Popular Government Without the Will of the People

Populism sees representative government as intrinsically elitist, preferring to think about democracy in terms of the will of the people, expressed through devices such as referendums. However, this view is not one that can be made sense of and seeking to pursue the will of the people is dangerous to democracy. Citizen engagement is important in a representative democracy, but this is best conceived on a model of civil society organizations undertaking practical public deliberation. A philosophical model of deliberation leading to choice is introduced, and the argument that such a theory is itself elitist is considered but found wanting.


The Problem
Many of those who are not generally drawn to populism are nonetheless sympathetic to those criticisms of representative government that it has become elitist in character, with remote politicians and technocratic officials pursuing plans and policies that serve their own interests and their own vision of political order, a vision that is unresponsive to the will of the people, whether that be in the area of international trade, management of the public budget, support for local industry or open borders allowing for free migration. On this analysis, whatever their excesses, populist movements are articulating genuine grievances about political representation in modern democracies and there is a rational kernel to their claim that in place of the dominance of elites, government should rest on the will of the people.
The starting point for this paper is that the idea of 'the will of the people' is a highly misleading and sometimes dangerous way of thinking about the reform of democratic government. Of course, it is possible to use the term in a modest way simply to refer to the result of a legitimate election or referendum. In this sense, the term resembles the phrase 'the measure was the will of parliament' used by people to legitimate a tax, law or policy by reference to a valid parliamentary vote, even if the measure passed by only a narrow majority. In a similar way, we can say that a referendum yields the will of the people, but in this use we are simply using the term a shorthand way of saying that the result emerged from a popular vote conducted under free and fair conditions. In this modest sense, there is no challenge to the principles of representative government, when referendums take place in an agreed constitutional order.
By contrast, there is a fundamental challenge to the principle of representative government, if the claim is made that it is representative government, as such, that is insensitive to the will of the people. That challenge, however, rests upon a meaningless assertion, since it presupposes that it is possible to discover the will of the people as though it were a natural fact of the world, whereas the will of the people is not independent of the way in which it is measured in the way that natural facts are. Suppose you wish to measure the diagonal length of a field. You may do that directly by running a measuring tape corner to corner. But there may be a large patch of mud preventing you walking directly from corner to corner, requiring you instead to estimate the length of the diagonal by measuring two adjacent edges and then using Pythagoras's theorem to obtain the length of the third side. Using two different processes for ascertaining the same measurement does not affect the result: both should converge on the same number. By contrast, the result of an election or referendum will depend upon the rules under which the votes cast are counted. With a given distribution of political opinion in the electorate at large, an election conducted under a first-past-the-post electoral system will often yield a majority party in parliament (sometimes quite a large majority), whilst an election conducted under a system of proportional representation, with the same distribution of preferences, may, and often does, yield an outcome in which no party in parliament has an overall majority. There is no will of the people waiting to be discovered independently of the way in which it is measured.
It is one thing to make the logical mistake of treating the will of the people as an objective fact to be discovered. It is another to put that logical mistake to political use. David Hume (1739, p. 319) once said that whereas the errors in religion were dangerous, those in philosophy were only ridiculous. So it might be thought that logical mistakes about the will of the people should simply be added to the collection of the oddities of speech like Malapropisms or most of what is written in the reports of management consultants. However, as I have tried to show elsewhere (Weale 2018), misuse of the phrase 'the will of the people' is dangerous, posing a threat to the principles and practice of constitutional democracy and the political culture of pluralism on which it rests.
