The Vicious Circle that is Mauritian Politics: The Legacy of Mauritius’s Electoral Boundaries

Abstract

In the lead up to independence in Mauritius two commissions were appointed by the British colonial authorities to determine the type of electoral system that would be used. The choices were between proportional representation (PR) and plurality voting, and larger versus smaller constituencies. PR was rejected as it was feared it would harden communal divisions and smaller constituencies were chosen to ensure adequate representation of the society’s different ethnic groups. A review of the 11 general elections that have taken place between 1967 and 2014 shows that plurality voting and smaller constituencies have had the very same effect that was imputed to PR. With the exception of the 1982 general election, government has always been formed by a Hindu-dominated political party and voting seems to be conducted primarily along ethnic lines. The paper argues that this level of predictability in voting patterns is due to the legacy of how Mauritius’s electoral boundaries were delimited and suggests change will only come once they are reformed.

Acknowledgements

We would like to thank the editors and anonymous reviewers for their assistance and suggested improvements to the paper. Also, the paper would not have been possible without the assistance of the following students who helped with the collection of some of the primary data: Dhirish Krishna Raghoo; Raveena Gunesee; Gabriella Batour; Mohini Jugessur; Mevina Ramasamy; Ashil Mohit; Marie Sandy Jennifer Comte. And lastly we would like to thank the political agents we interviewed for the paper but who asked to remain anonymous.

Notes

1 For ease of reference we use the dates of the reports—rather than their duration—to refer to the electoral commissions throughout this paper.

2 Evidence of this claim is usually illustrated by recourse to the practice of Hindu-dominated political parties awarding key political posts to ethnic minorities.

3 There are a number of reasons to question the transparency of Mathur’s work. Apart from those already mentioned, his claim that Hindus would be prepared to vote in a non-Hindu Prime Minister is highly questionable (Mathur et al., 1999, p. 99). It is well-known in Mauritius that all Prime Ministers have been Hindus, only come from the Vaish caste, and from the two family dynasties of the Ramgoolam and Jugnauth families. The only exception was when Paul Bérenger served as Prime Minister for a short spell from 2003 to 2005, owing to a pre-election arrangement with the Mouvement Militant Mauricien’s (MMM) coalition partner the Mouvement Socialist Mauricien (MSM). Equally concerning is the fact that prominent political theorists of plural societies, such as Arend Lijphart (2012) and Crawford Young (1994, 1999), have tended to rely on Mathur’s work for their assessments of the workings of democracy in Mauritius (cf. also Premdas, 2001).

4 These were considered to be the main ethnic groupings for the purposes of the 1958 and 1966 electoral commissions and their estimates were based on the 1952 and 1962 censuses respectively. Chinese Mauritians were included but were not given the same weightage as Hindus, the General Population and Muslims due to their smaller numbers. It is these very same ethnic ‘groupings’ that have been enshrined in the nation’s constitution.

5 The argument put forward by one of the LP representatives, Dr. Millien, during the first session of the Trustam-Eve public hearings (dated 12 September 1957) implied that the electoral system in place between 1948 and 1957 had allowed a number of Franco-Mauritian sugar estate owners to secure seats by bribing or intimidating voters in rural constituencies where they were an ethnic minority. The same charge was repeated by Abdool Razack Mohamed in the Banwell Commission public hearings, but without any supporting evidence.

6 The other important party to these discussions, the Comite d’Action Musulman (CAM), led by Abdool Razack Mohamed, advocated reserved seats and separate rolls.

7 See the fourth session of the Trustam-Eve Commission public hearing dated 16 September 1957.

8 The LP, PM and CAM were joined by two additional political parties at the Banwell Commission public hearings: the Independence Forward Block (IFB) and the All Mauritian Hindu Congress (AMHC). CAM repeated its call for reserved seats and communal rolls, while the IFB and the AMHC (as allied parties) called for 13 multi-member constituencies but opposed PR and block voting.

9 This paper will only refer to the 20 electoral constituencies of Mauritius and not Rodrigues because the electoral constituency of Rodrigues has had relatively little impact on the formation of coalition governments throughout Mauritius’s post-independence period.

10 The LP was founded by Dr Maurice Curé in 1936. Curé was a coloured Mauritian and like other famous Creole leaders of the LP such as Emmanuel Anquetil, fought for many years to organize and represent the interests of both Indian sugar estate labourers and Creole dock workers. Before joining the LP, Ramgoolam was already a member of parliament and apparently showed little interest at first in fighting for worker’s rights. Ramgoolam would eventually take over the leadership of the LP following the death of Guy Rozemont in 1956. Simmons—upon whom the aforementioned analysis is based—quotes a speech delivered in 1940 by Curé in which he is said to have warned that one day the party would be taken over by Indian intellectuals who would turn it into a party for Indians rather than labourers (1982, p. 77). Curé was eventually be expelled from the LP in 1967.

11 According to Simmons (1982, pp. 23, 152–157), the increasing communalization of party politics that became evident in Mauritius with the approach of independence was also reflected in Creole and coloured supporters shifting their allegiance from the LP to the PM at the beginning of the 1960s.

12 Downloaded http://electoral.govmu.org/English/electionresult/nasselec/Pages/default.aspx on 9 April 2019.

13 The Mauritian newspapers we consulted included: L’Express, Le Mauricien, Le Cernéen, Mauritius Times, Advance, Le Dimanche, Week-End, Le Défi Quotidien, Le Défi Plus and Star.

