SOURCE: Watt, W. Montgomery. “The World of al-Ghazālī.” In Muslim Intellectual: A Study of al-Ghazali, pp. 7-24. Edinburgh: The Edinburgh University Press, 1963.

[In the following excerpt, Watt provides political, religious, and intellectual background to the education of al-Ghazālī.]

1 THE POLITICAL BACKGROUND

In a sense the background of the life of any individual is the whole previous history of his civilization. For an understanding of al-Ghazālī it will be sufficient to glance briefly at the history of the Islamic empire or caliphate from the death of Muḥammad in 632 to the birth of al-Ghazālī in 1058. In these four centuries four main phases may be distinguished, which may be labelled: conquests; conversion; disintegration; reconstitution. These phases follow one another chronologically, but overlap to some extent.

(1) THE CONQUESTS.

As Muḥammad lay on his death-bed in Medina an expedition was being assembled on the outskirts of the town whose task was in fact to open the way for the conquest of Syria. For the next two years, however, the Muslim leaders were busy suppressing revolts in Arabia, but in the following ten years the small state with its centre at Medina wrested the rich provinces of Syria and Egypt from the Byzantine empire and that of Iraq from the Persian empire, besides sending the latter reeling to destruction. A hundred years after Muḥammad's death the sway of his successor extended from north of the Pyrenees, through North Africa and the Fertile Crescent to Central Asia (Transoxiana) and the Punjab.

The effective control of these vast territories after the amazingly rapid conquest was made possible by the simplicity of the central organization. The Arabs constituted themselves into a vast army. At the extremities of their domains they had the help of auxiliaries from such peoples as the Berbers, but otherwise the army of Arabs did all the fighting and all the garrison-work. The local administrations were taken over and continued to function much as before. All that the Arab provincial governors had to do was to have direct supervision over the army and then to see that the non-Arab local administration was effective and handed over the due taxes.

The head of this state was called the caliph or successor (sc. of Muḥammad), and had inherited the latter's administrative but not his prophetic functions; the state is correspondingly known as the caliphate. From the description given it will be seen that it is essentially an Arab-Muslim military aristocracy; or rather, only those who are Arabs and Muslims are full citizens, serving in the army and in return drawing an annual stipend. The non-Muslims were related to the Muslim government not as individuals but as groups, later known as millets, and usually with a religious basis; e.g. the Christians of Jerusalem or the Jews of Iraq. Such a group had internal autonomy under its religious head, who was responsible to the government for handing over the taxes. Since it was a matter of honour for the ruler to make the official protection of such groups effective, there was practically no religious persecution. Yet the suggestion that these “protected persons” were second-class citizens meant that there was a constant pressure on them to become Muslims. On the whole the system has worked well and made life tolerable for millions; but it has tended to “freeze” small groups and prevent their assimilation in the larger whole except at a very slow rate (by conversions to Islam). The present troubles with minorities in the Middle East are largely due to the breakdown of the millet system of the Ottoman empire.

(2) CONVERSION.

Islam was by tradition a missionary religion, and was, at least implicitly, of universal validity. Because of its Arabic origin, however, there was a tendency to think of it as primarily for Arabs. This tendency was reinforced during the first century of the caliphate by the desire of the Arab Muslims to retain their privileged position as first-class citizens. Little effort was made in the early decades to convert non-Arabs to Islam. When non-Arabs insisted on becoming Muslims, whatever their motives may have been, they had to be attached to Arab tribes as “clients”. This still had a suggestion of inferiority. As the number of non-Arab Muslims increased, their discontent with their status and demand for equality was one of the factors behind the movement which replaced the Umayyad caliphs of Damascus (who had ruled from 660 to 750) by the ‘Abbāsid caliphs of Baghdad. This change was not simply a change of dynasty; it was a change of the basis of the caliphate. The body politic was now more explicitly based on Islamic principles and regarded as a “charismatic community”;¹ and all Muslims, whether Arab or not, were full citizens. The establishment of the ‘Abbāsid caliphate thus reflected the fact that many non-Arabs had been converted to Islam.

