top of page

ISI Annual Lecture 2022

Copyright © 2023

Inclusive Society Institute PO Box 12609

Mill Street

Cape Town, 8000 South Africa 235-515 NPO All rights reserved. No part of this publication may be reproduced or transmitted in any form or by any means without the permission in writing from the Inclusive Society Institute D I S C L A I M E R

Views expressed in this report do not necessarily represent the views of

the Inclusive Society Institute or its Board or Council members.

Author: Olivia Main

Editor: Daryl Swanepoel



1. Setting the scene

2. Annual Lecture presented by Jørgen Elklit, professor emeritus, Department of Political Science, Aarhus University

3. Q&A session with Jørgen Elklit, professor emeritus, Department of Political Science, Aarhus University

1. Setting the scene

The Inclusive Society Institute (ISI) hosted its Annual Lecture in Cape Town on 8 December 2022. Jørgen Elklit, professor emeritus at the Department of Political Science, Aarhus University, in Denmark, was invited to lecture on the topic “Democracy and Electoral Systems as Tools to Promote Social Cohesion”.

In early 2022, the ISI published a report in which it made proposals concerning electoral reform, in light of the Constitutional Court judgement declaring the current Electoral Act unconstitutional. The Institute appointed an expert panel, including Professor Elklit, to undertake the work and mandated the panel to design an electoral model that will meaningfully give effect to the judgement, respect the boundaries set out in the Constitution, retain proportionality as a basis for representation in that it best promotes inclusivity, and which enhances representativity, accountability and transparency.

Professor Elklit is an expert on global electoral systems. Since 1990, he has been active as an advisor on democratisation and elections, electoral systems and electoral administration in a number of countries across the globe. Elklit was also an international member of the South African Independent Electoral Commission in 1994 and of the country’s Electoral Task Team 2002-2003 (aka the van Zyl Slabbert Commission), and was Secretary to the Independent Review Commission in Kenya in 2008.

Professor Elklit’s lecture focussed on how we can strengthen democracy and social cohesion through an improved electoral system. Elklit opined that this must be done within the framework of South Africa as a diverse population with various backgrounds. The understanding that the mere holding of elections entitles a country to be called a democracy is an electoral fallacy. Elections must also be fair, inclusive and integrous in order to gain voters’ trust in the election process and the outcome of such.

In a single seat constituency model – in other words, a winner-takes-all majority election system – you could conceivably have many provinces represented wall to wall by one political party, which, in South Africa, would most likely be the ANC.

Even though the system would result in general proportionality, those members would only come from the compensatory proportional list, which may or may not draw from members of a particular province. This does not promote inclusivity of the diversity in the country or do much for geographic representation. It's a problematic approach to adopt in a country such as South Africa.

Therefore, in order to achieve a true democracy and social cohesion, we need to adopt a system that not only results in general proportionality, as is the constitutional requirement, but also where representatives are drawn from the many different communities across the country – in other words, a multi-member constituency model. We also need to ensure that all eligible voters are included in the elections, by adopting a system such as automatic voter registration.

2. Annual Lecture presented by Jørgen Elklit, professor emeritus, Department of Political Science, Aarhus University

This lecture on “Democracy and Electoral Systems as Tools to Promote Social Cohesion” is an opportunity to engage in a much-needed conversation about matters of the utmost concern for all who want to see South Africa prosper, develop, and eventually become the country we want it to be – the country so many dreamt of during the Struggle, but also after 1994.

Democracy is a fluid concept, defined and interpreted differently by different people at different times and in different situations. But there are at least three elements that are necessary for a democracy to develop and consolidate: free and regular elections of good quality, the full complement of political rights for the citizens and, above all, the rule of law, which is often the weak link in emerging democracies.

Elections must come at regular intervals, and between elections citizens must be able to try to influence public policy through other instruments such as interest associations, social movements and civil society organisations, local groupings, and so forth. Only then can we see the development of a complex system of institutions, rules and patterns of incentives and disincentives which makes it evident that democracy has become “the only game in town”.

Not too many decades ago, some politicians – and even some political scientists – claimed that if a country conducted elections for its leadership, then it was a democracy, as its leadership was elected by the people, the demos – even if the elections were only of poor quality.

