Pulled by God: Sound and Altered State of Consciousness in the Hälqä-Sohbät Ritual of Uyghur Sufis

In Xinjiang, northwestern China, the Uyghur Sufi Muslims practice a localised samāʿ ritual called hälqä-sohbät, which involves dhikr, singing and dancing. Participants in hälqä-sohbät sometimes enter an altered state of consciousness called jäzba, which literally means pulled by God. The hälqä-sohbät includes various sounds such as the singing of hikmät, muqam and mäshräp, as well as instrument playing. However, Uyghur Sufis often deny the role of sound as a decisive factor in triggering jäzba, but attribute its causation to the practice of dhikr, especially silent dhikr. Based on ethnographic data from Khotan, southern Xinjiang, and inspired by Leman’s research framework for music, gesture and embodied meaning, I study the relationship between sound and jäzba in hälqä-sohbät in three perspectives: a first-person perspective from the point of view of the practitioners; a second-person perspective in which hälqä-sohbät is viewed as a way of social interaction in a “me to you” way; and a third-person perspective in which I, the researcher, try to disclose and analyse the various factors at work in triggering the altered state of consciousness, such as timbre, rhythm, and physical movements.

Mäkhsum 6 is an ishan (leader of a Sufi brotherhood) with whom I became friends during my fieldwork in Khotan. He organises gatherings regularly in a khaniqa founded by his grandfather Rashdin.
Rashdin received his title as an ishan from the khaniqa of Atawullakhan Khojam Ishan in Qaghiliq, which is in turn a branch of Tiräkbagh Khaniqa in Yarkand. Molla Niyaz Ishanim, the founder of Tiräkibagh Khaniqa in Yarkand, was initiated by Majzub in Namanga and died in 1889 in Yarkand (Zarcone 2001, 127).
Tiräkibagh Khaniqa belongs to the Naqshbandiyya-Jahriyya (Änwär 2013, 36). However, Mäkhsum says he is pricinpally affiliated with the Binuri, a sub-group of Naqshbandiyya which practices silent dhikr. The term Binuri is probably a corruption of al-Banūrī, a disciple of Sirhindi who died in 1643 (Bazmee Ansari 2012). At the same time, Mäkhsum teaches students the practice of Mujaddidiyya, another sub-group of Naqshbandiyya, and organises hälqä-sohbät, a ritual of collective dhikr. This is something that I found to be common during my fieldwork among Uyghur Sufis.
According to the prominent Naqshbandi shaykh Ahmad Kāsānī Dahbīdī (d. 1542), who had had great influence on Islamic practices and thought in Xinjiang, the choice between loud and silent dhikr is not so much a distinction between different orders, but a choice of the appropriate technique for a particular Sufi according to his abilities and along his spiritual progress (Papas 2014, 32). That is the reason why both silent and loud dhikr are practiced within the same Sufi community, as is shown in later parts of this article.

Experiencing hälqä-sohbät
One afternoon, Mäkhsum invited me again to join a gathering at his khaniqa, which is located in a back room of his house. When I arrive, a group of people are already there singing mäshräp songs. All of them are men, and most live nearby and have participated in previous gatherings that I attended. In Xinjiang, Uyghur women and men are segregated in many situations, especially for religious activities. Women do not go to mosques but only pray at home. Dhikr rituals are segregated too. Uyghur women do not -EJM, vol. 20/1 (2021) - ISSN 2504ISSN -1916 participate in men's dhikr (although they can take part by standing outside the khaniqa to recite dhikr alongside the men, as I have witnessed several times), but have their own gatherings. Harris describes a khätmä ritual of Uyghur women in a village of south Xinjiang, which resembles men's dhikr rituals in many ways, except that it is based in homes rather than in khaniqas like the men's gatherings (Harris 2013, 234). However, as governmental restrictions tighten up in recent years, most of men's dhikr rituals can only take place in private rooms that double as khaniqas, if at all, and there is almost no difference between men's and women's rituals in this aspect.
I am asked to sit (or rather, kneel) beside Mäkhsum, at the end of a long table (floor) cloth, on which food, tea, and fruit have been placed. On the left-hand side of us are eight musicians, who sit by the wall and mostly play the sapayi (a percussion instrument of wood sticks with iron rings), except Ikhlim who plays the tämbur lute. On the right-hand side are a group of listeners who sway their bodies to the beat. Mäkhsum plays a dap frame drum and sings together with the others. All these seem to be quite casual, until Ikhlim starts to solo on the tämbur and sing the free-rhythm Chäbiyyat Muqam, 7 his eyes closed, while others stop playing and listen attentively. After the Chäbiyyat Muqam comes to an end, Ikhlim starts the metered mäshräp songs that are often played after muqam. The musicians also play the sapayi and join him in singing. The first song is in the 'limping beat' (aqsaq, see the fifth section for more discussion) that is characteristic of Central Asian music and part of this area's unique musical-emotional vocabulary (Harris 2014, 353 When the song is over, they change to some other songs in a regular four beat pattern. The eight or nine participants who sit on the right-hand side are becoming more involved. They rock their hands, heads and torsos with the beat. One man who sits in the middle seems the most deeply immersed. Eyes closed, he moves his hands like a conductor of symphonic music. As the singing speeds up, his whole body swings more and more; he seems entranced. The others are also beginning to sway more, although there is one man who seems wholly unaffected by it, sitting there motionless. Suddenly, the man in trance falls into a backward roll. He thrashes around and rolls on the floor, before jumping on his knees back into line and stretches his legs uncontrollably. Some people prevent him from rolling on the table, lest he break cups and hurt the others. He moves to the empty space of the room. Other people on this side of the table also move gradually to the far side of the room, where -EJM, vol. 20/1 (2021) -ISSN 2504-1916 -178 they form a circle and dance. The unaffected man now quietly moves to a corner, as if this is not his party.
