4

Content versus Context in Samanid Epigraphic Pottery

Robert Hillenbrand

The epigraphic pottery found mainly, though not exclusively, in Afrasiyab (Samarqand) in Transoxiana and in Nishapur in eastern Iran is traditionally associated with the Samanid dynasty which ruled eastern Iran, and Transoxiana between 265/874 and 395/1005. It will be convenient to begin by sketching the broad historical context of these wares, and then moving to the manifold issues that they raise, before tackling the principal themes of this paper.

The enormous empire built up during the first century of the Arab conquests between 10/632 and 114/732 did not survive intact for more than a few years after the fall of the Umayyad dynasty in 132/750. The Abbasids who supplanted them lost, in quick succession, first Spain and then Morocco. Their control over the eastern periphery of their domains was also challenged, but in a subtler way. Thus in the course of the ninth century the eastern Islamic world was ruled by provincial governors who effectively enjoyed political independence of Baghdad but acknowledged their formal allegiance to the Abbasid caliph, honouring his name on the coinage and from the pulpit. The most powerful of these local dynasties was that of the Samanids. Their principal cities were Bukhara1 and the megalopolis2 of Nishapur.3 They claimed descent from the Sasanians, and during their heyday in the tenth century their court became a focus for what has been termed ‘the Persian Renaissance’. Here the first masterpieces in New Persian were written by such luminaries as Bal‘amī, the translator of the Ta’rīkh al-Rusul wa’l-Mulūk of al-Ṭabarī,4 and the poets Rūdakī and Daqīqī among many others.5 Indeed, Firdawsī himself began his career under Samanid patronage. The revival of pre-Islamic Persian traditions at this period – perhaps, among other things, a signal of anti-Arab feeling6 – takes several forms. In the visual arts it can be noted in architecture7 and coinage8 alike, but is perhaps most fully reflected in the figural ceramics of Nishapur, which perpetuate – admittedly in strikingly denatured form – hallowed Sasanian images, for example of hunting and feasting.9 These themes resonated far into the Islamic period. It might be argued that this choice of subject matter mirrored, in the realm of the visual arts, that independence of Baghdad and, more generally, of Arab culture, that can be detected in the literature of the period; Buyid hegemony in western Iran offers instructive parallels,10 not least in the visual arts.11 In the case of early medieval Central Asia, a Buddhist substratum, which expressed itself in literature12 as well as in architecture and numismatics,13 should not be forgotten.

Problems posed by Samanid epigraphic wares

This context confronts quite sharply the problems posed by the epigraphic pottery. Samanid epigraphic pottery – an umbrella term which has been subdivided on technical grounds into at least eight sub-groups14 – is widely regarded in the West as one of the high points of Islamic art as a whole. While the breath-taking beauty of these ceramics has long been acknowledged, often in repetitively rhapsodic terms; but scarcely one of them (Fig. 4.1) has attracted intensive analysis.15 And that is only one of the gaps in current knowledge. Troubling questions remain, and they come thick and fast. How do these wares relate to other uses of writing on Islamic pottery? Why were they made only in Khurasan and Transoxiana? Is their marked concentration in only two cities, cities 1100 km apart, an accident of archaeological discovery or a signal of cultural unity?16 Why did their production suddenly cease? What inspired these wares and what kinds of relevant models were available? In practical terms, how were they made? That is to say, by what technical means were the inscriptions themselves executed? And, more generally, how do these ceramics fit into the wider picture of the arts of Iran in the tenth and eleventh centuries, and of the culture which produced them? Why did they leave no significant legacy for later potters?

Another set of questions relates to what is written on them. The very idea of using pottery as a surface in this way for significant writing (as distinct from, say, bearing a signature) is worthy of note, for this was not a feature of Roman, Byzantine or Sasanian ceramics. But one need not be surprised at the link between proverbs or nuggets of wisdom – the most common content of these pieces – and the daily business of eating and drinking. The specious wisdom and empty rhetoric of the fortune cookie, or the trite message of the Christmas cracker, quickly come to mind. Both these types of messages are found in the context of a communal meal. Nor is it strange that the common themes of these Samanid inscriptions should move from eating and drinking to the related themes of hospitality and generosity – and, it must be said, greed – since the prime utilitarian purpose of these ceramics was to hold food.

Much less obvious, though – in fact, counter-intuitive – is the reason why Samanid epigraphic wares make exclusive use of Arabic in a society whose everyday language was Persian – Persian itself having no lack of proverbs of its own.17 The quintessentially Arab nature of its writing and content is patent.18 It is strongly impregnated with the ideals of pre-Islamic Arab society, notably their emphasis on generosity, hospitality, patience in adversity and the manly virtues in general19 – and pre-Islamic proverbs circulated widely, at least in the ninth century.20 How can all this be explained against a background of burgeoning ‘national’ sentiment21 and increasing independence from Arab rule? And while many of the intelligentsia operated easily in both Arabic and Persian,22 that is less likely to be true of the buyers of Samanid epigraphic wares, many of whom (as will be suggested below) came from a more modest social level. So were the citizens of Nishapur and Afrasiyab trying to turn the clock back? Were these wares made for diehard Arabophiles stubbornly resisting the winds of change? They present the paradox of high art in low dress, of a moral message in a modest medium. Could it be, indeed, that the medium is the message? That the proverbial, gnomic and aphoristic content of these inscriptions is almost literally grounded, like the humble earthenware of which these wares are made, in the life of the common man and of the homely wisdom by which he lives? That the cheapest material of all, the earth on which we live and move and have our being, can be transformed into a source of wisdom, ethical and spiritual guidance and moral uplift? This may sound far-fetched, but one should at least consider the possibility that an extended metaphor is at work here.

The status of these ceramics in that time and place poses another riddle. The high aesthetic quality of many (but by no means all) pieces suggests that they were made for elite patrons (Fig. 4.2), but the remarkably large numbers of sherds which have been found suggest otherwise, and indeed point to a high level of literacy, or aspirations to literacy, among the non-elite population. Even so, one is still left wondering who bought them, and why. That enquiry immediately generates another, namely what governed the choice of texts? Any attempt to understand the phenomenon of this Samanid production must consider the context of this pottery in other contemporary ceramics both from the same region and from further afield to both east and west. It should also note potential connections with higher-status media such as textiles, metalwork and the arts of the book. And it is important to remember that, however much these ceramics are prized today, they do not rate a mention in the medieval texts; al-Tha‘ālibī, for example, who indeed was a native of Nishapur, devotes three pages to the products of the city, ranging from rhubarb and edible earth to turquoises and textiles,23 but he has nothing to say about pottery.

One further general point needs to be made, for it is crucial to any detailed consideration of these wares. Of course many pieces have been restored to make them ‘complete’. At the very least, some are no doubt more repaired than others.24 Indeed, some that have gone on sale are, to put it bluntly, a lucky dip of several individually distinct shattered wares held together by modern material. That reflection imposes caution on any attempt to use a few isolated pieces as the basis for general remarks about this pottery. In short, the basic physical material on which all scholarship on Samanid epigraphic wares must rest is often badly flawed. Yet the larger the sample that is investigated, the weaker the impact of these deficiencies, and the sheer number of sherds (as distinct from complete pieces) with fine calligraphy suggests that the spectre of outright forgery – while it does occur in this field – should not be allowed to distort the assessment of Samanid epigraphic wares.25 So despite a degree of modern interference which lurks behind the current appearance of numerous ceramics of this genre, and which imposes caution in the analysis of many a given piece, that is no reason to avoid trying to highlight the intentions behind the choice of texts, to identify the various ways in which these ceramics were revolutionary, and to suggest what kinds of models might have inspired these ceramics. With that caveat constantly in mind, then, one may begin to examine them.

This is not the place to propose definitive answers to such a barrage of questions; that would require a book rather than an article. But at least such searching questions provide a framework for any close enquiry. And, for all the popularity of Samanid epigraphic pottery, they remain largely unanswered. This is because, despite their rare beauty, these ceramics have not attracted the amount of serious research that their quality would so amply justify,26 though Russian scholarship has by far the longest tradition here.27 The scholarship in Western languages has been slow to catch up.28 Moreover, scholars have devoted too little discussion to the content of these wares.29 Most references to them in the scholarly literature (except, of course, articles and books focusing specifically on the subject) and in auction catalogues recycle the same set of comments, mostly variations on Arthur Lane’s celebrated dictum: ‘Their beauty is of the highest intellectual order; they hold the essence of Islam undiluted’30 – a seriously erroneous judgment, incidentally, since it attributes to these wares a religious element which (with very few exceptions) their texts plainly do not possess.

Major themes for discussion

In what follows, it will be convenient to begin, after some brief remarks on chronology, by underlining the various ways in which this pottery is special. First and foremost, perhaps, comes its minimalist aesthetic, memorably but incorrectly termed ‘Islamic’ by Arthur Lane. A second theme is how these ceramics relate to other early Islamic epigraphy, and also how Samanid practice differed from the later uses of inscriptions on pottery. A natural subset of this theme is the sheer range of scripts employed, including pseudo-epigraphy,31 and the wider implications of this. A fourth topic concerns the various compositional devices employed; this subject will entail occasional detours into a discussion of the related wares which dispense with inscriptions altogether. The fifth section will focus on the ornamentation of letters in the wider context of Islamic epigraphy in other media. The next theme is the key issue of legibility, which includes an explanation for the use of prompts and addresses the language on these wares. It also highlights the silent battle between the desire to inform and the desire to decorate. There follows a brief and admittedly speculative discussion about how these ceramics might have been executed. Thereafter, in the rest of this paper, most of the space will be devoted to the eighth theme, and so to confronting the actual content of these inscriptions and the problems that it poses. A ninth and final section will consider the possible sources of inspiration for these wares.

Chronology

Close scrutiny of these wares dispels even the illusion of a chronology, and makes it plain that their attribution to the Samanid period,32 though tenable and indeed plausible, is as far as it is reasonable to go.33 There is as yet no question of developing a set of criteria that would allow this material to be sorted in some kind of chronological order, although there are pointers in that direction, for example the introduction, perhaps in the eleventh century, of outlining for letters.34 The meticulous work of Bol’shakov suggests that Central Asian ceramics in simple Kufic are frequently of ninth-century date.35 But the fundamental problem is that a reliable database for these ceramics simply does not exist. And a database is sorely needed, for these wares are found in their thousands, and that means that there is a sufficient quantity of data for patterns – of use, type, shape, content and epigraphic style, to mention only a few categories – to be discerned. But some issues are plain enough. Not a single piece of this pottery is dated, although impudent forgeries exist with dates in banner headlines.36 Also, the traditional association with the Samanids rests on the discovery of this kind of ware in the Samanid levels at Afrasiyab, that is Samarqand – the only location so far, apart from Nishapur, where they have been found in huge quantities.37 Indeed, excavations at Afrasiyab disclosed an entire potters’ quarter with at least fifteen households.38 Wares of this kind have also turned up elsewhere in the eastern Islamic world, in sites as widely scattered as Gurgān,39 Sirjān,40 Binkath (modern Tashkent),41 Marw, Ura-tyube,42 Paykand,43 Munchak-Tepe,44 Akhsīkath,45 Utrār46 and the province of Shāsh.47 The apparent absence of such wares in quantity from the principal Samanid capital, Bukhara, is noteworthy, though future excavations might alter this picture. In the past century, medieval Nishapur has been systematically wrecked by clandestine excavations, with the result that the official American excavations there were unable to produce anything approaching a reliable stratigraphy. Since then, hundreds of further pieces have come to light, but without an archaeological pedigree. These ceramics, then, are dated to the tenth century for convenience as well as by common consent. They could be a century or more later, perhaps even as late as the twelfth century therefore, or alternatively they could be as early as the ninth century. That prompts the reflection that some of them could have been produced under Qarakhanid rule,48 which extended over much of Transoxiana, and indeed the remarkably assured, elegant Kufic epigraphy of some Qarakhanid silver issues49 has obvious points of contact with the epigraphic wares. The fragments of Qarakhanid epigraphy in the frescoes found at Samarqand, although cursive, point in the same direction.50 Therefore, the comment by Spuler that these ceramics are better described as Turkistani rather than Samanid is worth taking seriously.51

A minimalist aesthetic

Samanid epigraphic wares must rank in the West as one of the most popular of Muslim ceramic types. The reason is not far to seek. Arthur Lane, for all his unfortunate use of words, was certainly on to something. At their best, these are the most uncompromisingly reductive wares that one could imagine to have come from a culture renowned for its emphasis on plentiful ornament. Their beauty resides principally in that reduction to essentials; less is more. The intense focus on the word, silhouetted against a neutral white ground, creates a powerful tension between solid and void. At times, that immediate impact owes little to the quality of the glaze on these wares, or their form, or their potting technique. Frequently that glaze is uneven and pitted, the form irregular and perhaps lumpy, the technique safe rather than daring, which means that at close quarters their heaviness belies the echoes of porcelain evoked at a distance by their white slip. But these wares were probably not intended to withstand such detailed scrutiny. If, on the contrary – as seems likely – they were meant to be viewed from a couple of metres away rather than a few centimetres, the defects of this surface would have disappeared as if by magic. At that distance the principal impact would have been the startling tonal contrast between the expanse of plain white- or cream-glazed body and the black or blackish-brown script. It seems quite likely that the intention of this contrast was to accentuate the whiteness of the body and thus to underline its affinity with Chinese porcelain, for centuries a symbol of excellence and prestige to Islamic potters. Thus at least some Samanid wares can be interpreted as precursors of the Seljuq stonepaste pottery in which the floriform treatment of the rim, the pierced body and the flowers or buds impressed on the surface all claim an association with Chinese porcelain, or with certain Song wares. There will be more to say about this later. One should also note that the very best pieces of Samanid epigraphic ware are remarkably large, originally possessed a glossy transparent glaze (usually denatured as a result of burial) and were very thinly potted. So the quality of these wares is distinctly uneven, ranging from the coarse to the superlative.

The mere notion of reduction is of course not enough to define the particular appeal of these wares. The real clue lies plainly in the arcane scripts which they employ. One might particularly stress ‘scripts’, in the plural, since this in turn is revealing. It shows that, quite apart from the specific dimension of ornament, mystery and difficulty were deliberately sought in these inscriptions. The proof lies in the bewildering variety of unusual, indeed highly mannered, hands. Many are, it seems, intentionally difficult to decipher, and often their idiosyncratic layout intensified that difficulty. These wares therefore appeal to anyone who is game for a challenge.52 Small wonder that radically different interpretations have been proposed for a given inscription.53 They announce that the content of their inscriptions is not everything. Their beauty is also important. But once the hard work of decipherment is done, one may regretfully conclude that to travel hopefully is perhaps better than to arrive. Like the messages tucked into the fortune cookies in Chinese restaurants, these aphorisms sound good to begin with, but sometimes, on reflection, one gradually suspects that the glib wording hides an essential banality and hollowness.

The same waywardness which characterises the rendition of the texts can be recognised in the way that some of the pieces are signed by the craftsman. Most striking of all, perhaps, is the utterly plain bowl in the Forughi collection whose only message is ‘made by al-Akhwal’, placed with modest ostentation on the lip of the bowl. On another bowl, however, the same craftsman cast discretion to the winds and signed himself squarely in the centre of the piece, again brooking no competition with any other decoration.54 Several pieces have the single word ‘Aḥmad’ at the centre (Fig. 4.3),55 and it is natural to assume that this is the name of the craftsman,56 though since this is usually unaccompanied by the standard formula ‘amal or its variants, it may not in fact be a signature at all.57 It could be interpreted to mean ‘most commendable’, and of course in that event it could be a pun with the additional meaning of Aḥmad as the name of the potter.58 Still other possibilities have been canvassed.59

That quality of understatement, that instinctive sense of interval which allowed these potters to make empty space so eloquent, was not confined to the epigraphic wares. It found expression also in wares with purely abstract designs, such as palmettes. And the way that these are conceived also reveals a close affinity with the epigraphic wares.60 Clearly the same principles generated both types, and yet others in this notably creative period.61 The preference for open bowls and dishes highlighted the interior rather than the exterior of these wares, and fitting an appropriate design into a circular format was a further challenge.62 The abstract ceramics make much play with radiating designs, with the constantly shifting interplay between the centre and the outer field. Occasionally they bear the mandala motifs for which a Buddhist connection has been proposed.63 Sometimes the centre is blank; sometimes it is marked by a small dot, with clusters of dots wheeling around it at intervals like planets in their orbits. Of course, the boundaries between the epigraphic and the non-epigraphic wares break down at times – for instance, when the birds that constitute the sole ornament themselves bear inscriptions or have a calligraphic character. Some have read certain versions of this motif as the word, or part of the word, ‘Allah’. Others have interpreted it as a bird.64 The interpretation depends on how the bowl is held. Sometimes an entire inscription takes on an avian character, in that the letters resemble a flock of swans.65 Or huge, richly textured palmettes, big with meaning, spread inwards from the rim of a dish (Fig. 4.4),66 perhaps in pairs to assert an axis, or in an equilateral triangle, or in a quartet to mark matching axes, such as the four cardinal points of the compass. This kind of antiphonal rhythm is also common in the placing of inscriptions (Fig. 4.5).67 In all such cases, symmetry is of the essence, but it is redeemed from any suspicion of being mechanical by the sense of free space which envelops the design. Indeed, empty space is an integral element of the layout. It is a matter for keen regret that by and large Islamic art turned away from the exploration of such felicities.

Early Islamic epigraphy

To assess accurately the role of Samanid epigraphic wares both in their immediate eastern Iranian context and, more generally, in the development of medieval Islamic art, it is vital to be aware of how revolutionary they were in their own time. Islamic art historians, schooled in the central role played by inscriptions in Islamic art in the later middle ages, are apt to overlook that, in the first two centuries of the Islamic era, inscriptions played a very subsidiary role indeed. It is true that the Dome of the Rock has lengthy inscription bands, but it is equally true that these are located high up in the building and that they are always difficult, in fact frequently impossible, to read with the naked eye. Moreover, they are quite remarkably repetitious, a feature which has not attracted the attention that it deserves. Most other surviving monuments of Umayyad and early Abbasid art are notably devoid of inscriptions. None of this is to deny that epigraphy played an honoured role in coins and – certainly from the ninth century onwards – in textiles. However, these are exceptions. One has only to look at the Islamic metalwork securely dated before, say, 1000 A.D. to realise that the notion of using epigraphy as a major element of the design rather than merely to inform had not yet caught on. As for making epigraphy the sole medium of ornament, that was simply unheard of at the time.

The same goes for the earliest Islamic glazed high-quality pottery, namely the various wares associated with Sāmarrā’. The earliest epigraphic wares are not Samanid but Umayyad and of course early Abbasid: on Umayyad oil lamps, on blue and white wares,68 and on relief ceramics.69 None have the elegance of Samanid wares, but they certainly habituated potters to the idea of writing on pottery, and then, by degrees, to writing better on pottery. And while the wealth of early Qurʾan manuscripts (maṣāḥif) indicates how strong the fascination with writing was in these early centuries, they demonstrate calligraphy, not epigraphy – though to term them epigraphy on vellum would not be entirely inappropriate. Thus, when seen from the vantage point of later Islamic art, these Samanid wares can be regarded as marking a watershed. They define a new role for Arabic inscriptions in early Islamic times, and they do so in a Persian-speaking context.

The main argument in this paper is that this breakthrough operates principally in the specific role of writing as ornament, but it actually goes for the use of writing per se as well. And the ramifications of Samanid epigraphic pottery extend further than this. These wares establish a new balance between decoration and emptiness, somehow making that emptiness active. And perhaps most of all, they document a new popular role for fine pottery, emancipating it from the taste of the court. With them, beautiful Arabic writing in a popular context comes of age.

It is precisely in the manifold later uses of epigraphy in Muslim ceramics that the originality, the discipline, the sheer sure-footedness of the Samanid calligraphers is highlighted. Not for them the hasty cursive scribble so frequently found on Seljuq lustreware. Not for them, either, the massed uprights of Mamluk honorific inscriptions, standing stiffly to attention like a guard of honour for the dignitary whom they extol. In the Seljuq pieces the writing is treated visually as little more than an afterthought; in the Mamluk ones, by contrast, the writing elbows aside everything else, creating a claustrophobically crowded field – full of power but also full of tension. Between these particular polarities there are naturally countless other ways of using inscriptions – bold headlines, somewhat coarsely carved, marching across the centre of a jug;70 deliberately etiolated or squat letters disposed with mechanical regularity around the rim of a plate, expressing repetitive litanies of good wishes as in twelfth-century Almohad platter from Murcia inscribed ‘peace’, al-salāma (Fig. 4.5);71 planned contrasts of cursive and Kufic hands; cartouches placed at intervals across the surface; and thin panels arranged concentrically, vertically, radially, in tiered or spiralling fashion and even in zigzags. All this affords eloquent testimony of how enthusiastically later potters took up the basic idea embodied in these Samanid wares and developed it in all kinds of directions. It is nonetheless significant that they virtually never recaptured the innate stylishness of Samanid wares. And the reason is simple: they were not content to leave well alone. Only rarely did later potters confine the decoration of their wares to epigraphy, and even more rarely were they ready to permit large tracts of empty space in the design. Indeed, by the twelfth century, the interests of potters had moved away from fine writing; they preferred to develop figural, vegetal and geometric themes. Finally, too, it has to be admitted that most types of cursive hands lacked the formidable presence of the best Samanid Kufic, while equally some of the Kufic found on later pottery is coarsened, simplified, even monotonous in comparison with its Samanid counterparts. By the twelfth century, too, pottery had definitively ceased to be in the forefront of experiment in Kufic hands. Qurʾans and architecture dominated the epigraphic field.

