the Massachusetts Institute of Technology and the editors of The Journal of Interdisciplinary History

Traditionalism in Islam: An Essay in Interpretation Author(s): William A. Graham Reviewed work(s): Source: The Journal of Interdisciplinary History, Vol. 23, No. 3, Religion and History (Winter, 1993), pp. 495-522 Published by: The MIT Press Stable URL: http://www.jstor.org/stable/206100 . Accessed: 02/11/2011 06:08
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.

The MIT Press and the Massachusetts Institute of Technology and the editors of The Journal of Interdisciplinary History are collaborating with JSTOR to digitize, preserve and extend access to The Journal of Interdisciplinary History.

http://www.jstor.org

Journal of Interdisciplinary History, xxIII:3 (Winter 1993), 495-522.

WilliamA. Graham

Traditionalism in Islam: An Essay in Interpretation

"No other community (ummah)has a continuous historical transmission (isndd) like that of ours-meaning this community."-Ibn Qutaybah (d. 276/889)1

Nowhere in the history of religion is the danger of interpretative generalization becoming reductionist or simplistic more acute than in the study of Islamic religion. No tradition, not even the Buddhist or Christian, has manifested itself in such widely varied geographical, historical, and cultural milieux with such diversity of particular manifestations and simultaneous continuity of generic social, religious, cultural, and political traits. It is a truism to say that there is no single entity called "Islam," only the various "Islams" of local contexts: to speak of Islamic society or civilization is to speak of myriad local or regional traditions of sharply differing forms and often rapidly changing historical circumstances. On the other hand, to speak of any particular Islamic society is also to speak of a shared tradition that is astonishingly recognizable across all of its regional divisions and historical eras. Consequently, while those engaged in Islamic studies dare not lose sight of the many variations and changes Islam has known as a global religious and cultural tradition, they must still press forward to generalization about the wider Islamic tradition and

William A. Graham is Professor of History of Religion and Islamic Studies at Harvard University. He is the author of Beyond the WrittenWord(New York, 1987); Divine Word and PropheticWordin Early Islam (Paris, I977). ? 1993 by The Massachusetts Institute of Technology and the editors of The Journal of Interdisciplinary History. I Laysa li-ummatinmin al-umamisnddmithla isnddind ya'ni hadhihi 'l-ummah,as cited from the Kamdlfi asmd' ar-rijdl, without specific reference, by Aloys Sprenger, "Ueber das Traditionswesen bei den Arabern," ZeitschriftderDeutschenMorgenldndischen IX Gesellschaft (I856), I.

496

WILLIAM

A. GRAHAM

its historical manifestations, if justice is to be done to the impressive continuities it has and does exhibit.2 In what follows, I tackle one sector of the problem of generalizing about the unity and continuity of Islamic tradition as opposed to emphasizing the disparity and discontinuity of discrete "local contexts" of the Islamic past and present. Specifically, I discuss a salient, if not primary, facet of what has been often characterized as a "traditionalist" bent in Islamic religion and culture, namely the foundational character, pervasiveness, and vivid elaboration of what I call the isnddparadigm.
"Tradition" is the pritwo different senses of the term mary notion; for our purposes are paramount. In its simplest, most specific sense, tradition is, as Shils put it, "a traditum,. . . anything which is transmitted or handed down from the past to the present." It is also anything perceivedto have been passed from one generation to the next; or, as Acton defined it: "a belief or practice transmitted from one generation to another and accepted as authoritative, or deferred to, without argument." Traditions in this sense are almost synonymous with "customs" or "customary institutions," in that they are actions, ideas, and written or oral texts received from the past and accepted as normal or normative for a given community.3 Second, in its general, collective sense, tradition is the modus vivendi of a society insofar as it is understood as congruent and continuous with the past; it is the "cumulative tradition" peculiar to any community large or small-"the way society formulates
"TRADITION" AND "TRADITIONALISM"
2

The Islamic calendar, dating from Muhammad's Hijrah, or emigration, of A.D. 622, is a lunar one in which every year is about eleven days shorter than the Julian and Gregorian solar year. July 13, A.D. 1991 (or I99I C.E.) marked the beginning of the year A.H. 1412 (Anno Hegirae). In this study, when two dates are given, the first is the Islamic, or Hijri, year, and the second the Western, Christian (or Common Era, c.E.) year. 3 Edward Shils, Tradition(Chicago, I98I), I2; Harry B. Acton, "Tradition and Some Other Forms of Order," Proceedings the AristotelianSociety, LIII (1952-54), 2. On tradiof tion, see also Jan Vansina, Oral Tradition:A Study in HistoricalMethodology (London, 1965); as idem, Oral Tradition History (London, I985); Raymond B. Williams, "The Sacred Thread: XXXVI (I975), 225-243. On the often imaginary Transmission of Tradition," Encounter, antiquity of "traditions" in our own Western world, see Eric Hobsbawm and Terence Ranger, The Invention of Tradition(Cambridge, 1983) (I see, however, no need to distinguish sharply, as Hobsbawm does [p. 2] between "tradition" and "custom"; see also Alan Bailey, "The Nature of Tradition," Folklore, C (1989), I53-I58.

TRADITIONALISM

IN ISLAM

497

and deals with the basic problems of human existence," as Heesterman put it. In this collective sense, "tradition" is the sum of a society's specific "traditions."4 Tradition in either of these two meanings is a descriptive, value-neutral term, even though it has been common in the modern West to see it as explicitly negative. This negative valuation derives primarily from the European Enlightenment, which encouraged the denigration of tradition as the excess baggage of the past, an impediment to "progress" toward a scientific, rational "modernity," and therefore to be jettisoned. However, as others have noted, tradition cannot be relegated, even in advanced industrial societies, to the past or discarded as something opposed to, or to be superseded by, reason, innovation, technology, or science, since all of these also depend on tradition.5 If tradition is understood in the two senses just described, "traditional society" may be applied to those societies in any age in which a majority see and explicitly value their current way of life as rooted in and continuous with a valued past (that is, with tradition in the general sense), even while newer values may openly compete for authority in the society. Traditionalists in such societies, even those undergoing rapid change, understand themselves to be conservators of values and what is valued from the past. They do not perceive recent history as a major rupture with the past that has decisively broken the authority of that past as normative for, and intimately linked to, the present. Hence derives the power of tradition as a rallying cry in such contexts, whether for retention of what is valued or for recovery of what has been lost.6
4 "Cumulative tradition" is a term developed by Wilfred Cantwell Smith with specific reference to religious traditions in The Meaningand End of Religion (New York, I962) and used in his many subsequent works, but the term is easily applicable to any cumulative tradition-religious, cultural, social, or political. The citation is from Jan C. Heesterman, "India and the Inner Conflict of Tradition," Daedalus, CII (1973), 97. 5 On the danger of setting up a dichotomy between "tradition" and "innovation" (or "change," "rationality," or "science"), see H. T. Wilson, Traditionand Innovation:The Idea of Civilization as Culture and Its Significance(London, 1984); see also the works in n.8. 6 "Traditional societies" in my usage includes what Shmuel N. Eisenstadt seems to have been the first to term "post-traditional societies," namely those non-European societies of the "third" or "developing" world that have had to deal in the past century or so with Western-style "modernism." My usage stresses the continuity of previous tradition in such societies; whether it is/was valid for certain of these societies (e.g., Japan and China) is an open question that need not trouble us here. On "post-traditional societies," see the

498

WILLIAM

A. GRAHAM

Traditional societies stand over against advanced industrial or so-called "modernized" societies in which a major, or perhaps only the dominant elites, already have experienced, or see themselves as having experienced, a decisive rupture with the past and thus perceive themselves as no longer "traditional." In the collective self-consciousness of these principally Western societies of the "developed nations," various watershed periods or new developments have served as symbolic, if not actual, breaks with a now devalued past: the Renaissance; the industrial, scientific, or technological revolution; the Enlightenment; the French or American revolution; the colonial or imperial experience; and/or the increasing globalism and pluralism of the nineteenth and twentieth centuries more generally. As a result, older traditions have been wrongly perceived as "dead" or declining and no longer relevant
today.7

While "modern" and "traditional" societies both possess traditions, the crucial distinction between them lies in their selfimages and the functional authority of tradition in each. I use "traditionalism" and "traditionalist" to refer to a person's or a group's strong preference for recourse to tradition (genuine or invented) as the primary source of authority. This is not to say that either traditions or traditional societies are immutable. On the contrary, as Levenson noted: "an idea changes in its persistence as well as in its rejection .... While iconoclasts relegate traditional
Daedalus issue of that name edited by Eisenstadt (CII [1973]), particularly Eisenstadt, "Post-Traditional Societies and the Continuity and Reconstruction of Tradition," 1-27. 7 The question of what is meant by "traditional society" is not so easy to answer as one might think. Western ethnographic and sociological study of non-Western societies, both literate, "great" societies, and nonliterate, "little" societies, long perpetuated a way of thinking about "traditional" and "traditionalism" that portrayed these as characteristics of all non-Western societies and contrasted them to "modern" and "modernity," which are used to characterize post-Enlightenment or even post-Renaissance Western Europe and its culturally derivative societies around the world-specifically to characterize them as "rational," "scientific," or "innovative" as opposed to "traditional," "customary," or "imitative" in what they accept as norms or as usages (Cf. Eisenstadt, Tradition,Change, and Modernity[London, 1973], 98-I12; Ernest Gellner, Traditionand Change [London, 1969]). By way of refutation of this commonly held position, see the last two works; Lloyd I. Rudolph and Susanne Hoeber Rudolph, The Modernityof Tradition:Political Development in India (Chicago, 1967); Joseph R. Gusfield, "Tradition and Modernity: Misplaced Polarities in the Study of Social Change," American Journal of Sociology,LXXII (1967), 351362; Heesterman, "India and Inner Conflict," 97-113; Wilson, Traditionand Innovation; Bailey, "Nature of Tradition," I43-16I; Marilyn R. Waldman, "Tradition as a Modality of Change: Islamic Examples," History of Religions, XXV (I986), 318-340.

