By Kate O’Regan
Justice of the Constitutional Court (1994 – 2009)


The Constitution of the United States of America is 7 000 words long. Nowhere does it mention political parties. And that was not because the question did not arise. James Madison, in The Federalist No 10, characterised “factions”, his reference to political parties, as a dangerous vice that tainted public administration. In his farewell address, George Washington too warned against the “baneful effects” of political parties and advised against their “insinuation” into American democracy.

Kate O'Regan

Speaking this week at a seminar on political parties, former Constitutional Court judge and Corruption Watch board member Kate O’Regan set out arguments for and against the requirement of full disclosure of political party funding.

Nevertheless, by the third presidential election in 1796, (George Washington stood unopposed for the first two), two political parties had emerged: the Federalist Party led by Alexander Hamilton and John Adams; and the Democratic Republican party led by James Madison and Thomas Jefferson. John Adams won the election in 1796 but in 1800 Thomas Jefferson won (well, after an intervention by Congress to resolve the status of Aaron Burr’s candidacy) and the Federalist Party never won another election. The election year of 1800 saw the first transfer of political power from one party to another in the US. In the more than 200 years since, it is accurate to say that two political parties (albeit not the same political parties) have dominated the politics of that country.

The US Constitution is not alone in the manner in which it treats political parties. Many democratic constitutions pay little attention to the role of political parties, despite the fact that it is now widely accepted that political parties play a crucial role in modern democracies. A significant exception is the German Constitution, which I shall discuss a little later.

Our Constitution does mention political parties, but it cannot be said to regulate them exhaustively. It starts firmly, in section one by asserting that a multi-party system of democratic government is a founding value of our Constitution, to ensure accountability, responsiveness and openness. Right up front, then, is the assertion that South Africa’s democracy will be a multi-party democracy. A second key provision is section 19 of the Bill of Rights, which entrenches the rights of citizens to form political parties and participate in their activities, including campaigning.

Thereafter, however, there are only a few provisions that refer to parties and they do so in what can perhaps best be described as a piecemeal fashion. These include the following: first, the rule that prohibits floor crossing by providing that members of Parliament and provincial legislatures will lose their seats if they cease to be a member of the political party on whose list they were elected; secondly, the provision that the leader of the largest opposition party in the National Assembly and provincial legislatures will be “the leader of the Opposition”; third, the Constitution provides that the rules of Parliament and the provincial legislatures may provide for financial and administrative assistance to political parties in proportion to their representation in the relevant chambers; fourthly, it provides that members of the security services may not prejudice or further the interests of a political party in the performance of their functions; and finally, it provides that to enhance multi-party democracy, national legislation must provide for the funding of political parties at both national and provincial level in an equitable and proportional basis.

There are thus no explicit rules regulating how political parties should function, whether their internal systems should be democratic, how they should appoint leaders and office bearers, how they should manage their relationship with their members, nor does the Constitution require auditing or disclosure of their finances. Although section six of the Public Funding of Represented Political Parties Act does require political parties to account for the moneys provided to them from the state purse.

As stated above, South Africa’s Constitution is not unusual in its relatively scant provision for political parties. What explains this relative ‘absence’ of regulation of political parties in democratic constitutions?

There are several possible explanations: first, there is arguably a lack of fit between the role political parties actually play in modern democracies, on the one hand, and current widely shared understandings of democratic theory – premised on the idea that democracy means government of the people, by the people and for the people, on the other. Where in this classical formulation of democracy is the space for the intermediary organisation that is a political party? So, thinking about political parties might unsettle our safe certainties about what democracy is, and may also give rise to intense contestation, so the response is to leave well alone.

A second related explanation for the failure to regulate parties more thoroughly may arise from the fact that it may seem to many that the way in which the law regulates political parties in most democracies, albeit somewhat odd, as I shall explain in a minute, seems to work and so on the “if it ain’t broke, don’t fix it” notion, we leave well alone.

Thirdly, it may be that the view is that the nature of political parties is transient and contingent, and that regulating them in a constitution will, in face of this evanescent quality, inevitably be unsuccessful and the project of constitutional regulation of political parties is therefore a flawed one.