Those who dance in a circle are chanting -nay, shouting -dhikr, "höm, höm…," which is one of the many ways Allah is called. While kneeling on the floor, they move anti-clockwise. Among these participants, the entranced man and the unaffected man seem to be the two extremes, with everyone else somewhere in-between. The musicians sing a few different songs and form a structure of a multi-part musical work together with the dancers' breath dhikr. The last song, Imam Husäynim, although a dastan song about the martyrdom of the Shia Imam Husayn ibn Ali, functions the same as other mäshräp songs here. 11 When Imam Husäynim ends, the dance stops. The musicians go on to sing two munajat, or supplicant hikmät, without any instrument, while many of the dancers cry, some beating their chests. We end the gathering by shaking hands, everybody with everybody else. This round of singing and dancing takes about 40 minutes.
Mäkhsum later told me that this gathering was a mixture of mäshräp and hälqä-sohbät. Indeed, this ritual consisted of mostly muqam and mäshräp rather than the hikmät of hälqä-sohbät that I attended here previously. However, the more obvious difference was the extent of trance involved. Although last time people stood up to dance and whirled at a high speed, nobody appeared to be so much in trance as Tokhtibakä, the man who rolled on the floor this time. The singing of mäshräp songs, the rhythmic breathing and reciting of dhikr, the plucking of tämbur and the clinking of sapayi rattle, the entrained collective sama dance, the crying…all these create an experience of a threshold between this and another world. In the following sections I will discuss how the altered state of consciousness happens, and what its relationships with the sounds and movements are.

First-person perspective: What does hälqä-sohbät mean to Uyghur Sufis?
I did not have a chance to talk to Tokhtibakä, the man who rolled on the floor, but I interviewed Mäkhsum, who is the master of Tokhtibakä and host of the gathering. I also talked to Nurmämät, one of the singers that day, and Niyaz, who did not participate in that gathering but has experienced many such events. Their opinions that I quote below are based on my interviews with them between 13 and 18 May 2016.

Jäzba: Pulled by God
When I asked Mäkhsum what happened to Tokhtibakä during the ritual, Mäkhsum replied that he was in jäzba (plurual form jäzibat) or was being 'pulled' (tartilish) by God, during which he could not control himself, and did not know what was happening. People had to prevent Tokhtibakä from breaking things or hurting the other participants. Niyaz says while in jäzba himself once, he broke two of someone's teeth without knowing it, and had to compensate him afterwards.
Jäzba comes from the Arabic word jadhba which means 'attraction.' According to Annemarie Schimmel, the mystical path has sometimes been described as a ladder, a staircase that leads to heaven, on which the salik slowly and patiently climbs toward higher levels of experience. But the Muslim mystics knew that there is another way of reaching higher experiences: it is the jadhba, "attraction," by which a person can be exalted, in one single spiritual experience, into a state of ecstasy and of perfect union (Schimmel 1975, 105).
For Uyghur Sufis, the behaviour of Tokhtibakä during the gathering were representations of getting closer to God, which is the aim of hälqä-sohbät. Niyaz compared the state of someone in jäzba to that of a man who falls in love with a girl but cannot see her. When someone becomes mad through longing for Allah, he does not care if people stab or kill him; those in jäzba are unaware of themselves and what is happening around them. Mäkhsum used the metaphors of a bird and cage to explain Tokhtibakä's behavior: a man's body is like a cage, and his soul a bird in the cage. When the soul is agitated, it tries to break out of the cage like the bird. That is why people do things like rolling on the floor. Ahmad Kāsānī Dahbīdī, the famous Naqshbandi Sufi introduced earlier, wrote in his Risāla dar samā' (Treatise on audition) that "In samā' the Sufi is deprived of power and is perturbed; he may scream as he feels as if he is hearing the covenant, and the voice of the singer permeates the Sufi's soul" (Papas 2014, 35). Kāsānī also used the metaphor of cage: "When his soul starts to take off, the cage of existence is closed; so he yells and cries; he wants to return to his origin but cannot. Several times, as a matter of prudence and wisdom, God has to close the cage of body" (Papas 2014, 35). It seems that screaming or In Harris's article on Uyghur büwi women ritual specialists, one büwi is quoted as saying "The oil is sizzling in the pot [qazan kizip ketti] . . . Allah's passion (ishqi) is like the hot oil in the pot, their passion for Allah is so strong" (Harris 2014, 354). Mäkhsum remarked that the state of sizzling is the beginning of jäzba, when you feel like boiling but you still know what you are doing. If it goes on, the water is not only boiling but also spilling, and you lose control of the rhythm of your movements. That is when jäzba happens. Compared to the women büwi who do not have access to formal religious training, Mäkhsum has specialised knowledge and vocabulary to talk about the altered state of consciousness as part of Sufi practice. He has acquired this knowledge through the education he received from his family lineage which he can trace back to Yarkand and further to the Ferghana Valley, as discussed in the first section.
Mäkhsum also observed that Tokhtibakä's behaviour (rolling on the floor) represents an elementary level of jäzba, called dawajit, apparently from tawajud, which Zarradi, a disciple of the foremost Chishti saint of Delhi Awliya (d. 1325), defines in his treatise Usul as-sama' as "graceful movement that voluntarily emanates from the listener when he is overcome by sama'" (Lawrence 1983, 97). It is noteworthy that there are Chishtis in Khotan, with whom Mäkhsum keeps a friendly relationship.