A telling example is taken from one specific incident in British politics in 2016. In the wake of the Brexit referendum in June of that year, the government of the day, led by the Prime Minister Theresa May, decided that it would initiate the Article 50 process of the Lisbon Treaty, by which a country could leave the European Union, not through a vote in parliament, but through executive action, known in the UK as an exercise of the Royal Prerogative. This decision was challenged at law by a private citizen, Gina Miller. The key decision was made by the High Court on 3 November 2016, which appealing to the principle that the 'subordination of the Crown (i.e. the executive government) to law is the foundation of the rule of law in the United Kingdom' (R (Miller) v. Secretary of State for Exiting the European Union 2016, §26), ruled that parliamentary approval was needed. On their front pages the following day the Brexit-supporting press attacked both the verdict and the judges who had made it. 'The judges versus the people' exclaimed the Daily Telegraph. 'Enemies of the people' pronounced the Daily Mail in chorus, adding for good measure that one of the judges was an 'openly gay' former Olympic fencer. (The Mail did not make clear whether it was being gay, being a fencer or being both that disqualified the judge from being a man of the people.) Worse still, no cabinet minister, including the Lord Chancellor, was prepared publicly to explain and defend the role of the courts in judging constitutional controversies, although the government itself had an interest in resolving the uncertainty about its constitutional powers on a question that it agreed was justiciable. Among much public disquiet, the incident even prompted a question in the House of Lords from Lord Lexden (a Conservative peer) asking the government which steps the Lord Chancellor was taking to protect the independence of the judiciary, a question followed by a number of other peers expressing their concerns along the same lines (Hansard 2016).
The failure of the UK government to defend the role of the judiciary in adjudicating constitutional matters was not surprising. At the Conservative Party conference in October 2016 Theresa May asserted that those who argued that initiating the process for leaving the EU could only be triggered after agreement in both Houses of Parliament were not standing up for democracy but trying to subvert it. For May the referendum of June 2016 had expressed the will of the people, and it was not the place of parliament to undermine that will. The will of the people trumps the authority of parliament, but does so by executive decree.
To assert that a referendum result or an election result embodies the will of the people such that constitutional process can be dispensed with is to fall prey to the errors of those forms of populism that hold that government policy should be decided by the people, directly mandating governments to do its will without the need for parliamentary debate or opposition. The terms 'populist' and 'populism' are widely used to characterize a wide variety of political movements and political parties of both the left and right ranging from European right-wing anti-immigrant parties in countries like France, Austria and Hungary to left-wing anti-austerity parties in Europe, like Podemos in Spain or Syriza in Greece or movements like Occupy Wall Street as well as Trump supporters in the US. Given this varying mix of political ideas and movements, it often seems as though the term populist is simply being used as one of abuse to describe the politics a speaker disagrees with. Yet, as Mudde and Katlwasser (2017) have shown, the term does pick out common elements in these otherwise very different movements, involving a thin ideology to which different causes are attached. Populist ideology sees the prevailing system of representative government as something that has been taken over by an elite. It looks to rectify this state of affairs through the direct involvement of ordinary people whose will should prevail in the making of policy. It holds that the will of the people alone should be the basis of government policy.
Yet, if the idea of the will of the people is as meaningless, with dangerous implications for the functioning of constitutional democracies, then there is an obvious problem. Democratic government differs from authoritarian government in making law and public policy responsive to the opinions of citizens. Democracy therefore requires, almost as a matter of definition, some form of systematic connection between the opinions of citizens and what governments do. Eighteenth century government in the UK was once characterised as 'despotism tempered by riot'. Democratic citizens should not have to riot in order to have their voices heard. But in the plethora of voices, some way has to be found of coming to a practical conclusion that has democratic authority, and it is this link that the idea of the will of the people is supposed to supply. So on the one side we have the dangers to constitutional democracy implied by the idea of the will of the people, and on the other side the claim that democratic government requires responsiveness to citizens' judgements and opinions. How are we to reconcile this seeming conflict? My suggestion, in part developed in this paper, is that we fill the space of public decision that some think should be occupied with the concept of the will of the people with the idea of practical public reasoning carried on under conditions of open competition among political ideas and compromise borne of pluralism.
The argument proceeds in two main stages. Firstly I examine the most obvious candidate for the claim that in place of representative government through processes of party competition, the will of the people requires the extensive use of referendums on matters of public policy. I seek to show that if referendums are proposed as a way to overcome the putative alienation implicit in representative government, the supposed cure is worse than the disease. Referendums easily lead to 'substitutionism', to take a phrase from Trotsky. I then go on to consider those alternative forms of representation, outside of voting, in which practical public reasoning takes the form of citizen engagement through representative civil society institutions or organizations articulating concerns in respect of specific areas of public policy. I then draw the argument back to the philosophical fallacy embodied in the idea of the will of people, presenting an alternative conceptualization of public choice as deliberation preceding choice, a conception of public choice that is distantly shadowed in citizen engagement through civil society organizations. I conclude by reflexively posing the question as to whether offering such an interpretation of democratic government is itself elitist.

