Democracy Is More Than Voting
National Political Committee member Amy W gives a primer on the purpose and function of deliberative democracy and proportional voting for the socialist party.
Amy W
Zhou Xiaojun, Shi Dawei, Wang Chunyan, The Fire Is Red
The case for parliamentary democracy
DSA’s democratic structures typically operate under parliamentary procedure; from the National Political Committee (NPC) and its subcommittees to chapter committees and other formal bodies. They usually use Robert’s Rules of Order Newly Revised (RONR), but some bodies instead use an alternative or derivative parliamentary authority such as Rusty’s Rules, a simplified RONR developed by the IWW which is adequate for small meetings. I prefer consistency and am most familiar with RONR, but in my opinion the exact authority is less important than the general shape of the practice.
One can make legalistic arguments about the why: RONR is our rules, that’s how we work, you have to follow them. But I’m a firm believer in the principle that the law was made for man, not man for the law, and instead hope to make a more rational and ultimately positive political argument.
Despite its name, parliamentary democracy when implemented in a mass organization is a form of participatory democracy. Participatory democracy means all members have equal rights, not just to vote (as in plebiscite democracy, including referenda and absentee voting), but to propose and shape motions. Through this method we can reach agreement on action collectively, and avoid both the minority rule of consensus methods (which allow a small group to hold up making a decision) and the constrained choice of referenda (which often represent binary choices devised by leadership).
We need an argument for why how we do democracy in DSA matters. Why vote in meetings? Why not simply poll members?
Often when arguing for parliamentary methods, like voting in meetings, we cite the importance of debate. And while it is important, (I’ve had my mind changed by debate many times), it's only one aspect of parliamentary procedure. The goal, especially in organizations like DSA, is to move from combative to collaborative. Proposals can be amended in response to criticisms, or delayed or referred if they require further input. The agenda and the meeting itself are subject to the democratic will of the membership.
Our model of organizing means we develop ideas collectively. We don’t exclusively depend on leaders to guide and members to follow; we discuss problems we’re facing and develop plans to address them together. This distinguishes us from bureaucratic organizations like most unions and other socialist parties. Although this process can also take place outside of meetings (collaboration is an important part of the resolution-writing process), taking proposals to the general body means that there is opportunity for all members to weigh in.
Participatory democracy also serves our strategic goals. Any organization is doomed to produce an outcome in its own image. If we seek to build a society that truly believes that every cook can govern, we have to put those beliefs into practice. Just as we argue for democracy in society or for participants in movements and coalitions, we need democracy for our members.
Through this process we develop members as political actors: not just as voters, but as confident participants in governing the organization. We help them build skills that go beyond DSA; that they can take home to their workplaces, apartment blocks, and other sites of struggle. Through both recruitment of workers into the organization and building member capacity and experience in decision-making, we develop a society and a working class ready for self-governance.
I will now discuss some common alternative voting methods in DSA and analyze them through this lens of democratic practice.
But sometimes it isn’t?
Many DSA bodies regularly practice absentee voting, including the NPC. Absentee voting, in the context of parliamentary democracy, means taking votes outside of a meeting by one means or another. It was adopted by many chapters and other bodies at the start of the COVID-19 pandemic because there was no immediately practical way to vote during an online meeting.
Some proponents argue that meetings are inaccessible, leading to the euphemism “accessible voting.” They claim that members who do not or cannot attend meetings are “disenfranchised.” Leadership bodies also treat it as a measure to expedite decision-making, taking votes using anything from Loomio to thumbs-up in Signal chats for items they are unwilling (or sometimes unable) to wait for a meeting, or call for a special meeting, to dispose of.
