With the advent of decolonisation, the greatest challenge for African leaders was in transferring their people’s tribal loyalties to the state, which would henceforward serve as a multi-ethnic unifying factor (Keller 2014, p. 4) . From the moment they became independent, many African states have experienced challenges in defining and managing their political identities. Kenya and South Africa are cases in point.
In the wake of recent cases of post-election violence, discussion of Kenyan politics is often carried out alongside that of ethnicity. This is because ethnicity influences political formations, competition and how public goods and national resources are allocated. Shilaho (2018, p. 2)  contends that since independence in 1963, members of the president’s tribe had disproportionately dominated the civil service. This is a view shared by Ajulu (2002, pp. 259-265)  who highlights the alternation of Agĩkũyũ and Kalenjin dominance of the civil service to date. However, indications are that in the period shortly before and after independence, there were instances when politics was shaped less by ethnicity and more by the leadership qualities of the politicians seeking election. A notable case was the 1961 triumph of Tom Mboya (a Luo) over Munyua Waiyaki (an Agĩkũyũ) in the race to represent the then Nairobi East constituency in the Legislative Council. This is in spite of the fact that the Agĩkũyũ constituted 64% of the electorate in the aforementioned constituency (Chege 1981, p. 76) . That said, since the legalisation of multipartyism in 1991 it has been noted that political leaders across the political divide have formed political parties and campaigned on the strength of ethnicity (Shilaho 2018, p. 2) .
Because of the legacies of the long periods of colonialism and Apartheid, South Africans have, in their political expressions, continued to struggle to make sense of who they are in relation to each other (Frueh 2003, p. 39) . As is the case with Kenya, discussions about politics in South Africa have included the relationship between ethnic groupings. However, the discourse on race relations has remained prominent because of its highly fractious nature. Guelke (cited in Frueh 2003 , p. 48) insists that the concept of race is used to explain many issues in South Africa including electoral politics and the debates over economic transformation. In other words, any social problem that is commonly discussed in South Africa always boils down to the dynamics of historical and contemporary race relations.
In a bid to uproot or balance the endemic inequalities sown by Apartheid, political debates especially among political parties have focused on the need for equitable transformation of the South African society in order to reflect the rainbow nation1  that it is. Political parties are the premier entities that front people’s representatives to the legislatures. Their composition and viewpoints are therefore microcosms of those of the country’s population and this can be examined from two points. Firstly, there are parties mainly supported by white South Africans and others by black South Africans with other groups spreading their support across the political divide. Secondly (using a left-right dimension), there are parties that are revisionist (left-wing) in their calls for economic transformation, those that are centrist, yet others that are right-wing in their postures (Rohanlall 2014, p. 136) . A look at the party representation in parliament reveals that race still plays a strong role in politics with most white South African politicians elected via centrist parties and the of majority black South African politicians via left wing parties (Parliament of the Republic of South Africa 2017) .
Group identity influences politics in Kenya and South Africa. With that in mind, this article assesses the linkages between ethnic and race relations on one side and on the other side, the activities of political parties in both countries. This is against the backdrop of the fact that in both countries, political parties: 1) are the most effective means of political socialisation, 2) provide avenues for pursuit of individual political aspirations, and 3) avail a collective of individuals tasked with legislating frameworks for social order (Gauja 2016, p. 44) .
This analysis has been developed from primary and secondary sources that document party politics and identity issues in Kenya and South Africa. Primary sources included questionnaires, blogs and interviews2. Primary data collection purposively targeted individuals with specialist and/or experiential knowledge of the workings of political parties in Kenya and South Africa. The targeted individuals were also those that were familiar with the triggers of political violence in the two countries. The interviews used open-ended questions designed to receive large amounts of content while providing the freedom for the respondents to express their substantive grasp of the subject matter. Closed-ended questions were also posed.
In order to obtain the specific themes focused upon in this article, the author used the process of concatenation in the review of similar studies. This was complemented by a trends analysis. This article discusses ethnicity and race relations through the lenses of: 1) legal frameworks governing race/ethnic dynamics in relation to political parties, 2) political party identity, 3) the strategies that political parties use in seeking electoral support and 4) the policies of political parties and how they allocate public goods and national resources. The analysis sets the stage for lesson learning from both countries.
3. The Concepts of Ethnicity and Race
The author considered three schools of thought when making distinctions between ethnicity and race. The first school belongs to those (such as Christmann 2009 , p. 546) who suggest that all humanity belongs to one race also known as the human race. Secondly, there are others such as Fenton (1999, p. 62)  as well as Ritter and Hoffman (2010, p. 7) . The latter make the distinction between the two concepts: ethnicity denotes the cultural factors that are used to classify a people and includes nationality, regional norms, ancestry as well as language, whereas race refers to the physical attributes that are used to cluster individuals e.g. bone structure, skin, hair, or eye colour. Thirdly, there are those who acknowledge the two terms but identify factual (Appiah 1998)  and epistemic (Zaibert & Milán-Zaibert 2000)  challenges with how they are used distinctly. This third group offers alternative ways of looking at race and ethnicity distinctly or as the same thing depending on the circumstance or context of their application .3
For this article, the author used the second school of thought, especially in the analysis of field data. This is because the respondents in South Africa emphasised the (physical) differences between black, white, coloured and Indian South Africans when discussing socio-economic imbalances. This emphasis of the differences is due to the history and legacy of Apartheid laws. In addition, respondents in both countries consistently pointed out in their responses the cultural and linguistic differences between black Africans.
4. Legal Frameworks Governing Ethnic/Race Dynamics in Relation to Political Parties
In Kenya and South Africa, policymakers in both countries have over the years established legal frameworks targeting different sectors affected by the ethnicity/race relations debates, including political parties.
