In India, Hindu–Muslim violence is commonly referred to as ‘communal’ violence, with the term ‘ethnic’ reserved for racially and linguistically distinct groups. I prefer ‘ethnic’ for two reasons. First, following Gupta (2002) and Varshney (2002), conflict among Indian castes is properly communal because the opponent's national identity is not questioned; upper-caste Hindus accept backward-caste Hindus as legitimate members of the nation. ‘Ethnic’ mobilization, by contrast, reasserts national identity and defines the opponent as foreign; Hindus thus associate Muslims with Pakistan. I describe Hindu– Muslim conflict as ethnic or religious, as distinct from conflict among castes. Second, I follow Weber's classical definition of ethnic groups:

The book, therefore, makes use of the terms ‘ethnic’ and ‘religious’ interchangeably. It follows then that the label ‘ethnic violence’ is applicable to acts of collective violence, in which the ethnic difference of the parties involved is ‘integral rather than incidental to the violence … (wherein) violence is coded as having been meaningfully oriented in some way to the different ethnicity of the target’ (Brubaker and Laitin, 1998).

Some clarity on the labelling of violence is also important. Characteristic to violence in India, the state camouflages rationally motivated acts of violence against minority groups as spontaneous outbursts of emotions, notably anger, giving these acts a legitimate form of political expression (Brass, 2003; Hansen, 2008; Spodek, 2011; Tambiah, 1986). One may even compare these legitimized acts to informal justice systems in apartheid South Africa (Knox, 2001). The terminology is crucial here: organized violence is camouflaged as spontaneous, or a clash between antagonistic civilian groups whose members are overwhelmed by emotions. Would it then be appropriate to define such camouflaged yet planned attacks as riots?