Abstract
Sociological research on inequality has increasingly moved beyond the examination of inequalities as they presumably exist to explore the generic narrative processes that perpetuate that inequality. Unfortunately, however, this research remains concentrated on either individual or ideological grand narratives and ignores the fact that the work narratives do, including the production and structuring of inequality, occurs at multiple levels: cultural, structural, organizational, and personal, and never exclusively at just one of these. In this study, we use Somali origin narratives to describe conceptually the ways in which narratives produced at different personal and societal levels—cultural, institutional, organizational—dialectically structure the generic processes that produce and perpetuate social inequality.
Over the past two decades, an increasing number of scholars have called for researchers to study social stratification from symbolic interactionist and particularly narrative perspectives (Anderson and Snow 2001; Chang 2000; Harris 2000, 2004; Maines 1992, 1993; Schwalbe et al. 2000). At the center of these calls is the argument that it is not sufficient to study social stratification and inequality because they presumably exist at the expense of how stratification and inequality are produced and reproduced in the first place (Harris 2001; Schwalbe et al. 2000). More important is the argument that social stratification and inequality are not a priori social fact, but, to borrow from Schwalbe et al. (2000:420), they “require attention to the processes that produce and perpetuate [them]” (see also Chang 2000; Collins 2000; Harris 2001).
The scholarly concentration on describing how inequalities are narratively produced instead of describing them as they exist has significantly contributed to our understanding of the nature of social inequality. However, most of the current research on narratives and inequality concentrates on either personal narratives (Harris 2000, 2004) or ideological level grand narratives (Chang 2000) and ignores the fact that the work narratives do, including the production and structuring of inequality, is dialectically produced at multiple levels—cultural, institutional, organizational, and personal—and never exclusively at one level (Loseke 2007).
In this study, using Somali origin narratives, we explore the ways in which narratives produced at the cultural, institutional, organizational, and personal levels dialectically structure the generic processes of othering, social stratification, and inequality. We define cultural level narratives as collective stories and representations that are embedded in a structured culture that constructs symbolic boundaries around types of social groups and individuals (Durkheim 1961; Lamont and Molnár 2002; Loseke 2007). By institutional narratives, we mean stories produced at the political and policy-making levels that construct and legitimize specific political and policy boundaries around groups and individuals. Organizational narratives refer to stories produced by organizations that are in the business of protecting vulnerable individuals and/or are engaged in human rights advocacy. Narratives produced at the personal level refer to stories produced as a result of individuals struggling to locate themselves in relation to the boundaries constructed at the cultural, institutional, and organizational levels (Loseke 2007).
We argue that the generic processes that produce, sustain, and perpetuate inequalities emerge from the continuous and reflexive interaction between narratives produced at the cultural, institutional, organizational, and personal levels. We specifically argue that with the continued interaction between narratives produced at different levels, they become fully entrenched as sui generis, social facts, to borrow from Durkheim, and ultimately become formulaic. And once formulaic, these narratives provide the generic formulas that structure the production and perpetuation of social stratification and inequality. However, since all narratives are incomplete in their hegemonic character (Maines 1993), they are continuously vulnerable to subversive stories (Ewick and Silbey 1995) and ultimately counter narratives that allow the excluded to find a mechanism for resistance. Counter narratives are therefore stories that rewrite the past in subversive ways. They defy the legitimization of the oppressors’ narratives and may even at times politically transform and set the conditions that generate counter hegemonic narratives (Ewick and Sibley 1995). For the purpose of this study, counter narratives are stories that continuously attempt to defy, even if unsuccessfully, the inequality structures that the Somali formula narratives erect and impose on those they exclude from the social boundary of Somaliness.
We begin with a brief discussion of the recent literature on symbolic interactionists’ understanding of the ways in which narratives structure the generic processes that produce and perpetuate social inequality. Second, we describe narratives and provide a conceptual framework for understanding how they become entrenched in institutional, organizational, and personal levels and ultimately become formulaic, a fundamental aspect of the social and cultural constitution of society. Third, we discuss how Somali formula narratives structure social boundaries to delimit who belongs and who does not belong in the social boundary of Somaliness and therefore the nature of social inequality. Fourth, we examine how the assumptions and logic of the formula narrative provide loopholes that allow those it excludes to come back with their own counter narratives to resist the social stratification parameters set against them by the formula narratives. We conclude with empirical and theoretical suggestions for incorporating narrative approaches in research on inequalities and reviewing policy implications.
Literature Review
The calls to explore social stratification and inequality from a symbolic interactionist or narrative viewpoint have generated a number of theoretical and empirical studies concerning the situations in which social inequality is produced, reproduced, and continuously contested (Chang 2000; Collins 2000; Harris 2001; Schwalbe et al. 2000). For example, following his longstanding discussion of the microfoundations of macrosociology, Collins (2000) argued that instead of using the normative sociological focus on the structured hierarchies of inequality measured in terms of aggregated differences in income, wealth, education, and occupation, a more meaningful way of understanding social stratification is to study the microfoundation instances in which stratification is produced through everyday interaction and to examine how these situations form the basis for macrostratification. According to Collins (2000), microsituational observations have conceptual priority because the construction of inequalities is fundamentally situational such that individuals “may rank one way on an abstract occupational prestige survey but the opposite when evaluated by their companions at certain social gatherings” (Collins 2000, quoted in Harris 2004:116). Even though, according to Collins, a “plumber may rank low in occupational prestige surveys, in practice, their income outranks many educationally credentialed white-collar employees, and this may translate into material resources to dominate most life situations; plumbers may sit in the box seats at the stadium while white-collar workers are in the remote grandstand” (Collins, quoted in Harris 2004:18). In other words, to understand the nature of social stratification fully, we need to explore the ways in which the actual distribution of advantages and disadvantages is or is not realized in situational practice.