Yet the change of dynasty also meant in various ways a return to Persian ideas of autocratic government. Under the Umayyads power had been shared between the new Islamic aristocracy (who received higher stipends because they or their ancestors had become Muslims at an early date) and sections of the old Arab aristocracy. At many points actions had been based on traditional Arab political ideas, derived from experience with tribes and confederations of tribes; but in several ways this was unsatisfactory, and unsuited for a vast empire. Under the earlier ‘Abbāsids power was almost exclusively in the hands of the caliph and his court. Since membership of the court was virtually in the gift of the caliph, this meant that power was in the hands of the caliph and one or two other men, such as the Barmakid viziers; how far the caliph had to share his power depended on his strength and capacity for controlling affairs. Within the court circle, that is, within the ruling institution, there was practically no check on the autocratic decisions of the caliph; and contemporary chronicles depict a naked struggle for power in which nothing was barred. On the other hand, the relations between the ruling institution and those ruled were largely determined by Islamic principles as stated in the Sharī‘a or revealed law. The general acceptance of Islamic principles outside the court circle produced during the next century or two a high degree of homogeneity in the vast and varied empire.

(3) DISINTEGRATION.

After the first enthusiasm had waned the ‘Abbāsids found it increasingly difficult to exercise effective control over their domains. Provincial governors had to be given large powers, including the command of considerable armies. If they disliked some order from the caliph, they could hardly be forced to obey it. They tended to present the caliph with a series of faits accomplis, such as the extension of the boundaries of their province, which he was obliged to ratify. At length demands came that a son should succeed to the governorship, and the caliph had to accede. Thus there came into being local dynasties, for all practical purposes autonomous, but making a formal acknowledgement of the supremacy of the caliph. This description is specially applicable to the east, where there are four dynasties which deserve to be mentioned.

(A)

Tāhirids. Five men (four generations) of the Tāhirid family maintained themselves as governors of Khurāsān from 820 to 872. From the standpoint of the present study it is worth noting that the Tāhirids, by making Nishapur their capital, gave a fillip to its development as an intellectual and cultural centre. Their downfall resulted not from any action of the caliph but from military defeat by the first of the Saffārids.

(B)

Saffārids. Three men of the Saffārid family, starting shortly before 868 from the governorship of Sijistān (roughly southern Afghanistan), extended their rule (by 872) to most of southern and eastern Persia up to the Oxus, and maintained themselves there until about 903.

(C)

Samānids. The Samānid family is reckoned as having ruled from 874 to 999, and has a complex history which need not be described here. The chief basis of their power was Transoxiana, and their eastern capital, Bukhārā, became a literary and cultural centre of great brilliance.² After they had wrested Khurāsān from the Saffārids (900-910) Nishapur became their second capital, not far behind Bukhārā in the splendour of its intellectual life.

(D)

Ghaznavids. The Ghaznavid dynasty (976-1186) was of Turkish race, being descended from officers in the Samānid armies. Subuktigīn became governor in the mountain town of Ghazna (about a hundred miles south of Kābul in Afghanistan), and extended his power both towards India and into eastern Persia. His son, Maḥmūd of Ghazna (regnabat 998-1030), repudiated Samānid suzerainty, was appointed governor of Khurāsān and Ghazna directly by the caliph, and made great conquests in India. Soon after the death of Maḥmūd, however, the dynasty began to be deprived by the Seljūqs of its domains in Persia and Transoxiana, so that from about 1050 its rule was restricted to Afghanistan and India.

Further were there were small dynasties which developed from provincial governorships and continued to acknowledge the caliph of Baghdad. In the west, however, there were also actual losses of territory. A few years after the overthrow of the Umayyad caliph by the ‘Abbāsids, a member of the Umayyad family became independent ruler of Spain, though without claiming to be caliph. Such a claim was first made by the Fāṭimids, a dynasty which established itself first in Tunisia in 909, and then in 969 transferred the seat of its power to Egypt. The Fāṭimid rulers claimed to be the rightful caliphs of the whole Islamic world, and sent emissaries into the ‘Abbāsid domains to preach revolution. No more need be said about the Fāṭimids here, since their propaganda (also known as Ismā‘īlite or Bāṭinite) became a major concern for al-Ghazālī (chapter IV).

