Article by Lara Wijesuriya.

What makes a riot?

The dictionary definition is ‘a violent disturbance of the peace by a crowd’[1]. These crowds are usually spontaneously gathered, usually have something in common which becomes the foundation of their riot, and are not organized. For this exhibition, we have included riots which were not directly led by organized groups (whether the government or militant groups). In short, according to our definition, riots are popular disturbances of the peace by crowds. Although the riots may not have been directly led by these groups, many of them have been accused of instigating these riots, and a certain level of premeditation has been noted by eyewitnesses and scholars of the riots. However, the links between organized groups and the rioting mobs is often not directly visible or provable. In some cases, such as the expulsion of the Muslims from Jaffna in 1990, an organized group directly led the violence. Our decision to exclude the  riots and communal conflict which have been proved to be led by these groups is out of a desire to differentiate between short-term, unorganized (or semi-organized) mob violence and longer-term, organized guerrilla action[2].

This exhibition covers the major riots that have occurred in Sri Lanka over the past 140 years. Naturally, this may not be an exhaustive list; but it is safe to say that it includes all the riots which have had far-reaching consequences. Most of these riots have been based on ethnic or religious divides. There are some features that these riots have in common, namely, the ways in which the violence spread, the frameworks in which they were portrayed or understood by the people in power and the newspapers of the time, and the people who made up the violent mobs.





The usual way in which mob violence and rioting spread is by word of mouth, or rumour. Drawing on the rumours that spread after an earthquake in India in 1934, J. Prasad compiled this list of conditions necessary for rumours to flourish: a condition that (a) sets up an emotional disturbance; (b) is of an uncommon and unfamiliar type; (c) contains many aspects unknown to the individuals affected; (d) contains several unverifiable factors; (e) is of group interest[3]. Of these, all except (b) are relevant to riots. In the absence or tardiness of official news, rumours are more likely to be believed. Countless examples from the ten riots covered in this exhibition prove this again and again. Although potentially dangerous at the best of times, in riot situations, rumours are dynamite.

There are certain rumours which are more explosive than others. One is rumours concerning disrespect to religion- for example, the rumour about blasphemous images being carried by the Buddhist procession during the 1883 riots; rumours that clergy had been attacked or killed (1883, 1956, 1977); and rumours that places of worship had been defiled (2014).

Rumours of rape and sexual violence are also capable of causing serious reactions. These rumours are very widespread, and have been documented with regard to almost all the riots covered in this exhibition. This includes both rape and forced sterilization (2019).

The final form of rumour is of interest today because it has played a great role in the more recent riots; that of poison. Some poisoning rumours involve ‘infertility pills’ (2018), and others well-poisoning (2019).

These rumours in particular are effective in transmitting violence because they strike at the heart of communities. In our society, religion plays a major role in many peoples’ lives. Religious and ethnic identities are often closer to people than their national identity; which makes a perceived attack on their religion an attack on their identified community and thus on themselves. The rumours of rape and sexual violence consist an attack on ethnicity- the concept of racial purity is tarnished by rape or other sexual violence, and this type of rumour usually arouses anger not because of the violent act itself but because of the perceived contamination of the race. This type of rumour was very visible in 2019, when many Sinhala women accused a Muslim doctor of forcibly sterilizing them.

The spread of these rumours used to depend on a chain of people who believed the message enough to remember (or misremember) it and repeat it to others; with the advent of social media the process of rumour-mongering has become much smoother and faster- and unbelievably more explosive. While the violence of 1915 spread fairly slowly along the railway lines as travellers carried the news and rumours, and rioting in 1983 sprang up in different towns according to how soon people travelled there from Colombo; the 2018 riots are at least partially attributable to rumours spread on social media. From being merely an effect and instrument in spreading violence, rumour has become a major cause.


