SOURCE: "Al-Bīrūnī and Islamic Mysticism," in Hamdard Islamicus, Vol. 1, No. 1, 1978, pp. 53-70.

[In the essay that follows, Lawrence contends that al-Bīrūnī's account of Islamic mysticism, specifically Sūfism, exemplifies his approach to knowledge in general—careful observation and documentation in an attempt to comprehend the unfamiliar.]

Al-Bīrūnī's acquaintance with the concepts of Muslim mystics (hereafter known as Sūfīs) may have been extensive or restricted. Historically, his immediate sources of information on Siufis and Siifism are not known. Yet al-Bīrūnī's encyclopaedic presentation of Indian thought, known as kitāb al-Bīrūnīfītahqīq mā lil-Hind min maqāla maqbūla fi'l 'aql aw mardhūla or simply the India,¹ leaves no doubt that whatever the scope of his acquaintance with tracts on tasawwuf their principal value for him is functional: to provide points of reference within a tradition already known to his Muslim readers, making possible the comparison of tangentially Muslim statements with references from Jewish, Persian and Christian but mostly Greek sources, for the ultimate purpose of paralleling and, hopefully, elucidating Indian religious beliefs. As a comparativist al-Bīrūnī is wide-ranging in his repeated references to non-Indian authors and their writing, but his single, constant focus in the India is Hindū thought and its comprehensibility to his Muslim readers. It is noteworthy, therefore, that the India does not include an independent treatment of Sūfism, nor does it provide an attempt to evaluate the principal tenets of tasawwuf vis-ā-vis the other branches of institutional Islam, Sunnī or Shī'ite.

Outside the India al-Bīrūnī makes few references to Silfism. In the Chronology there is a solitary exposition of the teaching of Husayn b. Mansiir al-Hallij,² and there is a further allusion to Sufi leaders in the Kitāb Bātanjal³ To determine al-Bīrūnī's use of characteristic Sufi expressions, therefore, one must consult the numerous but cryptic India passages pertaining to tasawwuf

Ideally our task ought to be twofold: (1) to examine explicit reference to Siufi thought—whether the name of a particular writer is provided, or, as in most cases, authorship remains anonymous; and (2) to determine the use of Sūfī concepts or technical terms in the general context of al-Bīrūnī's discursus on religious traditions. This second level of investigation would be especially profitable if directed not only to the India material but also to the Kitāb BātanjaL⁴

Yet, because of the frequency and complexity of al-Bīrūnī's references to Sūfism in analyzing Hindū thought, the present essay will concern itself mainly with the India passages where explicit reference is made to Sūfī patterns of belief. If we presume to suggest that human reason and historical imagination will lead us to understand what an author of nearly 1000 years antiquity and veneration intended to say about Sūfism, then at a later date we might assay the further task of laying bare Sūfi categories dimly reflected in the Kitāb Bātanjal as well as the India.

It is in the second through the seventh chapters of the India that we find most of the references to tasawwuf⁵ appearing in several contexts to illustrate a variety of doctrinal positions. The intention of these chapters is to summarize the Hindū outlook and thereby provide a frame of reference for the reader who ventures into the maze of mathematical and scientific data that follow. With respect to the span of topics included in the India, Chapters 2-7 appear to be straightforward and easy to understand. In themselves, however, they telescope a vast segment of medieval thought into a very few paragraphs, with the result that it is no minor feat to unravel the original context and intention of particular passages—whether they pertain, as in most cases, to the views of the ancient Greeks or, on occasion, to the views of the Sūfis.

The chapter headings supply minimal signposts for the unsuspecting reader. After a general introduction (Chapter 1). al-Bīrūnī explains the structure of Hindū belief under six topical rubrics:

1. —God (ch. 2)
2. —the interaction of intelligible with sensible objects (ch. 3)
3. —the connection of soul with matter (ch. 4)
4. —metempsychosis (ch. 5)
5. —cosmology (ch. 6)
6. —liberation (ch. 7)

The categories, in fact, overlap and the chapter divisions only roughly correspond to their content. It would be possible to label the same material as four aspects of Indian thought, with the following distribution:

GOD—as transcendent being (ch. 2)

GOD—as human creation (ch. 3a)

SPIRIT—as apogee of Samkhyan thought (ch. 3b)

SPIRIT—as corporeal and, hence, contaminated (ch. 4)

TRANSMIGRATION—its conceptual basis (ch. 5)

TRANSM[GRATION—its topography (ch. 6)

SALVATION—(ch. 7)

The glaring distinction of Hindū from Muslim belief occurs in the third category, and al-Bīrūnī has underscored and memorialized the distinction by his oft-quoted remark:…⁶ "Metempsychosis is the password of Hindū belief." Yet even on this topic Hindūs are not adjudged to be unique, for it is in his prefatory remarks concerning the Greeks that the Ghaznavid savant first applauds the ancient philosophers as those who endeavored to discover the truth⁷ and then compares them to Sūfīs and Christian sectaries as partners with the Hindūs in propounding the twin beliefs of transmigration and unification.⁸

Transmigration only becomes the central topic for consideration in two chapters of the India (5-6), while unification is not stressed till the discussion of salvation in Chapter 7. Together, however, they become the organizing principle for the material on Hindū doctrinal beliefs in the initial chapters of the India, and it is in relation to them that the excerpts from Greek sources assume a primary importance, while Süfism becomes an adjunct discipline, deriving even its valuation from the treatment of the Greek material.