Referendums and Substitutionism
Although referendums are often taken as the populist device par excellence revealing the will of the people, this is a mistake. One fundamental reason why this is so is that, in the absence of complete unanimity, there has to be some counting rule used to get to a result. However, there is no obviously right counting rule. Usually the rule is that a majority of those voting is a sufficient number to determine a result, but this rule can be called into question. Why not, for example, say that it should be a majority of those eligible to vote, even if some abstain, or some specified proportion of those eligible to vote? For example, in the referendum on devolution in Scotland in 1978, a majority was in favour but the legislation required 40% of those eligible to vote to be in favour, and in this respect the vote fell short. Had the same rule been in place for the UK referendum on Brexit, then it too would have failed, since support was less than 38% of eligible voters. We cannot simply say that the result of a referendum expresses the will of the people, since what the result is will depend upon how the votes of separate people are counted. The measuring instrument is not distinct from the phenomenon that is being measured.
The issues of principle involved in determining the correct counting rule are considerable. There are a number of reasons for requiring a threshold higher than simply a majority of those voting. In the first place, if a decision is being made about the constitutional status of citizens taken as a whole-a people-then some assurance is needed for those who transact with that people about the commitments that can be entered into. This is particularly important in the case of secession, where other political associations will need assurance that anyone who claims to speak on behalf of a people really does so, and this is not possible if the crucial vote rests upon less than, say, 40% of the people taken as a whole. Secondly, referendums on constitutional status can be impossible to reverse in anything other than the long term. To enter into long-term change with potentially large and uncertain consequences on what could be a narrow majority at any one point in time is an imprudent risk with respect to collective interests and assets. A threshold larger than a simple majority of those voting introduces a bias in favour of the status quo, but it is a justified bias, since it places the burden of proof on those who wish for the change to show that the adverse consequences of the decision do not outweigh the putative advantages.
There is a problem of recursive determination raised by these questions. Since there is no obvious rule for determining how votes in a referendum are to be counted, there needs to be a process of determining on what principles the counting rules are to be defined. To believe that the results of a referendum yield the will of the people, you have to believe that the counting rules are also what the people would will. Yet, since there is disagreement as to what those rules should be, you will need to determine what the rules are for determining the constitutional provisions that in turn determine how referendums are to be conducted. And to infer those constitutional provisions from some supposed will of the people in turn requires one to know the rules for how those constitutional provisions themselves are to be determined. And so on.
Behind this recursive problem is the simple fact that in a democracy there will be pluralism of viewpoints and opinion. Even if one says that it is not simply the fact that people disagree that is decisive-since some disagreements are prompted by prejudice, self-interest or ignorance-there is still the issue of reasonable pluralism emerging from the burdens of judgement. As Rawls says, reasonable pluralism is not something to be regretted but is the consequence of the operation of human reason operating under conditions of freedom and under the burdens of judgement (Rawls, 2005, xvi and 54-8). In a situation of reasonable pluralism, the democratic principle of political equality says that the voice of some cannot arbitrarily be substituted for the voice of all. What is needed are legitimate constitutional processes such that those whose voice is overridden have reason to accept. Otherwise there arises the danger of substitutionism.
This idea of substitutionism is taken from the social and political thought of Leon Trotsky, which offered an analysis of the centralizing and authoritarian tendencies of the Bolsheviks and the Leninist doctrine of the leading role of the party. As Trotsky put it: 'the party organization substitutes itself for the party, the Central Committee substitutes itself for the organization and, finally, a "dictator" substitutes himself for the Central Committee' (quoted in Knei-Paz, 1978, p. 199). An analogous form of substitutionism can be found when interpreting the results of a referendum. In the Brexit controversy, Theresa May was offering her own form of substitutionism, when she claimed that those who argued that initiating the process for leaving the EU could only be triggered after the agreement of Parliament were seeking to subvert democracy since the referendum of June 2016 had expressed the will of the people. The results of the referendum are substituted by the will of the people; the government substitutes itself for the will of the people; government by executive degree is substituted for government by elected representatives.