A final consideration might be that those who have most to lose by rethinking the manner in which constitutions and law regulates political parties are those who would have to take steps to address the problem (that is senior members of political parties and members of legislatures) and therefore there is no incentive for reconsideration and change to happen.

In the rest of my talk, I am going to consider the role of political parties in modern democracies, and conclude that political theorists are right – we do need political parties in our democracy. Then I am going to describe briefly the general approach to the legal regulation of political parties in South Africa, which, as I shall say is the approach adopted in many Commonwealth countries. Thirdly, I am going to describe what is arguably the leading alternative constitutional model, the one adopted in Germany. Fourthly, I am going to consider the lessons that we might draw from the German model, particularly in relation to rules relating to the disclosure of party finances, and the requirement of internal party democracy.

I am not going to propose conclusions on these issues: but merely outline the arguments, for and against, to put you in a position to decide for yourselves. My reason for not providing any answers to the questions is not because I don’t think these are important questions. They are. I am not going to provide answers, firstly, because there is a case pending before the Constitutional Court concerning whether our Constitution imposes a duty on political parties to provide information on campaign funding, which is related to one of the issues I am going to discuss, and the outcome of which I do not want to be understood to be prejudging in any way. Although I should add that the question before the court is different to the question I am addressing in that it is concerned with the meaning of our constitutional text whereas I am looking at a question of constitutional design.

The second reason is that, at least at the level of constitutional design, these are difficult questions with many cross-cutting considerations, which means that reasonable disagreement is likely. It seems to me that leaving such questions open for further thought and debate might be the best way to foster civic engagement.

The role of political parties

Political parties, operating at their best, make democratic government possible in large, complex and heterogeneous societies. In the classical formulation by the political theorist Schattschneider – “political parties created democracy and … modern democracy is unthinkable save in terms of the parties. [Political] parties are not therefore merely appendages of modern government: they are in the centre of it and play a determinative and creative role in it.”

In analysing and understanding the role of political parties, it is important to realise that a successful political party operates at three levels: within the party itself, within the broader community and within the structures of government.

At organisational level within the party itself (again when operating optimally), a political party recruits, selects and trains potential political leaders so “socialising them into the norms and values of democratic governance and thereby contributing to political stability”; and the party analyses policy choices and determines an appropriate electoral platform.

Within the broader population, both at election time and at other times, political parties mobilise members of the public to participate in elections and other political processes, educate the broader public about democratic processes and the values that underpin them and articulate and explain the policy choices that are at issue.

Within government, political parties seek to implement their identified policy choices, and to ensure that the administration of government works. Because, as classically understood, the electorate assesses a governing party on the performance of government, the party normally has a direct interest in ensuring that government works efficiently in implementing its policy choices, but also in carrying out the tasks of government that may be outside the areas of electoral competition, but nevertheless basic for a stable and successful state.

Of course, this is all something of an idealistic conception of the work of political parties. An ideal that is often not achieved. Indeed, in a recent study of political parties in emerging democracies, Thomas Carothers described what he called “the standard lament” about political parties in new democracies across the developing world. The lament goes like this: Political parties are corrupt, self-interested organisations dominated by power-hungry elites who pursue their own interests or those of their wealthy backers, and not those of ordinary citizens; they do not stand for anything, their policies are vague and insubstantial; they spend too much time in meaningless squabbles with one another for political advantage rather than addressing real problems; they only become active at election time when they are seeking votes; and they are ill prepared for running the country and do a bad job at it.

There are no doubt some of you who will find some of Carothers’ “standard lament” to have some resonance in South Africa and I am not sure if it will be heartening or not to know that he studied political parties in a wide range of emerging democracies and found that elements of the “lament” are to be heard in all of them. Nevertheless, Carothers himself concluded, “problematic, aggravating and disappointing though they are, political parties are necessary, even inevitable. No workable form of democratic pluralism has been invented that operates without political parties.”

Accordingly, a fundamental starting point of my discussion is that political parties are an integral part of a modern constitutional democracy, including ours. Another premise of my discussion is that political parties do not always operate optimally. The consequence is that an important aspect of any study of constitutional design and constitutional practice is how to structure a system to ensure that political parties operate optimally.

