Participatory Democracy in Southern Africa: Explaining Botswana’s Exceptionalism

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The manuscript makes a laudable effort to take African institutions--in this case the kgotla--seriously in political analysis. This is a vital and necessary frame for research on the continent, and in keeping with moves to decolonize social science. However, the manuscript makes sweeping claims based on insufficient data as currently presented. The empirical evidence is weak and the claims so broad (and often under-conceptualized/operationalized) that it does not appear yet ready for publication. Based on the data presented, the causal mechanism connecting the kgotla institution to ‘exceptionality’ is weak. Moreover, the key variables are not clearly conceptualized or operationalized, which results in sweeping statements that cannot be supported by the limited empirics. For example, the manuscript describes Botswana’s ‘exceptionality’ as social, political, and economic--though these immense categories are largely left undefined. These concepts are too broad and undefined to develop any meaningful causal relationship, or to act as variables of interest. Furthermore, the premise of ‘exceptionality’ is not adequately interrogated. The wealth of literature dissecting Botswana’s exceptionality narrative that is mentioned in a cursory manner and not weighed up meaningfully. The author is free to reject the critical interpretation of the exceptionality discourse, but they must, at a minimum meaningfully engage with this body of research. Additionally, the measurements of concepts are left under-theorized. For example, is the level of democracy judged by voter turnout? The author seems to want to make an argument to understand Botswana separate from analysis of a representative/parliamentary political system, but then uses those very metrics about the participation in the electoral system to validate claims? This is logically inconsistent. An improved manuscript would focus more specifically on the kgotla and what it does in practice that can be evidenced, perhaps on a smaller scale, rather than making a sweeping claim about broad and undertheorized concepts like ‘exceptionality’ or ‘good governance’.

 

The manuscript does not provide a plausible causal relationship between the empirical data presented (~20 interviewees and several days of participant observation) with the ambitious claims being made. I would encourage the author to approach the manuscript with greater humility in the analytical claim. The manuscript is overly reliant on anecdote and a very narrow empirical base. I am myself a qualitative scholar and am advocate for qualitative methodologies in political science and the social sciences more generally. Qualitative work can be just as rigorous as quantitative scholarship, but what is presented in this manuscript is lacking. The depth, breadth, and robustness of the data must match the claims made. The empirical base of this manuscript is far too narrow (and indeed at times anecdotal) to provide sufficient evidence. The scope of the data is mismatched to the scope of the proposed finding. With the narrow empirical base, the author should either temper the scope of the claims, or conduct further research to move the data from limited and anecdotal to comprehensive and systematic.

 

The manuscript is composed of several leaps of causality that are not rigorous nor evidenced. For example, the author states: “The causal link connecting dikgotla to these positive outcomes is clear, as dikgotla is a millennial practice whereas the all others variables have come later.” This is not a scientifically valid claim. The existence of the kgotla system does not render subsequent causal variables moot, nor does broad proposition account for the nuance that the practice of the kgotla was historically not always in widespread use across all the peoples and territories that become contemporary Botswana. What may hold true in the historical morafe does not necessarily hold true for other communities and regions of the country.

 

The manuscript does not meaningfully engage with the rich existing literature interrogating the kgotla (see scholars such as Gulbrandsen, Mompati, Chebanne, Nthomang, de Jager and Sebudubudu). As such, it does not account for the kgotla as a site of patriarchy, gerontocracy, and Tswana-domination. By failing to engage with these concepts, it presents a relatively one-sided analysis of the kgotla as an institution. For example, on page 7 the author comments in passing about gender without engaging the published scholarly literature that analyzes gender dynamics in kgotla, and other marginal populations like non-Tswana ethnic groups and youth. The manuscript can certainly provide a different interpretation, but it must grapple with existing scholarship that understands kgotla meetings as highly choreographed affairs that can limit freedom of expression. The author proclaims that the kgotla system holds political actors accountable. Again, this is presented as a self-evident fact without ever clarifying how it holds political actor accountable, or providing evidence of this accountability or a mechanism demonstrating this occurring in actual fact.

