‬‎).


Mosul is an ancient city situated in northern Iraq.


Taḥqīq (1958), pp. 15.14–16.3; Sachau 1910: I/21.


See also Maclean 1989: 5.


Watters 1904: I/202. In Al-āṯār al-bāqiya, al-Bīrūnī also mentions the decline of Buddhism in Central Asia (Sachau 1879: 188–189; quoted in Elverskog 2010: 51).


For instance, the situation was different in Sind and Balkh (Maclean 1989: 22–77; Arezou 2017: 44–47), and Buddhist traditions also survived for a longer time in lower than upper Sind (Maclean 1989: 52–57).


S.J. Heras (1951: 119–123) argues that the idea of God conveyed by al-Bīrūnī in the Taḥqīq mā li-l-Hind is connected with Advaita-Vedānta. According to al-Bīrūnī, educated Indians believe in a unique and all-pervasive God who adopts plural manifestations in the world, whereas uneducated Indians only see the manifestations and are not able to conceive an abstract notion of God. The concept of God that al-Bīrūnī attributes to educated Indians, however, does not necessarily reflect that of Advaita-Vedānta. As seen in Chapter 4 below, much of al-Bīrūnī’s interpretative work consists in reducing cultural gaps between his own culture and the culture he encountered. His interpretation of God in the Taḥqīq mā li-l-Hind is also indebted to this tendency according to which the Indian God would resemble Allah in some of His characteristics (see below for al-Bīrūnī’s interpretation of Īśvara, pp. –). Moreover, al-Bīrūnī never explicitly mentions a work related to Advaita-Vedānta. It is, however, possible that Advaita ideas were spread among the Indian thinkers whom al-Bīrūnī met. These thinkers would have thus transmitted such ideas indirectly, which al-Bīrūnī conveyed then in his book, without identifying these ideas as such.


Al-Bīrūnī generally employs the Arabic term al-ḵalāṣ (‮الخلاص‬‎), meaning “liberation” or “deliverance,” to refer to the Sanskrit terms mokṣa (liberation), kaivalya (isolation) or apavarga (emancipation). In the Taḥqīq mā li-l-Hind, he also provides a transliteration into Arabic script for the word “liberation,” i.e., mūkša (‮موكش‬‎). See Taḥqīq (1958), p. 53.8–9; Sachau 1910: I/70.


In Sachau’s edition, the reading is nāyabhāša (‮نايبهاش‬‎), instead of nāyayahaša (‮نايَيَهاَش‬‎) as in the Hyderabad’s edition.


Sachau reads the term as ğaymin (‮چيَمن‬‎), instead of chiyaman (‮جيمن‬‎) as in Hyderabad’s edition.


The Arabic Jupiter here stands for the Indian Bṛhaspati who is considered the founder of the Lokāyata school of thought.


Sachau does not identify this work (1910: II/300). However, the Arabic transliteration from Sanskrit (agasta mata, ‮آگست مت‬‎) seems to render Agastimata, which is the title of an Indian treatise on gemstones composed before the tenth century. Unless there was a further work with this title, al-Bīrūnī’s description does not fit the content of this treatise.


Canopus corresponds to Agasti in Indian astronomy.


Taḥqīq (1958), p. 102.1–10; Sachau 1910: I/131–132.


Al-Bīrūnī’s Viṣṇudharma is to be identified with the Sanskrit Viṣṇudharmottarapurāṇa; see Gonda 1951: 111.


Taḥqīq (1958), p. 200.3–6; Sachau 1910: I/242.


Taḥqīq (1958), p. 205.12–14; Sachau 1910: I/248.


Taḥqīq (1958), p. 196.14–15; Sachau 1910: I/238.


Taḥqīq (1958), p. 237.11–15; Sachau 1910: I/284.


Taḥqīq (1958), p. 195.1–2; Sachau 1910: I/237.


Taḥqīq (1958), p. 349.6–8; Sachau 1910: II/11. See the full passage above, pp. –.


Taḥqīq (1958), p. 18.5–7; Sachau 1910: I/24.


Taḥqīq (1958), p. 5.11–12; Sachau 1910: I/7.


On literate circles see Touati 2000: 108.

  • Collapse
  • Expand