A large community does not become governable by becoming harmonious. As participation expands, differences in opinion, passion, interest, and attachment multiply. Some disagreements concern the community’s purpose; others concern how its rules should be interpreted or enforced. The moderator’s central problem is therefore not how to abolish faction, but how to prevent faction from turning disagreement into domination.

That distinction matters because faction is not simply any group that disagrees with another. It is a majority or minority united by a passion or interest hostile to the rights of others or to the community’s lasting interests. A group can be intense, organized, and persistent without being factional in this sense. The relevant question is not whether a position is unpopular, noisy, or politically inconvenient. It is whether the group’s conduct threatens other people’s ability to participate or converts collective power into an instrument against them.

This gives moderation a different purpose from consensus-building. Consensus may be desirable in some disputes, but it cannot be the general remedy for pluralism. People differ naturally, and attempts to make everyone share the same opinions, passions, and interests are impracticable. An institution that treats disagreement itself as the danger will eventually have to choose between suppressing liberty and imposing uniformity. Either remedy destroys something the community needs in order to remain a community of free participants.

Yet unrestricted conflict is not a neutral alternative. Unmanaged faction produces instability, confusion, distrust of public commitments, and threats to private rights. In a large online community, the practical analogue is a setting in which repeated pile-ons, brigading, or factional campaigns make formal access meaningless. Members may still technically be allowed to speak, but participation becomes too costly or intimidating for some groups. The issue is not that every person must be protected from criticism. It is that one group’s superior force—whether numerical, organizational, or rhetorical—must not decide who can safely remain present.

The danger is especially acute when a faction is a majority. A minority may obstruct discussion or disturb the community, but ordinary procedures can often defeat its aims. A majority faction is more difficult because it can preserve the appearance of popular rule while sacrificing the rights of a weaker party. A vote, a flood of reports, or widespread approval does not by itself establish justice. Numerical strength can be the mechanism by which favoritism or exclusion becomes institutionally respectable.

The challenge, then, is to protect both the spirit and the form of popular participation. The answer suggested by the extended republic is structural rather than moralistic. It does not depend on every leader being wise, impartial, or unusually virtuous; enlightened statesmen cannot always be expected to be at the helm. Instead, it seeks to control faction’s effects through representation, variety, and barriers to coordinated abuse.

Representation is useful here not because representatives are automatically better than the people they represent. They may be factious, locally prejudiced, corrupt, or self-interested. Its value lies in creating a body responsible for considering broader interests rather than simply mirroring the loudest assembled crowd. For a community moderator, this implies that rules and decisions should not be determined solely by the most active participants in the immediate controversy. A decision process needs some distance from the dispute it is judging, because those who are both parties and judges have strong incentives to treat their own faction’s interest as the community’s interest.

Scale creates a second protection. A larger sphere contains more distinct parties and interests, making it less likely that one majority will share the same motive to oppress another group. Even when an oppressive motive exists, dispersed participants may find it harder to discover their strength and act in unison. This does not make large communities automatically fair. Scale can make coordination difficult, but motivated factions may still coordinate, and size can also weaken local knowledge. The point is narrower and more practical: diversity and dispersion can prevent one concentrated interpretation of the community from becoming the only legitimate one.

That insight also clarifies why minority protection is not favoritism. A safeguard against coordinated domination does not endorse every minority claim. It recognizes that a community’s rules must remain legitimate when control shifts between rival groups. The same standard should therefore distinguish protected disagreement from conduct that deprives others of participation. A minority faction can be wrong without being silenced; a majority faction can be popular without being permitted to make the space intolerable.

Durable moderation consequently requires more than deciding individual cases. It requires institutional arrangements that filter immediate passions, preserve room for competing interests, and make coordinated injustice harder. The arrangement must also acknowledge tradeoffs: too much distance can lose local knowledge, while too little distance can make decision-makers overly attached to the faction immediately before them. The aim is not perfect neutrality, which may be impossible, but procedures that remain defensible across changing disputes.

Faction cannot be removed without removing liberty or imposing uniformity. The responsible task is therefore to control its effects. For a large community, legitimacy does not come from ensuring that no one is ever frustrated, criticized, or outvoted. It comes from ensuring that no faction—popular or marginal—can convert its passion into a standing power to exclude others. Pluralism is not a problem to erase. It is the condition the institution must be designed to survive.