Are representations of the Prophet Muhammad permitted in Islam? To make or not to make images of the Prophet: that is the question I will try to answer. It is an unexpectedly burning question, as the newspapers regularly demonstrate. But both the answer to the question and the reasons for raising it require a broader introduction.

There have been many times in recent years when one bemoaned the explosion of media that have provided public forums for so much incompetence and ignorance, not to speak of prejudice. Matters became worse after September 11, for two additional reasons. The first is the propagation of a climate of fear, of ever-present danger from ill-defined foes, which led in the West, and especially in the United States, to a plethora of security measures ranging from reasonable and useful to ridiculous and demeaning. Penetrating and perverting institutions and individuals, this fear collided in the Muslim world with a complex ideological and psychological evolution that led many people in Muslim countries and communities to a reflexive and often self-destructive brutality in reaction to the slightest whiff of verbal or visual provocation.

The second reason is the exacerbation of a mode of judgment that is not new by itself but has in recent years acquired frightening dimensions. It consists in identifying the country--or religion, ethnicity, race, or any other general category of human association--of anyone responsible for a crime or misdeed, and then condemning the whole group for the action of a single person. The crimes and misdeeds, I should add, need not be recent ones. They can be--and often are--events of many years and even centuries ago. A cult of past and present horrors surrounds us. The paradoxical analysis of past evils according to contemporary norms has the effect of denying history, which has its own explanation of events.

Recently Yale University Press, one of the most distinguished university presses in America, agreed to publish The Cartoons That Shook the World by Jytte Klausen, an academically acceptable and well-researched study of the publication by a Danish newspaper, in 2005, of cartoons willfully showing the Prophet Muhammad in vulgar and politically charged ways, and of the turbulent aftermath of their publication. As is well known, several weeks after their appearance these drawings--which should be called caricatures rather than cartoons: a first example of the technical ignorance in the media's accounts of the story--were shown, and sometimes simply mentioned without being shown, in Muslim communities in Europe, and then in various parts of the Muslim world. This led to riots, with losses of life, in a few cities, and to the destruction and the boycott of Danish products.

Klausen, who provides a careful chronology of the events, is a Danish political scientist who teaches at Brandeis University. Her book was meant to include the images themselves (which are available on the Internet) as well as earlier, mostly Western, illustrations of the Prophet Muhammad in a variety of contexts, usually not in a terribly favorable light. But at the last minute, and in accordance with opinions provided by a wide variety of people, Yale University Press decided to drop all representations of the Prophet from a book whose subject is their impact. The argument of the press was that the images could be considered offensive by Muslims and lead to violence, to attacks on Yale and other American institutions.

The assumption that the masses in Karachi and Jakarta would have seen, or otherwise taken note of, a book from Yale is a bit presumptuous--unless, of course, they were prodded by the media's sensationalism, and its interest in stories of riots by uncouth youths worked up in their anti-American feelings (by this point Yale and its actual book would be long forgotten) by local purveyors of hate and destruction. Yale's decision is certainly a denial of free speech, though of course the argument can be made that a possible danger to people may compel restrictions in the expression of opinions and of facts. I am not persuaded by this argument about this book. And the deletion of the images is also--a far more important criticism in this instance--a gratuitous betrayal of scholarship, since many other books (including at least four published by Yale, two of them by me) do show images of the Prophet.

Here I must make a disclosure. Several years ago, in a book on the Dome of the Rock in Jerusalem that was published by Harvard University Press, I included a representation of a fourteenth-century Persian painting showing the archangel Gabriel bringing the city of Jerusalem to the Prophet Muhammad. The press requested that the section of the painting representing the Prophet be removed. First I objected and then I agreed, because its presence was not essential to my argument; but the episode left a bad taste in my mouth, a feeling of regret, especially in light of the fact that many learned books or journals, and even some popular ones, especially in Europe, publish pictures of the Prophet when such images are required by the text or proposed by their authors.

