Reimagine On-chain Dispute Resolution

Authors: LJ Huang, Heather Zhou, Xin Yan, EthSign

In July 2022, Vitalik shared the concept of "network states" proposed by former a16z partner Balaji Srinivasan in his new book The Network State on his blog. This concept refers to a collective that is not constrained by previous social conventions or existing laws, aiming to establish the most effective and member-satisfactory guidelines for "co-living." Both blockchain and network states share common attributes, as they both seek to "create a new cornerstone," which means that this "network state" also needs to establish its own social rules and judicial system.

Moreover, for a long time, the blockchain community has represented openness, freedom, resistance to censorship, and neutrality. However, this does not mean it is a lawless land; when disputes and controversies arise on-chain, the community also needs to establish institutions similar to courts to coordinate and adjudicate. So, can blockchain and network states seek smart contracts or more efficient systems to resolve on-chain disputes?

Perhaps we can first try to seek solutions from the original logic of judicial procedures.

A Thought Experiment from 76 Years Ago: How Does the Logic of Judicial Procedures Emerge from Nothing?

In a classic legal academic paper published in the Harvard Law Review in 1946, Lon L. Fuller discussed several important points about judicial procedures through a thought experiment. He asked readers to imagine "a group of castaways isolated in some corner of the Earth," all of whom have forgotten their "previous social existence and the rules and laws they once followed."

One person from this group is chosen as an arbitrator or judge to resolve all disputes arising in this "blank" society. Fuller argued that if this island judge is rational, intelligent, and wishes to promote the group's prosperity, he would not make judgments solely based on personal preferences, as he knows his decisions will be seen as precedents, and the outcomes of individual cases could give rise to a set of rules that the community will follow.

Thus, this judge would strive to discover the rules that people verbally adhere to from the perspective of promoting the welfare of the group. In other words, he would study and contemplate the logic behind the law. However, he would also realize that it is impossible to answer all potential disputes solely based on logic; some laws must be determined by his personal authority, as a representative of fairness and justice chosen by the community to establish rules. Therefore, the people in the community would adhere to these rules not only because they are "correct" (as they align with social reality) but also because they are established by the judge (reflecting legitimacy).

Thus, Fuller asserted that judicial procedures are a combination of reason (referring to the order discovered from the substance of life) and legal authority (referring to the order achieved through legal authority).

Source: Mike Lynch Cartoons

This 76-year-old thought experiment not only reconciles (or at least attempts to reconcile) the antagonistic relationship between natural law and positive law but also reveals how to interpret on-chain dispute resolution systems. In a sense, the vision of the Web3 community and network states is very similar to Fuller's imagined island community.

Current State of Web3 Dispute Resolution Solutions

In fact, there are already protocols attempting to establish a set of on-chain dispute resolution systems, but their logic is not perfect.

Currently, Web3 dispute resolution service providers like Kleros and Aragon simulate the American jury system by randomly selecting some arbitrators and requiring them to make judgments based on the evidence provided by both parties.

To address the potential abuse of power by jurors, selected jurors find it difficult to make judgments solely based on what they believe is right; they must consider the judgments of others and the majority of jurors. Jurors participate in this Schelling game-type gamble with financial incentives: they must place bets each time they judge a case, and if they correctly predict the outcome of the majority's decision, they will be rewarded; otherwise, they will lose their stake.

Source: Kleros White Paper

Within Fuller's theoretical framework, this game-theory-based system relies too heavily on "rationality" while underestimating "legal authority," leading to certain design flaws.

First, the basic assumption of this approach is that in a decentralized world, everyone is rational and will strive to maximize their own gains. Not only are each juror's actions rational, but they must also assume that all other jurors (who do not know each other) are similarly rational. However, as many behavioral economics studies indicate, this assumption does not hold. This method also overlooks the fact that often, for a complex case, there may not be a single correct answer, and due to certain community-specific values, the judgment outcome is debatable.

Second, after a group of jurors arbitrates a case, no principles are produced. When similar cases arise next time, a new randomly selected group of jurors will arbitrate, unbound by any decisions made by the first jury. They will start from scratch and hope to reach similar conclusions—though there is absolutely no guarantee of that.

