Part of review forum on “Playing with Fire: Parties and Political Violence in Kenya and India”

When and why do political parties resort to violence? In Playing with Fire, Aditi Malik offers a unique comparative analysis of Kenya and India, where she shows that what drives political violence use by politicians is party instability. Politicians in parties that are unstable have shorter time horizons and thus care less about the potential backlash from voters when they use violence. She builds out an innovative theory that brings together important insights from comparative politics, broadening our understanding of the drivers of political violence across the globe. She tests her theory using rich qualitative data from India and Kenya, carefully tracing the use of violence in both contexts to help readers understand the role of political party instability in fostering episodes of ethnic violence.

The book offers several strengths. First, it draws much-needed attention to the role of parties in understanding when elites use violence. In particular, she offers a supply-side framework that takes parties and their decision-making constraints seriously. She further contributes to our understanding of party violence by moving beyond election violence and building a theory of when parties foment violence more generally.

Secondly, in contemporary comparative politics, there’s been a noticeable shift toward single-country, micro-level studies—a trend I certainly appreciate, given my own work follows this approach. However, this shift has distanced us from the traditional emphasis on multi-country comparative studies. Aditi’s book reconnects us with that valuable tradition and does so exceptionally well. Specifically, one of the strengths of this book was how it not only compared two cases, but provided data and insights at two different levels—national and subnational—within those cases. This shows the breadth and depth of not only the argument, but Aditi’s skills in field research and the commitment to getting the theory and data correct in these complicated contexts. Further, her argument has real world policy implications: strengthening parties and increasing their time horizons may be the best method to reduce political violence.

There were some areas where I wish the book had done more. In particular, I wanted to know more about the menu available to political parties when they are deciding to use violence (or not). What other tools are available to them, and if there are none, is that when they resort to violence? In addition to time horizons, I wanted to understand better why politicians choose violence over, or in addition to, other options, such as electoral manipulation, intimidation, or fraud. Secondly, it would have been interesting to better understand how party instability affects individual politicians, not just parties. For example, in unstable party systems, politicians may switch parties or run as independents: does this affect their individual time horizons apart from the horizons of their parties and therefore their violence use?

I will conclude with a few reflections on how this outstanding book could influence future research. Aditi already offers a test of her argument in Kenya and India, and also Ghana and Turkey, demonstrating the portability of her theory. Future research could take it further by asking how the role of party loyalty, in both stable and unstable party systems, might affect whether politicians expect to be sanctioned for using violence. Loyalty conditions when supporters will use violence on behalf of politicians, but also whether voters will “look the other way” when violence is used. In other words, how does instability interact with voter-level and politician-level attributes to affect violence use?

This is a compelling and thought-provoking book that offers a wealth of meticulously developed data. It will appeal not only to specialists of Kenyan and Indian politics, who will appreciate its perspective on ethnic violence, but also to a wider audience interested in party system stability more broadly.