In thinking about these questions, we must start by acknowledging that the role of political parties differs from society to society and also within one society at different periods of time. There are a range of factors that determine the manner in which political parties function, including the history and socio-economic circumstances of the country, the nature of its Constitution including whether it is a federal or unitary state, the number of political parties, and the ideological range of the political system. One of the key factors that determines the character and role of political parties is the electoral system. In South Africa, we have a closed list system of proportional representation (PR), which means that in effect, voters vote for parties (or for the lists of candidates produced by the parties) and not for individual candidates. The closed list PR electoral system is one of the factors that determines the role of political parties – an issue to which I shall return shortly.

Before doing so, I should like to describe the manner in which the law currently regulates political parties.

The current regulatory framework

Political parties are viewed as associations in South African law (and in most of the Commonwealth, where they are generally considered to be ‘voluntary associations’). In South Africa, most political parties in most circumstances are likely to be constituted as a universitas personarum, what can loosely be called a voluntary corporation, as opposed to a voluntary association.

As distinct from voluntary associations, voluntary corporations have legal personality separate from their members and accordingly the capacity to acquire rights and incur obligations separate from their members, and the capacity to sue or be sued in their own names. They have what lawyers call perpetual succession in that they continue to exist regardless of changes in their membership. Whether an organisation is a common-law corporation rather than a voluntary association is primarily determined by its constitution.

There are some very large and important associations in South Africa, some of them are voluntary corporations and others are merely voluntary associations that have no legal personality apart from their members. They include churches, trade unions, and that perennial of the South African law reports, the Jockey Club.

Generally our law has taken the view that associations (whether incorporated or not) have a duty to act fairly towards their members, at least in disciplinary proceedings conducted by in-house or domestic tribunals. The extent to which the provisions of the Bill of Rights bind these different associations in their relationship with their members, or with third parties, is less clear, although the provisions of the Promotion of Equality and Prevention of Unfair Discrimination Act, would require them not to discriminate unfairly against members or others.

Apart from the requirement to act fairly in disciplinary proceedings, and the obligation to avoid unfair discrimination, the primary legal mechanism that regulates the relationship of associations with their members is contractual and the terms of the contract between them will be found in their constitutions.

Although the Constitutional Court has stated that the constitutions of political parties must be consistent with section 19 of the Constitution (the right to form and participate in the activities of political parties), so far it has not had to provide guidance as to what is required of a constitution in order for it to conform to section 19.

Save for the requirement of conformity with section 19, there is no explicit regulation of what an association’s constitution should contain. It is a matter that is largely left to the association. When disputes arise, as they do from to time, between members and the association, the dispute will turn in the first place on an interpretation of the relevant constitution. As a recent commentator observed:

“The common law brings with it no understanding of power relationships or imbalance. Instead a party’s rules and internal workings may be open, inclusive and membership-driven, or entirely hierarchical and repressive of membership involvement.”

As this remark suggests, the common law is agnostic as to the content of the constitutions of voluntary corporations. As the Constitutional Court suggested in Ramakatsa, it may well be that section 19 of our Constitution will provide a value-based framework for assessing substantive provisions of the constitutions of political parties, but what they may be remains inchoate.

In addition to the common law regulating voluntary corporations, there are special electoral rules that govern political parties, insofar as they seek to context elections. To do so, political parties must register with the chief electoral officer (an official of the Independent Electoral Commission). They must have a name (and an abbreviation of the name of no more than eight letters), a constitution, a deed of foundation signed by 500 registered voters and a logo. And that is about that.

An alternative approach: the German model

As mentioned above, there are some modern constitutions that regulate political parties and the leading example in this regard is the German Constitution. The German Basic Law was adopted after World War II, in the wake of the horrors of the Nazi regime. In this respect, the German Constitution shares a similar focus to the South African Constitution. Both are committed to the principle of “never again”, intent on turning their back on the evils of the past, and trying to build a better future. Of course, the Nazi party was initially elected by popular vote, a fact that undoubtedly concerned the drafters of the Basic Law. With this history, it is not surprising that the German Constitution was the first to regulate political parties, by asserting that political Article 21(1) of the German Basic Law states:

“The political parties participate in the formation of the political will of the people. They may be freely established. Their internal organisation must conform to democratic principles. They must publicly account for their assets and of the sources and use of their funds as well as assets.”