Generalities and disclosures aside, the substance of the dispute lies in the allegation made by Muslims, or at least some Muslims, and often repeated by the Western media, that representations of the Prophet are forbidden in Islam, and therefore that such representations as do exist, or have existed, within the Muslim world or beyond its borders are either sins or provocations. The conclusions to be drawn from such a view are obvious. Sins must be punished, and their repetition avoided; and provocations must be answered with vigor.

In reality, however, things are not so simple. In the past, and still today, pictures of the Prophet Muhammad have been produced, and are still produced, by Muslim artists for Muslim patrons. How do these images fit with the presumed existence within the Islamic world of a doctrine prohibiting all representations of living beings? To answer this last question, it is essential to understand the nature of a legal system that operates in the absence of an organization such as the church or of formal written codes of law accepted by the vast majority of those who claim to be Muslims.

From the very beginning of its existence, the Muslim world practiced and developed an elaborate legal system meant to control and to judge all aspects of life, but its totalistic ambition was often frustrated by its own sophistication and diversity. This system, known as sharia, was based on the Qur'an, an immutable divine revelation, and the hadith, a huge body of actions and statements attributed to the Prophet, whose authenticity--and reliability for believers--was discussed for centuries. The words of the Qur'an and the stories of the hadith were interpreted and re-interpreted for centuries by learned scholars and practicing judges, known respectively as fuqaha and qudat (the plural of qadi). Although a consensus was established on many issues, and was often adopted by the legal systems of Muslim states in our times, this consensus was not total or universal. With variations that arouse the passions of modern historians and politicians, the opinions and judgments of this tradition of legal interpretation can, in theory at least, range from absolute and constant to near-anarchical and open-ended.

The issue of the visual representation of human beings, and therefore of the Prophet too, belongs to the latter category. The Qur'an itself is silent on the subject. Only a single passage is usually quoted in discussions of the matter. This passage (3:43) relates the words spoken by God to Mary, the mother of Jesus, saying that her son will proclaim: "I come to you with a sign from your Lord. I will make for you out of clay the likeness of a bird, then I will breathe into it and it will become a bird, by the leave of God." This was understood by the majority of interpreters to mean that God alone can create life, and to imply that there is no point in representations other than to make them alive. Other passages that are sometimes adduced in discussions of representation refer to them as real or potential idols--which is to say, sinful less for what they are than for the behavior that they may encourage.

The fear of idolatry permeates the formative centuries (essentially the seventh and eighth of the common era) of Islamic culture, which is perfectly understandable when one recalls the importance of images in Christianity, Buddhism, Zoroastrianism, Hinduism, and whatever pagan traces had remained in the vast territory, from the Atlantic Ocean to the frontiers of China, taken over by a relatively small army of Arab Muslim conquerors and missionaries. The result of these contacts with a world replete with religious and other imagery was a refusal by Muslims to make images--what scholars now call aniconism; and the frequent substitution of writing for representation. There are occasional examples of the destruction of images, though in early times such iconoclasm is much rarer in Islamic lands than in Christian lands. Eventually--possibly as early as the end of the eighth century, according to a shaky scholarly consensus--the condemnation of all those who make images became the view of the majority of legal scholars. And yet a minority kept on maintaining that beauty pleases God and does not necessarily lead to idolatry.

The result of all these opinions and feelings was complicated: religious art, in mosques in particular, avoided and rejected images, while the secular art of princes, and later of wealthy city dwellers, ornamented their abodes and the things in their possession with all sorts of representations. In other words, and in perfect harmony with the rich legal culture of the time, a range of possible attitudes toward religious imagery was maintained. Abstinence dominated, but it never became the only Muslim attitude or practice.

On the whole, especially when compared to the contemporary Eastern Christian world, which was rocked by the crisis of iconoclasm, the question of images was secondary within the thinking of the legal scholars, largely because neither the bases on which Islamic thought rests nor the specific needs of the Muslim faithful gave it much consideration. Although I am not familiar with the legal or theological literature of later centuries, or with the jurisprudential discourse in legal and theological schools in our own day, I suspect that the same comparative absence of extended reflection on the subject of iconography remained the case until the twentieth century, when technology made visual images ubiquitous. And even then the subject provoked relatively few comments. The one exception may be the milieu of Saudi Arabian wahhabism, where a doctrine of aniconic prohibition in official and public circumstances co-existed with the relatively open practice of displaying images in the privacy of homes or as homages to ruling princes. A sort of "don't ask, don't tell" policy seemed acceptable to the ruling classes, although rumbles for change could easily be felt in the few countries with forceful restrictions.

