An ambiguous spirit dream and Tuareg-Kunta relationships in rural northern Mali/Targat n kel essuf enten kel tamajaq enten Kunta, dagh tenere n arewa n Mali/Une reve ambigue et les rapports entre le Touareg et les Kounta au milieu rural au nord du Mali/Um sonho esplrita amblguo e relagoes Tuareg-Kunta no norte rural do Mali

Citation metadata

Author: Susan Rasmussen
Date: Summer 2015
From: Anthropological Quarterly(Vol. 88, Issue 3)
Publisher: Institute for Ethnographic Research
Document Type: Article
Length: 11,083 words
Lexile Measure: 1540L

Document controls

Main content

Abstract: 

Spirit-themed dreams--whether visitations or soul travels--are altered states that pose special challenges to the anthropologist. They are rooted in subjective personal experience, but their meanings are shaped by culturally and socially constructed religious knowledge and power. This article offers a case study of one Tuareg man's dream of an elderly female spirit, based on my fieldwork in a small rural community in northern Mali, where residents include Tamajaq-speaking, Muslim, semi-nomadic, and socially stratified Tuareg, as well as Arabic-speaking Kunta, with whom the Tuareg residents have cultural ties but also long-standing tensions. This case study does not represent all local experience, but does illustrate an individual's interpretation of widespread social, religious, and political predicaments in that village and region. Many Tuareg, whose own cosmological and social systems are at odds with but also influenced by those of North African Islam, express ambivalence toward Kunta efforts to impose more "orthodox" Islamic practices. I show that this spirit dream expresses both religious influences and ambivalence toward them. I explore the meaning of reticence surrounding the spirit's name and analyze the spirit's multiple and interpenetrating identities. More broadly, the article shows how dreaming, though psychological and subjective, is also situated in wider contexts of historical, sociopolitical, and religious encounters. Further, it reveals how the cultural models pondered in dreaming and waking interpretations of dreams can involve cultural dissonance and mixed sentiments, rather than agreed upon models, and can refer to past, present, and future concerns. I argue that several possible "kaleidoscopic" identities emerge to describe this spirit figure, thereby reflecting interweaving and contentious models of gender and religion in the dreamer's community. The article shows the importance of ambiguity in dreams and ambivalence in dreamers as an index of intersecting and colliding meaning systems, contested social changes, and opposed interests. [Keywords: Dreams, spirit possession/mediumship, religion, gender, Africa]

Full Text: 

This article approaches dreaming through a case study revealing one dreamer's subjective experience of a community's wider social divisions and conflicts, which revolve around religion, gender, and politics. Specifically, I analyze a Tuareg man's dream about an elderly female spirit in a small rural village in northern Mali. There, most residents are Tuareg (sometimes called Kel Tamajaq, after their Amazigh or Berber language), predominantly Muslim, semi-nomadic, and socially stratified, and live alongside an Arabic-speaking group of Mauritanian origin, the Kunta. The latter are also Muslim but are more influenced by pan-Islamist reformist piety movements, and tend to be more economically prosperous and politically dominant than many Tuareg in that region; their Islamic scholars (popularly called "marabouts") interpret the Qur'an for some Tuareg residents in the village. Relationships between these groups--who view themselves as distinct--are historically and currently characterized by both conflict and cooperation.

The exact identity of the spirit in this man's dream is ambiguous; her precise name is never explicitly mentioned, only hinted at euphemistically. She is called an "old woman," and is classified under a generalized Tamajaq gloss as one of the Kel Essuf spirits of the wild, solitude, and nostalgia. This designation, I show, powerfully resonates with wider North African (Maghrebian) Sufi Islam and with Tuareg cultural symbolism and cosmology that diverges from "orthodox" Islam. It also conveys preoccupations with recent socio-political changes and encounters between these belief systems, which affect the dreamer's household and marriage during his Tuareg wife's absence. The old woman's actions, the dreamer's reactions, and his community's wider religious and political dynamics suggest that this spirit has several overlapping identities, which I argue are not necessarily mutually exclusive and which convey concerns the dreamer is reluctant to express openly (like other dreamers studied by anthropologists; see, e.g., Borneman 2011:235). I explore why this is the case for the Tuareg man, as well as what his circumstances might imply more broadly for anthropological understandings of dreams as altered states, at the interface of idiosyncratic private conflicts and contested public meanings (Rivers 1922, Stewart 2012).

This case study is not intended to represent all local residents' experiences. Rather, its value is to reveal psycho-social conflicts present in the region that powerfully affect many residents there, though residents respond to these conflicts in different ways. These conflicts stem from both large-scale and intimate, long-term and immediate encounters in the community, and this dream illustrates one person's response to prevalent local concerns.

Local cultural models--a term I borrow from Mageo's (2013:387) dream study--are recognized widely in the case analyzed here, expressed in locally prevalent but not agreed upon symbols. Cultural models expressed in this dream reflect the dreamer's personal predicament and mixed feelings in the context of conflictual models of gender, religion, and other factional rivalries in northern Mali. My data show how dreams' core metaphors do not necessarily always signify mastery of internalized cultural models by evoking a dreamer's past (Mageo 2013, Rivers 1922), but instead may signify attempts to balance contested cultural knowledge in confronting difficult choices and fear of the future. Data also suggest the value of additional third parties to interpretation when voice and authority are disputed (Borneman 2011), though here I do not consider this dreamer's own voice to represent a unitary cultural model, notwithstanding his self-identification as Tuareg. Nor do I intend to represent the Tuareg and the Kunta in rigidly opposed terms, though I do offer evidence of substantial differences and tensions between them.

The Case Study: Moukha's Dream

This dream was defined by the dreamer as visitational (i.e., as a visit by a spirit to his room), though many Tuareg believe a dreamer's own soul may travel as well. This dream contains elements of both. The complex spirit presence active in this dream was an unnamed old woman. She was only identified as some type of Kel Essuf, a non-Qur'anic spirit. Kel Essuf are spirits of the wild, solitude, and nostalgia, who possess individuals feeling isolated or alienated from their (human) moral and social communities (Hawad 1979; Rasmussen 1995, 2001, 2008). I show that this spiritual complexity, in effect, mirrored the growing spiritual complexity and political tensions of Tuareg-Kunta relationships, which have intensified through sporadic wars in their region.

The male dreamer, whom I shall call Moukha (pseudonym), was a primary school teacher in his small, semi-nomadic, socially stratified, and ethnically diverse village, where there were sharp social, economic, and religious differences. Approximately 45 years old, Moukha was married to a woman whom he loved very much. At the time of his dream, his wife had been absent for approximately six months, which was longer than usual for most Tuareg women. She was participating in a school personnel training program in the large town of Kidal, approximately 100 miles away over unpaved desert roads--a rare opportunity for most rural women, and particularly important to this formally educated couple.

Tuareg married couples often live apart for varying time periods due to men's and women's long-standing nomadic transhumance, men's caravans and other itinerant trade, labor migration, and refugee flight. While separations were usually briefer and more predictable in older forms of seasonal travel (Bernus 1981, Camel 1996, Nicolaisen and Nicolaisen 1997), more modern migrations of Tuareg--first caused by disruptions of the caravan trade by French colonial policies in the early 20th century and later by refugee flight and exile from recurrent droughts and wars--have made separations increasingly unpredictable and irregular in duration. Many local residents in northern Mali intermittently flee adversity, though Kunta women, unlike Tuareg women, are also not permitted to travel or visit freely during peaceful times.

Peace accords attempting to end several Tuareg rebellions against the central governments of Mali and Niger in the 1990s and the early 2000s promised semi-autonomy, political rights, and economic opportunities for the northern regions of these countries, but national governments' budgets limited these policies' implementation. (1) In 2012, Tuareg nationalist/ separatist leaders declared the northern Kidal region an independent state of Azawad, but Islamist reformist piety militants seized control over parts of the North. Even after intervention by French troops and the Bamako-based army, the region still simmers with political violence.