Article 21(2) continues by providing that political parties that by reason of their aims or the behaviour of their adherents seek to impair or destroy German’s constitutional order or to endanger the existence of the Federal Republic of Germany are unconstitutional.

The German Basic Law thus confirms the right to form political parties, subject to the substantive limit that political parties may not endanger the constitutional order; and subjects political parties to two procedural requirements that can loosely be abbreviated as a requirement of internal democracy and a requirement of financial disclosure. The German model has been followed by several other constitutional democracies, including Spain and Portugal. The Portuguese Constitution goes further than article 21 of the German Basic Law in providing that political parties must be governed by principles of transparency, democratic organisation and participation of all their members.

Is there anything to be learnt from the German model?

In the remainder of my remarks, I am going to consider whether there is anything we could learn from the German model, in relation to the express constitutional provision for the disclosure of party finances, and secondly the explicit obligation of internal democracy on political parties. My purpose is to outline the key arguments for and against so that you can make your own assessment.

In considering these issues, we need to start by recognising that there are at least three reasons to be cautious about proposals to regulate political parties by law. The first concern is concerned with the issue of constitutional regulation rather than ordinary legislative regulation. A constitution is by its nature relatively fixed and inflexible. It is not easy to amend. Good constitutional design requires us to be careful to include in a constitution only what is necessary and appropriate. Issues that may require regular reconsideration or amendment are generally not suited for constitutional regulation. Moreover, where detailed rules are necessary, it is often better for those rules to be provided by legislation. A mixed approach is possible with a general principle in the constitution coupled with an express (or implicit) requirement that the issue be regulated by legislation in a more detailed fashion, as our Constitution does in the case of administrative justice and freedom of information, for example. In Germany, for example, there is detailed legislation that regulates the obligations imposed by Article 21 of the Basic Law.

A second concern that one must bear in mind in relation to any legal regulation of the affairs of political parties is the concern that such regulation should not inhibit the substantive policy debates that must happen in political parties. Democratic processes are vibrant, contested and pluralistic; and unwieldy constitutional or legal rules should not stifle them.

A third concern with any form of legal regulation is the need to avoid, where possible, the relocation of legitimate political contestation within political parties into courts. Some relocation of such disputes into courts is inevitable, but there are many political disputes that are not really suitable for adjudication. Getting the line between the ones that are suitable and those that are not, is not easy.

These caveats should influence our thinking about the regulation of political parties, both its content, and whether if we consider some regulation might be appropriate whether it would be fitting to include it in the Constitution or only in legislation, or in some mixed system. Accordingly, they all need to be borne in mind in mind in the discussion.

I will consider separately the questions of campaign finance and internal party democracy. My purpose is to outline the arguments that might support, or gainsay, the introduction of obligations in relation to each of these two obligations.

A duty to disclose party financial statements?

There are probably at least three reasons that would support the introduction of a duty to disclose financial statements by political parties. The first relates to the right to vote. In South Africa, political parties are the mechanisms through which citizens exercise their political choices. Because we have a closed list proportional representation electoral system, the right to vote is yoked to political parties: citizens vote for political parties on the ballot paper. Political parties, of course, construct the lists of candidates who will be elected by those votes, but citizens have no right to be involved in the compilation of lists: that is a matter solely for political parties. The right to vote is thus effectively limited to the choice of a political party.

A conscientious voter choosing how to exercise her right to vote would find it useful to have access to all the material information relevant to making the choice between political parties. Key to that, of course, will be the parties’ policy platforms, but in addition, a voter would find it helpful to know whether there are material considerations that might undermine the parties’ relative abilities to carry out their platforms.

For example, a conscientious voter might well want to know whether there are conflicts of interest between the parties’ expressed policy choices, and their organisational dependencies. If a party has expressed a policy preference in support of renewable energy, voters might find it relevant to know whether the party is dependent financially on a renewable energy company (which might have influenced its choice) or whether it is dependent on a fossil fuel company (which might limit its commitment to its expressed choice). Thinking about the relationship between voting and political parties suggests that provisions for disclosure of party funding may well be relevant to voting decisions.

