Liquid Democracy, Delegation and VoteWrap
Liquid democracy sits between direct and representative democracy.
A participant may vote personally or delegate their voting power to someone they trust. In many liquid-democracy models, that delegate may pass the accumulated voting power to another person, creating a chain of transitive delegation.
Delegation may be general or limited to a topic. A voter might participate directly on education but delegate health policy to somebody they believe has greater knowledge or experience.
What liquid democracy contributes
Liquid democracy recognises two realities:
- people should retain the right to participate directly; and
- people cannot become experts in every subject.
It therefore allows democratic participation to move rather than remain fixed for an entire parliamentary term.
Potential advantages include:
- flexible delegation by issue;
- the ability to reclaim a vote;
- greater use of trusted knowledge;
- participation without requiring everyone to vote on everything;
- alternatives to permanent political representatives; and
- the possibility that capable representatives emerge through public trust rather than party selection.
These ideas strongly overlap with VoteWrap.
Risks within delegation systems
Delegation is not automatically democratic merely because it is voluntary.
Voting power may concentrate around a small number of prominent delegates. Participants may follow popularity rather than competence. Long delegation chains may weaken accountability because voters no longer know who ultimately exercises their voting power.
Cycles can also occur: one person delegates to another who eventually delegates back through the chain. Systems require rules for broken chains, abstaining delegates, conflicting delegations and reclaiming a vote.
Privacy presents another challenge. Publishing delegation relationships may expose political associations, enable coercion or encourage people to follow visible trends. Concealing relationships protects voters but makes public auditing more difficult.
Research has therefore produced both promising models and warnings. Experimental work has found that overdelegation can reduce the amount of independent information in a vote and may perform worse than universal majority voting in some conditions. Other research explores ranked or multiple fallback delegates to reduce vote loss and delegation failures.
How VoteWrap differs
VoteWrap is not only a delegation algorithm. It is intended as democratic infrastructure connecting people, representatives, communities, electoral geography, legislation and public outcomes.
Within VoteWrap:
- a person is connected to a verified electoral jurisdiction;
- votes aggregate through communities and electorates;
- a voter may self-represent;
- Bill Representatives may be selected for particular legislation;
- General Representatives may provide standing representation;
- Community Representatives may express a collective position;
- the voter remains able to change position or representation; and
- parliamentary decisions can be compared with verified electorate results.
A representative’s legitimacy would therefore not come simply from accumulating anonymous voting weight. It would exist within identifiable democratic relationships, jurisdictional limits and continuing accountability.
At the same time, the identity of a voter’s proxy or representative should not automatically be exposed publicly. VoteWrap’s emerging privacy model treats contact information and political relationships as private by default, while still preserving the records needed for verification, aggregation and audit.
Avoiding uncontrolled chains
VoteWrap does not need to adopt every feature commonly associated with liquid democracy.
In particular, unrestricted transitive delegation may create excessive concentration or make responsibility difficult to trace.
The system could instead use carefully defined forms of representation, such as:
- direct selection of a Bill Representative;
- a standing General Representative;
- ranked fallback representatives;
- limits on onward delegation;
- automatic return of voting power where a representative does not act;
- visible concentration warnings; and
- an overriding personal vote by the voter.
These design decisions should be tested rather than assumed.
What the approaches can learn from each other
Liquid-democracy research provides useful tools for studying delegation chains, concentration, cycles, fallback rules and the circumstances in which delegation improves or harms collective decisions.
VoteWrap adds a broader social and constitutional framework:
- verified participation;
- real electoral boundaries;
- local community relationships;
- bill-level and issue-level deliberation;
- representative roles with defined responsibilities;
- privacy protections;
- aggregation from community to jurisdiction; and
- comparison with the actions of elected officials.
Liquid democracy asks:
Who should exercise my vote on this question?
VoteWrap asks that question, but also asks:
How do we verify the democratic community, understand the legislation, deliberate across differences, protect political privacy, aggregate the result and hold public representatives accountable?
Delegation is therefore an important component of VoteWrap, but not the whole democratic system.
Sources and further reading
- Generalizing Liquid Democracy to Multi-Agent Delegation
- Liquid Democracy: Two Experiments on Delegation in Voting
- Liquid Democracy with Ranked Delegations
- The Fluid Mechanics of Liquid Democracy
- Anonymous and Copy-Robust Delegations for Liquid Democracy