 

The kgotla is a Tswana institution, albeit one that has become embedded throughout the country. However, the Tswana people are not the only people in Botswana and this manuscript under-recognizes this complex reality. The manuscript should be engaging with the rich literature regarding identity in Botswana (see the work of scholar/activist Nyati-Ramahobo, or anthropologist Nyamjoh). Some particularly useful studies examining the contestations about (and subsequent revisions to) the House of Chiefs in the late 1990s/early 2000s would vastly improve the analysis about the intersections between identity and traditional authority. For example, the author states “Particularly the north of today’s Botswana escaped major colonization and was thus able to maintain its own political institutions”--but the north (and west) of the country are largely inhabited by non-Tswana Batswana, so the importation and utilization of the kgotla itself is a political imposition of a kind.

 

The manuscript assumes that the kgotla operates uniformly throughout the country--that the one instance of participant observation in one community is sufficient data to conclude sweeping claims regarding the institution and its functions across the whole of the country. This is, to my mind, flawed analysis and represents too great a leap from limited empirics. If, for example, the participant observation occurred in a primarily San community in the country’s west, the findings may be substantially differently. Additionally, engaging with opposition figures may complicate the picture presented of the kgotla as largely apolitical (many opposition politicians argue they are de facto ruling party entities). The empirical base of the data collection is too narrow to incorporate different perspectives and there is simply not enough data to make the kind of all-encompassing argument presented by the manuscript.

 

There are some errors of fact in the manuscript. For example, it is simply not true that all citizens are able to freely access land and livestock--this is highly variegated by region, land tenure regime (ex: people living in conservation areas) and ethnic identity. Additionally, Botswana is far more diverse than the eight Tswana tribal groups, and the manuscript does not engage with the ethnolinguistic complexity of the country, in favor of a perception of Tswana homogeneity that is counter to facts on the ground. On page 8 the author mentions in passing “areas of the country without a traditional or dominant ethnic group” which completely erases the non-Tswana communities who often pre-date the arrival of the Tswana in the area. It seems to adopt the position that history in Botswana only begins with Tswana in-migration, and assumes that other co-national ethnic groups are somehow not ‘traditional’ in the country.

 

The manuscript says: “British colonialism weakened the power of traditional chiefs, dikgosi, it did not affect dikgotla, the village assemblies” This is a very definitive claim that does not provide evidence or citation. Such a sweeping statement would require substantiation. Moreover, it seems to elide the fact that many chiefs in the colonial era sat on Native Advisory Councils, and played an important role in the administration of the colonial state. In the postcolonial transition period and after, political elites were often drawn from the dikgosi (such as Seretse Khama and Bathoen).

 

I would encourage the author to re-engage with a broader swathe of existing scholarship, not just in political science/theory but also history and anthropology in order to more firmly ground the manuscript in the literature. I would also encourage further research in order to develop sufficient data to empirically support the ambitious scope of the manuscript.  

Author Response

Reviewer 1:

I have followed the advice of reviewer 1 and reformulated my claims. I now make narrower and more humble claims, as requested by this reviewer, and I have added a section on the limitations of this study, resultant from the limited scope of my empirical research. I hope to be able to return to the field with more time and resources to conduct a more systematic study - and I have stated this in the text.

I have integrated all the references this reviewer suggests: Gulbrandsen, Mompati, Chebanne, Nthoman, Jager and Sebudubudu, Nyati-Ramahobo, and Nyamjoh. 

I have expanded the theory section quite a bit, responding to the suggestions by reviewers 1 and 2, to make the relationship between deliberation and democracy clearer. I added a long quote from Bernard Manin on deliberation and relate it to the work of Bidima. 