The range of scripts on Samanid pottery

These ceramics testify to a sudden and intense fashion for epigraphic pottery that was clearly served by a number of workshops – as proved by the masses of sherds found so abundantly in Afrasiyab and Nishapur especially. It embraced any number of variations from plain or simple Kufic (as defined by Bolshakov),72 (Fig. 4.25) to what he terms ceramic cursive (Fig. 4.26),73 a style close to the broken cursive of tenth- to eleventh-century Qurʾans (figs 4.27–4.28). Like many such fashions, it fed off itself, creating its own momentum and taking off in all manner of directions. Many of these pieces have a distinctly experimental feel, almost as if they were exercises or doodles by acknowledged masters who were trying to see how far they could go in making writing beautiful but almost impenetrable. It is likely enough that, as in other crafts, the makers worked cheek-by-jowl with each other and were thus well placed to keep a sharp eye on the innovations devised by their colleagues and rivals. The material that provided the surface for the writing was as cheap as it could possibly be – fired earthenware – and thus encouraged experiment. Broken pots and wasters could have been used by calligraphers for doodling, practising, following through an idea, and simply inventing. After all, paper – though much less costly than parchment – was by no means cheap. And the maturity of the script used in many of these pieces argues long practice.

With this school of ceramic epigraphy, fine writing decisively emancipated itself from the constraints of a book hand, constraints that were particularly marked in the field of Qurʾanic calligraphy which accounts for the overwhelming majority of so-called Kufic hands in books. That dominance of Qurʾanic scripts lasted some three hundred years, from the later seventh to the later tenth centuries, and in that time it consistently imposed decorum as well as discipline on calligraphers.

That situation changed when they could put their skill to secular uses. One might argue that some top calligraphers, habituated to certain writing practices by a lifetime of copying sacred, chancery and literary or scientific texts, would be inhibited from experiment of the kind shown in the pottery, or might indeed disdain it. Others might embrace such an opportunity. And yet others, unable to make a living from such formal calligraphy alone, would be versatile enough to turn their skills to account with other commissions. One must also reckon with an environment where – as the quantity and variety of the sherd evidence suggests – there was a fever of experiment and of competition.

At long last, then, fine writing for the sake of fine writing, unconnected with the religious impulse, could strike out on its own, and plainly calligraphers revelled in their new-found freedom. Pots were on just the right scale for such experiment, unlike coins on the one hand – where small size curbed the imagination – and unlike architecture on the other, where the correspondingly large scale called for greater formality. And in both textiles and metalwork, where the scale was closer to that of ceramics, the difficulty of working the material itself discouraged the free flow of the calligrapher’s imagination. Moreover, the material of which ceramics were made was cheap and abundant. None of this is to deny that these inscriptions were often impressive, stately, formal. Their form alone contrived to suggest that their content too was significant. Thus the nuggets of wisdom dispensed in these striking calligraphic compositions were a kind of objective correlative to the stylish writing. Occasionally, one must admit, form trumps content, even to a faintly absurd degree, as if the message in a Christmas cracker were read in a dramatic style more suited to Shakespearean blank verse. This is bombast masquerading as philosophy.

Sometimes the inscription looks as if it were written with an instrument held at an angle guaranteed to stress thinness rather than thickness, so that occasional recourse to a thick stroke stands out with greater force. Other wares renounce such delicacy in favour of repeated thick strokes that evoke huge headlines, with the individual letters sometimes almost two centimetres thick. Such inscriptions (Fig. 4.6–7) were clearly executed using a thick brush,74 and the traces of its individual fibres can be detected at close quarters. But the variations in execution are so wide as to forbid any attempt to judge the main body of these wares by a single aesthetic standard. Most, in point of fact, are rather coarsely written and, as will be seen later, could very well have been executed by stencil.

There are similar polarities between a ductus that rigorously eschews ornament and one that is overloaded with it in the form of knotting (Fig. 4.29),75 arabesques or floral flourishes. Many inscriptions exploit the contrasts of filled and empty space, solids and voids, by dint of cramming certain letters close together and conversely stretching them far apart. Thus the calligrapher creates a variety of rhythms: measured, stately, staccato, syncopated and so on. In just the same way there is perhaps something musical in the contrast between extreme extension and extreme compression. Links with music can be proposed for other media of Islamic art, for example architecture.76 And a common device for creating a continuous rhythm is for the extended flourish of a terminal rā’, nūn or wāw to break into the following word (Fig. 4.3).77

Sometimes repeated words or phrases echo each other across the largely empty space of the interior of a bowl or dish, and that symmetrical repetition points a moral.78 Pseudo-epigraphy allows the artist to create continuous unbroken patterns broadly based on letter forms (Fig. 4.4),79 but it also creates split words, distortions of various kinds and, on occasion, may even have symbolic meaning.80 The astonishing, unprecedented variety of all these experiments becomes most apparent when the achievement of Samanid calligraphers is seen as a whole, and that is when one can recognise everywhere the impact of powerful individuality, playfulness, humour and sheer joie de vivre expressed through writing. It is in this humble medium, even more than in the far more austere and formal world of Qurʾanic calligraphy, that the limitless potential of ‘Kufic’ hands makes itself felt. It almost follows that many of these hands are found only on these wares. They are one-off. And their impact can be felt even in other media and in later times. Thus it is surely no accident that the Qarakhanid coins mentioned above, minted alongside and in the immediate aftermath of Samanid power, have some of the most elegant calligraphy in all of Muslim numismatics, in which the severely reduced scale of the coins scarcely challenged the virtuosity of the die-engraver.

Countless other irregularities are inflicted on the Arabic alphabet in these wares. Thus letters which ought to extend below the baseline are wrenched out of true so as to stay on it, while letters meant to extend above the line can flourish freely. Often the purpose of these elaborate calligraphic games is to set up rhythms at variance with that of the baseline. Indeed, it is the core text, so to speak, rather than the flourishes applied to it, that can best exploit blank space and inject rhythm into the design. These calligraphers knew not only how to fit a pint (and sometimes as little as a gill) into a quart pot, but could also manage the reverse with remarkably little strain. And of course they could operate with ease anywhere along the sliding scale between these two poles. In the spoken word the equivalent would be to allow long pauses to develop at intervals, or conversely to gabble, and throughout to place the stress wherever the speaker wishes. Thus the placing of words is closely akin to punctuation. Blank space serves to ‘underline’ words and to heighten their significance, an especially appropriate device for the gnomic proverbs which are the staple content of these ceramics.81 A visual antithesis thus mirrors one of content, for example in inscriptions which form a straight line right across the plate (Fig. 4.30).82 The text ‘He who talks too much errs too much’ divides neatly into one phrase or word for each cardinal point of the compass (Fig. 4.31).83 Here the extreme parsimony which marks the placing of the words accurately reflects the sense of the text. Or the key word may appear in the middle of the dish, with the rest of the text sandwiched above and below it (Fig. 4.7).84 A really short text – al-ḥurru barrun, ‘the free man is charitable’- lends itself to being repeated twice,85 and to even out the spaces the scribe has run the phrases into each other so that each group consists of three letters, in defiance of sense, not of two or four letters respectively (Fig. 4.32).86 Once again the effect is antiphonal – a matter of sound as well as spacing. A strikingly similar result can be recognised in certain Hispano-Moresque wares made in Manises in the fifteenth century which bear the repetitive legend Ave Maria in Gothic minims (Fig. 4.8).87 They serve as a reminder of how much more limited than Arabic that European script was. It is a matter for future research to determine the extent to which proverbs were chosen for the visual puns that they could suggest – for example, in Arabic words like ‘rest’, ‘relaxation’, ‘comfort’ and ‘patience’ (rāḥa, istirāḥa , ṣabr) can be given an extra length that mimics what they actually mean.88 Other proverbs contain assonances that can find expression in visually similar letters. The phenomenon of rhyming titles in Islamic books comes to mind in this connection.89

Compositional devices

The sense of absolute rightness in the best Samanid epigraphic pottery is of course not achieved by accident. Certain compositional devices were used to this end. Thus a single dot at dead centre served as a magnet for the satellite letters orbiting around it. In this way the design is given a focus. This centring dot was sometimes replaced by a rosette or a yin/yang combination, or a knotted cartouche, or even a bird. But the commonest form taken by this motif was a simple dot. There is no denying the change it makes to the design, though to analyse why this should be so involves trespassing into the domain of the psychologist. The dot does more than merely articulate space. It serves as a kind of magic eye. A more practical function for such dots will be discussed in detail below in the section on technique.

These inscriptions are notoriously some of the most difficult to decipher in the whole canon of Islamic art, and there is evidence aplenty that they were not easily read in their own time either. Often the calligrapher considerately provides a guide for the baffled would-be reader: a single inconspicuous motif,90 sometimes one that can easily be mistaken for a letter form but is not. It may be a red alif (here it is colour that draws attention to the clue), a dot, a palmette, a vertical attenuated oval, or conversely a fat one, or a horizontal one enclosing two dots,91 or a triangle whose sloping sides are lobed, a pellet, a pair of dots or a simple horizontal bar. Such motifs may also take the form of a bell or a pair of book-ends.92 Bol’shakov gives a selection of them (Fig. 4.34).93 Whatever their form, they mark a distinct break and thus serves to signpost, in a suitably subtle way, the beginning or the end of the inscription. Such prompts, then, are an aid to comprehension. Their frequency is a tacit acknowledgment that these inscriptions are hard to read. Yet all this also suggests that the inscriptions were indeed meant to be read, even if the literacy of the intended viewer might be somewhat shaky.

In many cases, the location of an inscription is an intrinsic element in its impact. Some inscriptions create a long horizontal band in the middle of the dish, bisecting it neatly. Or they cling close to the rim of the dish or bowl that they decorate, taking an area a few millimetres inside that rim as a notional baseline. This emphasises the circular shape of the piece. Normally this leaves a ragged edge facing the centre, which of course allows the letters free rein to develop. But the ascenders may be disciplined, indeed squashed, paradoxically by means of applied ornament, into a regular continuous line, so that their tops form an inner circle (Fig. 4.9). Circularity can be stressed still more insistently, and in more ways than one. Two concentric inscriptions may do the job (Fig. 4.10). On other occasions the potter uses a baseline about halfway down the surface of the bowl, with the shafts of the letters pointing to the rim rather than to the centre of the bowl. Thus the ascenders have to conform to its rim, with the baseline (being even) creating an inner circle. In this way both the upper and the lower confines of the inscription are strictly regulated (Fig. 4.11).94 This variation may have been inspired by coins.95

Sometimes the letters seem to glory in their slender proportions, with exaggeratedly long shafts96 ending in razor-sharp serifs or foliations. Alternatively the letters may stretch in notably long horizontal bars or sweeping curves hugging the baseline (Fig. 4.12). A third option – among many more – is for an inscription to be limited to a height of no more than a couple of centimetres despite all its applied ornament, as distinct from the aspirational quality of an inscription distinguished by its shooting vertical shafts whose length creates a radial pattern that is the dominant feature of a bowl (Figs 4.2 and 4.13). The pieces which operate horizontally may be a conscious exercise in virtuosity, an attempt to show how much variety can be crammed into a small low space; they are full of contained energy. Yet the actual content of the inscription has nothing to do with these visual exercises, some of which seem to be pruning devices.

The ornamentation of the letters

This theme has as many ramifications as the preceding one. Here, too, Samanid epigraphic wares are in the vanguard not just in Iran or even the eastern Islamic world, but in the lands of Islam generally. Most inscriptions of the first three centuries of Islam are found either on buildings or on tombstones, and the unyielding surfaces used for such epigraphy inhibited the free flow of fantasy in adding ornamental flourishes to the component letters of a text. As already noted, regarding Qur’anic manuscripts, convention severely restricted the amount of added ornament that was permissible for the most holy of texts. The principal forerunners of Samanid wares in this respect are Abbasid blue and white ceramics, which deploy a range of decorative devices for the letters of their inscriptions. These favour triangular or blob-like serifs and terminations, mashq , jagged spikes, extreme contrasts of thick and thin letters and S-shaped beginnings to the letter ḥā’.97 But the execution of these features looks rough rather than delicate, the result perhaps of using a thick and loaded brush. So in this respect, as in the devising of new variations of letter forms, the calligraphers working on Samanid pottery effectively had a blank canvas before them. They were the first to exploit intensively the rich potential of individual letters of the Arabic alphabet as vehicles for extra and indeed redundant ornament. In so doing they also opened up new horizons for the internal transformation of letters.98 Thus their range of options was wide. Knotting or plaiting was a favoured device along both the horizontal and vertical planes, and sometimes it greatly extended the length of certain letters, a device recorded in tenth-century numismatic inscriptions from this part of the world.99 It slows down the process of decipherment like a road-block, but also lays down a challenge to the reader and has the extra advantage of diversifying the otherwise over-extended stretches of some letters. A dense knot or a bulbous U-shaped protrusion might mark the centre of a vertical shaft. Rosettes proliferated within, say, ‘ayn, ghayn and mīm. Terminal wāw, nūn and rā’ acquired the graceful curves of a swan’s neck. Medial mīm or ‘ayn/ghayn might take the form of a helmet. The terminus of many a letter ends in a bud or a blossom. Small vertical channels are cut into horizontal letters as if to impede their flow. Some terminations bifurcate like leaves while others have shredded ends like fronds.

Quite aside from such tinkering with the forms of the individual letters themselves, the impulse to decorate takes several other forms in these ceramics. On many occasions, too, the calligrapher executes part of a letter in red, and this lends an extra chromatic dimension to the inscription as a whole, quite aside from imparting an extra rhythm to it by ensuring the regular beat of a flash of bright colour at set intervals (Fig. 4.9). Or words in red balance the words in black in a contrast between the inner and outer ring of an inscription.100

The decorative flourishes which embellish so many of these inscriptions deserve a word or two to themselves. This is a subject full of surprises. This ware is not the first to demonstrate the desire of the Muslim potter to decorate his inscription; Abbasid blue and white wares display, besides a variety of layouts for the inscription proper,101 a whole range of devices, even though the range of the messages is very limited, confined as it is almost exclusively to good wishes and signatures.102 But in Samanid wares, perhaps for the first time in Islamic epigraphy, redundant single or paired shafts are added to an inscription for purely aesthetic purposes, such as asserting rhythm or syncopation. Sometimes, in the contrast between a central field with the letters disposed horizontally and an outer circular inscription whose base is well within the rim, an echo of the standard numismatic layout can be sensed.103 In some inscriptions, certain types of ornament are reserved for specific letters, yet this usage is not consistent, so that sometimes the letter in question is plain while on its next occurrence it is embellished. It is plain that these technical inconsistencies are dictated by the overall design, which demands certain types of ornament at given locations.

It is quite common for a single inscription to be arranged in two-tier fashion so that the letters that convey information are on the lower tier, keeping close to the baseline, while often enough the upper tier exhibits a baroque exuberance, with complex compositions neatly divided from each other by the straight vertical shafts of the definite article, alif and lam. That upper tier, as already noted, may be disciplined into an even inner circle which mimics the broader circle of the baseline. Above all, the curvilinear quality of many inscriptions is manifest as long serpentine extensions (not required by sense) double back on themselves, sprouting leaves as they go. A terminal curve may end in a spreading bifurcated palmette which can form a kind of umbrella over an entire word, and many a letter can sprout subsidiary ornament in this way (Fig. 4.14). In many plates a substantial role is allotted to arabesque ornament, creating in the process a rhythmic beat – so much so that on occasion it threatens to squeeze the inscription off the plate. Indeed, the most pervasive change of all is the way that the free-flowing arabesque infiltrates much of the alphabet, frequently changing direction and creating continuous scrolling vegetal forms. Thus the calligrapher creates a dense thicket whose wilful complexity seems to imprison the letters in its toils. Many of these felicities were repeated by the designers of later inscriptions in more forbidding materials such as stone and terra cotta, wood or woven material, but it was the Samanid calligraphers who had blazed a trail. This emphasis on the transformative power of the arabesque is a reminder that plaiting is only one aspect of the decorative repertoire.

Legibility and language

The difficulty of reading many of these inscriptions, especially those executed with the utmost care, has long been generally recognised. A whole series of explanations could be adduced for this. It seems likely enough that some calligraphers delighted to showcase their skills, and a deliberate complexity was one way to do so. Another possibility is that the intention was to make the reader work hard so as to concentrate all the better on the meaning of the text, and to contemplate its significance. Sometimes, indeed, the text is almost self-referential in this respect, as when one reads that knowledge is bitter to the taste at first, but in the end is sweeter than honey.104 In that respect, the difficulty of the script could be interpreted as a direct challenge to the reader’s intelligence and level of literacy. In certain especially familiar groups of letters, notably baraka and li-ṣāḥibihi105 – perhaps the commonest inscription on these wares (Fig. 4.15), sometimes with an extra word such as ni‘ma (Fig. 4.6) – an element of abbreviation can be detected, with the letters sometimes suggesting these words rather than explicitly rendering them. Thus a symbolic element enters the equation;106 the scribe gives himself licence to play, knowing that the message will still get across. And of course, as already noted, there might be a deliberate connection between the elaborate script and the profound wisdom of the saying which it renders. There may even be an appeal to the pride and snobbery of those readers who could make sense of these riddling inscriptions.

There are many different ways in which these inscriptions are hard to read. Lisa Volov’s classic article plots the internally consistent evolution of many of the devices which these potters employed.107 A common pitfall that awaits the unwary reader is the playful use of mashq or extension, whereby letters are wrenched out of true and extended to the very limits of legibility. This extreme stretching may be so placed as to confound expectation, and there ensues a pause while the reader tries to assimilate what this unfamiliar concatenation of shapes might mean. Legible handwriting, after all, depends on the reasonable assumption that a more or less even space will be maintained between letters and also between words. When that law is flouted, chaos threatens.

Stretching is, of course, not the only way to disguise the normal form of a letter. Looping the upper part of a letter backwards, and even making it produce multiple loops, is another way of disguising its normal form. Or the whole inscription may be so flattened that it barely rises above the baseline, in which case almost every letter is hard to recognise. Or a horizontal stroke may change its thickness quite markedly over a couple of centimetres. It is common practice for an entire inscription to exploit the contrast between very thick and very thin strokes, and this can disorientate the reader. Letters may nest together like spoons (Fig. 4.16),108 or alternatively they may overlap. A common ploy is to attach a long line to the right of the opening letter of a word. This is not just redundant but positively misleading. A more elegant version of the same idea is the rightward sweeping curve of an isolated alif. This serves as a bridge between two words that would otherwise be set very markedly apart from each other, as sense requires. Or the elongated lower stroke of a rectilinear terminal dāl, nūn or rāʾ may perform the same connecting function, interlocking two words. A plate in the British Museum repeats this device four times in quick succession.109 Another device is to use different forms for the same letter in a single inscription. Or both the shafts of vertical letters and the bars of horizontal letters may suddenly generate sudden compressed S-bends, re-entrants, doglegs, bumps, hooks, dents, notches, nicks or bulges, which slow down the process of recognising the sequence of letters and therefore understanding what the text says. And sometimes the curvaceous sweep of an inscription judders to a halt as it is brought up short by an obstinately rectilinear letter such as rā’ or dāl. However, there is no doubt that the principal device of this kind is knotting, which can be single, double, or multiple; indeed, disparate letters can even be knotted together. Lisa Volov noted that contemporary coins from Rayy also displayed this feature,110 although it is interesting to note that this fashion is apparently not found in the Samanid-period coins of Nishapur itself.111 The eye is constantly enticed astray by the way that the upper parts of letters break into extravagant vegetal flourishes, foliations, blossoms or scrolls. In all these devices one may recognise easily enough the urge to decorate; but they also serve to confuse the reader. Hence the need for the prompts discussed earlier: the calligrapher is showing a modicum of mercy to the reader.

Finally, then, the language of these inscriptions. Arabic to the citizens of Nishapur and Afrasiyab was a foreign tongue, although of course there was a longestablished Arab minority in Khurasan. No doubt members of that minority were among the customers for the epigraphic wares, but most of the buyers would not have known Arabic, just as most medieval Europeans would not have known Latin. The parallel is just, in that, like Latin in medieval Europe, Arabic was not only a foreign language in the medieval Iranian world but also the language of culture and, of course, the language of religion. It therefore enjoyed a quite special cachet. To display pottery with maxims in that tongue was therefore to display one’s own knowledge and culture, even if they had somewhat insecure foundations. And when those maxims were written in a ductus that instantly called to mind contemporary luxury Qurʾans, and whose colour scheme of brown and cream resembled that of these manuscripts, the subliminal associations of these texts were with holy writ. So to have such ceramics in one’s home where visitors could see them was to lay silent claim to both culture and piety.