Robert’s Rules typically prohibits absentee voting unless specifically allowed in the bylaws, which it usually isn't (RONR (12th ed.) 45:56). RONR argues that it is “a fundamental principle of parliamentary law”—that is, of the practice of deliberative assemblies—“that the right to vote is limited to the members of an organization who are actually present.” Organizing bodies have various workarounds at their disposal; a chapter steering committee may call an advisory vote by email to take an action on behalf of the chapter if they can’t meet quorum, or a body may move to ratify votes taken outside of the meeting (which is not really what a motion to ratify is for (ibid 10:55), but that’s fine).
This is the negative, legalistic argument. But I think RONR 45:56 also supports the positive argument I laid out in the previous section: when proposals are voted on in meetings, the question can be modified through amendments, minds can be changed through debate, and members can make procedural motions (to table, to refer to committee, etc.) which affect the question being voted on. In other words, members play a much more active and decisive role in the process of making decisions.
Voting by online referenda obscures a very important question: who decides what the question being voted on even is? One possible answer is that any proposal with enough signatures is put to the membership, similar to ballot initiatives in states which allow them. It could be that the chapter steering committee or another leadership body makes the decision. Both of these options risk boiling decision-making down to binary choices made by atomized voters. It is possible that questions could be amended in a membership meeting, then the amended version put to a referendum, but this workaround undermines the argument that referenda are important because they make decision-making more accessible than meetings.
My core argument is this: voting only by mail (or email, etc) ceases to be parliamentary procedure. It ceases to enable participation in the democratic process. It gets rid of participatory democracy’s ability to develop members, to execute on the philosophy that every cook can govern. And that is reason enough to avoid it.
Another form of absentee voting is proxy voting. Proxy voting is a practice where one member can assign their voting power to another. This is discussed in RONR (12th ed.) 45:70-71 as a regulated practice, though 45:56 describes proxies as a form of absentee voting, which is generally discouraged.
I view proxy voting as often, but not always, harmless. The membership of Seattle DSA adopted it as a negotiated alternative to a proposal for absentee voting, but RONR makes an interesting argument against it. That is, in stock corporations proxies make some sense because shares are voting, not individuals, but it should be avoided in “nonstock” corporations where the voter is an individual. The reality is that this form of voting doesn’t provide representation for absent members; it instead provides the illusion of representation while over-weighting the votes of one or more present members.
At the 2025 Seattle DSA Convention, two opposing sides on a particularly contentious topic engaged in what one member called an “arms race” to get proxies for their side of the debate. It’s impossible to say whether this changed the outcome, but it over-weighted proxy holders’ votes on not only this question but the others we considered as well.
There is one exception where I think proxy voting is positive in DSA: National Convention. At Convention, delegates are elected to represent their chapter. They do so with qualifications: they represent both their own views and those of their faction in general. But each chapter is entitled to a certain number of votes. When some delegates are unable to attend (and alternatives are exhausted) it makes sense to allow delegates to assign a proxy so that the chapter still receives its proportional representation at Convention.
In the next and final section, I will briefly touch on electoral methods and make the case for Single Transferable Vote (STV), and proportionality.
Representative bodies should represent
I have until now focused on the case for parliamentary democracy in DSA, and critiqued other methods of voting and democratic practice common in the organization.
In the first section I discussed participatory democracy. Often this term is synonymous with direct democracy, which DSA members practice in chapters, committees, and caucuses where every member often votes on every question. But direct democracy runs into a problem of scale, not just in society but in our organization. It can be difficult to arrange a meeting of even a couple dozen people. While larger chapters frequently manage meetings of 100 or more members, it would be impractical to manage a meeting of DSA’s tens of thousands, never mind the logistics of getting them to one place at one time! It's also sometimes necessary to act or make decisions between meetings of the full body.
Representative democracy is (on paper, at least) the sort we are most familiar with in the US, and in most nations with parliamentary or congressional systems. Because every person cannot be present for every decision, they elect people to represent them. The US is particularly ineffective at this because between the Senate and the near-universality of first past the post voting, its institutions are particularly unrepresentative. But many other nations do better, and so does DSA.
