The findings presented in this paperFootnote 1 stand at the crossroads between three growing fields of study: Abbasid court practice, Turkish presence at the heart of the Abbasid empire, and Eurasian cultural interconnectedness. In this paper I hope to offer a new look on the practice of samāǧa performances at the presence of the Abbasid caliphs and a better understanding of the few Arabic sources on these masquerades that, in some cases, fittingly lent themselves to misunderstanding and confusion.
One such case is a description by Abū al-Faraǧ al-Iṣfahānī of a group of dancing farāġīna in al-Muʿtaṣim's (r. 218–227/833–842) court. In his study, Live Theatre and Dramatic Literature in the Medieval Arab World, Shmuel Moreh reads the collective name of this group of performers as ṣafāʿina.Footnote 2 To understand why, we must reconsider how samāǧa performances were introduced to the Abbasid courts.
Abbasid samāǧa and Tang sumozhe
Abbasid sources record in a handful of passages a small number of samāǧa performances, a custom falling into the category of “semi-theatrical and semi-carnivalesque phenomena” characterized by the presence of grotesque elements and socio-political meanings.Footnote 3 The details of this practice are not clear in the sources, nor does modern scholarship offer much insight into its precise features. Nevertheless, a few studies provide brief descriptions. M.M. Ahsan, in his monograph on social life in the Abbasid age, holds samāǧa to be “something like a masquerade”.Footnote 4 A more in-depth explanation can be found in Shmuel Moreh, “Performing artists”, in Medieval Islamic Civilization. Footnote 5 Moreh's main study on samāǧa performances, however, comes in a chapter in Live Theatre and Dramatic Literature in the Medieval Arab World. Moreh relates the entirety of the information available on samāǧa performances from Arabic sources, as well as situating the performances within the broader context of masquerades in the pre-modern Near Eastern Islamic world.Footnote 6
The history of this practice in the central lands of the Abbasid empire has received even less attention, in spite of the fact that most – if not all – samāǧa performances in the caliphal age seem to have taken place there. Chronologically, it is safe to say that samāǧa performances enjoyed a period of popularity among sectors of the Abbasid entourage in the third/ninth century, since the term cannot be found in reference to theatrical, farcical, masquerade, or otherwise playful performances in earlier or later centuries in the Abbasid context.
The research presented in this paper stems from the working hypothesis that this eclipse of the term reflects either the disappearance of samāǧa performances among Muslim elites, or that similar practices had come to be known under a different name. However, before dealing with the question of why the Abbasid courts ceased hosting samāǧa performances, we should ask ourselves why they began in the first place. In other words, who brought this practice to court?
On the basis of Arabic literary sources, we can argue that samāǧa performances came to the central lands of the Caliphate by means of Turkish slave imports from Central Asia. At the time, Nawrūz celebrations were gradually becoming an integral component of both the cultural background of Muslim elites and Abbasid political discourse on imperial legitimacy.
I should note in this regard that ʿirāqī Nawrūz celebrations, of whose existence we have knowledge already from under the Umayyads,Footnote 7 did not immediately become part of courtly customs after the so-called Abbasid revolution. Instead, the celebrations gained momentum in the first half of the third/ninth century, paralleling a macroscopic shift in the lifestyle of the new Muslim elite, who had begun to move from the countryside to the cities as their main source of income shifted from direct land ownership to the control of taxation.Footnote 8
In a recent article Norman H. Rothschild lays new ground for the study of Abbasid samāǧa,Footnote 9 allowing us to trace an ideal arc connecting samāǧa performances in Abbasid Iraq to Chinese sumozhe 蘇摩遮 in East Asia.Footnote 10 In his study, Rothschild engages the complex history of the “Cold-splashing Sogdian plays” that took place during the Tang dynasty in the first half of the eighth century. He provides context for the opposition of Confucian dignitaries, who held such customs in contempt and ultimately succeeded in having them banned.