Using content analysis of a number of published qualitative and ethnographic scholarly papers concerning one or another aspect of social stratification and inequalities, Schwalbe and colleagues (2000) proposed a framework for studying social stratification and suggested four generic processes as central to the reproduction of inequality: othering, subordinate adaptation, boundary maintenance, and emotion management. Othering, as a generic process that reproduces inequalities, takes two forms: oppressive othering and defensive othering among subordinates. Oppressive othering refers to situations in which one group advances its social, political, and economic position by defining members of another group as socially and morally inferior, while defensive othering occurs when those who are othered engage in an intracommunal process of othering members of their own social class in the hope of deflecting the stigma and dented identities they experience, such as the case of homeless individuals defining other homeless individuals as lazy (Snow and Anderson 1987).
Social stratification and inequalities are also reproduced and experienced through subordinate adaptation by trading power for patronage, forming alternative cultures (see also Ogbu 1978), and hustling or dropping out. Albeit in different ways and degrees, both othering and subordinate adaptation processes contribute to the reproduction of inequalities. Snow and Anderson (1987) explored the everyday situations in which stratification and differences manifest themselves at the micro level, the consequences of being the victim of stratification and inequality for those who are stigmatized, and the manners in which they manage the status affronts and subtle reminders of the inequalities they constantly tolerate in everyday life. According to Snow and Anderson (1987), homeless people use several strategies, including distancing by disassociating from similarly situated individuals and/or the institutions that serve them; narrative embellishment of past, present, and future expectations by fictive storytelling; and embracement by the expressive confirmation of the social roles and statuses associated with homelessness so as to avoid the everyday stigma associated with being a homeless person.
Other scholars have attempted to define inequality from a symbolic interaction and from narrative sociology to explore how married individuals define and understand marital equality and/or inequality (Harris 2001) and to examine the impact of ideological narratives on the transformation of class stratification in China (Chang 2000). One important application of the tenets of symbolic interaction and narrative sociology on social stratification is found in Chang’s (2000) analyses of how the 1970s post-Mao reform, which redefined China’s political and economic ideologies from class-based struggle and conflict to pragmatic economic and market development, resulted in the transformation of class stratification in China. Chang suggested that this process increased the number of classes in China from five to seven and increased the extent of inequality between classes more so than during the Maoist era. This transformation of China’s class structure, according to Chang (2000), resulted not from objective capitalist economic conditions in the sense of Karl Marx but from meaning making and particularly the ways in which the Chinese Communist Party, led by Deng and his followers, used reform ideological narratives intended to move China from Maoist collective-based economic system to a market-based economy in the late 1970s.
Harris (2000) examined the ways in which married individuals actually understand inequality within their own marital contexts and how they attach meaning to the source of such inequality. According to Harris, some individuals describe their inequality through differences in the economic power between them and their partners, while others understand equality or inequality with their partners in religious and biblical terms.
Conceptual Framework
In general, we draw on David Maines’ (1993:32) conceptualization of narrative sociology in the sense that “any of the standard interest claimed as central to sociology can be approached through a narrative ontology.” We specifically draw, however, on Loseke’s (2007) argument that a meaningful understanding of the work that narratives do requires explicit attention to the relationships among cultural, institutional, organizational, and personal identity narratives to explore how Somali formula narratives provide a framework for the production, reproduction, and continuous contestation of social stratification and inequality. Critical to Loseke’s framework is that narratives are produced at different contextual levels: cultural, institutional, organizational, and personal levels. Cultural narratives reflect widely circulating symbolic codes that organize and construct social and symbolic boundaries, describe binary categories of people, and ultimately prescribe specific social relations among categories of people. Institutional narratives, in contrast, refer to those narratives that enter into the policy-making process. These may be narratives that enter into government policies constructed around, for example, deserving/undeserving welfare recipients, the Vietnam veteran, or the disloyal Japanese American during World War II. In essence, institutional narratives provide the justification necessary to construct specific boundaries around those who are, and those who are not, privileged under a particular policy regime. Organizational narratives are primarily service provider stories that encourage victims seeking help to conform to the parameters of institutionally produced victimhood narratives. Personal narratives refer to stories that social actors tell to locate themselves into those narratives produced at the cultural, institutional, and organizational levels.
Narratives produced at the individual level filter up to the macro level, are recycled as cultural narratives, and filter down to the micro level in an unending recursive loop; develop as widely circulating narratives; and ultimately become entrenched in the cultural, institutional, organizational, and personal levels of society. The process through which narratives become formulaic occurs through a complicated web of interaction between members of the Somali dominant groups, their leaders, prestige bearers, officials, group agents, and ordinary laymen who engage in a collective disparagement of the subordinate group through talk, tales, stories, and gossip, and the narratives run through society until they become a collectively shared understanding (Blumer 1958). And once fully entrenched, they acquire a life of their own, become sui generis or social facts (Durkheim 1961), and ultimately become formulaic (Loseke 2007) to provide the generic formula that structures the production and infinite perpetuation of social inequality.
Regardless of the contextual level at which narratives are produced, however, they must, according to Maines, satisfy certain preconditions: First, events must be selected from the past for purposes of focus and commentary. Second, those events must be transformed into story elements. Third, a temporal ordering of events must be created so that questions of how and why events happened can be established and the “narrative element can acquire features of tempo, duration, and space” (Maines 1993:21). When events are transformed into a narrative story such that they acquire both narrative probability (coherence) and narrative fidelity (believability), they become inherently political and set conditions on social action and interaction that embody social stratification and inequalities. Moreover, narratives, according to Maines, enter into sociology’s domain primarily in two ways. First, narratives can enter the process of research as an object of inquiry and analyses, such that researchers can examine the different ways in which people rely on narratives in interpreting and making sense of their social world. Second, researchers themselves can become narrators and perform the analytical process through which these stories are given shape and provide meaningful coherence.