Even before 2012, during my two field research trips to northern Mali in the first decade of the 21 st century, many (though not all) Tuareg residents already feared and resented both the encroaching Islamist movements seeking to restrict women and ban public musical performances--important to Tuareg culturally--and the Malian army, which remained at bases in the North well after peace accords in the mid-1990s had promised its withdrawal. The peace accords also promised to include more Tuareg in the national army, universities, and jobs. By around 2006, many residents felt impatient about these largely unfulfilled promises and ambivalent about the benefits of neoliberal NGO restructuring policies emphasizing individual sacrifice, withdrawal of government aid, and privatization. Many, like Moukha's wife, began to search for alternative opportunities in more distant places.

Moukha was supportive of his wife's new training project. Indeed, he encouraged local parents to send their children, including daughters, to school. Much effort was directed toward convincing residents--both Kunta and Tuareg, who opposed schooling for different reasons (which I outline below)--that secular schooling would not challenge Islamic devotion or destroy their cultural values. But Moukha encountered some hostility, particularly from more conservative Islamic scholars/marabouts in his village, and felt politically isolated. He also felt very lonely personally.

I first became acquainted with Moukha when I collected data on village household composition, and continued visiting residents informally. During this period, I noticed marked differences between Tuareg and Kunta households: married women among the Kunta were more secluded inside their compounds, which their husbands owned; whereas married women among the Tuareg were free to receive visits from, and themselves visit, unrelated men and women, and could leave their compound for herding or other work at will. These marriages have important legal consequences. Tuareg married women always own their nuptial tents, though husbands tend to own the house beside it in each compound. Tuareg husbands, even following bridewealth payments--which, in this region, their mothers-in-law and wives control--must continue to provide periodic goods on demand to their parents-in-law, and must respect their mothers-in-law, in particular, who can break up a marriage if they do not like their sons-in-law. Kunta husbands, once bridewealth is paid, owe no obligations to their mothers-in-law.

One afternoon, as Moukha and I chatted over tea at the residence of a mutual acquaintance, we mentioned missing absent friends. Others present lamented the widespread dispersion of residents due to recent regional turmoil, and hoped for their return. Suddenly, Moukha became very upset. He broke out into a sweat and, with great agitation, veins bulging and eyes opening wide as though in a trance, warned in a shaking voice: "You know, it is extremely dangerous to sleep in a room or compound alone because the Kel Essuf spirits enter to fill empty spaces! This can cause nightmares!" To illustrate, he related a dream he recently had:


   Once, while my wife was away in Kidal at her school training
   session and I was alone, I dreamed that an old woman [in Tamajaq,
   tamghart] resembling one whom I had seen in my village came into my
   room. She had an all-white head, opened her mouth, and then lay
   down on the bed, beckoning me. I refused to sleep with her in the
   bed, remaining on my mat, but I could not sleep. So I got up and
   walked outside. Still, I remained frightened because this [act] is
   dangerous, too; for Kel Essuf haunt abandoned places at night that
   were full during the day, for example, the market, ashes from
   fireplaces, or blood in butchers' and sacrificial sites.

The old woman who visited Moukha in his dream, whom he referred to as simply an old woman resembling one he had seen in his village, in some respects resembles a local Tuareg, non-Qur'anic Kel Essuf spirit called Doguwa, who is never mentioned by her name, but almost always referred to euphemistically as either tamghart or as the "Old Woman of the Earth." She threatens a mother giving birth, and can also emerge during other crises (Rasmussen 2004). For example, once during a violent thunder and lightning storm, my hostess in that village approached me, announcing very solemnly, "The Old Woman (tamghart) has arrived; are you afraid?" Then she suggested we take food and alms over to an impoverished bella (former slave) woman to obtain our al baraka protection from the storm--a storm presumably sent by Doguwa. Many fear that evoking Doguwa directly will activate her dangerous powers (Worley 1992), reflecting a cultural model prevalent among many Tuareg valuing cautious expression, indirect speech (tangalt), and reserve (takarakit) (Casajus 2000).

As Doguwa the Old Woman of the Earth, the elderly female spirit from Moukha's dream resembles the mother-in-law, with whom the new bridegroom practices strict reserve and respect, and to whom he owes bridewealth and additional work and goods--difficult obligations, even in peaceful times in the uncertain economy. Ideally, it is the mother-inlaw (the bride's mother) who decides how worthy a husband is after a trial two-to-three year period following the wedding, and who gives her daughter permission to move her kitchen and disengage her herds from her mother's.

In my interpretation, the old woman spirit figure operates at one level as a negative reinforcement, or a warning to this man to respect the marital symbolic space of the tent, his wife's dowry and property, and also to respect her tent's symbolic extension; ehan denotes both "tent" and "womb." On another level, this spirit expresses this man's ambivalence toward older women more generally as respected authority figures, who in Tuareg culture are not juxtaposed onto sexual contexts without some psychic stress (Rasmussen 1997).

The elderly female spirit in Moukha's dream also resembles Aisha Qandisha, a spirit in Moroccan popular or pre-Islamic cosmology (Crapanzano 1973,1981) who can appear as either an old or young woman. Even as an elderly female, however, she is very libidinous, and attempts to entrap and enslave each man she visits, requiring obligations from him as from a husband. As Aisha Qandisha, the spirit undoubtedly represented the personal anxiety and dilemma Moukha felt between his loneliness and his love and respect for his wife (as manifested in his mentioning the solitary, "empty" space of her property, the tent), and the temptation that he perhaps felt (though did not mention) to approach other women during his wife's prolonged absence--recall that "spirits fill empty spaces." But Moukha and his confidants hesitated to explicitly name her, perhaps because they did not wish to openly alienate anyone in that already tense community.

Both Doguwa the Old Woman of the Earth and Aisha Qandisha, despite their predominantly aged imagery, are in this dream therefore associated with sexuality and, by extension, fertility. The longterm separation between Moukha and his wife interrupted the couple's fertility, and threatened to disrupt the couple's household stability, in both Moukha's own viewpoint and that of many of his neighbors.

Both these spirits, moreover, are disdained by most Islamist (Da'wa/ Salafist) piety movements, to which some Kunta marabouts are sympathetic, as "not truly Muslim." Even Tuareg marabouts tend to regard them ambivalently (Rasmussen 1995). The ambiguity in Moukha's own interpretation of the spirit's precise name thus reflected wider social and religious debates and Tuareg cultural values of caution, reticence, and indirect allusion in dangerous predicaments, as well as reserve about relationships between the sexes out of respect for women. Without naming her specifically, Moukha in his own interpretation conferred upon this "old woman" both ordinary human identity (as a woman he said he had seen) and superhuman identity (as a possible Kel Essuf spirit).

Dream Interpretation in Anthropology and in the Tuareg Ethnographic Setting

Anthropologists have pointed out that although dreams are considered ontologically real in many cultural settings, dreaming ultimately remains an inward, subjective experience that is not easily verified and, thus, the dreams of others cannot be directly observed (Barrett and McNamara 2007, Lohmann 2007, Tedlock 1987). I recognize this challenge, but do not lament it; that is, I consider this divergence potentially valuable for fine-grained insights. In the present article, I celebrate this "gap" rather than attempt to narrow it by including my own interpretations alongside the dreamer's interpretations of his dream imagery (though I do not consider my own interpretations to be more "objective" in the old etic/emic dichotomous sense). By implication, I contend that although Moukha's dream narrative may not be possible to decipher completely by either dreamer or anthropologist, this juxtaposition of interpretations may offer broader insights for anthropologists who explore the interface between personal and collective sentiments and experiences (Borneman 2011, Mageo 2013), as well as dissonant cultural models of religion and gender (McIntosh 2009, Mittermaier 2011).

First, in this section, I critically discuss anthropological theorizing and ethnographic findings on dreams cross culturally, in terms of their relevance to the Tuareg data. In particular, I am inspired by studies that highlight the importance of understanding local cultural distinctions among conscious and intentional dreaming in divination/mediumship healing, on the one hand, and more spontaneous, less consciously intentional dreaming by non-specialists, on the other. Drawing on these insights, I then discuss widely held beliefs concerning the meanings of dreams and dreaming in Tuareg communities. Finally, I analyze these concepts in relation to political, historical, and dream-specific interpretive contexts.