Rothschild's work evaluates two words employed in Chinese sources in relation to “cold-splashing plays”: huntuo, originally the name of a peculiar kind of skin or felt hat, which was later extended to a dance and song performed while wearing said hat; and sumozhe. According to Rothschild, the latter term's etymology may be traced back to Sanskrit, “to the term samāǧa (in Brahmi) that appears in the first Asokan Rock Edict”.Footnote 11 However, as he does not fail to note, the word samāǧa also appears in Arabic lexicography Footnote 12 where it is defined as both an abstract concept meaning “foul, unseemly, or ugly” – in a word, grotesque – and a kind of mask or masquerade. Sumozhe, therefore, is most likely a Chinese transliteration, referring to “a tune accompanying theatrical amusements from Kutcha”, a region in modern-day Xinjiang. These theatrical amusements lasted for several days and featured “dancers wearing animal masks”.Footnote 13 According to the Tang official Zhang Yue (663–730), this practice, meaning sumozhe, “originated in Persia” as a New Year custom and had strong ties to the imperial traditions of that land; it involved dances, the spattering of water, and beautiful – if not boisterous – masks. In truth, the aesthetic quality of such masks was debatable: they are sometimes described as a thing of beauty, other times as fantastic, representing “ghosts or hoary ancients” or “demons and animals”, but always as infectious and frenetic.Footnote 14 Be that as it may, sumozhe performances were part and parcel of Sogdian “cold-spashing plays” in Tang China.
Less than two centuries later, samāǧa performances made their appearance in Abbasid courts on the other side of the Eurasian continent. The earliest samāǧa known to Arabic sources took place at the court of the Abbasid caliph al-Muʿtaṣim (r. 218–227/833–842). The performance is described in the Kitāb Maqātil al-Ṭālibiyyīn, a 313/923 work by Abū al-Faraǧ al-Iṣfahānī.Footnote 15 According to this report, on Nawrūz of 219/834, al-Muʿtaṣim had the Zaydī imām Muḥammad b. al-Qāsim (also known as Ṣāḥib al-Ṭalaqān) brought before him.Footnote 16 Al-Muʿtaṣim celebrated the New Year's Day with a samāǧa performance, which al-Iṣfahānī vividly describes as follows:
[…] the aṣḥāb al-samāǧa were performing [yalʿabūna] in front of him, and the farāġina where dancing [yarquṣūna]. When Muḥammad saw all this, he cried and then said, “Oh God! You know that I would always want to change all this and remove it.” The farāġina began throwing dirt and rotten meat on the common people. Al-Muʿtaṣīm was laughing, and Muḥammad b. al-Qāsim glorified God and asked for His forgiveness, moving his lips, preaching. Al-Muʿtaṣim was sitting in the palace in Šammāsiyya,Footnote 17 while [Muḥammad b. al-Qāsim], looking at them, was standing still.Footnote 18
The aṣḥāb al-samāǧā Footnote 19 of al-Muʿtaṣim do not exactly dance, but rather seem to engage in some sort of playful performance. Dance proper is a job left for a different group: the farāġina, who then proceed to throw disgusting waste on the public. This is surely unusual, given al-Iṣfahānī’s clear disdain for this mockery of the nawrūzī water-sprinkling custom, known in Arabic sources as ṣabb al-māʾ. Footnote 20 Of course, it is not surprising that al-Muʿtaṣim is represented here in such a negative light; some of the most extreme details of the story may well be counterfeit or exaggerated better to convey the image of a wicked oppressor of the Alid family. Nevertheless, some other details that do not necessarily play into the hands of al-Muʿtaṣim's critics remain noteworthy.