Since formula narratives essentially construct symbolic codes that organize and construct symbolic boundaries, we also draw on Lamont and Molnar’s (2002) discussion of the importance of symbolic boundaries as a primary vehicle via which social actors characterize their social and physical environment. Symbolic boundaries are tools by which individuals struggle over and come to agree upon the definition of reality, a means through which social actors put individuals into groups and generate feelings of similarities and differences, and ultimately the essential means through which people acquire social status. Once collectively accepted and socially hardened, symbolic boundaries become social boundaries that produce objectified social differences manifested in the unequal distribution of resources, status, economic-driven patterns of interaction and association, and, most important, identifiable patterns of social exclusion and social inequalities.
Background: Somalia
The modern Somali Republic was created as a result of the unification of southern and northern Somalilands previously administered by the Italian and British colonial powers, respectively. The British granted independence to the northern region on June 26, 1960, and the southern region became independent five days later after being a United Nations Trust Territory administered by the Italian colony. After two successful parliamentary-based civilian administrations from 1960 to 1964 and 1964 to 1967, a military regime toppled the third administration in October 1969 and remained in power until 1990, when the whole country descended into a generalized civil war. Somalia remains without a consensually recognized national government to this day.
Social Organization
The primary category of social organization in Somalia is the clan system, which acknowledges four major clan families (Darood, Digil-Mirifle, Dir, and Hawiye) as constituting the primary Somali ethnic people. The Dir settled in both the northern and southern parts of the country. The Darood clan family primarily settled in the northeast and in some parts of southern Somalia. The Digil-Mirifle clan family settled in the regions between the Jubba and Shabelle Rivers in southern Somalia, while the Hawiye inhabit land in and around the Banadir, Hiran, Galgadud, and Middle Shabelle regions. These four clans are considered to comprise what is generally deemed the Somali ethnic identity.
Somali Minority Groups
The Somali social structure also acknowledges three major minority groups: 1) the Somali Bantu Jareer; 2) the Banadiri Reer Hamar; and 3) the occupational caste groups. In this paper, we focus on the Somali Bantu Jareer and the occupational/cultural caste groups since they represent the largest and most socially othered groups. The Somali Bantu Jareer are probably the largest ethnic minority in Somalia. They are primarily distinguished from the rest of the Somali population by their appearance, which, according to Somalis, is more African than the typical Somali. The Somali Bantu Jareer community consists of two clusters: indigenous groups and former slaves. The history and settlement of the indigenous Bantu groups can be traced back several thousand years as part of Shungwaya, a well-established ancient kingdom that preceded Somali settlements in the region. The second segment of the Bantu Jareer community resulted from the tens of thousands of slaves imported from Tanzania and Mozambique to southern Somalia as part of the nineteenth century Indian Ocean slave trade.
The occupational caste groups, according to Mohamed-Abdi Mohamed (1997:149), “are either the descendants of previously noble groups who were defeated and had fallen after a war against another group or descendants of the first inhabitants of the Horn.” Eno and Eno (2010) wrote that before the arrival of the Somali forefather, and prior to their defeat by the Somalis, the Gabooye people lived in most of the northern part of the country. Through interviews with members of the various outcaste groups, Eno and Eno (2010:122–3) revealed how these minorities became ostracized due to their resistance against foreigners and how “the Somalis took sides with the immigrants” to devastate the Gabooye Kingdom at the time.
However, unlike the Bantu Jareer, whose othering is achieved by way of perceived racial differences and a background in slavery, the occupational caste groups are not in any way distinguishable from the Somali population. They speak the local languages or dialect of the specific clan(s) with whom they live and are culturally and ethnically similar to them. They are different from the Somali clans with which they live in the sense that they are castes who have practiced occupations that the members of the major Somali clans have historically despised. The occupational caste groups are Gabooye, Tumaal, Yibir, and Midgaan. The Gabooye were historically hunters and leatherworkers. The Tumaal were blacksmiths, and the Yibir were ritual specialists who provided blessings for newlyweds and newly born children. The occupational caste groups are found in every region of Somalia.
Methods and Data
The history of when and how Somali origin narratives came to be is extremely difficult to establish with any certainty. However, they are part of the oral storytelling traditions of the earliest societies where narratives were used to address fundamental questions pertaining to issues of meaning and knowing (Hendry 2007, 2010). In addition, such narratives set the basis for the legitimation of existing power by dominant groups within society “through references to their descent from a mythical figure credited with establishing the existing social order” (Fuller 1988; Bauer 1991:7).
It appears that the geography of Somalia, being at the junction between the Islamic world and the rest of the East African region, has played an important role in the formation of the origin narratives. The early Somali origin narratives were part of the general ancestor myths that were prevalent in the east African region before the arrival of Christianity and Islam in the region (Cerulli 1922, 1933, 1957; Luling 1988). These ancestor narratives were part of that age-old metaphysical realm where humans engaged myths, fables, spirituals, rituals, and dance. These narratives, according to Hendry (2010:75), “served as forms of inquiry to explore and examine inevitable doubt and questions human beings experience in an unpredictable cosmos filled with mystery and awe,” including questions of creation. Early Somali communities found answers to these complicated questions by creating myths about ancestors who descended from the sky “for they saw it not so much as God’s dwelling place as God himself” (Luling 1988:333).