Did the pictures of the Prophet Muhammad exist within this double system? The historical evidence is very curious. During the first six centuries of Islam's existence, there is one very early coin type, an example of which is kept at the American Numismatic Society in New York, which may have represented the Prophet Muhammad; but this interpretation is far from being accepted by all scholars. And then, in the tenth century, there developed a fascinating story. It was said that there was an image of the Prophet in the possession of the Byzantine Christian emperor, which was shown by him to early Muslim ambassadors who came to convert him to Islam. The alleged image was in a collection of images of prophets from Adam to Moses, Jesus, and Muhammad. It served to demonstrate that Muhammad was indeed the last of the messengers sent by God for the salvation of man. One account of this image is that it was made by a Christian monk named Bahirah, who recognized prophecy in the youthful Muhammad accompanying a caravan of merchants from Mecca.

Until very recently, the sensuous representation of young Muhammad, as allegedly painted by Bahirah, could be acquired in Iran. It had been copied and reproduced in the many ways in which the commercial society of today distributes its most valued symbols. The image has undergone recent changes, and it no longer exhibits the slightly risqué sensuality of its early versions--whose original, according to the Persian inscription at the bottom of the image, was kept in a "western or Christian museum" (muze-i rum), a mysterious and perhaps mythical location which deserves its own philological and psychological analysis. The casuistic explanation by Islamic scholars for the existence of this image was that it belonged to a time that preceded the assumption of prophethood by Muhammad, and that therefore "Islamic" rules did not apply to it.

The alleged existence of such early images of the Prophet was fascinatingly developed among Chinese Muslims of later times. There is a story, told in various texts, that a Chinese emperor had heard of Muhammad and wanted to know more about him and his new beliefs. He invited Muhammad to visit China and to convert its people, but Muhammad thought the trip too complicated, and instead he sent a picture of himself. The Chinese emperor set it on his bed, and once the country had been converted, the picture disappeared. There are several variants to this story, which serves simply to show that, even when there is no sign of the existence of actual portraits, the idea of such portraits of holy figures existed, and was usually associated with others rather than Muslims, in this case with Western Christians or Chinese. As so often in every cultural system, foreigners are responsible for things that seem wrong or embarrassing.

What is more important, for our purposes, is that these stories--written by and for Muslims about non-Muslims--employed images to demonstrate the point of Muhammad's uniqueness within a divine revelation through a series of prophets that began with Adam. In a paradox that permeates much of early Islamic thought on this subject, this visual and theological uniqueness co-existed with the very human reality of a particular person's life and demeanor, which were described (and of course praised) in many accounts. As we shall see, this paradox affected the further history of images of Muhammad.

A major transformation took place in Islamic culture from around 1200 onward. The old centers of science and culture were profoundly affected by ethnic change (mass immigration of Turks), territorial change (the conquest of Anatolia and of northern India), social change (the growth of a feudal order and of an urban middle class), religious change (a complex symbiosis of Sunnism and Shiism, and the growth of mystical Sufi movements), and intellectual change (the emergence of a brilliant new Persian literature, and of vast theological and philosophical syntheses). In the Arab world, and especially in Iran, there was born a new art of book illustration. One of its earliest examples is the Persian manuscript of a mystical romance known by the name of Warqa wa Gulshah, now kept in an Istanbul museum, usually dated to the early thirteenth century and probably executed in Anatolia or in Iranian Azerbaijan.

In one of the miniatures in Warqa wa Gulshah, the Prophet Muhammad is seated on a princely throne and is surrounded by the first four caliphs, the so-called orthodox caliphs who succeeded the Prophet at the head of the Muslim community, who are shown like members of a feudal court, some carrying military symbols, others bureaucratic ones. Nothing distinguishes the Prophet from the other personages in the miniature. They all have a halo, a sign of honor probably picked up from Christian art. In another miniature, the Prophet is shown accomplishing the unique task of resurrecting two lovers who had already been buried. He is smaller than all the other figures in the painting, and is making a simple gesture of the hand to accomplish his task. Similar secondary appearances occur a few more times in fourteenth-century chronicles or romances, but not very frequently.