Mäkhsum said dawajit is like a messenger who gives some news of jäzba, while when a man is in real jäzba, his soul is separated from his body. As a result, he may lose consciousness (bihosh bolush) for one or two hours, and even die. It is similar to the people recorded by Mojizī in his Tawārīkh-i mūsīqīyūn ("Histories of musicians"), a rare treatise about musicians from the 19 th -century Xinjiang, who have died in extreme emotional states in music (Sumits 2016, 152-53), although Mäkhsum would deny that death in jäzba is caused by the sound. Mäkhsum has witnessed the death of several people in jäzba, and he considers death in jäzba to be the most elevated kind of death -to die with one's true feelings and iman (faith).
Mäkhsum has experienced losing consciousness during hälqä-sohbät himself, but he cannot recount the details because he lost his consciousness. That happened when he was dancing. When he is the hapiz, the lead singer, he has to control the ritual by singing hikmät and cannot engage himself in dance or totally lose his consciousness, although sometimes he cannot help crying while leading others singing hikmät.
For Mäkhsum, dance movements help one to enter jäzba, because when one's body suffers, one's soul will be released. The more intensely one moves one's body, the freer and happier one's soul will be.
On the contrary, musical sound does not affect jäzba, except through the implications of the lyrics of hikmät, muqam, or mäshräp songs. The hapiz can choose any piece from these repertoires to sing, as long as he can "match it well." Whatever the hapiz sings, it does not make a difference to the state of consciousness. Mäkhsum says no particular piece makes people more excited, although people certainly react to musical sound, as "even a bird reacts to a sound," but these are not important in hälqä-sohbät.
What is important is dhikr.

Loud and silent dhikr
Hälqä-sohbät contains collective loud dhikr, but one must practice silent dhikr (zikiri khufi) on one's own in order to achieve jäzba. Mäkhsum practices silent dhikr every day when he is alone. His master taught him to recite God's names or the shahada (profession of faith) at least 5,000 times a day. He started when he was 15, and has been doing it for more than 30 years.
Mäkhsum considers that while it seems people reach jäzba through singing and dancing, silent dhikr is the base. People must recite as much silent dhikr as possible, or they will not get the knowledge of Sufism, and will not reach jäzba, as was the case of the person who sat stiffly during this ritual. Although occasionally someone who has not done much silent dhikr can reach jäzba in hälqä-sohbät, usually one has to recite silent dhikr for years before he can enter jäzba. It is not easy, and among 100 people who do dhikr, only five or six get God's favour and reach jäzba. On the other hand, one can reach jäzba through silent dhikr without singing or dancing at all. Therefore, if you can reach jäzba, it is because of the silent dhikr you have done, not the tunes, instruments, or movements in hälqä-sohbät. If you have not recited silent dhikr enough, you will not be able to reach jäzba in hälqä-sohbät even if you sing and dance.
Tokhtibakä has reached jäzba because he has done a lot of silent dhikr.
Loud dhikr (zikiri jähri) developed out of the challenges of practicing silent dhikr. Participants would sometimes be distracted or even fall asleep, so some saints encouraged people to gather together and practice dhikr with sound in a circle. The term hälqä-sohbät evolved because people began to practice dhikr by sitting in a circle and talking. Mäkhsum, who is a knowledgeable and well-read Sufi, referred me to two texts about dhikr. First, in the Qur'an, Sura 33:41 says "yā-ʾayyuhā lladhīna ʾāmanū dhkurū llāha dhikran kathīra," "O you who believe, remember God often" (Jones 2007, 681). Second, according to the Hadith, Ibn Umar reported that the Prophet said: "When you pass by the gardens of Paradise, avail yourselves of them." The Companions asked: "What are the gardens of Paradise, O Messenger of Allah?" He replied: "The circles of dhikr. There are roaming angels of Allah who go about looking for the circles of dhikr, and when they find them they surround them closely." 12 Over time, people would come together and practice vocal dhikr as a way of disciplining and focusing the mind. Using this group approach, they would not get distracted. The more one longed for (seghinish) Allah, the more agitated one became in loud dhikr, and consequently, the more expressive one became. Therefore, when an individual practiced loud dhikr together with other people, the individual felt delight (huzurlinish) and excitement (ishtiyaqi örläsh) and could express his emotion more thoroughly.
When people gathered together to practice dhikr, the hapiz sang about heaven, hell, Allah, and Prophet Muhammad to preach or edify people. Their attention was focused on such contents, and they could better discipline their minds in loud dhikr.
So, in Makhsum's view, the individual obtained more knowledge of God through reciting dhikr.
When he had acquired enough knowledge and joined a hälqä-sohbät, the musical sound (vocalisations such as loud dhikr, hikmät, muqam, and mäshräp songs, and sounds of musical instruments if any are involved) helped him to concentrate on God and made his longing for God stronger. When doing dhikr, the individual's soul becomes agitated and wants to break away from the body, so he begins to dance. Bodily movements intensify his emotion, until it is boiling and spilling, and he cannot control his behaviour anymore. At that moment, he will not be able to move his body to the rhythm and may lose his consciousness. Mäkhsum considers jäzba to be decided by God. The individual is not able to enter jäzba simply if they desire to, neither can the individual refuse jäzba if they are in it. During hälqä-sohbät, whichever part of the body God's light falls upon, that part moves the most. If God's light falls on one's eyes, one cries. If God's light falls on one's mouth, one recites God's names the most. If God's light falls on one's legs or arms, those parts move the most.