With the introduction of Islam into the region, the idea of ancestors descending from the sky became incompatible with the teachings of Islam, which emphasized the separation of God from the human realm. For those who converted to Islam, it made much more sense to locate their ancestor in the birthplace of Islam, and hence the ancestor now comes from across the Red Sea in Southern Arabia. The narratives about the origin of the Somali people are expressed in oral form and firmly retained in the social memory and oral history of the Somali people from generation to generation, such that the majority of Somali people take these narratives as scriptures and retain them as part of their everyday psychology. In fact, the notion that the original Somali ancestor came from somewhere in southern Arabia, the birthplace of Islam, is, if not collectively agreed upon, a generally understood Somali formula narrative. It is a narrative that speaks to the nature of social structure, social relation, and ultimately how the Somali social and ethnic identity was shaped to its current structure. It is presented in the form of origin narratives.
The analysis presented here was undertaken with a set of narrative frames extracted from Somali origin narratives. The actual narratives were collected from historical and anthropological studies (Burton 1943; Cerulli 1922, 1933, 1957; Luling 1988) conducted in colonial and postcolonial Somalia and the authors’ extensive knowledge of Somali society. The Somali origin narratives also form part of contemporary Somali public memory. To examine the intersections between narratives and their social actors, we use a combination of analytical techniques derived from grounded theory (Glaser and Strauss 1967), narrative analysis (Reissman 2001), and frame analysis (Goffman 1975). We begin with open-coding by way of a detailed reading of the entire narrative followed by axial and selective coding (Charmaz 1990; Glaser and Strauss 1967).
In the open-coding phase, we read the entire narrative for the presence and/or absence of commonly accepted elements of narratives, the selective appropriation of past events and characters, the temporal ordering of events and characters, and the relation of events and characters to overarching societal structures, in this case the structures of inequalities. This process allowed us to identify broad narrative frames and compare them for similarities and differences. In other words, we used open-coding to fracture data so as to identify important narrative frames and their properties and dimensions. Once we identified those, we further examined the data for connections between narrative frames and subframes. Finally, by using selective coding, we compared all the narrative frames for similarities and differences until we found a central theme and frame. This resulted in the identification of Somali formula narratives that construct a social boundary between the descendants of the original ancestor and therefore those in the social boundary of Somaliness and those who did not allegedly descend from the original ancestor and therefore were located outside the social and ethnic boundary of Somaliness.
Narrating Social Inequality
In what follows, we present the Somali formula narrative, how the narrative delimits the social boundary of Somaliness, and the process by which it creates otherness. The Somali formula narrative creates a social boundary of Somaliness on the basis of who is, and who is not, considered a descendant of the original ancestor from Arabia.
Figures 1 and 2 present a conceptual model of the way in which the ontological assumption of the Somali formula narrative is organized and structured. On one hand, the narrative acknowledges (Figure 1) members of the four major clan groups, Darood, Digil-Mirifle, Dir, and Hawiye, as the supposed descendants of the original founding ancestor. They are essentially considered as forming the social boundary of Somaliness. On the other hand, it removes two groups (Figure 2) from the social boundary of Somaliness and essentially engages in what Somalis refer to as Heeb Sooc, or what sociologists call othering. The first group of the two communities in this context consists of Gabooye, Midgaan, Tumaal, and Yibir—who are removed from the social boundary of Somaliness on the assumption that they are socially polluted, as we will learn later, because their ancestor ate spoiled meat. The other group, the Bantu Jareer community, is removed on the grounds of racial differences. It assumes, as we will see later, that they are a mixture of non-Somali groups who occupied the land prior to the arrival and spread of the descendants of the original Somali ancestor and groups who were brought to Somalia as slaves from Tanzania and Mozambique during the nineteenth century Indian Ocean slave trade who were later incorporated into the social fabric of southern Somalia. They represent a significant percentage of Somali society today and are referred to as Somali Bantu Jareerwayne, a reference that they are more African-like, whatever that means, than the so-called pure Somalis. In what follows, we will start with the part of the Somali formula narrative which establishes how the original Somali ancestor came to lay the foundation of what is considered the primary source of what Somaliness is today.
Founding ancestor (Muslim and supposedly of Southern Arabia).
Excluded from the social boundary of Somaliness (Heeb Sooc).
Somali Origin Narratives
Narrative 1a: The Somali ancestor emigrated (exiled in other versions) from southern Arabia and settled in present-day Somalia. This ancestor later married a local woman and started what became the basis of the ethnic identity of the contemporary Somali nation. Narrative 1b: The Somali ancestor was found sitting on a tall tree by a young girl who was then grazing her flock. He refused to come down until the girl called her brothers (father in other versions) and the elders of the clan. He asked for three things: a camel, a slave, and the girl’s hand in marriage. On top of that, he asked to climb down on the back of the girl’s father. The father accepted the offer, and then the ancestor-to-be (for the Somalis) came down, married the girl, and laid the ancestral foundation for the first Somali family.
As shown in narrative versions 1a and 1b and in Figures 1 and 2, the Somali formula narrative provides the master frame through which the nature of social stratification and inequality is produced and maintained. It follows from the premise that the Somali principal ancestor originated from east of the Red Sea by way of a boat, logically from the Arabian Peninsula (narrative 1a) or the sky or heaven (narrative 1b), and married a local woman. This union, according to the formula narrative, started what later became the modern Somali identity of Muslim and not purely African. The religious identity of Islam and the ethnic identity of Arabness were used each as a basis for constructing the process of othering and boundary making. In the case of the occupational caste groups, the Islamic identity of the ancestor was used as a strategy to impose otherness on those who allegedly insisted on retaining their pre-Islamic identities and values. As a result, they were removed from the boundary of Somaliness and were forced into outcast status. In the case of the Bantu Jareer, the Arab identity of the ancestor was used to implement otherness, by accusing them of having African ancestry. In other words; the Somalia origin narratives divide the Somali society into two status groups: (1) descendants of the original ancestor who consider themselves as the true or noble Somalis, and (2) those unrelated to the original ancestor either by way of religious difference or racial distinction. The first group is collectively characterized as ignoble and is variously known as Gabooye, Midgaan, Tumaal, or Yibir. They are considered as the descendants of the man who ate the meat of the dead animal but refused to vomit and who ultimately became a ruthless king. Therefore, on the basis of those circumstances, they are considered biologically different and socially polluted. The second, the racialized group, is known as Somali Bantu Jareer. They are removed from the social boundary of Somaliness on the basis of racial difference and are discriminated against in all forms and facets of social and political life. In the next section, we analyze the ways in which minority groups are removed from the social boundary of Somaliness. We will start with the outcaste groups, followed by the racialized groups.