The aṣḥāb al-samāǧa
First, a few words should be allotted to the aṣḥāb al-samāǧa. As al-Iṣfahānī says little about their appearances or their contribution to the performance, we must resort to later sources for a description of their features. A few years later, Ibn al-Muʿtazz (247–296/861–908) describes them as similar to a group of ǧinn:
Drink on the morning of Nayrūz a pure cup! / May the days happily pass
The Ǧinn appeared to us in daytime, / some of them in line, others holding hands
Dancing, their bodies sway / as cypresses do in the wind
Ugliness covers their beauty / but in their samāǧāt lies grace.Footnote 21
That the aṣḥāb al-samāǧa wore masks has already been noted by Shmuel Moreh, who further states that Ibn Sīnā employed the term samāǧāt as a translation of Greek comical masks, while a fourth/tenth century commentator chose the term samāǧā to gloss a verse by Ğarīr mentioning the kurraǧ, a hobby-horse.Footnote 22 Caliphs themselves occasionally enjoyed this popular and courtly custom. Its origins, however, are highly disputed. According to Arabic lexicography, the term is derived from Persian, meaning young horse or mule. A. van Gepp argues for a Western European origin, while Gaudefroy-Demombynes disagrees. More recently, Moreh proposed to link the kurraǧ with Central Asian shamanism, concurring with the general hypothesis of J. Baumel, who sees in the kurraǧ “l'homme qui agite une tête de cheval au bout d'un bâton ou qui anime le corps du masque continue en effet à traduire inconsciemment un des plus anciens rites qu'ait connu l'humanité”.Footnote 23
From these elements we can assume that the masks worn by the aṣḥāb al-samāǧa could be described as grotesque, possibly with animal-like features, working towards a comical effect. These masks clearly resembled those employed in the sumozhe customs resented by Tang Confucian officials: possibly animal-like; surely wondrous, if not ugly; and intended to catch the eye.
For a better understanding of the role played by the aṣḥāb al-samāǧa, we may turn to al-Šābuštī’s (356–386/975–996) record in the Kitāb al-Diyārāt of a samāǧa performance for the Abbasid caliph al-Mutawakkil (r. 232–247/847–861). According to the Fatimid-period littérateur, Al-Mutawakkil's celebration of New Year's Day was interrupted by a very concerned chief of the guards, Abū al-Ḥūsayn Isḥāq b. Ibrāhīm:Footnote 24
On a Nawrūz day, Isḥāq b. Ibrāhīm came to al-Mutawakkil as the samāǧa was [being performed] in front of him. He was wearing a heavy, multicoloured robe. The asḥāb al-samāǧa had gathered around him in great numbers, so as to catch the dirhams he was scattering on them; they were so close that they could pull edges [of his robe]. When Isḥāq b. Ibrāhīm saw all this, he became angry and left, saying, “Why do we waste money paying for guards in this reign?!” Al-Mutawakkil saw him and said, “Bring back to me Abū al-Ḥusayn [scil. Isḥāq b. Ibrāhīm], who just left in anger”. The gatekeeper left [to do as he had been ordered], followed by the attendants. [Isḥāq b. Ibrāhīm] returned, speaking very harshly to Waṣīf and Zurāfa until he was in front of al-Mutawakkil, who asked him, “Why did you get angry, and why did you leave?” [Isḥāq b. Ibrāhīm] then answered, “Commander of the Faithful, do you think that the enemies of this reign are not as numerous as its friends? You stay in the company of such dogs, who even dare to pull the edges of your robe, each and every one of them disguised by a horrid mask! What if an ill-intentioned, evil-minded enemy of yours, having renounced his faith, hides himself among them, thus making his way to you? When will all this stop? When will you free the earth of such people?” Al-Mutawakkil answered, “Abū al-Ḥusayn, do not get so angry. I promise that you will not see me like this again”. Afterwards, he had a higher structure built for himself, so that he may safely enjoy the samāǧa. Footnote 25
The anecdote probably conveys a wider concern for al-Mutawakkil's safety by some sections of the Abbasid establishment, perhaps in response to the somewhat cavalier attitude of the caliph in choosing his closest companions.Footnote 26 Nevertheless, what matters here is the description of the aṣḥāb al-samāǧa and their performance. Apart from the confirmation that the performers wore “horrid” masks, the unruly, almost wild, nature of this performance emerges from al-Šābuštī’s writing. Al-Mutawakkil's samāǧa bears much resemblance to what happened at the court of al-Muʿtaṣim – quite different from the idealized image conveyed by Ibn al-Muʿtazz's verses. Notably, the showering of money on the aṣḥāb al-samāǧa corresponds to the less pleasant throwing of rotten meat and dirt described by al-Iṣfahānī.