The second argument in favour of a duty to disclose relates to building public confidence in the political system. Our own Parliament has recognised that public confidence in Parliament requires disclosure of their financial interests by members of Parliament. As has happened in many other open democracies, Parliament has adopted a code of conduct that imposes an obligation upon its members to disclose such interests. The code stipulates that its purpose is to “create public trust and confidence in public representatives and to protect the integrity of Parliament.” Accordingly, members of Parliament are required to register their financial interests annually, as well as the financial interests of their spouses, permanent companions and dependants. The code thus recognises the harm that may result where legislators have a conflict of interest between their work as the representatives of the South African electorate and their personal financial interests. In addition to a duty to disclose personal interests, the code requires legislators to resolve conflicts of interest in favour of the public interest. It would seem to follow that, like the disclosure of the financial information of legislators, the disclosure of political party financial information might also contribute to building public trust and confidence in our political system, particularly given the close relationship between the vote and political parties under our electoral system. This would thus be a second reason to support duties of disclosure.

A third reason might arise from the growing international legal assertion of the importance of transparency in relation to political party funds as a key mechanism for tackling corruption. Article 7(3) of the 2003 UN Convention Against Corruption, which we have ratified, provides that states parties should consider taking appropriate legislative measures “to enhance transparency in the funding of candidatures for elected public office and, where applicable, the funding of political parties”. Even more important is Article 10(b) of the 2003 AU Convention on the Prevention and Combating of Corruption, which we have also ratified, which stipulates that each state party shall adopt legislative and other measures to incorporate the principle of transparency into funding of political parties. The legal consequence of our ratifying of the African Union treaty, in particular, is that we bear an international law duty to adopt measures to require transparency of political party funding. Our international law obligations, of course, do not necessarily impose a duty on government in terms of national law to observe our international obligations. In this regard, however, it is worth recalling the Constitutional Court decision in Glenister in which a majority of the Court held that the Constitution is “the primal source for the duty of the government to fight corruption”.

Combating corruption is thus potentially a third beneficial effect of introducing an obligation upon political parties to disclose their financial information. In this regard, I should note that in the last ten years (between 2004 and 2014) South Africa has slipped more than twenty places in Transparency International’s Corruption Perceptions Index from 44th to 67th. This alarming slide needs to be arrested, and international experience suggests that a requirement of disclosure of political party finance might well assist in this.

Indeed, in India, as a direct result of concerns about corruption in politics, candidates for election are now required to provide an affidavit disclosing their educational and financial information, as well as whether they have a criminal record, or are facing criminal prosecution. In 2014, this information disclosed that 34% of India’s MPs were facing criminal charges. It is interesting that, in India too, in 2008 there was an important ruling under India’s Freedom of Information law that required parties to release their income and expenditure records publicly.

To summarise, the three key reasons that support a requirement to disclose campaign funding would thus be the following: first, it would assist voters to make decisions, secondly it might improve trust in the political system and, thirdly, it would assist in deterring and combating corruption. At least in relation to the last two, the result of the introduction of a duty to disclose finances might address some of the concerns that underpin the “standard lament” – especially in so far as political parties are seen as serving the interests of senior members of a political party and their wealthy backers.

So those are the arguments in favour. There are, of course, arguments against: and they are arguments often raised by political parties. The first is the argument that donors and political parties will evade transparency requirements if they are introduced. Of course, the risk of evasion of legal rules is ever-present whether one is speaking of, for example, a duty to pay tax, or obligations to observe speed limits, or prohibitions on trade in abalone, ivory or rhino horn. As these examples illustrate, the likelihood of evasion on its own is not ordinarily a sufficient reason for not adopting a beneficial rule. The question remains whether there are substantive reasons to introduce the rule.

There are two main substantive arguments that are raised to resist a duty of transparency in relation to the finances of political parties. The first relates to the question whether imposing duties of disclosure on political parties is improper because political parties are properly understood, private organisations.

An example of this proposition is to be found in an important 2010 set of guidelines for the regulation of political parties published by the Office for Security and Co-operation in Europe stated:

“striking the appropriate balance between state regulation of parties as public actors and respect for the fundamental rights of party members as private citizens, including their right to association, requires well-crafted and narrowly tailored legislation.”

