Sebastian Elischer makes two contributions in this book. His typology of political parties has the potential to strengthen comparative research on party formation in non-industrial low-income countries. Hitherto, an operational comparative framework has been missing (as has comparative research on political parties in Africa). His book is also a convincing corrective to the prevailing view among some scholars of comparative politics – and among Western observers generally – that ethnicity is the main driving force of politics and party formation across countries in Sub-Saharan Africa, and that ethnicity is always detrimental to democratic development.

The author starts out by designing a typology that can capture the variety of parties across the continent. It draws on the work of Diamond and Gunther, and on a critique of prevailing Eurocentric perspectives on political parties. Parties are classified using specific indicators for their goals, electoral strategy, organisational structure and social base. This leads to five ideal types: mono-ethnic parties seeking to promote the interests of an ethnic group; ethnic-alliance parties promoting diverse group interests; ethnic catch-all parties attempting to transcend ethnic politics; programmatic parties pushing for particular ideas or ideologies; and personalistic parties enhancing the political power of elite members of society.

Next, this typology frames in-depth analyses of party formation in Kenya, Ghana and Namibia between 1989 and 2012, and less detailed analyses in seven additional countries (Benin, Botswana, Burkina Faso, Malawi, Senegal, Tanzania and Zambia). Among Elischer's important findings is that African political party systems are considerably more diverse than is normally assumed. Taking the ten countries together, all five party types have existed at some point in time. Ethnic parties (mono-ethnic and ethnic alliance) have been roughly 25 per cent more prevalent than non-ethnic parties, catch-all, personalistic and programmatic. Personalistic and programmatic parties (which are very few anyway) ‘are not sustainable over time’ (p. 183). Only Kenya and Benin have had ethnic party systems throughout; mixed party systems exist in Malawi, Tanzania and Zambia while non-ethnic party systems prevail in Botswana, Burkina Faso, Ghana (from 2000), Namibia (from 2004) and Senegal. Finally, the in-depth analysis shows that parties and party systems become less ethnic over the course of successive multiparty elections. These either lead to an increase in non-ethnic parties or to enduring peaceful coexistence between ethnic and non-ethnic parties: ‘nothing suggests that multiparty competition fosters ethnic divisions over time’ (p. 221) – except in Kenya.

In a final step Elischer addresses a question that the political science literature has not answered convincingly: why do ethnic parties prevail in some non-industrialised democracies and disappear entirely in others? His evidence shows that this cannot be explained by economic growth at national level, economic inequality at regional level, the electoral system, democratic quality, or party system characteristics – variables regarded as important in the literature. Instead the main causal factor is ethnic structure. Countries without a core ethnic group foster non-ethnic parties when the ethnic fractionalisation index is low, but ethnic parties when that index is high. Countries with a core ethnic group and low ethnic fragmentation appear to foster non-ethnic parties and a low political salience of ethnicity. This surprising result is caused by a ‘bandwagon effect’: initially, a core group is typically divided among several parties but this provides smaller ethnic groups with an incentive to join them rather than to try to beat them. This hypothesis is supported for all ten countries except Zambia and Tanzania.

Two clear strengths of this book have already been mentioned. In addition, the book facilitates a more nuanced understanding of party dynamics on the continent and helps to track transitions among party type over time and across countries. Moreover, Elischer presents very elaborate, detailed and systematic evidence to support his arguments. The usual reliance on electoral and survey data is combined with information about party history and organisational arrangements, composition of party leadership and factions, party manifestos and rhetoric. The author also carefully highlights the limitations of his research arising from lack of information and research on political parties in Africa and from a limited sample of countries. More comparative research on political parties is clearly needed.

Elischer's typology also has weaknesses. Even using a barrage of indicators to classify parties, such decisions sometimes have to rely on debatable quantitative threshold criteria or subjective judgement. For exactly what is an ethnic party? Afrobarometer surveys show that a large proportion of people identify themselves by their occupation rather than by ethnicity in several countries (including Tanzania and Zambia analysed in this book). And what is a programmatic party in an African context? The author only found one example in his sample of ten countries (the National Patriotic Party in Ghana from 1996 to 2000), but more realistic criteria are needed. Programmatic assessment could usefully be widened to include the actual implementation of legislation and policies (not just formal legislation and policy statements, which Elischer focuses on). Parties with programmatic aspirations simply face huge financial and bureaucratic capacity constraints in low-income countries, compared with parties in richer countries.

In the case material there are some examples of parties moving from one party-category to another following the reintroduction of multiparty elections in the 1990s, but it is not convincing to use ethnic structure characteristics to explain such movements because the demographic distribution of groups changes very slowly. There is a need to find additional causal explanations for the transformation of parties over time. Finally, although issued by a reputable publisher, there are several typographical errors in the book and the index is somewhat rudimental. A more disturbing problem is, however, that table information sometimes differs from the text that seeks to explain it.

Nevertheless, Elischer has written a valuable book for students of party development and politics in Africa. It will inspire academic debate for, as the author writes, the ‘essence of political parties is difficult to capture, not only in Africa, but anywhere across the globe’ (p. 191).