It was, however, not difficult to challenge that view – especially based on the way elections were conducted in Latin America during that period, but also elsewhere, such as the Soviet Union, Eastern Europe, and in many countries in Africa. The understanding that the mere holding of elections entitled a country to be called a democracy was pejoratively termed the electoral – or electoralist – fallacy. People then started to look for other, more comprehensive definitions of democracy.

The inclusion of the entire adult population in elections was obviously not to be left out, on the contrary, but what else should go into the definition?

Robert Dahl, a leading US political scientist, gave us an important list of seven minimal conditions that must be present before we can talk of a country being a modern political democracy.

Dahl required that elections should not only be fairly conducted, but they must also be seen so by the electorate, as that makes it natural for all – losers as well as winners – to trust the outcome of the elections and to perceive the elections as being of integrity.

The first prerequisite for having elections of integrity is to have a trustworthy electoral commission. South Africa can take pride in having had one of the best electoral commissions on the continent – partly because of a competent staff component; partly because commissioners have only rarely brought their private party political inclinations with them to the office.

This is not always so in other countries – in Africa or elsewhere – and one can only express the hope that politicians will honour the need for competent and trustworthy electoral commissioners considered to be men and women of integrity. And the electoral staff, at all levels, must have the same qualities, so that voters can rest assured that electoral results are honest and reliable – and not big lies as some US voters and politicians still claim, against all evidence, that the 2020 elections were.

The second prerequisite for high quality elections to pay attention to, is that all eligible voters must be on the electoral register. Going beyond Dahl speaking about the importance of having the right to vote, experience shows that some voters also need help to get into a position where they can actually use their voting right.

Elections can be considered a kind of conversation among the citizens of the country, where it is discussed and decided who shall form the political leadership during the next term – a conversation that is important for the emergence of social cohesion. If some citizens cannot participate in that conversation, the political leadership will have less legitimacy.

In South Africa, at least 30% of the eligible voters are not on the electoral register, despite the various efforts by the Electoral Commission of South Africa (IEC). Think of it: 30% of the eligible voters cannot participate in the elections. That is more than just a pity, it’s a disgrace!

The IEC will probably say that they have done all they can to get all voters on board, and that they cannot force people to register.

That is obviously true, but a registration rate of about only 70% is still not good enough for a country that claims to be a democracy. The situation has not improved over recent years – despite many registration campaigns – so it is high time to realise that the only solution is to shift to automatic voter registration.

Automatic voter registration means that the state takes steps to ensure automatic registration of all eligible voters. It is done in different ways in different countries, but it is normally done by linking voter registration to the national ID or to the social security system.

Automatic voter registration is not compulsory voting, which one should avoid at all costs. Voters are still free to vote or not to vote – as they should be – but automatic registration means that the entire adult citizenry can go and vote, even if they only decide to do so on election day. They don’t need to have registered beforehand, and they can react to what is happening during the election campaign and go and express their opinion if they wish.

The third prerequisite for qualitatively good elections might challenge some. But my experiences with many countries have been convincing enough to show that it is much better for a country – any country – to use a proportional representation electoral system than to rely on some kind of majority system.

The basic difference between proportional representation (or PR) elections and majority elections is that losers in PR elections don’t lose everything. A losing party will still have at least some seats in the Assembly, and those who voted for that party will still feel included in the political process, as some of the elected representatives sit in Parliament because the voters wanted that to be the case and therefore voted for them.

It’s easy to understand that this is particularly the case when a country is characterised by a number of social and political cleavages, due to racial and ethnic differences, religious antagonisms, national identities or class, to mention only the most obvious ones.

This was also the conclusion of a high-powered international commission under the chairmanship of former UN Secretary General Kofi Annan a few years ago.

With majority elections, politicians have too much at stake, as the winner takes it all, and the loser loses everything. That’s why we see candidates in majority elections fighting so hard to avoid losing – and behaving so unpleasantly, as did Donald Trump in the US, or some voters in Kenya in 2008 or in Lesotho in 1998, even if they nevertheless don’t win.

This consideration is also highly relevant in this country, and that is exactly why the ANC National Executive Commission during the negotiations in the early 1990s eventually decided to go for proportional representation.