This problem of substitutionism is particularly acute in political systems in which there are no formalized constitutional provisions for referendums and in which a parliamentary majority can rest upon a minority of the electoral vote, both conditions being met in the UK. However, there are more general problems with relying upon referendums to define the basis of public policy that occur in any system of government. Consider the tasks of government that referendums cannot perform. Take, for example, what might seem a relatively simple case, namely the legalization of marijuana use. A referendum that legalizes marijuana use leaves open all sorts of questions. What age limits are to apply? Should there be regulation of outlets and product quality, and if so how should the task of regulation be organized? What are the implications for laws regarding driving offences or public order? Will use have implications for mental health services, and if so what needs to be done? Should advertising be allowed, and under what conditions? If the product is imported, what are the implications for customs and tariffs? Details of this sort matter for the operation and effects of the new policy, and they may well have significant implications for the way the policy works. Yet they are not ones that can be determined through a referendum, but necessarily fall to legislatures to decide. If this is true in the relatively uncontroversial case of marijuana use, it will be even more true in such morally controversial cases as abortion or physician-assisted dying. Even were one to allow that the idea of the will of the people made sense, any expression of that will in legislative practice would be so indeterminate in its practical implications that we could not say that the people governed.
The problems occur on a larger scale in the case of such matters as secession. A newly independent country needs to determine many policy questions. How is diplomatic representation to be organized? In which international agreements and organizations should the country participate, and what fees and contributions are appropriate? How are international debts to be settled? What currency should be used? How are disputes about land borders or territorial waters to be negotiated? What obligations does the new country have for reparations of historic wrongs? None of the above is an argument against the use of referendums in principle. In some cases, they may be essential, for example in cases of secession or withdrawal from a political association like the European Union. However, to say that the instrument of the referendum is essential is not to say that it is sufficient to determine legitimate public action. Many other acts of government have to occur besides. If one thinks that a distinction between the roles of government and the people is elitist as such, then referendums may increase the scope of elite behaviour by mandating governments and legislatures to interpret what are the practical effects of the referendum.

Civic Engagement Without the Will of the People
We have, in effect, arrived at Schumpeter's (1954) claim that democracy cannot mean that the people rule, but only that the people choose who is to rule them. If this is taken to mean that any feasible democracy is to be based on the principle of representation, then this statement is unexceptionable. It simply means that, outside of small-scale self-governing communities, democratic government needs to be based on the principle of representation. However, Schumpeter added two corollaries to his claim about the necessity of representation in democracy, both of which should be resisted. The first was that people drop down to a lower level of mental performance when they start to think about politics; those who in their professional life think rationally think only in an associative way when thinking about politics (Schumpeter 1954, p. 262). The second corollary was that citizens should avoid back-seat driving, leaving the elected leaders to govern (Schumpeter 1954, p. 295). Although the two propositions were clearly related in Schumpeter's account-one would not want infantile back-seat drivers-both provide poor guides to democratic practice, ignoring the extent to which civic engagement is educative. They also ignore the extent to which participation in flourishing civil society organizations plays a part in the well-functioning of democracy. In recent years there has been considerable interest in forms of political engagement that involve devices for fostering deliberative awareness, in particular deliberative polls and citizen assemblies (see, among many sources, Grönland et al. 2014;Smith 2005;Smith 2009; and Warren and Pearse 2008). However, one feature of these innovative methods of public engagement is that they often give a privileged position to the formation of the opinions of individuals on the model of public opinion derived from survey methods. Yet this is only one aspect of civic engagement. Of course citizens vote as individuals (we hope), and so being concerned with how as individuals they form their opinions is important. However, citizens also campaign and lobby with other like-minded individuals in forms of political engagement that are collective rather than individual, whether this be as patients, parents, church members, workers in a sector of the economy, members of unions, employers, nature lovers, residents of an area, members of minority groups, land owners, sports organizations or whatever.