Consider, for instance, Chapter 1. The pagan Greeks are discussed at length. According to al-Bīrūnī, they are like the Hindūs in two respects: (1) their upper and lower classes have more in common cross-culturally (Greek to Hindū, Hindū to Greek) than they do linearly (Greek to Greek, Hindū to Hindū), for in both cases the upper classes are philosophers and/or scientists, while the masses are ignorant rabble:⁹ as heathens, all Greeks and all Hindūs are alike since they represent a common deviation from the truth.¹⁰ The phrase 'deviation from the truth' … sounds like a canon of Muslim orthodoxy, and whether al-Bīrūnī genuinely holds such a conviction or is advancing it to please the monarch (Mahmiud of Ghazna) under whose patronage his India was completed¹¹ is a moot question. If all heathens are lumped together as deviates from the truth, then Sūfi as well as Hellenistic categories of thought are implicitly condemned. Such a judgment is given added weight when one realizes that nowhere in the India is Süfism cited except as a serial addition to examples either of Indian or Greek speculation.

It would not be amiss to say that the introductory remarks in the India concerning Greek thought lead to two seminal insights into al-Bīrūnīmf's perception of Süfism. First, while all Greeks, Indians and Süfis are condemned for their collective, corporate expression of False Belief, some individual Greeks but no Sūfīs are partially extolled for their aspirations to understand the Truth. There is no Süfi counterpart to the tribute paid to Socrates who, though suffering at the hands of his ignorant countrymen, died, according to al-Bīrūnī, as one who "did not recant from the Truth".¹² Secondly, it is al-Bīrūnī's own vision of the Truth which not only governs his life and work but determines the spirit of his inquiry into the Indian mind. For al-Bīrūnīmf the truth of Islam is above question; he does not subject it to the same scrutiny which prompts him to lay bare the premises for Indian belief in transmigration and unification. Both the India and the Chronology contain scattered references to the historical cleavages within the umma, but they are few and seldom polemical in intent.¹³ When excerpts from Süfi writings appear in the India, as they often do, their proponents are not castigated; only the Chronology depiction of extreme Süfism, as attributed to al-Halläj and his followers, seems to have piqued al-Bīrūnī.¹⁴ The generally ironic approach to sectarianism which characterizes his work appears to derive from an unquestioning acceptance of his inherited faith, coupled with a dogged pursuit of the highest Truth.

The larger number of references to tasawwuf rather than to other Muslim traditions—sectarian or orthodox—in the India cannot, therefore, be attributed to al-Bīrūnī's acceptance of Sūfi views; rather it is the fundamental parallelism which he perceives between Sūfīs, Greeks and Hindūs, especially on the pivotal topics of transmigration and unification, that prompts al-Bīrūnī to make frequent mention of the mystical branch within Islam. In Chapter 2, when the topic under surveillance is God as transcendent being, reference to Sūfism would be cumbersome at best and hence is omitted. Yet in introducing the Hindū view of the origin of matter, in Chapter 3 al-Bīrūnī can and does suggest a Sūfi parallel for the first time in the India. Unfortunately, the context obscures the exact referent which he had in mind. After setting forth the Greek postulate of the First Cause as a self-sufficient and independent entity, al-Bīrūnī adds, almost parenthetically, that the Sūfī also adhere to this belief¹⁵ But which Sūfis? In which writings? And to what extent? We are not told the answers to these questions. Subsequently, al-Bīrūnī refers to the Sūfi affirmation that the seeker who turns to the First Cause will after intermediate stages of ascesis become united with it. In all likelihood, this second reference to tasawwuf in Chapter 3 interprets the first, linking both to the central concept of unification … mentioned in the Preface as common to Sūfis, Christians and Greeks as well as Indians.

Wedged in between the two citations to Sūfi tenets in Chapter 3 is a lengthy digression the purpose of which is not immediately clear. After the initial mention of the word Sūfi, al-Bīrūnī elaborates its meaning. The excursus does not develop the particularity of Sūfi thought, at the same time that is remote from Hindū parallels. It does suggest, however, that in a mistaken line of reasoning the word Sūfi can be linked to Greek philosophy, thus implying that the Sūfis themselves felt dependent on the modes of rational discourse as well as the popular beliefs emanating from Athens. Such a negative valuation of tasawwuf as a tradition that derives not only its substance but its name from a non-Islamic source is partially offset by the concluding remark of al-Bīrūnī's etymological excursus. Abū'l Fath al-Bustī is quoted, and it appears that his view echoes al-Bīrūnī's own judgment, namely, that the word sūfī derives from a word-play on sāfī ('pure') and hence refers to 'a class of (pure) thinkers'.¹⁶

In the remainder of Chapter 3 of the India there is no further reference to Sūfism. Only in discussing the concept of God and particularizing the names by which various peoples refer to Him does al-Bīrūnī give as the canonical Muslim definition of God: … i.e., God is pure Truth. Elsewhere in both the India and the Kitāb Bātanjal al-Bīrūnī uses al-haqq as the conclusive and distinctly Muslim way of describing God. The emphasis is curious, and may exemplify the indirect influence of tasawwuf on al-Bīrūnī's vocabulary about which we spoke at the outset of this investigation.¹⁷ Though the term al-haqq recurs in the Hellenistic as well as the Sūfī traditions, as a name of Allāh it becomes emphasized in Sūfī vocabulary after the time of al-Hallāj (d. 922 A.C.) and seemingly as a result of his influence.¹⁸ While it is unlikely that al-Bīrūnī would trace his definition of the divine to al-Hallāj or any Sūfī author, it is of interest that he knows of al-Hallāj and quotes him directly in the Chronology and indirectly in the India,¹⁹

In the Preface to the India al-Bīrūnīl had declared his intention to use Sūfī teaching to explicate the concept of transmigration as it appeared in the Greek and Hindū traditions. The next references to Sūfism, accordingly, come at the end of Chapter 5 as doctrinal tenets corresponding to the philosophical bases for transmigration which up till then have been elaborated only with reference to Greek and Hindū material.²⁰ The remarks are brief but specifying. The Sūfīs, according to al-Bīrūnī, maintain that this world is a sleeping soul and that the (other) world is a waking soul. They also assert that God … is immanent in certain places—in heaven, for instance, in the seat and on the throne.²¹

There are several aspects to these observations that merit further comment.

