A DAO is a mechanism of governance enabled by smart contract execution of tokenized commands. Tokens send signals; smart contracts interpret those signals and deliver an outcome. In its most basic form, a DAO is naive: it is not necessarily flat, nor necessarily democratic. In order for a DAO to form, a group of people must agree on the initial architecture and answer: who holds tokens, what rights do tokens create, and what smart contracts exist at the beginning? Answering these questions and committing them to code creates the initial state, the genesis block of the DAO.
Source: https://i.imgur.com/KkpnFYw.png
Distinguishing Initial States
Each DAO will have a different initial state. Some set of facts leads the founding group to reach an agreement to develop a governance system that enables trustless execution of commands based on an initial set of assumptions. The agreement creates legitimacy amongst the founding community and may be memorialized in a white paper, documentation, or constitution that will serve as the baseline of governance from which all subsequent proposals and votes will extend. The scope of the assumptions that create the DAO can vary widely. The initial state can be a nearly blank slate or a well-developed governance plan.
If the founding group decides to begin with a blank slate, the community faces a difficult, granular road. They have to begin to propose and vote on the bureaucracy that will become their governance system (e.g. how, when, and where to make proposals). It is rare that a DAO begins with no predetermined mechanism for making proposals, but examples of the rigor it would take to do this after genesis come from several DAO governance revamps. For instance, Maker launched a set of proposals in 2020 in an effort to fully decentralize, which established a framework for making Maker improvement proposals “MIPs” that began with a proposal for the form and process of submitting more proposals, labeled MIP0. The proposal on how to submit a proposal (a meta-proposal) is an echo of the 2013 proposal to create a form of Bitcoin improvement proposals (“BIPs”).
It is more likely that a DAO begins with some ground rules established by the founding group. This may come in the form of suggestions set forth by the founding team that the community adopts by custom, or it may come as a more formal set of rules that the founding team designates as a condition for further participation. An example of the first approach comes from ShapeShift. After launching its DAO, ShapeShift explained to the community what the DAO could control and offered to the community a suggested format for governing by using a proposal framework and soft quorums.
Many DAOs launch with developed initial states. Recent launches, like Goldfinch’s DAO, include detailed proposal frameworks, populated initial governance councils, and presumed voting systems (i.e. quadratic voting). DAOs that launch on platforms like Superdao, Aragon, Colony, and DAOstack may choose from a broad range of initial states. For instance, DAOs launched on Colony can implement lazy consensus voting and reputation scoring; DAOs launched on DAOstack can bifurcate governance and economic rights through a separate reputation voting system and can implement voting by holographic consensus.
Comparison of Reputation and Token-based Voting (DAOstack)
Source: Daostack Comparison of Voting Methods
Distribution of Governing Power
Apart from the rules that create the initial state, there is much to learn from the characteristics of the genesis moment. Who are the initial token holders and how are the tokens distributed? If the DAO implements token-weighted voting and the initial team holds 60% of the tokens, would it be fair to call that DAO democratic? If the tokens are distributed such that the initial state is highly fragmented, will the rules permit active participants to obtain a higher concentration of voting power over time?
The two exhibits below show examples of initial token distributions that are quite divergent. Rarible made 40% of its token supply available at launch and reserved 60% of supply for marketplace miners (platform participants). Rarible’s structure suggests that early voting will amplify the voices of the investors and team over airdrop recipients by a ratio of 3:1, but, over time, active miners (the participants on the Rarible platform) would have a more dominant voice in governance matters than any other group of participants.
Rarible Initial Token Distribution
Source: https://rarible.medium.com/introducing-rari-the-first-governance-token-in-the-nft-space-5dbcc55b6c43
On the other hand, Bitdao allocated the bulk of tokens available at launch to Bybit and reserved a larger portion for Bybit (60% of total supply) over a longer period.
Bitdao Initial Token Distribution
Source: https://docs.bitdao.io/litepaper-1/tokenomics
As the Bitdao example shows, the rules of distribution can drive outcomes. Based on the initial plan for token allocation, participants can gauge the degree of influence their voting participation can have on the DAO’s governance both at genesis and as the DAO develops.
Nudging Behavior
Apart from the assumptions that are used to create the DAO and the initial distribution of governance, another key aspect of the initial state is the means by which founders set an agenda. A DAO could conceivably begin as a blank slate, but the founders could populate the governance forum with a series of pre-written proposals for the initial participants to vote on in rapid succession as the DAO comes to life. Agenda setting has the power of suggestion: if the DAO creators put forth a scheme of rules for an initial vote without a clear alternative, they are likely to select for participants that gravitate towards approving those rules. Proponents of quadratic voting are likely to gravitate towards early DAOs that have an initial proposal to approve quadratic voting at the outset, and opponents will be likely to stay away.
In addition to setting the agenda, the creators can design behavior pathways that are more likely to create certain governance outcomes. An example of this is the process a DAO uses to permit delegation of voting power. For instance, ENS launched with an airdrop that required token holders to vote on the clauses of the initial constitution and to opt-in or out of making a delegation of their voting power to persons who applied to be delegates. The presence of delegates creates the possibility that even with a fragmented distribution of tokens, outcomes may be determined by a subset of core contributors. Delegation can also create unwanted incumbency: holders may delegate and disappear from governance, enabling delegates to retain power even after they fall out of favor with the community.
Concluding
Governance begins prior to DAO formation when the founding group decides to commit a vision to code. This vision coalesces in a founding moment that sets the DAO into motion and creates a degree of path dependency for the community. To evaluate a DAO’s governance, it is important to understand the degree to which governance mechanisms are predetermined or engineered from the DAO’s initial state. A DAO can be a blank slate or a fully designed parliament from the beginning. Beyond the form of governance that has been reduced to code, the initial state depends on how governance is distributed among different constituents of the DAO. The engineering of member onboarding has a meaningful impact on the types of members that gravitate towards the platform and how they decide to share power with other members of the community. Taking store of the documentation, distribution, and onboarding pathways of the DAO will give strong insight into what forms of governance will follow.
In the next installments of this primer, we will survey common forms of governance that have taken hold in DAO communities and assess the future of DAO governance.