TRADITIONALISM

IN ISLAM

499

ideas to the past, traditionalists, at the same time, transform traditional ideas in the present." Hence an important caveat: traditionalism does not necessarily involve conservatism or opposition to change in political, social, or religious life; traditions may serve equally well as the bases of reform and innovation or as the grounds for defense of the status quo. Just as tradition is not the opposite of reason or innovation, traditionalism is not the opposite of modernism, although it may oppose modernism where the latter is perceived as destructive of important traditions in a society.8

Conversely, the "modernity" of advanced industrial societies, from the United States to Europe and Japan, has not meant the demise of many forms of traditionalism in these societies. It may well be argued that in these largely urbanized and industrialized societies there have been many fewer instances in which traditionalism has been a major societal force than has been the case in societies such as those of most of the Islamic world, in which greater value is still placed upon continuity with perceived traditional norms of great antiquity. Part of the reason is the former societies' positive valuation of change as a desideratum in and of itself. Nevertheless, none of these societies are without their own traditionalisms, for in them religious, political, national, regional, local, ethnic, and other traditions continue to carry authority for people's lives. Indeed, as noted above, our modern science and rationalism have their own traditions and traditionalism, even if these are invoked to discredit other kinds of traditionalism. Because of these ambiguities, I use "traditional"here to refer to those societal norms and institutions that a culture perceives as congruent with or continuing older precedents and values, and as important if not essential to its identity. Even though no society is without elements of modernism and traditionalism, I distinguish traditional societies in which tradition plays a visibly major role from the advanced industrial societies that have cultivated "modernism" and perceive themselves as nontraditional. Correspondingly, by "traditionalist" and "traditionalism" I refer to an
emphasis upon the historical authority of, and the continuity with
8 Joseph R. Levenson, Confucian China and Its Modern Fate (Berkeley, I958), I, xxvii. See also the important comments of Gusfield, "Tradition and Modernity," 351-353, on tradition and change. Cf. Shils, Tradition,213-26I.

500

WILLIAM

A. GRAHAM

or recovery of, norms and institutions basic to a particular cumulative tradition.

To speak about Islamic traditionalism is normally to speak about the widespread Muslim emphasis upon the primary, dual authority of the revelations of the Qur'an and the tradition or practice (sunnah)ascribed to the Prophet and the first few generations of Muslims (the "pious forebears," as-salaf). Indeed, since the Salaf themselves are the original codifiers and transmitters (and hence guarantors) of the Qur'anic text and the reports concerning Muhammad's words and practice, any traditionalist vision among Muslims can legitimately be termed salafiquite apart from the specific historical groups that have adopted the name for themselves-to refer to emphasis on the conjoined authority of Revelation, Prophet, and early community. I use, however, "Islamic traditionalism" in a slightly broader sense, albeit one rooted in these primary elements of Muslim "Salafism." I mean by the term the long-standing, overt predilection in diverse strands of Islamic life for recourse to previous authorities, above all the Prophet and Companions, but also later figures (whether an Abui Hanifah, Jalal ad-Din Rumi, Shaykh Wall Allah ad-Dihlawi, or Ibn 'Abd al-Wahhab) who are perceived as having revived (jaddada),reformed (aslaha), or preserved (hafiza) the vision and norms of true, pristine Islam, and thus as being in continuity and connection with the original community, or ummah. All such authorities function as the interpreters of God's revealed word and as paradigms for the present and future. An important concomitant of this attitude is a wariness or even abhorrence of any "innovation" (bid'ah)that runs counter to the perceived tradition. Thus Naqshbandi or Bektashi Sufi thought can be just as traditionalist as Hanbali or Maliki legal thought.9
THE ISLAMIC CASE

9 For two different but equally important scholarly approaches to the problem of traditionalism in Islam, see Johan Fiick, "Die Rolle des Traditionalismus im Islam," Zeitschriftder Deutschen Morgenldndischen Gesellschaft,XC (I939), 1-32; George Makdisi, "Remarks on Traditionalism in Islamic Religious History," in Carl Leiden (ed.), The and Modernismin the Muslim Middle East (Austin, I966), 77-87. Conflict of Traditionalism On the Salaf and salaf , see also Ignaz Goldziher, Muhammedanische Studien (Halle, I890),
II, 21-22.

TRADITIONALISM

IN ISLAM

501

One particular element of this Islamic traditionalism is pervasive, even indispensable: a "sense of connectedness," or to coin an Arabic neologism for this, ittisaliyah-the need or desire for personal "connection" (ittisal)across the generations with the time and the personages of Islamic origins-something that has been a persistent value in Muslim thought and institutions over the centuries. I do not contend that Islam is unique in valuing personal connectedness, for such valuation might well be taken as a fundamental, even defining, sociological trait of "traditional" as opposed to "modern" societies. I suggest rather that whereas Muslims have elaborated this emphasis in different ways, at different times, and in different sectors of their collective life, they have always done so in ways that are characteristic, identifiable, and central. Indeed, it is possible to discern a basic, recurrent pattern that is used to express their ittisaliyah, and hence their traditionalism. Muslim traditionalism has most clearly and consistently expressed its need for "connectedness," and specifically personal connectedness, in variations on a single model: the isnad paradigm. This paradigm derives from the central Islamic institution of the Hadtth, the collective corpus of traditional reports (ahadith, plural of hadith;or akhbar,plural of khabar)ascribed to Muhammad or others of the first generation of Muslims. The early collection of ahadith and the disciplines of study that grew up to authenticate their content have provided all subsequent generations of Muslims with thousands of brief texts (mutun; singular: matn) that claim to preserve the opinions and practice, or Sunnah, of Muhammad and the other members of the original, paradigmatic community, the Ummah, of pristine Islam. The Hadith reports have also been the basic textual unit of, and the fundamental literary form for, history, biography, scriptural exegesis, and legal and theological argument in the Islamic tradiTHE ISNAD PARADIGM

tion. 10
Io The terms ahadith, akhbar, and also sometimes dthar("remains" or "traces")and qisas ("stories" or "tales") should not be too sharply distinguished, especially in the early period (cf., on akhbar and hadith, Franz Rosenthal, A History of Islamic Historiography [Leiden, I968], II, pace Nabia Abbott, Studiesin Early ArabicPapyri [Chicago, 1957-67], I, 7), even though ahadith normally refers specifically to reports from or about the Prophet, while

502

WILLIAM

A. GRAHAM

The system of transmission of the Hadith has been based upon the use of the isnad, or "support," that accompanies the matn of every individual hadith. The isnad takes the form of a list or "chain" (silsilah) of individual transmitters who span the generations from the most recent reporter back to the Prophet or Companions. The isnads exist in order to substantiate the authenticity of a given report by confirming that it has come down via a silsilah composed of trustworthy persons-preferably a "high" chain made up of a minimum number of links. Thus the defining elements of the isnad paradigm are: (I) derivation of authority primarily or even, in extreme cases, solely from (2) linkage to a sacred, but historical, time of origins of the tradition through (3) a chain of personal transmission, the individual human links of which represent all intervening generations between that of the original source (ideally the Prophet or one of his Companions) and that of the latest reporter. This paradigm is, in turn, the mechanism or overt vehicle for the realization of ittisaliyah, the personal connectedness described above which authenticates a report as valid tradition.11 No student of a "traditional" nonliterate or literate society anywhere in the world would find anything particularly surprising in the general shape of this mechanism. Witness the gotra or lineage concerns of Hindu brahmans and their expression, for or example, in the guruparampara teaching lineage, which is central to all Indian learning; the Hasidic or Zen line of spiritual masters; the idea of Apostolic succession to the Holy Roman papal office; the Sikh line of gurus; and the lineage system of royal, priestly, or shamanic succession in many tribal societies. All of these are similar manifestations of a traditionalist concern for connectedness with past authority. Indeed, there is reason, as Horovitz argued long ago, to think that the isnad itself was influenced by Rabbinic models of transmission of the oral Torah. However this may be, the Islamic case has its own distinctive configuration and unusual
akhbar (like ithar) is usually a more general term for any traditional reports: See Gualtherus H. A. Juynboll, "Muslim's Introduction to His Sahih," Jerusalem Studies in Arabic and Islam, V (I984), 265, n.3; MuhammadJamal ad-Din al-Qasimi (ed. Muhammad al-Baytar), Qawd'id at-tahdfthminfunan mustalahal-hadith, (Damascus, 1380/1961; 2d ed.), 61-64. II A note on the "high" ('dli) isnad: such a chain depends upon transmission as much as possible at every stage from long-lived teachers in old age to the youngest possible students, thus shortening the number of links needed to reach back to the original reporter.

