What is Religious Syncretism?
Jacqueline Suthren Hurst and John Zavos identify that Religious Studies has been criticised and its methodology interrogated due to its ‘lack of focus’. The issue of South Asian religion has been at the forefront of this discussion. It has been argued that the category of Hinduism as a religion cannot adequately make sense of the plethora of ideas and actions that are regularly seen to be encompassed by it. Hinduism, Christianity, Islam and Sikhism are many religious systems that apparently have their own presence in South Asian social and political life. Scholars have recognised that the idea of religion in South Asia is implicated in the development of colonial knowledge, a modern, dynamic category associated with power relations. Oberoi and Ballard have demonstrated that ideas and practice in South Asia frequently cut across boundaries of different religious systems. The categories that we use often do not fit the ethnographical data that is collected, resulting in marginalisation. Syncretism as part of ethnography is a very radical research method based upon collection of fieldwork data, followed by descriptive and analytic conclusion.
Syncretism refers to the synthesis of different religious forms - it is the illusion of two entities joining, or the attempted union of two or more systems of belief. The term itself derives from the ancient Greek prefix ‘syn’ meaning with, and ‘krasis’ meaning mixture, to finally come together in words such as syngkrasis meaning a ‘mixing together’ or ‘compound’, and idiosyngkrasia meaning ‘(peculiar, individual) temperament’.
This essay investigates the problems associated with the application of syncretism, and explore the work of Roy and Assayag to investigate how ethnographers have tried to overcome these problems in the field. I conclude that although ethnography is problematic, we do find some elements of religious syncretism in ethnographic studies. It therefore needs to be used strategically - the ethnographer incorporates into his studies the terms used on the ground, and is aware of his or her own assumptions.
 Suthren Hirst, Jaqueline and Zavos, John, Riding a tiger? South Asia and the problem of ‘religion’, from Contemporary South Asia 14(1) (March, 2005) (Routledge, Taylor and Francis), p.4.
 I recognise, despite using the term Hinduism, that the term itself needs to be acknowledged as a Western construct that failed to recognise diversity within Indian religious traditions. The term was originally developed from ‘sindu’ which was used to denote inhabitants of the region around the Indus who were not Buddhist or Muslim.
 Suthren Hirst, Jaqueline and Zavos, John, (2005), p.5.
 I would regard ethnography as radical do to the presumptions that are made by the ethnographer before he or she enters the field.
 The Oxford English Dictionary defines syncretism as ‘an attempt to sink differences and effect union between sects and philosophical schools’.
 Historically, syncretism has many positive connotations – referring to a strategically practical, morally justified form of political allegiance. However, the sixteenth and seventeenth centuries reversed syncretism’s positive associations through protestant syncretic movements that were accused of the ‘jumbling together’ of religions which attacked doctrine and tradition, and arguably resulted in the ‘confusing’ of religions. Clearly this was a great problem for Christian Orthodoxy. See Shaw, Rosalind and Stewart, Charles, “Introduction: problematising syncretism”, in idem. [eds], Syncretism/Anti Syncretism: The Politics of Religious Synthesis (London: Routledge, 1994)
Amazon Price: $40.00 $16.88 Buy Now
(price as of Aug 25, 2013)
Evaluating Syncretism as an Ethnographical Approach
Tony Stewart outlines the basic model of syncretism as one which,
“assumes that two distinct entities – in these examples, ‘Islam and ‘Hinduism’, as if those were somehow truly monolithic entities – were brought together to form a new construction that shaped parts of both, and could be classified as neither.”
Classically, religious syncretism is the combining of two religions or belief systems into one whole, resulting in a ‘mix’ or ‘hybrid’. Similar terms used are ‘interculturation’ and ‘creolisation’. The basic model is problematic as it assumes at the outset of its own conclusions, by articulating the inappropriate alliance of two things that are in their pure or essential form mutually exclusive and distinct from one another. Furthermore, the unstated object of the model of syncretism is its end product, pointing to some kind of static entity that is inherently unstable. This model projects impure entities that cannot reproduce themselves. It polarises the pure and the impure, and re-enforcing the fact that there is such thing as a ‘true’ form of a religious tradition. As Shaw and Stuart put it,
“This uneasiness about ‘syncretism’ in contemporary anthropology may be due to the term evoking for some of us the existence of a ‘purity’ or ‘authenticity’ in contrast to which it is defined.”