The proposition is that because political parties are, in a sense, consortia that result from the exercise of ‘private’ political freedoms, legislation to regulate them must be narrowly tailored. Similar concerns are formulated by many political theorists: an example of such thinking informs the proposition that that there is a tendency to consider political parties to be “legitimate objects of state regulation to a degree far exceeding what would normally be acceptable for private associations in a liberal society”.

Are commentators right to assert that political parties are ‘private’ in some senses which should make us chary of regulating them? I am not sure. There is a conspicuous contrast between the manner in which the law regulates voluntary corporations and the manner in which it regulates companies. The new Companies Act has 225 sections and runs to over a hundred densely packed pages of Butterworths Consolidated Statutes. It provides for the formation, administration and dissolution of companies, takeovers and mergers, business rescue, shareholders rights, compulsory meetings, annual financial statements and the like. There is simply no comparable regulation of voluntary corporations and political parties.

Corporations are, of course, the archetypal private institution. Lawyers grapple with the question where to draw the line between private and public and often fall back on inarticulate premises that when exposed lack coherence and principle. I would suggest that any general principle that political parties should not be regulated because they are in some sense “private” does not on its own constitute a satisfactory reason for our refusal to regulate political parties more closely. It seems to me that the question requires a functional analysis of the role of political parties to determine what regulation is appropriate. In particular, what forms of regulation would be likely to produce the optimal performance of political parties, given their role and importance in any constitutional democracy.

In assessing the private nature of political parties, it seems to me that there may a continuum of privacy, with some extremely private organisations on one end, and some semi-public organisations on the other. On one end might be a book club or a gardening society or a Sunday afternoon cricket club. On the other might be organisations like churches, the Jockey Club and political parties. As we have seen, political parties operate not only in the sphere of their internal affairs, but in the public sphere and in government. The question then remains whether we should require them to disclose their finances.

The second substantive argument relates to the risk of harassment of donors of minority parties, in particular by governments and ruling parties, with the consequence that minority parties might lose significant sources of funding. These arguments are raised in many democracies and they cannot be dismissed out of hand.

Before turning from the subject of disclosure, I should add that disclosure is not the only mechanism that is used in modern democracies to regulate campaign finance. Some democracies impose both bans and caps on campaign finance contributions, as well as caps on campaign spending. An example of a very common contribution ban is a ban on donations from foreign sources. 68% of states have an absolute ban on foreign entities providing funding to political parties, while 51% of countries ban foreigners funding individual candidates. Anonymous donations are also banned by more than half of all countries. South Africa does not ban either donations from foreign sources or anonymous sources.

Contribution and spending caps are another common form of campaign finance regulation. In the UK, there is an absolute cap on campaign spending. In the US, of course, there is the complex situation where there are caps on both contributions and spending by candidates, but these caps do not apply to political action committees (PACs) and to the independent expenditure-only committees, the so-called SuperPACs, who are independent of candidates. Since 2010, as a direct result of the Citizens United v Federal Election Commission decision of the US Supreme Court, which held that that corporations and trade unions may spend unlimited amounts of money to support or denounce individual candidates in elections, campaign spending by PACs and SuperPACs has soared. In the 2012 presidential election, non-party outside spending tripled to in excess of $1-billion: of that SuperPACs spent $600-million. SuperPACs raise the bulk of their money from a small group of wealthy donors and corporations who make very large contributions.

There is no doubt that some of these additional mechanisms for the regulation of campaign finance are worth considering. Mechanisms like bans on foreign donors, as well as contribution and spending caps, may well enhance the integrity of our political system.

But the first question is whether we should introduce a duty of disclosure. I have outlined the arguments for and against, and will let you draw your own conclusions. I now turn to the question of internal party democracy.

Internal party democracy

As mentioned above, the German Constitution requires the internal organisation of political parties to conform to democratic principles, and several other European countries have followed suit by including requirements of internal party democracy within their constitutions. In addition to countries that have constitutional requirements of intra-party democracy, there are also a significant number that impose elements of intra-party democracy legislatively.

In considering the desirability of introducing a principle of internal party democracy into our constitutional order, we should bear in mind that there are already some legal requirements imposed upon the internal regulation of political parties, as I have explained above. In summary, political parties must act lawfully and consistently with their own constitutions; those constitutions must conform to section 19 of the Constitution, even though what such conformity requires remains uncertain; and political parties must act lawfully and fairly in relation to disciplinary proceedings against their members.