The constitution of Kenya prohibits the formation of political parties based purely on ethnicity (Constitution of Kenya 2010, art. 91 (2)) . In order to ensure practicality in the implementation of that constitutional provision, the Political Parties Act (2011, sec. 7)  obliges each political party to satisfy the Registrar of Political Parties that it has a following of at least 1000 voters from each of more than half of the 47 counties in Kenya. Failure to adhere to these provisions may lead to deregistration of a party.
Kenya’s policy frameworks and statutes have identified ethnic consciousness as a structural cause of many conflicts. This is because most Kenyans perceive many processes of allocating public goods and national resources to be focused on how different ethnic groups will benefit or lose (Shilaho 2018, p. 33) . The enactment of laws regulating political parties has therefore taken into account how the formation and management of political parties could contribute to the ethnicisation of politics. The fears over ethnicisation of politics are informed by ethnic clashes that have been a fixture in every general election since 1992 (Cowen & Kanyinga 2002 , p. 154; Rawlence, Albin-Lackey & Neistat 2008 , p. 19).
4.2. South Africa
The legal frameworks that govern ethnic and race issues in relation to the management of political parties in South Africa may be inferred to have their rationale in opposing the manner in which Apartheid compartmentalised society. Notably, the Apartheid laws identified South Africa as made up of whites (also called European), Coloureds4 , natives (also African or Bantu), or Asians (also Indian) (Population Registration Act No. 30 of 1950, sec. 1) . Until 1985, South Africans were also prohibited from forming political parties that had mixed membership from the aforementioned groups (Prohibition of Political Interference Act No. 51 of 1968, sec. 2) .
The laws that were enacted after the end of Apartheid therefore reflected the desire by lawmakers to break away from the past and encourage multi-ethnic as well as multi-racial political parties. To this end, political parties that seek to participate in elections can now be registered by the Chief Electoral Officer based on ethnicity, religion, regionalism, tribalism or advocacy of secession from the Republic (Electoral Institute for Sustainable Democracy in Africa 2006, para. 4) . Nonetheless, a party may be denied registration if among other reasons, it incites violence or hatred, or causes offence/denies one admission on the grounds of race and ethnicity (Electoral Commission Act 51 of 1996, sec. 16 (1)) .
It should however be noted that the registration of political parties in South Africa is done by the Independent Electoral Commission (IEC) only when the parties seek to participate in elections. A party may therefore exist without being registered as long as it does not participate in elections. This stems from the fact that political parties are viewed as associations or voluntary corporations. This means that it is the constitution of the respective political parties that govern how the parties relate with their members as well as how the members themselves relate. Be that as it may, there are laws that give some indication of the latitude allowed to the constitutions of political parties. These laws oblige political parties to act fairly and without any ethnic or racial discrimination towards party members or those that seek to join it (Constitution of the Republic of South Africa Act 1997, sec. 9 (5) ; Promotion of Equality and Prevention of Unfair Discrimination Act 4 of 2000, sec. 7-8) .
5. Political Party Identity
In the process of seeking to control and allocate public goods and national resources, organised groups in democratic societies use political parties as a fulcrum of identity politics. Alumona and Azom (2017, p. 297)  note that this is because of ideological differences that lead to the binding of like-minded individuals through subscription to the tenets of respective political parties that they identify with. Political parties therefore provide institutional structures for identity politics. They add that, in Africa, there are other factors that make people identify with political parties and these include, among other things, ethnicity and race. As examined below, the question of who joins or lends support to a political party is one that contributes to the definition of the party’s ethnic/race identity in Kenya and South Africa.
5.1. Membership Base and Party Control
Upon formation, political parties in Kenya and South Africa have often appealed to a majority of certain demographics of the society who then became members. In addition, the identities of the leadership of political parties might influence the power dynamics in terms of how different demographics in a party relate with each other.
The legal frameworks earlier alluded to have remained largely academic because the nature and operations of political parties reflect the ethno-regional calculations of politicians. Apart from that, the formation and management of political parties since the reintroduction of multipartyism in 1991 has been driven by the “charisma and deep pockets” of the party luminaries (Respondent Ten 2018) . This lends credence to assertions that members of political parties are not recruited based on their support for manifestos presenting parties’ fundamental principles and policies for the country (Mueller 2008, p. 200) .
When forming political parties, politicians often do so while banking on the belief that their ethnic compatriots would support such a party. Such support is then used to negotiate coalitions with other political parties. To this end, it is not uncommon to hear that the Orange Democratic Movement (ODM) is a party of the Luo ethnic group whereas the Jubilee Party is associated with the Agĩkũyũ based on the fact that they are led by Raila Odinga (a Luo) and Uhuru Kenyatta (an Agĩkũyũ) respectively (Respondent Nine 2018 ; Respondent Ten 2018) . It is no wonder that some political parties are synonymous with the ethnic groups or regions associated with their leaders (Ahere 2018, p. 32) . The general perception is that if members of one party violently clash with those of another party as happened in the post-election violence in 2008, then it is tantamount to one community fighting another. However, Respondent Eleven (2018)  does not think that political parties “… go [outright] to say that this ethnic group will be the base of this political party” but concludes that “it happens by default … you find that the parties are dominated by a few individuals who might know each other or who might be coming from a specific ethnic group.” He attributes this situation to the general lack of trust among Kenyans when it comes to political issues and the high likelihood of people speaking the same language trusting each other more and coalescing around one political party.
