The Newsletter 73 Spring 2016

Sri Mariamman worship in the Gulf of Thailand

William Noseworthy

<p>Sri Mariamman (alt.: Mariyamman) is generally understood to be a Tamil goddess associated with smallpox. She was in fact cursed with the disease in at least one version of her narrative, and so offerings to the goddess during puja are meant to ‘keep her away from your door’. Historically, her festival directly preceded the ‘hot season’ in Tamil Nadu, which coincides with the season of smallpox and other plagues. In Southeast Asia she was transformed, into an urban goddess of cholera prevention, as well as a ‘goddess of the soil’ for the Tamil diaspora.&nbsp;</p>
<p style=\"font-weight: normal!important; font-size:0.9em;\">Photo above: Sri Mariamman temple in Singapore &copy; William Noseworthy.</p>

Most scholars would agree that Sri Mariamman worship has its origins in śaktī worship, originating in southern India. In this context the goddess herself is generally clothed in green, has a pale complex and holds a dagger, which Younger associated with Kālī imagery.1 Younger, P. 1980. “A Temple Festival of Mariyamman”, Journal of the American Academy of Religion 48(4):493-517; see pp.504-505. . In one story, Paraśurāma ‘switched’ the heads of Mariamman and an ‘untouchable’ (Harijan; Dalit) maid. When the heads were switched, the Brahmanic and non-Brahmanic caste imagery became blended, as did the imagery of ‘right’ and ‘left’ handed castes.2 Younger 1980, pp.507-508; Younger uses the term ‘Harijan’ despite the fact that this term was rejected by many South Indian activists in favor of the term ‘Dalit’.  Regardless of the origins of the goddess, the picture that a historicization of records associated with Sri Mariamman worship paints is one where new diseases become associated with her figure, throughout the emergent Tamil diaspora in Southeast Asia in the 19th and 20th centuries.

First, there was already a smallpox vaccine in Southeast Asia that was being used in the 18th century. Although it was initially ineffective, we might rightly conclude that the utility of worshipping a goddess associated with smallpox prevention declined as the disease itself was effectively stamped out. However, mortality rates for smallpox itself remained high into the 19th century. There were still many outbreaks, with mortality rates exceeding 25%, although these outbreaks were increasingly in rural areas. Second, as new urban centers expanded, and disease climates shifted, it became possible to associate the goddess with diseases that appeared to afflict individuals in similar ways to smallpox, to have similar mortality rates, or similar causes. In Penang, for example, a cholera epidemic broke out in 1900 and 1911, with the 1900 epidemic resulting in a quarantine of over 2,000 individuals in a small depot that had a planned capacity of only 800.3 Arasaratnam, S. 1970. Indians in Malaysia and Singapore, Kuala Lumpur: Oxford University Press, p.53.

We should not take accounts as those above as evidence to portray Southeast Asia as ‘inherently diseased’ the way that some popular portrayals and scholarly works have done. To nuance this perspective, one could also suggest that the European fascination with Thai medical standards is not a ‘new’ phenomena. A farang visitor to Bangkok in the 1830s remarked at the ‘surprising cleanliness’ of the city.4 Sternstein, L. 1984. “The Growth of the Population of the World’s Pre-Eminent ‘Primate City’: Bangkok at Its Bicentenary”, Journal of Southeast Asian Studies 15(1):43-68; see p.60-62. . Still, smallpox was the number one complaint in the city for that decade; while cholera, smallpox and ‘the plague’ remained the most common deadly diseases in the city throughout the 1840s. Then, even though the cholera vaccine was instituted quite effectively in the 19th century, compared to other Asian and many European centers, outbreaks of cholera remained common in Bangkok until the vaccine was made mandatory after the 1911-1912 epidemic. Nevertheless, a trans-regional outbreak that centered on Malay ports in southern India ‘halted’ migration from South to Southeast Asia in 1919. In other words, although smallpox became less of a concern, cholera remained a concern for medical officials into the 20th century. This helps to explain Aiyappan’s  assertion that the worship of the goddess in Cochin State (southern India) was related to the prevention of both smallpox and cholera, as well as Mialaret’s assertion that Sri Mariamman, during the middle of the 20th century, had become associated with the prevention of smallpox, cholera, and even chickenpox.5 Aiyappan, A. 1931. “Myth of the Origin of Smallpox”, Folklore 42(3):291-293; Mialaret, J.P. 1969. Hinduism in Singapore, Singapore: Asia Pacific Press.