Yet without exception these Samanid wares, produced in an environment where Persian was the normal medium of communication, ostentatiously dissociate themselves from that everyday language. They are all in Arabic. And that phenomenon deserves closer scrutiny. Although Nishapur, as one of the major cities of the Samanid realm, no doubt had an Arab element in its population, it was located a thousand miles away from the linguistic frontier between Arabic and Persian. The epigraphic wares are far too numerous to be explained as being made specifically for such an Arab minority. The same argument excludes the suggestion that the patrons were that small minority of people who used both languages regularly.112 And while Arabs might have constituted a significant element in the city in the eighth century, its rapid growth under the Samanids, which seemingly led to its attaining a population of perhaps two hundred thousand, would surely not have involved a major expansion of this Arab element. Moreover, Persian had enough proverbs of its own and had no need to plunder the resources of Arabic for this purpose. So why was Arabic chosen? Indeed, not just Arabic, but Arabic showcased in a spectacularly difficult and decorative script?

Technique

The makings of a solution to this problem may lie, paradoxically enough, in that same austerity that has aroused such admiration in the West. Seen from the practical angle of employing time-intensive labour, this austerity means that the only decorated surface on many of these ceramics is the underglaze writing itself, set off in spectacular fashion by the colourless lead glaze coating it.113 If that writing could somehow be done expeditiously with no loss of quality, strikingly handsome wares would be produced at remarkable speed. And they would be relatively cheap. The implications of that finding will be explored in the next section, which deals with patronage. For the moment, it will be convenient to examine the technical side of that speed and cheapness. Broadly speaking, two possibilities present themselves: were these inscriptions produced largely or entirely by mechanical means, or were they executed freehand?

The first of these possibilities has been proposed by ‘Abdallah Ghouchani, who has probably devoted more time to this pottery than anybody else apart from Bol’shakov. He is of the opinion that the inscriptions were executed not free-hand but by ‘a flexible mould, probably of leather’ – in other words, a stencil. This seems a reasonable supposition, especially if the leather in question were soft, like chamois leather. Much follows from this. It helps to explain an otherwise puzzling aspect of these inscriptions, namely how an accomplished professional calligrapher, accustomed to writing on treated parchment or sized paper and thus on a smooth, flat surface, could replicate his normal high standard of execution on these ceramics. After all, their frequently gritty texture would tend to impede the even flow of the writing implement, especially if it were a reed pen. Similarly, the steeply plunging sides or convex surfaces of these wares would have necessitated constant and abrupt changes of angle in the process of writing. So even if a calligrapher repeated a text so many times that he could, as it were, write it in his sleep, these practical obstacles would remain.

A parchment original, cut to serve as a stencil, would sidestep these disadvantages, and would accommodate any angle of ceramic surface. Such a master copy, moreover, could continue to be used almost indefinitely. And the use of a stencil meant that a professional calligrapher could be employed by the potter,114 and suitably paid, without the need to pass on to the customer the full cost of his much sought-after expertise. Otherwise it would be difficult to reconcile the mass output of these wares with the participation of probably the most highly paid craftsmen in contemporary society. Given that most of these bowls or dishes probably took less than four minutes to make, and were thereafter rapidly covered in slip, to be decorated eventually in this mechanical fashion, the huge output of these wares – as witnessed by the thousands of known sherds – is readily explained. So these ceramics could not have been very expensive. Probably they were cheaper than a sheet of calligraphy on paper. Stencils would not be economic unless they did indeed save a lot of time, so it seems unlikely that they were time-consuming to use, with the colour being applied by brush or even a finger. If there was a slight fuzziness, as one might expect with a stencil, that could be tidied up right away with a knife or some blotting agent.115 That same knife could be used to make tiny scratches in the letters to suggest plaiting or other internal decoration.

Moreover, more than one kind of stencil is imaginable – at one extreme, a complete inscription; at the other extreme, a stencil that ran the full circumference of a bowl or dish but was limited to a few key markings. Such a simplified stencil would be ideal for subsequent manual titivation and embellishment. That way the broad outline of the epigraphic design would be certain to work; its main components would all be in the right place and of the correct size. Given that many inscriptions survive on multiple wares, it seems likely enough that those who wrote them would quickly have become proficient at executing them at speed, thanks to a few fixed points, quickly put in place on the ceramic by a stencil or compass, which would ensure that no part of the inscription would over-run or under-run.

This is where the central dot found on so many of these pieces makes good sense. Achaemenid metalworkers used a centring dot and a compass to lay out the decoration evenly.116 Some similar process, using peg, string and charcoal, probably operated in the production of these Samanid wares – for example, charcoal marks to draw the basic guidelines within which the inscription was to fit. Hence the frequent presence of a central dot. Heat would destroy such guidelines of charcoal and perhaps also the tell-tale traces of fibre in a brush.

What, then, of the second possibility, that these inscriptions were executed freehand? In some cases one might assume that a stencil was not used, for two reasons, one specific and one more general: first, the occasional mistakes in design cannot easily be reconciled with the use of a stencil, since the text would have been calculated precisely well in advance. So any piece with a gap, or with a filler word, should be considered at least as a possibility to have been executed freehand. The second, and more general, reason is that certain pieces – the very finest – are too free and assured to be readily explained by a stencil. The more precise and tinyscale knottings also point in that direction.

However, other arguments in favour of freehand execution could also be deployed. Thus the use of a thick slip would make the surface much smoother and therefore more suitable as a surface on which to write. Moreover, a craftsman seated cross-legged on the ground would be able to rotate the ceramic bit by bit, supported by his body, as he wrote; no assistant would be necessary to move it for him. Craftsmen typically sat on the floor to work. Nor should one underestimate the sheer awesome technical ability that a talented craftsman might develop over a long period of very specialised work. So to argue that the pieces with plunging sides were simply too awkward to bear the finest calligraphy written freehand is perhaps mistaken.

A related issue is the type of instrument employed for these inscriptions. Many of the so-called ‘pseudo-inscriptions’ (Fig. 4.4), for example, seem to have been done freehand with a brush and therefore would have taken longer, for all that their quality was less high. And one might propose still other scenarios. For these pieces are so very varied that plainly one explanation does not fit all of them. Some may have used a stylus or reed pen, others a brush. The finest knotting would be simply impossible with a reed pen, since the manual dexterity required would be too great and the lightest of touches would be required. Such knotting calls for a very fine brush. There is also frequently a sense of last-minute improvisation. It is also doubtful whether a reed pen could have held enough slip. Instead, it is perhaps more likely that something fibrous, something more suited to holding fluid, was used, perhaps something like a toothbrush. Paintbrushes could be of the most varied kind, from extreme thickness to extreme delicacy. The artists often worked with a fully loaded instrument, since so many pieces have the letters standing proud of the surface and this is instantly plain to the touch. Had that instrument been a pen, it would have needed frequent refills, more than a brush would, and the strokes would have had to be much more interrupted. The use of some kind of fibre or brush – a felt-tip pen is perhaps an appropriate modern equivalent – would allow the calligrapher to overcome the problem of a gritty surface. Such a surface would present a greater problem if he used a reed pen. The fact that the slip has congealed in little drops or pools, like impasto paint, is perhaps a further argument in favour of free-hand application. And still other possibilities present themselves. An unpublished jug in the Sarikhani collection (Fig. 4.17), for example, has an inscription read by Will Kwiatkowski as ‘the death of the (chivalrous) youth is preferable to the youth than miserliness, and miserliness is preferable to a miserly request’. It is of such clarity and purity of execution that one might consider another method of manufacture, namely outlining the letter forms with a stylus or qalam and filling them in with a brush. This jug would have been a difficult form to write on, with its marked convex bulge.

Perhaps the strongest argument in favour of freehand writing is the ability – scarcely to be underestimated – of a craftsman to achieve the hand-eye coordination necessary to repeat a given text ad infinitum until he could almost repeat it with his eyes shut. This would not exclude minor variations en route. Hence even a piece with a dense repeated group of letters which fit together in virtuoso fashion (Fig. 4.16)117 would present no difficulties after multiple repetitions. Even a schoolboy writing lines can manage that. And the finest craftsmen would develop a sixth sense of how to fit their inscriptions into the space available,118 allowing themselves a fair amount of discretion along the way. And the oftener they wrote a given text, the more confident they would become in creating visual riffs on it. These inscriptions often have a freedom, a liveliness, a fantasy that does not sit well with the discipline of pen-made writing and its emphasis on slow, steady execution. They have more in common with the individuality, the sense of personal expression, of Chinese calligraphy, and also with a sense of art for art’s sake. All this points to the use of a brush for most of the best pieces. Even so, a few of the grandest – and, one may note, the simplest and the least amenable to redundant ornament – may deliberately have used the pen in an attempt to lend the piece the look of a book.

Plainly, then, it is too crude a polarity to propose either an unmodulated stencil option or an equally unmodulated freehand one. It seems almost certain that there was an extensive middle ground in which mechanical means such as simplified stencils or charcoal markings were used to greater or lesser extent. Similarly, it would be premature to propose the exclusive use of either a brush or a qalam; it is likely that both were used.

Patronage

The preceding discussion suggests that many, perhaps most, of these pieces could not have been very expensive, since often enough the element of an artist’s individual contribution had been virtually removed. In this respect one may point to the contrast with the Nishapuri figural wares with their strong neo-Sasanian elements. Despite their somewhat coarse execution, many of these Nishapuri dishes, with their over-populated fields, represented hours of work for the painter, which in turn suggests that they were significantly more expensive than most of the epigraphic wares. While the spectacular calligraphic display pieces universally hailed as masterpieces were plainly made to be used on special occasions by well-to-do patrons, the majority of epigraphic wares are of a much lower quality. Bol’shakov noted that extended benedictions were found only on good-quality wares and that the use of baraka on its own was often associated with coarse, cheap wares, with a concomitant difference in the social status of the purchaser.119 Sometimes it seems that an inscription written by a calligrapher of indifferent quality is itself rendered still more defective because it is copied by an unskilled potter,120 perhaps himself illiterate.121 Could one then propose that it was the figural wares that expressed the aspirations of the Persian renaissance in Samanid times?122 Perhaps one could make the further deduction, on this particular body of evidence, that this renaissance was the concern of a more moneyed class than that represented by the epigraphic pottery.

In this perspective, the typical buyer of most epigraphic pottery would most likely have been someone of relatively modest social status rather than a rich merchant or official.123 Many such patrons would have been, in modern jargon, upwardly mobile. A few of them are identified by name or title: Abū Ja‘far b. ‘Arīb, whose name appears on a bowl of first-class quality (Fig. 4.18);124 Muḥammad Faḍl;125 and an anonymous amir.126 There is even a piece inscribed repetitively ‘to you’ (laka)127 – clearly a gift. Many pieces have repetitions of a single word like al-yumn (‘prosperity’; or ‘joy’, ‘good luck’),128 and here again one may assume that the purchasers were not of high social status.

To return to the choice of language, namely Arabic, the key factors may be the length of the inscription and the quality of the epigraphy. For it may be that these provide the best clue to the otherwise puzzling use of Arabic in a context where it was not required, as it would have been, say, in Qurʾānic studies. Knowledge of a prestigious foreign language is being flaunted, and all the more so when the inscription is long, complex in content and displaying abstrusely executed calligraphy of a bleakly cerebral kind. Someone is showing off. In plain terms, then, are we dealing with snobs? With the Samanid equivalent of modern people who lace their conversation with Latin or French or Italian tags whose meanings they do not fully understand? That would explain why a certain type of Samanid customer, like his latter-day counterpart struggling with a French menu in a non-French restaurant, needed a bit of help. Perhaps that help was of a more radical kind than the surviving evidence suggests – though if someone cannot even discover where the inscription begins, never mind what it says, that person is in a pretty bad way. One wonders whether the potter sold, along with the pot itself, a crib complete with translation. Then the owner, baraka to him, would be able to reel off the text on his pot to admiring visitors. Such ceramics, then, could have been fancy visiting cards on permanent display.

None of this is intended to exclude the possibility that the buyers of such ceramics – some of whom might indeed have belonged to the Arab minority – had a genuine loyalty to what might have been seen as traditional Arab culture, and that they expressed that loyalty in the choice of the pottery that they bought. Richard Bulliet has suggested that they may have been members of the earliest families to accept Islam and saw themselves as an elite group for that reason.129 Samanid material culture, and most particularly pottery, mirrors the clash of opposing attitudes and loyalties, between a conservative attachment to the Arab past, which would also embrace its prestige as the language of Islamic civilisation in general, and an equally conservative attachment to the more ancient Iranian civilisation. The evidence of that clash in material culture, moreover, is not confined to pottery. Standard Arab plans for mosques, and a number of mausolea whose tenants are identified, against all common sense, as the Prophet’s Companions, such as ‘Abdallāh b. Burayda130 and Qutham b. al-‘Abbās,131 contrast with local pre-Islamic structural and decorative modes. Alongside everyday coinage of standard pan-Islamic design and content are medallions whose iconographic roots are firmly in the eastern Iranian world.

This seductive scenario begs the question of how such visitors would ever see these ceramics, for the whole point of the purchase would be lost if they were not prominently on display. A lucky find by the American expedition which excavated medieval Rayy in the early 1930s indicated how this might have been managed. A merchant’s house was discovered in which one wall contained a series of plaster niches whose shapes mimicked the forms of contemporary ceramics.132 Much more fully preserved chambers of this kind, though of Safavid date, survive in Ardabil133 and Isfahan.134 They are the Persian equivalent of the Welsh dresser, but have this advantage over it – that the forms of the niches are tailored to the ceramics which they hold, so that every such collection is personalised. And the principle holds if only a single piece is displayed. This function of display goes far both to explain the large size of so many pieces and the fact that the inscriptions are frequently laid out so as to be seen all at once, not bit by bit.

Nevertheless, warnings against greed135 and gluttonous desires,136 wishes that ‘everything may be wholesome’137 and that the dish may be ‘a blessing to drink with’,138 the consoling reflection that ‘it is permissible and dainty to eat from a bowl that is always full’139 or that ‘the thankful eater is comparable to the one that fasts patiently’,140 all suggest that the purpose of these wares was not exclusively display. Of particular interest here is the energetic fourfold ‘kul!’ (‘eat!’)141 placed strategically at the base of the interior of a deep bowl, so that the diners see it only when they have nearly emptied its contents (Fig. 4.19). One can almost hear the voice of the attentive host urging his guests to finish all that has been served up to them. This takes up a theme already found inscribed on Abbasid blue and white pottery,142 Thus, for the most part, these wares were probably for use as well. But that use was, as it were, sanctified by the immemorial and honourable tradition of Arab hospitality, one of the linchpins of Jāhiliyya society.

The desire to show off, then, can be proposed as at least one tenable explanation for this pottery. So can the encouragement to eat. However, these are only two explanations. Another line of investigation might stress the special qualities of the Arabic language. There is no denying that it is a tongue eminently suited to proverbs: brief, sharp, concentrated. Did proverbs simply sound better in Arabic than in Persian, and were they in almost every mouth? Even today, several non-Arabic languages in the Muslim world contain a very high incidence of Arabic words and phrases.

Yet another explanation might be that at this period the very notion of inscribing an object in any language other than Arabic was not seriously considered. The prestige of that language was still unchallenged in the visual arts. It was, after all, not until the Timurid period that Persian inscriptions began to figure significantly in architecture143 and the so-called ‘minor arts’. By that reckoning, the Persian Renaissance was a phenomenon of the court and its sphere of influence was limited in the main to literature.

Thus the intended functions of these ceramics pose some knotty problems. So far, very few have been discovered with obviously personalised inscriptions naming a specific owner,144 and none mentioning a place or date. It seems fair to conclude that they were manufactured “on spec” for sale to the general public rather than to order. This suggests that they were aimed at a significantly lower level of patronage than, say, decorative metalwork or most textiles. They would have been the poor man’s metalwork – and Persian metalwork, too, it has to be remembered, has almost exclusively Arabic inscriptions in the early medieval period. The frequent repetition of good wishes in these inscriptions (especially baraka, ‘blessing’ [Fig. 4.15]), suggests that they may have been intended as gifts, perhaps on the occasion of the major festivals of the Muslim year such as the two ‘īds. It would have been attractive to propose that they could have served as a form of Nawrūz greeting, but the absence of Persian in this epigraphy excludes that possibility. Nevertheless, the generic similarity to Christmas cards can perhaps be accepted readily enough, and that might also explain the simpler designs with only a few words per plate. Perhaps these were the result of bulk orders.

Content

It is now time to confront squarely the problems posed by what these inscriptions say, and how they say it. The study by ‘Abdallah Ghouchani of a substantial sample of 140 pieces of this ware greatly expands the previous research by Bol’shakov, and, basing herself on this material, Oya Pancaroğlu has made it possible for the first time to suggest some preliminary conclusions about the nature and purpose of these inscriptions.145 It may be useful to begin by clearing away a few of the standard misconceptions. None of the pieces quote poetry, none use the Persian language, and there are no quotations from the Qurʾan, though some scholars have professed to detect occasional echoes from it. Nor is there an obvious Sufi strain, despite the important role that Sufis played in tenth-century Nishapur.146 Even the very word ‘Allāh’ occurs only rarely, notably in several pieces bearing the familiar motto al-mulk li’llāh – ‘power belongs to God’ (Fig. 4.20). One further piece expands this message – al-mulk li’llāh, al-wāḥid, al-qahhār, thus adding two attributes of God, the One and the Subduer. Perhaps this reluctance to employ overtly religious inscriptions may stem from the feeling that pottery is not an appropriately dignified vehicle for such texts. Three ḥadīths of the Prophet have been found, one of which occurs on two pieces.

Far more ceramics bear sayings attributed to ‘Alī b. Abī Ṭālib – a total of eleven found on 35 separate pieces. Thus some 27% of the sample analysed could be claimed to have religious associations, and that connection has a pronounced Shi‘ite tinge – and it was precisely the later tenth century that witnessed the public celebration of Shi‘ite festivals, even in Baghdad itself, the seat of the Sunni caliphs, but now controlled by the Buyids, a Western Iranian Shi‘ite military elite.147 One is immediately reminded of the concentration of Seljuq lustre tiles in Shi‘ite shrines in Iran – though there the percentage of ‘Shi‘ite’ tiles148 in relation to all lustre tiles is very much higher, whereas in this particular sample of Samanid epigraphic wares allegedly from Nishapur it is only a quarter of the total. Moreover, Seljuq lustre wares were manufactured in Kāshān, a town which in the late twelfth and thirteenth centuries was well known for its fervent Shi‘ite sympathies. Nishapur, like any other major Iranian city of the tenth and eleventh centuries, had its Shi‘ite community, but the work of Bulliet149 and Bosworth150 does not suggest that it accounted for more than a quarter of the population. And of course it would be a gross abuse of these statistics to interpret them as leading to the conclusion that this segment of the city’s population bought wares with sayings attributed to ‘Alī. Nevertheless, it is a factor to bear in mind in any general survey of Samanid Nishapur.

The remaining three-quarters of these wares, then – a very substantial majority – consists of proverbs, 40 in all, some repeated three, four, six, eight, fifteen or even thirty-five times on separate pieces. A high moral tone is adopted which somehow adds a gloss of religious respectability to these nuggets of practical wisdom. Hence the frequent references to reward and punishment, the Garden and the Fire. Oya Pancaroğlu has proposed three major categories for these proverbs, each with its subdivisions: generosity, virtuous conduct and knowledge.151 A few examples will suffice to give the flavour of these aphorisms.

Knowledge is bitter to the taste at first,

But in the end it is sweeter than honey

The free man is still free even if touched by harm

Patience is the key to joy and happiness152

Generosity and paradise are for pious men …

Impiety and parsimony are hell’s children.

With good health!

The noblest of riches is the abandonment of desire

Generosity is the custodian of honour and property153

Generosity is a quality of the people of paradise154

The free man does not break his promise

Safety lies in silence, and only speech

Will reveal the inner side of the man with faults155

All this is very elevating, and no doubt all of us have visited houses – perhaps some of us also live in them – where similarly uplifting sentiments are prominently displayed, with messages like ‘Better once in heaven than ten times at the gate’. The difference is merely that the medium of expression nowadays is more likely to be a poster than a pot. In an earlier age, samplers were a favourite repository for such axioms, and they were made to be put on display, not tucked away in a drawer somewhere. And even today, some of the sentiments inscribed on Samanid wares can be found on pottery – and not high-end pottery either. ‘Best wishes from Bognor’, ‘Greetings from Blackpool’. Or, as a direct equivalent to the frequent references to food on Samanid ceramics, the message on plates that read ‘Give us this day our daily bread’. No doubt similar homely messages seen over the years, expressions of popular culture, will come to mind in this context. The plaque by the front door that says ‘Home Sweet Home’. The doormat inscribed ‘Welcome’. That, one may suggest, is at least part of the context of these wares.