In my opinion, because of this polarisation between the pure and impure, syncretism should be viewed negatively, as it emphasises difference.
According to Van der Veer, syncretism is a descriptive term referring to the borrowing, affirmation, or integration of concepts, symbols or practices of one religion into another by a process of selection and reconciliation. Tony Stewart identifies problems with the term ‘borrowing’, which should be rejected on the grounds that these terms suggest that one groups’ members are not sufficiently independent enough to think for themselves. Also, for Stewartt, the idea of syncretism as a ‘cultural veneer’ implies the overlay of one alien culture on another, for example, an imported Islam overlaid on to a Bengal that is assumed to be Hindu. This wrongly implies a static end result, without recognising a process or dynamic. Furthermore, a ‘veneer’ is impermanent, while the permanent base continues to function as it always had.
Stewart argues that Syncretism “operates on the assumption that the language transparently and faithfully reflects the traditions behind the language, and not just their conceptual structures.” This analysis makes the shift from language to tradition, extrapolating the form of religion from its limited expression in texts. By assuming then, that language reflects the traditions behind it, Stewart is arguing that syncretism works through language, in the way that cultures or traditions modify their language to be more understandable from a different perspective to their own. This makes sense, yet I would argue that syncretism is not only at work through language. In the ethnographical works of Asim Roy and Jackie Assayag, we find differences in language remaining. For example, a demon is referred to as a bhut by Hindus and a djin by Muslims in Shahabandar village (Karnataka), yet both traditions share beliefs in demonology.
In order to assess the alternative proposed models of ‘Islamisation’ and ‘acculturation’, we need to identify what is meant by these terms. Friedmann identifies the ambiguous phrases associated with defining Islamisation, such as ‘they came under submission to Islam’. Contextually, people may have submitted to the Indo-Muslim state as opposed to the faith itself. Yet the regions where the most dramatic Islamisation occurred (i.e. Bengal and West Punjab) lay on the fringes of Muslim rule. Islamisation in India may also be explained in terms of political patronage and caste. ‘Acculturation’ views syncretism as ‘culture-contact’ leading to cultural change, or picking up dominant cultural strands when exposed to them for a long period of time. I shall discuss the further limitations of acculturation, and investigate how Jackie Assayag has tried to overcome them.
Roy provides an analysis of an Islamic syncretistic tradition that arose in Bengal in the nineteenth century. The ‘Islamic syncretistic tradition’ is a descriptive and analytical label for the religious tradition that Muslim cultural mediators consciously reconstructed, with their needs embedded in the cultural and social milieu of Bengal. This Islamic Syncretistic tradition that emerged had pre-occupations with mystic discipline, cosmonogy, eschatology and religious mythology. Although I recognise that Roy’s account is problematic in some areas, he does successfully demonstrate through example, areas of this emerging tradition that could be seen as syncretistic. Mythological accounts combine the characters of the Muslim tradition with the Hindu epics, and equating heroes between traditions, for example, the great war between Ali and Jaykum was compared to those of Ram and the pandavs of the Ramayana and Mahabharata. Circumstances of the construction of this syncretistic model point to an exogenous Islamic tradition based on Persio-Arabic literatures that failed to interact with the world of Bengali speaking Muslims. The ‘cultural mediators’ not only had to pull down the language barricade, but also had to make the Islamic tradition more meaningful to the Bengali converts in syncretistic and symbolic forms.
An emerging problem here is that it is not possible to decipher between cultural and religious phenomena within other societies. By culture we generally mean ‘way of life’. The famous anthropologist Geertz defined a cultural system as a ‘system of symbols’ which ‘establish powerful, pervasive, and long lasting moods and motivations’ and naturalise ‘conceptions of a general order of existence’. I would define religion as an ideological category of belief systems often based on the belief in a God or gods, though not necessarily. Werbner’s criticism of syncretism appears to be extremely helpful in an analysis of Roy’s work on Bengali Islam. Werbner limits the synthetic to religious or ritual phenomena and views religion as a culturally constructed Western category.
“Is attempting to do this full of assumptions about what constitutes ‘religion’ anyway? The fluidity and political contingency of such boundaries as ‘religion’ and ‘culture’ become part of the very subject matter of syncretism, rather than the impediments to it’s study.”
What may appear to be aspects of the Islamic religion may at a later date be seen as cultural phenomena (or indeed vice versa).