Indianization and localization

From her origins in Tamil Nadu in the first millennia, as a goddess associated with smallpox, to a goddess who became popular throughout Southeast Asia, it is clear that certain local conditions impacted Sri Mariamman worship throughout the epochs. Scholars of South and Southeast Asia tend to examine these trends on spectrums of ‘localization’ and ‘Hinduization,'6 The most influential early work on Southeast Asian process of localization may well be: Mus, P. 1933. “L’Inde vue de L’Est. Cultes indiens et indigenes au Champa” (Seeing India from the East: Indian and Indigenous Cults in Champa), BEFEO XXXIII, no 1. pp. 367-410. While the most influential early work on Hinduization may well be: Coedes, G. 1944. Histoire ancienne des États hindouises d’Extrême-Orient (Ancient History of the Hinduized States of the Far East), Paris: EFEO. although I would assert that we should continue to include discourses of ‘Indianization'7 Many scholars have pointed out that although Coedes’ original 1944 work was translated into English as the “Indianized states of Southeast Asia”, that the original title bore with it no such presumptions about any form of such a unified or ‘nationalized’ vision of the sub-continent. However, in my usage ‘Indianization’ refers to distinctly 20th century processes. See also: Mabbet, I. 1977. “The ‘Indianization’ of Southeast Asia: Reflections on the historical sources”, Journal of Southeast Asian Studies 8(2):157-158.  and ‘Re-Sanskritization’ within this discussion as well. Here, for simplicities sake, ‘Indianization’ refers to the cultural influences of the greater subcontinent cultural zone (including what is now Pakistan, Nepal, Bangladesh and India) on Southeast Asia. To presume the subcontinent as one cultural zone is problematic, and ‘sub-continentization’ may be more accurate, although an even more clunky neologism.

By contrast ‘localization’ tends to refer to any active Southeast Asian adaptation of ‘Indian culture’. For example, the Balinese goddess Dewi Majapahit is clearly a local adaptation of Sri Mariamman who brings ‘the gift of the gods’ – smallpox – to the Balinese. Local treatment of the disease includes lifting the afflicted – usually children – above the shoulders of those in a procession devoted to Sri Majapahit, and addressing ‘the children’ – or the afflicted – with the title ‘God’ (Dewa) or ‘Goddess’ (Dewi) for nine days, as a means of appeasing the goddess. Since Majapahit is also the name of the classical Javanese empire that ruled much of insular Southeast Asia, from the 13th through the 15th century, we may hypothesize that this particular localization involved a cheeky response to the Javanese conquest of Bali, under the auspices of Majapahit’s expansion and conquest of Bali in the 1340’s.

It is not clear if other highly localized forms of Sri Mariamman appeared in Southeast Asia throughout the epochs, although it is clear that Sri Mariamman worshippers were not set aside from the other major trends that influenced religious people throughout the region. For example, as modernist-traditionalist debates swept throughout both Buddhist and Muslim communities around the Gulf of Thailand, during the early part of the 20th century, it does not seem that Tamils worshipping Sri Mariamman were exempted from these questions, as a full report (dated 6/1/1924) of events at the Kuala Lumpur temple indicates that positions of those in the temple in such debates were aligned between those who supported the senior panchayat (‘senior priest’), and a ‘populist’ group that sought to make an appearance in front of the ‘electoral board’ (the urar).8 Arasaratnam 1970, pp.164-165. However, due to the lack of investigation by existing studies, into a broader ‘regionalist’ frame of Sri Mariamman temples in the Gulf of Thailand, we have little further knowledge about the particular policies under debate. Meanwhile, there is ample evidence to support the ‘Hinduization’, ‘Sanskritization’, and ‘Re-Indianization’ of Sri Mariamman temples in the later part of the 20th century.