The formula used for calculating each party’s share of the National Assembly seats is fine and the same can be said about South Africa having no formal electoral threshold. These two features – in combination with the size of the National Assembly – combine to make the National Assembly the most proportional parliament in the world, as measured by the very reliable and useful index developed by the Irishman Michael Gallagher.

This implies that any discussion about what is meant by the expression “in general proportional representation” in the Constitution is strange, as it is obvious that the National Assembly is more proportional than what we see in other PR systems. This also implies that the current PR electoral system can be changed, at least somewhat, without any risk of violating the Constitution.

Some change of the current electoral system is, however, seriously needed – not because the system is poor, but because South Africa has changed since 1994. The system, in other words the closed-list system, with 200 seats allocated in the nine provinces and 200 seats used as compensatory seats to reach full proportionality, was probably what was best in 1994, at the system’s inception.

But the political development since then has made many call for a system that makes it easier – some would even say possible – to hold parliamentarians more to account than is now the case. I agree with those who feel that way.

I had the honour of serving, as the only non-South African, on the van Zyl Slabbert Commission in 2002-2003. Our task was to develop an electoral system that was still compatible with the Constitution, of course, but that also introduced some element of accountability.

Our majority proposal suggested a number of multi-member constituencies – all in all 60-some such constituencies, with from three to seven parliamentarians elected in each of them, also by a PR system, so that local voters would be represented in proportion to the local strength of their parties. That would allow the current 200+200 system to be changed to a 300+100 system, with 300 seats allocated to the new and smaller MMCs, while 100 compensatory seats should definitely be enough to ensure that “in general proportional representation” could still be easily achieved.

Such small constituencies would allow a much closer relationship between represented and representatives than is now the case – or to put it differently: increase the level of accountability, especially when preferential voting would eventually be introduced. We did not propose preferential voting from day one, but we suggested a change from closed to open lists at some later point in time, above all, to allow a strong element of accountability to enter into the relationship between voters and representatives.

Our majority proposal had many qualities, but it was probably never considered seriously by the Cabinet, and therefore South Africa still uses the electoral system implemented in 1994.

We all know that the system does not allow independent candidates and that the Constitutional Court (CC) eventually had to consider the constitutionality of this fact – when the Constitution entitles all adult South Africans to stand for election to public office if they so wish, which is also one of Dahl’s seven conditions for a political democracy.

There has always been this incompatibility between the Constitution and the Electoral Act, so the CC judgement in June 2020 did not surprise me. It actually pleased me. What has surprised me, however, is that the National Assembly has been so slow to act.

And it has now even been necessary to apply for a further extension of the time frame for finalising the Electoral Amendment Bill, as more public consultations are suddenly needed. Will Parliament have completed the job in February 2023, or will it take even longer?

Various groups, civil society organisations, and even individuals have both studied the options and also presented a number of ways in which independent candidates could stand for election and thus comply with the views of the Constitutional Court.

The proposal presented in early 2022 by the Inclusive Society Institute is well worth considering, as it creates space for independent candidates, but not really at the expense of the established political parties. The only real change from the current system is that the number of lower-tier constituencies – which are now the nine provinces – will change to 66 smaller multi-member constituencies (MMCs), following already existing administrative borders.

These smaller MMCs will each elect a modest number of Members of the National Assembly, and that will bring with it a much closer rapport between voters and representatives elected in the local areas. This will eventually increase the level of accountability and can also be expected to increase the level of social cohesion, as voters in an MMC have something in common: the opportunity to elect their MPs, who might engage with the local voters and their various associations in defending the area’s interest in the National Assembly and within their respective parties. And this will happen even if an independent candidate is not among those elected in the MMC.

The ISI proposal would certainly be a good starting point for the work of the Electoral Reform Consultative Panel, which has now been suggested by the National Council of Provinces (NCOP).

And by the way: Experience from other countries tells us that very few independent candidates will eventually be elected. It’s good, and constitutionally required, that independent candidates can have a try, but there is no reason to fear that many of them will be elected.

It is in my opinion more important – but not because of constitutionality issues – to also use this opportunity to discuss the large number of political parties in South Africa. At the national level we saw 48 parties participating in the 2019 elections – and as many as 34, almost 75%, did not make it into the Assembly, despite there being no formal threshold.