The process of acculturation of the Muslim tradition was stimulated by attempts of the Bengali writers to set the characters, situations and stories in the natural geographical, social and cultural settings of the land. For example, the river Nile was introduced as the Ganges and the landscape of Egypt was recreated as a typical Bengali scene with blossoming mango and blackberry trees.
Roy finds almost all facets of a mystical system in Muslim Bengali writings, broadly classified as Mystical speculations (the path, the deity and The Guide) and the practices and techniques of a mystical discipline (The stages and stations and the Psycho-Physiological Culture – the microcosm). Yet Roy rightly admits that developing a syncretistic tradition based on Islamic and Hindu mysticism is problematic due to several unresolved issues. Firstly that an apparent similarity between Hindu and Islamic practices or beliefs may conceal a basic conceptual difference, for example the view of the body as a microcosm in both sufi and yogic traditions Secondly, even when agreement is substantial, the problem of independent origin arises. Did Islam influence Hinduism, or did Hinduism influence Islam? Finally, the interaction may be complicated by a double movement. Original Hindu influences may have passed over into Islam; the movement or process that resulted from this may in turn influence Hinduism (for example in the sufi practice of zikr).
A central issue in syncretistic ethnography is the necessity to acknowledge diversity. Roy states that,
“The classical formulation of these concepts as applied to India, basically enunciated in the context of a “single” great tradition of Hinduism, seems clearly inadequate for our investigation of the Islamic phenomenon of Bengal.”
Roy correctly attempts to acknowledge diversity and regional variation within Islam and Hinduism, identifying the fact that the exogenous Islamic great tradition was unable to interact with endogenous little traditions of the region, which had been culturally continuous with the regional Hindu tradition for centuries. Yet he still refers to Vaishnavism in his chapter on mysticism, without recognising that Vaishnavism itself is a problematic term. Islam in Bengal, according to Roy, presents us with the paradigm of one religion containing two great traditions juxtaposed to each other, one exogenous and classical, the other endogenous and syncretistic. Roy recognises the cultural dynamics of Islam in Bengal, identifying two alternative responses - orthogenetic (the response of the Bengali’s to the challenge of Islam) and the heterogenetic (the classical Islamic response to the same).
Roy concludes that the Islamic syncretistic tradition of Bengal, when viewed in its historical context, is a necessary stage in the process of Islamisation of the country, as integral as the reformist and fundamentalist movements. Furthermore, this dynamic phenomenon raised questions about the meaning of Islamisation. Conversion to Islam that occurred in Bengal was more to do with identity and fellowship than it was to do with spirituality.
The Islamisation taking place was divided into three stages. In the first stage, Islamisation was merely a change of commensal and connubial relations of the converts in a social sense. The second stage saw the emergence of the cultural mediators needed for the breaking down of the dichotomy between the exogenous great tradition and the endogenous little tradition of the converted masses. The syncretistic model held its ground until what Roy defines as the third stage, beginning from the early nineteenth century, when the fundamentalist and the revivalist forces of Islam attacked the syncretistic and acculturated tradition. This forced Bengali Muslims to return the heterogenetic model.
Richard Eaton is also aware of the problems associated with asking the question ‘Why did Hindu’s convert to Islam?’ It presupposes the pre-existence of Indians as Hindu. Furthermore, it implies that there was a motive for conversion, and that it was a deliberate conscious act. He also recognises that models of conversion share a vision of Islam as a monolithic entity. In Roy’s account, the Bengali Muslims were forced by fundamentalists to return to the heterogenic model. Each system holds authenticity for those within it. To claim that a syncretic form is not pure or correct must be avoided. In this sense, to discuss an Islamic or Hindu syncretic tradition is spurious. Ethnographers need to be aware of this, which I think Roy is, though not fully. If Roy has recognised the limitations caused by syncretism, I would argue that he should have used an alternative in his title, such as ‘acculturation’.
Roy was not fully aware of the limitations of acculturation. Shaw and Stewart inform us that,
“What is especially problematic about the concept of ‘acculturation’ is its teleological and quantitative assumptions, such that if a person is placed in a new cultural setting he or she will acculturate progressively proceeding along a continuum towards some ultimate completion. While people do ‘acculturate’ in the sense of picking up a dominant culture if they have to live in the midst of one for any length of time – they must do so, of course, in order to communicate – this does not happen in any necessarily logical, progressive way.” 