TRADITIONALISM

IN ISLAM

503

degree of institutionalization, and it is to this case that we now

turn. 12

The basis of the isnad paradigm is the Sunnah-the way, usage, practice, or tradition of the Prophet and his Companions. A great deal has been written about Sunnah in modern orientalist, as well as traditional Muslim, scholarship. It likely predated Islam in Arabic linguistic usage. Persuasive arguments have been advanced for Sunnah having served in the early decades of Islam to refer to the general usage of the first generation of Muslims, the Companions, both before and after Muhammad's death. This general usage then gradually gave way in the later first century to a more specific notion limited to the tradition or practice of the Prophet (sunnatan-nabi). The debates in both traditional Muslim and modern orientalist scholarship about the crystallization of the Sunnah concept and the authenticity of the Hadith as the major vehicle of transmission of the Sunnah need not be rehearsed. Nor need we explore the contested issue of the precise antiquity of the isnad as a formal element accompanying the matn,
or concrete text, of every hadith. Most authorities in both scholarly traditions would agree that since at least the beginning of the second/eighth century, the Sunnah of the Prophet has been seen as the primary content of the Hadith and the major extra-Qur'anic authority for Muslim life.13
12 On the question of Jewish models for the isnad, see Josef Horovitz, "Alter und Ursprung des Isnad," Der Islam, VIII (I918), 44-47. Cf James Robson, "The Isnad in Muslim Tradition," Transactions the Glasgow UniversityOriental Society, XV (I953-54), of 15-26.

13 On the hadith and sunnah, see Subhi as-Salih, 'Ulum al-hadithwa-mustalahuhu (Beirut, al-Fdsil 1969; 5th ed.), I-Io; al-Hasan ar-Ramahurmuzi (ed. 'Ajjaj al-Khatib), al-Muhaddith Studien, bayna 'r-rdwiwa 'l-wd'i, (Beirut, I39I/1971), I63-I8I; Goldziher, Muhammedanische II, 3-27; cf. Juynboll, Muslim Tradition (Cambridge, 1983), 30-39. For the range of meanings of sunnah, see Ibn Manzir, Lisdn al-'Arab (Beirut, 1956), XII, 224-225; cf. Fazlur Rahman, "Concepts Sunnah, Ijtihdd, and Ijmd' in the Early Period," Islamic Studies (Karachi), I (1962), 5-21. For a good example from the end of the first Islamic century of sunnahas the practice of the first generation (as-salaf), see the citation from 'Abd Allah b. im 'Abd al-Hakam in Tilman Nagel, Staat und Glaubensgemeinschaft Islam (Zurich, 1981), II, 55. Concerning the authenticity of the Hadith, the classical Western work is Goldziher's Muhammedanische Studien, II, 1-274. Goldziher questions the authenticity of the bulk of the Hadith as documents of Muhammad's times and demonstrates that many individual hadiths reflect political or doctrinal concerns one to two centuries after Muhammad's death. The radical skepticism of Joseph Schacht, The Origins of Muhammadan Jurisprudence (Oxford, 1950) carries Goldziher's perceptions to an extreme; cf. also Robson, "Muslim and Tradition: The Question of Authenticity," Memories Proceedings the Manchester Literary of and PhilosophicalSociety, XCII (1951-52), 84-102. For sometimes overly positive assess-

504

WILLIAM

A. GRAHAM

Correspondingly, "Qur'an and Sunnah" early on became the dual watchword of Islamic traditionalism as transmitted and elaborated by the 'ulama', or religious scholars. Whether identified as "orthodox," "orthoprax," or "scripturalist," the tradition of the 'ulama' has always been characterized by reliance on "the two sources" of scripture and sunnah. With the cessation of active prophetic-revelatory activity at Muhammad's death, the "Recitation," or Qur'an, remained the one impeccable source of authority in this world. It is not, however, principally or even substantially a book of laws and practical guidance. Thus, although its word was and is paramount, to articulate their ideal of personal and communal existence and to translate it into practice, Muslims have had recourse to the Sunnah, using it to interpret the Qur'an and to extend its spiritual authority to cover matters on which its text is silent. Indeed, to a great extent, as the one fully valid interpretive vehicle for the Word of God after the death of the Prophet, the Hadith that carried the Sunnah became as one (likely invented) hadith put it, "judge over the Qur'an, not the Qur'an over the Sunnah." Any interpreting agent ultimately controls the interpreted authority, and the role of the transmitted Sunnah of the Prophet as agent vis-a-vis the Qur'an as prime authority was no exception.14 One dimension of the Muslim notion of Sunnah deserves particular note here: the great stress placed upon the paradigmatic character of the earliest community, the Ummah-in the first instance, the Ummah developed during Muhammad's lifetime but also that of his surviving Companions after his death, specifically in the three decades (A.D. 632-661) of the first four comments of the reliability of the Hadith corpus, see: Abbott, Papyri, II, 1-83; Fuat Sezgin, Geschichte arabischen des Schrifitums,(Leiden, 1967), I, 53-84. More balanced treatments of the questions of Hadith authenticity, isnad dating, etc., are: Juynboll, Muslim Tradition; Harald Motzki, "The Musannaf of 'Abd al-Razzaq al-San'ani as a Source of Authentic Ahidith of the First Century A.H.," Journal of Near Eastern Studies, L (I991), I-21; cf. Josef van Ess, Zwischen Hadit und Theologie (Berlin, I975). On the history of sunnahas a development from pre-Islamic Arabic usage, see M. M. Bravmann, The Spiritual Background of Early Islam (Leiden, 1972), 123-I98.

14 Both Clifford Geertz, Islam Observed(New Haven, 1968) and Ernest Gellner, Muslim Society (Cambridge, 1981), use the term "scripturalism" as descriptive of mainstream "ulema piety." On the question of revelatory versus prophetic authority in early Islam, see Graham, Divine Wordand PropheticWordin Early Islam (The Hague, 1977), 9-48. The cited hadith is from 'Abd Allah b. 'Abd ar-Rahman ad-Darimi's Introduction to his Kitdb as-sunan(Cairo, 1386/I966), bdb49.

TRADITIONALISM

IN ISLAM

505

munal leaders, the so-called Rdshidun,or "rightly guided" caliphs. The paradigmatic quality of Muhammad as the model human being for every generation of Muslims over the entire sweep of Islamic history needs no iteration here, and one need only survey the popular books and pamphlets on the street corners of Islamic cities from Indonesia to Morocco to see how popular and vividly familiar the stories of the first caliphs and the other Companions are to this day.15 An important part of being Muslim is a developed and powerful consciousness of the pristine community of Medina in the time of Muhammad and his immediate successors as the prototype of, or paradigm for, human society properly ordered under God's ordained norms. The model Ummah actually existed, in the golden age of Muhammad and, for Sunnis, at least the Medinan caliphate after him. If the Shi'ls feel that the majority of Muslims went astray after the Prophlet's death, by failing to recognize as the first caliph Muhammad's cousin and son-in-law, 'Ali ibn Abi Talib, they too look back to the lifetime of Muhammad for an exemplary community. For them it is the community headed by the Prophet and his family, the earliest "people of the House"Muhammad, his daughter, Fatimah, 'Ali, and their two sons, Hasan and Husayn. Beyond these, to some degree the model Ummah is mirrored also in the small groups that recognized and gathered around 'All and his designated descendants, the true Imams,as the only valid, albeit rejected and persecuted, leaders of all Muslims.16 Muslim scholars in later times developed an explicit concept offasdd az-zaman, or "the degeneration of the times," to express
I5
720,

On the Rdshidan (to whom 'Umar b. 'Abd al-'Aziz, Umayyad caliph, 98/717-101/ is sometimes added as a fifth member), see Ibn Sa'd, Kitdb at-tabaqdt al-kabir(Cairo,
III, i, II-27, 36-58,
II9-152, 190-274.

1388/I968),

On Muhammad-piety

in Muslim

tradition, see, e.g., Abu Nu'aym al-Isfahini, Dald'il an-nubuwwah (Cairo, 1390/1970); Abi'l-Fadl al-Qadi 'Iyad, ash-Shifa bi-ta'rffhuqaqal-mustafd(Cairo, n.d.); Yusuf an-Nabfil-madd'ihan-nabawiyah(1903; reprint ed., Beirut, 1974); hani, al-Majmu'ahan-nabhdniyah Jalal ad-Din as-Suyfti (ed. Muhammad J. Hiras), al-Khasd'isal-kubrd(Cairo, n.d.); Tor Andrae, Die Person Muhammedsin Lehre und Glaubenseiner Gemeinde(Stockholm, I918); Annemarie Schimmel, And Muhammad His Messenger(Chapel Hill, 1985 [revised English is version of 1981 German ed.]); Constance E. Padwick, Muslim Devotions (London, I960). I6 On the centrality for Shi'ite piety of the Prophet and "the people of the House" (ahl al-bayt), especially his grandson, al-Husayn, see, e.g., Mahmoud Ayoub, RedemptiveSuffering in Islam (The Hague, 1978), 37-147. Cf. Said Amir Arjomand, The Shadow of God and the Hidden Imam (Chicago, 1984), io6, 152-53.

