Creator: Ong Cidade/Arquivo
A mostly ignored tool could improve democracy within legislative bodies and political organizations. Failures of democracy contribute to the current reactionary moment, and play a part in the rise of fascism and authoritarianism.
Many, but not all, of the policies that helped create the current conjuncture are deeply unpopular. In the United States, for example, majorities support strict gun safety rules (Edwards-Levy 2023), far higher taxes on the rich (Newport 2023), and stronger policies favoring renewable energy and efficiency (Tyson 2022; Roberts, 2018, Gallup 2023). Most voters also oppose the Republican “anti-woke” crusade (Page 2023), anti-trans bills (Marist Poll 2023), and abortion bans (Pew Research Center 2022). If majority preference drove policy starting today, the United States would still have long way to go to solve the crises it faces; democracy is not magic. But genuine democracy would make this nation a far better place than it is, and a more favorable terrain for struggle. In its absence, unpopular policy within a nominally democratic society is one of many drivers of people towards fascism, especially when that society provides tempting targets for scapegoating in the form of oppressed groups.
Similarly, flawed democracy within left movements and organizations contribute to their current weakness. For instance, members of groups and movements can become disillusioned when they discover that their nominal democracy conceals control by a minority with resources and time to prepare for and participate in meetings. The tendency of such minorities to gradually shift the focus of such groups in directions not corresponding to member priorities can also disappoint participants.
A tool which increased both citizen control of legislatures and member control of democratic voluntary organizations could be as revitalizing to grassroots groups and movements as to states. This would be especially true if this tool allowed an increase in democratic control without indefinitely prolonging debate, or piling meeting upon meeting.
Ironically, one of the most undemocratic of institutions, for-profit corporations, invented that tool, the voting proxy, at least as early as 1710 (Axe 1942, 41). A voting shareholder, unable to participate in a meeting, usually may delegate their voting proxy to someone who will attend. That attendee, in addition to casting their own votes, can also cast the votes of those whose proxies they hold. Shareholders also have the right to revoke proxies. There is no reason someone who may rightfully vote in a democratic organization’s meetings should not be allowed to similarly delegate voting proxies to another voting member. This is not a new idea. Charles Dodgson (Lewis Carroll, author of Alice in Wonderland) first proposed it in 1884, calling it liquid democracy (Dodgson,1884). Although proxies themselves are a simple concept, how they can transform democracy is not.
Proxies in Representative Democracy
Focusing on representative forms of democracy will help explain many of the implications of proxies. One well known flaw in representative democracy is that politicians break campaign promises (Shalom 2021, 7-14). Some democracies do have recall processes, but these are often limited to criminal or extreme misbehavior. Regardless of formal limits, gathering signatures, and winning a recall election is too burdensome a process to provide routine accountability for broken campaign promises. In a legislature consisting of proxy holders, each lawmaker would act as proxy-holder for a number of citizens, and would cast that number of votes within the legislative body. If a legislator acted contrary to the will of those citizens, voters could simply shift their proxies to a different legislator. The ability to revoke and redelegate proxies serves as a low-barrier recall.
This raises a question. If a legislative election consists of voters delegating proxies to proxy-holder candidates, what prevents this resulting in legislative bodies of tens of thousands or even millions of members? The answer: proxy-based legislatures, like conventional ones, would consist of a fixed number of lawmakers. A 101-member legislature, for example, would only seat the 101 candidates receiving the most proxies. Candidates placing below the 101st ranking, in number of proxies received, could redelegate their proxies through a negotiated process. Voters who disapproved of the result could, in turn, redelegate their proxies, leading to a new round of negotiations.
Legislative elections would then consist of several steps. Initially, voters delegate proxies, by mail or in voting booths; then a negotiation process places proxies of candidates who did not win seats. After that, voters dissatisfied with the results revoke and redelegate their proxies; a final round of negotiations follow.
The ability of proxy holders to revoke and redelegate proxies on the individual voter level not only provides more accountability than recalls in conventional representative bodies, it also resolves a contradiction between the democratic accountability recalls provide and democratic representation.
Election of representatives can easily result in a minority of voters electing a majority of representatives due to gerrymandering. One example, Shalom gives, “…in 2012, Democrats received 1.4 million more votes than Republicans for the House of Representatives, but Republicans won more seats in the House of Representatives, 234 to 201” (2021, 9).