Narratives and Caste Stratification
Narrative 2a: Before the arrival of the founding ancestor, a vicious magician king known as Bucurbacayr (Bu’urbaeir) ruled the country. He was a ruthless king who terrorized the people, raped women, killed innocent children, and, in general, exploited the people until the founding ancestor, with the help of Saint Aw Barkhadle, caused two mighty hills to close on him.
Narrative 2a removes all outcaste groups (Gabooye, Midgaan, Tumaal, and Yibir) from the social boundary of Somaliness. It is derived from the Somali formula narratives (1a and 1b), specifically the part that limits the social boundary of Somaliness on the basis of those who embraced Islam and those who allegedly did not. The above narrative presents one version of how the Somali formula narrative created a Somali occupational caste group. Since the Somali mythical ancestor is narratively considered to have migrated from southern Arabia or from heaven, and because Islamic values dictated the new moral values of the society, the first task of the narrative was to engage in the removal of non-Islamic cultural traces and heritages and their retainers from the social boundary of this new Somaliness (Kusow 2004). Subsequently, it constructs a narrative about the defeat of a king of those who retained pre-Islamic cultural values. According to this narrative, the defeat of the vicious, non-Muslim indigenous king and the victory over him by the Somali saint created the line that later distinguished the Somali society as noble versus the others who were lower caste.
Another version of the narrative that removes the Somali occupational caste groups (Gabooye, Midgaan, Tumaal, and Yibir) from the social boundary of Somaliness goes as follows: Narrative 2b: The ancestors of both the noble group and the nonnoble castes were two brothers. Before setting out on a long journey, their father advised them that in case they became hungry at any time during the journey, they could eat whatever they found, even the meat of a dead animal. However, the father warned that upon reaching their final destination, they should force themselves to vomit in order to cleanse their bodies from the spell and impurity of the dead animal. As the narrative goes, midway through the journey the brothers became so hungry that they ate the meat of a deceased animal. After they reached their final destination, the younger brother followed his father’s advice and forced himself to puke out the dead animal’s meat, while the older brother disobeyed it. After that incident the descendants of the younger brother ennobled themselves, while those of the older brother became classified as outcasts.
Narrative 2b provides a similar process of removing those who retained pre-Islamic values from the social boundary of Somaliness on the assumption that Islamic values that prohibit eating the meat of improperly slaughtered animals (not Halal) made the descendants of the younger brother biologically impure and therefore unfit to remain in the social boundary of Somaliness. Both narratives (2a and 2b) have been very successful in effectively marginalizing a significant portion of Somali society as having an unholy origin. Consequently, the occupational groups of Gabooye, Midgaan, Tumaal, and Yibir remain outside the boundary of Somaliness to this day. They are not allowed to intermarry with the rest of society. In some situations, they cannot shake hands with their self-ennobled cousins, or eat with them, simply because they are considered as socially polluting.
Narratives and Racial/Ethnic Stratification
Narrative 3: As the descendants of the Somali original ancestor multiplied and spread into Southern Somalia, they come into contact with non-Somali communities, some of whom they pushed further into the west, but those who remained were incorporated into the Somali social fabric as inferiors. These groups were later augmented with thousands of slaves imported into the Banadir coast and in general into southern Somalia as a result of enslavement. Consequently, the Bantu Jareer community does not have the claim to Arab ancestry as the Somalis do. They are ethnically Africans and considered all as descendants of slave ancestors who do not have physical resemblance to claim Somaliness. Based on these accounts of disqualification, the Bantu Jareer are considered as a community inferior in status and therefore ethnically lower than the Somali proper. These dissimilarities are sufficient reasons to determine the Bantu Jareer as degradable people unequal to the Somalis.
The second major Somali community narratively removed from the social boundary of Somaliness is the Bantu or Jareer groups in (Narrative 3). The narrative involving the marginalization of the Bantu Jareer was situated on the premise of their African descendance and physical features, which are allegedly more African and therefore easily distinguishable from the physical characteristics of the Somalis. Due to those racial and physical distinctions, they are classified with abusive epithets like Adoon (slave) or Jareer (kinky hair). In other situations, they are referred to as Gosha (people of the forest) or Shabelle (people of the Shabelle River). Despite the variations in nomenclature, all such references are derogatory in nature.
Beyond the alleged physical characteristics, though, the Bantu Jareer people as a group are considered to have descended from slaves exported from the eastern and southeastern regions of Africa and are therefore seen as having no legitimate claim to Somaliness as an identity or the land they occupy. This argument is, of course, not empirically true because we have considerable research showing that some of the Bantu groups have been settled in the region for much longer than the so-called Somali nobles (Besteman and Cassanelli 1996). Ultimately, however, the Bantu Jareer Somalis have suffered an enduring social, political, and economic marginalization. The narrative surrounding the dehumanizing of Bantu Somalis has been reported by a number of academics and by relief agencies and human rights organizations (Ingiriis 2012; IOM 2002; Owens 2004). Drawing on the vulnerability of the minorities (Bantu Jareer included) in preconflict Somalia, Hill (2010:3) emphasizes how “regardless of the conflict,” the minorities had “already suffered marginalization and exclusion from mainstream economic, social, and political life.”