The ethnic nature of the aforementioned parties’ membership influences the choice of party leaders. It is rare that a party whose membership is dominated by one ethnic group would have among its higher echelons, someone from another ethnic group. When that has happened5 , it has been because the “party owners”6  installed such officials ostensibly to give the parties a more heterogeneous outlook hence hoping to garner more votes in elections. Respondent Ten (2018)  contends that if the members of the parties were allowed to elect top officials, they would likely choose their ethnic compatriots to lead the parties. These practises have had adverse impacts on internal party democracy to the extent that Respondent Ten (2018)  refers to them as a situation in which “… we have political party leaders choosing the members [and] not the members choosing the leaders”.
5.1.2. South Africa
Whereas in Kenya the debate on ethnicity is more prominent when it comes to who controls political parties and how their membership bases are defined, South Africa presents a contrasting view. In South Africa, debates about party membership bases are dominated by racial differences as is revealed by examination of some of the main political parties, defined here as those that performed well in recent elections. Respondent Three (2017)  posits that one of the most notable debates on race relations in South Africa is the postulation that the white and coloured South African voting blocs mainly support the Democratic Alliance (DA) even as the African National Congress (ANC) is perceived as having most of its support from black South Africans. He adds that Indians are perceived to mainly support the Minority Front party. It is, however, noteworthy that the DA has in recent years made efforts to repackage itself as a party that celebrates the country’s ethno-racial inclusivity (Southern 2011, p. 296) . In a bid to attract support from more black South Africans, the party allocated more senior positions to black South Africans and followed through by electing Mmusi Maimane as the party’s first black African Federal Leader in May 2015. The aforementioned dominance of the debate on race has, however, not suppressed that of ethnicity.
Among black South Africans, ethnicity has played a role in the perceptions of the membership base and the control of some political parties such as the ANC, the Inkatha Freedom Party (IFP) and the National Freedom Party (NFP). For example, Respondent Seven (2017)  notes that during the presidency of Thabo Mbeki (a Xhosa), it was common to hear the term Xhosa Nostra in many hushed debates within the ANC. The term meant that the Xhosa ethnic group were calling the shots in the party. While acknowledging the prominence of luminaries like Nelson Mandela, Oliver Tambo, Govan Mbeki, Thabo Mbeki, and Walter Sisulu (all Xhosas) in one era of the ANC’s leadership, Frankental and Sichone (2005, p. 278)  contend that Xhosa Nostra is not a term that is a true representation of the distribution of political power in post-Apartheid South Africa. The rationale of their contention is that in the provinces that the ANC (just like the other parties) has the right to choose a Premier based on its election performance, the Premiers chosen have been those whose ethnic backgrounds tie them closely to their respective provinces.
This debate is based on the notion that the largest ethnic groupings―Xhosa and Zulu―have, during different periods, been perceived as solid bases of some political parties. During the struggle against Apartheid, the Xhosa were associated with the ANC whereas the Zulu with the IFP (Frankental & Sichone 2005, p. 210) . Respondent Seven (2017)  hypothesises that it is during the presidency of Jacob Zuma (a Zulu) that KwaZulu-Natal, a province dominated by the Zulu, became a stronghold of the ANC at the expense of the IFP.
Piombo (2009, p. 34)  presents another perspective on ethnicity and political parties in South Africa. She holds that the strategies that opposition parties have used to mobilise supporters have either avoided narrow ethnic appeals or shied away from focusing on any social cleavages that would lead to ethnic politics and/or violence. In essence, whereas the DA has obtained support based on the overlap of race and class divisions, other parties such as the IFP have made attempts to move beyond a purely ethnic appeal (Piombo 2009, p. 33) .
5.2. Electoral Performance
The perception of a party’s ethnic and/or race identity may or may not determine how the party performs during elections. This is especially relevant in this era where populism, social media, opinion polls and/or “Fake News” have been used to project the popularity (or lack of it) of some political parties that end up performing dismally or securing landslide victories contrary to pre-election expectations. An examination of recent electoral performance of the main political parties in Kenya and South Africa reveals some useful insights.
While there seems to be general consensus that ethnicity is an endemic determining factor of electoral outcomes in Kenya, the reality is that no political party can win an election if different ethnic groups vote strictly as blocs and only support candidates from their ethnic groups (Long & Gibson 2015, p. 1) . Leaders of parties must therefore form alliances with other parties whose leaders represent other ethnic groups and campaign across the country outside of their ethnic bases. This is because the most coveted prize in an election in Kenya―the presidency―has a constitutional threshold that requires a candidate to not only garner more than 50% of the votes cast nationally, but also obtain more than 25% of the votes cast in each of more than half of the 47 counties in the country (Constitution of Kenya 2010, art. 138 (4)) . This is against the backdrop of the fact that no ethnic group constitutes more that 23% of the population of Kenya (Nyambura 2017, para. 15) .
A 2008 study found out that even though Kenyans resist defining themselves exclusively on ethnic terms, they will, to a larger extent, be ethnically conscious when they vote. This is a defensive act due to mutual mistrust and fears that other ethnic groups will organise themselves along ethnic lines (Bratton & Kimenyi 2008, p. 287) . The situation is compounded by the zero-sum nature of the Kenyan political system, which has in the past seen communities in power (symbolised by their ethnic kin who won elections) marginalise and persecute those that lost.
Table 1 lists the voting patterns for the counties in which there were more than 400,000 votes cast, and the performance of the top two candidates who vied for the presidency in the 2017 general elections. Uhuru Kenyatta was fronted by the Jubilee Party7 whereas Raila Odinga by the National Super Alliance (NASA)8. Each candidate obtained the most votes in the counties where they or their running mates have strong ethnic support, or due to their association with coalition partners with ethnic linkages in certain counties. This is in light of the fact that the delineation of counties is almost matched with the locations in which individual ethnic groups reside or originate. The only exception is Nairobi, which is cosmopolitan, and where each of the candidates garnered almost half of the votes cast.
