Arbitration DAO

An Arbitration DAO (arbDAO) is a group of judges (or arbitrators) who resolve disputes that arise in the course of immediate business. In any DAO which employs a Work smart contract (WSC) under DGF, when one or more of the parties is dissatisfied with the execution of the contract, they may trigger an arbitration clause, which transfers the assets encumbered in the WSC to an arbitrator in the arbDAO. The judge then resolves the dispute according to the standards set by the arbDAO.

Arbitration is the immediate type of judicial governance. Importantly, the arbitration clause must be triggered before the WSC is concluded. Otherwise the dispute can only be resolved by the lengthier review of Forum revaluation.

Overview

The basic type of judicial governance is arbitration. Arbitration allows a DAO to review the automated outcome of a work smart contract (WSC). The "code is law" assumption of Web3 makes such review difficult in a primary DAO. Nevertheless, the code is law standard is necessary when a DAO allows open access to pseudonymous members.

However, rigid contracts following code is law are not good for business. Business contracts demand all parties perform in a predictable manner under assumptions about the future business environment. But no one can predict the future. Business must be able to continue when an eventuality occurs that was not anticipated in the smart contract. Parties to the smart contract must be able to come to some consensus on what is fair in those circumstances. The existence of a meaningful reputation system motivates the parties to behave as well as possible. Nevertheless, sometimes, even if both parties are behaving with the best intentions of honoring the spirit of good business, irreconcilable differences in perspective about what is fair are inevitable. In such cases a third party is necessary to adjudicate and resolve the dispute.

System design

To achieve this type of judicial governance in a primary DAO while respecting the standard of code is law, the WSC must therefore include all the logic necessary to satisfy business parties before the dispute arises. To anticipate every possible scenario and include protocols for dealing with them in the WSC is not possible. Instead, the WSC merely needs to include the ability to transfer the regulatory logic to a second arbitration smart contract which evolves with the arbDAO to have more sophisticated mechanisms to serve the needs of their jurisdiction.

Trigger

Each WSC which chooses to employ an arbDAO includes a clause that allows any party in the smart contract to trigger a sub-contract, called an arbitration work smart contract (arbWSC). If arbWSC is triggered before the conclusion of the basic WSC, then the WSC sends all its encumbered REP tokens, fees, and other property to the arbWSC, which then follows the basic work flow of DGF.

arbDAO workflow

Specifically, the arbWSC then calls arbitrator(s) from the arbDAO to do the work of the arbWSC. The arbitrators are selected (usually randomly weighted by arbREP) by the arbWSC from the currently functioning arbitration availability smart contracts (arbASCs) from the arbDAO bench.

The work of the arbWSC means the arbitrators follow the established arbDAO protocols for resolving a dispute between parties to the WSC. Once the work of arbitration is finished off chain, the arbWSC concludes by posting evidence of its work to the Forum for recording, then initiates a Validation Pool for the post to disburse the fees and mint new arbREP.

The arbWSC will typically itself include a clause which allows unsatisfied parties to trigger another arbWSC, recursively. Assuming the issue is not resolved to the satisfaction of the parties, these subsequent arbWSCs can escalate in intensity, such as expanding the number of judges to give greater attention to the matter, perhaps escalating to the larger attention of the DAO.

The incentive for each party to corruptly exploit the arbitration clause is everpresent and will need to be policed by sophisticated legal thinking when designing the protocols in each particular arbWSC. This needs to be continually updated as markets change, so the arbitrators themselves will improve the protocols with the DGF legislative governance process.

Example arbitration chain

Each stage is meant to facilitate a harmonious resolution to the dispute. This may be intended to redress a harm, but it may be resolved by instituting a future plan for further business.

1. Negotiation: Basic WSC is paused. The parties attempt to resolve their difference informally. If this fails, one or more party triggers the next step.
2. Mediation: One judge (mediator) is chosen to facilitate formal communication and resolution following an established process of information discovery. If this fails, one or more party triggers the next step.
3. First Arbitration: One judge (arbitrator) is called. Parties submit their cases. Arbitrator proposes the disbursement of assets. If this fails, one or more party triggers the next step.
4. Arbitration Appeal: A new arbitrator is called. Repeat 3 a fixed number of times. If this fails, one or more party triggers the next step.
5. Court Case: Larger body of judges is called from the DAO to resolve the case democratically. This finalizes the WSC.