In some societies, dreams are rarely recounted to others (Jedrej and Shaw 1992, von Grunebaum and Caillois 1966); in other societies, for example in Syria, there are local distinctions among ordinary dreams during sleep, unpleasant nightmares, and daydreams that affect (and reflect) the role of ethnographer as interlocutor in mutual constructions of ethnographic knowledge during interactions with informants (Borneman 2011:234-249). Most Tuareg willingly, even voluntarily, describe their dreams to others (including this outside ethnographer) and do not often distinguish between dreams and daydreams; both "genres" may convey wishes and fears. To many in the Sahara, dreaming while sleeping approaches other altered states such as spirit possession and mediumship and also can include soul travel, which refers to both literal geographic travel while awake and travel during sleep, when one's character is believed to change and great creativity is possible. These altered states and the spirits they dramatize in interior dialogues are much debated in their significance and legitimacy, particularly by Islamic scholars (both Tuareg and Kunta), who attempt to retain moral control over dream interpretation (Rasmussen 2001,2006). To some, particularly more devout Muslims and Islamic scholars, dreams are controversial. Even more so during intermittent waves of reformist piety movements' efforts to establish what they view as their own "stricter" interpretations of the Qur'an, as promoted by many powerful leaders and respected marabouts in Moukha's village. For example, many Islamic scholars, both Tuareg and Kunta, regard with skepticism a legal testimony based on "just anyone's dreams"--once legitimate in the past. Non-Qur'anic techniques of dream divining must adhere to higher standards of proof and moral legitimacy (Rasmussen 2001, 2006). In general, many Tuareg take dream interpretation more seriously when it is based on the divination of marabouts who sleep with Qur'anic verses beneath their pillows, rather than that of non-Qur'anic specialists, who communicate with non-Qur'anic spirits through offerings. Many prefer to consult marabouts for divination before making important decisions, such as negotiating a marriage, selecting a terrain to camp in, and sinking a garden well.

Many Tuareg make a major distinction between, on the one hand, the conscious and intentional dreaming of a diviner/medium (Qur'anic or non-Qur'anic) in medico-ritual practice for counseling and, on the other, the spontaneous, less consciously intentional dreaming of a non-specialist, such as Moukha, whose dreams are interpreted in private consultation with the former specialist. In rural communities, Tuareg cultural interpretations of Islam interweave with Qur'anic law, medicine, and rituals, though there are also debates and conflicts with some more devout members of the community who advocate pan-lslamic identity. Long-standing and widespread matrilineal influences in many Tuareg groups, which protect women's property (2) and address other legal matters, have encountered Arabic, Qur'anic, and nation-state patrilineal legal influences, particularly in semi-sedentarized villages and sedentarized urban centers (Bouman 2003; Keenan 2004; Kohl 2009; Oxby 1990; Nicolaisen and Nicolaisen 1997; Rasmussen 2010a, 2010b; Worley 1988, 1992). Furthermore, ecological disasters (droughts), regional political violence (invasions and armed conflicts with colonial and post-colonial state governments), and national unemployment have altered relations between the different social strata within Tuareg society. (3) Dreams, attitudes toward them, and their interpretation and moral legitimacy (or lack thereof) are all affected by these upheavals.

When I discussed non-Qur'anic mediums' dreams and divining with a rural-based Tuareg research assistant, a member of a respected maraboutique and icherifan family (tracing descent from the Prophet, with special healing powers), and his nephew, a student at a town CEG (middle school), both agreed that, "the type of person and the moral character of whoever dreams are important for credibility and control over future events." They associated the moral legitimacy of dream interpretation (i.e., the power to control whether a dream has a good or evil outcome) with Qur'an-based divination. They also expressed doubt about the moral legitimacy of just any dream or dreamer. "For example," they said, "if a good Muslim dreams about someone's death, this will postpone the death for a longer time" (Rasmussen 2001:90). A person who dreams and divines by non-Qur'anic means with tutelary spirits may be credible, provided that he or she otherwise practices Islam through prayer and generosity.

The same research assistant also felt that many dreams "are like tangalt, they are metaphorical [using shadowy, indirect speech], rather than literal in their significance." To illustrate, he went on to relate that he had recently dreamed that someone was killed by a hyena on a nearby mountain, but he did not take this dream very seriously as a literal prophecy; rather, he explained that it meant that "the man was probably bitten by an insect." Or alternatively, he surmised, "this dream could refer to another, similar event, for example, I killed a snake in the dried riverbed early that morning."

These examples illustrate the importance, in Tuareg culture, of the symbolic analysis of dreams, albeit with some caution in attributing final authoritative meanings. Moral authority in dream interpretation is not automatically conferred on just any dreamer or any specialist diviner. Also important here is a suggestion, in Moukha's dream, of subtle local, not universal, sexual symbolism, in keeping with shadowy speech and reserve about sexuality (the old woman in his dream is somewhat incongruously associated with the bed and sexual relations; a sexually demanding and morally ambiguous female spirit is also said to arise if a man enters into a pact with her). On another level, property symbolism populates the dream (the bed, like the tent, is the property of the married woman, reminding Moukha that he was, in effect, a guest in his wife's home). All of these symbols convey the moral limits set by specialist diviners on creative subjectivity in dream interpretation.

Similarly, herbal medicine women, who occasionally practice dream-based divination, also emphasize the subjective symbolic creativity in dreaming as well as its moral limits. Many, like the Moroccan religious specialists Crapanzano (1973, 1981) studied, recognize false dreams, which they say are caused by the Devil, Shitan, or Iblis. There are no clear criteria to distinguish false dreams, but the Tuareg medicine women I studied (Rasmussen 2006) made intense efforts to discern such criteria in their moral commentary on dreamers. One medicine woman asserted, "a good person is less likely to have false dreams." Also, a dreamer who divines should ideally counsel a client only if the latter is "good." These herbal medicine women diviners, who gather from trees that are abodes of matrilineal spirits, practice ritual restrictions so as not to alienate Islamic scholars. This suggests that, since healing with tutelary matrilineal spirit pacts can become a significant competing force for Islamic scholars, the former is carefully monitored.

This undercurrent of competition explains the greater ambivalence of Islamic scholars/marabouts and some more devout Muslims toward non-Qur'anic dreaming and divination. It was in the interest of Islamic scholars/ marabouts in Moukha's community to discourage or contain potentially competing authoritative knowledge from a Tuareg teacher in a "modern" school with secular power, especially one with added spiritual power from non-Qur'anic sources; by combining the two, Moukha would have effectively become doubly empowered. It was no surprise, then, when cautionary rumors flew about another local man who had "married" a Kel Essuf spirit and later became widowed several times, but whom human women henceforth avoided marrying, fearing his "marriage" to that jealous spirit had caused his wives' deaths. That unfortunate man had to pay a marabout exorbitant fees, including eventually his adobe house. Destitute, he moved in with his brother.

Thus men, as well as women, may become possessed by different types of spirits. But most men's spirits are diagnosed as being Qur'anic djinnoun, who seize men inside mosques. Although a few men also become possessed by non-Qur'anic Kel Essuf spirits outside of mosques, and may undergo possession ceremonies to address them, this is ridiculed as "effeminate" by some other Tuareg men (Rasmussen 1995), thereby rendering men who follow pacts with non-Qur'anic spirits as non-conforming with regard to gender. Ideally, it is women's spirits who are supposed to be Tuareg non-Qur'anic--some nature-related and others matrilineal, all usually glossed as Kel Essuf. Regardless of the gender of the human host or of the spirit, however, calls of these Kel Essuf and other superhuman beings are powerful, and transport the person away from the social and moral community, plunging individuals into psycho-social limbo (Hawad 1979, Rasmussen 2008). The spirit can take over vulnerable persons who are already in solitude or isolation, as Moukha was during his wife's absence, and can push them over an abyss into recurrent spirit possession requiring exorcism, or even causing permanent insanity.