Acculturation is clearly not as simple and straightforward a process as Roy suggests. Jackie Assayag, in At the Confluence of two Rivers focuses his study in Karnataka, Southern India, attempted to ‘explain the tenuous but dynamic nature of a composite culture’ of Hinduism and Islam. Unlike Roy, Assayag is keen to avoid the use of the term ‘syncretism’, instead asserting the term ‘eclectic’ when discussing the shrine of the fakir-yogi Rajabag Savar, a Hindu Muslim Shrine in Yamanur village. Assayag tries to understand the Hindus and Muslims of South India through the notion of dynamic acculturation, which takes Roy’s concept of acculturation one stage further. Assayag states that,
“The reason for pointing out thus the changing forms of a culture is to prevent them from being reduced to an organic model in which all cultural manifestations have a fixed and unchanging place.”
Assayag’s recognition of culture and religion as dynamic and unfixed should be applauded, as no religion(s) are ever static entities that remain constant, they are always in flux changing with varying social, economic and political influences. He rightly identifies that use of the term community implies each group has always been cohesive and stable, which are represented by an ideal. This can be demonstrated by the hostility between the Lingayats and Hanabars - it forces us to forget the tactical changes taking place within a social set up that has been created by theses groups jointly, and at the same time we also to ignore the extreme diversity of practices that are likely to express an identity displayed through more or less institutional channels. He also recognises the fact that we should not limit ourselves to textual norms or to the viewpoint of the great traditions.
“Different cultural forms are not watertight compartments, but the products of dynamic processes and mutual adjustment. Thus when it comes to notions that oppose the gross categories Hinduism to Islam, polytheism to monotheism, or even egalitarianism to hierarchy, one must be careful not to project them anachronistically in history, or consider them as values on which contradictory conceptions are founded.”
He avoids the trap of categorisation through world religions much more successfully than Roy. Jackie Suthren Hurst and John Zavos are extremely aware of the pitfalls of the world religion model, which underpins South Asian politics. Not only does it force religion into categories, but the model itself is a Western Construction.
Throughout the second half of the nineteenth century, syncretism was used not only to characterise confusion, but as an imperialist strategy of control. This links in to the argument of Van der Veer, as a Foucoldian view of power struggles – the world religion model can be equated to the power of the West. Van der Veer identifies syncretism to be crucial for the project of the contemporary Indian Nation State, particularly with reference to the aftermath of partition resulting in continuing violence between Hindus and Muslims in many parts of India. The national congress party has tried to contain this violence by projecting a secular ‘multiculturalism’, allowing for the possibility if peaceful co-existence of different ethnic and religious communities under the umbrella of a secular state. Van der Veer goes on to say,
“The notion of the pluralist, tolerant nature of Indian civilisation is not only held by India’s nationalist leaders. It is complemented by the idea that there is a ‘folk culture’ or a ‘popular religion’ in India which is at the grassroots level ‘pluralistic and tolerant’.”
Should pluralism or multiculturalism be more acceptable terms than syncretism or acculturation?
Nandy distinguishes between faith (religion as a way of life) and ideology (religion as a sub-national, national or cross national identifier of populations contesting for or protecting non religious interests). Through the work of Gandhi as a prime example of religious tolerance, Nandy identifies religious violence as the work of fanatic religious and political operators. Yet Nandy’s pluralistic view comes with its own problems. Firstly it denies agency to the people who are involved in the violence. Secondly, the tolerance and pluralistic spirit of India is essentially Hindu, effectively denying that Muslims have a religion which is different to that of the Hindus.
Tony Stewart argues that the biological model of syncretism is probably the most persuasive. Yet when the offspring’s characteristics are identifiable to a parent, the ‘mixture’ metaphor is invoked and the dominant or ‘real’ features come through. If the product is blended in such a way that dominant features from both parents are incorporated, it is a ‘hybrid’ or ‘half breed’. Still Stewart acknowledges that this model, along with many others, adequately characterise the process by which the religious practitioner actually encounters the other and addresses it, leaving those processes to the imagination invoked by the metaphor itself, while focusing on its analysis and metaphoric end-product.
 Stewart, Tony K., ‘In search of Equivalence: Conceiving the Muslim-Hindu Encounter through translation and Theory’, in Easton, Richard, M., [ed.] India’s Islamic Traditions: 711-1750, (New Delhi: Oxford University Press, 2003), p.371.