(4) RECONSTITUTION.

The word “reconstitution” is not altogether satisfactory as a description of the fourth phase of the caliphate, but it is convenient to have a single word. In this phase the caliph loses most of his remaining power, though he retains his position as a figurehead with certain official functions and dignities. Real power passed into the hands of a series of warlords, who eventually came to have the title of “sultan”. The first of these war-lords was Ibn-Rā'iq, who entered Baghdad at the head of an army in 936 and simply took over the machinery of government from the caliph's vizier. As a Muslim historian puts it:³

“From this time the power of the viziers ceased. The vizier no longer had control of the provinces, the bureaux or the departments; he had merely the title of vizier, and the right of appearing on ceremonial days at the Palace in black with sword and belt.”

Ibn-Rā'iq held this lofty position for less than two years, but in 945 Baghdad was captured by the Buwayhid (or Būyid) family—chiefs of a warlike highland tribe from Daylam, at the south of the Caspian Sea—who assumed the reins of government, and held them, though latterly with a slackening grip, until 1055. Their direct rule extended over Iraq and a large part of Persia, but provinces were entrusted to different members of the family, and these did not always see eye to eye.

The Buwayhids eventually fell before another family of war-lords, the Seljūqs, who, supported by Turkish tribesmen, first made themselves masters of Khurāsān, and then in 1055 established themselves in Baghdad. At its widest extent their empire was much greater than that of the Buwayhids, including Syria in the west and Transoxiana and the whole of Persia in the east. This was the situation during the maturity of al-Ghazālī, but before his death in 1111 the central government was weakening and it eventually disintegrated in 1157. This is as far as we need follow the history of the caliphate.

This phase of reconstitution has various aspects. While in one way it was the end of the rule of the caliphs, in another way it was a restoration to the central government of the territories directly under the caliph. In this new central government the place of military power was more explicit. The early conquests had been made by a citizen army, but in course of time a citizen army was shown to have disadvantages. In any case, after conversion became frequent there were too many citizens for the army. In practice it was found more satisfactory to have mercenaries, though this meant that the officers of the mercenaries might have undue power. It was becoming clear that political power depended on military backing. Those who were successful in the struggle for power, like the Buwayhids and the Seljūqs, were groups of men—not isolated individuals—who had effective military support that was in part independent of monetary payments. Political power partly also depended on the acquiescence of the citizens, and this was gained by recognition of the Islamic basis of society—acknowledgement of the caliph, participation in worship on certain occasions, continuation of courts applying the Sharī‘a. In major political decisions, however, and in the functioning of the court Islamic principles counted for nothing.

Despite this apparently unsatisfactory state of affairs (at least from a theoretical standpoint), the earlier part of the Seljūq period, especially the reigns of Alp-Arslān (1063-72) and Malik-Shāh (1072-92), was a time of comparative peace and prosperity and of great cultural achievement.⁴ To this happy condition the wise and efficient vizier of these two sultans, Nizām-al-Mulk, made an outstanding contribution. Though nominally subordinate to the sultan, he was practically all-powerful during these thirty years.

2 THE RELIGIOUS AND INTELLECTUAL BACKGROUND

The religion of Islam in its earlier forms was adapted to the social and intellectual needs of Mecca, Medina and Arabia.⁵ But the framework of material circumstances in which it had to function even under the Umayyad caliphs was entirely different from that of Muḥammad's closing years.

The first phase of development, the conquests, quite apart from the effects on the subject peoples, involved a vast social upheaval for the Arabs, that is, the Muslims. The old tribal and clan system broke down; and, since it was through the tribe that a man's life became meaningful, this led to a religious as well as a social crisis. An important section of the Arabs dealt with this crisis by substituting for the tribe the Islamic community. Life became meaningful for them through membership of this community, since it was divinely founded and was living in accordance with divinely-given mores. But the question of how to deal with those who transgressed God's commands proved intractable, and there was much bitter argument before it was solved. In the end, however, a way was found by which the whole community, despite the presence of sinners in it, could be regarded as a “saving sect”, so that membership led to everlasting bliss.⁶