506

WILLIAM

A. GRAHAM

the increasing temporal (and similarly, moral) distance from the time of the Prophet and the model Ummah. However, this idea was never elaborated into anything like the Hindu notion of the Kaliyuga, or final age of full devolution of the universe. The original Ummah is not simply a source of nostalgia for something irrevocably lost, but also a historical model for every society in every age that would be properly muslim, or submissive to the will of God. The utopian blueprint is available as a realized era of the historical past, which makes it all the more compelling as a realizable possibility in every age. It is a major source of what can legitimately be called a pervasive and recurrent "traditionalist" bent in Islamic thought, one that looks to this original, ideal model for renewal. In this regard, the Sunnah of the Prophet represents for all practical purposes the sunnah of the original Salaf.17 The Hadith genre and the concept of Sunnah, when joined to the fundamental fact of the Qur'an, supply the functional and ideological bases of Muslim traditionalism. It is, however, the isnad system of documenting the authenticity of all transmission of knowledge that provides the specific model for various forms of unbroken "connection" with the first generation of the Salaf, the idealized Muslim Ummah. This key concept of trustworthy linkage with the Prophet and his Companions focuses through the mechanism of the isnad on the trustworthy faithfulness of muslimindividuals linked personally with one another in an unbroken chain through every generation. It is ultimately a concept of the authentic faithfulness of the ongoing Ummah as the only ultimately reliable guarantee of right knowledge and action, since both stem from the experience and example of the pristine Ummah and its prophetic-revelatory authority.
I7 On fasdd az-zaman, see (with reference to Sufi attitudes) Richard Gramlich, Die schiitischenDerwischorden Persiens (Wiesbaden, I981), II, 155-156, nn. 855, 856. Cf. the comments of Madeline C. Zilfi, The Politics of Piety: The Ottoman Ulema in the Postclassical Age (1600-18oo) (Minneapolis, I988), 134: "The community of Medina has been the most glorious and compelling memory of the Sunni community. But the memory has also been a painful one. Every age that passes since that of the One True God's revelation to His last Prophet brings a dreadful distancing from the ideal. With time come changes and accretions. Whether large or small, matters of ritual, or dress or social ceremony, differences are inherently consequential for a faith that holds all human activity to be a sacred concern.... In a salvationary sense, far from healing all wounds, time is itself wounding." Cf. Goldziher, Muhammedanische Studien, II, 21-22.

TRADITIONALISM

IN ISLAM

507

What is crucial here is the fundamental presupposition that truth does not reside in documents, however authentic, ancient, or well-preserved, but in authentic human beings and their personal connections with one another. Documents alone, without a line of persons possessed of both knowledge and righteousness to teach and convey them across the years, are useless as instruments of authoritative transmission. It is "the golden chain of sincere Muslims" that guarantees the faithful copying, memorizing, reonly those of the Hadith, citing, and understanding of texts-not but those of the Qur'an and all subsequent works of Muslim piety and learning. With this in mind, let us look more closely at how the isnad mechanism can be the paradigm for various Islamic institutions and traditions that exemplify and sustain the "traditionalist" dimension of a Muslim world view.18 The need to transmit faithfully the Sunnah WORKS RIJAL the medium of the Hadith was the motive for the birth through of the isnad and its subsequent development as a mechanism of personal contact with the sources of tradition. We can see in the early development of Hadith collections known as musnads a focus on the individual guarantors of each report. These musnad collections were organized according to names of the Companions who first transmitted a given hadith from or about Muhammad. This form endured, even though it was largely displaced by musannaf works arranged more practically "according to chapters" that were distinguished by their subject matter rather than their sources. In general, with each generation's greater remove from the time and persons of the first Ummah, an increasing need to distinguish authentic from forged or faulty reports about Muhammad and the Companions led to a gradual systematization of the formal criteria of the isnad, of the biographical data about its
THE

I8 I am indebted to Wilfred Cantwell Smith for his suggestion to me many years ago that a sacred text is not itself the locus of truth for Muslims or, indeed, most religious in persons. To the point here are Smith's own remarks on tasdfq "Orientalism and Truth," public lecture in honor of T. Cuyler Young (Princeton, 1969). I take the phrase "golden chain of sincere Muslims" loosely from Schimmel, "The Golden Chain of 'Sincere Muhammadans,'" in Bruce B. Lawrence (ed.), The Rose and the Rock (Durham, N.C., 1979), I04-I34. Her title is a combination of "The Golden Chain" title of a poem by Jami and a phrase from a poem in which Mir Dard calls himself a "sincere Muhammadan." Note also that the Dhahabiyah tariqah describes itself as "the golden chain," silsilat adh-dhahab: Gramlich, Derwischorden, 15. I,

508

WILLIAM

A. GRAHAM

transmitters, and of the use of hadiths alongside the Qur'an as proof-texts in virtually all contexts. Slowly there emerged from beginnings in the late first and early second century A.H. what came by the third/ninth century to be called "the science/s of Hadith," the focus of which was the evaluation of isnads and of rijal, the "men [and women]" whose names appeared in the isnads
as transmitters. 19

In connection with this systematization of hadith- and specifically isnad-criticism, one development is especially significant: the almost uncanny prolixity of the early and continuing production of works on the "science of the 'men"' ('ilm ar-rijdl)-biographical dictionaries covering first the generations of the Companions and their early "Successors/Followers," then gradually all the men and women of note in a given period or region, most often those involved in the transmission of learning, and, in particular, the specialists in Hadith. The rijal genre reached its maturity in the fifth/eleventh century, although it had begun much earlier, as witnessed in one of the major comprehensive rijal
works, the Great Book of Classes of Ibn Sa'd (d. 230/844). This

compilation provides biographical data on well over 4,000 Muslims down to the author's lifetime, arranged by "classes"of Companions and Followers, and within these by other subcategories (such as Medinans or "wives or the Prophet"). Later compendia focused more narrowly on particular professional or regional groups, for example, mystics (Sufis) or legists of one of the major schools of law, or prominent persons of a particular city, and were usually arranged alphabetically. The purview of these books eventually extended to include persons of note in virtually all of the sciences and professions, from poetry to medicine. Even geof nealogical dictionaries such as the Compendium Ibn al-Kalbi (d.
204/819), which served as the basis of both the never-completed Genealogies of the Nobility of al-Baladhuri (d. 279/892) and the

famous Compendiumof the Genealogiesof the Arabs by Ibn Hazm

I9 On the chronology of early Muslim tradition, seeJuynboll, Muslim Tradition, 1-76. On the development of the sciences of Hadith, see 134-217 of the same work and the still excellent overview of Muhammad Z. Siddiqi, Hadith Literature:Its Origin, Development, Special Featuresand Criticism (Calcutta, I96I), 164-204. Cf. also Muhammad M. Azami ['Azami], Studiesof Hadith Methodologyand Literature (Indianapolis, 1977), 32-67; Muhammad Abdul Rauf [Ra'ufj, "HadithLiterature: The Development of the Science of Hadith," in Alfred F. L. Beeston et al. (eds.), Arabic Literatureto the End of the Umayyad Period (Cambridge, 1983), 271-288.

TRADITIONALISM

IN ISLAM

509

of Cordoba (d. 456/1064), were effectively genealogically arranged rijal works. The culminating examples of encyclopedic rijal works are the four compendia of Ibn Hajar al-'Asqalanl (d.
852/1449) that treat, respectively, the Prophet's Companions,

Hadith specialists, Hadith specialists and other scholars, and the notables of the eighth/fourteenth century.20 It is hard to overestimate the importance of the biographicaldictionary genre in the history of Islamic learning and culture, especially as an index of the ongoing Muslim emphasis on knowing the individual men and women upon whose learning and piety rests the soundness of the transmission of the "core" tradition through every generation. Gibb underscored this when he emphasized that "the conception that underlies the oldest biographical dictionaries is that the history of the Islamic Community is essentially the contribution of individual men and women to the building up and transmission of its specific culture."21 At the same time, another of still more general importance related to the Hadevelopment dith and the isnad mechanism was the very early extension of the use of the brief, hadith-style report and of the isnad well beyond the Hadith literture per se to virtually any form of writing in any discipline of the religious and historical sciences. Both early and late, the use of the matn-isnad format as the basic building block of a text was characteristic of scriptural exegesis (tafsir),prophetic biography and the history of the Prophet's era (sirahand maghazi),
THE TRANSMISSION OF KNOWLEDGE
20 On the rijal genre, see Hamilton A. R. Gibb, "Islamic Biographical Literature," in Bernard Lewis and Peter M. Holt (eds.), Historiansof the MiddleEast (London, I962), 5458; as-Salih, 'Ulam al-hadtth, 335-400; Tarif Khalidi, "Islamic Biographical Dictionaries: A Preliminary Assessment," Muslim World,LXIII (I973), 53-65; Rosenthal, Historiography des (see n. Io), 93-95, Ioo-Io6; Otto Loth, Das Classenbuch Ibn Sa'd (Leipzig, 1869); idem, Gesell"Ursprung und Bedeutung der Tabakat," Zeitschriftder DeutschenMorgenlandischen schaft, XXIII (I869), 593-614; Fritz Krenkow, "The Tarikh of the Imam al-Bukhari," Islamic Culture, VIII (1934), 643-648; Siddiqi, Hadith Literature,I65-I88; Robson, "AlDjarh wa'l-ta'dil," Encyclopaedia Islam, New Edition, s.v. The rijal works cited are: Ibn of al-kabir (Beirut, I957-58); Ibn al-Kalbi (ed. Werner Caskel and G. Sa'd, Kitab at-tabaqdt Strenziok), Jamharatan-nasab,Leiden, 1966); al-Baladhuri (ed. 'Abdal'aziz ad-Duri et al.), Kitdbansabal-ashrdf(Wiesbaden, 1978); Ibn Hazm (ed. Evariste Levi-Provenqal), Jamharat ansdbal-'arab, Cairo, 1367/I948); Ibn Hajar, al-Isdbah tamyiz as-sahdbah f (Calcutta, I32931); idem, Tahdhib at-tahdhib (Hyderabad, 1325-27); idem, Lisdn al-Mizdn (Hyderabad,
I329-3I);

idem, ad-Durar al-kamina (Hyderabad,

I348-50).