I, therefore, was very pleased when I saw that one of the changes proposed by the National Council of Provinces to the electoral Amendment Bill was to require that parties not already presented in Parliament or the Provincial Legislatures, must also submit a substantial number of seconding signatures in order to be allowed ballot paper access.

The requirement is suggested to be 20% of the relevant quota at the previous election, which is exactly the same requirement as suggested for independent candidates.

This is a constructive solution to the problem with the first versions of the drafted Bill, namely that it would be much more difficult for independent candidates to get on the ballot paper than it would for new and formerly unrepresented political parties – who could even hope for access to the pool of compensatory seats.

So, suddenly both problems – (1) the differential treatment of independent candidates and new parties, and (2) the mushrooming of new parties with virtually no chance of getting elected – have found a common solution!

This constructive and welcome addition to the current electoral system does not violate the right to associate freely, which is so important in a democracy. But it makes it clear that access to the ballot paper does not come almost automatically to any small association or publicly known personality who can collect as little as the 1,000 signatures required for registration of a party.

Parties allowed on the ballot paper must of course be able to prove that they deserve this honour, and it is completely legitimate in a democracy to require parties to do so.

Parties that were elected in the previous election – and are still represented in Parliament – can be allowed to participate in the elections without further ado and have their performance as parliamentarians evaluated by their constituents, as that is how parties are being held to account.

The newly suggested requirement will definitely reduce the number of parties on the ballot paper considerably – and consequently, make it easier for serious parties to make it into the National Assembly.

It makes sense to see both the requirements for ballot paper access and for winning a seat in the National Assembly as important gatekeeping tools. Some kind of gatekeeping is necessary at Parliament’s gate to restrict entrance to those trusted by a reasonable number of voters. All to avoid votes being scattered over so many parties that none of them make it into Parliament.

One can even argue that the NCOP proposal is too kind to the previously unrepresented parties, but I will not go into that debate here. I shall only in passing mention that in my own country, in Denmark, the similar requirement for new and currently unrepresented parties is to provide as many seconding signatures as is the equivalent of a full electoral quota at the latest election, not only a modest 20%!

It has, however, surprised me that I have not seen or heard any public debate about this constructive suggestion that will definitively decrease the number of parties on the next ballot paper considerably – something that many South Africans will certainly see as a fine step forward, as I also do.

Social cohesion is about social integration and inclusion, in local communities and in society at large. I’m strongly convinced that the issues I have addressed in relation to the electoral system and the Electoral Amendment Bill will all have a positive bearing on social cohesion both locally and in the South African society as a whole.

My seven take-aways cannot be very surprising:

  1. Democracy is much more than elections, but as good quality elections is a necessary part of it. It is okay to pay special attention to elections.

  2. Democracy becomes stronger and more consolidated, the higher the quality and the more integrity the electoral process has.

  3. In South Africa, the competence and the integrity of the IEC has proven important since the first IEC. It is of the utmost importance that the perception of the IEC as competent and politically unbiased is not allowed to change.

  4. Automatic voter registration must be introduced as a matter of urgency.

  5. The ISI proposal for a new South African electoral system can be a good starting point for the work of the Electoral Reform Consultative Panel, which has been proposed.

  6. The number of signatures required for a national party to get on the ballot paper has been suggested to be the same as the requirement for independent candidates. That is a very welcome improvement to South Africa’s electoral system – it might even be a good idea to set the requirement for political parties somewhat higher than for independent candidates.

  7. My suggestions will all increase social cohesion, both in local settings – the MMCs – and in society at large, by reducing exclusion and by reducing the costs of taking responsibility for the attainment of goals agreed upon.

3. Q&A session with Jørgen Elklit, professor emeritus, Department of Political Science, Aarhus University

  • Q. The automatic voter registration will obviously bring down the costs for the IEC, because it's just a data merging between Home Affairs and the IEC. But at the same time, it would also most probably lead to a lower voter turnout, in the sense that, at the moment voter turnout is registered on 70%. Whereas, with the new system it would be registered on 100%. The problem in South Africa is with the young voters that are not coming on board. How do we motivate the young people to start talking and to exercise their democratic right?