Institutionalizing Narratives
The symbolic boundaries of othering created by the formula narratives have been institutionalized at the constitutional/political level. In other words, the Somali formula narrative’s cultural processes of narrating exclusion and the reproduction of inequality in Somalia have been institutionalized at the national level through a political system now known as the “4.5 formula” of political power sharing, which Eno (2008) criticized as “endorsing apartheid” at the national level. The formula gives equal political representation to the four so-called major clans (Darood, Digil-Mirifle, Dir, and Hawiye) that supposedly share a common Somali pedigree and gives half of a share to what the Somali power brokers referred to as others; a connotation that indicates less Somaliness, in other words, those with no claim of ethnic connection to the so-called Somali founding ancestor.
The 4.5 formula for political power sharing represents the social fiber that encapsulates and feeds from the Somali founding ancestor narratives. It was incubated in Sodere, Ethiopia, in 1997 and introduced in 2000 during the 13th Somali reconciliation conference in Arte, Djibouti. In other words, Somali political and cultural elites incorporated the social distinctions created by the formula narratives into the political and nation-making processes in a way that is accessible to organizations and both victim and victimizer. The participants divided the Somali society into superior and inferior groups: descendants of the so-called Somali original ancestor (Darood, Digil-Mirifle, Dir, and Hawiye) as superiors and the nondescendants (Bantu Jareerwayne and occupational caste groups) as inferiors.
Organizational Level Narratives
In most situations, and in almost all societies, organizations in the business of protecting vulnerable individuals and/or engaged in human rights advocacy come to the aid of the stigmatized and, in general, vulnerable individuals and groups. However, these organizations need their own narratives to justify their actions. They create what Ann Swidler (1986:273) calls a “tool kit of symbols, stories, rituals, and world-views” that encourages victims to engage in the process of turning their plight into globally audible stories. It is those audible stories that allowed human rights advocates to make a case for Somali Bantu Jareer families in refugee camps in Kenya to be resettled in the United States. For example, before the Bantus’ and outcaste groups’ personal narratives were audible to the international organizations, all Somali refugees, oppressed and oppressor alike, were invariably hosted in the same refugee camp spaces. However, once their narratives became audible, the United Nations High Commission for Refugees (UNHCR) had to reformulate its policy toward the different Somali groups, based on the evidence of the existence of marginalization among the Somalis in the camp. As a result, the Bantu and the caste groups were put in separate spaces in the camp away from the noble-claiming Somalis. Later, the Bantus had to be relocated to another camp about 1,500 kilometers away from other camps where members of the so-called noble clans recreated the social division that had existed between Somali majority and minority groups. All international organizations, the UNHCR, the International Organization for Migration (IOM), Care International, and other service providers followed this division in terms of all their provision of assistance to Somalis in the refugee camps, categorizing the groups as Somalis, Somali outcasts, and Somali Bantu.
In time, the separation of all Somalis in the refugee camps into oppressor and oppressed and the kind of victim narratives provided by Bantu and outcast individuals seeking resettlement to the United States and other parts of the West became standardized case reports as filed in the organizational immigration records. The policy of separation was used to protect the marginalized groups from the imminent domination of their oppressors who had immediately upon arrival in the refugee camps created leadership structures that obscured the equal rights of the minorities. More significantly, the policy empowered the minorities to establish their own independent leadership structures to liaise directly with the international organizations providing service to the refugees. It also resulted not only in the allocation of separate settlements but also provision of separate schools for the minority children who were facing extreme forms of marginalization when sharing classrooms with their noble-claiming peers in the refugee camps.
Moreover, even in the regional offices and suboffices of these service-providing organizations, identities such as “Somali Bantu,” or “Somali outcast” have become very effectively constructed to the point where they became synonymous with “oppression,” marginalization,” and “vulnerability.” In the course of time, the organizational symbols or narratives of identity paved the way for individual and community-wide consideration for resettlement in many countries in the West, mainly in the United States, as a durable solution to the persecution they had been facing over the years and the possibility of suffering further persecution upon returning home. Understanding the interaction between client and advocacy organizations is extremely important in that the construction of group identity based on victim narratives alone could not create normatively audible stories without the active participation of the listeners, the organizations that had the access to influence. As a former refugee from one of the camps in Kenya admitted in a discussion with one of the authors (Mohamed) in Louisville, Kentucky, in 2008, “Without high level advocacy campaign by the UNHCR and other service providers, we would still be rotting in the refugee camps in Kenya.”
Narratives and Discrimination
We must note, however, that the work narratives do is not confined to the processes that produce and reproduce inequality but also leads to both individual and institutional discrimination. In other words, narratives become actual social boundaries and the basis of institutional and individual discrimination against those removed from the social boundary of Somaliness. Once collectively accepted and socially hardened, Somali formula narratives become social boundaries that produce objectified social differences manifested in unequal distribution of resources and, more importantly, in exercising discrimination. In the following section, we examine the institutional and personal ways in which Somali Bantu and Somali outcaste individuals (Gabooye, Midgaan, Tumaal, and Yibir) articulate the kind of discrimination and social exclusion imposed on them through the formula narratives. This discrimination is not confined to either the cultural, institutional, organization, or personal levels alone but encompasses all levels. The information for this section is derived from a report by Martin Hill (2010) for the Minority Rights Group International on individuals of the minority groups.