On Ibn Hajar, see Encyclopaedia

of Islam, New Edition, s.v. Ibn Hadjar al-'Askalani. 21 Gibb, "Islamic Biographical Literature," 54.

510

WILLIAM

A. GRAHAM

all forms of history (ta'rzkh),and even legal discussion (fiqh), in addition to the Qur'anic and Hadith sciences. Ultimately, the isnad came to function also as the authoritative mechanism of transmission of all learning and scholarship as well as of guild and craft traditions, futuwwahs (young men's organizations), and Sufi affiliations (tartqahs).In this way the unbroken personal linkage or "connection" with the Prophet and/or the Salaf was symbolically if not actually guaranteed in every domain of the tradition. The importance of the isnad paradigm for transmission of knowledge in Islamic cultures can best be seen in one specific concept that grew up in connection with the development of the isnad as a technical mechanism and the evaluation of the rijal as transmitters. This is the concept of mutawdtir ("possessing unbroken succession [in transmission]"), a status achieved by any transmitted material considered to have so many guarantors in every generation of Muslims reaching back to the Prophet that there could be no possibility of an error having been introduced into the transmitted report. To say that a report is mutawatir is to say that it possesses the highest level of reliability: it has been transmitted by many different chains of transmission with trustworthy guarantors at every link, and its veracity cannot be doubted by any faithful Muslim. This concept was and is absolutely essential to the assurance of the integrity and perfection of not only "sound" hadiths but also and especially the text of the Qur'an. Only widespread memorization, study, and teaching of the Qur'an by many capable, trustworthy, and pious Muslims in every generation could protect the divine text from the textual corruption that had struck every previous scripture revealed to other peoples. The eventual growth of seven recognized mutawatir traditions of recitative Qur'anic transmission remains the basis for the authoritative text and accepted variant readings to the present day. The best text of the Qur'an was established on the basis of this recitative tradition, not on the manuscript fragments from the first centuries of Islam.22
I am concerned only with the theoretical and the perceived functioning of the isndd system, not with a critique or defense of it as to the degree to which forgery or error was possible-a complex issue dealt with for centuries by Muslim scholars and repeatedly by modern orientalists. On khabarat-tawdturand its opposite, khabaral-dhdd,see Al-Khatib al-Baghdadi, al-Kifayahft'ilm ar-riwdyah (Hyderabad [Deccan], 1357/1938), I6-17; as-Silih, 'Ulum al-hadith, I47-15I. On the science of Hadith, see the classical handbook of Ibn as22

TRADITIONALISM

IN ISLAM

5II

THE IJAZAH SYSTEM

Historically, the most important extension

of the isnad paradigm was into the fundamental framework of education and scholarship, a phenomenon that has both characterized and sustained Islamic culture wherever it has spread. The basic system of "the journey in search of knowledge" that developed early in Hadith scholarship, involved traveling to specific authorities (shaykhs), especially the oldest and most renowned of the day, to hear from their own mouths their hadiths and to obtain their authorization or "permission" (ijazah) to transmit these in their names. This ijazah system of personal rather than institutional certification has served not only for Hadith, but also for transmission of texts of any kind, from history, law, or philology to literature, mysticism, or theology. The isnad of a long manuscript as well as that of a short hadith ideally should reflect the oral, face-to-face, teacher-to-student transmission of the text by the teacher's ijazah, which validates the written text. In a formal, written ijazah, the teacher granting the certificate typically includes an isnad containing his or her scholarly lineage of teachers back to the Prophet of Companions, a later venerable shaykh, or the author of a specific book. This expresses vividly, as Nashabi pointed out, "a consciousness on his [the teacher's] part of the fact that knowledge is essentially cumulative." In addition, as Makdisi remarked, "the numerous certificates of audition written and signed by the authors of books, or by persons duly authorized in succession, attest to the perennial personalism of the Islamic system of education."23
Salih (d. 643/1245), al-Muqaddimah 'ulum al-hadith (Damascus, n.d.); Siddiqi, Hadith ft Literature,164-204; Yahya b. Sharaf an-Nawawi (d. 676/1278), Taqrtbfann 'ulam al-hadith (Cairo, I388/I968) (French trans. William Marcais, Journal Asiatique, XVI (I900), 315346, 478-53 ; XVII (I90I), IOI-I49, 193-232, 524-540; XVIII (I90I), 61-146); the modern summation of as-Salih cited just above. On the Qur'anic science of recitative transmission, see Graham, Beyond the Written Word (Cambridge, 1987), 79-I 5; works cited there, especially Labib as-Said (trans. Bernard Weiss, Muhammad Abdul Rauf, and Morroe Berger), The RecitedKoran (Princeton, 1975); Kristina Nelson, The Art of Recitingthe Qur' an (Austin, 1985). 23 On ijazah, Franz Taeschner's comments on the spread into diverse arenas (mystical, theological, legal, and popular) of the ideal of an isnad reaching back to the Prophet or Companion ("Das Futuwwa-Rittertum des islamischen Mittelalters," in Richard Hartmann and Helmuth Scheel [eds.], Beitrdgezur Arabistik, Semitistikund Islamwissenschafi [Leipzig, Studien, II, I88-I93 (also contains examples 1944], 364). Cf. Goldziher, Muhammedanische of abuse of ijazah without personal contact); Georges Vajda, Ignaz Goldziher, and S. Adrianus Bonebakker, "Idjaza," Encyclopaediaof Islam, New Edition, s.v., especially the bibliographies; Aloys Sprenger, "Uber das Traditionswesen bei den Arabern," Zeitschrifi

512

I WILLIAM

A. GRAHAM

The journey, or rihlah, tradition of personal study with outstanding teachers, wherever they might be, also rendered an intangible service to the continuity of Islamic tradition across the centuries: it hallowed and thus encouraged doing what a famous, if likely spurious, hadith urges: to "seek knowledge, even though [it be] in China." As a result, Muslim scholarship has always been and remains international. The lines of connecting isnads crisscross the geographical and political divisions of the Islamic world just as Muslim students and scholars have done for centuries. The goal has always been to build the strongest links with past learning, which has always meant going to the most venerable teachers. Even after the schools of religious learning, or madrasahs, developed in the fourth/tenth and fifth/eleventh centuries, the teaching and certification system remained individual and personal, not corporate as in the Western university, and students continued to travel in search of teachers. That is to say, one acquired scholarly credentials not by a diploma certifying a completed course of study, but by acquisition of a series of personal certifications of "permission" to transmit and to teach specific texts and knowledge learned at the feet of particular scholars. The large and important genre of works known under the rubric of mashyakhah,or "catalogue of shaykhs," amount to the amplification of the teaching isnad into personal biographical dictionaries of all of the teachers from whom one has obtained an ijazah to transmit and teach. They testify to the importance throughout Islamic history of the solid intellectual and religious repute of the persons from whom one has one's learning, wherever they may have been found.24
der DeutschenMorgenldndischen Gesellschaft,X (1856), 1-17; al-Baghdadi, Al-Kifayah (see n.22), 311-355; 'Abdullah al-Fayyad, al-Ijdzdt al-'ilmfyah 'ind al-muslimin(Baghdad, 1967); Ahmad Shalaby, History of Muslim Education(Beirut, 1954), 147-149; Muhammad 'Abd ar-Rahim Ghunaymah, Ta'rikh al-jdmi'dtal-isldmiyahal-kubrd(Tetuwan, 1953), 219-231; Johannes Pedersen (trans. G. French), The Arabic Book, (1946, Princeton, 1984), 24-36; Hisham Nashabi, "The Ijaza: Academic Certificate in Muslim Education," Hamdard Islamicus, VIII (1985), 7-15; Makdisi, The Rise of Colleges (Edinburgh, 1981), 140-146 on the ijazah. A fine new study is Jonathan Berkey, The Transmission Knowledgein Medieval of Cairo: A Social History of Islamic Education(Princeton, 1992). On the personal nature of the ijazah connection, see 21-43. See also Louis Massignon, "Etude sur les 'isnads' ou chaines de temoinages fondamentales dans la tradition musulmane hallagienne," Melanges Felix Grat (Paris, 1946), I, 385-420 [partially trans. by Herbert Mason in Massignon, The Passion ofal-Halldj (Princeton, 1982), III, 245-277], on oral vs. written transmission. 24 On international links of learning, see John O. Voll on eighteenth-century Meccan-