I realize that the empirical foundations of this article do not live up to the standards of cultural anthropologists or historians. I am, however, a political scientist. Most of my colleagues never venture into the field. I hope to be able to return to Botswana with more time and resources in order to expand my empirical basis. I have stated as much in the article. I am particularly interested in observing dikgotla in the north of the country, where the Tswana are not dominant.

I have addressed the ethnic diversity reality, following the suggestions of reviewer 1. 

 

Reviewer 2 Report

Comments and Suggestions for Authors

This is a review of Participatory Democracy in Southern Africa: Explaining Botswana’s Exceptionalism for the journal Social Sciences.  This is an excellent paper that is well written and addresses an important research question.  I enjoyed readings it and found its qualitative approach, including intensive field work from Botswana, to be compelling and instructive.  Based on my reading of the paper, I have a few comments and questions.  I present these primarily in the order I read the paper. 

-Is Botswana more generally truly an exceptional democracy?  Since independence, it has been dominated by the BDP which is not indicative of a consolidated democracy.  The sign of democracy is the transition of power from competing parties, and, as we can see, many countries that have had elections, like Niger, are far from being true democracies as they cannot undergo a true transition in power.  Why is this not true for Botswana? 

-The HIV prevalence of 20%, as cited in the article, is deeply troubling from the perspective of an argument of good governance.  I realize that many factors in this case are out of the control of the government, but there are counter-examples, like Senegal under Diouf, that did a much better job of controlling the disease.  I think more should be said about this point.

-The discussion of diamonds and resource wealth as a possible counter-argument is a bit of a straw-man.  Unlike cases like Sierra Leon, diamonds in Botswana are found deep in mines and thus accessing them is a capital-intensive process.  This effectively eliminates the conflict mechanism inherent to non-excludable resources as people have no incentive turn to violence in order to preclude others from accessing diamonds because they are inherently excludable.  I think more needs to be said about this because the dynamic nature of collaboration with DeBeers certainly seems like an incentive to govern effectively. 

-Building on this last point, could we say if the Dikgotla was not present, i.e., the counterfactual scenario, with all of other aspects of Botswana remaining the same (ethnic homogeneity, small population, regional location, deep diamonds), that the outcomes would suffer as a result?

-What is precisely the mechanism between Dikgotla and the outcomes of good governance?  Or are there multiple mechanisms?  I was unclear exactly what was driving it, and found possible explanations being an information effect, a participation effect, but also an interaction effect between leaders and their constituents.  A more detailed explanation about what the Dikgotla does would be helpful, perhaps adding some sub-sections in the theoretical section. 

-Is traditional rule in Botswana really democratic?  Is it not more of an example of competitive authoritarianism?  The selection of the chiefs, which the article notes was hereditary and from a royal line, is typical of traditional life in Africa, and while it leads to good governance that does not make it democratic.  Too often it seems good governance is taken to mean democratic rule and that might not be the case at all levels of governance.  Why is this then not another case of good governance in an authoritarian system or a ‘communocracy’ as the paper cites?  For what is it worth, I disagree that colonial rule, as Mamdani asserts, turned local rulers into despots across Africa.  They were always despots it’s just that the constraints they faced changed with colonial rule.  From Mansa Musa in Mali to Osei Tutu in the Asanti Empuire to Shaka Zulu in Zululand, traditional leaders in Africa have always been authoritarian. 

-The argument that traditional African institutions promote good governance seems to work in Botswana’s case.  This is not necessarily a novel argument.  In political economy, Botswana is often cited as the shining star in Africa, and it is noted as an example that institutions matter for development (Acemolgu, Johnson and Robinson 2002; Acemoglu and Robinson 2010).  But, the question with Botswana is if it is an exception.  It has a small and ethnically homogenous population that is exceptional compared to the rest of Africa.  In my own field work, I have seen how traditional rule in ethnically diverse countries, like in West Africa, has worked and produced good governance for those sharing the ethnic identity of the traditional rulers but for those on the outside or out-group members it has led to ineffective governance and in some cases the worst forms of governance.  To what degree, then, is Botswana an exception?  Does this model only work in small, ethnically homogenous societies?  Indeed, the author/s effectively make this point at the end of the paper with the discussion of Lesotho.  I thus think more discussion of either the generalizability of this case or what others can take from it would be good to offer. 