A. Although the automatic voter registration would lead to a decrease in the official voting turnout number, I don’t think this matters, because the present voter turnout figure is artificial. The IEC says that the participation rate is about 70%, but on which participants are they basing this on? If it’s 70% of the elligible voters, then the current turnout in South Africa is actually only about 50%. But the IEC is always very proud of claiming that it is about 70%. What really matters is that the entire eligible population can easily exercise the right to vote without having to register or being registered some time ago, so I don't buy that argument at all.

In reference to the question about the young voters – bearing in mind that some female voters are also underrepresented in the current situation – the issue with young voters voting less than others and registering less than others is not only a South African problem, it is seen all over the world and it is, in particular, seen among underprivileged young voters of different kinds. They have to be motivated and one can do different things in order to get them to register, but it is a complicated issue and I think the main reason is that the parliamentarians are not delivering on their promises. If parliamentarians made society function better, so that not only young voters but all of us could see that it made sense to go to the polls and vote – because that would influence what happened in our local society, in our village, in our country – then it would be easier to motivate young voters to go and vote. It's something which cannot be remedied very easily, as one would have to change the entire political culture and deliver in a different way.

  • Q. I want to follow on from the previous question. The issue about automatic voter registration is that it’s great in a country where the systems work well, but when you have a Home Affairs that is struggling to ensure the integrity of its data, then what you’ll have going forward is a well functioning electoral system that is in doubt. And so, we are actually going to go backwards. In my opinion, there's a prerequisite to automatic voter registration, given our context. What is your comment on that?

A. Yes, the more serious problem is, of course, how can one introduce automatic voter registration when Home Affairs is functioning, or rather, not functioning the way it is? Therefore, I understand completely that it is not something that could be immediately implemented or even set up long before the next election. Rather, it requires that structures be put in place that can handle it. It would not necessarily have to be Home Affairs, it could be some other body, but it would have to be capacitated and funded to do so. It could even be something the IEC could take on.

It would require legislation and decisions of a very complicated nature. I understand that. But the current situation, where more than a quarter of the population is not able to vote is, from a democratic point of view, a serious problem, and one that has to be dealt with, in my opinion. It will be complicated to salvage, but as I was just saying, it is what should be done because it cannot continue as it is.

  • Q. How do we deal with the issue of fraud and having ”dead” people coming into the electronic voting with automatic voter registration?

A. The way the automatic voter registration would function to avoid fraud within the system, is that when a person dies, they are deleted automatically from the register. And therefore, also from the voter register. So, a person cannot come with a claim, for example, for their deceased grandmother, because she would have already been deleted from the voter register automatically. And on the other end of the spectrum, when a person turns 18, they will automatically be registered as a voter. I think if one could get an organisation to prepare the framework for this system, and it works, it would probably solve the problem of fraudulent voters. That's one of the advantages.

I understand that it's difficult to get to that point, because it’s expensive. But in my experience in Denmark, I have never heard of a person succeeding in using a deceased person’s voter ID, and that’s simply because they are no longer on the register. I'm not saying that fraud cannot happen. I’ll be honest with you, the only way it can happen is that sometimes a person’s voter card gets stolen. The system automatically sends out a voter ID card by mail with all your information on it, which you take with you when you go to the voting station. In theory, a person may be able to go and vote for another person using their stolen voter ID, but as I understand, it's very rare.

  • Q. I agree that it’s ridiculous to have such a long ballot paper. How do we effectively reduce the number of parties on the ballot paper, and guard against the counterattack that says this would be an assault on our democracy because people/parties are being suppressed?

A. I think the suggestion in the NCOP's changes to the Amendment Bill is a good one. If a political party not only has to register but also collect a certain number of signatures to support its being on the ballot paper, this will result in a natural screening of parties on the ballot paper, because some of the weaker parties that do not have the appropriate structures in place will not be able to deliver the number of signatures required.

I think one can easily argue that if a so-called political party trying to get into Parliament cannot even organise the collection of a certain number of seconding signatures, then why should it go to Parliament? It will not have the capacity, as we have seen in recent elections, to get enough votes to be elected. The outcome of the current situation is that you have ballot papers as long as my arm, which confuses voters and some voters might decide not to go and vote because they can’t find a party to vote for. So, by reducing that number considerably, those problems would be solved.