A woman from one of the majority/nobility-claiming clans shares some of the difficulties her friend faced when she attempted to marry a man from the outcast groups: I know a girl from the Hawadle clan who got married to a Midgan man. They were neighbors in Beletweyne [in central Somalia], but her family did not accept her choice. She has five children, three boys and two girls. Her parents no longer consider her as their daughter and severed contacts with her. She loves her parents and wants to visit them, but she fears they might harm her for her choice of husband. Realizing the ordeal, her loving husband decided to divorce her so that her “dignity is restored.” (Hill 2010:15)
The marginalization of the Somali outcast groups extends to the institutional level as well. One outcast female interviewee shares the difficulty she faced in attempting to secure employment: The Issaq will never give you a job, and they will always call you names and say, “Why are you letting your parents pay so much for an education which will not lead you anywhere? Why don’t you stay at home and help your mother?” (Hill 2010:15)
This stigmatization of outcast groups is, in fact, part of the Somali public school culture as the following interviewee discusses an incident she encountered in Somaliland: When I was at school my teacher did not know that he had a Midgan in the class, so he went on with his lesson on minority clan culture and tradition. He said that the Midgan are different from the rest of the society. They belong to an inferior culture. . . . They lack language skills and eat different, bad quality food. I did not react at all, but all my schoolmates were shocked. I just waited ’till the end of the lesson and then ran home. I told my mother and sister. The teacher did not realize the moral damage that his speech had on me. He probably never thought he could have a minority student in his class. (Hill 2010:17–8)
The Gabooye Madhiban minority communities also face significant victimization as seen in the following account of a 50-year-old Madhiban female in Bossaso: I usually go early in the morning to the market. Five years ago I was with my 14-year-old daughter. We were walking down the road to the market when a car approached us. Six men in military uniform forced my daughter to get into their car and beat me as I tried to stop them. . . We found her at the same place the following day. It seemed that she was not even alive; she was like a dead body. I reported it to the police and was told that I was lucky that she was still alive. My husband was also beaten up in the police station for insulting the police authorities and wrongly accusing them. He started suffering from high blood pressure as a consequence of the physical and psychological injuries and died one year later. Six of them raped her brutally and repeatedly ’till she fainted. I did not have money for her hospitalization. . . . I also tried to contact journalists to denounce the case, but I had no evidence and I was overwhelmed with those problems. After five years she still has problems while urinating. She married last year, but she has not given birth yet. My daughter is seriously injured, and she needs medical intervention. If we had been from another clan we would have been given compensation but we are just “poor Midgan, who nobody cries for.” (Hill 2010:20–1)
A 42-year-old Somali Bantu woman in Bossaso describes her ordeal this way: The first time in 2002, four men raped me. I at first tried to escape but then I was beaten. I lost some of my teeth as a result of the beating. I felt humiliated and did not report it to the police. Another time in 2004, six men in army uniform attacked me and raped me. I was with other women, but they managed to escape. I had physical problems as a result of the brutal rape, but I did not have enough money to seek medical treatment. I just took medication from the pharmacy. I did not report it to the police because I was too afraid. I got pregnant. I now raise my four-year-old daughter; her father is one of those six men who raped me. I love my daughter. Sometimes the bad memories come back, and I cry in silence (Hill 2010:20).
Another 40-year-old Somali Bantu driver in Mogadishu had the following to say: Life is hard for everyone in Mogadishu, and our lives are continuously under threat. But we also have the burden of discrimination and daily humiliation. I am somehow used to it, but I feel anger and sadness when I think that even my children go through this discrimination. We are used to the heavy jobs that others do not want to do. If you are a Jareer, you have to work hard because nobody will help you in this society. If I have an accident with another car, it is always my fault and I am insulted. My wife works as a maid. Once, she was arrested and beaten up in jail. She was accused of stealing a gold chain, and there was no evidence. They finally released her, and the gold chain was eventually found somewhere else. This is our life; we are discriminated against in our own country. (Hill 2010:22)
Overall, the nature of discrimination and victimization presented above shows that Somali formula narratives are not just symbolic boundaries but also social boundaries that carry significant social, political, economic, and quality of life consequences for those who are removed from the social boundary of Somaliness.
Counter Formula Narratives
As shown in the preceding discussion, the primary assumption of the Somali origin narrative, that the founding ancestor came from outside the Somali region (narrative 1a) or from the sky or heaven by way of a tree and descended on the back of an indigenous person (narrative 1b), creates a formidable generic process of social stratification and inequality by dividing the Somali society into two binary groups: (1) a settler group that claims descent from an original immigrant ancestor and (2) indigenous groups that do not make such claims and who are therefore removed from the social boundary of Somaliness.
However, since “all narratives are ultimately incomplete” (Maines 1993:21), they potentially create loopholes with which those who are excluded from the narrative can continuously modify, challenge, and discard the narrative (Loseke 2007) and at times can create counter formula narratives. The Somali counter formula narratives do just that: shift the ontological assumption of the ancestor-based priorities with territorial and moral ones. In fact, they are primarily an inversion or a subversion of the Somali formula narrative. Specifically, the counter formula narratives (1a, 1b, 1c and Figure 3)construct the social boundary of Somaliness in terms of either territorial priorities (which group is the rightful owner of the land) or moral priorities (which group observes respect and human dignity as opposed to those who do not respect morality and prefer anarchy) as the following: Territorial-based counter narrative 1a: Since the original ancestor of the dominant Somali clan groups—Darood, Digil-Mirifle, Dir, and Hawiye—originally came from outside Somalia, they should return to their ancestral land (i.e., Arabia) and leave Somalia for those indigenous to the land.
Counter narratives.
The Somali counter narrative also engages in an exercise of moral subversion with the suggestion that the hegemonic and warlike tendencies of the descendants of the original ancestor are responsible for most, if not all, of the social ills of the country and should not be tolerated anymore. The moral priority-based narratives specifically emphasize the degree of anarchy displayed by a clan or found in a regional setting as revealed by a dialogue involving the late Mohamed Rajis, a renowned Somali lawyer, and a certain warlord at a Somali reconciliation conference in the 1990s.