For diagnoses and advice related to dreaming (in Tamajaq, targat) and psycho-social non-organic illnesses, there are several different types of diviners/mediums: predominantly male Islamic scholars, popularly called "marabouts" throughout Mali and Niger regardless of ethnic origin; specialists of either sex known as "friends of the Kel Essuf," who have pacts with non-Qur'anic spirits; and some female herbal medicine specialists gifted with additional dream-divination powers (Rasmussen 2001,2006). The Islamic scholar healers/diviners specialize in written Qur'anic verse healing. Although many women and men see all types of specialists, more men tend to see Islamic scholars. In Moukha's community, Kunta marabouts dominated over Tuareg marabouts because of the economic and political power they acquired when colonial French forces in the early 20th century armed the Kunta in order to subdue Tuareg resistance against the French, and elevated the powers of the Kunta traditional leaders there. Most Tuareg are cool toward Islamist reformist movements but, over the past few decades, a few Tuareg leaders have been pressured to become more devout and follow the reformist piety movements, at least superficially. Most nationalist projects of the Tuareg rebellions from the 1990s through approximately 2006, still sporadically simmering, have tended to be clan-based and secular, but some later factions aligned with the wider Ansar (E)dine and other Islamist movements (Lecocq 2010:53).

The point is that this dreamer may have avoided direct, explicit reference to Doguwa or Aisha Qandisha out of political caution as well as personal reserve or embarrassment. Many afflictions, including "illnesses of the heart and soul" or love-related feelings, also prominent in possession ritual cases (Rasmussen 1995), are alluded to indirectly for this reason. The Islamic scholar/marabout whom Moukha consulted discouraged him from naming the old woman (in her human and/ or spirit identities) more precisely. Like many marabouts, this specialist was discreet, and emphasized confidentiality to gain clients' trust. The marabout interpreted Moukha's dream as literal physical endangerment from being apart from his traveling wife. Fie confirmed that, "Kel Essuf [the gloss for non-Qur'anic spirits] may attack those in solitude or those who feel lonely." Fie did not specify which non-Qur'anic spirit this old woman was, mentioning neither Doguwa (the Old Woman of the Earth) nor the North African Aisha Qandisha, thereby discouraging Moukha from embarking on a spirit pact which would have involved marriage to the spirit--much disapproved of by more devout Muslims there, including the marabouts. Fie did not mention Qur'anic djinnoun spirits, either. Thus, the marabout's interpretation, like Moukha's, pointed to loneliness and solitude, but importantly, the marabout encouraged a strengthening of the human marriage bond. Many Tuareg tacitly ignore extra-marital affairs, but Islamic scholars (whether Kunta or Tuareg) condemn them, and more militant Islamist reformists in 2012 introduced flogging as punishment for these transgressions in parts of northern Mali. That said, the marabout did not council Moukha to contract a polygynous marriage.

There is therefore strong evidence that Moukha did not wish to name the woman in his dream because he felt that his more religiously-conservative Kunta neighbors would disapprove of identification with a Tuareg non-Qur'anic spirit or a local variant of the North African one, and/or he feared neighbors would suspect he felt attracted to a human woman in the village. As a school teacher, this dreamer faced predicaments that were potentially or actually awkward, shameful, and disruptive to his prestige--even dangerous--given the conflicts in his village. For example, some men ridiculed a husband who allowed his wife to travel for extended time periods.

I contend, therefore, that Moukha's dream motif and his reticence surrounding it convey mixed sentiments and an ambivalent deferral of meaning, more than a straightforward dramatization of a unitary cultural model in the face of cultural dissonance. The kaleidoscopic meanings of this dream, conveyed by local spirit symbolism rather than universal "Freudian" sexual symbolism, derive from the dreamer's stress concerning both his immediate household and the wider community dynamics.

Spirit motifs recur in many Tuareg expressive forms (for example, possession and mediumship trances, poems, songs, dreams, and divination), and often convey psychological reactions to distance, separation, and loss--prominent experiences in nomadic (and now, also, migrant labor and refugee) life which are reflected on through altered states of consciousness and creativity. Many Tamajaq songs and poems mention essuf (solitude, wild, and nostalgia), and call for the return of migrants, refugees, and exiles to their home regions. Moukha felt nostalgic for his wife. Many Tuareg further assert that most creativity takes place in a state of solitude and/or nostalgia, in "the wild"--a state that is both a physical space and a psycho-social condition (Hawad 1979; Rasmussen 1995, 2001,2008). As Dragani (2012) points out, dreams also play a role in poetic inspiration among some Tuareg groups in Niger. Thus, dreams and other altered states need not always be solely afflictions; they can also be considered creative spaces. Moukha's dream transported him to the "wild," expressing his solitude and nostalgia, but also enabled him to express these sentiments indirectly, albeit not in a poem but in an altered state. Indeed, spatial and temporal aspects of dreams need not have a fixed reference point.

There is also a non-linear aspect to dreams. Emphasizing re-play of the past, Mageo observes that dreams "are a stage on which people replay the dramas of the day" (2013:387). But in these data from northern Mali, more occurs: dreams also enact dramas of current, and even future-oriented ambivalence, fear, and struggle, recalling W. H. R. Rivers's (1922:12) critique of a Freudian emphasis on dreams as arising from infantile wishes and also supporting Stewart's (2012) finding that altered states can contain powerful imagery of the past, but not in a strictly linear sense. Rather, past imagery may "barge into consciousness and create affective tensions and identifications between the past and the present" (Stewart 2012:7). Therein reside dreams' ambiguous meanings. I agree with Mageo (2013) that dreams dramatize not solely personal, but also cultural models, but that the latter are not consensual or unitary, nor are they always fully realized dramas of the day. The dreamer in the present case study is not a microcosm of his entire community experience, but rather is someone who grapples with several competing cultural models in his community. Widely-held Tuareg religious and gender-related models derived from earlier institutions, preceding conversion to Islam influenced by Sufism, allow considerable freedom in sociability between women and men. Arabic-influenced reformist piety (Salafist, Da'wa) movements' interpretations of Islam seek to restrict conduct (and contact) between the sexes, the latter being more influential among the Kunta. Moukha experienced these conflicting cultural models both at the intimate level of his household dynamics and in the wider political "winds" sweeping his region. The sexual symbolism (in the old woman beckoning Moukha to approach her on the bed), moreover, indirectly expresses this local conflict.

In Moukha's community (4)--in the region between Kidal and Gao--long-standing tensions between several factions, already escalating between 2006 and 2007, culminated in 2012 in full-blown battles between several factions, including: the predominantly secular Tuareg separatist rebels (called the MNLA, or the Movement National pour la Liberation de Azawad); the Bamako-based Malian army (long resented by many residents for atrocities some soldiers committed in the northern region earlier in the 20th century); and the da'wa (Salafist)-inspired Islamist-reformist piety forces (comprised of predominantly though not exclusively Arabic-speaking fighters), who extended their control over parts of northern Mali and imposed their strict interpretation of Qur'anic law there until they were driven out by French and Malian troops in 2013. Most Tamajaq-speaking residents opposed the Islamic militants. As of this writing, peace negotiations and sporadic outbreaks of fighting continue, and there are also AQIM (Al Qaida of the Maghreb) forces in remote border areas. Local residents have been well aware of these gathering dangers, collapsing past and present experiences and future fears into their preoccupations and conversations.

Thus, Moukha's dream represented not only past struggles, but also kaleidoscopic time and space travel in its past and present personal (marital) and political (religious) preoccupations and conflicts, conveying his fears of the future.

Historical and Ethnographic Background to Tuareg-Kunta Relationships: A Closer Look

A historical and ethnographic overview of relationships between Tuareg and Kunta residents in northern Mali is instructive here. The Kunta (sometimes Kounta; singular, Elkentawi, also called Alkanata)--a historically Arabo-Berber, but now Arabic-speaking people originally from northern Mauritania--now self-identify as Arabs in northern Mali (Boilley 1999). They are a large religious clan whose relations are the product of fissions and fusions in response to pressures such as invasions and droughts. From the 15th century onward, Qur'anic scholarship became for the Kunta "a means to wealth and therefore power as controllers of the trans-Saharan trade from Morocco to Timbuktu" (Brett and Fentress 1997:152); such scholarship enabled them to keep the peace for caravans and to control Qur'anic education. Kunta have, for centuries, nomadized around Timbuktu and to the north around the Adragh-n-Ifoghas region near Kidal (Boilley 1999). From Timbuktu, the Kunta exerted an enormous influence on the development of Islam in West Africa. Kunta and Tuareg have been enmeshed for at least a millennium (Lecocq in Kohl and Fischer 2010:53).