5.2.2. South Africa
Many studies argue that the outcomes of post-Apartheid South African elections may be referred to as “racial censuses” due to the fact that, overall, a vast majority of white South Africans vote for one set of parties while black South Africans vote for another set of parties (Ferree 2010, p. 1) . Nonetheless, bearing in mind the arguments of Piombo (2009)  in section 5.1.2 of this article, there is a contention that the constituency that tends to vote for the ANC are the poor uneducated black South Africans whereas the more educated and privileged black South Africans tend to vote for white minority parties such as the DA (Anyangwe 2012, p. 55) . This means that beyond race, there are other identities such as class that influence election outcomes, especially among black South Africans who constitute approximately 75% of the electorate (Nhlapo, Anderson & Wentzel 2017, p. 3) .
During the 2016 municipal elections, the ANC, DA and the Economic Freedom Fighters (EFF) emerged as the biggest winners (or losers in some cases). In Table 2 there is a summary each of the aforementioned parties’ overall performance in each of the nine provinces. IFP has been included in KwaZulu-Natal due to its performance and historical dominance in the province.
The aforementioned results indicate that the DA obtained double-digit percentages of the vote in all but one province. Given that white South Africans constitute approximately 9% of the population (Nhlapo, Anderson & Wentzel 2017, p. 3) , the aforementioned results infer that the DA received a significant number of votes from non-white South Africans. This lends credence to the argument of Piombo (2009, p. 34)  that the DA’s support comes from an overlap of race and class divisions in the South African society. The results also back the intimations of Anyangwe (2012, p. 55)  that the ANC’s support among black South Africans comes from those who are uneducated, most of whom who live in rural areas.
Table 1. Voting patterns during the 2017 general elections in Kenya.
Source: Independent Electoral and Boundaries Commission (2017) .
Table 2. Performance of South African parties in the 2016 municipal elections.
Source: Electoral Commission of South Africa (2016) .
Indeed, outside of the Western Cape Province and the big cities, the ANC received most of the votes in rural areas. As surmised by Table 3, the ANC performed dismally (in comparison to past elections when it won by majority) in five of the eight major cities. Nevertheless, the ANC remains the biggest party in every municipality in Free State, Mpumalanga, Limpopo, North West, and the Northern Cape provinces; winning some of the municipalities with huge margins (Roodt 2016, para. 6) . Some opinions in the media have argued that South African voters have started to look beyond race in deciding the party to vote for, as exemplified by poor black South Africans’ support for the DA in Nelson Mandela Bay (Brock 2016, para. 2) .
Table 3. Performance of South African parties in the metropolitan municipalities during the 2016 municipal elections.
Source: Electoral Commission of South Africa (2016) .
It is noteworthy that the DA performed spectacularly well in Western Cape where black South Africans constitute just 33% of the population in the province (Nhlapo, Anderson & Wentzel 2017, p. 5) . This supports the beliefs of Respondent Three (2017)  that most white and coloured South Africans vote for the DA. However, the claims of Respondent Three (2017)  that most Indians support the Minority Front party are not backed by the numbers in Table 2 and Table 3 especially going by the Minority Front’s poor performance in KwaZulu-Natal where Indians constitute 7% of the province’s population (Frith 2017, para. 3) . The DA was most likely the biggest beneficiary of the Indian voting bloc judging by the party’s ability in the past to secure at least 70% of the votes in the Indian-dominated suburbs of Chatsworth and Phoenix while securing at least 61% of the Indian vote nationally (Mkhize 2016, para. 3) .
6. Strategies of Seeking and Consolidating Electoral Support
Having examined the support base and electoral performance of political parties in Kenya and South Africa, this section analyses the approaches that parties have used in order to secure and retain votes from various demographic groupings in both countries.
6.1. Geographical Focus in Kenya
The electoral boundaries that define the counties and constituencies of Kenya coincide with those that define the geographical locations in which ethnic groups reside or originate. Banseka (2005, p. 206)  traces this coincidence to the British colonial system of divide and rule, wherein they kept different ethnic groups separate from each other and administered them separately. The defining point was in 1962 when the Regional Boundaries Commission divided Kenya on the basis of either ethnic homogeneity (one tribe per district) or compatibility where there was more than one tribe per district or province where they were deemed happy to coexist (Makulilo et al. 2016, p. 71) .
After independence in 1963, subsequent governments built upon the British colonial administrative boundaries to strengthen and hang on to political power. A case in point was during the de jure one-party system era when President Daniel arap Moi created more constituencies in the districts where he had strong ethnic affiliations by virtue of his own ethnicity or those of his closest political allies (Cowen & Kanyinga 2002, p. 131) . This gave his Kenya African National Union (KANU) party an advantage in the 1992 and 1997 elections because there were more parliamentary seats in the districts where the party had massive support. This legacy has influenced the regional approach of campaigns by political parties.
Some political parties have adopted militancy in the guarding of regions that they consider their strongholds. They have done this through inciting their followers against other parties or mobilising quasi-militia groups to unleash violence on supporters of other parties. In past elections it was therefore not uncommon to hear of groups such as Mungiki, Taliban, Jeshi la Mzee, Baghdad Boys, and the Kosovo Boys, among others, who were used by politicians and their parties to antagonise rival parties and their supporters (Gecaga 2007, p. 80) . Respondent Eleven (2018)  points out that for most political parties, violence has become a “currency” of negotiation and it escalates because “… when it happens and my party is a beneficiary I want it to continue …” This has led to the association of electoral cycles with vicious cycles of political violence.