For one, Sinha, who has published two studies of the Sri Mariamman temple in Singapore,9 Sinha, V. 2010. “Unravelling ‘Singaporean Hinduism’: Seeing Pluralism within”, International Journal of Hindu Studies 14(2/3):253-279; Sinha, V. 2009. “‘Mixing and Matching’: The Shape of Everyday Hindu Religiosity in Singapore”, Asian Journal of Social Science 37:83-106.  points to the particular ways that new worship of Sri Mariamman in Singapore’s cosmopolitan atmosphere is tied to a broader trend of Hindu revivalism. These trends, however, are not precisely ‘new’ but rather ‘ongoing’, with a burst of enthusiasm that re-emerged during the past decade. In the 1960s, for example, in addition to the Thai Poosam ceremony, the Sri Mariamman temple boards ensured that the Mariamman Thiruvali ceremony, specific to the veneration of Sri Mariamman herself, was held at “all state temples” in Malaysia and Singapore. An example of ‘Sanskritization’ occurred each year, according to Mialaret’s account,10 Mialaret 1969, p.22.  as during the ceremony of Thimithi (roughly October) – in Singapore at the very least, and perhaps at other peninsular temples – the Mahabaratha is recited in Sanskrit; although at least one individual remains employed to act as a live translator to translate the epic into Tamil. In more contemporary times, the Singapore temple has become even more an example of ‘cosmopolitan syncretism’. Sinha, for example, describes the ‘Hinduization’ of Chinese deities such as Guan Yin and ‘the Laughing Buddha’, as an expression of “everyday religiosity in practice”, identifying specific sites, refreshing historical memory, and adopting other ubiquitously Daoist deities. However, inasmuch as the particular examples of Guan Yin (a Chinese, gender shifted, adaptation of the classical Boddhisattva, Avalokitsvara) and ‘the Laughing Buddha’ (Pu Tai; a manifestation of a historical event wherein many Chinese became convinced that a monk was indeed the ‘next Buddha of our kalpa’, Meitreya) are Chinese adaptations of Buddhist figures, these adaptations represent a ‘Re-Indianization’ of Meitreya and Avalokitsvara, which also involves their re-incorporation into the accepted pantheon of South Indian Hinduism.

Tamil-Gulf of Thailand, and global networks

The Sri Mariamman temple in Singapore is just one example of the ‘new trends’ that are associated with the worship of what we may now better term, ‘the goddess of urban disease’. Sri Mariamman has been historically important outside of Southeast Asia, in other South Indian diasporic communities, in South Africa, Guyana and the Caribbean. In South India itself, Sri Mariamman was less and less attended as a ‘village goddess’, and more and more associated with ‘urban elites’. For some, this was because ‘the goddess’ as a figure brought common perceptions of “sanctity and prosperity” in the face of rapidly increasing class tensions and “rapid social change”.11 Waghorne, J. 2004. Diaspora of the Gods: Modern Hindu Temples in an Urban Middle-Class World, New York: Oxford University Press, p. 133.  The prosperity helped the central Sri Mariamman temple in Tamil Nadu become one of the three richest temples in India. Now nationalized as ‘Indian’ and hence less regionalized as ‘Tamil’, Sri Mariamman temples were built in London’s Tooting neighborhood [1996], Soneleigh [Sri Rajarajeswary Amman], and in the United States. However, the earlier mention of the ‘modernist-traditionalist’ debates that circulated the greater Gulf of Thailand and Indian Ocean networks throughout the 20th century, are now found in reference to these neighborhoods as well. “As recently as 1997”, Waghorne recorded that she was “told by an educated [individual] well settled in Britain that the less educated were, of course, continuing to build temples to Kali, Durga, and other such goddesses”.12 Waghorne 2004, p.173.  The implication is two-fold: 1) that education eliminates the need for the construction of temples, and 2) that Sri Mariamman is understood in a direct parallel to consorts of Siva, because of the ‘hot’ goddess connection.13 TIn Hindu cosmology gods and goddesses can be either ‘cool,’ protective, creative deities, or ‘hot’ destructive, violent deities. However, the association of a deity with either ‘cool’ or ‘hot’ properties does not necessarily preclude that they never embody the qualities of the dual-opposite.  The point adds weight to the most bold claim of this short research piece: that the exact trends that Younger and Waghorne have found as the ‘urbanization’ and the ‘Brahmanization’ (so to speak) of the goddess Sri Mariamman, have their roots in the Tamil diasporic urban enclaves of late 19th and early 20th century Southeast Asia.