Mäkhsum, Nurmämät, and Niyaz have all experienced jäzba and have lost their consciousness in it several times, but overall, jäzba is rare. Most of the time, Mäkhsum does not reach it, although every time he practices hälqä-sohbät, he feels strong emotions in it, and after one or two hours of hälqä-sohbät, he is in a good mood for several days and does not feel distracted by worldly affairs.

Second-person perspective: The passing of knowledge and social interaction
In Leman's methodology of research on music, gesture, and embodied meaning, the second-person perspective tries to look at gesture in terms of other people's engagement with music. In this aspect, gesture appears as a mediator for music-driven social interaction or as the vehicle through which a 'meto-you' relationship is established in space and time, through musical engagement. Thus, gesture is seen as the expression of a communicative act, rather than an expression of 'my own' personal experience (Leman 2010, 142).
This suggests a helpful perspective on the social interaction that happens in the process of hälqä-sohbät.

Learning and social interaction in hälqä-sohbät
During this gathering in Mäkhsum's khaniqa, musicians sat on one side, while dancers sat on the other side. According to Niyaz's division, they were "those who know how to sing" and "those who don't" respectively. It is true that those dancers did not participate in the singing, and in the first part of the ritual they were basically listening to the musicians' singing in silence. I am not sure whether they knew how to sing those songs, however, it seems that most of them, if not all, were Mäkhsum's students. As such, they are in a lower position in the hierarchy compared to the musicians who are mostly Mäkhsum's friends.
The dancers' relatively low status is also shown by the fact that they did not stand up to dance, because the singers did not stand up. That is the rule -if the musicians do not stand up, the dancers cannot stand up either. I remember that during the previous gathering I attended at Mäkhsum's khaniqa, it was Mäkhsum who first stood up at one point, then the others followed suit to stand up and dance. In the several hälqä-sohbät that I attended were always adepts and neophytes among the dancers. It seems that for the neophytes the most direct and accessible way to learn about sounded Sufi practice is to imitate others' actions at a hälqä-sohbät, including reciting dhikr, dancing, crying, and possibly, trancing.
Thus, each hälqä-sohbät is also a learning session for the student about how to behave. The student is often given some cues about what to do during the gathering. For example, the student should follow the hapiz's physical position. When Mäkhsum stands up, it means that the student should also stand up, or when Mäkhsum does not stand up, the student should remain seated. The rhythm can be another cue for the student's bodily movement. During the unmetered muqam, he should remain quiet with some gentle bodily movements, while the four-beat mäshräp songs are a cue for more obvious bodily movements and sama. The end of mäshräp songs indicates the end of the sama dance, while the subsequent singing of munajat means that crying is a proper behaviour at this point.
Sometimes the cues come directly from the master, sometimes they may come from people associated with the master. For example, after Tokhtibakä went into trance, it was Mäkhsum's friend Tursunturäm who indicated that the others should move away from their original line and form a circle to dance together. Such a cue may also come from other students. For example, when Tokhtibakä went into trance, that could be taken as a cue for other participants that jäzba was a permissible reaction to the ritual at this stage, although that was not necessarily everybody's reaction. At other points there were no obvious cues, but the participants would probably be acting according to past teachings from the master, for example, when preventing people in trance from hurting themselves and others or helping those still in strong emotional states at the end of the ritual.
The hälqä-sohbät is also a process of social interaction from the perspective of coordinating sound and movement. While the dancers' movements largely depend on the hapiz's singing, the hapiz also decides his singing according to the state of the dancers. Mäkhsum observed that if he sees that dancers can dance fast, he will sing fast, and if not, he will sing more slowly, as togetherness is important for Sufis. From the perspective of social interaction, hälqä-sohbät provides an opportunity for these Sufis to enhance the solidarity of their community.
Of course, rituals are not the only way for the student to learn knowledge. As Mäkhsum says, what is more important is the daily practice of dhikr, and one has to practice it every day under the guidance of the master. A lot of such knowledge is clandestine and only shared between the master and his disciples. I am very grateful to Mäkhsum for having shared some of his knowledge with me.

The barrier in my learning
Mäkhsum primarily belongs to the Naqshbandiyya but has studied the practices of all the four Sufi orders of Khotan. This makes him particularly knowledgeable about dhikr. According to him, there are three kinds of dhikr: zikiri jähri, the loud one; zikiri khufi, the silent one; and zikiri sir or "secret dhikir," which is considered to be the highest level of zikiri khufi. When one does zikiri khufi, one should remain quiet and not even move the tongue, but use the mind to do dhikr. This kind of dhikr is called lisanihal, an Arabic term which means language of the mind. In zikiri khufi of Naqshbandiyya, one can recite either Allah's names or the shahada in the mind. In the latter, one says "lā ʾilāha ʾillā-llāh" ("there is no god but God") for twenty-one times without breathing, then exhales and says "muhammadur-rasūlu-llāh" ("Muhammad is the messenger of God"), then inhales. This part is called näfi isbat. Although this is a practice of silent dhikr, I still find it to be embodied as it involves breathing. In addition, it involves conceptual embodiment. To reach zikiri sir, one should recite Allah's name in the mind language and use it to fill eleven parts of the body. This is the first step, called khätmä sir. Then one says Allah. This step is called säyri ilällah. Then one does näfi isbat. From this step onwards, one is in zikiri sir. When I wanted to ask more questions about dhikr, Mäkhsum considered that he had said too much and that very few people knew about these practices.
This situation happened to me several times with Mäkhsum, Nurmämät, and Niyaz, that they did not want to answer my questions. Sometimes it might have been because of the ineffability of the subject.