-The conclusion was outstanding especially discussion of Western academics upholding a colonial order and mentality.  I agree completely with the author/s conclusion that the solution and expertise come from within these places, but to some degree I feel like pointing out that it is “democracy” in traditional life falls into the same trap.  For me, whether local rule is democratic in a Western or any sense is irrelevant, but rather that Africans have the institutions to promote good governance. 

 

Minor Points:

-Line 141: believe should be belief

-Line 192/448/584: Batswana I think should be Botswana

-Line 214: What is perduring?

-Line 304/305: spend should be spent.

-Line 533: delete “the”

 

Citations:

Acemoglu, Daron, Simon Johnson, and James A. Robinson. "An African success story: Botswana." Available at SSRN 304100 (2002).

Acemoglu, Daron, and James A. Robinson. "Why is Africa poor?" Economic history of developing regions 25, no. 1 (2010): 21-50.

Author Response

Reviewer 2:

I am very grateful for the comments I have received from reviewer 2. I have made all the minor changes suggested, with the exception of line 192/448/584: Batswana is correct. The citizens of Botswana are called Batswana.

I have addressed the issue about exceptionalism and HIV / AIDS prevalence and expanded on the diamonds issue and overall economic performance by integrating the work of Acemoglu et al, 2001 and 2010. I found those two articles very helpful. They support my argument.

I have added more text to the theoretical section about the causal link between dikgotla and good governance, as suggested by Reviewer 2, and highlighted Botswana's exceptionalism, as requested. 

I do not agree with the de-emphasis this reviewer suggests on democracy, but I have expanded on this issue, particularly in the conclusion, hopefully to the satisfaction of this reviewer. 

As a result, my article is now quite a bit longer. All additions are in red.

I want to thank reviewer 2 for a very helpful review.

Reviewer 3 Report

Comments and Suggestions for Authors

I am so happy to read this very interesting article on Bostwana. I always said that there were very original African experiences that need to be known and this case-study is very precious for the subfield of participatory democracy, I really recommend its publication with some points that can be considered for improvement.

It could be good to relate more the kgotla to other political experiences in West Africa even though colonialism marginalized the visibility of these practices. I think about the notion of palabre (Jean-Godefroy based his philosophy on this word and his work could be mentioned in the theoretical framework).

Make sure to present the methodological framework in a more rigorous way. Did you make a participatory observation or just an observation  of the kgotla? Give all the details on this observation, the scene, your role and so on. The list of interviews has to be presented. It is not enough with "around 30 minutes", did you record the interviews? What were the questions? I would suggest the author to add an appendice with all these details. It would help to interprete the originality of the findings. Do you have the transcripts of the interviews?

Moreover, I think that you can take more quotes from your empirical investigation. You can even deepen some of the remarks. When an interviewee said that a kgotla was like a court, I think that here you have a very original mode of citizen jury that is worth being commented. Take advantage of your valuable empirical data. Maybe, a transcript of these data could be provided if possible (an appendice or a link to a repository). Thank you for such a good work, there is a lot of potential.

Maybe, the last part of the conclusion could enlarge the focus and call for comparative work in Africa on participatory practices, this is needed in a field where there is an overproduction of research on participatory democracy in Europe and in the Us.

Author Response

Reviewer 3:

I found the comments by Reviewer 3 very encouraging and helpful. I have integrated the work of Bidima. I am grateful for this suggestion, as it strengthens my argument and article. 

I have expanded a little bit on the methods section explaining my methods some more, and I added a little bit more empirical information (to my field notes taken during the Modipane kgotla I observed). I am, however, not able to add full transcripts as an addendum. I hope this is acceptable to this reviewer.