Due to ethno-regional factors that make voters in some regions hostile to some politicians they consider outsiders and the militant protection of political turfs by some parties, political party leaders vying for the presidency have been somewhat restricted from campaigning freely in all regions of Kenya as much as they might want to. For instance, before the 2017 general elections Uhuru Kenyatta of Jubilee Party organised lesser political rallies in NASA strongholds of Luo-Nyanza in comparison to areas where his party was perceived to have massive support. This is despite the fact that he was running for office as an incumbent and could therefore use state security machinery to assure his security by preventing disruptions of his rallies. On the same token, Raila Odinga of NASA also held fewer campaign rallies in Jubilee Party strongholds of Central Kenya.
The selective nature of how political parties have campaigned with focus on certain regions is also strategic. Some regions have more registered voters than others and political parties consider them as crucial to their victories (Wanga 2017, para. 7) . The parties therefore direct more resources to campaigns in such areas at the expense of the regions with fewer registered voters.
6.2. The Design of Campaign Messaging in South Africa
In recent elections, political parties in South Africa have adopted strategic campaign messages to woo new voters while seeking to consolidate the votes of their traditional supporters. Three main factors have influenced this approach in recent elections. Firstly, South African political parties have a history of being driven by ideology unlike their counterparts in Kenya that are driven mainly by personalities. Secondly, young black African voters have emerged who are more cognisant of the (non) performance of political parties while in power hence interested in the betterment of their lives as opposed to the focus on the history of the struggle against Apartheid (Graham et al. 2017, para. 22) . Lastly, the burgeoning of information and communications technology has provided political parties with more innovative platforms to reach many more potential voters than before.
The rural black South African voters’ bloc, however, remains huge and crucial to any electoral victory by any political party. Notwithstanding recent socio-economic challenges faced by the country, rural voters cherish the freedoms they are now accustomed to and which they continue to associate with the political organisations that fought for the end of Apartheid. Some political parties have sought to ride on the fears of the country reverting to the past (Apartheid)10  and packaged their messages accordingly. The ANC for example once warned its rural supporters in Seshego, a large township in Limpopo Province, that if they didn’t vote (for the ANC) in the elections “the Boers (DA) will come back to control us” (Johnson 2015, p. 143) .
Some parties, including the ANC, have packaged their campaign messages with the aim of sustaining racial polarisation since it gives them an advantage in winning the rural black South African vote. The leadership of the DA has also enabled this polarisation strategy to be feasible due to comments attributed to its leaders and which Respondent One (2017)  considers racist11 .
There are parties (e.g. the ANC and COPE12) that have used their liberation credentials to woo rural black South African voters. As a result, Macdonald (2015, p. 173)  insists, “the DA and other parties that lack ties to the anti-apartheid movement [continuously] strive to break down South Africa’s racial polarisation in order to win a majority.” The latter’s campaign messaging has therefore sought to link the ANC with political malfeasances that have stalked the government, especially during the presidency of Jacob Zuma. Troy Martens, a former Head of Communications of the ANC Women’s League, however, contends that the DA does not compete with the “ANC on an ideological and political level,” and holds that it is “a party that stands for nothing except attacking the ANC” (Martens 2012, para. 3) . This is a view that is close to those of Ngalo et al. (2018)  who wonder how the DA will define itself if the ANC does not have a scandal.
South Africa’s challenges in dealing with massive unemployment, socioeconomic inequalities, corruption, dysfunctional service delivery and crime have offered some parties the opportunities to develop radical campaign messages. The most notable of such parties is the EFF, which proposes radical socioeconomic transformation, expropriation of land without compensation and the nationalisation of mines (Beresford 2014 , para. 13; Mamdani 2019, para. 20 ).
7. Party Policies and Allocation of Public Goods and National Resources
The strategies that political parties in Kenya and South Africa use to seek and consolidate their electoral support are also linked with the social and political discourses around implementation of policies that affect the allocation of public goods and national resources. In other words, political parties package their policy proposals and manifestos based on the pronounced social problems that exist in society at a particular time. In both countries ethnicity and race feature overtly or covertly in the disparities between policies sold to the electorate during campaigns for votes and the implementation of the policies upon winning elections.
7.1. Inclusivity in Governance in Kenya
In Kenya, the conflicts arising from recent elections have brought to the fore the question of inclusivity in governance. The ethnic groups that are associated with the political parties that fail to win the presidency in the general elections often feel, rightly so, that they are marginalised in the allocation of state resources during the term of government of the winning party.
In an op-ed published during the tumultuous weeks immediately after the 2017 general elections in Kenya, Makau Mutua argues that the never-ending “political circus” in the country is all about exclusion and ethnic contempt. He infers that there are ethnic groups and the political parties that they are affiliated with, which “believe Kenya belongs to them and the rest of us are nothing but peons” (Mutua 2017, para. 5) .
The aforementioned sentiments help to underscore how political parties in Kenya implement policies as soon as they ascend to power and the extent to which such actions allow other political parties (read: ethnic groups) to have a sense of belonging or not. For instance, upon winning the 2013 elections, the Jubilee Alliance13  filled most key positions in government with Agĩkũyũ and Kalenjin individuals in a country that has more than 40 ethnic groups (Shilaho 2016, p. 117) . This kind of action implemented by different regimes has often disillusioned many Kenyans, especially the youth, who periodically became conflicted about their patriotism and apathetic about politics. Wa-Mungai (2007, p. 47)  reminds us that at one point this type of situations led to the popular 2006 satirical phrase “navumilia kuwa Mkenya”, which is Kiswahili for “I just tolerate being a Kenyan”. This phrase represented the pessimism that a section of Kenyans felt about the poor quality of their lives and the limited opportunities they had in the wake of the government’s rallying patriotic call “najivunia kuwa Mkenya” (Kiswahili for “I am proud to be Kenyan”). After the disputed 2017 general elections, the feelings of exclusion led the NASA coalition leaders to trigger debate among its supporters about the feasibility of secession from the Republic, alleging that most Kenyans felt marginalised by the Jubilee Party government which catered for the Agĩkũyũ, the Kalenjin and a few other communities (Barasa et al. 2017, para. 2) .