More interesting representations of the Prophet appear in two other manuscripts, both of which happen to be housed in Edinburgh. One is a copy of a celebrated History of Ancient Times by the great polymath Al-Biruni, who died around 1050. The manuscript dates from around 1300. One striking miniature in the manuscript shows Muhammad surrounded by his daughter Fatimah, her husband Ali, and their two children Hasan and Husayn--the first family of Shiism--greeting a Christian delegation from Najran in southwestern Arabia, an illustration of a well-known half-legendary event in the Islamization of the Arabian peninsula. All these personages are dressed in fancy clothes modeled on courtly vestments of the time, without any distinguishing or sacralizing sign.

The same is true of an even more remarkable series of representations of the Prophet found in the other manuscript of the time, dated 1314–1315, also in Edinburgh. It is a copy (only fragments remain from it) made in Tabriz of a world history, a Compendium of Histories, gathered by Rashid al-Din, the great vizier of the Mongol regime. There are no less than six preserved illustrations of the life of Muhammad, including his birth, his dramatic visit to Mecca as a young adult, the archangel Gabriel informing him of his vocation as a Prophet of God, and a stunning representation of his ascension into heaven, about which more will be said below. None of these images are given any formal, codicological, or iconographic features that would separate them from standard illustrations of historical texts.

This sort of inclusion of Muhammad within a broader body of illustrated material continued over the centuries, and in the sixteenth century it acquired a new vehicle in the Qisas al-Anbiya, or Stories of the Prophets, originally written in Arabic, and frequently illustrated in Persian or Turkish versions, many copies of which exist in most collections of manuscripts. Illustrations dealing with Muhammad are usually small in number, when compared with those that depict Moses, Joseph, or Jesus and many minor prophets from what is now called the "Abrahamic" tradition; and nothing distinguishes them in particular. They play a minor role in a genre that itself turned out to be relatively secondary in the tradition of painting in Muslim lands, which developed so brilliantly in the fifteenth to eighteenth centuries. But the important fact is that they existed, and there is no evidence that these images were criticized within the presumably religious milieu that created them. They must have fulfilled some purpose in the piety and the instruction of some faithful, though we do not know what that purpose was.

Something else also began in the early fourteenth century. In an album in Istanbul there is a remarkable series of large and beautiful miniatures depicting, without a remaining explanatory text, several episodes from a story that developed quite early within the Muslim religious system--the story of the Prophet's Night Journey to the masjid al-aqsa, the "farthest mosque," usually identified with Jerusalem, and then his ascension (mi'raj) to heaven and his encounter with God. At roughly the same time, in the early fifteenth century, a text known primarily by its Persian name as the Mi'rajnameh, or Book of the Ascension, was produced, based on earlier--and mostly Arabic--versions that have not been preserved or have not yet been discovered. The Persian text exists in several copies. The most spectacular of these, in the Bibliothèque Nationale in Paris, was written in 1436 in chaghatai, the Turkic language of the military class of Turks and Mongols who ruled over most of the Islamic world east of the Mediterranean. It depicts every detail of the Prophet's journey: his meeting with all other prophets, his seeing the tortures inflicted in hell as well as the rewards of Paradise, and his eventual encounter in a cloud of gold with the divine presence. The images are clear and direct, and of very high technical quality. The Prophet is consistently depicted with a crown on his head and riding the mythical beast Buraq, and he is preceded everywhere by the flying archangel Gabriel.