1. The metaphor of the two worlds as sleeping and waking souls respectively is most frequently cited in the literature of Hellenistic and Iranian gnosticism. Both the Great Book of the Mandaeans and the Corpus Hermeticum resonate with references to sleep as the condition of the soul in matter; its moral turpitude is underscored by the sequential depiction of earthly existence as numbness, drunkenness and oblivion.²² In contrast, awakening from sleep and recovering from sloth is spurred by the call from without. Indeed the first effect of the call is always seen as wakefulness. Note, for instance, the command to the Primal Man:

   Shake off the drunkenness in which thou hast slumbered, awake and behold me!²³

2. While the general distinction between sleeping and waking does exist in tasawwuf, it is expressed with neither the frequency nor the poignancy characteristic of the Hellenistic/Iranian tradition. Early Sūfism stressed, instead, tawba or repentance as the initial, essential step … to attaining the Truth. It is noteworthy that The Recital of Hayy b. Yaqzān was the topic of a philosophical/mystical discourse by al-Bīrūnī's contemporary, Ibn Sīnā,²⁴ but it seems unlikely that al-Bīrūnī would have generalized as Sūfī in inspiration a solitary detail from a man whose works he otherwise appears to have ignored.²⁵

   In sum, lacking information about al-Bīrūnī's immediate source for the metaphor of the sleeping and waking souls, we must entertain the possibility that he is ascribing to tasawwuf a tradition that only vaguely pertains to the genre of mystic thought.

3. The subsequent reference to the seat and throne of God is also problematic—not for its location in Sūfī literature, which is secure by the time of al-Bīrūnī, but for its applicability in the sense which al-Bīrūnī suggests. Both terms, of course, derive from Qur'ānic observations about the Throne of God, in most instances described as 'arsh but in two passages (Q. II:255 and XXXVIII 3:4) as kursī. The discrepancy has invited extensive speculation which has exercised the imaginations of Qur'ānic commentators since the early centuries of Islam.²⁶ The deeper question concerns the interpretation of the Throne as a literal or metaphorical abode of God. In Sūfī parlance its characteristic intention seems to be metaphorical. Consider, for instance, Abū Yazīd al-Bistāmī, whose shatahāt al-Bīrūnī knew and quoted at least once in the India.²⁷ When Abū Yazīd reached the Divine Throne, it is said, he found it empty and threw himself upon it; in other words, sitting on the Throne was an allegory for his discovery of the divine element in himself.²⁸ The Throne can only be interpreted to connote hulūl al-haqq, or the indwelling of God, insofar as awareness of the Godhead potentially exists in the soul of every human being. Al-Bīrūnī, by contrast, implies that the Sūfīs believed the Throne and/or the Seat literally embodied the Divine Being, and in this sense his ascription is less appropriate to Sūfi teaching than to the views of the anthropomorphists which by the fourth century A.H. had come to be condemned as unorthodox.²⁹
4. The context of al-Bīrūnī's succinct references to tasawwuf at the end of Chapter 5 uncovers some of the ambiguity surrounding both sets of images. 'The same doctrine' of which al-Bīrūnī speaks at the outset of the final paragraph, is the doctrine that the transmigrating force of the soul impels rebirth from one body to another. Yet neither of the allegedly Sūfī tenets suggests metempsychosis so much as the desirability of ascent from the lower to the higher world. The frame of reference could as easily be a single as a multiple existence for sentient beings. Why then does al-Bīrūnī include these random offerings of purportedly Sūfī origin? It is possible that the preceding series of quotations from Greek writings, which are far more extensive and detailed than Sūfi references, here as elsewhere in the India, proved to be misleading. The two excerpts from Plato³⁰ clearly elaborate the manner in which souls migrate from to form, but in the final passage Proclus simply speaks of properties (knowing and forgetting) which govern the relationship between soul and body. In all probability, it was this last thought which, in its cleavage of soul from body, reminded al-Bīrūinī of the ascetical tradition within Sūfism, prompting him to conclude his introduction to Indian views on rebirth with some summary images that bear only a distant relationship either to actual Sūfī teaching or to the concept of metempsychosis.

When we proceed to Chapter 6 of the India, however, we find a lucid exposition of the mechanics of transmigration that defies comparison with any medieval Muslim text on the subject. One especially graphic quotation concerns the view of a mutakallim on the four stages of transmigration: naskh, maskh, faskh and raskh.³¹ Al-Shahrastānī mentions the same four stages without elaboration and attributes their professions to an extreme Shī'ite group, the Kāmilīya.³² Edward Sachau, however, assumes that the person to whom al-Bīrūnī alludes is a Sūfi, translating mutakallim as 'theosoph'.³³ Since the word has an ambiguous history,³⁴ the mutakallim in question could have been either Sūfi or Shī'ite, and al-Bīrūnī does not provide further clues as to the identity of his source. Yet the Sūfī affiliation seems questionable, especially since belief in transmigration did not appeal to certain, major expositors of tasawwuf till after al-Bīrūnī's time.³⁵

If we omit from consideration this last reference in Chapter 61 or minimize its relevance to Sūfism, we are left with no major statements of Sūfī doctrine in the India which parallel either Greek or Hindu views on metempsychosis. In itself this conclusion is not astounding, though it mildly contradicts the prefatory statement about the value of Sūfism for understanding the Hindū tradition. At the same time, it does point to the importance of that other aspect of Hindu thought for which Stfi examples were to be adduced, namely, unification. Briefly mentioned in Chapter 3³⁶ the topic of unification is not broached again till Chapter 7. The case for or against al-Bīrūnī's understanding of Sūfism rests squarely on the citations of this chapter, and we will now turn our attention to its contents.³⁷

Chapter 7—concerning Liberation or Salvation—is the final, the lengthiest and the most decisive chapter in the section of the India dedicated to expositing Hindū beliefs. Two structural features separate the tasawwuf material in this chapter from that which preceded it. Portions of two verses from the Qur'ān are explicitly cited, together with the Sūfi commentary on their content. Neither of them is identified with a particular author, but elsewhere in the same chapter two prominent Sūfis are mentioned by name, Abū Bakr al-Shibl and Abū Yazīd al-Bistāmī, and a sampling of the shatahāt ascribed to them is set forth. The inclusion of these amplified references to tasawwuf is in itself sufficient to underscore the importance of Chapter 7 for al-Bīrūnī: in describing the Indian doctrine of release or salvation, he wants to provide his Muslim readers with the most cogent and lucid examples from the cross-cultural material available to him.