Sometimes I probably touched upon some knowledge that is not shared with outsiders. It is common that Naqshbandi masters do not talk about issues of spiritual practice openly. For example, Kāsānī considered that Sufi teachings are not appropriate or even helpful for all disciples, and he often did not describe the spiritual states of Sufism during his conversations with his disciples (Papas 2014, 32 When I asked why Tokhtibakä danced more intensely than others, Nurmämät, who was singing at the time, said, "only he himself knows." And when I asked Nurmämät if he was in a similar kind of state, he replied "one thousand people have one thousand feelings." These Uyghur Sufis have been very generous in sharing with me their knowledge, but it must be difficult for them to express their mystical experiences in words, especially to someone who is not a practising Sufi. Similarly, Abū Sa'īd b. al-'Arābī viewed the discussion of wajd (a state of ecstasy, see the next section for more) as problematic: "How can one describe something which has no description other than itself, and to which no witness can testify other than itself?" (quoted in Avery 2011, 28). Sufi knowledge must be achieved through everyday practice of dhikr, the master's guidance, and one's observation, imitation, and experience from rituals like hälqä-sohbät.

Third-person perspective: How does this all happen?
Earlier in this article I described the ritual from a third-person perspective according to my experience as an observer. Next, I will analyse the ritual from a third-person perspective, with a focus on the sound and movement signifiers.

Sound and audition
People like Mäkhsum tend to relegate the impact of musical sound in Sufi rituals to a minimal degree, saying that jäzba does not have anything to do with musical sound, but results from one's practice of silent dhikr. However, it is hard to imagine dhikr that exists without any sound. Dhikr, whether of Allah's names or of the shahada, is based on the most important practices of Islam -Qur'an and salah (the five daily prayers), which have always existed foremost in sounded forms. The Qur'an, meaning 'recitation,' was initially transmitted orally from Prophet Muhammad to his disciples. Much of the Muslim's religious life is spent in sounded practices, which enjoy a higher status than vision. As During writes: Vision particularly refers to the "imaginative" world and is hardly taken into account as part of the world of the senses, since in the Islamic gnosis there is no meditation on icons, mandalas, or plastic forms.
Contrary to this, hearing, which is the other fundamental axis to Revelation, was privileged as a kind of contemplation, particularly in relation to the Sufi practice of the samâ', the spiritual concert. (During 2010, 552) The meaning of dhikr, as well as hikmät and mäshräp songs, has been embedded in sound. It is probably particularly because of the close ties between these texts and sound that Sufis tend to take the factor of sound for granted and overlook its impact. The daily practice of silent dhikr (which is recited internally but is still 'recited' with imagined sound) has molded the practitioner's acceptance of God, intensified his longing for God, and laid the foundation for an altered state of consciousness. During the hälqä-sohbät, collective musical sounds and bodily movements constitute important sensual stimuli that magnify the sound factor and embodiment that have always existed in Sufi practice of dhikr, thereby magnifying the original stimuli. As Kāsānī wrote, "men's hearts are like flints for fire, which have to be struck by fine sounds (alhān-i tībat) since these fine sounds are the sounds of Truth" (Papas 2014, 35). Like striking flints for fire, musical sounds and bodily movements catalyse the altered state of consciousness.
Signifying 'hearing,' the Arabic term samāʿ indicates that the action of listening is of utmost importance in the ceremony, because it is a means of revelation. In Qawwali, a South Asian form of samāʿ, the focus is on the listener and on his ability to draw spiritual benefits from the sound, rather than on the sound itself (Qureshi 2006, 108). That is the reason why Qawwals or singers of Qawwali are of a lower status, as they are considered to be paid service professionals, who are only a medium through which the listener gains spiritual benefit. In this hälqä-sohbät under discussion, however, the musicians are of higher status in the hierarchy, not because of their music, but because the musicians are led by the religious leader and ritual organiser and consist of his associates.

Sama: bodily movement in hälqä-sohbät
In hälqä-sohbät, the student's/dancer's bodily movement while reciting dhikr is called sama, which sounds very similar to the Arabic term samāʿ. In the gathering under discussion, people did sama while kneeling on the floor, but in most other gatherings that I witnessed they also stood up to whirl. In addition, I have also seen people do sama at a wedding. Sama is essentially a form of dance, but the way people talk about it indicates that a distinction is made between sama and common dance.
In the Uyghur language, ussul is the common word for dance, but it is not used to describe sama.
Sometimes Mäkhsum uses räqs, a term borrowed from the Arabic word for dance, raqs, to describe sama, but most of the time he simply uses sama as if it is self-explanatory. Räqs is not commonly used in the Uyghur language, and may sound more solemn because of its Arabic root, although the Arabic word raqs does not have any religious association. 13 Selish, the Uyghur verb for sama, differs from the common Uyghur verb for dance, oynash, which means to 'play.' Selish means to put, to place into. Although sama has the form of dance, it is not regarded as dance per se because of the ideological distinction between bodily movements for religious reasons and dance as entertainment. Similarly, the musical sound in hälqäsohbät is not considered to be music per se.
13 I am grateful to Iskandar Ding who pointed this out to me. There are mainly two opinions about the origin of the Uyghur term sama. One is that sama as a dance was formed during the period of time when Islam spread to Kashgar and is based on the movements in the rituals of the Yasawiyya order (Zhongguo 1998, 100). The second view is that the sama dance emerged from the dance movements of shamanism, which was practiced by the ancestors of Uyghur people (Djumaev 2002, 973;Wang 1996, 81). I see the first opinion as more credible, as sama is still practiced in hälqä-sohbät, which is indicative of its ties with Sufi rituals. Most Chinese scholars who take the second view that the sama dance is related to shamanism do so primarily because of similarities of the terms sama and "shaman" (saman in Chinese) (Wang 1996). However, the Tungusic term "shaman" is not used in the Uyghur language. The kind of spirit medium among the Uyghurs that is closest to a shaman is called a bakhshi or pir, and the dance they do in healing ceremonies is called pir rather than sama.