 Hybridisation implies seeing culture as fragmented
 Culture as a subversive hybrid invention
 A term borrowed from linguistics which currently enjoys ‘favoured concept status’ (Hannerz 1987). For further information and discussion of these terms, see Shaw, Rosalind and Stewart, Charles, “Introduction: problematising syncretism”, in idem. [eds], Syncretism/Anti Syncretism,
 Stewart, Tony K., (2003) p.372.
 Shaw, Rosalind and Stewart, Charles, “Introduction: problematising syncretism”, in idem. [eds], Syncretism/Anti Syncretism, p.2.
 This question of separating the ‘pure’ and the ‘impure’ will be discussed later in the essay with reference to the work of Asim Roy.
 Van der Veer, Peter, ‘Syncretism, multiculturalism and the discourse of tolerance’, in Shaw, Rosalind and Stewart, Charles, [eds], Syncretism/Anti Syncretism (1994), p.196.
 Stewart, Tony, K., (2003), p.372.-373.
 Stewart, Tony, K., (2003), p.373.
 Stewart, Tony, K., (2003) p.371.
 For example, sharia may altered to ‘custom’ or ‘law’ when a Muslim is addressing a non-Muslim.
 Assayag, Jackie, At the Confluence of two Rivers: Muslims and Hindus in South India, (Delhi: Manohar, 2004), p.96.
 Eaton, Richard M (2003), p.15.
 , Rosalind and Stewart, Charles, “Introduction: problematising syncretism”, in idem. [eds], Syncretism/Anti Syncretism, p.5-6.
 Roy, Asim, The Islamic Syncretistic Tradition in Bengal, (New Jersey: Princeton University Press, 1983) p.xii.
 Roy, Asim, (1983), xiii
 Roy, Asim, (1983), p.92.
 Roy, Asim, (1983), p.249.
 I recognise the limitations of this definition of culture, but do not have time to discuss them in this essay. For further discussion, see Mark Hulsether, ‘Religion and Culture’ in Hinnels, John R.[ed.], The Routledge Companion to the Study of Religion, (Routledge: London and New York, 2005), p.489-508.
 Geertz, Clifford, (1973) as quoted in Hinnels, John R.[ed.], The Routledge Companion to the Study of Religion, p.491.
 I distinguish that religion does not necessarily need a God as there is no God in Buddhism, but I would still identify this as a religion. However, I recognise here the difficulties that arise in defining both religion and culture. Anthropologists often confuse religion with ritual. Fitzgerald identifies religion as ‘an ideological category, an aspect of modern Western ideology, with a specific location in history, including the nineteenth centrury period of European colonialisation.’ (2000). Ninian Smart (1973) introduced a dimensional model of religion, which included experience, social, narrative, dogmatic, ethical, ritualistic and materialistic elements. Fitzgerald, however was critical of Smarts account, as to categorise was to impose a Western model. I understand Fitzgeralds criticisms of Smart, but still do think that such categorised are helpful in analysis. For further discussion, see Fitzgerald, Timothy, The Ideology of Religious Studies, (Oxford: Oxford University Press, 2000) and Hinnels, John, R., ‘Why tudy Religions?’ in Hinnels, John, R. [ed.] The Routledge Companion to the Study of Religion, (2005)
 Shaw, Rosalind and Stewart, Charles, “Introduction: problematising syncretism”, in idem. [eds], (1994), p.11.
 Roy, Asim, (1983), p.105.
 By mystical system I mean to refer to a system of thoughts and practices enhancing a profoundly spiritual relationship with the divine, in the case a system grounded in Sufism or ;mystical Islam’.
 Roy, Asim, (1983), p.142.
 Roy, Asim, (1983), p.192.
 Roy, Asim, (1983), p.250.
 Roy, Asim, (1983), p.251.
 In many parts of Bengal, a Hindu was seen as a ‘Bengali’ and a Muslim as a ‘Musalman’, and Bengali Islamic Cultural mediators aimed to close this gap of identity. Use of the term Persian Musalman may be referring to the meaning ‘Muslim on the ground’. Yet it may just be a reference to a different caste group. Many Muslims or ‘Musalmans’ were usually butchers as Sharia law allowed them to do that.
 Roy, Asim, (1983), p.253.