The phase of “conversion” was a piece of social adjustment following on the incorporation of vast territories and their inhabitants in the Islamic empire. While some material self-interest may have been a factor in conversion, the major factor was perhaps the religious one—the attractiveness of the dynamic image of the Islamic community as a charismatic one. Men felt they wanted really to belong to this, not just to be loosely attached to it. The conception of the Islamic community as charismatic, originally developed for Arab tribesmen whose tribe had broken down, was further developed by the non-Arab Muslims. The distinctive excellences of the community, especially its possession in the Sharī‘a of a divinely-revealed law or rather set of practices, were linked with its charismatic nature. Zeal for the charismatic community was an important factor behind the incredible intellectual efforts expended in the elaboration of the Sharī‘a.

In the course of elaborating the Sharī‘a something else was also done. Many of the new converts came from a higher cultural level than the Arabs, and naturally retained most of their culture. The pious scholars in whose hands the Sharī‘a took shape not merely developed the principles found in the Qur'ān by adding to them the Traditions, that is, anecdotes about Muḥammad's words and practices. Somehow or other, almost without any conscious deception, these scholars managed to include among the Traditions much of the inherited wisdom of the Middle East, transmitted through Christian, Jewish, Gnostic and other sources. To the modern student this is all the more remarkable since Muslims had a complex system of criticism of Traditions. Careful examination, however, shows that this system was not aimed at ascertaining objective historical fact, but at excluding the views of the eccentrics or “lunatic fringe”; and this it largely succeeded in doing. The effect of systematic criticism was in fact to stabilize the Islamic religion on a new ideational basis, namely, that amalgam of Qur'ānic principle, early practice and older lore which had come to be accepted by the main body of Muslims round about the year 800. This amalgam, it is to be noted, did not include the higher learning of the Middle East, such as Greek philosophy and science; and the correct attitude to these “foreign” sciences is one of the problems which al-Ghazālī had to tackle.

By these ideational developments the religion of Islam adapted itself with considerable adequacy to the changes of the first two phases of conquest and conversion. The point where its adaptation had been least adequate was within the ruling institution. There Persian traditions of autocracy and the unprincipled use of power had become dominant, even though in the relations of the rulers to the ruled Islamic principles continued to be respected. In the succeeding phases this impotence of Islamic principles in the topmost political levels—so curious in view of Islam's reputation in Europe of being a political religion⁷—contributed to the difficulties of the intellectual class, and so to the major problem al-Ghazālī had to solve.

It would be convenient to describe with similar brevity the religious and ideational repercussions of the third and fourth historical stages (of disintegration and reconstitution); but unfortunately it is not possible. These repercussions have not yet been properly investigated from the standpoint of this study. Moreover, their investigation cannot be altogether separated from the problem of al-Ghazālī himself. As our understanding of this great man increases, we get more light on what had been happening in the two centuries or so before his birth. The economic, political, social, intellectual and religious happenings of these centuries made the setting in which his life had to be lived. It is part of the aim of this study to discover the salient features of that setting and what had most contributed to making them what they were. At this preliminary stage in the investigation three points may be noted.

(A)

The standard Islamic ideational system had taken root nearly everywhere. The war-lords were under the necessity of recognizing it publicly in all their dealings with the populace. Consequently the disintegration of the caliphate under the war-lords led not to a diminution of Islamic intellectual culture but to its encouragement in numerous local centres. Among the most vigorous of these centres was Nishapur and the surrounding region, where al-Ghazālī's early life was spent.

(B)

In the fourth phase, and also in the third phase though less obviously, supreme rule belonged to superior military force. This happened in a community which had hitherto been regarded as charismatic or divinely-constituted. Did it mean that the community lost its charismatic nature? Was the difficulty a serious one for the men of the time?

(C)

Al-Ghazālī's abandonment of the standard career of a religious intellectual or scholar-jurist⁸ suggests that there was something wrong with this career. Was it that it implied subservience to godless rulers? Were the intellectuals trying to find the significance of their lives in a framework in which Islam was irrelevant? Was the difficulty that the Sharī‘a, whose ostensible purpose was to direct the affairs of the body politic, obviously did not do this?