Given the ties of sama with Sufism and its similarity in pronunciation to samāʿ, one may wonder why the word sama is used by Uyghur Sufis to designate only the bodily movement, not the listening or sound element of the ritual that is associated with the Arabic word samāʿ.
In Kashgar, a variation of the sama dance has developed into a large-scale group dance performed by local men on festival days outside the famous Häyitgah mosque and has been recognised by the Chinese government as a national-level Intangible Cultural Heritage (without the religious association of sama being mentioned). 14 Harris depicts one of a series of pieces performed on the naghra (paired kettle drums) and sunay (shawm) to accompany the circling sama dance in Kashgar. According to Mämät Tokhti, a musician Harris interviewed, whose family has been playing music for the sama dance outside the Häyitgah mosque for over 100 years, the Kashgar sama tradition started with a Sufi mendicant called Täykhan Khojam from Namangan in the Ferghana valley who performed the sama dance outside the Häyitgah mosque after Friday prayers. 15 The musician did not mention the music that was played with sama at that time. Given the fact that Sufism spread to Xinjiang from the Central Asian regions to its west, is it possible that Uyghur Sufis adapted the samāʿ ritual to local music, creating the localised hälqäsohbät ritual, in which only the dance from the original ritual was kept, together with the name sama, and then the dance also developed into a public outdoor practice for festivals? More comparative study between Uyghur and other Central Asian Sufi rituals needs to be done to shed light on this question.
There is a difference between samāʿ and dhikr, a ritual of remembering God through reciting his names, regarding the participation of the subject. The former is about passive listening, while the latter more about active remembering (During 2010, 555). The ritual of collective dhikr probably appeared a few centuries after samāʿ, and the two have often been combined with each other in various forms (During and Sellheim 2012). The hälqä-sohbät under discussion is an example of such a combination -in the first section the musicians sing and play instruments while the 'audience' are relatively passive listeners, while in the second section the 'audience' not only listen to the sound but also actively recite dhikr and dance.
However, it is not easy to divide the samāʿ and dhikr in a combined ritual, as During suggests that samāʿ and dhikr complement each other, and "in ritual practice, hearing is also remembrance and active invocation that frequently incites to movement and dance, just as dhikr also is the hearing of invocation" (During 2010, 555).
Dance seems to have often been a part of samāʿ, although not always in a prescribed way. Among the elements that the great twelfth-century Muslim scholar al Ghazālī lists, regarding the etiquette for listeners, is the need for stillness, and that one should not "look at others, clear the throat, yawn, clap, dance or talk. If one is overcome by wajd resulting in involuntary movements, one should return to a quiet state as soon as this passes" (Avery 2011, 46). Likewise, Ghazālī's brother Majd al-Dīn (d. 1126) recommended that when a listener comes into ecstasy, he should refrain from rising until he is overpowered (Avery 2011, 189). Here dance is only permitted when it is an involuntary act, which means that if dance occurs, it is the result of an altered state of consciousness rather than the cause. Of course, when one is already in an altered state of consciousness, it is very possible that dance will intensify the state.
In the case of Qawwali, "conspicuous or unrestrained self-expression" of the uninitiated or spiritually less committed is frowned upon (Qureshi 2006, 121). Qureshi lists three categories of expressive responses of Indo-Muslims. The first category, manifestations that symbolise the Sufi attitude, consists of five actions, from bowing the head to rubbing the face and touching the eyes. The second category, standard manifestations of enthusiasm or mild arousal, consists of seven actions from moving the head to involuntary movement, such as twitching. The third category, standard manifestations of strong arousal, consists of nine actions, which are, in order of increasing intensity, "sudden, uncontrolled movement, twitching, jumping", "weeping", "arms raised -both", "shout", "stand up", "dance", "walk", "fall down", and "die" (Qureshi 2006, 121). As can be seen from the last category, dance is one of the most intense manifestations of strong arousals, only before "walk", "fall down", and "die." Thus, it has to be dealt with very carefully. That is probably the reason why dancing at a hälqä-sohbät usually happens under some kind of direction from the hapiz and his associates.
If samāʿ is primarily devoted to listening and does not inherently involve dance, bodily movements seem to be more legitimate in the dhikr, a ritual of remembering God through reciting his names. Dhikr involves actions more active than listening, especially in collective dhikr rituals when a group of people recite God's names in coordinated ways which leads naturally to bodily movements.

Sound and altered states of consciousness in hälqä-sohbät
Since the early period of Sufism, altered states of consciousness have been sought after by Sufis because these were seen as signposts or way stations on the mystic's path towards divine unitary experience (Avery 2011, 3). Accordingly, jäzba is what Uyghur Sufis hope to achieve through hälqä-sohbät. The term jäzba has taken a local set of meanings compared to its Arabic root, jadhba. Among Uyghur Sufis, jäzba is used to describe not only the action of 'attraction' but also a state of consciousness caused by that attraction, which is referred to in Arabic as wajd, a euphoric state of trance resulted from the samāʿ ritual (Lucas 2012, 111). Wajd is not used by Uyghur Sufis, except in the form of dawajit, which comes from tawajud, an inflected form of wajd with the meaning of "expression of wajd". Tokhtibakä's rolling on the floor is a form of dawajit, showing that he has received some message of jäzba.