14 The subject as to whether Hindus as a proportion of the population are in decline is often a point of discussion among Mauritians, it being thought that this might affect their dominance of politics. Evidence of this is said to be reflected in the greater numbers of Hindus that are purportedly converting to Christianity, or the extent to which Hindu sub-ethnic groups are asserting their political autonomy from the dominance of the Northern Hindu majority. But the former trend has never been substantiated statistically, nor is there any reason to believe that efforts at differentiation among Hindus are incompatible with the forging of broader political alliances during general elections that can be mutually beneficial to these groups and the Hindu-dominated parties they choose to align with.

15 I.e. the Parti Socialiste Mauricien. This political party was a dissident group created by Harish Boodhoo which broke away from the Labour Party due in part to accusations of corruption within the party.

16 The 1982 general election result has been interpreted by a number of political observers as marking a short-lived shift from communal to class politics in Mauritius’s post-independence era (Bowman, 1991, p. 81; Dubey, 1997, pp. 217–219; Mukonoweshuro, 1991, p. 218; Selvon, 2005, pp. 388–425; Srebrnik, 2002, p. 280). But as many of the same observers have noted, in order to remain politically relevant, the MMM were forced to tone down their socialist rhetoric and increasingly had to play the game of ethnic politics. Some have even suggested the reason the MMM were able to win the 1982 general election was because they appointed Anerood Jugnauth (a Hindu) as Prime Minister and relied on the PSM to secure victory in the rural constituencies (Bowman, 1991, pp. 79–80; Dubey, 1997, pp. 133, 192; Mukonoweshuro, 1991, pp. 214, 218; Selvon, 2005, pp. 458, 469).

17 Significantly whenever the LP and the MSM have formed a pre-electoral alliance, as happened in 1983, 1987 and 2010, they were able to form government.

18 See Kasenally and Kadima (2005) for a more thorough treatment of the history of coalition governments in Mauritius since 1967. 1976 was the only year when a coalition government was formed after the outcome of an election and not before.

19 This is an important aspect of Mauritian politics that is rarely commented upon by foreign observers but noted in the work of Mauritian scholars seemingly more familiar with the inner workings of the society, particularly its ethnic relations (Jahangeer-Chojoo, 2010, pp. 125–127, 130; Kasenally & Kadima, 2005, p. 155; Nave, 2001, p. 100; Sithanen, 2003, pp. 5, 10).

20 There are five major caste groups in Mauritius, which are not identical to India’s caste system. They include the low caste groups the Ravived and Rajputs, the middle caste group the Vaish, and the high caste groups the Babujee and Maraz. However, these caste groups are only recognized by Northern Hindus and a majority of Mauritians are not familiar with their names, even though they can have a significant influence on politics (for further reading see Benedict, 1967; Claveyrolas, 2015; Hollup, 1994). There are four Hindu sub-ethnic groups in Mauritius. These include the majority Northern Hindus as well as the minority groups the Tamils, Marathis and Telugus who are sometimes referred to as the non-Hindi speaking groups. The Hindu sub-ethnic groups are differentiated by a number of factors including language, region of origin in India and cultural customs including perceived religious differences (for further reading see Couacaud, 2016; Dinan, 1986; Eisenlohr, 2006; Hollup, 1994). The category of Gujarati is the outlier in the list of categories mentioned in Figure 5. It refers to Muslims of non-indentured background who came to Mauritius as traders (for further reading see Hollup, 1996; Jahangeer-Chojoo, 2010).

21 Mauritius’s small size as an island and in terms of population number obviously makes this task more feasible in comparison to say a large, populous, continental country for example.

22 The agents may help the main political parties determine candidates with the right ethnic profile for a constituency, but as the agents repeatedly stressed in our interviews, the decision always rests with the party leadership. For a more thorough treatment of the organization of Mauritius’s main political parties, see Kasenally and Bunwaree (2005).

23 We only sought to speak to agents of the main political parties as minor political parties do not seem to engage in the practices associated with scientific communalism. Accordingly, we interviewed seven different political agents in constituencies 9 and 16. In constituency 16 we interviewed agents from the four main political parties, but in constituency 9 we could only interview agents from the LP, MSM and MMM, as the PMSD had no discernible presence, which appears to tie in with what we said earlier about their poor record in rural constituencies.

24 Mumbai is the point of embarkation in India from which the majority of Marathi immigrants to Mauritius are said to have left from.

25 A LP agent we spoke to in constituency 9 put it this way: ‘It is very difficult to change this system … Everything will depend on the new generations and [we] hope that they won’t vote based on caste and religion. They should vote with values and intelligence and not like the elder generation. However, people will always vote according to their caste because the parties itself put their candidates according to this system’. An MMM agent from the same constituency said: ‘The demand for nominating a candidate from the Hindu community of Vaishya, Rajput and Ravived castes comes from the people  … This practice has always existed. It has become a culture and difficult to change. All the political parties follow the same trend of nominating candidates on ethnic grounds as it is in their advantage’. While a LP agent in constituency 16 told us the following: ‘The main problem in Mauritius is that election is highly ethnicized  … People are more likely to vote for a candidate from the same community and caste  … Now candidates may not declare their community, but the electorates will inquire about the candidates’ community and caste because people tend to vote for candidates from the same ethnic group’.

26 For more in-depth studies of Mauritius’s socio-cultural associations and their place in politics see Hollup (1994, 1996) and Eisenlohr (2006).

27 The Electoral Boundary Commission is mandated to review the boundaries of Mauritius’s electoral constituencies every 10 years. It has released four reports in 1976, 1986, 1999 and 2009. Only two of these reports and their recommendations were passed by parliament in 1986 and 1999 and the changes were minimal owing to the fact that a 75% majority is required in parliament. Consequently, gerrymandering has never become as a source of political debate in Mauritius although it was a frequent accusation at the time the 40 single and 20 3-member electoral constituencies were being delimited during pre-independence negotiations.