There are a number of points to make about these forms of collective civic engagement. The first is that they rest as much on the principle of representative government as does the party system. For example, although some nature protection organizations are local, and rest on volunteering and the direct involvement of individuals, others are large-scale organizations, not just social movements. As organizations, they have a corporate life employing staff and occupying physical premises with governing bodies that are elected by their members. As such they make a representative claim, advocating policies that are supposed to embody their members' best understanding of what public policies are required. In some cases, as with the UK's Royal Society for the Protection of Birds, they conceive of their scope as international (birds not respecting national boundaries) and are involved in the working and development of international regimes of protection.
Secondly, from the point of view of individual citizens participating as members in these organizations, the political and policy concerns are issue specific. By contrast with political parties, which campaign across the full range of public policy, civil society organizations focus upon particular areas of interest and concern. Some may have an interest in more than one sector of public policy. Trade unions or trade associations, for example, will have relatively wide interests, including occupational health and safety regulation, tax and commercial policy, equal opportunities and diversity policies or land use regulation. However, even those civil society organizations that have such wide engagement with a range of public policies will not claim to represent their members, say, in respect of foreign and defence policy, justice and criminal policy or the charitable status of confessional organizations, since what is common to their members in respect of their areas of core concern may not be common to them outside of those areas.
Thirdly, in acting as representatives for the particular concerns of citizens, civil society organizations speak for individual citizens on matters that are of special concern to them. Those who care about nature protection may have no special interest in occupational health and safety; they want their nature organization to speak about species protection. Those who care about language teaching in schools may have no special interest in defence procurement; they want the language teaching association to press the case for more provision. Those who care about human rights may have no special interest in museum provision; they want their organization to lobby members of parliament. Those who care about lifeboat provision may have no special interest in state aid policy; they want a new lifeboat station. There is no reason in the theory of citizenship why every individual should care equally about everything. Indeed, given the complexity of social life, it is impossible for each person to care equally about everything. Moreover, when people have specialized interests, they are better able to contribute to the flourishing of the civic culture. If I am a public-spirited citizen, and I hear about a controversy about land use development, I need to hear from those who have a specialist understanding of natural ecology as well as those who understand the housing shortage; if I hear about a controversy in transport policy, I need to be able to hear from those who have an understanding of the possibilities of road and rail. This epistemic division of labour is the precondition for a flourishing civic culture.
Fourthly, some worry that the engagement of groups and associations in the democratic process will be a source of bias and distortion in the making of policy. Knowledge is power and specialist knowledge gives special power. Farmers will pursue the special interests of farmers; pharmaceutical companies will pursue the special interests of pharmaceutical companies; churches will act to protect their privileges; nature organizations will protect the wealthy who want no one in their back yard. This is the Madisonian problem of faction. But as Madison (1787) pointed out, the principle of free association makes faction ineradicable, and the solution is to put in place countervailing measures, including transparency in financing and reporting requirements, clear declarations of interest, obligations on the media to report impartially and the like. Moreover, if, in a democracy, one cannot listen to those with specialized knowledge on a topic, then one cannot listen to anybody. The danger, given the suspicious glance of a deflationary political culture, is a tendency for people to discount specialized knowledge, rather than gullibly accept it. There is a parallel here to what might be termed the Rawls-Humboldt principle, following Rawls's invocation of Humboldt's principle of human complementarity (Rawls 1999, §79, especially p. 463). Although no one person can realize the full range of human goods, each can appreciate and enjoy that range of human excellences as it is displayed in others. Moreover, there are some activities and forms of life that are not individually realizable, for example singing in a choir. From this point of view, the epistemic division of labour is not a burden to be overcome but the condition for enjoying a collective form of life, such that individuals come to value the complementary contribution to civic understanding that others make (compare Gauthier, 1986, p. 336). In the light of this principle, all can see the collective activity as producing civic understanding.