Framing Riots

Like rumours, unusual phenomena are often understood in familiar terms. Because of this, the way riots are understood depends on the political, social and economic climate of the time. In looking at the riots covered here, a wide gap is noticeable between the discussions about riots that occurred during colonialism and those that occurred post-independence. While the riots themselves don’t seem to have changed much, the ways in which they are framed does. The Times of Ceylon, reporting on the 1883 Riots, wrote ‘It is nothing short of disgraceful that in Colombo we should have to call on the military to quell a religious riot of this sort.’ In discussing the riots of 1915, the Governor, Robert Chalmers, said that ‘Is there… a single man who does not today deplore the tragic folly which we have witnessed? Is there a single man who does not fear that the hands of the clock may be set back, and that the march of progress may be arrested in the island?’[4] The chief theme in these two quotes is that of progress. The Times deplores the fact that even in (progressive and developed) Colombo the military should have to quell religious riots; and Chalmers is even more direct. Both look forward in time, and regard communal and religious riots as ‘backward’.

There is a marked contrast to this in the reactions to post-independence riots. Debates in Parliament tend to hark back to ‘2500 years of ethnic harmony’ and how ‘in the long history of this island…we have faced many grave issues…’[5] At first glance, this disparity in the framing of riots seems unusual. Surely looking forward should be the more ‘modern’ response? The reasons behind this change in attitude are possibly both the British mandate of the ‘white man’s burden’ (to bring their subjects into the light of progress)[6] and the concept of nation- the imagination of a community that is perceived as being ancient while being objectively modern[7].


Who Riots?

For quite some time, people involved in violent mobs were regarded as coming from the ‘criminal classes’[8]. These classes were probably the urban poor or working classes, who were regarded as having a higher tendency to crime than other classes. This would nicely explain the extreme violence and lack of empathy shown by rioting mobs. But is this actually the case? Kumari Jayawardene disagrees. She argues that, in the case of the 1915 Sinhala-Muslim Riots, rioting was carried out by skilled workers who used the religious/ethnic-based riots to express their dissatisfaction with working conditions[9]. Talking about the 1958 riots, N.M. Perera argued the same point in Parliament. He spoke of the rioters as ‘people roused by the basest passions but who, under normal circumstances…are decent citizens’[10]. This conception of violent mobs is, admittedly, worrying. It is pleasanter to imagine that horrible acts of violence that occur in any riot are done by psychopaths who are already criminals. Knowing that ordinary people are capable of such bestiality should cause us to re-examine ourselves. Would I, in that situation, act in that way? Is Freud correct when he suggests that these new (extra-violent) characteristics displayed by mobs are not in fact new but merely already-existing individual characteristics that are usually repressed[11]?

This exhibition is designed to present a linear narrative of these ten riots in an (as far as possible) unbiased way. We also hope that viewers will reflect on these riots, on the causes and effects of such communal violence, and be more aware of what leads to such senseless violence. 



[1] Oxford Dictionary of Current English, 3rd ed. (Oxford, Oxford University Press, 2001)

[2] It’s arguable that guerrilla actions are also the will of the people (Guevara, Ernesto. ‘Guerrilla Warfare: A Method’, 1963), but guerrilla groups are almost by definition organized with a clear long-term goal; rioting mobs, on the other hand, tend to lack a clear-cut purpose and organizational structure.

[3] Prasad, J. ‘The Psychology of Rumour: A Study Relating to the Great Indian Earthquake of 1934’, Journal of Psychology, 1935.

[4] Speech to Legislative Council, 6th August 1915, Hansard 1913-1916, Sri Lanka National Archives.

[5] Speech in Parliament, 4th June 1958, Parliamentary Session Book 1958, Sri Lanka National Archives.

[6] Cain, Peter J. ‘Character, ‘Ordered Liberty’, and the Mission to Civilise: British Moral Justification of Empire, 1870–1914’, Journal of Imperial and Commonwealth History, 2012.

[7] Anderson, Benedict Imagined Communities, revised ed. (London: Verso, 2006)

[8] Jayawardena, Kumari. ‘Economic and Political Factors in the 1915 Riots’, Journal for Asian Studies, 1970.

[9] Jayawardena, 1970.

[10] Speech in Parliament, 4th June 1958, Parliamentary Session Book 1958, Sri Lanka National Archives.

[11] Freud, Sigmund. Group Psychology and the Analysis of the Ego, trans. James Strachey. (London: The Hogarth Press, 1949)