Yet, there are also cultural differences and political tensions between Tuareg and Kunta based on economic inequalities beginning in the early 20th century, when the French armed the Kunta in order to subjugate local Tuareg groups who resisted colonial domination. Kunta descent groups of prestigious marabouts dominating some parts of northern Mali economically, religiously, and politically offer higher bridewealth than many of the generally poorer Tuareg men can afford; the latter often resent Kunta men for "taking" all the most beautiful local women and supporting them grandly, with minimal need for these wives to perform arduous domestic labor (Rasmussen 2010b). Tuareg women who marry Kunta Arabic-speaking men do not leave their husbands' compounds, veil much more closely than other Tuareg women (who wear a more abbreviated form of covering resembling a long scarf/shawl), and follow Qur'anic inheritance (in which male heirs inherit twice the amount of property females do) with less access to the Tuareg matrilineal property endowments. Some Tuareg women regard these marriages as prestigious and advantageous, however, since these arrangements free them from arduous physical labor.

In Moukha's rural village, residents of two neighborhoods are predominantly Tuareg (i.e., persons who self-identify as Tuareg or Kel Tamajaq linguistically, culturally, and socially), and though of diverse social origins --including both the aristocratic and (formerly) servile--they view themselves as a community distinct from the Kunta Arabs. In a third neighborhood of that village are the households of the Kunta Arab shopkeepers and marabouts, the latter specializing in Qur'anic law and psycho-social healing, speaking predominantly Arabic, and following more patrilineal Arabic and Qur'anic inheritance and descent patterns.

Regional wars, invasions, and Tuareg-Kunta intermarriage and trading are therefore also relevant to the dream motif. So, too, are competitions over resources (like women and water) and Tuareg-Kunta cooperation. Since residents have suffered from droughts and water shortages, there is pervasive concern with the fertility of humans, livestock, and land. When rain falls, many local residents cup their hands together, bring the water droplets to their foreheads, and murmur, "Al baraka n aman" the blessing of water, to express their gratitude. Since Moukha's Tuareg wife's lengthy absence is permissible among most Tuareg, this was not a contested issue between husband and wife but, nonetheless, the couple's fertility was threatened, and many Kunta disapproved of Moukha allowing his wife to travel and embark on a career in education. This Tuareg-Kunta social conflict is mirrored in his dream's spirit motif, which also corresponds to images in the following oral history of his village's origin.

In a mythico-historic account I collected relating the founding of Moukha's village, a Tuareg Islamic scholar/marabout and an Arab competed over water and, by extension, over fertility of land and livestock. The village founder, a Tuareg Islamic scholar in the Chamanamass descent group, divined for water by dreaming of a water source or, in another variant, by communicating with a crow. That marabout soon came into conflict with the Arab, who threw sand at him when the two fought over this precious water source. The origin tale ultimately promotes cooperative relations between these groups as necessary for their survival. The structural conflict in this oral history was mediated by the Tuareg marabout's ability to bring rain and find wells, thanks to his al baraka blessing/benediction power. Tuareg and Kunta relations are not only characterized by economic inequality and political tensions, but also by mutual dependence and close cooperation in trading, intermarriages, and Qur'anic consultations between Tuareg and their neighboring Kunta marabouts. Yet, these relationships are also ambivalent since the Kunta marabouts condemn extramarital affairs (with humans and/or spirits) and thus discouraged Moukha from pursuing a pact (long-term mediumistic relationship) with a non-Qur'anic spirit.

While they collaborate in many areas, therefore, the Kunta Arab and the Tuareg men compete over women to marry and, by extension, for the fertility of humans, crops, and livestock. They also compete over the "dissemination," as it were, of Islamic knowledge: in mixed marriages, it is the Kunta father who decides which religious route children will take. Thus, the dream and myth in this case study are mutually illuminating. Many thinkers have conceptualized the relation between myth and dream as one of homology. W. H. R. Rivers (1922:24, 26), following Victorian evolutionism, held that dream formations would shed light on the "rites and customs of savage men," including myth. More recently, anthropologists have argued that similarities between myths and dreams arise because both partake in a visual and emotional form of thought (Kracke 1987) or are structured by a common dialectic problem-solving logic as identified by Levi-Strauss (1963; see Kuper 1979). Others (Tedlock 1987:464, Stewart 2012:105) move beyond conceiving of myth and dream as homologues by considering the historical articulations and interactions between them.

Imagery in the oral history of this village's origin, I argue, is at least indirectly relevant to imagery in Moukha's dream, though I do not argue here that they are exact homologues. The important point is that Tuareg and Kunta have long competed over interpretations of Islam, water, and women's fertility. Although the symbols in the myth and those in the dream are not identical, they nonetheless express similar preoccupations. The sand thrown in the myth conveys extreme disrespect in Tuareg etiquette; moreover, sand represents dryness, whereas the water represents al baraka benediction/ blessing and life--imagery opposing the Arabic and Tamajaq-speaking disputants, as well as the desert, to fertility. (Within Tuareg concepts of reproduction, dryness is associated with infertility and post-menopausal status, while wetness is associated with youth, fertility, and childbearing [see Rasmussen 2006, 2010b].) The elderly female spirit in the dream, on one level, enacts Moukha's anxiety over his (young, fertile) wife's absence and her future transformation into an older, post-childbearing woman over time. On another level, as already noted, she symbolizes a human or a spirit woman with whom it was possible to have an affair or other connections (like a spirit pact), while perhaps also reminding him of his mother-in-law--placing a moral limit on his temptation. Thus, this dream motif encapsulates the social and religious complexity in Moukha's village.

In some respects, there was harmony in village diversity. Not all Tuareg were entirely excluded from prestigious or remunerative positions in that community. The village mayor, a Kunta Arab and son of the traditional Kunta chief of the local faction, was generally respected and well-liked. The mayor's secretary was of Tuareg and Songhay background. In other respects, however, by the first decade of the 21st century, there were resurging tensions. Local Tamajaq oral histories, poems, and songs lauded Tuareg male and female warrior heroes who had fought against the Arabs and the French. Women recited poems to me about female battle heroes who misled Arab fighters--an indirect way of expressing their dislike of policies promoted by many Arabic-speaking and culturally Arabic-identifying Kunta, such as polygyny and patrilineal inheritance. There was also sharp conflict over girls' education, which is relevant to interpreting Moukha's dream since the debate over girls' schooling influenced his domestic predicament: because of some local opposition to secular education, Moukha's wife was obliged to leave their rural village and attend training sessions in the town of Kidal, where there was greater support for such a program.

Historically, secular education has been opposed by some persons of both Tuareg and Kunta background, but for very different reasons. Many Tuareg feared schools as potential ethnocide and teachers as potential harassers of female students. But rural Tuareg do not discourage women from travel and, increasingly, many Tuareg are beginning to value secular education as a route to jobs (Rasmussen 1997). Nevertheless, some still fear that women far from home are vulnerable to harassment from outside men. Kunta marabouts and their traditional leaders also opposed secular education, particularly for girls, but for different reasons from the Tuareg: Kunta feared that educated women would demand more participation in "official" religion and governance, and more covertly, feared competition from secular schools with their Qur'anic education. Rumors flew that, once educated, "our women may even seek to become imams [leaders in the call to prayer]!" Others warned that girls in school would drop their traditionally modest dress. Through speeches at village meetings, children's songs, and written pamphlets, school officials and NGOs reassured residents that these problems would not occur; for example, posters throughout Mali and Niger exhorted, "Send your daughters to school; I sent my daughter, and she is now paying for my trip to Mecca!"

As a teacher, Moukha stood at the epicenter of this polarizing issue. His spirit dream subject, the old woman, did not directly represent a young female student or teacher, it is true, but this debate nonetheless caused him stress--a factor contributing to the construction of the spirit's kaleidoscopic identity as elderly, but also sexual (perhaps he repressed an image of a young woman). Moukha also worried about his wife's actions in the town, widely considered a place of temptation and immorality by more conservative residents who opposed women's travel without accompanying male relatives. Perhaps, Moukha also worried that his wife would find a lover while away, though he did not mention this openly out of dignity (imojagh) and reserve (takarakit) surrounding love sentiments (Rasmussen 1995). Influential leaders could gossip about a traveling woman, ruin her reputation, and embarrass her husband.