Gilbert Rouget, in his encyclopedic book on music and trance, distinguishes "trance" (obtained by means of noise, agitation, and in the presence of others) and "ecstasy" (attained in silence, immobility, and solitude; Rouget 1985, 7), and, under trance, "induced" trance (caused by an external action) and "conducted" trance (caused by one's own action; Rouget 1985, 286-90), based on data on altered states of consciousness from different cultures across the world. For Rouget, the Sufi collective dhikr involves conducted trance, because in order to enter trance the subject must recite the dhikr, sing, and dance, or, in other words, he has to be his own musicant. On the other hand, in classical samā', people are induced to trance because they are musicated rather than being the musicant (Rouget 1985, 288).
However, reality is not always so clear-cut. For example, in the hälqä-sohbät under discussion, some participants were both musicated (when they listened to the singing and instrument playing) and musicants themselves (when they began to actively recite dhikr). In addition, the explanation of jäzba by Uyghur Sufis relates to their practice of silent dhikr when they are alone, thus linking "ecstasy" with "trance." It is also noteworthy that the silent dhikr, categorised as "ecstasy" by Rouget because of its "immobility," is actually done with the movement of breathing, which does have an embodied dimension and physiological effect. It seems difficult to interpret the cultural-specific practices of altered states of consciousness of Uyghur Sufis with a universally generalised theory, and the meanings of different practices have to be found in their particular contexts.
Although some people enter jäzba while chanting dhikr and dancing in hälqä-sohbät, Sufis like Mäkhsum deny that musical sound has any impact on the state except through semantic meanings of the lyrics. Such an opinion may have been based on Muslim societies' ambivalent attitudes towards music.
In her article about music, musicians and Muslim law, al Faruqi presents a table of the hierarchy of genres of handasah al ṣawt (sound-art expression) of Muslim societies. At the top of the hierarchy is the category These are not designated as mūsīqā so as not to be confused or associated with unsanctioned types of handasah al ṣawt. Below non-mūsīqā are three other categories of legitimate (halal) forms of mūsīqāfamily/celebration music, 'occupational' music, and military music. Lower on the hierarchy are controversial forms -vocal/instrumental improvisations, serious metered songs and instrumental music, and music related to pre-Islamic or non-Islamic origins. At the bottom of the hierarchy is sensuous music associated with unacceptable contexts, which is illegitimate (haram).
Like in other Muslim societies, among the Uyghurs religious chants such as Qur'an and hikmät are not considered muzika, the Uyghur word that originated in the Arabic term mūsīqā. Mäkhsum's definition of muzika is the sound of musical instruments, which is often profane in nature and thus not to be encouraged. Vocalisations in hälqä-sohbät, on the other hand, are mostly religious and thus not muzika. Hikmät, for example, is sung poetry, while dhikr, recitation of God's names, is even further away from muzika. The tämbur lute was allowed in the hälqä-sohbät under discussion because of the influence of Chishtiyya, one of the four Sufi orders in Khotan, who includes instruments as a part of dhikr and makes use of them to serve religious purposes.
Given the often low opinion of music in Islamic societies, it is not surprising that Sufis like Mäkhsum would tend to neglect the importance of musical sounds in religious assemblies except with regard to their semantic meanings. Lyrics of hikmät, muqam, and mäshräp songs are certainly important in helping to concentrate the participant's mind on religious thinking. At the same time, hälqä-sohbät is a way for Uyghur Sufis to both recollect God and to recollect their past, in the verses of such Central Asian poets as Ahmed Yasawi and Shah Mashrab. However, does musical sound really have nothing to do with the altered state of consciousness in hälqä-sohbät?
Avery proposes that the quality of voice or the purely rhetorical element of the reciter's delivery of the verse might be a causal factor, as "the qualities of a voice such as resonance and sonority can have dramatic emotional effects on the listener" (Avery 2004, 93-94). He illustrates his point with a story about a man who fell unconscious upon hearing a Qur'anic verse subsequently regaining his consciousness through hearing a recitation of the same verse. Although Mäkhsum does not refer directly to the impact of the voice, he does say that a good voice is important for a hapiz.
My voice is so so. But God has given Osman Haji and Niyaz Qarim Haji (two famous hapiz of Khotan) perfect voices to be hapiz. Some hapiz don't have great voices, but it is certainly an advantage for a hapiz to have a good voice. The books of Rumi say that longing for Allah lies somewhere between thick and thin voices. Some say a good voice comes from the breathing of God. 16 talking about voice in an aesthetic sense, as that is not what he or a hälqä-sohbät is concerned with. Rather, he is talking about the voice as a means to deliver the religious message, so the quality of the voice matters in hälqä-sohbät, whose aim, in Mäkhsum's view, is to express one's longing for God and get closer to God.
A voice "between thick and thin" seems to him suitable to deliver the message of God, so that, in a way, it is like "the breathing of God." Mäkhsum's voice may be not perfect, but his resonance and sonority, which must have been tempered by decades of practice in reciting the Qur'an and himkät, makes him a good hapiz.
The second aspect of the musical sound in hälqä-sobät that I propose to discuss in relation to the altered state of consciousness is rhythm. The hälqä-sohbät under discussion started with the unmetered Chäbiyyat Muqam, which continued for about six and a half minutes and developed into metered mäshräp songs. The first mäshräp song was in the 'limping beat,' which can be transcribed into seven-eight beat with hemiola, for example, by Chinese ethnomusicologist Zhou Ji (Shinjang 1993, 96). However, rather than a seven eight beat, it is more natural to conceive of this rhythmic pattern as consisting of two parts with different time units. One beat in the latter part is about one and a half times the duration of one beat in the former part. Because this music actually alternates between two different beats, it can feel like floating between two worlds. The in-between state created by such an alternation may contribute to an altered state in the listener, as it provides a blurred feel.