Al-Ghazālī himself in his autobiography speaks of four groups of men who were trying to find an adequate response to the situation, and we can do no better than follow his guidance and investigate the attitudes of these four groups: the philosophers; the Bāṭinites or Ismā‘īlites; the theologians (among whom we may make a further distinction between Ash‘arites and Hanbalites); the ṣūfīs or mystics.

It remains to say a word about a fifth possible response to the situation, a response in which al-Ghazālī might have been interested but in fact was not—the Persian renascence. Before the Arab conquest of Persia the Zoroastrian clergy, to preserve their power as an intellectual class, had become closely allied with the rulers and subservient to them. In so doing they had largely become cut off from the ordinary people. When the phase of conversion began, therefore, it was not surprising that many Persians became Muslims. The Persian Muslims had much to do with the establishing of the ‘Abbāsid dynasty, and in return the equality of all Muslims, Arab and non-Arab, came to be generally recognized. After a time there was a movement among the secretary class or civil service which maintained the inferiority of the Arabs; but this Shu‘ūbite movement, as it was called, was chiefly a literary movement, it would seem, without much political influence. Other forms of Persian self-assertion are connected with Manichaeanism and with certain sects of Shī‘ite Islam.⁹

The real awakening of the Persian spirit, however, did not come until after the phase of disintegration. Local or provincial dynasties, especially the Samānids, were a focus for hopes and aspirations. It should not be supposed, of course, that there already was a Persian nationalism comparable to the nationalisms of the nineteenth century. There was potentially something similar to these nationalisms, but it had to become conscious of itself. The chief part in bringing about this national self-awareness was played by Firdawsī (d. 1020-1025). His great epic, the Shāh-nāma, welded many local traditions into a unity and gave men of Persian descent a renewed enthusiasm for the perennial mission of Iran—defence of civilization from the inroads of Turan, the Turkish “barbarians” from the great steppes. This was a mission which could be combined with membership of the Islamic empire, though one imagines that the Persians would have found it difficult to go on for centuries serving these two masters, Persian secular aggrandizement and the extension of Islam.

In favourable circumstances this Persian movement might have grown and become of much political significance. Circumstances were against it, however. Before Firdawsī had completed his great poem the sun of the Samānids was setting, and in the ascendant was the star of a Muslim Turkish general, Maḥmūd of Ghazna. Indeed, Maḥmūd became Firdawsī's patron, though it is not surprising in view of his Turkish origin that he and the poet fell out.¹⁰ He was soon followed by the Seljūqs, more Muslim Turks. With Persia largely under Turkish rule Firdawsī's conception of the roles of Iran and Turan had become no more than a political mirage. Persians had become weaker politically, and in their place Turks were now the military defenders of Muslim civilization. This was the position from a few years before al-Ghazālī's birth, and it is thus understandable that, though he must have had much Persian blood in his veins, he never seems to have been attracted by a “Persian” solution of current problems or even to have shown special interest in things Persian.¹¹

3 AL-GHAZāLī'S EARLY LIFE

The central figure of this study was born in 1058, four and a half centuries after the migration of Muḥammad from Mecca to Medina, and three years after the establishment of Seljūq rule in Baghdad. His birth-place was the town or district of Tūs, near the modern Meshed in north-east Persia. His name was Muḥammad, and he was son of Muḥammad, son of Muḥammad; he had the honorific title (kunya) of Abū-Hāmid, meaning father of Hāmid but not necessarily implying that he had a son of this name (certainly only daughters survived him). He is best known as al-Ghazālī, the Ghazālite, possibly meaning the man from Ghazāla, an otherwise unknown village in the region of Tūs;¹² he is sometimes also called aṭ-Tūsī, the Tūsite. He had one brother, Aḥmad, who became a distinguished scholar and mystic, and several sisters.