The thread of cohesion linking together the Sūfī references in Chapter 7 is unification or ittihād. On the one hand, unification is treated in relation to the miraculous feats which precede it but are also superceded by it; on the other, unification is examined as an epistemological process for which the only imagic equivalent is light.

The two passages pertaining to siddhis, i.e., extraordinary powers which issue in miraculous acts, are separated by several pages in the India text,³⁸ yet they constitute a unit from the perspective of al-Bīrūnī's use of Sūfī categories and hence will be treated together.

The first passage appears at the beginning of Chapter 7, after al-Bīrūnī has elaborated a number of siddhis from the third chapter of the Yoga-sūtras.³⁹ His intention is to clarify the nature of detachment (Ar., istighnā, Skt., vairāgya): only the yogin who has achieved the siddhis may become detached from them and attain a still higher level of realization.⁴⁰

Al-Bīrūnī sees a parallel between this teaching and the Sfti description of the gnostic … and the stage of gnosis.… According to the Stffs, in his view, there are functions and qualities appropriate to the two souls … the one eternal, unalterable, performing miracles, the other human, changing, coming into existence. Nothing further is said about how the two souls of tasawwuf relate to the yogin siddhis, and once again we have to search for Stfi antecedents outside the text of the India. The ruhan to which al-Bīrūnī refers are usually depicted in Sūfī manuals as the spirit … and the lower soul or nafs.⁴¹ Both are said to be subtle substances existing in a single body, and both are linked to apposite functions not dissimilar from those included in al-Bīrūnī's brief vignette,⁴² except for the performance of miracles. Early Sūfī writings do not stress wonder-making powers, and even if one were able to find a passage or two from accounts by pre-tenth century A.C. Sūfī enthusiasts which make mention of extraordinary feats, how would they measure up to the extensive faculties which are systematically elaborated as benefits of samyama throughout Book Three of the Yoga-sūtras? Clearly the two traditions of tasawwuf and Yoga have disparate emphases on the development of psycho-physical powers, and one might be led to question al-Bīrūnī's sound judgment as a comparativist if there were not a second passage in Chapter 7 of the India which also pertains to the siddhis and Sūfīs.

In the second passage,⁴³ as in the first, al-Bīrūnī is trying to explain how the transformed body of the yogin is said to become a vehicle for the soul instead of a barrier to it. Again, he reverts to Sūfism for substantiation of his basic point. Instead of drawing the contrast between the two levels of the soul, however, as he had done earlier, al-Bīrūnī explains the point analogically by citing a Sūfī story and then a verse from the Qur'ān (XVIII: 84), together with its commentary. The data is uneven. While the story itself is a mystery, being untraceable as well as anonymous, its effect is clear: to show that a student of tasawwuf in the state of intense prayer can experience a unification … of body and soul which allows him to die at will. The Qur'ānic verse, on the other hand, is clear in its origin and initial context but its applicability to Sūfism is less certain. The unnamed subject of the verse is Dhū'l-Qarnayn or Iskandar, and although it is apparent from the Chronology that al-Bīrūnī was aware of both Dhū'l-Qarnayn and his legendary exploits, there is no mention of details which would link him to the Sūfī tradition.⁴⁴ Perhaps in a commentary such as the Haqā'iq al-tafsīr of al-Sulami (d. 1021 A.C.),⁴⁵ the legendary properties ascribed to Iskandar were transferred to the Sūfī adept.… But it seems strange both that a commentator would have generalized the extraordinary qualities of a rare and mythical person such as Iskandar and that al-Bīrūnī would have accepted such attribution without question or notation. Moreover, as we mentioned above, reliance on magical feats did not appear to be a common element in the self-conception of early Sūfī communities.

Yet the question concerning al-Bīrūnī's source for this bloc of Sūfī material cannot obscure the fact that its inclusion in Chapter 7 places the earlier reference to siddhis and their Sūfī equivalent in a new perspective. What appeared in the first passage as an awkward or extraneous detail is here explicated with reference to one of the most miraculous figures in the Islamic tradition: Dhū'l-Qarnayn. Whatever the actual relation of the passage to tasawwuf, it is an ingenious citation on the part of al-Bīrūnī which at the least amplifies the ambiguity of his first reference to 'miracle-working' among the Sūfīs and safeguards his reputation as an innovative comparativist.

Unification as an epistemological process, the literary focus of which is light, describes the remaining catena of Sūfī references in Chapter 7 of the India.⁴⁶ They appear at the end of this chapter,⁴⁷ and seem to be shrouded by an initial error. After quoting the tree dialogue between Arjuna and Vasudeva, which could only hark back to Bhagavad Gītā, Ch. 15 al-Bīrūnī begins the first of his unprecented string of six Sūfī quotations by declaring: ilā tarīq Bātanjal, that is, 'related to the teaching of Patanjali is, is, etc.' Sachau says nothing about this seeming confusion of the Bhagavad Gītā and the Yogasūtras.⁴⁸ While it is possible that al-Bīrūnī made an outright error, it seems more probable that he is speaking of the Sāmkhyan doctrine common to all three of the major Hindū doctrinal texts known to him.⁴⁹ An essential element of Sāmkhyan metaphysics is the assumption of two worlds, prakriti and purusa (matter and spirit), which interrelate in the endless round of rebirth but separate in the state of release. In Patanjali's Yoga-sūtras as well as in the Bhagavad Gītā a frequent metaphor to describe the state of release is light, and it was perhaps the mention of al-anwār al-ilāhīya ('the divine lights') at the conclusion of the Gītā quotation which brought to mind Patanjali and the Sūfīs, producing the apparent error in the Arabic text.