TRADITIONALISM

IN ISLAM

513

Here again, we see a keen awareness of the intensely personal character of knowledge, or 'ilm: a book or a school of thought from a master in a given field is of no use without the personal instruction in, and certification of mastery of, its content that can only be given by a teacher at the end of a chain of scholars going back ideally to the author of the book or founder of the school of interpretation. Any such author or founder-figure stands also at the end of a chain of transmission going back to Muhammad and the Salaf themselves. The major schools of jurisprudence, for example, trace their teaching and textual traditions via isnads going back to eponymous founder-figures such as Ibn Hanbal
(d. 241/855) for the Hanbalis or Malik ibn Anas (d. 179/794) for

the Malikis; each of these figures is in turn linked through a short isnad of trustworthy persons to the Prophet and the pristine Ummah. Legitimacy has always been a key issue, for, as Trimingham remarked, "a sound isnad can support a multitude of heresies," and theologians as well as legists and mystics have been ever eager to ground their ideas in the "orthopraxy" of Muhammad and the Salaf.25
Medinan teaching connections to Africa and South, Southeast, and Central Asia, e.g., "Hadith Scholars and Tariqahs: An Ulama Group in the i8th Century Haramayn and Their Impact in the Islamic World," Journal of Asian and African Studies (Toronto), XVI (I980); idem, "Linking Groups in the Networks of i8th Century Revivalist Scholars: The Mizjaji Family in Yemen," in Voll and Nehemia Levtzion (eds.), Eighteenth Century Renewal and Reform(Syracuse, I987); idem, "Muhammad Hayya al-Sindi and Muhammad ibn Abd al-Wahhab: An Analysis of an Intellectual Group in Eigheenth Century Madina," Bulletin of the School of Oriental and African Studies, XXXVIII (1975), 32-39. On Muslim education, including numerous references to isnad and ijazah systems, see Roy M. Mottahedeh, The Mantle of the Prophet(New York, i985); the works of Makdisi, Ghunaymah, Shalaby, and Berkey cited in n.23; Abdul LatifTibawi, "Muslim Education in the Golden Age of the Caliphate," Islamic Culture, XXVIII (1954): 418-438; Makdisi, "Muslim Institutions of Learning in Eleventh-Century Baghdad," Bulletin of the School of Oriental and African Studies,XXIV (I96I), I-56; the response to this by Tibawi, "Origin and Character of al-Madrasah," Bulletin of the School of Orientaland African Studies, XXV (1962), 225-238. On the mashyakhah and (also termed barndmaj) ijizah, and for translated examples of same, see Georges Vajda (ed. Nicole Cottart), La transmission savoir en Islam (London, I983); du dans les manuscrits arabes de la Bibliotheque idem, Les certificatsde lectureet de transmission Nationale de Paris (Paris, I956); idem, Le "Dictionnaire autorites"de 'Abd al-Mu'min addes de Dimydti (Paris, I962); Jacqueline Sublet, "Les maitreset les etudesde deux traditionnistes l'epoquemamelouke,"Bulletin d'EtudesOrientales, XX (I967), 7-99; Pedro Chalmeta, "Le Barnamag d'Ibn Abi '1-Rabi," Arabica, XV (I968), I83-208. Cf. Arthur J. Arberry's translation of a list of books received by audition or ijazah by Radi ad-Din al-Qazwini (d. 590/II94), A Twelfth-Century Reading List (London, 1951), which gives for each book listed an isnddgoing back usually to the author. 25 J. Spencer Trimingham, The Sufi Ordersin Islam (Oxford, 1971), 5.

514

WILLIAM

A. GRAHAM

Traditionally, there has been no true education, scholarship, or authority in any field without personal engagement in a "golden chain" reaching back to Muhammad and the Companions, and this principle extends across a broad spectrum of fields and concerns in Islamic societies. There are, in particular, three further major and widespread traditions of Muslim faith and practice in which the isnad paradigm plays a central role: the Sufi tradition, the Shi'i tradition, and what, for want of a comprehensive technical term, can be loosely termed "sharifism" or "legitimation by birth," to refer to the widespread Muslim tradition of tracing genealogical descent back to the Prophet, his immediate family, or to his tribe of Quraysh.
SUFI AFFILIATIONS The Sufi tradition is one of experiential and internalized "mystical" piety that has historically played so major a role in shaping Muslim values, worship, theology, conversion, and social order that it has functioned almost as an equal partner with the "'ulama' tradition" of "normative" religious and legal learning in defining actual Muslim piety and practice. Together, these are the two primary strands of Muslim religiousness that have characterized Islam among both Sunnis and Shi'is. Their divergences have often been a source of conflict and tension in the Ummah. The 'ulama' have been the guardians of what Hodgson has called "shar 'ah-mindedness," or the shar'i Muslim vision of life lived according to the dictates of the Shari'ah, or Muslim "law." Typically they have been chary of Sufi excesses and critical of antinomian strains in sufism, just as many Sufis have been critical of what they perceive in shar'l piety as superficiality and lack of sufficient stress on inward spirituality. Yet for all of the varied forms that Sufi religiosity has taken and the often sharp critique Sufis have leveled against average piety and rote observance, major Sufi thinkers and leaders have seen themselves not so much as opponents of Sharl'ah-minded piety championed by the 'ulama', but more as proponents of an inner, esoteric, or simply more spiritual understanding and observance of the more "external," exoteric, or orthoprax traditions of the Shari'ah as codified and cultivated by the 'ulama'. Similarly, many of the greatest legal and religious scholars within 'ulama' scholarly traditions have been themselves also members of Sfil orders and authored works important to both camps. Thus we must con-

TRADITIONALISM

IN ISLAM

515

stantly guard against portraying the shar'i and Siifi traditions as mutually exclusive, closed communities on the order of competing sects or "churches."26 As a major strand in Islamic religiousness, sufism did develop from pietistic, usually individualistic, beginnings in the first three or four Islamic centuries into an array of discrete teaching traditions or "paths" (tariqahs). Around these teaching paths there crystallized from about the fifth/eleventh and sixth/twelfth centuries onward formal soldalities or orders (ta'ifahs)with their own centers and hierarchies. Each of these orders perpetuated a particular inner or esoteric teaching traced back to a master teacher from whom the tariqah or ta'ifah often takes its name (for example, the Badawiyah or Ahmadiyah, after Ahmad al-Badawi [d. 675/1276]; the Qadiriyah, after 'Abd al-Qadir al-Jilani [d. 56I/ I166]). As we shall see below, behind this founder-figure, the isnad of every tariqah is normally traced still further back, to famous early exemplars of Muslim spirituality, especially among the Prophet's Companions (most often Ali ibn Abi Talib), and so to the Prophet himself. For the Sufis, prophecy (nubuwwah)was replaced after the death of the last Prophet with sainthood or the state of nearness to God (wilayah), which has its own succession. Thus we find at the center of systematized Sufi thought the key concept of an initiatory isnad of spiritual guides or masters (murshids, shaykhs, pirs) that is most commonly referred to as a "chain," or silsilah. The spiritual power (barakah)of charismatic religious figures is thus perpetuated in a line of spiritual descent that links each new generation of adepts and lay followers of an order to the spiritual authority of the Prophet-and ultimately to God-through the Companions and the best of their successors. So pivotal is the notion of the silsilah that it is used generally in
26 On sufism, the best English survey is that of Schimmel, Mystical Dimensionsof Islam (Chapel Hill, N.C., 1975). For the fullest picture of classical Sufi thought, see the major Sufi handbooks, e.g., Abfi'l-Qasim al-Qushayri (d. 466-7/1074), Risalahfi'ilm at-tasawwuf (Cairo, 1368/1948); Abf Nasr al-Sarraj (d. 378/988) (ed. Reynold A. Nicholson), Kitab al-luma'fi't-tasawwuf;(Leiden, 1914); Abu Talib al-Makki (d. 386/996), Qtutal-qulub(Cairo, 1310/1892-93); Abf Bakr al-Kalabidhi (d. c. 380/990), (ed. Arberry) at-Ta'arrufli-madhhab ahl at-tasawwuf(Cairo, I934) (trans. Arberry, The Doctrineof the Sufis [Cambridge, 1935, I977]). On "shart'ah"-mindedness," see Marshall G. S. Hodgson, The Ventureof Islam (Chicago, 1984), I. Makdisi has pointed out how Hanbalites such as Ibn Qudamah (d. 620/1223) were both ulema and members of Sufi orders; see, e.g., "L'isnad initiatique soufi de Muwaffaq ad-Din Ibn Qudama," Cahiersde L'Herne (Paris, 1971).

