Governance, whether of a company, state, or blockchain, fundamentally boils down to who is a stakeholder and the rights to which those stakeholders are entitled. If you fail to clearly define stakeholders, or fail to establish and enforce their rights, governance has likely already failed. Establishing stakeholders and their rights is not sufficient to have effective governance, but it is a necessary prerequisite.
In practice the enforcement of rights is difficult, as rights are generally properties of individuals, but group decisions are most practically deferred to consensus-seeking political processes.
Rights must be enforced for individuals within political systems despite the fact that individuals wield only marginal political power. This challenge is often framed in the context of the Tyranny of the Majority, where a majority position is used to create policies which negatively impact those in minority positions.
One approach to avoiding the tyranny of the majority is to aim for unanimity, or as close to that as is possible. As the group grows, and the types of decisions become more diverse, the ability to reach near unanimity becomes increasingly more challenging and eventually results in political gridlock.
In some instances political gridlock is desirable, and certain types of changes should not be easy. However, it is critical for organizations of all types to be able to make simple decisions like allocating resources towards common goods or adapting rules and policies to thrive in the face of changing market conditions or technological advancements.
The most historically successful approach to protecting individual rights is through the creation of courts. Courts are tremendously useful institutions, as they serve to interpret and enforce the rules that define individual stakeholder rights while remaining neutral to the political aspects of creating and changing rules. However, courts only work if they can credibly establish and maintain neutrality.
In public legal systems, judges are often appointed and confirmed infrequently. The confirmation process is intended to insulate the judiciary from short-term shifts in political power, and helps to slow the rate of change to important foundational rules. Similarly, the use of randomness when selecting jurors helps to divorce the judicial process from political bias.
In private law, consenting parties can select their own arbitration providers on a per contract basis. Rather than appointing judges, market forces help establish the credible neutrality of arbitration providers. Private arbitration is incredibly common, particularly for resolving and enforcing contractual rights across jurisdictions.
Private arbitration works exceptionally well in commercial situations where parties have many interactions across many counter-parties, but may be less appropriate for long-term multi-party agreements, as it may be difficult to reach consensus on changing arbitration providers in the event the neutrality of the arbitration provider is questioned. This type of long-term multi-party agreement is common in the context of organizations, where a shareholder agreement of a for-profit company or the charter of a non-profit is used to establish rights of participants when the company is initially founded.
This particular challenge is something that we have thought a lot about in the process of researching a solution for decentralized dispute resolution for use within the Aragon Network’s Court. An approach we are considering incentivizes jurors to fairly resolve disputes through a system of random selection and appeals, with the final authority on any dispute ultimately determined by a futarchy decision market. This approach enables us to resolve most disputes simply by randomly selecting a small number of staked jurors, but provides users with the security and collusion-resistant assurances associated with a futarchy decision market. We will go into more detail about the importance of futarchy decision markets and what they are exactly later in the post.
Sortition and Appeals
In much the same way we don't immediately jump to the supreme court to resolve trivial disputes, disputes handled by the Aragon Court will initially be resolved by a process that has significantly less authority, then only if necessary, can be escalated to processes that have more authority through appeals.
Our goal is to design a mechanism that encourages randomly selected jurors to align their ruling with what they believe would be the result in the event a dispute is appealed to the point of a decisive verdict. We have approached this by using an iterative game divided into multiple rounds, with the number of jurors participating in the game increasing in each round. Between each round, external input (whether appeal fees are present) is used to determine if the dispute should be escalated to the next round or have the verdict finalized.
Since jurors are only rewarded after a verdict is finalized and only if their ruling agreed with the final verdict, they are incentivized to make their ruling based on their expectation of the final verdict regardless of what round they are drafted. Given this property, it's reasonable to expect that the verdicts returned by early rounds are likely to be representative of the final ruling, in these cases the amount of time, effort, and energy required to resolve most disputes is minimized.
With this mechanism alone we can resolve disputes fairly effectively if we assume that a majority of jurors are likely to reliably produce a reasonable verdict. Producing an obviously unreasonable verdict would have a huge impact on the future potential revenue of the court and all the jurors involved. This reasoning is similar to how private arbitration works currently, where firms compete to attract dispute resolution services in a market and are incentivized by market dynamics to rule fairly.
However, we actually don't need to entirely depend on an honest majority of jurors to determine the final verdict.
Futarchy Decision Markets
When a dispute is appealed to the final round, and all jurors have committed to a specific ruling, rather than resolving the dispute according to a majority of jurors we can defer the ruling to a futarchy decision market. If you are unfamiliar with futarchy, I recommend starting at the source.
However, if you just want a quick summary, the basic premise is that if you have a decision which needs to be made, and you can relate that decision clearly to some objective success metric, you can use prediction markets to pick the option which is predicted to be more positively associated with your success metric. Anyone can participate in the prediction market, but attempting to manipulate the outcome of the decision by manipulating the prediction market represents a large cost so long as the success metric used to settle the market cannot be easily manipulated.
ANT is required to stake in the court, and demand to stake in the court is driven by the court’s fee volume. If the market loses confidence in the court because a decision was resolved unfairly, it will negatively impact fee volume and therefore also the price of ANT. Similarly, if the fork choice rule picks the option that is generally preferred by the market we would expect demand for ANT to increase or remain constant. This makes the price of ANT a good success metric for determining the outcome of disputes.
Well constructed futarchy decision markets can be highly collusion resistant. However, futarchy decision markets become less reliable when making less impactful decisions, as the less impact a decision has on a market's success metric the less signal there is to predict on and the more risk participants take on by participating, and the fewer participants in the market the easier it is to manipulate. This is why we don't want to use a futarchy decision market for every decision or to resolve every dispute! Instead we use the futarchy market as a trust anchor to establish a strong incentive for jurors to adjudicate honestly and with appropriate care and avoid using the futarchy decision market unless a dispute has been escalated through a series of multiple appeals. By the time a dispute has been escalated to the futarchy decision market, there will be a lot at stake for jurors, and there will be lots of attention on the dispute, and we can expect that the outcome will have a significant impact on our success metric.
The Aragon Court is still under active research and development, and may change significantly before release. You can follow our progress on our forum. We’ve also recently updated our whitepaper to more closely align with recent research and development. The main court protocol including Sortition and Appeals is being actively developed by Aragon One here and the Futarchy implementation for Aragon organizations is being developed by Level K here.