Of the six quotations which follow four have been scrutinized by L. Massignon,⁵⁰ and it is valuable to consider their import with reference to his observations:

The first text (he states) is from Hallāj, and we have studied it in its place for its theory of the shahāda. It is more than the samādhi of Patanjali (as Bīrūnī implies), since it is, in addition to the renunciation of the soul (i.e., of the body by the soul), the actual transformation (of the soul) in God.

The second text, which is anonymous and probably later, brings to mind a commentary on the Anā'l-Haqq from the school of al-Hallāj.

The third, from Shiblī, is equally an elliptical condensation of the thesis of Hallāj.

The fourth, from Bistāmī, inspite of its extreme brevity, is monist only in appearance, though one may find some Hindū analogies in its method.

For the initial text, Massignon has provided a substantive and incontrovertible identification. It is of special interest that this Hallājian commentary on Q. III: 27 appears in al-Sulamī's Haqā'iq al-tafsīr,⁵¹ increasing the likelihood that the two Qur'ānic passages, with commentary, which are cited elsewhere in the India to illustrate Sūfī tenets⁵² may also be traceable to al-Sulamī.

At the same time, however, Massignon's analysis of the text itself seems heavily weighted in favor of al-Hallāj's theological independence from the Hindū parallels which al-Bīrūnī adduces. The issue at stake is the ultimate condition of the saved soul. Both Massignon and J. Houben⁵³ argue that Hindū ascesis provides for renunciation but no transformation. The evidence of the Yoga sūtras themselves, especially I:41 and IV:23, suggests the equivocal nature of such a judgment, and in the Kitāb Bātanjal⁵⁴ as well as the India al-Bīrūnī indicates that he, too, has perceived the transformational quality in the soul which Yoga mandates.

With respect to the second passage in this series, Massignon, by focussing on the source critical problem, neglects the distinctiveness of al-Bīrūnī's contribution as a student of comparative religions. The Sūfī narrative appears to describe the distinction between normative and meditative consciousness, but it can also be shifted, as al-Bīrūnī implies, to explain the difference between return into another bodily existence and release from rebirth, hence relating both the first and second passages to the interlocking Hindu conception of samsāra/moksa.

Similarly, both the third and fourth quotations appear to supplement and reinforce each other as still another explanation of the Sūfī approach to release. Both stress tajrīd or renunciation. While the excerpt from al-Shiblī provides a general description of the transformation that takes place through tajrīd, the shath of Abū Yazīd aphorizes the ultimate condition of the saved soul.

It is the example chosen by Abū Yazīd, the serpent casting off his skin, which has sparked much conjecture concerning Hindu influences on early Sūfī thought, and is worthy of brief exploration here. Even Massignon allows the possibility that Abū Yazīd's teaching may represent an instance of cross-cultural borrowing:⁵⁵ Zaehner, for his part, eliminates all semblance of scholarly modesty by asserting that in this shath Abū Yazīd is quoting his spiritual master, Abū 'Alī al-Sindī, who knew the Upanisads first-hand.⁵⁶ The geographical argument against Zaehner's unqualified assertions is lodged already in Massignon's footnote to the effect that there are two Sinds, one in Iran, another in India, casting doubt upon the allegedly Indian origins of Abū Yazīd's mysterious teacher.⁵⁷ The theological counter-argument is advanced by S. H. Nasr, namely, that certain aphoristic dicta 'are so universal and so deeply rooted in the texture of reality that they do not need to point to any historical borrowing whatsoever.'⁵⁸

The particular shath of Abū Yazīd al-Bistīmi cited in the India poses an additional, literary critical problem for Zaehner's postulation of Vedāntin influence on early Sūfism. The Brhadāranyaka Upanisad text, as allegedly summarized in the shath, is not contiguous. The verses which Zaehner cites are BU 4.4.7, 12. It is their fragmented juxtaposition which cloaks the dominant stress of this and virtually every Upanisadic passage: the search for the awareness 'I am He' comes from no other motive than the fear of rebirth. The shatahāt of Abū Yazīd, by contrast, fail to indicate that he supported belief in transmigration, any more than did al-Shiblī or al-Hallīj. Later Sūfis did occasionally subscribe to such belief, as we indicated above,⁵⁹ and al-Bīrūnī is already aware of the possible elision between Sūfi teaching and acceptance of rebirth, e.g., in the second of the six passages under consideration. However, it is a recurrent error on the part of Zaehner and others⁶⁰ who wish to connect Vedānta to early Sūfism that they overlook the absence of Sūfi references to the fundamental stress on samsāra in Indian thought, while simultaneously exaggerating the significance of parallel textual assertions pertaining to unification.

The final two quotations of Chapter 7 in the India are omitted from Massignon's analysis, and they also appear to have baffled Sachau.⁶¹

Once again, it seems probable that the source of this Sūfi commentary is traceable to al-Sulami's Haqā'iq al-tafsīr or to an early handbook on tasawwuf of comparable quality.

The dominant metaphor in the commentary and in the passage which follows is light: the heart becomes alive 'by the lights of knowledge' …, while it is attainment to 'the stations of light' … that brings the security from which there is no return of fear, i.e., no rebirth into the endless go-round of human existence.