Al-Mutawakkil's pivotal role in the performance deserves a few words. In each of the three prose descriptions of samāǧāt recorded in Abbasid sources, the sitting caliph is present. This is true for al-Muʿtaṣim, for his son al-Mutawakkil, and for al-Muʿtaḍid, as we will see shortly. In fact, the caliph emerges as the central figure of the Abbasid samāǧāt. This is consistent with the general purpose of courtly Nawrūz celebrations: a moment of recognition and renewal of political authority and imperial order.Footnote 27 In principle, we cannot exclude that samāǧāt may have been performed outside the context of Nawrūz celebrations, but the only known Abbasid samāǧa performances took place, without fail, during courtly Nawrūz celebrations.
The farāġina
Let us now consider the second group mentioned by al-Iṣfahānī, the dancing farāġina. Moreh contends that the text should be read as ṣafāʿina, i.e. slapstick actors. Even though he does not explain his decision, the reason is still quite clear: it makes more sense to have slapstick actors participating in grotesque, even comical, performance than people or masks from Farghana, whose presence awaits explanation. Nevertheless, the report contained in Kitāb Maqātil al-Ṭālibiyyīn may well be correct and reliable, provided that we can elucidate the link between Farghana and samāǧa performances. This latter option would be preferable, not just for philological reasons, but also because it would provide context for a later source on the samāǧa performed for al-Muʿtaḍid, described in the Kitāb al-Ḏaḫāʾir wa-l-Tuḥaf as follows:Footnote 28
On Nawrūz, in the year 282[/895],Footnote 29 Qaṭr al-Nadā, the daughter of Abī al-Ǧayš Khumārawayh, presented to al-Muʿtaḍid bi-l-llāh a gift containing twenty gold trays, […] The cost of the masks for the samāǧāt she had provided for Nawrūz amounted to thirteen thousand dinars. Thirty maids of honour came out of the palace and danced with the farāʿina. Footnote 30
The term farāʿina found in this description, which would literally translate as “Pharaohs”, is puzzling. Moreh proposes to correct the Kitāb al-Ḏaḫāʾir wa-l-Tuḥaf in the very same way as he did for the Kitāb Maqātil al-Ṭālibiyyīn, reading farāʿina as ṣafāʿina. This remains unconvincing due to the graphic distance between the two words (فراعنة and صفاعنة).
In her English translation, Qaddumi transcribes the Arabic word as is, suggesting that the presence of masks inspired by Ancient Egyptian lore among the gifts Qaṭr al-Nadā presented to her husband may be explained by her Egyptian origins.Footnote 31 It is a weak connection – and Qaddumi is not fully convinced of her hypothesis – because, while Qaṭr al-Nadā’s father Khumārawayh did in fact rule over Egypt and the Šām, he was a Turkish military commander born and bred in Samarra. Moreover, while we know of Egyptian samāǧa performed during the Coptic Nayrīz of 364/974–5, we have no mention of Pharaonic masks.Footnote 32
On this matter, Muḥammad Ḥamīdullāh, the editor of the 1959 Arabic edition of the Kitāb al-Ḏaḫāʾir wa-l-Tuḥaf, seems to be correct. He suggests amending farāʿina to farāġina, since a mere dot distinguishes the two Arabic spellings. Ḥamīdullāh was probably unaware of al-Iṣfahānī’s passage on the samāǧa performed by dancing farāġina for al-Muʿtaṣim – a passage that gives much strength to his argument.