5I6

WILLIAM

A. GRAHAM

Persian and also in many Sufi traditions throughout the Islamic world to designate an entire order and/or its tradition of teaching and initiation. Thus it serves widely as a synonym of ta'ifah, tariqah, or even khirqah(the "cloak" of the master, passed on as symbol of investiture and succession in the tariqah).27 In Sufii as in other Islamic religious and intellectual traditions, one acquires a teaching silsilah only through a formal ijazah from one's shaykh-a practice in place apparently since even before the emergence of formalized tariqah orders. Although scholars often cite as the oldest Sufii silsilah the isnad traced from the mystic Ja'far al-Khuldi (d. 348/959) back to the Companion, Anas b. Malik, it was much later, in the specific context of the major growth and spread of Sufi orders from especially the seventh/ thirteenth century forward that the mystical isnad flourished and became a fixed and defining institution in Sufi tradition, even if it was often in some part legendary (primarily at its older end). Used ever since as an indication of initiation into the devotional exercises-especially the ritual chanting (dhikr)-and the mystical, often esoteric teaching of a particular order or branch order, the Suifi silsilah became the emblem of allegiance to the all-powerful spiritual master who is the one true "door" to spiritual truth for the seeker or disciple. A major part of the master's claim to this position of absolute spiritual authority is his or her place in a silsilah of teaching and initiation reaching back to the Prophet.28 The specific silsilahs of the various orders and their branches are both as simple and as complex as genealogical "trees" (in
On the rise of Sufi orders, see Gramlich, Derwischorden, 14I-143. See also Louis II, Massignon, "Tarika," Encyclopaediaof Islam, New Edition, s.v. On the use of silsilah to mean ta'ifah or tariqah, see Gramlich, Derwischorden, 2. On the importance of the silsilah I, in demonstrating the spiritual pedigree of the shaykh, see Christian Snouck Hurgronje, Mekka in the LatterPart of the 1gth Century(Leiden, 193 I), 209. On khirqah, see Jean-Louis II, Michon, "Khirkah," Encyclopaedia Islam, New Edition s.v.; Gramlich, Derwischorden, of 173-175; Seyyed Hossein Nasr, Sufi Essays (London, 1972), io6-IIo. 28 For examples of transmission of a Suffi ijazah from different Muslim lands, see Snouck Hurgronje, Mekka, 276 (in Mecca); Edward B. Reeves, The Hidden Government(Salt Lake City, I990), 59 (in upper Egypt); Michael Gilsenen, Saint and Sufi in ModernEgypt (Oxford, 1973), 214, 237 (in lower Egypt); Muhammad S. Siddiqi, The BahmaniSufis (Delhi, 1989), 52, 92, I27-n.22, 156 (in India). On the Khuldi initiatory isnad, see Massignon, Essai sur les origineslexique techniquede la mustiquemusulmane(1922, Paris, 1968; 2d ed. rev.), 128; but cf. Gramlich, Derwischorden,II, 171-172, who sees the formal silsilah as older. On traditions about the oldest Sufi silsilahs, see Trimingham, Sufi Orders,261-263; cf. 1-30. On the often fictive links in Sufi initiatory isnids, see Massignon, Essai, 128-134.
27

TRADITIONALISM

IN ISLAM

517

Arabic as in English, shajarah,"tree," is used verbally and pictorially to represent family genealogies and sometimes mystical or other teaching lineages as well). The investiture of the Sufi into a particular "path" receives its legitimacy and authority through the silsilah into which he or she gains acceptance. Since the fifth/ eleventh century, almost all tariqahs have traced their silsilahs back to Muhammad through his son-in-law and cousin, 'Ali ibn Abi Talib, most often through the famous Muslim pietist, alHasan al-Basri (d. 110/728) and the famous "orthoprax" Sufi, alJunayd al-Baghdadi (d. 298/9Io). The inclusion of such giants of the tradition in a chain going back to Muhammad guarantees the dignity and power of the spiritual lineage, especially when one takes into account, first, the immense importance throughout medieval and all later Islamic history of Muslim piety focused on emulation of and devotion to the Prophet, and, second, the persistent idea in Muslim tradition, Shi'i and Sunni alike, that Muhammad's esoteric knowledge (usually identified as derived from his experience of the divine presence on the night of his fabled Mi'raj, or Ascension) was passed on to 'Ali in particular.29 The silsilah lineages of mystical investiture came in time to be bound up intimately with genealogical lineages as well. The Mawlawiyah order stemming from Jalal ad-Din Rumi (d. 672/ 1273) was hereditary virtually from the start, and more commonly from the fourteenth and fifteenth centuries, the cloak of leadership in many other Sufif orders came to be passed on in a family line as a hereditary transmission, a development that only increased the importance of the silsilah even as it altered its traditional form and meaning. Thus arose the distinction between a silsilat ad-dam or "blood lineage" (or silsilat an-nasab, "genealogical lineage"),
I 29 For examples of specific tariqah silsilahs, see: Gramlich, Derwischorden, (on the major Iranian Shi'i orders: Dhahabiyah, Ni'matullahiyah, and Khaksar, all tracing their lines through the early Shi'i imams to 'Ali and then the Prophet, but all with their peculiar silsilahs); Hans J. Kissling, "Aus der Geschichte des Chalvetijje-Ordens," Zeitschriftder DeutschenMorgendlidischen Gesellschaft,CIII, (I953), 283-285 (on the Kahlwatiyah silsilah back to 'Ali via links including al-Junayd and al-Hasan al-Basri); Nasrollah Pourjavady and Peter L. Wilson, Kings of Love (Tehran, 1978), 246-248 (on the Ni'mutullahiyah affiliation going back to the Prophet); 'Abd al-Qadir as-Sufi, The Way of Muhammad (Berkeley, 1975), frontispiece (on the Habibiyah-Shadhiliyah silsilah back to the angel Gabriel, through 'Ali); Dale Eickelman, MoroccanIslam (Austin, 1976), 185-186 (on the spiritual lineage of the Sherqawiyah, traced to the Prophet through a marabout chain); Vincent Crapanzano, The Hamadsha (Berkeley, 1973), I-4, 69-71 (on the Hamdushiyah claim to connection with the Prophet, in part by blood descent).

518

WILLIAM

A. GRAHAM

and a silsilat al-barakah "spirituallineage." Muslims and students or of Muslim mysticism have often seen in the development of hereditary leadership of Suifi hierarchies throughout the Islamic world the degeneration of sufism as a vital movement. Any such judgment has to be made on a case-by-case basis. As Gilsenan noted, most members of orders would "insist that descent is not the principle" on which the spiritual leadership is based, even if blood lineage (nasab) is also prominent in a silsilah. Nor is the issue of emphasis upon nasab only a matter of familial lines of local Sufi orders; the mixing of spiritual leadership of many tariqahs with claims to blood descent from the Prophet, especially through 'All, further complicates the picture. The entire issue of the joining of spiritual isnads to genealogical nasab is an important and insufficiently investigated aspect of Muslim social and religious history.30
SHI I ATTACHMENT TO THE PROPHET'S LINEAGE

A closely related

phenomenon is the nexus of Sufi and Shi'i tradition in attachment to blood descent from the Prophet, something that exemplifies Shi'i piety and figures more than occasionally in Sufi as well as in Muslim popular piety. The majority of Shi'is look for fully authoritative guidance to the direct descendants of the Prophet through 'All and Fatimah; some Shi'i groups, however, have followed indirect descendants such as the line from 'All and a second wife, "the Hanafi woman," through their son, Muhammad ibn al-Hanafiyah, or even collateral lines of "the people of the House [of the Prophet]," ahl al-bayt. All of these traditions of emphasis on prophetic nasab are based upon a different form of the isnad paradigm, namely one forged by genealogical ties rather than by teaching or spiritual initiation and transmission alone. Yet even here, not all is vested in biological connection: Shi'i emphasis upon blood descent from Muhammad or his clan is tied inextricably to the transmission of divine wisdom (hikmah) and divine designation (nass) from the Prophet to his descendants, the line of the true Imdams. Whereas it is the sanctity of nearness to
See, e.g., Gilsenan, Saint and Sufi, 72-8I; Siddiqi, Bahmani Safis (Delhi, 1989), 2935, 19I-I94, and the many genealogical charts of Deccan Sufi silsilahs; Trimmingham, Sufi Orders, I5, 6i, 71-73, 83, I73-174, 255; Gramlich, Derwischorden, portrays in great detail and with lucidity and erudition the entire panorama of Shi'i Sufi traditions in which descent from 'Ali and the Imams is a given.
30

TRADITIONALISM

IN ISLAM

519

God (wilayah) that succeeds prophecy (nubuwwah)for the Sffi, for the Shi'i it is the Imamate (imamah),the true leadership of all Muslims, that is passed down in a silsilah distinguished by simultaneous claims to prophetic nasab and to spiritual gnosis that ensure divine guidance and religious temporal preeminence. Whether a given Shi'i tradition traces Imamah through twelve Imams only, like the Imami "Twelver" tradition of Iran, or through an unbroken chain from Muhammad to the present-day Imam, like the Isma'ili "Sevener" tradition of the Aga Khan's community, all look to connection with the Prophet's line as the key to religious truth and guidance.
SHARIFISM