Although the principle that political parties should conduct their internal affairs democratically has intuitive appeal, consideration of how it would work in practice uncovers a range of intractably difficult questions. What do we mean by internal party democracy? Is it only party members who may participate in political party decision-making? Is there an inevitable tension between internal party democracy and democracy as a national project? How should we respond to the claim that internal democratic processes in political parties may empower a relatively small and unrepresentative group of citizens to make decisions that will be of national importance? How do we balance considerations of internal democracy with other important principles, like gender, race or regional representivity? All of these are difficult questions that a simple constitutional requirement of party political internal democracy does not answer.

The difficulties that may be encountered are highlighted by the current Labour Party leadership campaign in the UK. How would we assess that process against a requirement of internal party democracy? Is it inappropriate that thousands of people have joined the Labour Party in order to vote in the leadership election? Has the dramatic increase in eligible voters enhanced or weakened internal party democracy? It may be that these are issues best left to the political process.

The difficulties are also illustrated by the jurisprudence that has emerged in the US around primaries. The introduction of compulsory primary elections had its origins in the attempts at the turn of the twentieth century to eradicate pork-barrel politics and to weaken the hold on power of “corrupt, local boss-led party machines”. In a short period of time, more than 40 states had enacted compulsory primary requirements for the selection of candidates for political office. Primaries can take different forms: closed primaries (open only to members of political parties); open primaries (non-members may vote but only for one selected party); and blanket primaries (in which non-members may vote in the primaries of all political parties). State laws compelling primaries appear not to have been challenged at the time, and are now widely accepted as legitimate imposition of forms of intra-party democracy.

The doctrinal difficulties were exposed, however, in the decision of the Supreme Court in 2000 California Democratic Party v Jones, in which the adoption of Proposition 198 by nearly 60% of voters in California had resulted in the introduction of a compulsory blanket primary system. The new law permitted voters to participate in the selection of candidates for office in all political parties, without having any prior institutional commitment to the parties. By a majority of 7:2, the Court struck the requirement down on the basis that:

“In no area is the political association’s right to exclude more important than in the process of selecting its nominee. That process often determines the party’s positions on the most significant public policy issues of the day, and even when those positions are predetermined it is the nominee who becomes the party’s ambassador to the general electorate in winning it over the party’s views.”

Can you hear the echoes of the British Labour Party debate? And one might well instinctively agree, but we should pause for there is room for reasonable disagreement too. And that disagreement runs deep. A blanket primary system allows ordinary voters, who have no party affiliation, to influence who should be nominated as leaders rather than just party members. Indeed, the arguments for Proposition 198 had stated that “the current closed primary system produces candidates at the ideological extreme who spend more time fighting with each other than addressing the state’s most pressing problems.” Nevertheless, both the Democrats and the Republicans opposed Proposition 198. Is it so clear that the Court was right? What is democracy after all, but the right of the people to participate in the choice of their leaders? Why should we privilege the party members in that choice over non-party members? And just when you think you have a different view, what of the effect of blanket primaries on small parties that is seeking to challenge the two dominant parties? The effect of Proposition 198 will in all likelihood be that supporters of the majority parties will select the leaders of small parties.

Now the terrain of the intractably contested is terrain commonly inhabited by constitutions, and constitutional courts. So the likelihood of intractable contestation is not on its own sufficient for us to conclude that the matter is not fit for constitutional regulation.

But we should also bear in mind that there are costs to legalising political processes, which as I mentioned above, are often contested and pluralistic. At times, it is better to leave contestation to be managed in the political process. After all, that is at root one of the reasons why we have a democracy. Not all problems can be determined by the application of neutral legal principles.

There are thus arguments that run in both directions here, and again I leave it to your own evaluation.


The one conclusion I shall draw tonight is that political parties are a necessary part of modern constitutional democracies. It is also indisputable that they do not always operate optimally. How they should be regulated raises important and difficult questions of principle and constitutional design. I hope that I have at least touched on the key forms of regulation that might be considered, as well as the arguments for and against, and I leave it to you to discuss these issues further tomorrow. A democracy, after all, depends for its success on civic engagement on matters of national importance. Civic engagement is precisely what this conference is all about, and for that, we should welcome it.