And yet it is only one part. The kind of message just discussed is always couched in everyday language. It is not in a foreign tongue. That foreign tongue distances the sentiment expressed, no matter how homely it is, and removes it to a higher plane. It also imparts an extra authority and formality to the message. This brings us back to the notion that these wares embody aspirations to greater social distinction.

Some of these pieces exert a special fascination because something seems to have gone wrong. It is probably safer to say ‘seems’, since the evidence is too ambivalent to convict the calligrapher of an outright mistake. Is it a mistake, for example, when the scribe adds one or two unrelated words, such as al-ni‘ma (‘delight’), al-yumn (‘prosperity’), al-salāma (‘good health’) or baraka (‘blessing’) to an otherwise self-contained text? One cannot generalise. Sometimes the extra word is required not for its sense, but visually, because the alif and lam of the definite article fill an otherwise painful gap in the regular succession of ascenders. Sometimes it is only the alif that is so stressed. Or the redundant element may be a single letter, such as a dāl which, in the Kufic ductus typical of these wares, can be prolonged for several centimetres without appearing unduly distorted. In some cases the redundant letter or word can be understood in a humorous sense. Lisa Volov has noted the comic contradiction embodied in the carefully written text ‘Deliberation before action protects you from regret’ – we would say ‘Look before you leap’ – which is then followed by the otiose letter kāf, proving that deliberation did not occur in this case.156 Sometimes the calligrapher seems to be playing with his readers, leading them on only to disappoint them. Take a famous piece in the Freer Gallery of Art (Fig. 4.33),157 one of the most accomplished of the entire series. The outer inscription reads ‘He who believes in recompense is generous. And to whatever you accustom yourself, you will grow accustomed. Blessing to the owner’. The inner one repeats the proverb exactly except for the crucial last word. That leaves the sense hanging in the air, as if the scribe had written: ‘What then?’ So what at first sight looks like a virtuoso technical display – the repeating of the identical inscription, executed with equal grace and fluency, but on a much smaller scale – suddenly turns, at the very last moment, into an enigma. One is left contemplating the central eye, and asking oneself ‘so what does happen when you have accustomed yourself to anything?’ Given the flawless execution of this inscription, it is perhaps more likely than not that the scribe was producing his own ironic commentary on the proverb by the very way he wrote it.

A similar irony could be detected in an inscription running across the centre of a plain dish: ‘O he that seeks the universe when it is Death that seeks him’ (Fig. 4.21).158 The text begins not at the far right of the bowl, where one might legitimately expect it, but well in towards the centre, surrounded by plenty of free space – perhaps suggesting thereby the ample aspirations of the seeker with the whole world to conquer. But the end of the text is treated very differently from the beginning, for the second half – ‘it is Death that seeks him’ – climbs sharply up the plunging surface of the bowl and is so laid out that the final suffix -hu (i.e. ‘him’) ends right on the rim. It is as if Death himself had finally nailed the speaker despite his frantic attempts to escape his allotted span, his appointment in Sāmarrāʾ.159 There may even be a punning reference to ‘Alī b. Abī Ṭālib, to whom the proverb is ascribed, in the double repetition of the verb ṭalaba, ‘to seek’. A further refinement of meaning may be intended by the way the inscription begins halfway across the dish, as if to suggest that the ambitious seeker after the universe has less time left than he thinks. All this may seem rather fanciful, but it does provide at least a possible explanation for the otherwise puzzling irregularity of the layout.

What is one to make of the sudden guillotine applied to the loquacious and rambling text which runs, in the manner of some Persian Polonius, ‘money is what you spend, not what you collect, and crops are not what you harvest, but what you cultivate, and it is said’? Could that last phrase be poking sly fun at the ponderous and repetitive banality of the sentiments that precede it? As if the speaker had still more examples up his sleeve but was cut short in mid-sentence? Humour can also be detected, not just in the actual content of these inscriptions, or more precisely in the relationship between what they say and the way the writing comments sotto voce on the message, but also in the idiosyncrasies of the writing itself, independent of the meaning it puts across. These examples indicate that humour makes more than just a guest appearance in Samanid epigraphic wares.160 A bowl from Utrār even has a caricatured lugubrious face on its base.161

The possible sources of Samanid epigraphic wares

A major problem still remains to be tackled at this stage: why was the writing on these ceramics of this particular kind? This question highlights the problem of what sources might have inspired these wares. Hitherto, the discussion of Samanid pottery has focussed, reasonably enough, on the mere fact that these pieces exalt the decorative potential of writing. This approach is fine, so far as it goes. But it does not go far enough, and it does not sufficiently grapple with the problem of why this particular use of writing is so different from the other ways in which epigraphy figures on Islamic pottery. For while Samanid pottery is no stranger to display, it is also not averse to the gentle art of suggestion. Indeed, what it suggests, in several different ways – and that word ‘suggests’ is significant – is a connection with the Qurʾan. It may be, in fact, that this was the decisive factor in its popularity.

The sheer range of Kufic scripts encountered in Samanid pottery is enough to prove beyond doubt that this was a golden age of experiment. Surely it is no accident that it coincides in time, though not in space, with the increasingly bold development of what has been variously known as Carmathian or East Persian or New Kufic, or more recently as broken cursive,162 and with the celebrated innovations of Ibn Muqla and Ibn al-Bawwāb163 – innovations memorialised above all in the Kufic Qurʾans of the tenth and early eleventh centuries and in the earliest naskhi Qurʾans. Thus fine writing had come to imply the Qurʾan. Indeed, it is not impossible that some of the varieties of Kufic found on these Samanid wares reflect more or less closely some of the styles developed by the leading calligraphers of the age. If so, that would indicate an unheralded democratisation of this most prestigious art. Indeed, these wares provide proof positive that the appreciation of calligraphy was not confined to court circles. On the contrary, there was a large popular market among the less moneyed classes for this finest of Islamic fine arts – even if the medium was unusually modest.

And surely the element of competition, of emulation, is not to be underrated; twenty-six of the 140 pieces analysed by ‘Abdallah Ghouchani are signed by craftsmen. Though the craft is not defined, this quantity of signatures sets a new benchmark of pride of achievement in Islamic pottery.164 Moreover, since these wares were presumably for sale in the bazaar, the calligraphers would have been able to keep an eye on each others’ work. Calligraphers working on or for pottery could allow themselves a freedom unthinkable in the context of a Qurʾan, where propriety would demand a consistent performance. Moreover, each of these wares represented a one-off commission (even if the stencil of the inscription was made and was used many times). Samanid ceramic epigraphy thus shines a spotlight on workshop practice in the field of calligraphy at an unprecedentedly popular level. No wonder, then, that so many of the styles on Samanid ceramics are not known from any other source. In a sense, they can be interpreted as exercises – not mere doodles, but consciously intended as experiments. It is as if an artist’s sketchbook had survived, complete with mistakes and dead ends. Calligraphers could sharpen their graphic skills and earn money at the same time – and, not least, they could have fun. They could play with grotesque distortions of familiar letters and words, with rhythms too bizarre to be maintained for long on the written page, with baroque incrustations of applied ornament. Of course these endless and unpredictable variations in style made such wares eminently collectable.

However, the main point to stress here is that some of the finest of these inscriptions – unlike those on practically every other kind of ceramics in the Muslim world – seem, to judge by their very high quality, to have been the work of professional calligraphers, not of potters. And the most prestigious task of such calligraphers was of course to copy the Qurʾan. This, then, is where the art of the book and the art of the pot meet – not for the last time in Persian ceramics, though sadly the later interplay of book writing and pottery inscriptions was of a decidedly lower standard. At the most obvious level, this connection is asserted, as already noted, by the type of Kufic employed. It is important to stress here that it was by no means self-understood, whether in earlier or later periods or indeed in Samanid times, that the style of writing used on pottery should have anything to do with that used for Qurʾans. Indeed, earlier potters, and many Samanid ones (as noted earlier) had relinquished any claim to such a connection by using a brush for their inscriptions,165 and thereby sacrificing the razor-sharp outline made possible by the reed pen, the qalam. The craftsmen responsible for many Samanid ceramics, on the other hand, did employ the qalam,166 either via a stencil or directly. Inscriptions executed with the brush may have their full measure of impetuosity, power, freedom from restraint; but in many cases their fuzzy, even ragged outlines simply cannot match the precision and delicacy of pen-made writing. In the context of ceramics, such writing had all the shock of the new. And its immediate association would have been with the Qurʾan. The contrast between extreme thickness and extreme thinness (as found on the Sarikhani jug – Fig. 4.17) is typical of certain Qurʾans (Fig. 4.28).

That association is driven home by a number of other features. Chief among them is the colour scheme adopted in these ceramics. For the most part this comprises a cream-coloured slip, often with a slightly brownish or buff tinge, against which the dark chocolate-brown of the epigraphy stands out dramatically. These are precisely the dominant colours in contemporary east Persian Qurʾans (Fig. 4.27). And as with these Qurʾans, the exclusion of other colours focuses attention on the writing and charges it with extra energy. Occasional grace notes of red pigment used for single letters or for some of the ornamental flourishes echo the sparing use of touches of red in Qurʾans, where, for example, red dots mark shaddas, short vowels, verse endings or difficult words.167 The colour scheme of brown on white can be reversed; moreover, the manganese pigment used for brown can shade into purple or black, the red can turn pink, the ground can be pale lilac, and sometimes an ochre inscription stands out against a red ground. So the palette has several sub-sets.

In addition, as noted earlier, many of these wares contain extraordinarily complex and densely composed palmettes which bear a strong generic similarity to the palmette or little tree (shujayra) so often encountered in early Qurʾans, including those from tenth-century Iran. And just as the palmettes in Qurʾans are placed at the furthest lateral extremity of the page, perhaps among other things to energise the empty space of the margin, so too these palmettes on pottery extend to the very edge of the plate, their huge bulk well removed from the inscription itself. Placed at strategic intervals along the circumference of a dish, they fulfil a secondary function as space-dividers.

Letters may be given a plain cream background which broadly follows their shape and has a red border, while the rest of the available space around the letters thus enclosed is filled with brown dots. The result is that an abstract pattern correlated roughly to the form of the inscription acts not just as a contouring device but also as a foil to the text and as a subsidiary focus of interest. This is a means of superimposing the writing on the ornament and, in order to ensure that there is no conflict of interest, to give each letter or group of letters its own surround of plain white. That basic idea is already found in a less developed form in Sāmarrān lustreware and – more significantly in the context of an association with Qurʾanic manuscripts – in the colophon of the famous Qurʾan of Ibn al-Bawwāb of 391/1000.168 The effect is comparable to that of a halo, and indeed that association can perhaps be recognised in some later Qurʾans, where the text sent down from heaven is appropriately enveloped in a nimbus of star-spangled cloud. These ceramics, then, offer some of the earliest examples of what was later to be called abrī painting.169 Some of its most powerful expressions can be seen in fourteenth-century Mamluk Qurʾans where the entire text (as distinct from individual letters or words) appears to float on a delicately spotted cloud which encases its general form; and abrī painting persisted for centuries thereafter.170

It seems, then, that this catalogue of resemblances between Samanid pottery and the Qurʾans of this period stretches the long arm of coincidence well beyond breaking point, and that the connection must have been intentional. But why was that connection devised in the first place? The extreme reverence accorded to the Qurʾan sufficiently explains why its actual words are not inscribed on dishes which at least in theory could be used to hold food or drink. But these numerous parallels with fine Qurʾans – for instance, the subtle use of elongation of letters – indicate that these ceramics were intended to partake in some measure of the sanctity enveloping that book. They radiate a residual holiness. The proverbial wisdom of their texts acquires an extra resonance conferred by the baraka of Holy Writ. And all this is subliminal: implied, not stated. The implied connection with religious faith becomes explicit in some cases where the content is a saying of the Prophet or of ‘Alī, and this can be linked to the religious controversies of the later tenth century in the eastern Islamic world.171 Perhaps, too, those who bought these wares saw them as a kind of amulet, to be displayed in the house rather than worn on the person – and such a belief could explain why the single word most commonly found on them is precisely baraka,172 a word which in this context takes on the lineaments, the magical efficacy, of a talisman.173 Nor is this the only motif with potential talismanic significance.174 Such reflections provide still another explanation of why the texts had to be in Arabic: for this was the language of the Qurʾan and of the Muslim faith.

There is an instructive footnote to the discussion above. This Samanid pottery offers an early example of what later became a standard practice of Islamic art: the migration of ideas and designs from one medium to another. And since a switch of this kind usually involves changes in scale, execution, colour, texture and of course function, the result is apt to be a transformation rather than a mere transference. The connection between Samanid ceramics and contemporary silverware is an obvious case in point;175 it extends from double concentric inscriptions or the use of aphorisms or similar forms to such details as interlaced geometric designs or floral motifs.176 In China, potters created porcelain imitations of silverware, and in similar fashion Julian Raby documents how an increase in the value of silver in the tenth century in the Samanid realm was matched by a burgeoning of fine pottery with a similar aesthetic.177 Thus it is noticeable that one of the most characteristic methods employed by Muslim craftsmen to regenerate tired designs and to devise new approaches by seeking inspiration from within – as distinct from outside – the Islamic world was already well developed in eastern Iran by the tenth century. Of course it could be taken to extremes, perhaps under the influence of undue regulation in court workshops. This is shown in particular by Safavid art, where the same visual clichés infiltrate pottery and textiles, frescoes and miniature paintings, lacquer and wall tiles, carpets and bookbinding. In that case one might argue that the various media are not enriched but impoverished. But such abuses, if they may be so termed, lay far in the future. In the Samanid ceramics under discussion one sees potters looking beyond the obvious parallels of metalwork or foreign pottery, and thinking laterally. The result was to give their wares an entirely new dimension. The choice of manuscript Qurʾans as a source of inspiration, moreover, conferred a far greater weight of meaning on their wares than did, say, the widespread use of Sāmarran stucco motifs on glass, woodwork and the like. In that sense their introduction of an entirely different art form, complete with its own set of distinctive associations, into the much more humble medium of pottery, opened new vistas for Islamic art.

The connection with the Qurʾan, however plausible it may be, is not the only possible explanation for these wares. Earlier scholarship has tended to underplay or even discount the possibility that Chinese elements were at work in them. Yet the historical situation makes it likely enough that Samanid culture as a whole was open to ideas and artefacts from China. The Samanids acted as a bridge between east and west, and were the obvious middlemen for all goods travelling by land between China and the Islamic world in both directions.178 Their eastern frontier was with China or with states politically dominated by the Chinese. Chinese princesses were despatched to marry Central Asian rulers;179 Chinese painters, it seems, worked at the Samanid court,180 and they are at any rate a topos in Samanid poetry.181 Embassies were exchanged between the two sides.182 The golden peaches of Samarqand were only one of the many kinds of Muslim exotica coveted by the T’ang court,183 while Chinese mirrors and other metalwork found a ready market in Transoxiana.184 As it happens, pottery represents perhaps the best documented of all these forms of cultural and commercial interchange. Imperial Chinese porcelain as well as many other less valued Chinese wares, such as celadon and white wares, reached the Abbasid court through the intermediary of the Samanid rulers,185 and T’ang splashed ware (which also on occasion bore inscriptions in ceramic cursive186) as well as celadon was widely copied at Nishapur.187 Sherd evidence from Afrasiyab has also yielded two types of wares with Chinese connections.188 Indeed, the whole subject of the interplay of Chinese and Persian modes in ceramics has generated a substantial scholarly literature, expertly surveyed by Oliver Watson some twenty years ago.189

But this discussion has not dealt in detail with inscribed wares, and it may be that that there are still some discoveries to be made in this area. A probably ninth-century bottle of Changsha ware bears what may well be intended as an inscription, although it strains credibility to recognise it as Arabic;190 and this is only one of several such pieces. And most telling of all, Chinese wares with no decoration other than Chinese characters were produced at least as early as the ninth century. Thus a money jar of Changsha ware dated 858 A.D. displays Chinese inscriptions executed in brownish-black on a buff ground and arranged in repeated vertical columns (Fig. 4.22),191 and bowls of the same period, similarly decorated, are known.192 Specimens of Changsha ware found their way to the Muslim world and have been recorded at sites as far distant from each other as Sāmarrā’, Sīrāf, Nishapur and Fusṭāṭ, among many other sites.193 Thus there is no denying the possibility that potters in Central Asia could have seen Chinese wares with no decoration other than writing, and have found them inspirational. These were, after all, pieces from pre-Samanid or very early Samanid times, so that there would have been enough time for these ideas of connecting writing with pottery to have taken root over an extended period. The connection was not limited to Changsha wares. Indeed, one variety of Song ware, the Cizhou/T’zu-chou type which was popular for centuries, uses precisely the cream and chocolate colour scheme of Samanid wares. Sometimes, like some Samanid dishes, it favours highly stylised creatures, as the principal decoration;194 more often it features vegetal or floral ornament; and occasionally it bears inscriptions. In some pieces, indeed, the writing is the only decoration,195 and this continued into the Yuan period, using free calligraphy that reads, for example, qingjing (‘purity’, ‘peace’, ‘tranquility’) or, more plainly, jiu (‘wine’) in Chinese cursive script (Fig. 4.23). Maxims for moral behaviour were both common and commonplace,196 as in Samanid pottery, as were simple references to the content of the piece.197 The inscription on a wine jar of Cizhou type of about the eleventh century,198 for instance, proclaims ‘purity, clarity and endurance’ in strong, outlined calligraphy against a background of tiny circles. Cizhou wares were for everyday use and not made for an elite, and many display bold, flowing, sweeping designs; in all these respects they have clear points of contact with Samanid epigraphic wares. They may even have passed through the Samanid domains, though the vast body of Chinese wares exported to the Middle East made their way west by sea.

The uncertain chronology of Samanid wares prohibits any definite statement as to whether it was the Islamic or the Chinese potters who first experimented seriously with writing on ceramics in this particular combination of colours in the ninth and tenth centuries, though the balance of probability points to the Muslim craftsmen as the innovators. Certainly, they experimented much more widely and boldly in this genre of pottery, and even, it seems, imitated Chinese characters, but so badly that they were meaningless.199 In just the same way, Chinese wares with allegedly Arabic inscriptions depict marks which might suggest Arabic epigraphy to those not familiar with it, but are not easily recognisable as such.200 However, the key ideas could first have travelled in either direction. At all events, the striking consonance in form, colour and content between Chinese epigraphic wares and Samanid ones that seem to be slightly earlier cannot lightly be set aside, especially when one remembers how greatly early Islamic pottery was indebted to the example of China. If, then, these Samanid epigraphic wares drew on the prestige of both Arab and Chinese culture, they would appropriately express the wide cultural horizons of Samanid Central Asia.

Conclusion

It seems plausible, when one surveys these inscribed ceramics as an entire genre, to conclude that they take to an extreme the conflict between the desire to inform and the desire to decorate. There can be no doubt that ornament trumps meaning. Indeed, to this day the popularity of these wares depends overwhelmingly on their beauty, which is obvious to all, rather than on their meaning, which is hidden from all except the privileged few. That mystery has much to do with their unique status among the ceramics of the Muslim world. In their own time, as this paper has tried to show, they served a variety of purposes; they gave aesthetic pleasure, they ministered to snobbery, they held food and, by extension, promoted hospitality, they proclaimed loyalty to traditional Arab values, and they lent themselves to public display – all this at the same time as providing moral uplift, sometimes with a distinctly religious flavour201 and even subliminal echoes of the Qurʾan. No wonder that they were so popular.

It is the great achievement of these Samanid calligraphers to have been the first – at least so far as surviving evidence shows – to exploit, beyond the privileged domain of the book, the expressive potential of the Arabic alphabet. Some say that in this respect it is much richer than our Roman one. I wonder. One has only to walk down a typical High Street in almost any British town to see a weird assortment of script styles, including pastiches of Chinese, Celtic, Greek, Russian and Arabic scripts, advertising anything from hairdressers to greengrocers. The difference is that for the Roman alphabet such freedom from convention has been achieved only in the last century or so; the Muslim world is a thousand years ahead of the West in that experiment. All the more sad, then, that the first great peak of attainment in the Samanid period was also arguably the high point of the entire development – though modern calligraphers in Iraq, Algeria, Egypt and Pakistan are using the Arabic alphabet as a foundation for all kinds of pictorial experiments. Ottoman and Qajar calligraphers had already explored some of this potential a century and more ago. Nevertheless, in the sphere of non-monumental inscriptions whose message is intended to be legible, these Samanid wares reached a level of technical and aesthetic excellence that was to remain unsurpassed.

Notes

I should like to thank Dr Barbara Brend, David Gye and most especially Dr Melanie Gibson for their helpful comments on this paper, and Kathy Judelson for translations of the articles by Bol’shakov. Much of this paper was written while I was a Senior Research Fellow at the Museum of Islamic Art at Doha. I am most grateful to Dr Thalia Kennedy and her staff for facilitating my work at every turn, and for permission to publish some of the Museum’s Samanid pieces, which are still relatively little known.