One solution, seldom used in the United States, though often adopted by parliaments, is called Proportional Representation (PR). In PR, seats are allocated in rough proportion to votes cast. All forms of PR count votes across multiple legislative seats, then allocate those seats according to formulas which differ between types. These formulas ensure that a party winning 30 percent of the vote in a legislative election wins roughly 30 percent of the seats. How roughly depends upon the specific implementation. But conventional forms of PR make conventional recalls, as weak a form of accountability as they are, somewhere between impractical and impossible. Allocation of votes across multiple seats prevent fairly determining who should vote in a recall election for a single member. Fairness would require re-running the whole election, or at least a good part of it.
Proxies reconcile accountability and representation. While providing a low-barrier form of recall, they also constitute a type of PR. Unlike conventional forms of PR, where seats are allocated in rough proportion to votes, representatives cast the exact number of votes within the legislature that they received. If a member of a legislature received 30,000 votes, either directly or via redelegation from other candidates, that member holds 30,000 proxies and can cast 30,000 votes for or against bills and motions. If 60 percent of voters in a legislative election voted for strong supporters of gun safety regulation, then 60 percent of the voting proxies in that legislature will be held by strong supporters of gun safety regulation. Unlike conventional PR, if legislators try to back off on that commitment, then voters can simply revoke their proxies, and shift their support to a different member of that body—in extreme cases even to one who is not currently seated.
Proxies mitigate another problem in representative democracy: the “lesser evil” paradox. Suppose a voter cares about ten issues. There are 1,024 (210=1,024) possible combinations of pro and con on those ten issues. That number rises drastically if more views are possible on those issues than simply pro and con, or if the voter cares about more than ten issues. Thus, the vast majority of voters will end up supporting someone with whom they seriously disagree on some policies. The simple mathematics of choosing from the limited selection of candidates available in a conventional representative democracy make this almost unavoidable, aside from other major contributing factors.
The lesser evil paradox, though serious, is not quite as structurally implacable as the mathematics imply. Policy positions relate to one another. Someone who supports gun safety likely supports worker and consumer safety rules as well, and probably protection for the environment. Social, economic, and foreign policy views don’t always connect to one other, though such disconnects happen less often than pundits sometimes pretend. Policy preferences mostly form around bundles of connected issues. Looked at this way, the overwhelming majority of the population probably hold no more than four bundles of policy preferences,* with a spectrum of perhaps three positions (left, center, right) on each one. That yields around eighty-one (34=81) possible policy combinations, still too many for conventional representative systems to reflect. It is not just a matter of comparing number of seats to eighty-one possible combinations. In district-based systems, views held by a percentage of the population that in theory could win a seat or two would be divided among too many districts to elect any members. Systems not using districts, party-list forms of PR, would exclude those same views via threshold requirements—minimum percentages parties must obtain to win seats. Proxies, however, would allow the reflection of eighty-one different viewpoints and even a bit more.
Take the earlier example of a 101-member parliament, quite small for a national or state/provincial level legislature, but larger than required to elect more than eighty-one points of view. Most importantly, quite small portions of the population will be represented. It is quite possible that one or two candidates will be popular enough to win 20 percent of the vote. It is almost certain that one or two will win 5 percent. Others will win 3 percent or 2 percent or 1.5 percent of the vote. And others will win that 1 percent. Thus, it is near certain that the member winning election with the lowest number of votes will represent a fraction of a percent of the voters, probably quite a small fraction. That becomes even more certain with larger legislative bodies.
This does not quite eliminate the lesser evil paradox, but it reduces it to something bearable. To be sure, some may deceive themselves that they decide issues case-by-case but, as documented in the note, four bundles is generous. Few people hold views that map beyond two dimensions, let alone four. However, three choices per bundle is probably too small. Views on bundles lie upon a spectrum, rather than dividing neatly into left, right, and center. Still that division does capture the ends and middle of the spectrum. Thus, a large majority will find candidates who represent their views exactly, and most of the remainder will find candidates close enough. And even views shared by too small a portion of the population to achieve direct representation will have a chance to extract concessions during the negotiation process between failed and successful candidates. That is important in legislative bodies such as city councils, which often are composed of fewer than a dozen members.; though at this level there are probably fewer policy “bundles” possible, and the 3-viewpoint spectrum is probably closer to reality than at the state or national level.