Emphasis on the luminous quality of saving knowledge has as its immediate referent in the India the Bhagavad Gītā depiction of al-anwār al-ilāhīya⁶², at the same time that it invites comparison with the passage on al-Hallāj in the Chronology. According to al-Bīrūnī, al-Hallāj describes himself and is described by his followers chiefly with reference to light imagery. He is at once 'the beaming, shining light' … and 'the niche of light' …, of whom his followers could say: …

The ancient Creator, the Giver of Light, Who reveals Himself in every time and age, and in this our time (He has revealed Himself) in the form of al-Husayn ibn Mansūr.⁶³

Reference to light as an image of divine transcendence exists both in Sūfism and in ancient Iranian religious thought independent of Islam or Sūfism. The language of the final quotation, in particular, could be referred as easily to Manichaean or Zoroastrian antecedents as to any tenet of tasawwuf. Yet its inclusive quality provides the fitting culmination to an exposition of the Indian concept of salvation. Liberation or salvation as a goal has many expressions, i.e., stations of light. The specific image of stations of light suggests the hieratic pattern of stages in the Sūfī path, but it is equally true that any point of progress toward the light holds out the prospect of removing darknes, defeating death and, above all, avoiding rebirth.

In sum, the final catena of quotations in Chapter 7 of the India not only draws selectively upon major expositions of tasawwuf it also summarizes the intent of the preceding excerpts from a variety of Greek and Hindū authors. Unification is the answer to the body/soul, matter/spirit dilemma for Sūfis as well as for their philosophical counterparts in Athens and Benares. In the earlier pages of the India, where references to Sūfism are linked to belief in transmigration and miracle-making, the tone of al-Bīrūnī's comparisons is not as convincing as at the end of Chapter 7 when unification becomes the sole topic of explication and comparison.

Yet the daring of al-Bīrūnī's undertaking pales any effort to dwell on the imprecise edges which we may now detect in details of his analysis. That the India exists is itself testimony to the fertile mind of a scholar whose search for the truth knew few restraints, and the specific investigations of al-Bīrūnī's approach to Islamic mysticism, as reflected in the India passages explicitly linked to tasawwuf, must eventuate in a mixture of wonder and humility. For what we have attempted is the understanding of glimpses into al-Bīrūnī's use of Sūfī categories and quotations. It is evident, moreover, that al-Bīrūnī's references to Sūfism are themselves but glimpses into the vast tradition of Islamic mystical thought as it evolved up to and throughout his lifetime. Can we then claim to have achieved a modestly accurate portrait of al-Bīrūnī's comprehension of tasawwuf, or does honesty compel us to acknowledge that our efforts have produced—to adapt a phrase from Mullā 'Abd al-Rahmān Jāmī—lawā'ih-i lawā'ih, glimpses of glimpses?

Notes

¹ The quotations which follow will be given both in Arabic (Hyderabad, 1958) and in English (E. C. Sachau, trans., Alberuni's India, London, 1910, reprinted, Delhi, 1964).

² See E. C. Sachau, ed. Kitāb al-āthār al-bāqīya 'an al-qurūn al-khāliya (Chronologie orientalischer Volker von Alberūni), Leipzig, 1878, pp. 211-12; trans., The Chronology of Ancient Nations, pp. 194-195.

³a'lmmat al-sū fīya. See H. Ritter, "Al-Bīrūnī's Ubersetzung des Yoga-sūtra des Patanjali', Orlens 9 (1956), p. 167, and S. Pines/T. Gelblum, 'Al-Bīrūnī's Arabic Version of Patanjali's Yoga-sūtra', Oriens 20 (1967), p. 310.

⁴ Pines/Gelblum, op. cit., pp. 309, 317, 319, 321-2 (fns. 58, 137-8, 165, 188, 199) provide some examples of al-Bīrūnī's implicit reliance on the language of Muslim mystics, but the examples could be multiplied with reference to the other three chapters of al-Bīrūnī unprecedented—and unsurpassed—venture into the realm of Hindu-Muslim metaphysics.

⁵ A solitary exception is the brief allusion to Sūfism in the India, ch. 35 for the purpose of explaining the Hindū view of Purusa. See Hyderabad, p. 296; Sachau I, p. 351.

⁶ Hyderabad, p. 38, Sachau I, p. 50

⁷ (wa-in) taharr al-tahqīq. Hyderabad, p. 5, Sachau I, p. 8.

⁸ibid. Note that the terms are hulūl and ittihād for transmigration and unification respectively. Each word had a long history in Islamic thought, and the Christians undoubtedly meant something different by hulūl than did either the sūfīs or Hindūs (see L. Massignon—(G. C. Anawati), "Hulūl", EI² III, 570-71), but in this context hulūl appears to mean 'transmigration' or 'reincarnation' and to be linked with tanāsukh. See also in Kitāb Bātanjal, H. Ritter, op. cit., p. 167; Pines/Gelblum, op. cit., p. 309.

⁹ In Chapter 11 of the India al-Bīrūnī brings out the same distinction between educated and uneducated strata of ethnic/religious groupings in discussing Hindu idol worship. Within Islam a point of contrast to al-Bīrūnī's depiction of ignorant idolaters may be found in the "Arā' al-hind" section of al-Shahrastānī's Kitāb al-milal wal-nihal; see B. Lawrence, "Shahrastānī on Indian Idol Worship", in Studia Islamica, XXXIV, 1973.

¹⁰ Hyderabad, p. 18; Sachau I, p. 24.