Moreover, Moukha, like many other Tuareg, believed that since one's soul (iman) travels during sleep, one's character (tesney) also changes, as during travel. By implication, then, Moukha's soul and character during his dream "travel" were vulnerable to transformations, as was his wife's character during her literal travel. Recall that as Doguwa, the Old Woman of the Earth Kel Essuf spirit, the dream figure on one level reminded him of his wife's continuing potential childbearing and challenges to it over time. On another, equally powerful and juxtaposed with this reminder, was the image of the empty home, signifying an empty womb or infertility--a symbol, like the sand opposed to well and water in the origin myth of the village, of lack of human life. The term ehan, denoting tent and marriage in Tamajaq, is also used to denote womb (e.g., ehan n barar denotes "the child's tent," slang for womb); one "makes a tent" when one marries (when the nuptial tent belonging to the wife is built by the wife's elderly female relatives, a section of it is taken from her mother's own tent). The wife's extended travel did interrupt fertility, and she and her mother (the latter considered "elderly"; see Rasmussen 1997) did, in effect, haunt the tent, explaining why the husband could never completely be at home in it. Doguwa's "Janus-faced" identity is also present in this fertility motif: though she is old, Doguwa is, as noted, present at childbirth--she is the evil twin of the good midwife, as she emerges from the earth and threatens the birthing mother and her newborn child. Other Kel Essuf spirits try to pull the child back into their world until the official name day held a week following birth.

On another level, the "old woman" stands as a symbolic transformation of a younger woman the dreamer had seen in his village whom he may have felt attracted to, but was unwilling to identify by name, since this would cause embarrassment. Perhaps, the old woman was a disguise, repressing his feelings. In effect, during his consultation with the marabout, this potential illicit (extramarital) human lover--perhaps, a young student in a "daydream" repressed by Moukha, metamorphosized into an elderly mother or mother-in-law figure. In this latter form, the spirit expressed the dreamer's preoccupation with his marital and domestic obligations, and her effect was to remind him of his obligations to his marriage and household and what they meant symbolically, as well as materially: payment of bridewealth, generous gifts to both his wife and her parents (for example, providing trade goods and garden produce for their storehouse or, in cases of a husband earning a regular salary, monetary generosity), and marital fidelity, at least ideally, all convey respect for the wife.

Without his wife's presence at home, Mouokha was, in effect, in essuf (solitude; feeling nostalgic), as well as in the wild (a limbo or liminal state of moral indeterminacy and danger) from the perspective of family members. In this state, spirit pacts can push one over the edge, far from the shelter of the Tuareg tent and the center of the human moral community. Hence the connection between cultural, psychosocial, and physical reproduction. On many occasions, Tuareg indicated that men, as well as women, suffer greatly (and are sometimes blamed and mocked) for childlessness or having few children. Thus, on this level, Moukha's dream conveyed his fear that uncertain or interrupted fertility would result from the uprooting of his family.

Spirit Possession and Spirit Contracts

Spiritual motifs in waking and sleeping life were therefore imbued with gendered and religious moral commentary, as many local residents, including Moukha, were pulled asunder by battling moral forces. Women in Moukha's rural community also expressed fears of spirit visitations to me when their spouses were traveling extensively--for example, one woman requested that I allow her to sleep inside my guest compound during such an absence "in order to avoid the spirits that fill empty spaces." Spirits may also attack women traveling alone, as for example when a woman left for home after an evening musical event, allegedly felt a slap or blow, and then became possessed; in local parlance, she was struck by a spirit (perhaps, a euphemism, out of embarrassment, for a male attacker). She needed the marabouts' treatments for three months following the incident. These comments use the idiom of spirits to interpret increasing violence, unpredictable strangers, prolonged travel, the separation of families, and uncertainty at night (the latter a long-standing Tuareg apprehension, newly intensified with fears of bandits and soldiers).

Moukha's non-Qu'ranic spirit visit may be experienced by either sex, but, as noted, is experienced more often by women. Thus, there are strong hints this dreamer felt ambivalent about the spirit's precise identity and its possible "gender-bending"--also discouraged by many Islamic scholars, though traditionally tolerated in prevalent Tuareg cultural attitudes. This ambivalence, again, recalls Crapanzano's (1973) findings concerning Aisha Qandisha, the female spirit in Morocco--which is the original home of several marabouts who converted Tuareg to Islam. If a man sleeps with Aisha before discovering her identity, "he becomes her slave forever, and must follow her commands [to sacrifice and give offerings]" (Crapanzano 1973:144). In Morocco, she requires her followers, as husbands married to her, to wear red, black, chartreuse green, or some combination of these colors. The members of the Hamadsha, an Islamic religious order or possession "cult" of adepts, are her special devotees in trance. Yet male devotees are somewhat atypical, for there, too, mostly women have relationships with named spirits.

In Tuareg pacts with non-Qur'anic Kel Essuf spirits, the medium is called a "friend" of the Kel Essuf (Nicolaisen 1961), and is also considered married to such a spirit (Rasmussen 2001). Such a pact brings marriage-like household responsibilities: if the medium/diviner is single, he or she cannot marry a human without the spirit's authorization and, if already married, the medium/diviner cannot easily remain married to a human spouse. The tutelary spirit is jealous and possessive, since this pact brings difficult restrictions and obligations: regular offerings to the spirit, wearing the color blue, applying henna to the hands as brides and grooms do at their weddings, and being sexually faithful to the spirit spouse, or at least, in the words of one Tuareg diviner/medium, "one must love one woman only." Islamic scholars/marabouts and other more devout Muslims tend to disparage this spirit marriage and its non-Qur'anic divination.

Recall that the spirit in Moukha's dream was represented in the form of not just any woman, but a very elderly woman, who nonetheless approached him for sexual or marital relations. This is, in effect, a Tuareg cultural variation on the North African spirit Aisha Qandisha's age transformations, which, I contend, reinforces my kaleidoscopic interpretation even if the dreamer was not conscious of this: this age-based shape-shifting, manifest in both the Aisha Qandisha and Doguwa figures, is a Tuareg cultural "twist" on the traditional spirit imagery. The social dynamic reflected here is the interface of cultural encounters--and "love/hate" relationships--between Tuareg and Maghrebian interpretations of Islam, imported via holy men from North Africa. The dream's symbolism and religious imagery, more crucially, provide ambiguity and leeway for individuals to save face in the consequent religious and political conflicts. In fact, neither Moukha nor the marabout openly identified this spirit figure as a Qur'anic spirit, either. There were no openly practicing Hamadsha in Moukha's region, as Sufibased Tuareg Islam was disapproved of by more radical Islamists. Moukha did not conduct intensive Qur'anic study, was not an adept in public spirit possession/exorcism rites in which women predominate, and did not practice the mediumistic powers in a Kel Essuf spirit pact.

There were several reasons why Moukha refrained from deeper involvement with these spirits. Spirit possession in the Kidal region is even more controversial than in some other Tuareg regions. In the 1990s, these rituals were banned in Kidal because a possessed person in trance reportedly falsely identified a functionary as a thief. Also, by 2012, a non-Qur'anic medium/diviner who held a spirit pact earlier in another Tuareg community where I conducted research was no longer practicing his profession since, local residents explained, "fewer people consulted him." Some residents viewed the decline in (though not disappearance of) non-Qur'anic spirit possession and mediumship as due to "more marabouts present in our region," though a few women there insisted that they still held these rituals. In Moukha's case, a spirit pact conferring mediumship would be difficult because it would contradict his professional status as "modern," or "following what is new" (i.e., secular rather than Qur'anic education), and would also prompt more direct confrontations with powerful Kunta residents whom he was trying to win over to secular education.