I'm not saying that new parties shouldn’t be allowed to stand for election, but rather, if they can organise the collection of the allotted number of signatures in time and get approved by the IEC, which could be the controlling organisation, then they might also be able to run a good election campaign, get out to the people and get the votes so that they would be represented. If they cannot organise the collection of signatures, then they would probably also not be able to get elected, and then there's no need to have them on the ballot paper.

So, I think it’s a set of regulations that would be very helpful, not only for the IEC, but also for all voters, who could more easily find solid, substantial parties on the ballot paper to vote for. I think that's a good way forward and I think one could even be less kind to the new and unrepresented parties, because there's no need to have them on the ballot paper if there’s no chance that they will get elected. If they have the support of the people, they will get the numbers they need to get into Parliament.

  • Q. In relation to the creation of a space for independent candidates. We are currently emerging from a very fragmented environment. I think we've been working hard for quite a long period of time on uniting as a country, so that we act as a collective in dealing with the national challenges. Especially at the national level, the very purpose of creating space for people to represent the interests of this society as individuals is to encourage unity and cohesion of South Africans, to act together in dealing with the country’s problems. What is your comment on this?

A. I think it's interesting because we talk about creating space for independent candidates a lot, but at the same time, many of us feel that we need the political parties as lightposts in developing the political policy proposals we need going forward, with candidates and all the functions of political parties. And that under normal circumstances it should be enough to have the political parties to organise the political space.

The problem is that sometimes individuals come forward who have strong opinions, people who could make a difference in Parliament with their views, but it happens very rarely. The idea is that, if such people do come forward, they can be tested by the voters, and if they can convince the voters in sufficient numbers of their abilities to influence the political debate in the country in a positive way, then they should be allowed to be tested, at least. I think that's the idea of the formulation in the Constitution of South Africa, and in the constitution of many countries, that if a person believes that they have something to offer in political life, there should be a possibility for testing whether the voters would trust that person. But as I said, it happens extremely rarely that people come forward who have convinced the voters that they have something to offer that is over and above what comes from the political parties. And that is why I said in my presentation that it is extremely rare that independent candidates are elected.

Something which has surprised me in relation to South Africa, is that the debate about independent candidates has been so intense from some corners. For one thing, there is the constitutionality problem, and for the other, what will happen in actual practice? Have we come across people who we would like to see as independents in Parliament? That can hopefully be tested in the next election, but my guess is that it would be very difficult.

It has only happened on one occasion in Denmark, and I know that the Danish political system is very different from the South African system, but on one occasion an independent candidate was elected. It was a comedian who was very popular at the time. He was not your ordinary political comedian; he was just your normal comedian who decided to give it a try. After a couple of elections, I think in the third, he got elected as an independent. And it so happened that he became the person in the middle numbers who commanded the majority in Parliament, because on one side there was the socialist bloc and on the other the bourgeoisie bloc, and then there was Mr Haugaard in between.

What he realised was that to hold that position as a relatively ordinary person was extremely complicated. He hated his life in Parliament for that period. I suppose he needed the money, so he stayed for the full term. But then he decided never again, because it was too demanding in all ways for him. And I think we see the same happening in many other countries. So, I think the option must be there, commanded by the Constitutional Court and the Constitution, but there's no reason to think that it's a big issue. The political parties are the building blocks of modern political society and should be so, in my opinion.

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

This report has been published by the Inclusive Society Institute

The Inclusive Society Institute (ISI) is an autonomous and independent institution that functions independently from any other entity. It is founded for the purpose of supporting and further deepening multi-party democracy. The ISI’s work is motivated by its desire to achieve non-racialism, non-sexism, social justice and cohesion, economic development and equality in South Africa, through a value system that embodies the social and national democratic principles associated with a developmental state. It recognises that a well-functioning democracy requires well-functioning political formations that are suitably equipped and capacitated. It further acknowledges that South Africa is inextricably linked to the ever transforming and interdependent global world, which necessitates international and multilateral cooperation. As such, the ISI also seeks to achieve its ideals at a global level through cooperation with like-minded parties and organs of civil society who share its basic values. In South Africa, ISI’s ideological positioning is aligned with that of the current ruling party and others in broader society with similar ideals.

Phone: +27 (0) 21 201 1589



bottom of page