¹¹ On the matter of al-Bīrūnī's personal religious belief, as reflected in his extant writings, see, for instance, the reasoned conjecture of E. S. Kennedy ("Al-Bīrūnī, Abū, Rayhān Muhammad ibn Ahmad" in C. C. Gillespie, ed. Dictionary of Scientific Biography, vol. II, p. 156):

As for religion, Brifirni was doubtless a sincere Muslim, but there is no firm evidence of his having been an adherent of any particular sect within the faith. In the Chronology (trans. pp. 79, 326), written at the court of Qābūs, are passages that have been interpreted as betraying a Shī'i (hence anti-Arab and pro-Persian) bent. On the other hand, the Pharmacology, compiled under Ghaznawid patronage, represents the author as an orthodox Sunnī. Probably these two situations reflect no more than the fact that the two patrons were Shī'i and Sunnī, respectively. While agreeing with the tone of Kennedy's remarks, we might note that the high praise accorded the Zaydīs and 'Alīds in the Chronology (see trans. pp. 98, 69 and 183 as well as 79, 326) is balanced by a condemnation of superstitious practices and ignorrant calculations among certain Shī'a (see trans. pp. 69,182-3 and 294). Other accounts of al-Bīrūnī's personal religious belief are to be found in S. H Barani, 'Ibn Sina and al-Beruni', Avicenna Commemoration Volume, Iran Society (Calcutta), 1956, p. 7 and S. H. Nasr, An Introduction to Islamic Cosmological Doctrines, Cambridge, Massachusets, 1964, p. 114.

¹² Hyderabad, p. 19: qadā nabhah ghayra rāji'in 'an al-haqq. (Sachau I, p. 25).

¹³ Exceptions are the ahl al-tashbīh or anthropomorphists and the Jabrīya against whose views he inveighs in the India (Hyderabad, p. 23); Sachau I, p. 31) as well as the uninformed Shī'a mentioned in the Chronology (see supra, fn. 11). In each case, however, it is not the sectarian posture but the superstitious idiocy of the individuals or groups which irritates al-Bīrūnī.

¹⁴ See trans., pp. 194-95; Ar., 211-12.

¹⁵ Hyderabad, p. 24; Sachau I, p. 33.

¹⁶ The Sūfīs themselves disagree about the appropriateness of using the word 'safā' except as an alliterative to explain the origin of their name. See, e.g., al-Kalābādhī, Kitāb al-ta 'arruf li-madhhab ahl al-tasawwuf (A. J. Arberry, trans., The Doctrine of the Sūfīs, Cambridge, 1935, pp. 9-10), where the link to safā is refuted. Yet al-Hujwīrī (R. A. Nicholson, trans., The Kashfal-mahjūb, London, 1911, pp. 30-41), while admitting that safā may not be linked etymologically to tasawwuf, avers that there is propaedeutic value in contrasting pure spirituality … with impure affections.…

Concerning the pun of sūfī and sophia, which suggests a direct derivation from Hellenistic sources, it is ironic that Sachau (I, 33) gives sūfvice sūf as the stem; for al-Bīrūnī begins by spelling the word with a sīn rather than a sād (see the doubli enstance of sūfīya' and sūfī' in Hyderabad 24), suggesting that he understood the etymological subtlety later elaborated by Noldeke, namely, that "the Greek sigma regularly became sīn (and not sād) in Arabic and that there is no Aramaic intermediary between sophos and sūfi" (See ZDMG, XLVIII, p. 45 and Shorter Encyclopaedia of Islam, "Tasawwuf", p. 579).

The other two 'false' derivations mentioned by al-Bīrūnī, namely, ahl al-sūffa and sūf have contrasting positions in the academic study of tasawwuf: while the first is still regarded as a transparent play on words to link sūfis with the Companions of the Prophet, the latter is generally acknowledged as the probable root for the term denoting Muslim mystics. See, e.g., the extensive tracing of textual references to sūf, sūfī and sūfīya in L. Massignon, Essai sur les origines du lexique technique de la mystique musulmane, Paris, 1954, pp. 153-56, where he argues that the reference to sūf as wool predates the Muslim use of the other two terms, though all three developed independently till they became interrelated in the sūfī vocabulary of the 4th cent. A.H.

¹⁷ See supra p. 2 and. n. 4.

¹⁸ See L. Massignon, Lexique technique, p. 203 and La Passion d 'al-Hallāj, Paris, 1922, pp. 472 and 625.

¹⁹ The pertinent passages from both works are discussed, pp. 23-8.

²⁰ There is also a reference to Mānī's exposition of metempsychosis, but it is treated as a direct development of Hindu concepts which Mānī learned during his alleged stay in India. See Hyderabad, pp. 41-2 and Sachau I,54-5.

²¹ Al-Blrfin goes on to hint at a strain of pantheistic belief among Sūfīs, but the remark is too general to invite further speculation about either his sources or his intentions. See Hyderabad, p. 44; Sachau I, pp. 57-8.

²² H. Jonas, The Gnostic Religion, Boston 1958, pp. 69-71.

²³Ibid, p. 83.

²⁴ See H. Corbin, Avicenna and the Visionary Recital (trans., W. R. Trask), New York, 1960; pp. 3-192.

²⁵ S. H. Barani, op. cit., p. 12 only found one instance where al-Bīrūnī, in his extant writings, acknowledged Ibn Sīnā, while the latter nowhere makes reference to his illustrious countryman and fellow scientist.

²⁶ The debate has been amply summarized in the article "Kursi", Shorter Encyclopaedia of Islam, pp. 286-87. The shift in orthodox views on this topic is addressed by A. J. Wensinck, The Muslim Creed, Cambridge, 1932, p. 116.

²⁷ Hyderabad, pp. 66-7; Sachau I, p. 88.

²⁸ See A. Badawi, Shatāhāt al-Sūfīya, Pt. I, Cairo, 1949; p. 128; and also F. Rosenthal, "The Empty Throne", Studia Islamica, XXXI 1970, pp 237-38. The list of references to the Throne elsewhere in the extant sayings of Abū Yazīd and other Sūfis could be extended, but their function is similar to the one here selected.

²⁹ See A. J. Wensinck, op. cit., p. 116.

³⁰ Hyderabad, pp. 43-44; Sachau I, pp. 56-7.

³¹ Hyderabad, p. 49; Sachau I, p. 64.