Fifthly, well-functioning democratic governance will necessarily involve processes of consultation and collaboration in the making of public policy. No government can know the information that is necessary for regulating all the sectors of public policy for which it is responsible. In order to regulate, there is a need for detailed knowledge of such matters as process and product standards or the qualifications necessary to undertake various forms of work. Typically these needs are met by consultative arrangements that maintain communication between civil society actors and governments. By the same token, elected representatives in parliament are dependent upon information from civil society actors in the crafting of legislation and in reviews of the operation of policy, sometimes formally supplied through evidence to parliamentary committees and sometimes informally supplied in private briefings. A general election can only be fought on a limited number of issues. Regular consultation and evidence gathering widens out the agenda of interest.
Notice that I have developed the argument above entirely within a traditional view of political participation, in which the domain the political is the authoritative allocation of values or who gets what, when, where and how and in which lobbying is a supplement to voting. In offering this characterization, the argument does not take sides on the question of how far the political domain extends from voting and lobbying, through such activities as street protests and consumer boycotts to raising funds for voluntary organizations, volunteering or life-styles expressions in diet and dress, matters on which there is extensive debate (see, for a useful review, Marsh and Akram 2015). Independently of these extended forms of political activity, the core of such activity involves a communicative intent to change the character of the collective organization of society. Since social organization is saturated with the influence of government, there are few activities that could not take on this communicative intent. In one sense, the concept of civic engagement advanced here is similar to that of Habermas's (1996, Chapt. 8) account of the role of civil society in the discursive formulation of collective organization, but without the commitment to a two track division between the public sphere and formal political process, and without the differentiation of function that Habermas ascribed to each. Democracy is deliberative, collective and plural.

In Place of the Will of the People
I have invoked David Hume's observation that, whereas the errors in religion were dangerous, those in philosophy were only ridiculous. I hope that the preceding argument offers some reasons to doubt Hume's claim, and to think that at least some philosophical errors can be politically dangerous. But what precisely is the philosophical error?
When there is an act of collective choice, say a vote in a parliament or a referendum, we are witnessing an event. It is tempting to think that such an event is an expression of some prior event in the consciousness of those voting, with each willing how he or she will vote. On this interpretation, what has happened is that voting registers a prior act of will. Rousseau (1762), as usual, captured this line of thought perfectly when he said that, if one finds oneself in a minority as a result of voting on the general will, then one should realize that one has simply mistaken that general will. The majority vote reveals that what one thought was the general will was not. Note that implicit in this view is a Cartesian concept of the will, in which the essence of the will is an inner occurrence in the mental life of persons, a view that Rousseau seems to have inherited through Malebranche (see Wokler 2001: 85). It is also a view of the will that explains why Rousseau thought that people could not be politically represented. If my will is an inner motion of my mind, then no one can have the same motion, since that motion will occur in their mind, not mine. (There is an interesting question here as to whether Rousseau has simply confused type with token, but I shall pass over that question in this discussion.) This was a view of the will that was effectively criticised by Ryle (1949, pp. 61-67) as involving the idea of the 'ghost in the machine'. Ryle offered a number of arguments against this view, but one of the most effective was the suggestion that it necessarily involved an infinite regress. If the exercise of the will were an inner event, it would have to be induced by another inner event, but that event would have to be induced by another event and so on ad infinitum. Populists who offer a view of democratic politics in which representative institutions repress the will of the people are in effect assuming that there is a popular democratic ghost in the political machine, to which the reply is that ghosts should be exorcised. Democratic choice is not the product of a collective will understood as the collective of events in the minds of citizens.
An alternative view of choice to that of the ghost in the machine is provided by Kenny's idea of the will as a way of designating an ensemble of intelligent action. As Kenny (1975, p. 26) put it, volitions are not motions of the mind, but states of the mind, states in which reasoned deliberation determines the choices that people make. On this account individual choices are the uncoerced affirmation of a reasoned choice. So the picture of agency that we are presented with in this conception is that of agents who deliberate in a practical way, considering reasons for action and who act according to the state of reasons that they are entertaining at any particular time. Thus, in a state in which I am thirsty and being thirst gives me a reason to get a drink, I will get a drink, but if I am in a state in which I am not thirsty, I will not get a drink. There is no need to add the thought, 'I now will myself to get a drink'.