The building of Sri Mariamman temples in Southeast Asia is linked directly to the spread of migrant Tamil urban enclaves. Furthermore, there is one family, the Pillai, who appear to be responsible for strongly encouraging the spread of the goddess. The 19th century ‘Gulf of Thailand revival’ begins with the Georgetown shrine in 1801. The Pillai then sent one of their sons, Naraina Pillai with Raffles to Singapore in 1819. There were no other Tamils present in Singapore at the time. Four years later, however, Naraina was responsible for organizing a land grant, donating the land to the Sri Mariamman temple board, which completed the Singapore temple just four years later, with thanks to predominantly convict labor. The Pillai were then responsible for upgrading the original shrine in Penang in 1833 [renovated in 1933]. The Pillai lineage in Singapore, however, continued to grow as well. Born in Singapore in 1850, Thamboosamy Pillai [Singlish alt. sp.: Pillay], became one of the founding members of the Kuala Lumpur Sri Mariamman temple board in 1873. At the same time, the Pillai also reshaped Tamil migration for British Malaya. Recruiting in South India was only done through two firms, and one of them was the ‘Ganapathi Pillai and Co.’ firm. In other words, the spread of Sri Mariamman worship into Southeast Asian urban enclaves was intimately tied to the desires of one of the, if not the most elite Tamil clans. It follows that the subsequent construction of Sri Mariamman temples in Bangkok [1879], in Saigon [1870s-1890s] and Medan (Sumatra, Indonesia) [1881] were all intimately tied to the spread of urbanized migrant labor and capitalizing cosmopolitan Tamil elites.14 Singh-Sandhu, K. 1969. Indians in Malaysa: Some Aspects of their Immigration and Settlement (1786-1957), Cambridge: Cambridge University Press, pp. 80, 139, 178; Mialaret 1969, p. 40

Although Tamil migration was initially encouraged by British policies, explicitly to counter balance the power of Hokkien, Hakka and other overseas Chinese in the Straits Settlements, encouraged migrations would wane in the 1920s. At the same time, nationalization and ‘Indianization’ ran high as anti-Imperialist sentiments mounted. There were, however, a few explicitly Tamil responses of the day. Tamilaham was a Tamil weekly that was published out of Kuala Lumpur. The serial actively discouraged Tamils from taking positions on British rubber plantations, and emphasized above all else the consolidation of Tamil (not Indian) identity in the straits. As an explicitly Tamil goddess, therefore, Mariamman continued to gain followers, with temples being constructed in Singapore at 126A Mandai Road [1947, by families of Nee Soon village], at Dunearn Road at the 6th milestone of Bukit Timah [1954, for the employees of the Turf Club] and on Coronation Road, as well as a secondary shrine at the Subramanian temple of Singapore Naval Base [shortly after 1966]. The placement of the Sri Mariamman temples initially followed what Younger highlighted as a practice of establishing ‘village shrines’. Younger draws from an earlier Anthropologist, Becker, who claimed that the ‘village shrines’ were indicative of the non-Brahmanical ‘right hand’ Sudra-landowning caste practice, of blessing new land with new temples.15 Mialaret 1969, p.45; Younger 1980, pp.498-99.  In other words, through worship in Southeast Asia, the goddess Sri Mariamman underwent another transformation, from a ‘hot’ goddess to a ‘goddess of the soil’.