1.For Samanid Bukhara at its apogee, see Richard N. Frye, Bukhara. The Medieval Achievement (Norman, Oklahoma, 1965), pp. 50–84. For a survey of the dynasty as a whole, see Richard N. Frye, ‘The Samanids’, in Richard N. Frye (ed.), The Cambridge History of Iran,Vol IV: From the Arab Invasion to the Seljuqs (Cambridge, 1975), pp. 136–61. For useful summaries of Samanid politics and culture see A.C.S. Peacock, Mediaeval Islamic Historiography and Political Legitimacy: Bal‘amī’s Tārīkhnāma (London, 2007), pp. 16–21 and 35–48. A substantial up-to-date monograph on the Samanids is long overdue; meanwhile, see W. Luke Treadwell, ‘The account of the Samanid dynasty in Ibn Zāfir al-Azdī’s Akhbār al-duwal al-munqaṭi’a’, Iran xliii (2005), pp. 135–73.

2.But this term is relative. C. Edmund Bosworth, The Ghaznavids: Their Empire in Afghanistan and Eastern Iran 994–1040 (Edinburgh, 1963), p. 162, estimates the size of Nishapur at 30,000–40,000. For a more detailed analysis of this issue, see Richard W. Bulliet, ‘Medieval Nīshāpūr. A topographical and demographic reconstruction’, Studia Iranica 5 (1976), pp. 67–89; and, more recently, Richard W. Bulliet, Cotton, Climate and Camels in Early Islamic Iran. A Moment in World History (New York, 2009), p. 4 (with an estimate of 200,000).

3.For an in-depth study of Nishapur in the tenth and eleventh centuries, see Bosworth, Ghaznavids, pp. 145–202.

4.Peacock, Mediaeval Islamic Historiography, especially pp. 49–140.

5.Jan Rypka, History of Iranian Literature, ed. Karl Jahn (Dordrecht, 1968), pp. 139–71.

6.See the discussion of the Shu‘ūbiya by Ignaz Goldziher, ‘The Shu‘ūbiyya’ and ‘The Shu‘ūbiyya and its manifestation in scholarship’, in Muslim Studies. Muhammedanische Studien, ed. Samuel M. Stern, tr. Samuel M. Stern and C. Renate Barber (London, 1967), I, pp. 137–63 and 164–200; Roy P. Mottahedeh, ‘The Shu‘ubiyah controversy and the social history of early Islamic Iran’, International Journal of Middle East Studies vii/2 (1976), pp. 161–182; and Susanne Enderwitz, ‘Shu‘ūbiyya,’ EI2, ix, pp. 513b-516a.

7.Thus the celebrated Tomb of the Samanids could be seen as a reincarnation of the standard Sasanian fire temple or chahār ṭāq, and so could its nearest Qarakhanid equivalent, the tomb popularly known as that of ‘Ā’isha Bībī (Monique Kervran, ‘Un monument baroque dans les steppes du Kazakhstan: Le tombeau d’Örkina Khatun, princesse Chaghatay?’, Arts Asiatiques lvii [2002], pp. 5–32).

8.W. Luke Treadwell, ‘A unique portrait medallion from Bukhara dated 969 A.D.’, The Ashmolean xxxvi (1999), pp. 9–10. This is one of several closely related variants on Sasanian and Central Asian models; I am indebted to Dr Treadwell for this information. See also W. Luke Treadwell, ‘Shāhānshāh and al-Malik al-Mu’ayyad: The legitimation of power in Samanid and Būyid Iran’, in Farhad Daftary and Josef W. Meri (eds), Culture and Memory in Medieval Islam. Essays in Honour of Wilferd Madelung (London and New York, 2003), pp. 328–30 and 336–7 and Shamsiddin Kamoliddin, ‘On the Religion of the Samanid Ancestors’ Transoxiana 11 (July 2006), unpaginated, §§19–25 (www.transoxiana.org/11/kamoliddin-samanids.html), accessed 21 July 2014. Kamoliddin extends still further the comparanda for this fascinating medallion.

9.This aspect of Nishapuri figural wares is fully explored in Teresa Fitzherbert,‘Themes and images on the animate buff ware of medieval Nīshāpūr’ (unpublished MPhil dissertation, University of Oxford, 1983), especially pp. 29–64; cf. also Richard Ettinghausen, ‘A case of traditionalism in Iranian art’, in Oktay Aslanapa and Rudolf Naumann (eds), Forschungen zur Kunst Asiens. In Memoriam Kurt Erdmann 9. September 1901 – 30. September 1964 (Istanbul, 1969), pp. 88–110 and Robert Hillenbrand, ‘The Islamic re-working of the Sasanian heritage: two case studies’, in Patricia L. Baker and Barbara Brend (eds), Shifting Sands, Reading Signs. Studies in Honour of Géza Fehérvári (London, 2006), pp. 1–14.

10.For the Būyid ‘renaissance’, see Adam Mez, The Renaissance of Islam, tr. Salahuddin Khuda Bakhsh and David S. Margoliouth (Patna, 1937), passim but especially pp. 15–31, 379–408 and 418–29; Bertold Spuler, Iran in frühislamischer Zeit. Politik, Kultur, Verwaltung und öffentliches Leben zwischen der arabischen und seldschukischen Eroberung 633 bis 1055 (Wiesbaden, 1952), pp. 342–56 (though some of this material refers to eastern Iran); Wilferd Madelung, ‘The assumption of the title Shāhānshāh by the Būyids and “The reign of Daylam (Dawlat al-Daylam)”’, Journal of Near Eastern Studies xxviii (1969), pp. 84–108 and 168–83; Heribert Busse, Chalif und Grosskönig. Die Buyiden im Irak (945–1055), (repr. Beirut, 2004), pp. 154–6 and 203–26 and Joel L. Kraemer, Humanism in the Renaissance of Islam (Leiden, 1992).

11.For some characteristic examples of the visual expression of this process, see Mehdi Bahrami, ‘A gold medal in the Freer Gallery of Art’, in George C. Miles (ed.), Archaeologica Orientalia in Memoriam Ernst Herzfeld, (Locust Valley, N.Y., 1952), pp. 5–20. For a general survey of the visual arts in this period, see Ernst Kühnel,‘Die Kunst Persiens unter den Buyiden’, Zeitschrift der Deutschen Morgenländischen Gesellschaft cvi (1956), pp. 78–92, an account which badly needs updating at length; meanwhile, cf. Jonathan M. Bloom, ‘The expression of power in the art and architecture of early Islamic Iran’, in Edmund Herzig and Sarah Stewart (eds), The Idea of Iran, Vol V: Early Islamic Iran (London, 2012), pp. 102-19.

12.Assadollah S. Melikian-Chirvani, ‘L’évocation littéraire du bouddhisme dans l’Iran musulman’, Le Monde Iranien et l’Islam ii (1974), pp. 34–65.

13.Kamoliddin, ‘Ancestors’, §§1–25. He illustrates mandala designs, to which he attributes a Buddhist origin, both on Samanid coins (§§12, 16–18; see too Fig. 1, a copper fals of 359/969–70 with an 8-petal design, and Fig. 2, a mandala design of 3 super-posed squares) and on buildings (§13 – the reference is to the spandrel design over the entrance arch of the Tomb of the Samanids, which is similar to his Fig. 2 – and Figs 3–6, featuring a complex multi-layered polygonal mandala, with a central spiral motif, on the tomb of Astāna-Bābā).

14.Peter Morgan, ‘Samanid pottery. Types and techniques’, in Ernst J. Grube, The Nasser D. Khalili Collection of Islamic Art. Volume IX (General editor Julian Raby), Cobalt and Lustre. The First Centuries of Islamic Pottery (London and Oxford, 1994), pp. 55–9.

15.This is the most famous of them all, the large bowl (diameter 39.3cm) in the Freer Gallery of Art in Washington (57.24), brilliantly analysed by Sheila Blair (Text and Image in Persian Art [Edinburgh, 2014], pp. 11–16 and 48).

16.Some of the differences between the two groups have been identified, but the problem of distinguishing one from another is most challenging precisely in the case of the wares with a white ground (see Charles K. Wilkinson, ‘The glazed pottery of Nīshāpūr and Samarkand’, Bulletin of the Metropolitan Museum of Art, N.S. xx/3 [1961], p. 109; cf. Blair, Text and Image, pp. 22–3). While some groups can safely be associated with either Afrasiyab or Nishapur, others were made in both cities. So far as design and calligraphic quality are concerned, it is often hard to tell the products of these two centres (not to mention other sites) apart.

17.Rudolf Sellheim, ‘Eine unbekannte persische Sprichwörtersammlung’, Der Islam lxxxi (2004), pp. 243–8. This was produced for the ruler of Sīvās in 693/1294 and contains material stretching back to pre-Islamic times. For further information on ancient Persian wisdom literature (andarz), see Wolfhart Heinrichs, ‘Orientalisches Mittelalter’, in Klaus von See (ed.), Neues Handbuch zur Literaturwissenschaft (Wiesbaden, 1990), v, pp. 346–65 and Mohsen Zakeri, ‘‘Alī ibn ‘Ubayda al-Rayḥānī), and a forgotten belletrist (adīb) and Pahlavi translator’, Oriens xxxiv (1994), pp. 78, 81 (on the collection of proverbial sayings from the Avesta by the ninth-century scholar ‘Alī b. ‘Ubayda al-Rayḥānī) and p. 91 (on the popularity of his book of Avestan proverbs, al-Maṣūn, among the Persians of Khurasan). Zakeri suggests that ‘many of the Avestan dictums’ were ‘filtered through the medium of Arabic translations’ (‘ar-Raiḥānī’, 92). Al-Rayḥānī apparently translated from Pahlavī some of a collection of maxims made by the sixth-century Zoroastrian mobad of Khusraw Anūshīrwān, Mihr Aḏar Gušnasp, as had Ibn al-Muqaffa‘ before him (Ibid, p. 101). Cf. too the remarks by Walter B. Henning on such axioms to the effect that what is long-winded and cumbersome in Middle Persian is elegant, stylish and pointed in Arabic (‘Eine arabische Version mittelpersischer Weisheitschriften’, Zeitschrift der Deutschen Morgenländischen Gesellschaft cvi (1956), p. 75. But see Laurence P. Elwell-Sutton, Persian Proverbs (London, 1954), pp. 3–4, for examples full of rhyme and rhythm, pungency and wit. And it is worth noting that al-Maydānī, perhaps the most famous medieval collector of Arabic proverbs, was a Persian who died in – of all places – Nishapur; see Hartmut Bobzin (ed.), 1001 Alt-arabische Sprichwörter. Deutsch von Friedrich Rückert. Aus dem Nachlaß ausgewählt, herausgegeben und eingeleitet von Hartmut Bobzin (Wiesbaden, 1988), p. 15.

18.The Rückert translation published in Bobzin, Sprichwörter, is from Freytag’s edition of al-Maydānī (Georg W. Freytag, Arabum Proverbia, vocalibus instruxit, latine vertit, commentario illustravit et sumptibus edidit G.W.Freytag i-iii (Bonn, 1838–43, repr. Osnabrück 1968), and is itself a masterpiece of conciseness and ingenious rhyme. Balance, assonance, alliteration and rhyme are as commonplace in Arabic proverbs in recent as in medieval times; see John L. Burckhardt, Arabic Proverbs (London, Dublin and Totowa, N.J., 1972), nos. 203, 205, 314, 399,415, 437 and 566.

19.For an overview of the characteristics of the oldest Arabic proverbs, and especially their roots in pre-Islamic Arab society, see Toufic Fahd, Etudes d’Histoire et de Civilisation Arabes et Islamiques II (Istanbul, 2006), pp. 85–99. For a useful bibliography of works on Arabic proverbs, see Bol’shakov, ‘Nadpisi’, XVI, 1963, pp. 50 and 52.

20.Shlomo D. Goitein, ‘The present-day Arabic proverb as a testimony to the social history of the Middle East’, Studies in Islamic history and institutions (Leiden, 1968), p. 368. It was, he notes, particularly Persians who collected these ancient Arabic proverbs.

21.Which was not confined to the Samanids. For the same process at the Saffārid court, cf. Samuel M. Stern, ‘Ya‘qūb the Coppersmith and Persian national sentiment’, in C. Edmund Bosworth (ed.), Iran and Islam. In Memory of the Late Vladimir Minorsky (Edinburgh, 1971), especially pp. 537–49.

22.Edward G. Browne, A Literary History of Persia. Vol. I: From the Earliest Times until Firdawsí (repr. Cambridge, 1977), pp. 453–4; Rypka, History, pp. 143 and 166. Cf. n.48 below.

23.C. Edmund Bosworth (tr.), The Book of Curious and Entertaining Information. The Laṭā’if al-Ma‘ārif of Tha‘ālibī (Edinburgh, 1968), pp. 131–3.

24.The images in Galina V. Shishkina and Ludmilla V. Pavchinskaja, Terres secrètes de Samarcande. Céramiques du VIIIe au XIIIe Siècle (Paris, 1992), pp. 56–9 and 90, offer a salutary reminder of the actual condition of many of these wares; see too Kirsty Norman, ‘Restoration and faking of Islamic ceramics: case histories’, in Oliver Watson (ed.), Ceramics from Islamic Lands (London, 2004), pp. 69–89.

25.Bol’shakov’s material, which comes almost exclusively from Afrāsiyāb, consists almost entirely of sherds, and this, taken with the lack of a developed commercial market in Central Asia in the 1950s, effectively renders his database bullet-proof against any suspicion of faking or forgery. This cannot be claimed for the material allegedly from Nīshāpūr, and in recent decades much suspect material has entered the market; see Oliver Watson, ‘Fakes and forgeries in Islamic pottery’, in Barbara Finster, Christa Fragner and Herta Hafenrichter (eds), Kunst und Kunsthandwerk im Islam. 2. Bamberger Symposium der islamischen Kunst 25. – 27. Juli 1996, Oriente Moderno xxiii (lxxxiv)/2 (2004), pp. 517 and 521–8.

26.A first attempt in this direction was made by Samuel Flury, ‘B. Ornamental Kūfic inscriptions on pottery’, in Arthur U. Pope and Phyllis Ackerman (eds), A Survey of Persian Art from Prehistoric Times to the Present (London and New York, 1938– 9), 1751–5. It should be pointed out that Samanid epigraphic wares get astonishingly short shrift in this pioneering survey of Persian Islamic pottery; only two of them – the majestic Louvre piece (Fig. 2) and an indifferent bowl, formerly in the Debenham collection – figure among the 474 pieces illustrated in the 256 coloured or black and white plates, in which Seljuq lustre and mīnā’ī wares dominate. Even in Maurice Pézard’s groundbreaking survey La Céramique archaïque de l’Islam et ses Origines (Paris, 1920), limited in scope as it is to wares which were made before 1000 A.D., only eight of the 198 Islamic wares in that time-span depict Samanid epigraphic pottery. Clearly, for Western scholars its time had not yet come. Sheila Blair, in Text and Image, pp. 11–56, gives the fullest, most nuanced and best contextualised study of these ceramics to date.

27.Here the pioneer is Vera Kratchkovskaya, especially her article ‘Evoliutsiya Kūficheskogo pisma v Srednei Azii’, Epigrafika Vostoka III (1949), pp. 3–27. Pride of place goes to Oleg G. Bol’shakov, who in a series of articles took scholarship in this area to a new level; all were published in various issues of Epigrafika Vostoka – XII (1958), pp. 23–38; XV (1963), pp. 73–87; XVI (1963), pp. 35–55; XVII (1966), pp. 54–62; and XIX (1969), pp. 42–50 – under the same title: ‘Arabskie nadpisi na polivnoy keramike Sredney Azii IX-XII vv.’ He did not include the material from Nishapur in his analysis. Thereafter work has continued apace, thanks to studies by Vakturskaya, Lunina, Akhrarov, Uzmanova, Tashkhodzhaev, Saiko, Shishkina, Brusenko, Bryashimova, Stolyarova, Baipakov and Erzakovich, but these have been published only in Russian and in books and articles that are very difficult of access. It has therefore not made its full impact on Western scholarship in this field (for bibliographical details, see Jangar Y. Ilyasov, ‘Exotic images: on a new group of glazed pottery of the 10th and 11th century’, Journal of the David Collection iv [2014], pp. 85–7).

28.A precocious pioneering article by Lisa Volov (Golombek) (‘Plaited Kūfic on Samanid epigraphic pottery’, Ars Orientalis vi [1966], pp. 107–33) brought the challenges of these wares to a much wider circle of readers; it has remained a landmark in the field. Her work has been supplemented in Western languages by important contributions by Charles K. Wilkinson (Nīshāpūr: Pottery of the Early Islamic Period [New York, 1973], pp. 90–178); Grube, Cobalt and Lustre, pp. 51–3, 66–7, 76 – 80, 82–3, 86–7, 91, 94–6, 98–105 and 109; and Oliver Watson’s concise survey of the relevant material in the Al Sabah collection, which includes many intriguing insights (Ceramics, pp. 91 and 205–19).

29.Apart from Bol’shakov’s pioneering work, see Giovanna Ventrone, ‘Iscrizioni inedite su ceramica samanide in collezioni italiane’, in Maurizio Taddei (ed.), Gururājamañjarikā. Studi in onore di Giuseppe Tucci (Rome, 1974), pp. 221–32, and Oya Pancaroğlu, ‘Serving Wisdom: The contents of Samanid epigraphic pottery’, in Rochelle L. Kessler et al., Studies in Islamic and Later Indian Art (Cambridge, Mass., 2002), pp. 59–75 – an important step forward thanks to the categories which she proposes for the content of these wares. See too Oya Pancaroğlu, 2005, ‘Functions of literary epigraphy on medieval Islamic ceramics: Part One: Samanid epigraphic pottery’, http//islamicceramics. ashmolean.org/Samanids/oya-part-one-.htm, accessed 19 August 2014. This, besides offering a brief but carefully considered historical context for these wares, has a useful appendix of 41 proverbs, aphorisms and ḥadīths found on Samanid pottery. But the key work so far has indisputably been that of ‘Abdallah Ghouchani: Inscriptions on Nishabur Pottery, tr. Martin Charlesworth (Tehran, 1986), which brings together no less than 140 pieces of Samanid epigraphic ware, each presented in full-page plates (28 in colour) and – more important still – with an accompanying transcription and translation of each inscription.

30.Arthur Lane, Early Islamic Pottery (London, 1947), p. 18.

31.This is a topic in itself; for the argument that it mirrors an economic decline at the end of Samanid rule, see Djamal Mirzaaxmedov, ‘La production céramique du Maverannahr du IXe au début du XIIIe siècle’, in Étienne de la Vaissière (ed.), Islamisation de l’Asie Centrale. Processus Locaux d’Acculturation du VIIe au XIe Siècle (Paris, 2008), pp. 64–8.

32.It is noticeable that Bol’shakov’s parameters for the Afrasiyab material are the ninth to twelfth centuries, when in fact Samanid rule ended with the death of Ismā‘īl II b. Nūḥ in 395/1005.

33.See Bol’shakov, ‘Nadpisi’, XII, 1958, p. 26, quoting Vera Kratchkovskaya to the effect that the wares with ‘simple’ Kūfic date to the ninth century.

34.Bol’shakov, ‘Nadpisi’, XII, 1958, p. 32.

35.Bol’shakov, ‘Nadpisi’, XVI, 1963, pp. 38–9 and 43.

36.Such as the bowl which bears, in defiance of all norms long established for these wares, a central inscription with both a Qurʾānic text and a date (300 A.H.); see Géza Fehérvári, Ceramics of the Islamic World in the Tareq Rajab Museum (London and New York, 2000), pp. 56–7. This piece is no longer on display.

37.Kurt Erdmann notes that in a group of 20,000 miscellaneous sherds found in Afrāsiyāb and sent to Berlin in the early twentieth century there were ‘not less than eighty wasters’ of tenth-century ware, plus several hundred whose glaze was defective, whereas Nishapur produced little evidence of local production (‘Afrasiab ceramic wares’, Bulletin of the Iranian Institute vi/1–4 and vii/1 [1946], pp. 102–3). He adds (p. 107) that the sherds included some 6,000 fragments of the epigraphic wares in black on a white ground, mainly bowls and plates. However, in a recent joint French-Iranian venture, excavations in the citadel of Nishapur yielded 67 sherds which petrographical analysis revealed to be of local manufacture. See Rocco Rante and Annabelle Colinet (with contributions by Rajabali Labbaf Khaniki, A. Bouquillon, Y. Coquinot, C. Doublet, Y. Gallet, A. Genevey, E. Porto and A. Zink), Nishapur Revisited: Stratigraphy and Pottery of the Qohandez (Oxford, 2013).

38.Galina A. Pugachenkova and Eduard V. Rtveladze, ‘Afrāsīāb’, EIr, i, p. 577b.

39.M. Yusuf Kiani, The Islamic City of Gurgan (Archaeologische Mitteilungen aus Iran. Ergänzungsband 11) (Berlin, 1984), pp. 42–3; but he does not illustrate these wares (his pl.39 does not correspond to his description of them).