Furthermore, monetary incentives can be said to undermine the legitimacy of the system—each decision is driven by the desire to make money and is based on guesses about others' judgments, rather than attempting to arrive at a result that could help the community improve efficiency.

Alternative Suggestions

A dispute resolution system that better embodies Fuller's concepts of "reason and legal authority" can be achieved through DAOs and their principles.

DAOs can serve as a "court," providing a set of dispute resolution systems for members both within and outside the organization. The proposal of the Nation3 court can serve as an example, utilizing smart contracts as arbitrators to implement certain "default dispute resolution logics," although their solution differs from our proposed principle-based jury system.

In our proposal, before establishing a DAO, DAO members must first collaboratively propose some principles to be adhered to during case hearings and reach consensus based on these decisions. These principles can initially be very basic or even abstract—such as "parties should fulfill contracts under all circumstances." As the DAO hears more cases, these principles will be further developed, refined, and constrained (for example, "parties should make reasonable best efforts to fulfill contracts, but under unforeseen circumstances, the obligation to fulfill the contract may be affected"). In this way, the DAO embodies "legal authority," while these principles serve as the foundation for legal reasoning.

Because each dispute resolution DAO has its own principles, people can negotiate and pre-designate a specific DAO to resolve disputes when signing contracts (similar to jurisdiction clauses in the Web2 world, but can be automatically executed using smart contracts when disputes arise). Theoretically, this will lead to the DAO with the most effective rules becoming the most popular "jurisdiction," or DAOs specifically addressing certain issues will emerge, dedicated to adjudicating specific types of cases (like the Kleros court tree).

Key Infrastructure

In our proposed dispute resolution system, the lifecycle of a case and the necessary technical infrastructure can be summarized as follows.

• Bets from the parties: To ensure the enforcement of the dispute resolution DAO's judgment, both parties must first stake tokens equivalent to the amount in dispute when they mutually agree to submit the case to a certain DAO. The stakes will be placed in an escrow account and distributed to the winning party after the final judgment.
• Selection of jurors: Jurors will be randomly selected but entirely from DAO members. For newly established DAOs, the proportion of new or first-time jurors can be limited to ensure the legitimacy of the jurors. Jurors must pass a proof of human test and receive endorsements from other qualified members to qualify as jurors. These measures can prevent Sybil attacks, where multiple accounts are created as false identities to manipulate the selection and decision-making process of the jury.
• Jurors' judgments, appeals, and impeachment: Jurors will adjudicate the cases assigned to them based on the agreed-upon and continually updated principles. Each juror will provide a brief explanation for their vote. The side that receives a simple majority (or overwhelming majority—regardless of how DAO members decide) wins the case. This decision can be appealed to a new randomly selected pool of jurors, with a limited number of appeals. If it can be shown that the jurors did not judge the case based on the agreed-upon principles, parties can also impeach the jurors. If the impeachment is successful, the jurors will be disqualified.
• Evidence submission: Specific rules regarding evidence submission should be established to ensure that the evidence submitted by both parties is relevant and authentic. Contracts and agreements signed and properly verified through products like EthSign Signatures should be prioritized, as this product can provide detailed information and a history of documents, including signature generation, signers' addresses, and other relevant information.

Source: EthSign Team

• Juror compensation and precedents: Regardless of the case outcome or how they vote, jurors will receive a fixed compensation for each case they adjudicate. Additionally, the final judgment of each case should be public and can serve as a reference for future cases. As more cases come to the court, a system should be established to select cases factually similar to previous rulings, allowing the jurors of the current case to use these cases as useful precedents.

Looking Ahead

As Vitalik suggested in his article, a few DAOs need to take on certain responsibilities of sovereign states, such as court systems. In the coming years, similar to the jurisdictions of Singapore and Dubai, which are favored by Web3 startups in the physical world, a few DAO-led courts may begin to gain the trust of the community and serve as on-chain dispute resolution centers for specific issues, such as asset tokenization, DeFi, NFT ownership, insurance, and more. In the event of a dispute, projects will be able to pre-designate a DAO as the venue for dispute resolution. This decentralized, on-chain, DAO-operated dispute resolution system will be highly beneficial for the prosperity of numerous Web3 communities and network states in the coming years.