Image removed.

Sri Mariamman’s own transformation has included an alteration in the worshipping community and of practices in urban Southeast Asia. The emphasis on the willingness of worshippers to undergo extreme pain appears less common in the temples in Bangkok and Ho Chi Minh City, for example, strongly contrasting with observations regarding Tamil celebrations, particularly centered around the Thai Poosam festival. Devotees may insert hooks into their body, suspending weights from them, connected to wires. These ceremonies remain common throughout the temples of Peninsular Malaysia and Singapore, as well as to a lesser extent, Bangkok. Thai Poosam is a much more modest occasion in contemporary Ho Chi Minh City. Furthermore, the Bangkok temple has been nationalized through the inclusion of adorning decorations that feature the Thai royal colors and, yes, statues of Buddha and Thai style mini-Brahmanical shrines. On any given day, the worshippers at the Thai temple are more of a mix between Thais and Tamils. In Ho Chi Minh City, all of the worshippers surveyed over the course of several visits during three years were either Vietnamese or Khmer Krom. Spirit possession is occasional, but rare.

In the case of Vietnam, the temple is furthermore not examined, generally, as an ‘active’ site, but rather as an example of the city’s ‘colonial past’ – of a once present Hindu population that has now virtually disappeared. It is, in other words, commonly recast through that Orientalist lens that viewed the Hindu towers dotting the Vietnamese coast as purely antiquities. Contrast this with the local revivals of Sri Mariamman worship in Singapore in the 2000s, recorded by Sinha, and the ‘nationalization’ of Sri Mariamman temples in the Gulf of Thailand in Southeast Asia becomes even clearer. While the temples in Singapore and Malaysia are ever present examples of one of the four acceptable forms of diversity (Tamil/Indian; Malay/Muslim; Mandarin/Chinese; and English speakers), the temples in Vietnam and Thailand have been incorporated into Vietnamese goddess worship (Thanh Mau) and a particularly Thai strand of Brahmanism, respectively.


In summary, the narrative of the spread of Sri Mariamman temples in Southeast Asia and the Gulf of Thailand challenges existing notions about the dynamics of mono-directional cultural influences extending between South India and Southeast Asia. The dynamics of Sri Mariamman as an emergent urban, elite, goddess, rather than a ‘village goddess’ were already present in Tamil urban communities in Southeast Asia in the 19th century. The association of Sri Mariamman with these communities further shifted the properties of the goddess. Originally associated with the prevention of smallpox, in the 19th century the goddess became associated with the prevention of cholera as well. By the 20th century, chickenpox was added to the list. Occasionally Southeast Asians have worshipped potentially localized forms of Sri Mariamman, such as in the case of the Balinese goddess Sri Majapahit; or have adapted, variously toned down, and nationalized practices of Sri Mariamman worship, such as at the temples in Bangkok, Ho Chi Minh City and Singapore. Southeast Asia has also been a site where Hindu temples have been able to ‘(re)Indianize’, ‘(re)Sanskritize’, or ‘(re)Hinduize’ Chinese Buddhist deities, just as well as presentations of Tamil ceremonies, all of which are conformed to ideas about what it means to be ‘Tamil’ and ‘Hindu’. In this way, the case of Sri Mariamman worship in Southeast Asia in the 19th and 20th centuries, virtually turns the meaning of ‘diaspora’ on its head. From many identities and locations in the sub-continent, the Tamil diaspora emerged into the singular network in the greater Gulf of Thailand area, centered on the worship of a goddess who might have just been able to fend off the most feared urban diseases of the day.

William B Noseworthy, University of Wisconsin-Madison (