The role of men or masks relating to Farghana in Abbasid samāǧāt could be explained by taking into account the Sogdian origin of Tang sumozhe, thus positing Sogdiana, which encompassed the Farghana Valley, as the epicentre of the diffusion of samāǧa from Central Asia into China and later into Iraq. In fact, customs similar to those of the Abbasid samāǧā performances at Nawrūz and those of the Tang “cold-spashing” sumozhe can be found in Central Asian contexts, for example the Turkic Kūsa ritual of Saya/Saḏaq, particularly in relation to the customs of the New Year King.Footnote 33 Krasnowolska, in her study of Iranian calendrical mythology, writes that “the elements that Kuse and Mir-e Nouruzi share are numerous and, although the origins of both ritual units are apparently distinct, the borderline between them is, sometimes, obliterated”.Footnote 34 We know from the work of Pūr Karīm on Central Asian festivals that Turkish Saya celebrations revolve around a clash of opposites, characterized by black and white, beautiful and ugly, old and new.Footnote 35 Of course, Saya celebrations are closer to Saḏaq than to Nawrūz. Nevertheless, as Cristoforetti notes in his study on Saḏaq customs, Saya celebrations feature an “element that, in light of an Iranian influx that seems to us undeniable, can be said to be nawrūzī. It is the relation that this festival entertains with the descent of the souls of the Dead.”Footnote 36 Moreover, there was a strong influx of nawrūzī elements in the Saḏaq festival during the third/ninth century, most likely due to the calendrical proximity of the two festivals at the end of the eighth century of the Common Era.Footnote 37
The participation of the dancing farāġina, therefore, gives strength to the hypothesis of a Central Asian origin for Abbasid samāǧa performances. This does not mean that the dancers called farāġina were necessarily from Farghana Valley, as the association with the Farghana Valley could be read as general brand, rather than precise geographical origin. First, Arabic sources tend to present significant discrepancies with regard to referent terms like maġāriba (Arabs from the Maghrib),Footnote 38 farāġina, and turk, that can assume both an ethnic and/or a military meaning.Footnote 39 Secondly, we should not discard the hypothesis that the term farāġina did not have the same meaning here as it did in military contexts. Instead, it could be used to indicate a particular set of costumes and moves: a character with its own role in a structured play.Footnote 40 Nevertheless, it is surely plausible that these dancing farāġina were indeed part of the farāġina corps established by al-Muʿtaṣim.Footnote 41 Given the paucity of available sources, it does not seem possible to provide a clear-cut answer on the matter.
At least from a chronological point of view, the adoption of these Central Asian masks in caliphal Nawrūz celebrations coincided with the growing presence of Turkish elements in Iraq.Footnote 42 This well-studied phenomenon had already begun during the reign of al-Maʾmūn (r. 196–218/812–833), and al-Muʿtaṣim (r. 218–227/833–842).Footnote 43 The fact that they retained cultural elements, such as customs and rituals, of their Central Asian forefathers, is clearly discussed by Gordon in his analysis of an anecdote (which also serves as the title of his book): the breaking of a thousand swords performed at the Funeral of the Turk Amir Abū Naṣr Muḥammad by his troops.Footnote 44 According to a brief account provided by al-Ṭabarī, this anecdote was performed by Turkish troops in Samarra. Al-Ṭabarī writes that it was common among Turks to perform the ceremony “when a chief died”.Footnote 45 Further, as noted by Gordon, the ceremony bears resemblance to a practice in which the guard corps committed collective suicide following the death of a military leader.Footnote 46 In other words, the relocation of Turkish men and women to Iraq led, to some extent, to the introduction of Central Asian cultural patterns to the Abbasid capitals. Samāǧā performances may have been among such traditions, which subsequently merged with pre-existing Abbasid celebrations of Nawrūz and, in turn, became common ground between the newcomers and the customs of the Abbasid court.