40.Andrew Williamson, ‘Regional distribution of mediaeval Persian pottery in the light of recent investigations’, in James Allan and Caroline Roberts (eds), Syria and Iran. Three Studies in Medieval Ceramics (Oxford Studies in Islamic Art, IV) (Oxford, 1987), p. 18 (on p. 19 he mentions related wares at Jiruft and Zarang, but does not specify that they used epigraphic decoration; see too his Map 5 on p. 20).

41.The production of epigraphic ware outside Afrasiyab and Nishapur has been highlighted by Saida Ilyasova and her colleagues (Saida Ilyasova and Nadeshda Wischnewskaya, ‘Glasierte Keramik von Binket (Taschkent) aus der Sammlung des Staatlichen Museums für Orientalische Kunst’, Tribus li [2002], pp. 114–26 and Saida Ilyasova and Rawschan Imamberdyev, ‘Eine Sammlung glasierter Keramik aus Taschkent’, Tribus liv [2005], pp. 91–102) and, most recently, Jangar Y. Ilyasov (‘Exotic images’, pp. 50–87).

42.Bol’shakov, ‘Nadpisi’, XIX, 1969, p. 47.

43.Bol’shakov, ‘Nadpisi’, XII, 1958, p. 29; Bol’shakov, ‘Nadpisi’, XV, 1963, p. 76.

44.Bol’shakov, ‘Nadpisi’, XII, 1958, p. 35.

45.Mirzaaxmedov, ‘Production’, p. 90 and pl.III-5, 6 and 7. See now Christina M. Henshaw, Early Islamic Ceramics and Glazes of Akhsiket, Uzbekistan (unpublished PhD dissertation, University College, London, 2010), pp. 67-71.

46.Karl Baipakov and Lev Erzakovich, Keramika srednevekovogo Otrara – Ceramics of Medieval Otrar (Alma-Ata 1991), pls. 105, 108, 111, 114–5, 119 and 126.

47.Maria Baskhanova, ‘Flowers, calligraphy and the potter’s wheel. Glazed ceramics from the Mawarannahr’, in Mikhail Baskhanov, Maria Baskhanova, Pavel Petrov and Nikolaj Serikoff, Arts from the Land of Timur. An Exhibition from a Scottish Private Collection (Paisley, 2012), p. 169, item 350. Cf. also the Samanid ceramics of Khulbuk in Tajikistan; see M. Pierre Siméon, Étude du materiel de Hulbuk (Mā wārā al-nahr Khuttal) de la conquête islamique jusqu’au milieu du XIe siècle (90/712-441/1050). Contribution à l’étude de la céramique d’Asie centrale (Oxford, 2009), pp. 226–7 and 322–6.

48.Peter B. Golden, ‘The Karakhanids and early Islam’, in Denis Sinor (ed.), The Cambridge History of Early Inner Asia (Cambridge, 1990), pp. 354–61; Omeljan Pritsak, ‘Die Karachaniden’, Der Islam xxxi (1953–4), pp. 25–34.

49.The most accessible survey outside the rich Russian literature (see especially the work of Elena A. Davidovich) is the useful but seriously outdated one by Richard Vasmer, ‘Zur Münzkunde der Qarahaniden’, Mitteilungen des Seminars für Orientalische Sprachen xxxiii (1930), pp. 83–104. See also the brief remarks of George C. Miles, ‘Numismatics’, in Richard N. Frye (ed.), The Cambridge History of Iran. Vol. IV: From the Arab Invasion to the Saljuqs (Cambridge, 1975), pp. 376–7 and pl.29/2–3. For a representative set of images in colour, see Bashkhanova, Arts, pp. 313–17, items 596–616.

50.Yury Karev, ‘Qarakhanid wall paintings in the citadel of Samarqand: First report and preliminary observations’, Muqarnas xxii (2005), pp. 66 and 69 and figs.26–8.

51.Spuler, Iran, p. 281.

52.Venetia Porter once told me that a group of enthusiasts with whom she read inscriptions was dubbed ‘The Headache Club’.

53.Thus the inscription on a splendid dish in the Keir collection has been read both as ‘When a creature is overcome by misfortune [he calls upon his Lord]’, a quotation from Qurʾan 39: 8 and 50 (Ernst J. Grube, Islamic Pottery of the Eighth to the Fifteenth Century in the Keir Collection [London, 1976], p. 97) and as ‘The free man is still free even if touched by harm’ (Ghouchani, Inscriptions, p. 90). The latter reading is preferable.

54.Ghouchani, Inscriptions, pp. 168–9.

55.Wilkinson notes (‘Glazed pottery’, p. 113) that this inscription indicates a Nīshāpūri provenance; see too Blair, Text and Image, p. 34.

56.As David Storm Rice believed (Bol’shakov, ‘Nadpisi’, XVII, 1966, p. 58).

57.See Don Aanavi, ‘Islamic Pseudo Inscriptions’, unpublished PhD thesis, Columbia University, 1969, pp. 77–8, for the suggestion that it may have mystical significance. On the other hand, the word Aḥmad preceded by ‘amal does occur on Abbasid blue and white ware (Ghouchani, Inscriptions, pp. 136–7).

58.For a recent discussion of this problem, see Watson, Ceramics, p. 211.

59.Ventrone, ‘Iscrizioni’, pp. 224–5.

60.This is especially clear in the case of wares with interlaced abstract designs; see Sharip S. Tashkhodzhaev, Khudozhestvennaya Polivnaya Keramika Samarkanda IX – nachala XIII vv. (Tashkent, 1967), Fig. 32.

61.For a useful survey of most of the major types, with an emphasis on the Central Asian material, see Baskhanova, ‘Flowers’, pp. 322–8. For the corresponding case of Nishapur, see Wilkinson, Nishapur Pottery.

62.Cf. Oleg Grabar, ‘Notes on the decorative composition of a bowl from Northeastern Iran’, in Richard Ettinghausen (ed.), Islamic Art in The Metropolitan Museum of Art (New York, 1972), p. 96.

63.By Kamoliddin, ‘Ancestors’, §12. See Tashkhodzhaev, Keramika, Figs 21, top right and left and bottom left; 25, top right and left and bottom left; 35, bottom right; 37, centre; 38, bottom left; pl.4, centre right; pl.6, bottom left; pl.11, centre right; and pl.12, top left.

64.For example, Eric J. Zetterquist, Persian Ceramics and Related Materials, (New York and London, 1993), pl.1 and 35th unpaginated page; Wilkinson, Nishapur: Pottery, pp. 170–1; cf. also Tashkhodzhaev, Keramika, Fig. 27.

65.Grube, Cobalt and Lustre, p. 96.

66.For heavy palmettes of this kind, see Tashkhodzhaev, Keramika, Fig. 23 and Fig. 24 top right and left.

67.See also Ghouchani, Inscriptions, pp. 85, 95, 113, 115, 127, 143, 153, 173, 177, 193, 207, 227, 245, 251, 271 and 295.

68.Vera C. Tamari, ‘Ninth-Tenth Century White Mesopotamian Ceramic Ware with Blue Decoration’, Oxford University, MPhil. thesis, 1984, pp. 37–51.

69.Ralph H. Pinder-Wilson, ‘A lustre relief dish of the early Islamic period’, British Museum Quarterly xxvii/3–4 (1963–4), pp. 91–4; for a colour plate, see Carel J. Du Ry, Art of Islam, tr. Alexis Brown (Baden-Baden, 1970), p. 43.

70.Irène Momtaz, Momtaz Islamic Art. Ornament and inscription, ed. Ralph Pinder-Wilson (London, 2004), pp. 12–13.

71.See Balbina Martinez Caviró, Cerámica Hispanomusulmana Andalusi y Mudéjar (Madrid, 1991), p. 72.

72.Bol’shakov, ‘Nadpisi’, XII, 1958, p. 25, Fig. 1.

73.Bol’shakov, ‘Nadpisi’, XII, 1958, pp. 24 and 26–7 and Fig. 2; Ilyasova and Wischnewskaya, ‘Binket’ (=Binkath), pp. 121 and 124 and Fig. 5.

74.For a piece of this kind in the Seattle Art Museum, see Irene Bierman, ‘Near East Gallery’, Arts of Asia xxii/3 (May-June 1992), pp. 120 and 122, which twice repeats an invocation to God – yā huwa – yet where the third repetition should be (between three and seven o’clock) the text is different, with an alif, long ya and waw. This section may not belong with the rest of the bowl.

75.Ghouchani, Inscriptions, pp. 184–5; here the motif of knotting is so obsessively employed that some letters resemble branches of trees laden with fruit.

76.As in the façade of the Court of the Lions in Granada, whose ‘rythme subtil’ is analysed by Georges Marçais, ‘Remarques sur l’esthétique musulmane’, Mélanges d’Histoire et d’Archéologie de l’Occident Musulman. Tome I. Articles et Conférences de Georges Marçais (Algiers, 1957), pp. 99–102.

77.Bashkhanov et al., Land of Timur, plate on p. 122.

78.Thus in the proverb man kathura kalāmuha kathura saqṭuhu (‘He who speaks too many words has too many faults’), the admonitory kathura is at twelve o’clock and six o’clock of the bowl (Ventrone, ‘Iscrizioni’, pp. 227–8).

79.Aanavi, Pseudo Inscriptions , pp. 92–3; Lane, Pottery, pl.19b.

80.Aanavi, Pseudo Inscriptions , pp. 58, 69–70, 74–78; Don Aanavi, ‘Devotional writing: “Pseudoinscriptions” in Islamic Art’, The Metropolitan Museum of Art Bulletin (May, 1968), pp. 352–8.

81.Arthur Lane, Islamic Pottery from the Ninth to the Fourteenth Centuries A.D. in the Collection of Sir Eldred Hitchcock (London, 1956), pp. 12 and 19–20 and pl.4 (al-‘afw taḥta lisānihi / baraka wa khayr, tentatively rendered as ‘Forgiveness under his tongue. Blessing and well-being’). See Ventrone, ‘Iscrizioni’, p. 229.

82.Ghouchani, Inscriptions, pp. 124–5. Note how the single word baraka at twelve o’clock and six o’clock contrives to bring all the empty space into play.

83.Ibid, Inscriptions, pp. 84–5.

84.Ibid, Inscriptions, pp. 36–7.

85.For another example, tajuz barran (‘may you be well rewarded’), see Oya Pancaroğlu, Perpetual Glory. Medieval Islamic Ceramics from the Harvey B. Plotnick Collection (Chicago, 2007), pp. 64–5.

86.Ghouchani, Inscriptions, pp. 142–3.

87.Du Ry, Art, p. 139.

88.As in the case of the ṣād of ṣabra in Fig. 15. For further comments on puns in Samanid wares see Blair, Text and Image, pp. 33–4.

89.Arne A. Ambros, ‘Beobachtungen zu Aufbau und Funktionen der gereimten klassischarabischen Buchtitel’, Wiener Zeitschrift fur die Kunde des Morgenlandes 80 (1990), pp. 13–57.

90.Sometimes there are two such devices, as in the celebrated Freer bowl (Blair, Text and Image, p. 15).

91.Baskhanov et al., Land of Timur, 127, item 255.

92.Grube, Cobalt and Lustre, pp. 100–101.

93.Bol’shakov, ‘Nadpisi’, XII, 1958, p. 27, Fig. 3.

94.Ghouchani, Inscriptions, pp. 86–7. Note how the central image of the bird is mimicked in the ascenders of the inscription.

95.‘Abdallah Ghouchani Rey Hoard of Nishapur Dinars (Tehran, 1383), passim.

96.Bol’shakov. ‘Nadpisi’, XII, 1958, p. 25 notes that the ratio of high to low letters may rise to 6:1.

97.Tamari, Ceramic Ware, pls.58–64, with 28 figures.

98.Volov, ‘Plaited Kūfic’, pp. 121–6.

99.As shown by Volov, ‘Plaited Kūfic’, pp. 119–20, 129–30, Fig. 5 and pl.6.

100.Margaret S. Graves and Benoît Junot (eds), Treasures of the Aga Khan Museum – Arts of the Book & Calligraphy (n.p. , 2010), p. 33, cat. no.9.

101.See the diagrams in Tamari, Ceramic Ware, p. 43. Note that not a single piece has its inscription round the inner rim or on the exterior of the piece. In both these ways, therefore, the Samanid wares offered major innovations. On the other hand, no Samanid piece takes up the arrangement of an inscription in three lines, one above the other, at the centre of the pot, as in Abbasid blue and white ware.

102.Tamari, Ceramic Ware, two-page list between plates 57 and 58.

103.Yet there is no parallel for the design of a Bukharan fals of 353H. whose central inscription (an abbreviated shahāda) is arranged centrifugally around a circle, while the outer inscription gives the bismillāh, the mint and the date (Richard Plant, Arabic Coins and How to Read Them [London, 1980], p. 44).

104.The most famous example of this text is on the celebrated Louvre dish (Fig. 2); it recurs on five Samanid ceramics in Russian collections alone, though with minor variations in the text (Bol’shakov, ‘Nadpisi’, XVI, 1963, pp. 36–7).

105.The earliest discussion of these terms in the context of eastern Islamic epigraphic pottery is in Samuel Flury, ‘Une formule épigraphique de la céramique archaïque de l’Islam’, Syria V (1924), pp. 53–62.

106.Tamari, Ceramic Ware, p. 41.

107.Volov, ‘Plaited Kūfic’, especially pp. 109–15. Her analysis has stood the test of time.

108.E.g. a bowl inscribed al-anāt qabl al-ra’y (‘deliberation before [voicing] opinion’); see Irène Momtaz, Memories from Islamic Lands (London, n.d.), pp. 22–3 and 222.

109.For a good colour plate, see Giuseppe Scavizzi, Maioliche dell’Islam e del Medioevo occidentale (Milan, 1966), pp. 20–1 and pl.7.

110.Volov, ‘Plaited Kūfic’, pp. 119–20 and 129–30, text Fig. 5 and pl.6, Figs 13–14. These coins are from Rayy.

111.‘Abdallah Ghouchani has published a substantial hoard of medieval dīnārs minted in Nishapur: Rey Hoard. Out of 296 Samanid-period dīnārs minted in Nishapur and belonging to a hoard found in Rayy, only two used plaiting. The great majority used mashq for the final dāl of Muḥammad, and other epigraphic flourishes include a forked terminal nūn (p. 281, no.191, 315H.) and a hanging hook for terminal rā’ or nūn (p. 285, no.205, 321H.). Interestingly enough, the use of plaiting in the dāl of Muḥammad which does occur is very modest; it is found in two dīnārs in the name of Abu ‘Alī b. Sīmjūr (Ghouchani, Rey Hoard, p. 262, nos. 435 and 436, both of 393H). The conclusion that imposes itself here is that, so far as plaiting is concerned, there is practically no correlation between the Samanid gold dīnārs of Nishapur and the epigraphic wares produced there. It may be that the larger space available on silver dirhams may point to different conclusions, but the huge empty flans of Samanid macrodirhams and their relatively cramped interior design suggest otherwise (Gilles Hennequin, ‘Macrodirhams sāmānides inédits (collection particulière), Annales Islamologiques 20 [1984], pls.XXXV-XL and Gilles Hennequin, ‘Grandes monnaies Sāmānides et Ghaznavides de l’Hindū Kush, 331–421 A.H. – Étude numismatique et historique’, Annales Islamologiques 9 (1970), pls.XII-XVI; none of the smaller dirhams that he discusses uses plaiting). Flury, ‘Formule’, speculates that plaiting originated in the eastern Islamic world, and analyses an example of a plaited kāf on a tenth-century vase attributed to Rayy (‘Formule’, pp. 58–9 and Fig.3).

112.Mikhail Zand, ‘Some light on bilingualism in literature of Transoxiana, Khurasan and Western Iran in the 10th century A.D.’, in Jiři Bečka (ed.), Yādnāme-ye Jan Rypka (Articles on Persian and Tajik Literature) (Prague, 1967), p. 162, speaks of a ‘unified literature in two complementary languages’ whose authors and readers were bilingual. Examining those poets who produced work in both languages, he makes a particular distinction (p. 164) between who are known especially for their Persian verse (Shahīd, Kisā’ī and Āghājī) and those noted especially for their work in Arabic (Qābūs and Abu’l-Fatḥ al-Bustī). Cf. Richard N. Frye, ‘Development of Persian literature under the Samanids and Qarakhanids’ in Bečka, Yādnāme, pp. 70–3. Cf. n.15 above.

113.For a clear step-by-step description of how these ceramics were made, see Blair, Text and Image, pp. 12–13.

114.Cf. Shishkina and Pavchinskaja, Terres secrètes, pp. 54 and 56.

115.For the suggestion that the ‘drawing of the letters was refined with a sharp instrument’, see Barbara Brend, Islamic Art (London, 1991), p. 88. Cf. Bol’shakov, ‘Nadpisi’, XV, 1963, p. 76.

116.I am grateful to Dr John Curtis for this information. Barbara Brend makes a similar suggestion in her analysis of a splashed-ware bowl from Nīshāpūr (Islamic Art, p. 86).

117.Momtaz, Memories, pp. 22–3; for other examples, see Ghouchani, Inscriptions, pls.10, 19, 62, 65, 85, 96, 98–9, 117–8, 136, 138 and 140.

118.This can readily be observed in the minor adjustments made line by line in the celebrated Qurʾān copied by Ibn al-Bawwāb; see, for example, David S. Rice, The Unique Ibn al-Bawwāb Manuscript in the Chester Beatty Library (Dublin, 1955), pls.VI-VIII and Elaine Wright, Islam. Faith. Art. Culture. Manuscripts of the Chester Beatty Library (London, 2009), pp. 128–30.

119.Bol’shakov, ‘Nadpisi’, XII, 1958, p. 30.

120.Bol’shakov, ‘Nadpisi’, XV, 1963, p. 75.

121.For examples of this see Bol’shakov, ‘Nadpisi’, XVI, 1963, pp. 45–6, with a discussion of the progressive distortion of letters that resulted from frequent copying of a text by those who did not understand it.

122.See Richard W. Bulliet, ‘Pottery styles and social status in medieval Khurasan’, in A Bernard Knapp (ed.), Archaeology, Annales and Ethnohistory (Cambridge, 1992), pp. 80–1, for similar conclusions based on proposed dating of this material and on the kind of food served in many buff ware bowls.

123.Pace the arguments put forward by Mirzaaxmedov, ‘Production’, pp. 61–2, which gloss over the implications of the sheer quantity of sherds of this ware that have come to light.

124.Blair, Text and Image, pp. 40–1; her reading seems preferable to the ‘‘Azīz’ mentioned by Mirzaaxmedov, ‘Production’, p. 63.

125.Blair, Text and Image, pp. 40 and 42.

126.Mirzaaxmedov, ‘Production’, p. 63.

127.Momtaz, Memories, pp. 40–1 and 222.

128.Bol’shakov, ‘Nadpisi’, XII, 1958, pp. 31–4; Bol’shakov, ‘Nadpisi’, XV, 1963, pp. 79–82. He ascribes a date in the eleventh or twelfth century to most of them (p. 79); the spelling of the word yumn is often defective, perhaps in the interests of symmetry (Bol’shakov, ‘Nadpisi’, XV, 1963, p. 82).

129.Bulliet, ‘Pottery’, p. 81.

130.Thomas Leisten, Architektur für Tote. Bestattung in architektonischem Kontext in den Kernländern der islamischen Welt zwischen 3./9. und 6./12. Jahrhundert. Materialien zur Iranischen Archäologie, Band IV (Berlin, 1998), pp. 34 and 164.

131.Lisa Golombek and Donald Wilber, The Timurid Architecture of Iran and Turan (Princeton, 1988), i, pp. 233–4. Cf. Oleg Grabar, ‘The earliest Islamic commemorative structures, notes and documents’, Ars Orientalis vi (1966), p. 39. A study by Dr Venetia Porter on medieval Central Asia mausolea attributed to other Companions is in progress.

132.Arthur U. Pope, ‘Discoveries at Harun ar-Rashid’s birthplace’, The Illustrated London News, clxxxvi (22 June 1935), pp. 1122–3.

133.For the pre-restoration state of the Chīnīkhāna in Ardabīl, see Friedrich Sarre, Denkmäler Persischer Baukunst (Berlin, 1910), pp. 41–4, figs. 41–2 and pl.LII; for its current appearance, see Kishwar Rizvi, The Safavid Dynastic Shrine. Architecture, Religion and Power in Early Safavid Iran (London and New York, 2011), pp. 143–55 and colour pl.17.

134.Sussan Babaie, Isfahan and its Palaces. Statecraft, Shi‘ism and the Architecture of Conviviality in Early Modern Iran (Edinburgh, 2008), p. 148 and colour pl.13; Rizvi, Shrine, colour pl.18.

135.Grube, Cobalt and Lustre, p. 78, notes five examples of the saying ‘Greed is a sign of poverty’.

136.Ghouchani, Inscriptions, pp. 144–5.

137.Ibid, Inscriptions, pp. 94–5.

138.Ibid, Inscriptions, pp. 118–9.

139.Ibid, Inscriptions, pp. 28–9.

140.Ibid, Inscriptions, pp. 58–9.

141.For a colour plate, see Du Ry, Art of Islam, p. 59.