 Eaton, Richard, M., (2003) p.14-27.
 Shaw, Rosalind and Stewart, Charles, “Introduction: problematising syncretism”, in idem. [eds], Syncretism/Anti Syncretism, (1994), p.6.
 Assayag, Jackie, (2004) p.25.
 Assayag, Jackie, (2004), p.131.
 Assayag, Jackie (2004), p.254.
 Both are Hindu groups, the difference is in caste.
 Assayag, Jackie (2004), p.254.
 Not only did Asim Roy limit himself to textual traditions, he limited himself to the writings of one Religious tradition, namely Islam. The lack of Hindu writings on the ‘syncretistic tradition’ or ‘Islamisation’ makes his account extremely one sided.
 Assayag, Jackie, (2004), p.252.
 This is an issue that I shall return to later in this essay.
 Suthren Hurst, Jacqueline and Zavos, John (2005) p.5.
 Shaw, Rosalind and Stewart, Charles, “Introduction: problematising syncretism”, in idem. [eds], Syncretism/Anti Syncretism (1994), p.3.
 An example of the world religion model as a development in colonial power relations is exemplified by the British aim to textualise Sati during the 1820s. For further information on this see Lata Mani, Contentious Traditions, Mani, L., Contentious Traditions: the debate on Sati in Colonial India, (London: University of California Press, 1998)
 Van der Veer, “Syncretism, multiculturalism and the discourse of tolerance” Shaw, Rosalind and Stewart, Charles, [eds], Syncretism/Anti Syncretism (1994), p199-200
 A prime example of political agitation of a religious issue is that of the Babri Masjid of Ayodhya, thought by Hindu’s to be the birthplace of Ram. The campaign launched by Hindu nationalists in 1984 to replace the Mosque with a Hindu temple resulted in many Hindu-Muslim riots.
 The biological model of syncretism is that in which two or more contributing ‘parents’ produce through miscegenation an offspring which cannot be classed with either parent.
 Stewart, Tony K., (2003), p.374.
Amazon Price: $51.95 $36.00 Buy Now
(price as of Aug 25, 2013)
It becomes evident through the scholarly studies investigated that the problems with syncretism lie in its application. Syncretism should be rejected on the grounds that it creates boundaries between the pure and impure, heightening religious tensions. However, I acknowledge that whilst terms such as ‘Hindu’ and ‘Muslim’ are problematic, this dichotomy still exists in reality, as people still use these terms on the ground. In fact, some societies work extremely well with distinct religious and denominational boundaries, for example Kerala’s diverse Christian community.
The terms ‘Islamisation’ and ‘dynamic acculturation’ have been suggested to overcome these problems, yet even these terms to some extent remain problematic. Clearly an alternative to syncretism needs to be found in modern scholarship. We need to recognise a rural-urban difference in the way that these terms are used due to political influences of the ‘bigger-picture’, such as ideas based on the differences between ‘Hindu’s’ and ‘Muslims’ embedded in the history and politics of partition. Yet I would like to retain assumption-less and strategic ethnography on a local scale, in which the ethnographer is aware of his or her own position, and when referring to syncretism does so with reference to the terms used on the ground. As I have chosen to discard the term syncretism due to reasons stated in this essay, I would like to explain why I feel more compelled to use the term cosmopolitanism. Cosmopolitanism catches something of need to ground our sense of mutuality in conditions of mutability or cultural transition. Three closely related forces are at work here; nationalism, globalisation and multiculturalism. As cosmopolitanism is used within urban settings, it could be seen as problematic when one tries to carry out an investigation in Shahabandar village, for example. I would suggest a ‘rural cosmopolitanism’ as a way forward in ethnography for future study.
 For further reading on this see Dempsey, Corine G., Kerala, God’s own country (Oxford: Oxford University Press 2006) available online at www.oxfordscholarship.com
 Whilst I am happy to use this term, I would like to acknowledge that this again is not one without its limitations. For further discussion on this see Benckridge, C. A., Pollock, S., Bhaba, H. K., and Chakrabarty, D., ‘Cosmopolitanisms’ in Breckenridge et al, Cosmopolitanism, (London/Durham. N.C: Duke University Press, 2002)
 Benckridge, C. A., Pollock, S., Bhaba, H. K., and Chakrabarty, D., ‘Cosmopolitanisms’ in Breckenridge et al, (2002), p.4.