Does this advantage require eliminating local districts? No, but it necessitates letting voters sometimes bypass them. Ballots can list only local candidates from within a voter’s district. But they would allow write-in votes for candidates in any district. The premise: in (say) a national election, a voter will prefer a representative who represents their views and also knows their local conditions; but they will prefer a distant national representative who supports their national policy priorities to a local who opposes them.
One important warning about proxies. Since, within democracy, they serve as a form of ballot, they should be treated like ballots. If mail is used rather than voting stations then, like vote-by-mail ballots, blank proxies should only be sent directly to the voters (or official guardians and caretakers where appropriate). No one should be able to print out blank proxies and set up tables in the public square or go door to door and pressure people to fill out a proxy for them immediately, just as nobody can (in most cases) do that now with blank mail-in ballots.
Participatory Democracy and Proxies
Arguably, the proxy system described so far gives citizens enough control to border on a participatory democracy. It is at least delegatory. However, it does not create institutions where citizens meet in small groups to engage in discussion and democratic decision-making face to face. Advocates of participatory democracy consider it essential that actual policy making and selection of delegates occur at such meetings. They argue that face-to-face deliberation and decision-making acts as education, and trains people to think politically, in a way that simply choosing candidates on a ballot does not.
What would this look like in practice? Steve Shalom proposes what he calls “nested councils” (Shalom 2021, 16-24). These combine deliberative direct and delegatory self-governance. Councils making decisions that affect smaller geographies elect recallable delegates to councils managing larger areas. Direct democracy occurs at the smallest local level where people deliberate and vote on issues critical to their lives. Neighborhood assemblies discuss and debate issues and allow modification of proposals in response to discussion before they are voted upon. (To increase concision, this article will refer to such local councils as “neighborhood assemblies,” and larger councils as “city,” “county,” “state,” and “national” councils. Neighborhood assemblies are far from the only possible basis for a nested council system)
Neighborhood assemblies practice delegatory democracy by electing delegates to city councils to resolve issues which affect multiple neighborhoods. These city councils, in turn, elect delegates to county councils, and so on. All delegates are subject to recall by the councils that elected them. Participatory democracy has seldom been practiced beyond the city scale. Shalom’s proposal, unlike many nested council proposals, does not call for delegates to be mandated how they must vote—for good reason. Mandates would prevent the next level of council from deliberating on mandated policy. In addition, legislators in California proposed about 2,000 laws during 2022; the governor ultimately signed about 1,000 of them, including a state budget (Kamal 2023). The U.S. Congress typically passes between 2 and 3 million words of legislation each year (GovTrack 2023). For obvious reasons, extensive staff do most policy work. Basing a political system larger than the municipal level on voter mandates would face serious information constraints. Note that the same constraints, plus the need for deliberation, means that democratic proxies, unlike many corporate ones, should not include the ability to mandate how proxy holders must vote.
Most decision making under nested councils is still delegated. The only decisions made via direct participation are neighborhood issues and selection of delegates to city councils. All other decisions, including the selection of delegates to county, state, and national legislatures are made indirectly, sometimes very indirectly. The idea seems to be that at each level of decision making, council members will personally know those they are delegating further decision making to. One problem with relying on this is that personal acquaintance with a candidate can cloud judgement as much as it clarifies; “what a wonderful person” can trump disagreeing with them on a multitude of issues, when the multitude of issues is what is at stake. If people personally knew all their delegates, direct and indirect, the point would at least be arguable. But it is hard to see a huge advantage to a citizen if their city council member personally knows their county council delegate, nor if their county council delegate knows their member in the state council.
Another problem: the lesser evil paradox may be greater in a nested council system than in conventional representative democracies. Citizen input into policies at all levels of legislative governance are packed into the choice of city council delegates from the small set of people attending a single neighborhood meeting. What are the chances that most attendees at a neighborhood meeting will find a candidate from that tiny universe who matches their policy preferences closely at all levels from municipal to national? The extreme indirection also presents an information problem. In the party game “Telephone,” if the starting phrase “San Andreas Fault” is passed from person to person, the phrase at the end of the chain could be something along the lines of “Sam and Dre’s fall.”