142.Tamari, Ceramic Ware, p. 44 and pl.4: kul hani’an wa mali’an, “eat with pleasure until you are full”. An almost identical formula occurs on Samanid wares and also on medieval metal spoons (Bol’shakov, ‘Nadpisi’, XV, 1963, pp. 85–7 and Fig. 16). Cf. Ghouchani; Inscriptions, pp. 94–5.

143.Bernard O’Kane, The Appearance of Persian on Islamic Art (New York, 2009), pp. 117–34; for earlier examples in Central Asia and Afghanistan, see pp. 17 and 20.

144.For two exceptions, see Blair, Text and Image, p. 40 and Figs 2.18 and 2.19.

145.Pancaroğlu, ‘Serving wisdom’, pp. 59–75.

146.Margaret Malamud, ‘Sufi organizations and structures of authority in medieval Nīshāpūr’, International Journal of Middle East Studies 26 (1994), pp. 436–8.

147.Reuben Levy, A Baghdad Chronicle (repr. Philadelphia, 1977), pp. 159–60 and 168.

148.That is to say, tiles whose location (as distinct from their content) has a Shī‘ite connection. Their content, however, in the large majority of cases, has nothing distinctively Shi‘ite about it. See O. Watson, Persian Lustre Wares (London, 1985), pp. 150–6.

149.R. Bulliet, The Patricians of Nishapur. A Study in Medieval Islamic Social History (Cambridge, Mass., 1972), pp. 14–15, 30 and, for the case of a single ‘Alid family, pp. 234–45.

150.Bosworth, The Ghaznavids, pp. 194–200, gives the situation in immediately post-Samanid times (but cf. p. 166 on the situation c.980); C. Edmund Bosworth, ‘Nīshāpūr. i. Historical Geography and History to the Beginning of the 20th Century’, Encyclopaedia Iranica online (http://www.iranicaonline.org/articles/nishapur-i) (last updated September 17, 2010), accessed 13 August 2014.

151.Pancaroğlu, ‘Serving wisdom’, pp. 72–5.

152.Bol’shakov, ‘Nadpisi’, XVI, 1963, pp. 52–5, records nine versions of this saying, including variants, in Russian collections. It is attributed to ‘Ali.

153.Bol’shakov, ‘Nadpisi’, XVI, 1963, pp. 46–52, records fifteen examples of this aphorism in Russian collections.

154.This saying has been recorded on 35 objects (Grube, Cobalt and Lustre, p. 76). Bol’shakov records 24 of them, including minor variants in the text, in Russian collections (Bol’shakov, ‘Nadpisi’, XVI, 1963, pp. 37–46).

155.For a slightly different translation of this inscription, see Volov, ‘Plaited Kūfic’, p. 133.

156.Volov, ‘Plaited Kūfic’, p. 117.

157.Accession number 52.11; see Esin Atıl, Ceramics from the World of Islam (Washington, D.C., 1973), pp. 28–9 and Bol’shakov, ‘Nadpisi’, XIX, 1969, pp. 48–50.

158.Ghouchani, Inscriptions, pp. 72–3. This piece, though attributed to Nishapur, has much in common with Abbasid blue and white wares.

159.Cf. the story with that title by W. Somerset Maugham (1933), itself reworking a tale found in the Babylonian Talmud.

160.Grube, Cobalt and Lustre, p. 77 (a misplaced ‘he said’ – why not yuqāl? And qāla should begin a quotation, not be tacked on at the end) and p. 78 (‘Be pious, and’). The way that inscriptions end with something unresolved has perhaps a teasing quality. It may of course be simple incompetence, but that sits ill with the finished appearance of these inscriptions. This does not happen with inscriptions on buildings or coins. But on Samanid wares these inscriptions, whether unresolved or simply unfinished, seem to have something of a sting in their tail, and this occurs too often to be an accident. It may be intentionally humorous. Cf. Volov, pp. 118 and 128, n.21 and Blair, Text and Image, pp. 33–4.

161.Baipakov and Erzakovich, Ceramics, pl.117.

162.Sheila S. Blair, Islamic Calligraphy (Edinburgh, 2006), pp. 145–57.

163.Ibid, Calligraphy , pp. 157–67.

164.For a preliminary discussion of this topic in the context of pottery, see Marilyn Jenkins, ‘Muslim: an early Fatimid ceramist’, The Metropolitan Museum of Art Bulletin (May, 1968), pp. 359–69.

165.Especially for what Bol’shakov calls ‘ceramic cursive’ (‘Nadpisi’, XII, 1958, p. 24).

166.As suggested by Julian Raby, ‘Looking for silver in clay: a new perspective on Samanid ceramics’, in Michael Vickers (ed.), Pots & Pans. A Colloquium on Precious Metals and Ceramics in the Muslim, Chinese and Graeco-Roman Worlds, Oxford 1985 (Oxford Studies in Islamic Art, III) (Oxford, 1986), p. 187.

167.Some of the many uses of the red dot are outlined by Y. Dutton, ‘Red dots, green dots, yellow dots and blue: some reflections on the vocalisation of early Qur’anic manuscripts: Part I’, Journal of Qur’anic Studies i/1 (1999), 115, 117–20 and 123 and ‘Red dots, green dots, yellow dots and blue: some reflections on the vocalisation of early Qur’anic manuscripts: Part II’, Journal of Qur’anic Studies ii/1 (2000), pp. 11–12.

168.As noted by Richard Ettinghausen, ‘Abrī painting’, in Miriam Rosen-Ayalon (ed.), Studies in Memory of Gaston Wiet (Jerusalem, 1977), p. 350; but he adds (p. 351, n.22) that the beginnings of this idea can be traced back to far earlier Qurʾāns, such as an example attributed to ninth-century Baghdad (Annemarie Schimmel, Islamic Calligraphy [Leiden, 1970], pl.IVa). For a colour plate of the Chester Beatty colophon, see Wright, Islam, p. 131.

169.For the history of this technique, see Ettinghausen, ‘Abrī painting’, pp. 345–56. For its connection with metalwork, see Bol’shakov, ‘Nadpisi’, XII, 1958, p. 28 and Raby, ‘Silver’, p. 199.

170.Nan B. Freeman and Walter B. Denny, Ebrû Art: Marble on Paper, The Work of Feridun Özgören (Bahrain, 2001).

171.Bol’shakov, ‘Nadpisi’, XIX, 1969, p. 46.

172.Bol’shakov, ‘Nadpisi’, XV, 1963, pp. 73–9.

173.This is perhaps especially the case when abbreviations are used; baraka is in itself an abbreviation, for the blessing comes from Allāh (the complete form of the commonest inscription would be barakat Allāh li-ṣāḥibihi), and another example is the word takfa’ which is an abbreviation of tawakkul takfa’ (‘trust [in Allāh]: you will be rewarded’). See Bol’shakov, ‘Nadpisi’, XV, 1963, pp. 82 and 85, and, for a further example, so abbreviated as to be meaningless, Bol’shakov, ‘Nadpisi’, XVII, 1966, p. 61.

174.Note, for instance, two examples of the seal of Solomon (khatam Sulaymān): Tashkhodzhaev, Keramika, p. 60, Fig. 12.

175.Raby, ‘Silver’, pp. 179–203.

176.Ibid, pp. 188, Figs 11–12; 189–90; 191 and 194, Figs 22–3; and 195, Figs 24–7 and 198–9.

177.See Jessica Rawson, ‘Chinese silver and its influence on porcelain development’, Cross-Craft and Cross-Cultural Interactions iv (1989), pp. 275-99 and Raby, ‘Silver’, pp. 191-8.

178.Frye, ‘Samanids’, p. 148.

179.For a detailed account of this theme, of which the Chinese chronicles record several examples over many centuries, see Toh Sugimura, The Encounter of Persia with China. Research into Cultural Contacts based on Fifteenth Century Persian Pictorial Materials, in Senri Ethnological Studies, No.18 (Osaka, 1986), pp. 71–131, especially pp. 114–19, an analysis of a famous fifteenth-century painting in Istanbul; for a good colour plate, see David J. Roxburgh (ed.), Turks. A Journey of a Thousand Years, 600–1600 (London, 2005), pl. on p. 252 and p. 431. For a discussion of such a marriage between a Chinese princess and a Bukharan or Eastern Turkish monarch within the period 572–626 see Josef Markwart, Wehrot und Arang. Untersuchungen zur mythischen und geschichtlichen Landeskunde von Ostiran (Leiden, 1938), p. 152 and n.3.

180.See Minorsky’s translation of the older preface to the Shāhnāma, which mentions that Naṣr b. Aḥmad asked his minister Khwāja Bal‘amī to translate Ibn al-Muqaffa‘’s Arabic text of the Kalila wa Dimna into Persian; that done, Rūdakī turned it into verse and ‘the Chinese added images to it so that the seeing and reading of it should please everybody’ (Vladimir Minorsky, ‘The older preface to the Shāh-Nāma’, repr. in Iranica [Tehran, 1964], p. 266 and n.3). Cf. Frye, Bukhara, p. 83.

181.Browne, Literary History, i, p. 454, quoting Abu Shu‘ayb Ṣāliḥ b. Muḥammad of Herat.

182.Richard N. Frye, The History of Bukhara: Translated from a Persian Abridgement of the Arabic Original by Narshakhī (Cambridge, MA, 1954), p. 109; for example, a certain Abū Dulaf acted as an ambassador to China for the amīr Naṣr and wrote up his travels (Frye, ‘Samanids’, p. 143).

183.See Edward H. Schafer, The Golden Peaches of Samarkand (Berkeley, CA, 1963).

184.Yuka Kadoi, ‘Translating from jing to mir’at/a’ina: medieval Islamic mirrors revisited’, Art in Translation v/2 (2013), pp. 256–9 (online journal: http://www.bloomsbury.com/uk/journal/art-in-translation/), accessed 22 August 2014.

185.Lane, Early Islamic Pottery, p. 10.

186.Bol’shakov, ‘Nadpisi’, XII, 1958, p. 29 (the word baraka).

187.Wilkinson, Nishapur Pottery, pp. 54–89.

188.Erdmann, ‘Afrasiab’, p. 109.

189.Oliver Watson, ‘Chinese-Iranian Relations XI. Mutual influence of Chinese and Persian ceramics’, in Ehsan Yarshater (ed.), Encyclopaedia Iranica V (Costa Mesa, CA, 1992), pp. 455a-457b. The list of exports to the Chinese court from Central Asia in the 7th-10th centuries assembled by Schafer is astonishing. It includes ceramics (p. 4), Turkish slaves (p. 44), pygmies (p. 48), music (p. 52), actors, flautists and oboists (p. 54), marionette plays (p. 54), male and female dancers (pp. 54- 5 and 106), including the much sought-after ‘Western Twirling Girls’ (p. 56), blood-sweating horses (p. 61), Arab horses (p. 64), camels and cameleers (p. 71), fat-tailed sheep (p. 75), hunting dogs (p. 77), lions (p. 85), cheetahs (pp. 87-8), dyed red deerskins and boots (p. 106), saffron (p. 125), wine (p. 144), the astringent myrobalan (p. 145), fragrant drugs (p. 148), dance mats (p. 148), embroidered carpets (p. 148), sugar (‘stone honey’) (p.153), cotton (p. 201), “hair brocade’ (p. 202), indigo (p. 212), black salt (p. 217), gems (p. 222), a jade finger ring (p. 226), carnelian (p. 228), lapis lazuli (p. 231), gold (p. 254) and silver (p. 256), ostrich-egg cups (p. 258), an incense brazier (p. 259), a jewelled couch (p. 259), and chain mail (p. 261). Cf. John Guy, ‘Rare and strange goods. International trade in ninth-century Asia’, in Regina Krahl, John Guy, J. Keith Wilson and Julian Raby (eds), Shipwrecked. Tang Treasures and Monsoon Winds (Washington, D.C., 2010), pp. 19-27.

190.Hsueh-man Shen, ‘Chinese polychromes in the Indian Ocean trade during the 9th century’, in Lorenz Korn (ed.), Beiträge zur Islamischen Kunst und Archäologie IV (Wiesbaden, 2014), p. 111 and Fig. 3. Here the ‘Arabic’ inscription is so garbled that it is illegible. It is a pleasure to acknowledge the help that I have received from Dr Shen.

191.See Wu Yuejian, Tang feng miaocai: Changsha yao jingpin zhan (Marvelous Color and Tang Style: Exquisite Wares of Changsha Kiln) (Changsha, 2008), I, p. 272. The inscription reads: ‘The government of Tanzhou [in Hunan province] approved for [sic] the reconstruction of the Daolin temple. There were 2,500 donors, each sponsoring 1,000 characters, for the production of 5,000 volumes of the Buddhist sutras. The sutras were preserved at the Yihe stupa. The completion of the sutras granted common happiness. Monk Shuyan reported this in the seventh month of the 12th year of the Dazhong reign’ [Tang dynasty, 859 A.D.]. I am most grateful to my friend Dr Anita Chung for this reference.

192.Shen, ‘Familiar differences’, p. 121, Fig. 6.

193.For further information on these links, see Krahl et al. (eds), Shipwrecked, pp. 4, 27, 69, 118, 170 and 187.

194.William Watson, The Arts of China 900–1620 (New Haven and London, 2000), pp. 35 and 38–9.

195.For example, a bowl from Northern China, datable 1100–1200, with an inscription in a triple concentric circle in brown slip on white slip beneath a clear glaze (Oxford, Ashmolean Museum, Ingram gift EA 1956.1313).

196.Such as ‘Speak little in a crowd; if it’s no business of yours then go home’ (Li Zhiyan, Virginia L. Bower and He Li [eds], Chinese Ceramics From the Paleolithic Period through the Qing Dynasty (New Haven and London, 2010), pp. 288–9). Other wares of the Song period also bore inscriptions, sometimes lengthy ones, but on the base of the piece and therefore out of sight (Chin Hsiao-yi, China at the Inception of the Second Millennium. Art and Culture of the Sung Dynasty, 960–1279 [Taipei, 2000], pp. 146–7, 151 and 159).

197.Thus a somewhat later wine bottle in the British Museum bears an inscription in Phagspa script proclaiming ‘a bottle of good wine’. The language is not that of a modern wine buff but the meaning comes across clearly enough. I owe this reference to the kindness of Dr Anita Chung.

198.Charleston, World Ceramics, p. 49, Fig. 132.

199.Aanavi, Islamic Pseudo Inscriptions, p. 96 (see p. 85 for a possible example of pseudo-Hebrew script on Samanid ware); ‘Devotional writing’, p. 355.

200.Such as two examples of glazed Changsha stoneware of the ninth century, namely a money box and a covered box from Hunan province (Wu Yuejian, Marvelous Color, i, pp. 275 and 280). I am most grateful to my friend Dr Anita Chung for this reference.

201.Such as ‘modesty is a branch of faith’; this occurs in various forms, and is a saying attributed to Muḥammad (Bol’shakov, ‘Nadpisi’, XIX, 1969, pp. 42–4; on p. 46 he mentions a similar utterance found – always in distorted form – in more than twenty examples. These distortions betray a consistent lack of understanding of what the text says.

Bibliography

Aanavi, Don, ‘Islamic Pseudo Inscriptions’, Unpublished Ph.D. thesis, Columbia University, 1969.

Aanavi, Don, ‘Devotional writing: “Pseudoinscriptions” in Islamic art’, The Metropolitan Museum of Art Bulletin (May, 1968), pp. 352–8.

Allan, James and Caroline Roberts (eds), Syria and Iran. Three Studies in Medieval Ceramics. Oxford Studies in Islamic Art, IV (Oxford, 1987).

Ambros, Arne A. ‘Beobachtungen zu Aufbau und Funktionen der gereimten klassischarabischen Buchtitel’, Wiener Zeitschrifi für die Kunde des Morgenlandes lxxx (1990), pp. 13–57.

Aslanapa, Oktay and Rudolf Naumann (eds), Forschungen zur Kunst Asiens. In Memoriam Kurt Erdmann 9. September 1901 – 30. September 1964 (Istanbul, 1969).

Atıl, Esin, Ceramics from the World of Islam (Washington, D.C., 1973).

Babaie, Sussan, Isfahan and its Palaces. Statecraft, Shi‘ism and the Architecture of Conviviality in Early Modern Iran (Edinburgh, 2008).

Bahrami, Mehdi, ‘A gold medal in the Freer Gallery of Art’, in George C. Miles (ed.), Archaeologica Orientalia in Memoriam Ernst Herzfeld (Locust Valley, N.Y., 1952), pp. 5–20.

Baipakov, Karl and Lev Erzakovich, Keramika srednevekovogo Otrara – Ceramics of Medieval Otrar (Alma-Ata, 1991).

Baker, Patricia L., and Barbara Brend (eds), Shifting Sands, Reading Signs. Studies in Honour of Géza Fehérvári (London, 2006).

Baskhanova, Maria, ‘Flowers, calligraphy and the potter’s wheel. Glazed ceramics from the Mawarannahr’, in Mikhail Baskhanov, Maria Baskhanova, Pavel Petrov and Nikolaj Serikoff, Arts from the Land of Timur. An Exhibition from a Scottish Private Collection (Paisley, 2012), pp. 322–30.

Baskhanov, Mikhail, Maria Baskhanova, Pavel Petrov and Nikolaj Serikoff, Arts from the Land of Timur. An Exhibition from a Scottish Private Collection (Paisley, 2012).

Bečka, Jiři (ed.), Yādnāme-ye Jan Rypka (Articles on Persian and Tajik Literature) (Prague, 1967).

Blair, Sheila S., Text and Image in Persian Art (Edinburgh, 2014).

Bol’shakov, Oleg G., ‘Arabskie nadpisi na polivnoy keramike Sredney Azii IX–XII vv.’, Epigrafika Vostoka xii (1958), pp. 23–38; xv (1963), pp. 73–87; xvi (1963), pp. 35–55; xvii (1966), pp. 54–62; and xix (1969), pp. 42–50.

Bosworth, C. Edmund, The Ghaznavids: Their Empire in Afghanistan and Eastern Iran 994–1040 (Edinburgh, 1963).

Bosworth, C. Edmund (tr.), The Book of Curious and Entertaining Information. The Laṭā’if al-Ma‘ārif of Tha‘ālibī (Edinburgh, 1968).

Bosworth, C. Edmund (ed.), Iran and Islam in Memory of the Late Vladimir Minorsky (Edinburgh, 1971).

Bosworth, C. Edmund, ‘Nishapur. i. Historical Geography and History to the Beginning of the 20th Century’, EIr online (http://www.iranicaonline.org/articles/nishapur-i), accessed 13 August 2014.

Bierman, Irene, ‘Near East Gallery’, Arts of Asia, xxii/3 (May–June 1992), pp. 120–7.

Bloom, Jonathan M., ‘The expression of power in the art and architecture of early Islamic Iran’, in Edmund Herzig and Sarah Stewart (eds), The Idea of Iran, Vol. V: Early Islamic Iran (London, 2012), pp. 102-19.

Brend, Barbara, Islamic Art (London, 1991).

Browne, Edward G., A Literary History of Persia. Vol. I: From the Earliest Times until Firdawsí (repr. Cambridge, 1977).

Bulliet, Richard W., The Patricians of Nishapur. A Study in Medieval Islamic Social History (Cambridge, MA, 1972).

Bulliet, Richard W., ‘Medieval Nishapur. A topographical and demographic reconstruction’, Studia Iranica, v (1976), pp. 67–89.

Bulliet, Richard W., ‘Pottery styles and social status in medieval Khurasan’, in A. Bernard Knapp (ed.), Archaeology, Annales and Ethnohistory (Cambridge, 1992), pp. 75–82.

Burckhardt, John L., Arabic Proverbs (London, Dublin and Totowa, N.J., 1972).

Busse, Heribert, Chalif und Grosskönig. Die Buyiden im Irak (945–1055) (Beirut, 1969, repr. 2004).

Caviró, Balbina Martinez, Cerámica Hispanomusulmana Andalusí y Mudéjar (Madrid, 1991).

Charleston, Robert J. (ed.), World Ceramics: An Illustrated History (London, 1977).

De la Vaissière, Étienne (ed.), Islamisation de l’ Asie Centrale. Processus Locaux d’ Acculturation du VIIe au XIe Siècle (Paris, 2008).

Du Ry, Carel J., Art of Islam, tr. Alexis Brown (Baden-Baden, 1970).

Dutton, Yasin, ‘Red dots, green dots, yellow dots and blue: some reflections on the vocalisation of early Qur’anic manuscripts: Part I’, Journal of Qur’anic Studies i/1 (1999), 115–40.

Dutton, Yasin, ‘Red dots, green dots, yellow dots and blue: Some reflections on the vocalisation of early Qur’anic manuscripts: Part II’, Journal of Qur’anic Studies ii/1 (2000), 1–24.

Elwell-Sutton, Laurence P., Persian Proverbs (London, 1954).

Enderwitz, Susanne, ‘Shu‘ubiyya’, EI2, ix, pp. 513–516.