Other commentators have taken note of how in the execution of the principal function of monopolising violence, respective governments have ethnically profiled citizens under the guise of maintaining law and order. An example could suffice. During public protests after the 2017 general elections, supporters of NASA and Jubilee Party were treated differently. While NASA supporters bore the brunt of excessive use of force by the police, mass protests by perceived Jubilee Party supporters did not receive similar attention from the police even when they turned violent or caused public anxiety (Gitonga 2017 , para. 4; Kenny & Ahere 2017 , para. 5; Maichuhie 2018 , para. 7).
7.2. The Discourse on Socio-Economic Transformation in South Africa
Many debates in the recent past have dwelt on the concept of transformation and the progress made in its achievement in the post-Apartheid era as defined and perceived by different social and political groupings. Transformation has become the currency of most political parties to the extent that there is “a narrative that they create for example, that our candidate is fighting for radical economic transformation, [he/she] is fighting against capital, [and] wants to fight for the mass marginalised” (Respondent Three 2017) . This is against the backdrop of raging public debate on the need to rein in white monopoly capital and redistribute land so as to benefit the landless millions. A look at the land question may help in understanding the standpoints of political parties when it comes to policies vis-à-vis the question of race. The importance of the land question cannot be overstated as its arbitrary annexation by white-dominated governments was the basis of the Apartheid system.
It is noteworthy that after taking cognisance of the mood in the country, in April 2018 the ANC supported an EFF-sponsored motion in the National Assembly that led to a Resolution proposing a look into the feasibility of land expropriation without compensation. The Resolution also calls for the exploration of the possibility of reviewing Section 25 of the constitution that deals with right of property ownership. The Resolution explicitly states in its text “that the African majority was only confined to 13% of the land in South Africa while whites owned 87% at the end of the apartheid regime in 1994” (Crouse 2018, para. 7) . Lange (2017, para. 1)  however contends that as at 2017, black South Africans already owned more than half of all agricultural land in two of South Africa’s most fertile provinces i.e. the Eastern Cape and KwaZulu-Natal.
Despite the fact that the ANC’s support of the Resolution can be said to be a follow-through of its Declaration at the 54th National Conference in December 201714 , it can also be inferred that the timing of party’s move on land expropriation was a reactionary one aimed at appealing to the poorer majority black African voters ahead of the 2019 general elections and staving off the rise of EFF’s popularity. This inference is valid when the position of the EFF is considered and given both parties’ strong opposition of each other on most policy issues.
Since its inception in 2013, the EFF has been steadfast in its call for, among other radical things, the expropriation of land without compensation, and has used its calls to target poor black South Africans who are increasingly impatient with the progress of socio-economic transformation more than two decades after the end of Apartheid. Gerber (2018, para. 22)  notes that other parties that support the expropriation without compensation principle as documented in the National Assembly Resolution are the IFP, United Democratic Movement (UDM), NFP, Agang South Africa (Agang SA), African Independent Congress (AIC) and African People’s Convention (APC).
The DA does not support the push for land expropriation within the framework alluded to above and voted against the Resolution in the National Assembly. Using the Western Cape Province as a case where the party claims to have had a six out of ten success rate, the DA advocates for schemes whereby workers are given equity shares in the farms they work in. While claiming that it is committed to promoting justice in land reform, the party is opposed to allowing the process to be used as a “decisive and racially-charged lightning conductor to pull public attention away [from] the failures of government” (Masilela 2018, para. 4) . Its centrist positioning on issues as well as the fact that it represents the very constituency (white South Africans) that is perceived to significantly control three factors of production (land, capital and entrepreneurship) may inform the party’s stand.
As the country headed towards and beyond the 2019 general elections, the discourse on transformation continued to target many sectors including banking, finance and mines. It is highly likely that the pattern created by respective parties’ support for the Resolution on land expropriation will continue to play out during debates on other sectors that will be targeted for transformation policy wise. These include the contentious debate of whether or not to nationalise banks and mines.
8. Lessons Learnt
This article set out to assess the linkages between ethnic and race relations and the activities of political parties in Kenya and South Africa.
The regulatory mechanisms that have been put in place in both countries have taken cognisance of the reality of ethnic and race relations and how that can be used to ethnicise and/or racialise political parties. In spite of the legal frameworks in place to ensure that political parties have a national outlook, political parties in Kenya have found ways to circumvent them by focusing on the ethnic bases of their leaders when mobilising support. This has led to the perception that a political party is an institutional representation of an ethnic group based on the ethnicity of its top leadership.
South African laws leave open the possibilities of formation of political parties based on ethnicity or race. The same laws, however, prohibit the registration of parties that use ethnicity or race to incite hatred or violence and offend or deny membership to potential members. Given that the formation and registration of parties are two distinct processes, a legal lacuna allows South Africans to associate politically based on ethnicity and/or race. In a country that has a fractious past, this lacuna may be an impediment to any efforts aimed at breaking the legacies of Apartheid.