Even though Nawrūz celebrations remained an Abbasid courtly custom well into at least the fourth/tenth century, there are no explicit references to samāǧa performances after the caliphate of al-Muʿtaḍid. Even conceding that the few verses by Ibn al-Muʿtazz could, in principle, be dated later than al-Muʿtaḍid's death, samāǧa performances did not survive the third century Hijri. This begs the question of what shifts occurred in the social settings that used to harbour Abbasid samāǧa performances.
The Samarran system
A possible answer lies in the history of the Abbasid family's management of the Turkish military. In his Samarcande et Samarra, Étienne de la Vaissière rejects the thesis espoused by Patricia Crone and others that the Mamluk system had already been created and planned under al-Maʾmūn and al-Muʿtaṣim.Footnote 47 Instead, de la Vaissière argues that the period from the first purchases until al-Muwaffaq's regency (261/875–278/891) and al-Muʿtaḍid's caliphate saw a different relationship between the royal house and its Turkish corps. He further posits that the Mamluk system arose from the ashes of the so-called “anarchy” that followed al-Mutawakkil's assassination, precisely as an effort to fix the shortcomings of the Samarran system.Footnote 48 A specific feature of this Samarran system is the isolation of the Turkish military community, which may well have allowed for the persistence of Central Asian Turkish customs – including the aforementioned funeral rite and samāǧa performances – within the city of Samarra.Footnote 49
According to de la Vaissière, the nature of the arrangement in Samarra differed from the later Mamluk system.Footnote 50 Whereas the Mamluk came to be based on slavery to enforce obedience in young boys, the Samarran system relied “sur l'utilisation des hiérarchies pré-islamiques qui s’étiolent loin de l'Asie Centrale face aux hiérarchies du pouvoir ou de l'argent au coeur de la vie politique abbasside”.Footnote 51 Accordingly, under the Samarran system every opportunity to strengthen loyalty between the caliph and its Central Asian Turkish troops was considered all the more precious. Nawrūz celebrations, specifically serving as festivals of the Iranian winter cycle, provided such occasions. The Central Asian contribution was the introduction of samāǧa performances in which the caliph acted as a pivotal participant. It does not seem far-fetched to imagine that samāǧa performances may have lost their function, and even their potential performers, because the last generation brought up under the Samarran system acquiesced to the Mamlūk system after the rule of al-Muʿtaḍid.
In conclusion, a close reading and contextualization of sources available to date on Abbasid samāǧa performances allows for the reconstruction of a significant picture of its history. The appearance of samāǧa performances at the Abbasid court coincided with the rise of the Turkish military at the beginning of the third/ninth century. As a result their features can be traced back to Central Asia, and to the Turko-Sogdian milieu in particular. Such performances came to be associated with courtly Nawrūz celebrations, whose presence in Mesopotamia largely pre-dates the Abbasid caliphate, in addition to bearing a striking resemblance to the “cold-splashing days” (sumozhe) brought to eighth-century Tang China by Sogdian merchants. The survival of this custom in the Abbasid heartland can be explained on two levels. On one hand, there is the usefulness of Nawrūz as a shared, hierarchy-confirming celebration well known and respected by all of the main parties involved in Abbasid politics, from newcomers from the Turkish military, to the abnāʾ al-dawla of mostly Iranian or Arab-Iranian descent, and to the Mesopotamian landed elites. On the other hand, the isolation of the Turkish population (and of the farāġina corps) under the Samarran system in the third/ninth century allowed Central Asian customs to survive until the rise of the Mamlūk system.
In short, the appearance and disappearance of samāǧa celebrations in the Abbasid courtly celebration of Nawrūz closely follows the development of the Turkish presence in third/ninth-century Mesopotamia, lending strength to the thesis that the Mamluk system was not in place before the end of the century. Instead, a system based on pre-Islamic Central Asian hierarchies was probably in place. The Turkish military played a major role in the struggle for political legitimacy and a military hold on power in the third/ninth century, notably under al-Muʿtaṣim, al-Mutawakkil, and al-Muʿtaḍid. The history of samāǧa performances allows us a rare glimpse into the nature of those crucial relations.