The Shi'i tradition of attachment to Muhammad's

descendants takes its own unique form but is by no means limited to Shi'ism. Nor is the socioreligious importance of blood descent in Islamic societies limited to Shi'ism and Suifi hereditary lineages. Quite apart from either Shi'i or Sfif'i traditions, a major social fact of Islamic societies in virtually every age and place has been the special social status attached to anyone who could successfully claim prophetic nasab (also qarabah),or blood descent from Muhammad, his family, or his tribe of Quraysh-the common mark of "nobility" in Islamic societies, loosely termed "sharifism" (from the Arabic sharif, "noble").31 Traditionally, claim to sharifian lineage in the strict sense has meant either Hasani or Husayni descent, referring to lineal descent from one of the Prophet's two grandsons (the sons of 'Ali and Fatimah), al-Hasan and al-Husayn. Here we see the close ties of sharifian ideas to Shi'i notions of the special status of Muhammad's descendants. Like sharif, the honorific sayyid, "lord, master" (plural, sadah), applied in the sources to both Hasan and Husayn, has traditionally referred to descendants of either of these, although if a distinction is made, it is between ashrdfor
31 The most comprehensive modern overview of sharifism is Cornelis van Arendonk, Encyclopaedia Islam, New Edition, s.v. Sharif Michael Winter, "The Ashrdf and Niqdbat of al-Ashrdfin Egypt in Ottoman and Modern Times," Asian and AfricanStudies,XIX (I985), I7-4I, provides both a succinct introduction to the general phenomenon and a lucid case study of sharifan status in Mamluk and Ottoman Egypt. Both of these articles contain excellent bibliographies on the subject and have served as my principal sources for the present overview. On the unique case of Morroccan sharifism, see Mohamed Kably [Muhammaad Qabli], Societe, pouvoir, et religion au Maroc a la fin du "moyen-dge" (Paris, 1986), 219-338.

520

WILLIAM

A. GRAHAM

shurafa(plurals of sharif) as descendants of al-Hasan and sddahas descendants of al-Husayn. In some contexts, such as India, sayyid is used more commonly than sharif to refer to any Prophetic descendant; more commonly, as for example in the Magrib, sharif (dialect plural shurfa')is the term of choice for this general usage. In the case of Sufi, Shi'i, or Sunni emphasis upon the nobility conferred by Prophetic blood, it is difficult to assess just how much, if any, of the strong pre- and early Islamic emphasis upon Arab lineage has had its influence. Alongside nobility conferred by Prophetic descent there exist also national, regional, dynastic, or ethnic nobilties conferred by Arab, Iranian, or other specific parentage. Arab descent, in particular, has played an important social and political role in diverse Islamic societies from Southeast Asia to Spain. Despite the consciously egalitarian social thrust of the Islamic message, the importance in Islamic societies past and present of Arab descent specifically from the tribe (Quraysh) or family (BanufHashim) of Muhammad is clear.32 In the course of Islamic history, sharaf al-'Arab, or Arab nobility, like that of other national or ethnic nobilities that had challenged it, was superseded by Qurashi or Hashimi blood descent from Muhammad's relations as the final measure of true noble ancestry in Islamic societies. In this regard, Mottahedeh cited from as-Suli's Akhbar a telling claim to true (that is, Prophetic) nobility ascribed to the 'Abbasid caliph ar-Radi (A.D. 934940): "The Prophet is one of us, with whom no one, either Arab or non-Arab, can be compared." Similarly, the claim ascribed to Muhammad and cited by Ibn 'Abd Rabbihi, "I am the best of you in family and the best of you in genealogy," neatly legitimizes his blood descendants as the true "nobles" of Islam. From local socioeconomic elites and ruling families to Shi'i communities and Sufi orders, prophetic sharaf has often been merged with whatever other factors and traditions exist-economic, local aristocratic, tribal, political, and spiritual-to strengthen a given claim to leadership or special privilege in this world (and even, as some traditions have it, at the Last Judgment). From early on, special tax relief and stipends from the state were given the ashraf; later,
32 On the importance of ancestry among the Arabs before Islam, see Goldziher, Muslim Studies, I, 4I-5o. For genealogical works on the Arabs, cf. those of Ibn al-Kalbi, alBaladhuri, and Ibn Hazm in n.2o.

TRADITIONALISM

IN ISLAM

521

likely only from the eighth/fourteenth century, in many Islamic societies they were either ordered or allowed to wear some form of green clothing or marking to distinguish them visually. A formal office of "marshal (naq^b)of the ashraf," was typically established or recognized by Islamic governments, even though its actual importance has varied greatly. However great the prerogatives it confers, the strong consciousness of sharifian descent can be documented across the Islamic world to the present day.33 In sharifism we can see a genealogical reflex of the importance of a chain of persons, or isnad, connecting later generations to the Prophet and Companions. Working on Islam, we find at every turn passing reference to the ashraf, especially in the ethnographical literature, so that their importance in diverse Muslim settings, east and west, can easily be seen.34 The isnad paradigm informs and undergirds a Muslim traditionalism that is characterized by what I have termed ittisaliyahthe sense of connectedness of each new generation with both the beginnings of the Ummah formed under God's final Prophet and all the intervening generations of faithful "servants of God" who have sustained the traditions and ideals of that original Prophetic
33 Mottahedeh, Loyalty and Leadershipin an Early Islamic Society (Princeton, 1980), I04 (see 198-204 on the phenomenon of noble nasab generally in the fourth and fifth centuries A.H.); Ibn 'Abd Rabbihi, al-'Iqd al-fartd (Bfilaq, 1293), II, 247, cited in van Arendonk, "Sharif," 325. Any full exploration of the proposed isnad paradigm would have to include investigation of the importance of the ansdb al-'Arab, or "Arab genealogies," to which many extensive works of Muslim scholarship have been devoted. On the sociopolitical importance of the sharifi lineage of saints in Morocco, see, e.g., Ernest Gellner, "Doctor and Saint," in Nikki Keddie (ed.), Scholars, Saints, and Sufis (Berkeley, 1972), 307-326. On special privileges of the ashraf, see, e.g., Zilfi, Politics of Piety, 95. On special divine li-dl (Cairo, pardon for Ahl al-Bait, see Yusuf b. Isma'il, ash-Sharafal-mu'abbad Muhammad 138I/196I), II-13, 45, 76-79. On the importance of one's nasab on the Day ofJudgment, see al-Majlisi, Bihar al-anwar (1883-87), III, 259-262. On green markings for clothing or turbans, see, e.g., Edward W. Lane, Mannersand Customsof the ModernEgyptians (1836;
reprint ed., New York, 1973), 31, 132, 140.

34 Sharifism as a typical element in diverse Muslim social contexts has not been as thoroughly studied as it deserves, even though we have from van Arendonk and Winter, respectively, a superb general and a superb specific article on the topic. There is still a need to examine the extensive overlap of emphasis upon Prophetic nasab with other forms of Muslim socioreligious status as a further instance of the "connectedness" that I am suggesting inheres in the various forms the isnad paradigm takes. For specific examples of the ashraf's social and religious roles, see, in addition to van Arendonk and Winter: Adam Mez, Die Renaissancedes Islams (Heidelberg, I922), 144-152; Gellner, Saints of the Atlas (London, I969), 70-80; Crapanzano, Hamadsha, I-4; "Mrs. Meer Hassan Ali," Observationson the Mussulmaunsof India (1832; reprint ed., Delhi, 1973), 6-io; Reeves, Hidden Government,I6--8, 58, 130; Zilfi, Politics of Piety, 114, 215.

522

WILLIAM

A. GRAHAM

community even in eras and places in which these were far from realization. It is difficult to exaggerate the power and importance of the widespread sense of this historical-biographical connectedness among Muslims down to the present moment, or the degree to which it is expressed in the form of the root metaphor
of the isnad paradigm.35

For most Muslims, religious traditionalism has been most readily expressed through specific historical connection to a past formed by connected persons. The isnad is a "continuous support" only insofar as it is an unbroken chain of trustworthy persons whom one can name, and whose personal authorization, or ijdzah, confirms the reliability of whatever tradition has been transmitted through so many generations. It is in this that the Muslim spirit of traditionalism lies, not in some imagined atavism, regressivism, fatalism, or rejection of change and challenge-especially since this same traditional ittisaliyah has served modernists as well as reactionaries as authority for their ideas. What the isnadparadigm reminds the student of Islam, and, at some deeper level, what it presumably reminds the Muslim specifically, is that a personally guaranteed connection with a model past, and especially with model persons, offers the only sound basis in an Islamic context for forming and re-forming oneself and one's society in any age. How one chooses to do that-and one must do it willy-nilly-is one's own responsibility; Islamic history reveals myriad ways in which it has been attempted. Whatever its forms, being religious
"Islamically" has meant taking history seriously, and ultimately so, for history or, still more, biography (the history of persons) represents the prime medium through which authority and truth have been transmitted and thus made available for each new generation.
35 The present work has attempted not to prove, but to offer the hypothesis that this paradigm may be seen as a "deep structure" in those cultures linked by their Islamic impress. It is intended to serve as a stimulus to further investigation and refinement or emendation of this idea; much work remains to be done on the presence and forms of the paradigm in the various spheres of Muslim life.