Moral-based counter narrative 1b: According to this narrative, one of the warlords informed Rajis during one of the deliberations that since members of his clan did not take arms during the war, he had no right to be in their company, let alone be part of the power-sharing processes. The learned Rajis responded rather ironically, “If this is the only criterion for one’s inclusion in the social boundary of Somaliness and the right to take part in this reconciliation conference, then I will go back to Somalia, kill a few innocent people and come back, at which point, I am sure you will accept me as a certified member of the community.”
The moral plot of the counter narrative is also revealed in the well-known dialogue between a nomad burglar and his victim, a Banadiri Reer Hamar, urban dweller in Mogadishu, the capital city of Somalia.
Moral-based counter narrative 1c: The narrative relates that one evening a burglar snuck into the house of a well-to-do Banadiri/Reer Hamar/Reer Baraawe family in the middle of the night to steal their cow. While he was still untying the cow, the owner’s wife saw him through the window and informed her husband of the unfolding event, to which the owner replied, “It is OK. Don’t worry. He will bring it to the market tomorrow, and there I will buy it from him.”
In the eyes of Somali formula narrative, the anecdotes in counter narrative 1c are part of the national comedy circulated and shared in the local tea houses. They characterize not just the reaction of the owner of the cow but also, by extension, the civility and pacifism of those who do not follow the bellicose nomadic lifestyle and culture as cowards and morally weak and therefore unfit for Somaliness.
The counter narratives, in contrast, interpret the above dialogue of the encounter between the burglar and the owner of the cow from a different ontological framework. At the outset, it attempts to reveal what it perceives as the moral shortcomings of turbulent republicanism by accusing it of inverted morality; a state in which right becomes wrong and vice versa. Based on this ontological assumption, the moral priority narrative asserts that it is, in fact, the burglar and the warlord’s comment in the Rajis narrative mentioned above in counter narrative 1b, and by extension the central principles of the ancestor-based Somali formula narratives, that suffer from lack of the necessary values for the constitution of moral social community.
Beyond the moral priorities, though, the counter formula narratives also construct the social boundary of Somaliness on the basis of territorial priorities. Here, the counter formula narratives shift the categories of social understanding from ancestry to territorial priority. They construct the boundary of Somaliness in terms of who owns which regional territory and how. Specifically, the process of boundary construction used by the counter formula narratives is informed by the degree to which a particular clan owns a collectively acknowledged and clan-delimited ancestral homeland rather than ancestral blood relation through what it refers to as Mudnaan or Asal (indigeneity or priority to ownership of the land) versus Xuraysato (appropriators of the land by means of force). This description is closely related to what Barnes (2006) referred to as U-Dhashey versus Ku-Dhashey. U-Dhashey indicates one as being the son or daughter of the land (also called son of the soil), while Ku-Dhashey means one who is merely born in it but not of it, an immigrant, if you will.
Conclusion
By using Somali formula narratives as a case study, we set out to outline the ways in which narratives construct generic boundary processes that organize the construction of social exclusion and stratification and inequality. The formula narratives divide Somali society into two distinct social strata: (1) the so-called four major clans who claim direct descent from a so-called Somali original ancestor from southern Arabia and (2) those who present no claim of relation to the original ancestor and are therefore permanently removed from the social boundary of Somaliness. This process constructs the social boundary of Somaliness by establishing an important ontological assumption: an original, Muslim, and nonindigenous founding ancestor. The Islamic aspect of the narrative removes the occupational caste communities from the social boundary of Somaliness on the grounds that they allegedly retained pre-Islamic cultural traditions and values. The implied nonindigenous character of the ancestor also removes those allegedly accused of having African-like physical features, whatever that means, from the social boundary of Somaliness.
It is important to note, however, that those removed from the social boundary of Somaliness do not accept the logic of this stratification system and enact their own formula narratives by shifting the category of stratification and exclusion from an ancestor-based narrative to one constructed on either morality or territoriality. The territorial-based counter narratives take advantage of a loophole in the main plot of the Somali formula narrative that the founding Somali ancestor came from outside the Somali region and suggest, if that is the case, then the descendants are not logically the rightful owners of the land and therefore should go back to their ancestral home.
Taken together, our analyses of the ways in which Somali narratives create generic processes that produce social stratification and inequality will further contribute to the recent calls (Chang 2000; Collins 2000; Harris 2001; Loseke 2007; Maines 1993) that narratives are important sites for the social construction of inequality and boundaries of exclusion. More important, though, the results of our analyses have significant implications for Somali political and social elites as well as nongovernmental organizations and other human rights groups working in Somalia and in Somali refugee camps in neighboring countries. This study specifically shows that unlike the egalitarian assumptions prevalent among Somali political elites, scholars, and international organizations, Somali society is organized along caste lines and is fundamentally racialized. Consequently, Somali society and all organizations associated with Somalia and Somalis everywhere must acknowledge this feature of Somali society and organize their understanding of Somalia accordingly. Of course this broader view of a stratified Somali society has also been witnessed by international organizations, which as a result put in place mechanisms to protect Somali minority groups in refugee camps. However, more needs to be done. The Somali social and political elites should, at a minimum, dismantle the 4.5 formula that characterizes the current political power-sharing system. Social historians, social anthropologists, and students of the mythical self-same Somalia should portray Somalia as what it is—a society with levels of social exclusion and othering, rather than the untenable egalitarian Somalia they misleadingly perpetuated. This new perspective should prompt international scholars and organizations to interrogate the hidden stratification and inequalities that underpin Somali social structure.
Footnotes
Acknowledgements
We benefited greatly from comments on earlier drafts by Scott Harris, Donileen Loseke, and David Schweingruber. We are also thankful to the three anonymous reviewers and the editors for their critical and constructive comments, and Renea Miller for helping with the design of the charts.