This song of 'limping beat' lasted about five minutes. The ensuing songs were in the four-beat, with accelerating tempo and reached the climax during the song Imam Husäynim, which marked the end of the mäshräp songs and collective dhikr. These four-beat songs lasted about 24 minutes. Tokhtibakä lost control of himself and started rolling on the floor during the third mäshräp song, about seven and a half minutes into the four-beat songs. I did not observe any obvious cue for Tokhtibakä to go into trance, and the fact that he was the only one that went into trance at this moment suggests this was due to his particular subjectivity. However, the timing of his paroxysm was not random. He had been dancing throughout the duple beat and his dancing had gradually become more intense. He went into his paroxysm at the point when the beat had consolidated into a stably accelerating four-beat. He had been moving his body since the unmetered muqam, but his movements became more regular once the metered mäshräp songs began, especially in the four-beat part, which became the most effective rhythmic stimulus.
This is easily explainable as the four beat is easier to coordinate regular movements with than the free rhythm of muqam and limping beat of the first mäshräp song.
This stable four beat is similar to that of dhikr, whether in the form of repeating "Allah" or of the four-syllable shahada "lā ʾilāha ʾillā-llāh" and can probably help to induce the mentality of Tokhtibakä from when he was reciting silent dhikr. As this beat went on, Tokhtibakä waved his one hand and then both hands, back and forth. Listening to the religious verses sung and played on monotonous beats apparently helped him to concentrate his mind on the transcendent and created religious arousal in him, resulting in his bodily movement, which in turn intensified the arousal. The first four-beat song was at the tempo of about 80 beats per minute, which rose to about 95 beats per minute when Tokhtibakä fell into trance and began to roll on the floor. As people formed a circle to dance in a group, the mäshräp songs continued to increase in speed, until the last song, Imam Husäynim, reached about 125 beats per minute, after which this round of singing ended.
The acceleration of music tempo is a triggering factor in trance in many music cultures around the world (Rouget 1985, 81-84). In happens also in dhikr in Aleppo, Syria, where Shannon notes that rhythmic acceleration contributes to the sense of spiritual transformation by altering the participants' perception and experience of temporality (Shannon 2004, 388). There is no universal meaning of a musical process, but that does not mean that music plays no role in trance in a particular setting. In each cultural context, music provides a system of meanings upon which people whose thinking has been shaped by that culture can draw interpretations of musical processes, which are then taken as cues for body movements and state of consciousness.
As analysed earlier, this ritual is a combination of samāʿ of more passive listening and dhikr of active reciting in a very embodied way. This embodiment refers to their dance and to the very act of recitation -däm zikiri or breath dhikr, in which God's names are vocalised in a guttural way, with exaggerated breathing. Rouget sees breath dhikr as a very particular form of autoexcitation, since it makes use of breathing, a certain overstimulation of the vocal cords, a very accentuated rotary movement of the neck and head, and a whole variety of physical movements that must certainly consume (or liberate?) a great deal of energy (Rouget 1985, 317). To this very physical process of recitation I will add another overstimulation that may contribute to the altered state of consciousness in the hälqä-sohbät -the playing of the sapayi, which is an instrument favored by Uyghur Sufis and played at almost all of their musical activities. In the ritual under discussion, every musician played a pair of sapayi, except Ikhlim who played tämbur and Mäkhsum who played dap. Sapayi is loud and played close to the players' ears. In a ritual that may last a few hours, it is very possible that the deafening volume of sapayi will have physiological effects on people's state of consciousness, especially for the players of the instrument.

Conclusion
Hälqä-sohbät is a localised form of samāʿ and dhikr practiced by Uyghur Sufis, who live in the remote area of Xinjiang in Northwestern China. In the musical sounds and bodily movements that are performed in hälqä-sohbät, participants seek to enter jäzba, an altered state of consciousness in which one feels the attraction of God.
From a first-person perspective interpretation of the mechanism of jäzba and, from the point of view of the practitioners, it is years of practice of silent dhikr which prepares one for jäzba. In hälqä-sohbät, the collective recitation of loud dhikr and sama movements help one to get rid of distractions from worldly affairs, and to concentrate the mind on God, which enables some participants to enter the state of jäzba.
From a second-person perspective, the hälqä-sohbät is a way of social interaction in a 'me to you' way, a means for both the master to impart knowledge to the disciple and for fellow Sufis to support each other. It presents a kind of social embodiment that strengthens the hierarchy and solidarity of the Sufi community.
From a third-person perspective, there are various factors at work in the hälqä-sohbät in triggering the altered state of consciousness. Uyghur Sufis tend to deemphasise the role of sound in reaching jäzba because sound is not an end in itself, but a means of getting closer to God. This points to the difference between sound as a Sufi practice and as music. However, musical analyses can reveal particular parts of sound that may contribute to practitioners' emotional changes. The limping rhythm, for example, which consists of regular interchanges of two different beats and time units, can suggest a feeling of being between two worlds. The timbre of the the hapiz's voice is an important vehicle in conveying the religious message. The acceleration of tempo may intensify expectations, while people's physical movements when reciting the dhikr and dancing, alongside the loud volume of the sapayi rattle, are sonic cues that can all have physiological impacts on the practitioners.
Overall, hälqä-sohbät is a ritual that combines local musical culture and influences from regions to the west of Xinjiang from where Uyghur people received Islam. It represents a means by which the Uyghur Sufis, who live on the margins of the Islamic society, can make sense of their religion through a localised practice of the samāʿ ritual.