³²Kitāb al-milal wa 'l-nihal, ed. by W. Cureton, London, 1842, vol. I, p. 133; Religions partheien und Philosophen-Schulen, trans., T. Haarbrucker, Halle, 1850, pp. 201-02. See also E. L. Dietrich, "Die Lehre von der Reincarnation im Islam", ZRGG IX (1957), p. 148, fn. 55. The mnemonic value of these four terms is transparent; however, they have no parallel in the schemes of rebirth worked out in Hindu religious texts.

³³ Sachau I, p. 64.

³⁴ For a detailed exposition of the elusive Islamic references to the mutakallim, see S. Pines, "A Note on an Early Meaning of the Term Mutakallim", Israel Oriental Studies I, Tel Aviv, 1971; pp. 224-40 passim but espec. pp. 225-28. The plural form, mutakallimūn, appears in a technical usage once in the Kitāb Bātanjal (Ritter, op. cit., p. 169; Pines/Gelblum, op. cit., p. 312), but the referent is equally ambiguous.

³⁵ See Margaret Smith, "Transmigration and the Sufis", The Muslim World 30 (1940), pp. 351-57.

³⁶ See supra, p. 9.

³⁷ The other mention of doctrine in Ch. 6 is too scant to merit separate analysis; see Hyderabad 47, Sachau I,62. The purport is to stress the ethical quality of tasawwuf, which al-Bīrūnīri deffly conveys by describing the Divine as al-khayr al-mahd, the Absolute Good, rather than al-haqq al-mahd (Absolute Truth; see supra, p. 11). There is no value to citing this text in trying to ascertain the relation of Sūfism to transmigration in al-Bmrini's mind, but al-khayr al-mahd invites comparison with Vasudeva's instruction to Arjuna in the Bhagavad GE, A which begins: in kuntu turdiu al-khayr al-mahda, etc. See Hyderabad, p. 60; Sachau I, p. 79.

³⁸ See Hyderabad, pp. 52-3; Sachau I, p. 69 and Hyderabad, pp. 62-3; Sachau I, p. 83.

³⁹ Hyderabad, pp. 52-3; Sachau I, p. 69.

⁴⁰ Vyāsa's Commentary on Yoga-sūtra III. 51 holds up as the supreme yogin the one 'who has passed beyond that which may be cultivated', i.e. the siddhis. See J. H. Woods, The Yoga-System of Patanjali. repr., Delhi, 1966; p. 285.

⁴¹ See al-Hujwīrī's analysis of the entrapment arising from nafs and hawā (op. cit., pp. 196-207), in contrast to his later emphasis on the transcendent value of rūh.

⁴² Al-Kalābādhī (op. cit., p. 52) declares rūh to be "a light, a fragrant breath … through which life subsists, while the soul … is a hot wind … through which the motions and desires exist".

⁴³ Hyderabad, pp. 62-63; Sachau I, p. 83.

⁴⁴ See especially the translation of Chapter 4 (pp. 43 f.) entitled "The Different Opinions of Various Nations Regarding the King called Dhū al-Qarnaini or Bicornutus".

⁴⁵ L. Massignon (Lexique technique, pp. 359-412) includes excerpts from al-Sulamī's work, but unfortunately omits the Qur'ānic verses (XVIII: 84 and II: 73) which al-Bīrūnī cites in the India as instances of Sūfī tafsir which correlate with Hindū belief. A manuscript of the full text was not available to me.

⁴⁶ Omitted from consideration is the minor reference in Chapter 7 to the Sūfī view of 'ishq or love, which occurs in Hyderabad, pp. 57-8 and Sachau I, pp. 75-6. Its purpose is to confirm and extend the value of chastity in the Indian and Greek traditions. Though al-Bīrūnī uses the term in another sense than al-Hallāj (see L. Massignon and L. Gardet, "Al-Hallādj", EI IV. p. 103a, where he is alleged to have perceived 'ishq as a divine attribute of Essence), the point of his analogue is still valid, viz., that tasawwuf eschews 'ishq in the sense of passion or hawā since passion, as an activity of the lower soul or nafs, binds man to the realm of imperfection. See al-Hujwīri, op. cit., pp. 207-10.

⁴⁷ Hyderabad, pp. 66-7; Sachau I, pp. 87-8.

⁴⁸ See India II, p. 289, where the source is correctly given as the Bhagavad Gītā without further comment.

⁴⁹ In addition to the Bhagavad Gītā and the Yoga-sūtras of Patanjali, the book Sāmkhya, as al-Bīrūnī call[s] it, is frequently cited in the India As is the case with the Kitāb Bātanjal, al-Bīrūnī's text on Sāmkhya does not correspond with any of the classical texts relating to that school, but instead seems to hav[e] incorporated a commentary with the text and been set forth in dialogic form. See Sachau II, pp. 266-68.

⁵⁰ See Lexique technique, pp. 97-8.

⁵¹ See L. Massignon, La Passion d'al-Hallāj, II, p. 643.

⁵² See Sachau I, pp. 83 and 88.

⁵³ See his "Avicenna and Mysticism", in the Avicenna Commemoration Volume, Calcutta, 1956; pp. 209 and 215.

⁵⁴ See H. Ritter, op. cit., pp. 176-7 and 296.

⁵⁵ See Lexique technique, p. 98.

⁵⁶ R. C. Zaehner, Hindu and Muslim Mysticism, New York, 1969; pp. 98-9.

⁵⁷Lexique technique, p. 98, f.n. 3.

⁵⁸ S. H. Nasr, Three Muslim Sages, Cambridge, (Massachusetts), 1964; p. 158.

⁵⁹Supra, p. 17, f.n. 35.

⁶⁰ See, for instance, M. Horten, Indische Stromungen in der Islamischen Mystick, Heidelberg, 1928; passim.

⁶¹ See India II, pp. 289-90, where he resorts to speculation on the origin of the Qur'ānic verse before repeating, without elaboration, the tafsīr as given by al-Bīrūnī.

⁶² See supra, p. 23.

⁶³ Trans., pp. 194-5; Ar. text, pp. 211-12.