Political parties in both countries have been used as fulcrums of identity politics. They therefore define political leanings of different demographic groups in Kenya and South Africa where the allocation of public goods and national resources have historically taken ethnic and race angles. As a result, political parties have constantly exposed the fissures between different identities.
These fissures have manifested differently in each of the two countries. In Kenya, such fissures have led to cyclic episodes of ethnic and political violence at the height of political party activity during elections. This is exacerbated by the fact that Kenyans mainly vote along ethnic lines. To this end, the clash between political parties is another way of describing a clash between two or more ethnic groups. In South Africa on the other hand, even though there have been many incidences of political violence, they have rarely taken an ethnic or race angle. Piombo (2009, p. 33)  attributes this to the fact that South African political parties have avoided narrow ethnic appeals or avoided focusing on social cleavages in a manner that can cause ethnic/race violence. It can also be surmised that because ideology has historically played a role in the posturing of political parties in South Africa, there is less focus on the ethnicity/race of the person standing for public office and more on the issues that his or her party stands for.
When it comes to formulation and implementation of policies for allocation of public goods and national resources, ethnic and race dynamics have played a role in how political parties posture themselves. In Kenya, the implementation of government policy depends on the political parties (read: ethnic groups) in power. Wrong (2009, p. 64)  aptly describes the situation as “it’s our turn to eat”. Ethnic groups whose parties are not in power are therefore often marginalised from the eating. In South Africa, the discourse on post-Apartheid transformation is one that has race connotations. This is indicated by how the DA and other centrist parties that have the support of white South Africans, are in direct opposition to the left-wing parties that have the support of black South African voters when it comes to radical economic transformation.
Ethnicity and race are endemic to the existence and nature of operations of political parties in Kenya and South Africa. Although there is recognition in both countries of the need to provide some regulatory mechanisms in order to prevent ethnicisation or racialisation of politics, political parties continue to define the identities of respective citizenry due to how different ethnic and/or race groups support parties and vote during elections. As a result, most political parties formulate policies and allocate public goods and national resources along ethnic and racial lines in a bid to satisfy their support bases so as to achieve and/or maintain political power.
The author wishes to thank Dr. Marty Branagan and Dr. Rebecca Spence, both of the University of New England, Armidale, New South Wales, Australia, for their support and initial review of this paper. The author also appreciates partial financial support from the French Institute for Research in Africa (IFRA-Nairobi) which enabled more extensive field data collection in Kenya.
1The term “rainbow nation” was coined following the end of Apartheid to symbolise the new South Africa in which there is recognition of the unity of multi-culturalism and convergence of people of many nations within a country once plagued by discrimination (Mapungubwe Institute for Strategic Reflection (MISTRA) 2014, p. 76).
2Due to ethical concerns, pseudonyms are used to mask identities of interviewees and questionnaire respondents. Interviews and questionnaires were conducted in Durban and Pietermaritzburg (South Africa) as well as in Nairobi (Kenya) between November 2017 and February 2018.
3For more on the alternative ways of thinking about race and ethnicity please see Gracia (2007, pp. 4-10).
4In South Africa, the term “Coloured” denotes the people of mixed-race parentage as opposed to indigenous Africans or whites of European ancestry (Christopher 2002, p. 405).
5Respondent Twelve (2018) notes that ahead of the 2013 general elections, Onyango Oloo (a Luo) was installed as the Secretary-General of The National Alliance (TNA) party led by Uhuru Kenyatta (an Agĩkũyũ). Given the party’s massive membership by the Agĩkũyũ, the party members constantly questioned Oloo’s loyalty. The same suspicions stalked Raphael Tuju (a Luo) who, ahead of the 2017 general elections, was also installed as Secretary-General of the Jubilee Party of Kenya, which is perceived as a party of the dominant tribes of Agĩkũyũ and Kalenjin.
6In this context, “party owners” are highly resourced individuals who fund the operations and campaigns of political parties and generally have a big say on who becomes a party official (Respondent Eleven 2018).
7The Jubilee Party of Kenya was formed in September 2016 after the merger of 11 political parties. It brought together Uhuru Kenyatta and William Ruto who command the support of the two largest ethnic voting blocs in Kenya: GEMA (a conglomeration of the Agĩkũyũ, Embu and Meru) and the Kalenjin. It also brought together other leaders and/or parties perceived to have sway over the votes of their ethnic compatriots; leaders who could use that sway to negotiate a stake in a coalition government.
8NASA is a political coalition made up of several opposition parties and was formed in April 2017. The key parties are Raila Odinga’s ODM, the Wiper Democratic Movement led by Kalonzo Musyoka, the Amani National Congress led by Musalia Mudavadi, Ford Kenya led by Moses Wetangula and Chama Cha Mashinani led by Isaac Rutto. In congruence with its leadership, the coalition brings together the Luo, Kamba and Luhya ethnic groups, and a section of the Kalenjin community respectively.
9“Local and metro council section” in this instance means the category A, B and C municipalities found in a province as prescribed by sections 7 to 10 of the Local Government: Municipal Structures Act, 1998 (Act 117 of 1998).
10A 2013 cell phone survey found that 52% of black South African respondents aged between 15 - 34 years believed that if brought to power, the DA would bring back Apartheid (Mail & Guardian 2013, para. 3).
11One of the comments that Respondent One (2017) alludes to is attributed to Helen Zille, a former DA Federal Leader and who until 2019, was the premier of Western Cape Province. After her March 2017 visit to Singapore and appreciating its good infrastructure, Ms Zille made comments through her Twitter account implying that colonialism was not just negative as it benefited South Africa’s infrastructure (Corcoran 2017, para. 3). The comments invoked public ire, as they were deemed racially insensitive in a country trying to heal from its Apartheid past.