In addition, remember how in conventional legislatures without PR, a minority of voters can elect a majority of representatives? The same type of gerrymandering applies to this council system. Shalom allows plebiscites to overrule council decisions if they are made by a narrow majority or if councils from smaller geographies call for a direct vote. That would help if uneven distribution of opinions happened for only a few issues. If that gerrymandering reflects fundamental splits affecting views on large numbers of issues, then nested councils would run into the same constraints as plebiscitary systems; too many referendums for voters to keep track of.
In spite of these flaws, nested council systems hold one important advantage. At least policy decisions at the neighborhood level are made through a deliberative process by those choosing to attend the meetings. Proxies can allow modification of the nested council proposal to keep its advantages, while eliminating some of the disadvantages and compensating for others. If democracy must rely on extreme indirection, let that indirection take the form of passing along bundles of proxies. In a proxy-based version of nested councils, a vote for a city council member also passes along the voter’s proxies for the other councils: county, state, and national. City council members select county council members by redelegating the county council-level proxies they received from individual voters, and also pass along proxies for the other councils to the county council. County council members vote for state council members by using the individual state-level proxies passed to them, and pass along individual national-level proxies to the state-level members they elect. As a final step in the process, state council members use their bundles of national council proxies to select national council members.
As in the representative process discussed in the previous section, though default choices will be within districts, voters can elect candidates from any district in the body they are voting for. The city council ballots that voters cast at neighborhood assemblies will contain only candidates running within their neighborhood. But if nobody in their neighborhood satisfies them, they may write in someone running in a different neighborhood. The same applies for county, state, and national councils.
This leaves the problem of extreme indirection, the game of telephone. But the process of passing along individual proxies suggests a solution. In addition to (for example) a city council having the right to recall a county level delegate, individuals could also revoke and redelegate proxies. If a citizen objects to the policies or character of their county or state or national legislative proxy-holder, they can revoke that proxy directly and redelegate it. The use of indirect proxies for initial selection, and both indirect and direct redelegation of proxies thereafter gives nested councils the same advantages proxies provide in representative democracy; PR combined with low barrier recall for true accountability.
I think though, the nested council system complicates delegation unnecessarily. The main advantages of nested councils over the delegatory system I outlined in the first section are neighborhood assemblies. Let neighborhood assemblies decide neighborhood level issues and delegate proxies to city council members. Let citizens delegate proxies to all other levels of government directly, as in the delegatory system outlined in the representative democracy section.
Hopefully, the advantages of proxies in choosing delegates are clear at this point. But the question arises, should participatory democracy allow citizens to delegate proxies to neighborhood assemblies, where direct deliberative democracy takes place? The performance of nested council systems, which have been tried on a small scale, suggests the answer is yes.
Small-scale participatory democracy has success stories. Porto Alegre’s participatory budget process greatly reduced poverty and corruption. New England, with its town meetings, was a center of the abolitionist and women’s suffrage movements, and avoided some (though by no means all, or even a majority) of the recent craziness rampant in other parts of the United States. The Zapatistas, surrounded by a hostile Mexican government and pressured by a hostile United States, still halved maternal and infant mortality in areas they control, and made similar gains in literacy. But all of these examples have failed miserably to reach one key goal of a participatory democracy. No attempt at participatory democracy has ever succeeded in what the name implies is its primary goal: – having the majority of citizens voluntarily participate in governing themselves over a long period of time.
The Porto Alegre case, which so many find inspirational, at its peak consisted of 17,200 citizens (Sintomer 2012) out of a population of 3.6 million (Macrotrends 2023), less than one half of one percent of the population. The subset of New England open town meetings which allow direct voting on issues and modification of proposals in response to discussion average turnout rates from 2 percent (Rolheiser 2020, 32) to 20 percent (Preer 1986). No more than 35 percent of the population of Venezuela ever took part in neighborhood citizen assemblies under Chavez (Hawkins 2010, 42), before it became a dictatorship under Maduro. Those assemblies often had trouble reaching their quorum (Azzellini 2016, 209) of 30 percent.