The account I have offered of civic engagement through representative associations who articulate reasons for public action from their own point of view is consistent with this idea of choice through deliberation. What emerges in public action is the product of the interaction of reasons offered by different, and sometimes competing, groups. Of course many everyday processes-faulty media reporting, inadequate presentation of a case, missing data, prejudice against certain points of view or certain groups, sentimental attachments, unreasoning partisanship, vested interests, poor public record keeping, pervasive logical error in the form of inadequate heuristics and so on-mean that no actual democratic process comes anyway near a well-functioning public sphere, let alone an ideal speech situation. However, the model of political choice as emerging from processes of representative public reasoning, interacting with the formal sphere of parliamentary deliberation, points in the right direction of the reforms that democracy needs. Calling for more unmediated public participation points in the wrong direction.
In a symposium devoted towards identifying the philosophical characteristics of elite theory, it seems appropriate to consider whether the foregoing argument is simply another iteration in that theory. Is what I have presented an essentially elitist conception of politics, in which the elitism of reason is substituted for the expression of lived experience? Ever since the emergence of modern theories of deliberative democracy, this charge has been made. In what is still the best exposition of an anti-elitist critique of deliberative democracy Sanders (1997, pp. 347-348) makes the point that the endorsement of deliberative democracy among political theorists was not itself the product of a public deliberative process and that democratic theorists are a select group who are not representative of ordinary people. Privileging of deliberation over testimony may simply be the theory of the theory class, an attachment that needs unmasking.
To this worry can be added a further concern. In the form in which I have presented civic engagement through associations, the burdens of organization seem to be added to the obligations of deliberation. Yet the burdens of organization will be more easily discharged by some types of interests and points of view than others: more easily by concentrated producer interests than by dispersed consumer interests; more easily by mobile capital than rooted labour; more easily by those who can afford the best lawyers and publicists than those who have to write their own submissions to public inquiries; more easily by the better educated than the less well educated; more easily by those who enjoy the self-confidence of privilege than those who experience the anxieties of discrimination; in short more easily by those who already enjoy social advantage than those who are the least advantaged. And, in any case, as Michels (2001, p. 241) pointed out (writing of the pre-war German Social Democrats who adopted various measures against hierarchy) 'who says organization, says oligarchy'. At least with voting in a referendum, it is one person, one vote.
To these charges there is no simple answer, but there are a number of points that need to be made. In the first place the right to vote, including the ability to exercise that vote in practice, is not something that can be assumed. In political systems in which voter registration is not automatic, ensuring that the vote is obtained requires organization, as Stacey Abrams showed to spectacular effect in Georgia in both the 2020 Presidential election and the 2021 run-off elections for the Senate (Fedor 2021). Secondly, representative organization is the only way politically for members of a group to compensate for their individual powerlessness. A rent strike by one person leads to eviction; a rent strike by many persons may lead to better housing. What is true of some need not be true of all. Thirdly, testimony cannot take you very far. To improve conditions, circumstances and policies requires the development of actionable programmes. The German Social Democrats at the end of the nineteenth and beginning of the twentieth centuries used to distinguish between their minimum and their maximum programme. But they realized that to get to the maximum programme, you had first to go through the minimum programme. Fourthly, democratic activity is only sometimes characterisable as the interests of the many against the interests of the few. More often than not, it involves the interests of some of the many against the interests of others of the many: consumers and producers, the young and the old, the mobile and the settled, the childless and those with families, the secular and the religious and so on. In that situation, some explanation and justification is needed by one group with respect to another as to why public power should be used to certain purposes and not others.

Conclusion
In his history of government Finer (1997, p. 1025) points out that the most durable political invention in government coming from the middle ages are not the city republics but the representative assemblies of nascent European nations that first started acquiring authority in the thirteenth and fourteenth centuries. The challenge of political legitimacy faced by the descendants of those representative assemblies in modern party systems is a considerable one. To meet those challenges requires both responsive and responsible political representation, and the accountability through public reasoning that makes possible such representation. In securing those conditions, the idea of the will of the people should be consigned to history.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.