If Step 5 fails, the code-is-law standard prevents the WSC from being tampered with. However, one or more parties which are still unsatisfied can make their final appeal to judicial redistribution of REP tokens, by making a proposal for Forum revaluation. Weighted democracy by participation of the entire DAO determines the outcome.

Domain specific arbREP

Jurisdiction is a complex issue. There are many different arbDAOs, one for each culture in the DAO ecosystem. Different DAOs have different values and different standards for resolving disputes. This requires different expertises in arbitration, requiring multiple different knowledge bases for each jurisdiction. Each DAO is responsible for deciding which arbDAO to use, and for monitoring that arbDAO's function.

The fact that arbitrating a dispute is a separate skill from the primary work a typical DAO performs, means an arbDAO will typically have a separate REP token, called arbREP. Therefore an arbDAO can be seen as a technically separate DAO with which the primary DAO negotiates in order to secure their work. This negotiation concerns the percentage given by each WSC to arbDAO. The result of the negotiation will depend on the type of work and the demand it has for arbitration as well as the supply of arbiters.

Two natural approaches to paying for the work of arbitrating disputes arise: paying for arbitration at the moment the trigger is pulled, or paying a general tax for every smart contract that includes the trigger. Both approaches open opportunities for abuse of process. We generally favor the latter approach of a general tax from every contract, since it distributes the cost to the entire DAO. This encourages parties to participate in arbitration whenever there is a genuine dispute. This can lead to abuse of process of frivolous litigation. However this can be addressed with sanctioning. If instead, however, a large fee is associated with pulling the trigger, this would inhibit the weaker party's choice to seek resolution. This is a threat to decentralization as it enables an economy of scale. It also encourages the use of the threat of malicious prosecution or vexatious litigation by either party, encouraging competition and destabilizing the harmony of the DAO.

Arbitration tax

Given the arguments in the previous paragraph, the typical primary DAO will pay for arbitration by including a tax in each WSC, which acts like insurance, giving everyone who uses the WSC the right to trigger the arbitration clause. This tax can be collected in a treasury which is tapped each time an arbitration clause is triggered. Each time the treasury is tapped the money is sent to the arbWSC as a fee, after which arbDAO workflow is followed.

The size of the tax will depend on the DAO and changing market conditions. Given the fact that the tax gives 0 transaction cost for inititating arbitration, if frivolous claims are not policed, the arbDAO will be expensive and ineffective. The more effective the arbDAO is, the less it will be used, and the lower the tax will be. Therefore arbitrators get more money and work if they do a worse job. This effect encourages the legal system to be ineffective and costly, explaining one type of institutional creep which is universally experienced in pure governmental bureaucracies. However, an arbDAO has different incentives than pure governmental bureaucratic institutions given the market price negotiations that occur between DAO and arbDAO, because there is a choice of different arbDAOs available (since arbDAOs can be cloned easily).

Arbitrator recruitment

The way arbitrators are initiated in a primary DAO is to recruit them from the group using their work REP. A potential arbitrator signals their availability to serve, by encumbering their REP in an arbitration availability smart contract (arbASC). If they demonstrate their mastery of the protocols, by successfully serving as an arbitrator, they receive arbREP to encumber in future arbASCs for future arbWSCs. This stage of recruitment requires careful monitoring by existing arbitrators with existing arbREP, with review using Forum revaluation.

Once sufficient arbitrators exist, arbitration ASCs can prevent new recruitment by preferencing arbREP of existing arbitrators to basic work REP from potential new arbitrators. New arbitrators would only be selected for a case if no arbREP was encumbered in arbitrations ASCs.

Failure of arbitration

Arbitration may fail to produce justice for many reasons. First, the WSC can conclude before a party notices an injustice. Then the WSC has no power to trigger the arbitration in the first place. Second, the WSC is not perfect. There will inevitably arise business situations that the WSC did not anticipate. The executive governance encoded in the WSC can be injustly written for certain eventualities. Third, the larger functioning of the DAO can be flawed. The legislative governance the DAO operates under requires eternal improvement. For all of these reasons, arbitration will sometimes fail to establish justice. Therefore, the final backstop is to rely on the transcendent values of the DAO – the motivations for why the group exists, which ultimately govern how the group behaves toward one another. These values can be served by one final type of judicial governance, when the other approaches fail: Forum revaluation.

See Also

• Policing DAOs