Erdmann, Kurt, ‘Afrasiab ceramic wares’, Bulletin of the Iranian Institute vi/1–4 and vii/1 (1946), pp. 102–10.

Ettinghausen, Richard, ‘A case of traditionalism in Iranian art’, in Oktay Aslanapa and Rudolf Naumann (eds), Forschungen zur Kunst Asiens. In Memoriam Kurt Erdmann 9. September 1901 – 30 September 1964 (Istanbul, 1969), pp. 88–110.

Ettinghausen, Richard, ‘Abrī Painting’, in Miriam Rosen-Ayalon (ed.), Studies in Memory of Gaston Wiet (Jerusalem, 1977), pp. 345–56.

Fahd, Toufic, Etudes d’Histoire et de Civilisation Arabes et Islamiques II (Istanbul, 2006)

Fehérvári, Géza, Ceramics of the Islamic World in the Tareq Rajab Museum (London and New York, 2000).

Fitzherbert, Teresa, ‘Themes and Images on the Animate Buff Ware of Medieval Nishapur’. Unpublished MPhil dissertation, University of Oxford, 1983.

Finster, Barbara, Christa Fragner and Herta Hafenrichter (eds), Kunst und Kunsthandwerk im Islam. 2. Bamberger Symposium der islamischen Kunst 25–27 Juli 1996 in Oriente Moderno XXIII (LXXXIV)/2 (2004), pp. v–vi and 355–539.

Flury, Samuel, ‘Une formule épigraphique de la céramique archaïque de l’Islam’, Syria v (1924), pp. 53–66.

Flury, Samuel, ‘B. Ornamental Kufic inscriptions on pottery’, in Arthur U. Pope and Phyllis Ackerman (eds), A Survey of Persian Art from Prehistoric Times to the Present (London and New York, 1938–9), pp. 1743–69.

Freeman, Nan B., and Walter B. Denny, Ebrû Art: Marble on Paper, The Work of Feridun Özgören (Bahrain, 2001).

Frye, Richard N., ‘Development of Persian literature under the Samanids and Qarakhanids’, in Jiři Bečka (ed.), Yādnāme-ye Jan Rypka (Articles on Persian and Tajik Literature) (Prague, 1967), pp. 69–74.

Frye, Richard N., Bukhara. The Medieval Achievement (Norman, Oklahoma, 1965).

Frye, Richard N., ‘The Samanids’, in Richard N. Frye (ed.), The Cambridge History of Iran. Vol. IV: From the Arab Invasion to the Saljuqs (Cambridge, 1975), pp. 136–61.

Frye, Richard N. (tr.), The History of Bukhara. Translated from a Persian Abridgement of the Arabic Original by Narshakhī (Cambridge, MA, 1954).

Ghouchani, ‘Abdallah, Inscriptions on Nishabur Pottery tr. Martin Charlesworth, (Tehran, 1986).

Goitein, Shlomo D., ‘The present-day Arabic proverb as a testimony to the social history of the Middle East’, Studies in Islamic history and institutions (Leiden, 1968), pp. 361–79.

Goldziher, Ignaz, ‘The Shu‘ūbiyya’ and ‘The Shu‘ūbiyya and its manifestation in scholarship’, in Ignaz Goldziher, Muslim Studies. Muhammedanische Studien, ed. Samuel M. Stern, tr. Samuel M. Stern and C. Renate Barber (London, 1967), I, pp. 137–63 and 164–200.

Golden, Peter B., ‘The Karakhanids and early Islam’, in Denis Sinor (ed.), The Cambridge History of Early Inner Asia (Cambridge, 1990), pp. 354–61.

Golombek, Lisa and Donald Wilber, The Timurid Architecture of Iran and Turan (Princeton, 1988)

Grabar, Oleg, ‘The earliest Islamic commemorative structures, notes and documents’, Ars Orientalis vi (1966), pp. 7–45.

Grabar, Oleg, ‘Notes on the decorative composition of a bowl from Northeastern Iran’, in Richard Ettinghausen (ed.), Islamic Art in The Metropolitan Museum of Art (New York, 1972), pp. 91–98.

Graves, Margaret S., and Benoît Junot (eds), Treasures of the Aga Khan Museum – Arts of the Book & Calligraphy (n.p., 2010).

Grube, Ernst J., Islamic Pottery of the Eighth to the Fifteenth Century in the Keir Collection (London, 1976).

Grube, Ernst J., The Nasser D. Khalili Collection of Islamic Art. Volume IX. Cobalt and Lustre. The first centuries of Islamic pottery (general editor: Julian Raby) (London, 1994).

Guy, John, ‘Rare and strange goods. International trade in ninth-century Asia’, in Regina Krahl, John Guy, J. Keith Wilson and Julian Raby (eds), Shipwrecked. Tang Treasures and Monsoon Winds (Washington, D.C., 2010), pp. 19-27.

Heinrichs, Wolfhart, ‘Orientalisches Mittelalter’, in Klaus von See (ed.), Neues Handbuch zur Literaturwissenschaft V (Wiesbaden, 1990), pp. 346–65.

Hennequin, Gilles, ‘Grandes monnaies Sāmānides et Ghaznavides de l’Hindū Kush, pp. 331–421 A.H. – Étude numismatique et historique’, Annales Islamologiques ix (1970), pp. 127–77.

Hennequin, Gilles, ‘Macrodirhams sāmānides inédits (collection particulière), Annales Islamologiques xx (1984), pp. 197–221.

Henning, B. Walter, ‘Eine arabische Version mittelpersischer Weisheitschriften’, Zeitschrift der Deutschen Morgenländischen Gesellschaft cvi (1956), pp. 73–7.

Henshaw, Christina M., Early Islamic Ceramics and Glazes of Akhsiket, Uzbekistan (unpublished PhD dissertation, University College, London, 2010).

Herzig, Edmund and Sarah Stewart (eds), Early Islamic Iran (The Idea of Iran, Vol. 5) (London, 2012).

Hillenbrand, Robert, ‘The Islamic re-working of the Sasanian heritage: Two case studies’, in Patricia L. Baker and Barbara Brend (eds), Shifting Sands, Reading Signs. Studies in Honour of Géza Fehérvári (London, 2006), pp. 1–14.

Hsiao-yi, Chin, China at the Inception of the Second Millennium. Art and Culture of the Sung Dynasty, 960–1279 (Taipei, 2000).

Ilyasov, Jangar Y., ‘Exotic images: On a new group of glazed pottery of the 10th and 11th century’, Journal of the David Collection iv (2014), pp. 50–87.

Ilyasova, Saida and Nadeshda Wischnewskaya, ‘Glasierte Keramik von Binket (Taschkent) aus der Sammlung des Staatlichen Museums für Orientalische Kunst’, Tribus li (2002), pp. 114–26.

Ilyasova, Saida and Rawschan Imamberdyev, ‘Eine Sammlung glasierter Keramik aus Taschkent’, Tribus liv (2005), pp. 91–102.

Jenkins, Marilyn, ‘Muslim: An early Fatimid ceramist’, The Metropolitan Museum of Art Bulletin (May, 1968), pp. 359–69.

Kadoi, Yuka, ‘Translating from jing to mir’at/a’ina: Medieval Islamic mirrors revisited’, Art in Translation v/2 (2013), pp. 251–72 (online journal: http://www.bloomsbury.com/uk/journal/art-in-translation/), accessed 22 August 2014.

Kamoliddin, Shamsiddin, ‘On the religion of the Samanid ancestors’, July 2006, www.transoxiana.org/11/kamoliddin-samanids.html, accessed 21 July 2014.

Karev, Yury, ‘Qarakhanid wall paintings in the citadel of Samarqand: First report and preliminary observations’, Muqarnas xxii (2005), pp. 45–84.

Kervran, Monique, ‘Un monument baroque dans les steppes du Kazakhstan: Le tombeau d’Örkina Khatun, princesse Chaghatay?’, Arts Asiatiques lvii (2002), pp. 5–32.

Kessler, Rochelle L. et al., Studies in Islamic and Later Indian Art (Cambridge, Mass., 2002).

Kiani, M. Yusuf, The Islamic City of Gurgan (Berlin, 1984).

Kraemer, Joel L., Humanism in the Renaissance of Islam (Leiden, 1992).

Krahl, Regina, John Guy, J. Keith Wilson and Julian Raby (eds), Shipwrecked. Tang Treasures and Monsoon Winds (Washington, D.C., 2010).

Kratchkovskaya, Vera, ‘Evoliutsiya kuficheskogo pisma v Srednei Azii’, Epigrafika Vostoka iii (1949), pp. 3–27.

Kühnel, Ernst, ‘Die Kunst Persiens unter den Buyiden’, Zeitschrift der Deutschen Morgenländischen Gesellschaft cvi (1956), pp. 78–92.

Lane, Arthur, Early Islamic Pottery (London, 1947).

Lane, Arthur, Islamic Pottery from the Ninth to the Fourteenth Centuries A.D. in the Collection of Sir Eldred Hitchcock (London, 1956).

Leisten, Thomas, Architektur für Tote. Bestattung in architektonischem Kontext in den Kernländern der islamischen Welt zwischen 3./9. und 6./12. Jahrhundert. Materialien zur Iranischen Archäologie, Band 4 (Berlin, 1998).

Levy, Reuben, A Baghdad Chronicle (repr. Philadelphia, 1977).

Madelung, Wilferd, ‘The assumption of the title Shāhānshāh by the Buyids and “The Reign of Daylam (Dawlat al-Daylam)”’, Journal of Near Eastern Studies xxviii (1969), pp. 84–108 and 168–83.

Malamud, Margaret, ‘Sufi organizations and structures of authority in medieval Nishapur’, International Journal of Middle East Studies xxvi (1994), pp. 427–42.

Marçais, Georges, ‘Remarques sur l’esthétique musulmane’, Mélanges d’Histoire et d’Archéologie de l’Occident Musulman. Tome I. Articles et Conférences de Georges Marçais (Algiers, 1957), pp. 93–104.

Markwart, Josef, Wehrot und Arang. Untersuchungen zur mythischen und geschichtlichen Landeskunde von Ostiran (Leiden, 1938).

Melikian-Chirvani, Assadollah S., ‘L’ évocation littéraire du bouddhisme dans l’Iran musulman’, Le Monde Iranien et l’Islam ii (1974), pp. 1–72.

Mez, Adam, The Renaissance of Islam, tr. Salahuddin Khuda Bakhsh and David S. Margoliouth (Patna, 1937).

Miles, George C. (ed.), Archaeologica Orientalia in Memoriam Ernst Herzfeld (Locust Valley, N.Y., 1952).

Miles, George C., ‘Numismatics’, in Richard N. Frye (ed.), The Cambridge History of Iran. Vol. IV: From the Arab Invasion to the Saljuqs (Cambridge, 1975), pp. 364–77.

Minorsky, Vladimir, ‘The older preface to the Shāh-Nāma’, repr. in Iranica. Twenty Articles (Tehran, 1964), pp. 260–73.

Mirzaaxmedov, Djamal, ‘La production céramique du Maverannahr du IXe au début du XIIIe siècle’, in Étienne de la Vaissière (ed.), Islamisation de l’Asie Centrale. Processus Locaux d’Acculturation du VIIe au XIe Siècle (Paris, 2008), pp. 59–70.

Momtaz, Irène, Momtaz Islamic Art. Ornament and Inscription, ed. Ralph Pinder-Wilson (London, 2004).

Momtaz, Irène, Memories from Islamic Lands (London, n.d.)

Morgan, Peter, ‘Samanid pottery. Types and techniques’, in Ernst J. Grube, The Nasser D. Khalili Collection of Islamic Art. Volume IX (General editor Julian Raby), Cobalt and Lustre. The first centuries of Islamic pottery (London and Oxford, 1994), pp. 55–9.

Mottahedeh, Roy P., ‘The Shu‘ubiyah controversy and the social history of early Islamic Iran’, International Journal of Middle East Studies vii/2 (1976), pp. 161–182.

Pancaroğlu, Oya, ‘Serving Wisdom: The contents of Samanid epigraphic pottery’, in Rochelle L. Kessler et al., Studies in Islamic and Later Indian Art (Cambridge, MA, 2002), pp. 59–75.

Pancaroğlu, Oya, 2005, ‘Functions of literary epigraphy on medieval Islamic ceramics: Part One: Samanid epigraphic pottery’, http//islamicceramics.ashmolean.org/Samanids/oya-part-one-.htm

Pancaroğlu, Oya, Perpetual Glory. Medieval Islamic Ceramics from the Harvey B. Plotnick Collection (Chicago, 2007).

Peacock, A. C. S., Mediaeval Islamic Historiography and Political Legitimacy: Bal‘amī’s Tārīkhnāma (London, 2007).

Pinder-Wilson, Ralph H., ‘A lustre relief dish of the early Islamic period’, British Museum Quarterly xxvii/3–4 (1963–4), pp. 91–4.

Plant, Richard, Arabic Coins and How to Read Them (London, 1980).

Pope, Arthur U., ‘Discoveries at Harun ar-Rashid’s birthplace’, The Illustrated London News CLXXXVI (22 June 1935), pp. 1122–3.

Pope, Arthur U. and Phyllis Ackerman (eds), A Survey of Persian Art from Prehistoric Times to the Present (London and New York, 1938–9).

Pritsak, Omeljan, ‘Die Karachaniden’, Der Islam xxxi (1953–4), pp. 17–68.

Pugachenkova, Galina A. and Éduard V. Rtveladze, ‘Afrāsīāb’, EIr, i, pp. 576b–578b.

Raby, Julian, ‘Looking for silver in clay: A new perspective on Samanid ceramics’, in Michael Vickers (ed.), Pots & Pans. A Colloquium on Precious Metals and Ceramics in the Muslim, Chinese and Graeco-Roman Worlds, Oxford 1985 (Oxford Studies in Islamic Art, III) (Oxford, 1986), pp. 179–203.

Rante, Rocco and Annabelle Colinet (with contributions by Rajabali Labbaf Khaniki, A. Bouquillon, Y. Coquinot, C. Doublet, Y. Gallet, A. Genevey, E. Porto and A. Zink), Nishapur Revisited: Stratigraphy and Pottery of the Qohandez (Oxford, 2013).

Rawson, Jessica, ‘Chinese silver and its influence on porcelain development’, Cross-Craft and Cross-Cultural Interactions iv (1989), pp. 275-99.

Rice, David S., The Unique Ibn al-Bawwāb Manuscript in the Chester Beatty Library (Dublin, 1955).

Rizvi, Kishwar, The Safavid Dynastic Shrine. Architecture, Religion and Power in Early Safavid Iran (London and New York, 2011).

Roxburgh, David J. (ed.), Turks. A Journey of a Thousand Years, 600–1600 (London, 2005).

Rypka, Jan, History of Iranian Literature, ed. Karl Jahn, (Dordrecht, 1968).

Sarre, Friedrich, Denkmäler Persischer Baukunst (Berlin, 1910).

Scavizzi, Giuseppe, Maioliche dell’Islam e del Medioevo occidentale (Milan, 1966).

Schimmel, Annemarie, Islamic Calligraphy (Leiden, 1970).

Schafer, Edward H., The Golden Peaches of Samarkand. A Study of T’ang Exotics (Berkeley, 1963).

Sellheim, Rudolf, ‘Eine unbekannte persische Sprichwörtersammlung’, Der Islam lxxxi (2004), pp. 243–8.

Shen, Hsueh-man, ‘Chinese polychromes in the Indian Ocean trade during the 9th century’, in Lorenz Korn (ed.), Beiträge zur Islamischen Kunst und Archäologie iv (Wiesbaden, 2014), pp. 107–22.

Siméon, M. Pierre, Étude du materiel de Hulbuk (Mā wārā al-nahr Khuttal) de la conquête islamique jusqu’au milieu du XIe siècle (90/712-441/1050). Contribution à l’étude de la céramique d’Asie centrale (Oxford, 2009).

Sinor, Denis (ed.), The Cambridge History of Early Inner Asia (Cambridge, 1990).

Spuler, Bertold, Iran in früh-islamischer Zeit. Politik, Kultur, Verwaltung und öffentliches Leben zwischen der arabischen und seldschukischen Eroberung 633 bis 1055 (Wiesbaden, 1952).

Stern, Samuel M., ‘Ya‘qūb the Coppersmith and Persian national sentiment’, in C. Edmund Bosworth (ed.), Iran and Islam. In Memory of the Late Vladimir Minorsky (Edinburgh, 1971), pp. 535–55.

Sugimura, Toh, The Encounter of Persia with China. Research into Cultural Contacts based on Fifteenth Century Persian Pictorial Materials, in Senri Ethnological Studies, No.18 (Osaka, 1986).

Taddei, Maurizio (ed.), Gururājamañjarikā. Studi in onore di Giuseppe Tucci (Rome, 1974).

Tamari, Vera C., ‘Ninth-Tenth Century White Mesopotamian Ceramic Ware with Blue Decoration’, Oxford University, MPhil thesis, 1984.

Tashkhodzhaev, Sharip S., Khudozhestvennaya Polivnaya Keramika Samarkanda IX – nachala XIII vv. (Tashkent, 1967).

Treadwell, W. Luke, ‘The account of the Samanid dynasty in Ibn Zāfir al-Azdī’s Akhbār al-duwal al-munqaṭi’a’, Iran xlii (2005), pp. 135–73.

Treadwell, W. Luke, ‘Shāhānshāh and al-Malik al-Mu’ayyad: The legitimation of power in Samanid and Buyid Iran’, in Farhad Daftary and Josef W. Meri (eds), Culture and Memory in Medieval Islam: Essays in Honour of Wilferd Madelung (London and New York, 2003), pp. 318–37.

Treadwell, W. Luke, ‘A unique portrait medallion from Bukhara dated 969 A.D.’, The Ashmolean xxxvi (1999), pp. 9–10.

Vasmer, Richard, ‘Zur Münzkunde der Qarahaniden’, Mitteilungen des Seminars für Orientalische Sprachen xxxiii (1930), pp. 83–104.

Ventrone, Giovanna, ‘Iscrizioni inedite su ceramica samanide in collezioni italiane’, in Maurizio Taddei (ed.), Gururājamañjarikā. Studi in onore di Giuseppe Tucci (Rome, 1974), pp. 221–32.

Vickers, Michael (ed.), Pots & Pans. A Colloquium on Precious Metals and Ceramics in the Muslim, Chinese and Graeco-Roman Worlds, Oxford 1985 (Oxford Studies in Islamic Art, III) (Oxford, 1986).

Volov, Lisa, ‘Plaited Kufic on Samanid epigraphic pottery’, Ars Orientalis 6 (1966), pp. 107–33.

Watson, Oliver, Persian Lustre Wares (London, 1985).

Watson, Oliver, ‘Chinese–Iranian Relations XI. Mutual influence of Chinese and Persian ceramics’, EIr, v, pp. 455a–57b.

Watson, Oliver, ‘Fakes and forgeries in Islamic pottery’, in Barbara Finster, Christa Fragner and Herta Hafenrichter (eds), Kunst und Kunsthandwerk im Islam. 2. Bamberger Symposium der islamischen Kunst 25. – 27. Juli 1996, Oriente Moderno XXIII (LXXXIV)/2 (2004), pp. 517–39.

Watson, Oliver, Ceramics from Islamic Lands (London, 2004).

Watson, William, The Arts of China 900–1620 (New Haven and London, 2000).

Wilkinson, Charles K., ‘The glazed pottery of Nishapur and Samarkand’, Bulletin of the Metropolitan Museum of Art, N.S. xx/3 (1961), pp. 102–15.

Wilkinson, Charles K., Nishapur: Pottery of the Early Islamic Period (New York, 1973).

Williamson, Andrew, ‘Regional distribution of mediaeval Persian pottery in the light of recent investigations’, in James Allan and Caroline Roberts (eds), Syria and Iran. Three Studies in Medieval Ceramics (Oxford Studies in Islamic Art, IV) (Oxford, 1987), pp. 11–22.

Wright, Elaine, Islam: Faith, Art, Culture. Manuscripts of the Chester Beatty Library (London, 2009).

Wu Yuejian, Tang feng miaocai: Changsha yao jingpin zhan (Marvelous Color and Tang Style: Exquisite Wares of Changsha Kiln) (Changsha, 2008).

Zakeri, Mohsen, ‘ ‘Alī ibn ‘Ubaida ar-Raiḥānī: A forgotten belletrist (adīb) and Pahlavi translator’, Oriens xxiv (1994), pp. 76–102.

Zand, Mikhail, ‘Some light on bilingualism in literature of Transoxiana, Khurasan and Western Iran in the 10th century A.D.’, in Jiři Bečka (ed.), Yādnāme-ye Jan Rypka (Articles on Persian and Tajik Literature) (Prague, 1967), pp. 161–4.

Zetterquist, Eric J., Persian Ceramics and Related Materials (New York and London, 1993).

Zhiyan, Li; Bower, Virginia L.; and Li, He (eds), Chinese Ceramics From the Paleolithic Period through the Qing Dynasty (New Haven and London, 2010).

You can support our site by clicking on this link and watching the advertisement.

If you find an error or have any questions, please email us at admin@erenow.org. Thank you!