DSA’s highest body—our equivalent of a party congress—is the biennial National Convention. This is a representative body: delegates are elected to represent their chapters and vote on direction and leadership for DSA as a whole. This is not too different from most parliamentary systems. Likewise, the NPC, chapter steering committees, and often committee or working group steering committees are representative bodies that make decisions and steer the organization in between meetings of the membership.
The most important role of representatives is to represent their electorate. This may sound tautological, but it’s important to keep in mind. That’s why recall is a crucial right for electors to a standing body, and why I’ll argue for proportional representation in nearly every situation. But before talking about proportionality, I need to talk about voting methods.
There are a variety of methods to elect representatives. Broadly they can be categorized into single-winner and multi-winner. Multi-winner methods can then be further divided into proportional vs majoritarian, while single-winner methods are (more or less) majoritarian.
We have almost exclusively single-winner elections in the US, and DSA has a lot too. Single-winner methods are intended to select a winner that best represents the preferences of a majority of the electorate as a whole, which inherently means that minority preferences are not represented. This discrepancy is widely abused in the US through gerrymandering, which is impossible in DSA where the electorate is a whole body.
DSA also has elections which are inherently multi-winner, such as the NPC, at-large chapter steering seats, and Convention delegates. Which leads us to the distinction of majoritarian vs proportional.
Proportionality is key to making sure every DSA member gets represented. Let’s imagine an election for 10 open seats, where three slates are running. Slate A attracts 60% of the vote, and B and C each attract 20%. Each slate runs 10 candidates, that is, enough candidates to fill all the seats. For the sake of argument we’ll say each individual member fills their ballot with only one slate, although this is rarely true.
Majoritarian voting methods like Approval and Score approximate a result where every elected seat represents the majority preference of the electorate. These methods usually make the election very simple to process. In Approval and Score, all you have to do is tally the “score,” and the highest scores win. But they also strip minorities of representation as long as the majority runs enough candidates.
In this example, a majoritarian voting method fills all 10 seats with candidates from slate A. The seats represent the majority of the electorate, but supporters of slates B and C have no representation. If slate A ran less candidates there would be room for others, but majoritarian representation offers no guarantee.
Voting methods designed for proportional representation like Single Transferable Vote (STV), on the other hand, are designed to represent the diversity of the electorate. Rather than minority perspectives being squeezed out, they get representation according to their level of support. Proportional voting methods tend to be more complicated to process. I don’t know of any that don’t require multiple rounds of calculation, except party list methods which don’t work in DSA. But we generally use voting software like Opavote, which, while complicated, uses simple math.
Under STV, slate A wins 6 out of 10 seats, while B and C each get 2. Slates generally do not benefit from running more candidates. The elected body represents the electorate as a whole.
The result from majoritarian voting methods is not really representative at all, and defeats the purpose of multi-winner elections. If the goal is only to represent a majority, why not elect a single leader and allow them to appoint a cabinet?
Some proponents of majoritarian methods seem to embrace this. They argue variously that the majority should rule, that leadership bodies are primarily executive and not political, or that representing the diversity of the electorate just leads to strife and deadlock. While there is a kernel of truth to these arguments—leadership bodies in DSA have a significant executive function and diverse bodies sometimes have no clear majority on a course of action—the more important point is that every decision in DSA is political, and leadership bodies necessarily make political decisions.
DSA members deserve proportional representation for their leadership bodies and Convention delegations. Minority perspectives deserve to be heard and have the opportunity to change minds on representative bodies just as they do in the general body.
All of these structures—meetings for deliberative decision-making, synchronous voting systems, and representative bodies elected by proportional representation—may at first seem superfluous, or hard to understand. Each can all too easily be discarded in the heat of the organizing moment. But this is to our detriment. They all fulfill the same principle that underpins the socialist project: the working class must first govern its organizations before it can govern society. Every member must have equal right to engage in the decision-making process in all its