The Zapatistas in Chiapas are an exception, with very high turnout rates. That is, in part, because attendance, at least until fall 2023, was mandatory (for good, culture specific, feminist reasons). In most contexts, “join our liberatory project; attendance at monthly meetings will be mandatory” does not seem like a good selling point for participatory democracy. In the absence of such forced attendance, only proxies ensure the majority gets a voice and a vote at such meetings.
Immediate and Short-Term Implications
The value of proxies transcends legislatures, as they can be useful for democratically-run voluntary organizations as well. A large labor union, or a good-sized socialist organization such as Democratic Socialists of America, would probably need proxies if they wished to run them via participatory democracy. Almost certainly, a minority with time and resources to show up at critical junctures would dominate any attempt by that size organization to implement participatory democracy without them. The necessity of proxies probably applies not only to large and medium groups, but even to tiny groups of as few as five dozen members, if they wish to avoid minority rule. Usefulness to groups below that size would depend upon goals, ambition, structure, resources, and circumstances.
Two last points:
Proxies are an important tool to help democracies in name become democracies in practice. But proxies also respond to a criticism sometimes made of participatory democracy as a goal: it requires too many meetings.
Because proxies make member day-to-day influence easier and more practical, they bring an important and difficult question to the forefront. Who should have voting rights? Geographical political bodies and labor unions both answer that question in well-known ways. In a political body governing a geographic region, at minimum every citizen within that region over a certain age resident for a minimum period of time, and probably non-citizens as well, should have the right to vote. Similarly, anyone hired into a bargaining unit, or participating in a union drive under certain conditions specified by the union, should (as now) be able to vote. That is because membership in both these cases connects to social roles, and anyone occupying those roles has a moral right to a say in such organizations. But who should constitute the voting membership of a voluntary group where membership does not tie so closely to well defined social roles?
One practice that goes back at least to the 1960s, and is still found in organizations with roots in that era, is clearly wrong. No one should get a vote in a voluntary organization five minutes after they show up in a meeting for the first time. While I have no comprehensive answer, I can outline some general principles.
Anyone joining a truly voluntary organization as a voting member should:
share the principles, goals, and ambitions of that organization on a general level, with plenty of room for disagreement on non-core principles and upon details.
have a stake in that organization and have contributed to it in some way.
have been part of that organization long enough to understand it in a practical sense.
If these seem unsatisfactorily general, it is because they are. Specifics are best decided on a case-by-case basis. Especially, striking a balance in the first principle is difficult. Steering between the sectarianism of “democratic centralism” or the “value-based” narrowness of some Green groups, and the silly putty vulnerability of Occupy requires a difficult balance. My personal experience of groups too slack in voting requirements include the following:
An organization (admittedly one that was already dying) taken over by hostile forces, systematically looted and dismantled.
The frustrating of an organization’s ability to accomplish anything by participants who were there for the fun of debate rather than to advance the group’s goals.
An organization destroyed by a faction of people who joined it shortly after formation in order to change the organizational goal, who opposed the intent with which it was founded.
Any democratically-run membership organization faces these risks. But the very advantages of proxies, that they are a path to participation by everyone, becomes a disadvantage if membership requirements do not protect against both hostile takeovers and influxes of chaos agents. Proxies make more urgent a need faced by all democratic organizations.
Conclusion
Any truly democratic polity, and any truly democratic membership organization, even one as tiny as five dozen, probably needs proxies; they may sometimes prove useful for even smaller groups. They are important for participatory democracy, but also for any form of democracy that seeks to carry out the will of the majority of people, rather than merely obtaining their tolerance of decisions made by a minority. In participatory democracy, they allow the practical maximum of direct participation while ensuring that delegates both represent and are accountable to those who select them. The sheer volume of political decisions required in complex democratic societies ensures the need for such accountability; time costs and information constraints prevent voters from making most decision directly.
Works Cited
Achterberg, P., & Houtman, D. (2009). “Ideologically Illogical? Why do the Lower Educated Dutch Display so Little Value Coherence? Social Forces, 87 (3), 1649-1670.
Ashton, M. C., Danso, H. H., Maio, G. R., Esses, V. M., Bond, M. H., & Keung, D. K. Y. (2005). “Two Dimensions of Political Attitudes and their Individual Difference Correlates: A Cross-Cultural Perspective.” Culture and Social Behavior: The Ontario Symposium, Vol. 10 : 1-29.