There were additional difficulties. As noted, the demands of spirit pacts often interfere with household subsistence and (human) marriage. Maintaining a pact with the spirits demands not only material resource sacrifices, but also some gender "shape-shifting." Tuareg male non-Qur'anic diviners dye their hands with henna, which normally only women do as a regular cosmetic practice, and other Tuareg men do only at their weddings. Moukha and the Islamic scholars/marabouts he consulted felt ambivalent about all these demands since residents in that village were being pressured to practice Islam more devoutly in the hope of greater socioeconomic mobility, recalling some observations by Masquelier (2009) concerning Hausa Bori practitioners in central Niger, and by McIntosh (2009) concerning Giriama possession by Muslim spirits in coastal Kenya.5

Conclusions and Broader Theoretical Implications

How do my data inform theories of dreaming in anthropology, and how does the theoretical literature on dreaming inform these findings? Often, a dreamer is incapable of or reluctant about explicitly articulating the conflict expressed in dreams in terms of his real-life situation (Lohmann 2003, Mageo 2013). The marabout Moukha consulted encouraged him to quietly reflect on the predicament and his feelings, rather than pursue a spirit pact. In Tuareg communities, as elsewhere, dreams can symbolically articulate psycho-social, religious, and political conflicts simultaneously, in some respects confirming the findings of other valuable anthropological studies of dreams (Borneman 2011:234-349; Dragani 2012; Graham 2003; Lohmann 2003, 2007:35-71; Mageo 2013:287-411; Mittermaier 2011). Some dreams may articulate, but not resolve, a conflict or provide perfect closure (Carpanzano 1973, 1981). As Lohmann (2003) observes, dreaming knowledge everywhere is filtered through multiple cultural possibilities, left open and ambiguous in interpretations and reformulations by the dreamer, translated into a performance presented for a specific reason to a particular audience, and subsequently heard and evaluated by listeners, including the anthropologist. These different audiences express mixed sentiments, as well as caution and perhaps also dread of what the future holds, as reflected in this Tuareg man's dream in northern Mali.

Regardless of its specific identity, this spirit motif expresses cultural contradictions regarding the anomalous mixing of categories of lover, wife, and affine, in ambiguous, ambivalent sentiments concerning the husband's relationship to his wife, mother-in-law, and neighbors over time. Thus, this dream encapsulates long-term problems of gendered spiritual power experienced by both sexes, as well as the dreamer's immediate loneliness while his wife was absent. Certainly there was some gender reversal or inversion by this couple in the eyes of their more conservative neighbors: usually, men travel more widely than women, migrating to nearby towns and beyond, and Tuareg women, even while herding and visiting independently, tend to remain in or near rural camps and villages. By traveling to attend an educational training program, Moukha's wife was transposing the more long-standing role of Tuareg mothers as educators inside their tent into the context of modernity--the focus of both praise and ire in that region. Moukha stood at the center of this controversy and at the nexus of more long-term psychological, religious, and political tensions. Thus, his dream was conflict-focused, recalling some of Rivers's (1922) findings in his research on dreamers: in sleep, repressed attitudes reassert themselves, though not in the sense of universal symbols of early wish fulfillment. There is no single, obvious iconic meaning to dreams' elements. Moukha's spirit dream held a subjective, creative, contextual meaning.

Dreams arise in particular moments, sometimes of contestation and anxiety (Stewart 2012), and can reveal unresolved disputes in a community as well as in the psyche of the dreamer. As Stewart (2012:12) points out, dreams may also convey worry over not only past, but also present and future traumas, thereby constituting a valuable source of cultural knowledge derived from an altered state of consciousness. And, as Mittermaier (2011:239) argues, "dream visions" tend to embrace ambiguities and can resonate with multiple meanings, even in silence or reticence. These insights require ethnographers to make analytical distinctions among different interpretations (those of local residents and of outside researchers), but also to recognize the fluid, overlapping, and ambiguous identities in the dream symbols, which reflect social relationships in the community.

More generally, dream narratives resemble those of other altered states of consciousness, even those associated with public rituals, such as possession and mediumship. Despite their often public ritual manifestations, both are often characterized by subtle, rather than direct or overt, meanings. In this article I have attempted to closely examine the connections between these states. Yet, in contrast to public rituals, access to dreams is indirect, achievable only through dreamers' narratives (Crapanzano 1973; Graham 1995; Mittermaier 2011; Obeyesekere 1981, 1992; von Grunebaum and Caillois 1966).

I have proposed what I term "kaleidoscopic" identities and meanings in this spirit dream, to accommodate diverse viewpoints and voices including, but not limited to, those of the dreamer, and to show how dreams may often express mixed feelings and respond to cultural dissonance, rather than refer to a unitary or consensual cultural model or even a single conflict, and may also often resist neat linear temporal sequences. I hope this concept is helpful in anthropological interpretations of dreams and other altered states more generally, as well as to efforts to re-center the subject in anthropology (albeit not solely grounded in individual psychology, but also in ethnography; see, e.g., Borneman 2011; Jedrej and Shaw 1992; Lohmann 2003, 2007; Mageo 2013). The notion of kaleidoscopic subjective meanings of dreams may be useful in analyses of personal and cultural experiences of gender ambiguity and religious power in altered states of consciousness.

Acknowledgments:

I gratefully acknowledge support for field research projects over nearly 30 years, between 1983 and 2012, in rural and urban Tuareg communities of Niger and Mali, from Fulbright-Hays, the Wenner-Gren Foundation, the Social Science Research Council, the National Geographic Committee for Research and Exploration, Indiana University, and the University of Houston, on the following topics: spirit possession, gender, aging and the life course, artisans, healing specialists, verbal art performance, and youth cultures.

Endnotes:

(1) Upon Mali's and Niger's independence from France, new state governments inherited colonial policies that tended to favor settled farming peoples in their southern regions. Due to these policies, there have been several Tuareg rebellions: in northern Mali around 1963, over food relief distribution and taxes; in the northern regions of both Mali and Niger between 1990 and 1996, over some northern residents' economic and political marginalization and disillusionment with progress in integrating more Tuareg into national infrastructures; resurgences of armed conflicts in Mali briefly in 2006 (during one of my field residences) and sporadically since 2012; and in northern Niger between 2007 and 2009 over uranium mining contracts.

(2) Most Tuareg groups compensate women for male bias in Qur'anic inheritance, which leaves two-thirds of property to male heirs and one-third to female heirs, with endowments called akh ihuderan ("living milk"), passed to daughters, sisters, and nieces, which cannot be sold, but in recent years these endowments have been challenged by some Islamic scholars and state laws.

(3) Today, most Tuareg in northern Mali, Niger, and southern Algeria and Libya are semi-nomadic or semi-sedentary, due to pressures by colonial and post-colonial independent government policies and natural disasters such as droughts, which have devastated many herds. Many (though not all) tend to combine stockbreeding with oasis gardening, market and itinerant trade, artisanry, and labor migration. Occupations were formerly associated with inherited social categories based on descent, but most persons now practice diverse types of work. Pre-colonial Tuareg social categories were hierarchical, but also flexible and negotiable. More nomadic aristocratic descent groups (often called "nobles" in English, and called imajeghen or imoujagh in Tamajaq) owned and controlled most large livestock, weapons, and the caravan trade, and placed subordinates under their protection in exchange for tithe-like tributes, taxes, and rents. Tributary and client groups (imghad and ighawalen) raided and traded for their noble overlords, enjoyed usufruct rights over animals' offspring, and kept part of the booty from war, while others gave portions of their harvests to nobles. Artisans (inadan), formerly attached to noble families, still manufacture jewelry, tools, weapons, and, in some regions, also perform verbal art and act as go-betweens and ritual specialists. Islamic scholars (ineslemen or "marabouts"), in some regions Tuareg, in others, Arabo-Berber, interpret the Qur'an. Iklan, persons of varying degrees of servitude owned by some aristocratic elites until freed around the mid-20th century, performed most manual domestic, herding, gardening, and caravanning labor.

(4) In order to protect identities of individuals (whose names are omitted here or replaced with pseudonyms), I do not give the real name of Moukha's village, one of several in northern Mali where primary schools were established. Kidal is the real name of a much larger town and also of the region surrounding it.

(5) In the Hausa and Giriama cases, as described by Masquelier (2009) and McIntosh (2009), greater religious devotion is associated with upward socioeconomic mobility. In the Giriama case, for example, nonMuslims are more economically marginalized than their Muslim Swahili neighbors.

References:

Barrett, Deidre and Patrick McNamara, eds. 2007. The New Science of Dreaming: Cultural and Theoretical Perspectives, Volume III. Westport: Praeger.