12The Congress of the People (COPE) is a South African political party formed in 2008 by former members of the ANC.
13The Jubilee Alliance was a coalition mainly built around The National Alliance (TNA) party and the United Republican Party (URP), respectively led by Uhuru Kenyatta (an Agĩkũyũ) and William Ruto (a Kalenjin). It also included the National Rainbow Coalition (NARC) party of Charity Ngilu (a Kamba) and the Republican Congress (RC) party of Najib Balala (part of the Kenyan coastal communities that have mixed ethnic descents) (Mérino 2013, p. 76).
14The Declaration indicates that as a matter of policy, the party would pursue expropriation of land without compensation (African National Congress 2017, p. 11).
 Mapungubwe Institute for Strategic Reflection (MISTRA) (2014) Nation Formation and Social Cohesion: An Enquiry into the Hopes and Aspirations of South Africans. Real African Publishers, Johannesburg.
 Appiah, K.A. (1998) Race, Culture, Identity: Misunderstood Connections. In: Appiah, K.A. and Gutmann, A., Eds., Color Conscious: The Political Morality of Race, Princeton University Press, Princeton, 30-105. https://doi.org/10.1515/9781400822096-002
 Zaibert, L. and Milán-Zaibert, E. (2000) Universalism, Particularism and Group Rights: The Case of Hispanics. In: Gracia, J.J.E. and De Greiff, P., Eds., Hispanics/Latinos in the United States: Ethnicity, Race, and Rights, Routledge, New York, 167-179.
 Gracia, J.J.E. (2007) Race or Ethnicity? An Introduction. In: Gracia, J.J.E., Ed., Race or Ethnicity: On Black and Latino Identity, Cornell University Press, London, 1-18.
 Cowen, M. and Kanyinga, K. (2002) The 1997 Elections in Kenya: The Politics of Communality and Locality. In: Cowen, M. and Laakso, L., Eds., Multi-Party Elections in Africa, James Currey Ltd., Oxford, 128-171.
 Alumona, I.M. and Azom, S.N. (2017) Politics of Identity and the Crisis of Nation-Building in Africa. In: Oloruntoba, S.O. and Falola, T., Eds., The Palgrave Handbook of African Politics, Governance and Development, Palgrave Macmillan US, New York, 291-306.
 Ahere, J.R. (2018) When a Compatriot Becomes a Foe: Political Parties and Violent Elections in Kenya and South Africa. Journal of African Elections, 17, 25-48.
 Southern, N. (2011) Political Opposition and the Challenges of a Dominant Party System: The Democratic Alliance in South Africa. Journal of Contemporary African Studies, 29, 281-298. https://doi.org/10.1080/02589001.2011.581478
 Anyangwe, C. (2012) Race and Ethnicity: Voters’ Party Preference in South African Elections. International Journal of African Renaissance Studies Multi-, Inter- and Transdisciplinarity, 7, 38-58. https://doi.org/10.1080/18186874.2013.774697
 Nhlapo, M.S., Anderson, B.A. and Wentzel, M. (2017) Trends in Voting in South Africa 2003-2014. University of Michigan Population Studies Center, Ann Arbor.
 Roodt, M. (2016) South Africa’s Local Elections: Reasons for Each Party to Be Happy and Sad. African Arguments.
 Brock, J. (2016) ANC Shaken to Core as South African Voters Look beyond Race. Thomson Reuters. https://www.reuters.com/article/us-safrica-election-race-analysis/anc-shaken-to-core-as-south-african-voters-look-beyond-race-idUSKCN10G1BJ
 Mkhize, N. (2016) The Indian Vote Is the Prize Political Parties Are Chasing in KwaZulu-Natal. Business Day. https://www.businesslive.co.za/bd/politics/2016-07-03-the-indian-vote-is-the-prize-political-parties-are-chasing-in-kwazulu-natal
 Makulilo, A.B., Ntaganda, E., Aywa, F.A., Sekaggya, M. and Osodo, P. (2016) Election Management Bodies in East Africa: A Comparative Study of Electoral Commissions to the Strengthening of Democracy. African Minds, New York.
 Gecaga, M.G. (2007) Religious Movements and Democratisation in Kenya: Between the Sacred and the Profane. In: Murunga, G.R. and Nasong’o, S.W., Eds., Kenya: The Struggle for Democracy, Zed Books Ltd., London, 58-89.
 Mail and Guardian (2013) Sample Survey of Young Blacks Believe DA Would Bring Back Apartheid. Mail and Guardian.
 Shilaho, W.K. (2016) Ethnic Mobilisation and Kenya’s Foreign Policy in the Face of the International Criminal Court (ICC). Journal for Contemporary History, 41, 103-125.
 Wa-Mungai, M. (2007) “‘Is Marwa!’ It’s Ours”: Popular Music and Identity Politics in Kenyan Youth Culture. In: Njogu, K. and Oluoch-Olunya, G., Eds., Cultural Production and Change in Kenya: Building Bridges, Twaweza Communications, Nairobi, 47-62.
 Barasa, L., Ahmed, M., Fredrick, F. and Apollo, S. (2017) Uhuru Warns against Secession Campaign as Nasa Vows Defiance. Daily Nation.
 Gerber, J. (2018) Land Expropriation without Compensation: What You Need to Know. News24. https://www.news24.com/SouthAfrica/News/land-expropriation-without-compensation-what-you-need-to-know-20180228
 Masilela, B. (2018) DA Resolute against Land Expropriation without Compensation. The Citizen. https://citizen.co.za/news/south-africa/1884367/da-resolute-against-land-expropriation-without-compensation