Nothing is known for certain about his family except that he had a grand-uncle (or less probably uncle), also called Abū-Hāmid al-Ghazālī, who was one of the scholars of Tūs and died about 1043. The family was thus in touch with intellectual circles, as is also shown by the father's anxiety that his two sons should receive the fullest possible education. The assertion in some sources that the theologian's father was a spinner and vendor of wool is to be rejected, since it appears to be an inference from the less probable spelling and derivation of the name Ghazālī. It may be accepted, however, that the father was comparatively poor. On his death he left as much money as he could with a ṣūfī friend, charging him to see that the two boys were well educated. When the money was exhausted the friend made arrangements for them to go to a college or madrasa where they could receive free board and lodging as well as instruction. This very brief glimpse of al-Ghazālī's family shows that the family background was not without its influence on his later career. His father would be characterized by the simple piety of ordinary Muslims, based no doubt on a considerable knowledge of the Qur'ān and the Traditions which could be gained by attendance at the lectures given freely in the mosques. Towards the end of his life al-Ghazālī wrote a book in which he advocated prohibiting ordinary people from attending lectures on theology,¹³ but this must be taken to apply only to the abstruse rational theology of the time and not to the more concrete forms of religious instruction.

No dates are recorded for the earlier part of al-Ghazālī's education. The normal age to begin schooling was eleven, and he would be eleven in 1069.¹⁴ In 1077 he went to an important school or college at Nishapur, the capital of this part of Persia, to study under the most distinguished theologian of the age, al-Juwaynī.¹⁵ In the intervening years he pursued his studies mainly at Tūs, apart from a visit to Gurgan (Jurjān) at the south-east corner of the Caspian Sea. (Nishapur is about fifty miles from Tūs, Gurgan over three hundred, the road passing through Nishapur; these were comparatively short journeys for a great scholar.)¹⁶ The story is told of how the caravan in which the young student was travelling back from Gurgan was set upon by robbers. Among the goods they seized were the notebooks, with the harvest of his study in Gurgan. He went after the robbers and pled for the return of his notebooks, which contained, as he phrased it, the knowledge he had gained at Gurgan. The robber-chief scoffed at this alleged knowledge which could be taken away so easily, but gave back the notebooks. The visit to Gurgan cannot have been later than 1074, since al-Ghazālī on his return spent three years committing his “knowledge” to memory.

In these years of study at Tūs, Gurgan and Nishapur, al-Ghazālī followed the standard curriculum of Islamic higher education. This had a predominantly legal slant. The basis was the study of the Qur'ān and Traditions, together with the commentaries on these. Jurisprudence was derived mainly from the Traditions. Then there were ancillary sciences such as Arabic grammar, differences between the recognized legal rites, and biographical knowledge of the transmitters of Traditions. In al-Ghazālī's case, at least until he went to Nishapur, the chief emphasis was on Traditions and jurisprudence. In these subjects the standard of instruction in Tūs and Gurgan may well have been high. For over a century Nishapur and the neighbouring regions had been in the forefront of educational development, doubtless owing to the virtual independence of the Samānids and their patronage of learning and the arts.

Instruction in the “Islamic sciences” had originally been given in mosques without any fees, and this practice continued. Gradually, however, special institutions were created. At first they may have consisted merely of a room or hall and a library. In course of time living-quarters for the students were added, and funds made available for their support. To this latter form of institution the name madrasa is given, which may be rendered “college”. The first such college seems to have been founded in Nishapur before 960, and this was followed within the century by several others. The movement of college-founding was vigorously encouraged by Nizām-al-Mulk, the great Seljūq vizier (in power from 1063 to 1092). One source suggests that he was the first to provide “scholarships” for the students; but some earlier cases are known.¹⁷ What is certain is that he founded at least nine Nizāmiyya colleges, scattered from Mosul to Herat, and that they were lavishly endowed. In 1077 Nishapur had enjoyed relative peace under the Seljūqs for nearly forty years, whereas Baghdad had been the scene of strife, which must have made academic work difficult, till after the Seljūq occupation in 1055. It might, therefore, be expected that the level of academic attainment in the region of Nishapur would be among the highest in the Islamic world.