This manuscript is unique for the quality and the quantity of its images, but other "Books of Ascension" exist, some with a few illustrations. Yet there was one image created for the Mi'rajnameh that appeared in manuscripts of many other texts, sometimes with a minimum of inspiration from whatever passage seems to have been illustrated. This image shows a brilliantly lit sky, sometimes with clouds and sometimes with stars, and against the sky there is the Prophet on his mythical beast rising up into the heavens, with a host of angels holding lamps, crowns, and various gifts surrounding him. It is a heroic and brilliant procession which, owing to the consistent repetition of its details and the consistency of composition and colors, comes as close to being an icon as can be imagined within the Islamic world, even though it is found most of the time within the context of secular or mystical poetry.

Sometime in the fifteenth century, and certainly by the sixteenth century, it became customary to veil the face of the Prophet. The reasons for this veiling have not been clarified, but they seem to have arisen less from a formal decision by legal scholars than from a piety that, while not averse to the representation of human beings, concluded that the Prophet's uniqueness could best be expressed by making his face invisible. This procedure was then extended to all Shiite leaders, such as Ali and Husayn, and occasionally to other figures as well. What all of this seems to mean is that the representations of holy figures were distinguished from all other representations by the hiding of their faces. Again, the decision to do so was certainly not a legal or learned one, but the expression of a religious desire to emphasize the uniqueness of the person by concealing his physical presence. It was only then, in the sixteenth century, that such images of the Prophet visually separated him from the rest of mankind.

The sixteenth and seventeenth centuries were also the times when, especially

in India, mementos of holy figures, especially Shiite ones, increased drastically. Images of Muhammad existed among them, but neither in number nor in aesthetic quality did they resemble the images of Ali or his children. Yet something quite different transpired in the Sunni world, especially in the Ottoman empire. Relics of the Prophet were collected (and are still kept in the Istanbul museum located in the imperial palace), though nothing comparable in size or brilliance to Christian or Buddhist relics. There also developed a special genre of highly decorated sheets of paper, known as hilyes, which contained descriptions of the Prophet's features and qualities, and praises for his beauty and his nobility.

Some of these sheets of paper included, on the upper part of the page, a small portrait of the Prophet, usually shown as a distinguished man of middle years with a simple turban. These portraits were not the norm, but there does not seem to have been anything argued against them. Hilyes were hung on walls or kept in private treasuries as signs of piety and as ways to recall the unique physical form, personality, and character of the Prophet. It is probable that, in our own times, contemporary technology has been used to continue the manufacture of such souvenirs.

What, then, can one deduce from all these examples? For a start, there can be no doubt that, especially from the thirteenth century onward, the Muslim world accepted the existence of representations of the Prophet. This iconography was not common, and was usually restricted to the accompaniment of a narrative text, or to serve as pious reminders of an exemplary life. There seems to have been no judicial ruling approving or opposing them, but their relative paucity may reflect the long-standing cultural aniconism of a system of faith that was surrounded by religions and cultures in which images played a major role. It should be added that these images, and the references to them, are far more frequently Shiite than Sunni, and very few of them come from the Arab world.

Two details merit special attention, even though we do not quite understand the intellectual and psychological mechanisms that led to them. One is the striking development of ascension (mi'raj) images as works of art rather than as mere illustrations. It is perhaps not an accident that some scholars over a century ago saw in these images of the Prophet, and in the story on which they are based, an inspiration for Dante's Divine Comedy, although this suggestion is less popular in recent scholarship. The other detail is the eventual predominance of images in which the Prophet's face is veiled, thereby suggesting that images of people are acceptable but that the face of holiness must not be shown. This phenomenon is of considerable interest for any general theory of ways to represent the holy.

To the extent that the argument against the so-called cartoons has centered on the legal propriety or impropriety of representing the Prophet Muhammad, it has been a pointless argument. Of course it is possible to question the Danish caricatures on grounds of taste, or social or political intent; but the lack of taste is not a legal category, and mischievous or even evil intent is difficult to discern in the absence of clearly stated moral and philosophical principles. The only certain lesson to draw from the sad story of the Danish cartoons is the almost universal prevalence of ignorance and incompetence--and that everyone, from writers and pundits to the leaders of mobs, should learn more before making a judgment or starting a riot.

Oleg Grabar is professor emeritus at Harvard University and the Institute for Advanced Study in Princeton. He is the author most recently of Masterpieces of Islamic Art: The Decorated Page (Prestel).