Bernus, Edmond. 1981. Touaregs Nigeriens: Unite d'un People Pasteur. Paris: Editions de I'Office de la Recherche Scientifique et Technique de I'Outre-Mer.

Boilley, Pierre. 1999. Les Touaregs Kel Adagh: Dependances et revoltes: du Soudan francais au Mali contemporain. Paris: Karthala.

Borneman, John. 2011. "Daydreaming, Intimacy, and the Intersubjective Third in Fieldwork Encounters in Syria." American Ethnologist 38(2):234-249.

Bouman, Anne Marie. 2003. Benefits of Belonging. Utrecht: University of Utrecht.

Brett, Michael and Elizabeth Fentress. 1997. The Berbers. Oxford: Blackwell.

Camel, Florence. 1996. "Les relations touaregs/sedentaires a travers le regard du colonisateur francais." In Helene Claudot-Hawad, ed. Touaregs et autres Sahariens entre plusieurs mondes, 199-214. Aixen-Provence: Edisud.

Casajus, Dominique. 2000. Gens de la Parole. Paris: Editions la Decouverte Texts a I'Appui/Anthropologie.

Claudot-Hawad, Helene. 2002. Voyages d'un point de vue nomade. Paris: Presse d'Anthropologie.

Crapanzano, Vincent. 1973. The Hamadsha: A Study in Moroccan Ethnopsychiatry. Berkeley: University of California Press.

--. 1981. Tuhami. Chicago: University of Chicago Press.

Dragani, Amalia. 2012. Interno Tuareg: Ethnographia Partecipativa dei Poeti Nomadi LED Niger. Rome: Aracne.

Graham, Laura. 2003. Performing Dreams: Discoveries of Immortality Among the Xavante of Central Brazil. Austin: University of Texas Press.

Hawad, Mahmoudan. 1979. "La Tagdudt." Tisuraf: Groupe d'Etudes Berberes 8(3):79-82.

Jedrej, M. C. and Rosalind Shaw. 1992. Dreaming, Religion, and Society in Africa. Leiden: E. J. Brill.

Keenan, Jeremy. 2004. The Lesser Gods of the Sahara: Social Change and Contested Terrain Among the Tuareg of Algeria. London: Frank Cass.

Kohl, Ines. 2009. Beautiful Modern Nomads: Border-crossing Tuareg Between Niger, Algeria, and Libya. Berlin: Reimer.

Kohl, Ines and Anja Fischer. 2010. Tuareg Society Within a Globalized World: Saharan Life in Transition. London: I. B. Tauris Publishers.

Kracke, Waud. 1987. "Myths in Dreams, Thought, and Images: An Amazonian Contribution to the Psychoanalytic Theory of Primary Process." In Barbara Tedlock, ed. Dreaming: Anthropological and Psychological Interpretations, 31-54. Cambridge: Cambridge University Press.

Kuper, Adam. 1979. "A Structural Approach to Dreams." Man 14(4):645-662.

Lecocq, Baz. 2010. "Tuareg City Blues: Cultural Capital in a Global Cosmopole." In Ines Kohl and Anja Fischer, eds. Tuareg Society Within a Globalized World: Saharan Life in Transition, 41-58. New York: I. B. Tauris Publishers.

Levi-Strauss, Claude. 1963. Structural Anthropology. New York: Basic Books.

Lohmann, Roger. 2003. Dream Travelers. New York: Palgrave/MacMillan.

--. 2007. "Dreams and Ethnography." In Deidre Barrett and Patrick McNamara, eds. The New Science of Dreaming: Cultural and Theoretical Perspectives, 33-71. Westport: Praeger.

Mageo, Jeanette. 2013. "Dreaming and Its Discontents: US Cultural Models in the Theater of Dreams." Ethos 41 (4):387-411.

Masquelier, Adeline. 2009. Women and Islamic Revival in a West African Town. Bloomington: Indiana University Press.

McIntosh, Janet. 2009. The Edge of Islam. Durham: Duke University Press.

Mittermaier, Amira. 2011. Dreams That Matter: Egyptian Landscapes of the imagination. Berkeley: University of California Press.

Nicolaisen, Ida and Johannes Nicolaisen. 1997. The Pastoral Tuareg: Ecology, Culture, Society. Copenhagen: Carlsberg Foundation, Rhodos International Science and Art Publishers.

Nicolaisen, Johannes. 1961. "L'essai sur la Religion et la Magie Touaregues." Folk 3:113-160.

Obeyesekere, Gananath. 1981. Medusa's Hair: An Essay on Personal Symbols and Religious Experience. Chicago: University of Chicago Press.

--. 1992. The Apotheosis of Captain Cook: European Mythmaking in the Pacific. Princeton: Princeton University Press.

Oxby, Clare. 1990. "The Living Milk Runs Dry." In P. T. W. Baxter and Richard Hogg, eds. Property, Poverty, and People: Changing Rights in Property and Problems of Pastoral Development, 222-228. Manchester: Manchester University Press.

Rasmussen, Susan. 1995. Spirit Possession and Personhood Among the Kel Ewey Tuareg. Cambridge: Cambridge University Press.

--. 1997. The Poetics and Politics of Tuareg Aging: Life Course and Personal Destiny in Niger. DeKalb: Northern Illinois University Press.

--. 2001. Healing in Community: Medicine, Contested Terrains, and Cultural Encounters Among the Tuareg. Westport: Bergin & Garvey.

--. 2004. "'These are Dirty Times!': Transformations of Gendered Spaces and Islamic Ritual Protection in Tuareg Herbalists' and Marabouts' al Baraka Blessing Powers." Journal of Ritual Studies 18(2):43-60.

--. 2006. Those Who Touch: Tuareg Medicine Women in Anthropological Perspective. DeKalb: Northern Illinois University Press.

--. 2008. "The People of Solitude: Recalling and Reinventing Essuf (The Wild) in Traditional and Emergent Tuareg Cultural Spaces." Journal of the Royal Anthropological Institute 14(3):609-627.

--. 2010a. "The Slippery Sign: Cultural Constructions of Youth and Youthful Constructions of Culture in Tuareg Men's Face-veiling." Journal of Anthropological Research 55(4):471-493.

--. 2010b. "Re-making Body Politics: Dilemmas Over Female Fatness as Symbolic Capital in Two Rural Tuareg Communities." Culture, Medicine, and Psychiatry 3A(A):6t 5-632.

Rivers, W. H. R. 1922 [1918]. Conflict and Dream. New York: Harcourt.

Stewart, Charles. 2012. Dreaming and Historical Consciousness in Island Greece. Cambridge: Harvard University Press.

Tedlock, Barbara, ed. 1987. Dreaming: Anthropological and Psychological Interpretations. Cambridge: Cambridge University Press.

von Grunebaum, Gustave E. and Roger Caillois. 1966. The Dream and Human Societies. Berkeley: University of California Press.

Worley, Barbara. 1988. "Bed Posts and Broadswords." In Richard R. Randolph, David M. Schneider, and May N. Diaz, eds. Dialectics and Gender: Anthropological Approaches, 273-288. Boulder: Westview Press.

--. 1992. "Where All the Women Are Strong." Natural History 101 (11):54-64.

Foreign Language Translations:

An Ambiguous Spirit Dream and Tuareg-Kunta Relationships in Rural Northern Mali

[Keywords: Dreams, spirit possession/mediumship, religion, gender, Africa]

Targat n Kel Essuf enten Kel Tamajaq enten Kunta, dagh tenere n arewa n Mali

[Awal: targat, imijiwen n kel essuf, medden enten chadoden, Afrique]

Une Reve Ambigue et les Rapports entre le Touareg et les Kounta au milieu rural au nord du Mali

[Mots clefs: les reves, les amis des Kel Essuf, la divination, la religion, le genre, Afrique]

Urn Sonho Esplrita Amblguo e Relagoes Tuareg-Kunta no Norte Rural do Mali

[Palavras-chave: Sonhos, possessao esplrita/mediunidade, religiao, genero, Africa]

Susan Rasmussen, University of Houston

Source Citation

Source Citation   

Gale Document Number: GALE|A428175779