Axe, Leonard H., (1942). “Corporate Proxies.” Michigan Law Review 41 (1), 38-65.
Azzellini, Dario N. (2016). Communes and Workers’ Control in Venezuela: Building 21st Century Socialism from Below. Boston: Brill.
Davoudi, Leanardo; McKenna, Christopher; Olegario, Rowena (2018). “The Historical Role of the Corporation in Society,” Journal of the British Academy, 6 (s1), 17–47.
Dodgson, Charles (1884). The Principles of Parliamentary Representation. Harrison and Sons.
Edwards-Levy, Ariel (2023). “Most Americans want stricter gun control, but they’re divided on whether guns make public places safer.” CNN (May 26).
Eysenck, Hans (1975). “The Structure of Social Attitudes.” British Journal of Social and Clinical Psychology 14 (4).
Feldman, Stanley and Christopher Johnston (2013). “Understanding the Determinants of Political Ideology; Implications of Structural Complexity.” Political Psychology 35 (3).
Gallup (2023). “Energy: Gallup Historical Trends.”
GovTrack (2023). “Congress/Bills Statistics and Historical Comparisons.”
Hawkins, Kirk A. (2010). “Who Mobilizes? Participatory Democracy in Chávez’s Bolivarian Revolution.” Latin American Politics and Society 52, (3, 31-66.
Hellwig, Timothy (2014). “The Structure of Issue Voting in Postindustrial Democracies.” Sociological Quarterly 55 (4).
Kamal, Sameea (2023). “In California Legislature, 500 Bills Beat the Deadline.” Cal Matters (Feb. 21).
Macrotrends (2023), “Porto Alegre, Brazil Metro Area Population 1950-2024.”
Marist Poll (2023). “Biden & the Issues Facing the Nation NPR/PBS NewsHour/Marist National Poll.” Marist Institute for Public Opinion (Mar. 29).
Newport, Frank (2022). “Average American Remains OK With Higher Taxes on Rich.” Gallup: Polling Matters, Aug. 12.
Page, Susan. (2023). “A GOP War on ‘Woke’? Most Americans View the Term as a Positive.” USA TODAY/Ipsos Poll finds. USA Today Mar. 8.
Pew Research Center (2022). “Public Opinion on Abortion.” Pew Research Center (May 17).
Preer, Robert (1986). “Town Meetings Don’t Work.” The Washington Post, (July 13).
Roberts, David (2018). “Utilities have a Problem: the Public Wants 100% Renewable Energy, and Quick.” Vox, Oct. 11.
Rolheiser, Lyndsey; Saiz, Albert (2020). “Is Participatory Democracy Representative? A Survey of Engaged Town Meeting Voters.” New England Journal of Political Science, XII (1), 30-66.
Ru, Carlos Cortez (2004). “Social Strategies and Public Policies in an Indigenous Zone in Chiapas, Mexico.” Institute of Development Studies Bulletin, 35, (2), 76-83.
Shalom, Stephen (2021). “Decision-Making in a Good Society: The Case for Nested Councils.” The Center for Postcapitalist Civilization’s Working Papers.
Sintomer, Yves; Herzberg, Carsten; Röcke, Anja; and Allegretti, Giovanni (2012). “Transnational Models of Citizen Participation: The Case of Participatory Budgeting.” Journal of Public Deliberation, 8 (2), 9.
Tyson, Alec and Cary Funk and Brian Kennedy (2022). “Americans Largely Favor U.S. Taking Steps To Become Carbon Neutral by 2050.” Pew Research Center, Mar. 1.
Wilde, Matt (2017). “Contested Spaces: the Communal Councils and Participatory Democracy in Chavez’s Venezuela.” Latin American Perspectives 44 (1), 1 -31 of pdf linked, 140 -158 of print edition.
*Please note that 4 “bundles” is a generous estimate. There is a great deal of literature on multidimensionality in policy preferences: much of that literature that compares such theories to empirical data find a two-dimensional model yields a good match to actual public policy polling data (Ashton, et.al 2005; Achterberg, and Houtman 2009; Feldman and Johnson 2013, 7-10; Hellwig 2014; 596-624). One of the early pioneers in this analysis eventually added a third dimension (Eysenck 1975) based upon further analysis.
Leave a Reply