In particular, when al-Ghazālī went to Nishapur in 1077 it was to the Nizāmiyya college he went, attracted by the fame of the great theologian, Abū-'l-Ma‘ālī al-Juwaynī, known as Imām-al-Haramayn, “the imam of the two holy places” (Mecca and Medina). Al-Juwaynī was the son of a professor or lecturer at Nishapur, but was admitted by all to be more brilliant than his father. He was primarily a theologian, and introduced al-Ghazālī to theology, perhaps the most difficult of the Islamic sciences. Al-Ghazālī remained at Nishapur until al-Juwaynī's death in August 1085, and latterly helped with teaching. Then he went to the camp of Nizām-al-Mulk, and was received by the vizier with honour and respect, though still only twenty-seven. Though one would have expected him to go on teaching in Nishapur, the records suggest that he spent the whole of the next few years at the camp, until his appointment as professor at the Nizāmiyya college in Baghdad in July 1091.¹⁸

Thus we see that al-Ghazālī had an education as good as any to be had in the Islamic world. Al-Juwaynī was the first theologian of his time. His teachers in Tradition were not so eminent, but his inexactitude in quoting Traditions and his use of uncanonical Traditions are probably due mainly to his own slackness and unorthodoxy. Education, too, had struck deep roots in the region round Nishapur and Tūs, and had influenced many classes of society. This meant that al-Ghazālī, while gaining an excellent education, was not cut off from the simple but well-informed faith of the ordinary people. Al-Ju-waynī is reported to have made a statement which indicates how the younger man was moulded by the older in this point and in others.

“I heard Abū-'l-Ma‘ālī saying, I had read thousands of books; then I left the people of Islam with their religion and their manifest sciences in these books, and I embarked on the open sea, plunging into the literature the people of Islam rejected. All this was in quest of truth. At an early age I fled from the acceptance of others' opinions (taqlīd). But now I have returned from everything to the word of the Truth, ‘Hold to the religion of the old women’. If the Truth does not grasp me by the grace of His justice, so that I die in the religion of the old women and the result of my life is sealed at my departure with the purity of the people of Truth and the word of sincerity, ‘There is no god but God’, then alas for the son of al-Juwaynī (that is, himself).”

Notes

1. Cf. Integration; also “The Conception of the Charismatic Community in Islam”, Numen, vii. 77-90.

2. Cf. E. G. Browne, Literary History of Persia, Cambridge, 1928, i. 365 f.

3. Miskawayh in H. F. Amedroz and D. S. Margoliouth, The Eclipse of the ‘Abbasid Caliphate, Oxford, 1921, i. 352 (iv. 396).

4. Cf. B. Spüler, Iran in früh-islamischer Zeit, Wiesbaden, 1952, 129.

5. Cf. Muhammad at Mecca, Oxford, 1953; also the opening chapters of Integration.

6. Cf. Integration, 102.

7. Cf. N. Daniel, Islam and the West: the Making of an Image, Edinburgh, 1960.

8. See ch. VI below.

9. But the identification of Persians with Shī‘ite doctrines in the early period must not be exaggerated; the identification has only been close since the sixteenth century.

10. Browne, Literary History of Persia, ii. 129-41.

11. He is said to have written some books in Persian, such as Kīmiyā' as-Sa‘āda, and appears to have written letters in Persian (Faḍā'il al-Ānām). In Maqṣad, 73, he uses the form Kurkānī (Gurgānī) for Jurjānī, although the scholar Yāqūt (Mu‘jam al-Buldān, s.v.) says the K form is never used in Arabic.

12. Some authorities say it ought to be Ghazzālī, the descendant of the ghazzāl or spinner, but this is less likely. Cf. 183 below.

13. Iljām; cf. p. 148 below.

14. A. Mez, The Renaissance of Islam (Eng. tr.), Patna, 1937, 182.

15. I have not been able to find the source of this date 470/1077, given, presumably independently, by Bouyges and M. Smith.

16. Mez, 191.

17. Cf. J. Pedersen, art. “Madrasa”, IV, in EI(S). Also Subk. iii. 135-45, esp. 137. He may have been imitating the seminaries of the sectarian Karrāmites and Quarmaṭians (cf. Massignon, Passion, i. 166); and his aim was doubtless to counter sectarian views; cf. p. 106 below.

18. We do not know where the camp was; it presumably moved. During this period he may have travelled to Zūzan in Afghanistan; he is said to have studied under a traditionist there (cf. SM, i. 19 f.; Subk. iv. 114).