Ever since Jacob Horne bombastically demanded we all apply mathematical models of thinking to cryptoeconomic structures sometime in early 2022, I have been deep in thought. If I’m being honest, it’s truly just been utterly captivating to me. Horne’s structured philosophy of hyperstructuralism is both elegant in its premises, easily adaptable to new paradigms within the cryptosphere, and endlessly scalable in modeling ever-growing structures. That said, I approach with a singular critique that perhaps the name is a bit too grand for what it accomplishes as it stops more or less at the protocol: No meta-entity could ever truly embody the label of being a hyperstructure by definition, as the network effects of the DAO may necessitate governance over non-hyperstructures simultaneously with their true cousins. I contend that hyperstructures can only ever exist at the DAO specific level (that is, that entity governing a protocol) and could never exist outside that: They are too narrowly defined to extend to higher levels and thus, I find the hyperstructure model to be limited in application beyond the singular, individual DAO. In this article, I hope to introduce a more defined set of models for metagovernance, what that looks like, how it might work, and thereby define the next logical step: Metastructures.

Since the original thesis on hyperstructuralism was introduced, the landscape and thought leadership around DAOs has shifted. Though this was originally applied to be protocol specific, I believe that as more and more protocols are governed by DAOs our mental models around the protocols should also extend toward the architectural models that govern them. Thus, hyperstructuralism should, in theory, apply equally to both the protocol and its governing DAO (which it tends to). What hyperstructuralism fails at is in the extended fringe: It cannot logically define the “DAO of DAOs” as the grail of my field has been recognized as. The properties of hyperstructures are defined by Horne as follows:

1. Unstoppable: the protocol cannot be stopped by anyone. It runs for as long as the underlying blockchain exists.

2. Free: there is a 0% protocol wide fee and runs exactly at gas cost.

3. Valuable: accrues value which is accessible and exitable by the owners.

4. Expansive: there are built-in incentives for participants in the protocol.

5. Permissionless: universally accessible and censorship resistant. Builders and users cannot be deplatformed.

6. Positive sum: it creates a win-win environment for participants to utilize the same infrastructure.

7. Credibly neutral: the protocol is user-agnostic.

I posit that while hyperstructures are a desirable state, not all DAOs (nor their protocols) can be considered within this framework. A working example may be FREE DAO which not only fails property 3, it takes an active stand against the creation of value at all. In the days after the formal creation of the DAO actions were deliberately communicated to reduce the value and create a lock in social value instead. It is a case study in removing exitable value within a DAO for the purpose of social good and welfare, but also means that the FREE DAO and the associated token are not hyperstructures. They are, however, DAOs, still decentralized and run organically. I believe, then, that a higher model is needed, not as replacement theory, but rather as a unifying theory that allows hyperstructures to exist within the broader ecosystem of DAOs alongside other protocols that may not fully conform to a hyperstructure model.

Before going any further, it is important to understand why this is needed. DAOs are an organic architecture that extends from a collaborative interface only and builds up from there, eventually adopting governance through a protocol. They are a bottom up exercise in collaborative governance. At this stage, the early stage of a budding DAO, one can envision them as a flat surface level of hierarchy of stakeholders, the so called Core. These stakeholders exist in equal on a plane and are connected at all nodes. Eventually, this Core begins to expand. How this expansion is modeled is essentially the definition of governance, and it can look like many different things depending on the governance framework the DAO has adopted. Certain models may simply broaden the circle of nodes, adding on layers of concentric nodes and creating flat, solar geometries. Other models will branch decision trees, with nodes acting as hubs and spokes for collaborative meshes. Whatever governance framework drives the ultimate model, this is the model of what a DAO “is” per se: This is the DAO mesh, a collaborative network centered around a Core executing the outcomes of a self governing framework. The nature of what that looks like matters not for our purposes here.

Now, if the reader has been paying attention to DAOs at all over the past several years then none of this is going to be revolutionary at all so far. In fact, this is the architecture of a hyperstructure’s governance model, more or less. While a hyperstructure explicitly defines the underlying protocol, the scope of that definition ends here: At the DAO mesh, neatly governing a protocol hyperstructure. Congratulations: We have reductively reasoned our way from a hyperstructure to a DAO! But what happens if I push the model a bit further?

DAOs are naturally equal and decentralized through token ownership. This is what makes them a flat mesh: Anyone can swap into the token at any time and thereby gain governance powers (assuming a purely optimistic protocol for the sake of ease). This is an acceptable hyperstructure as it is purely equal in nature. But what if we don’t think of the DAO as only a mesh. What if we push the model into three dimensional space where each node within the DAO can be zoomed into and out from with equal outcome. For example, from the Core with its mesh of nodes, one might pick any singular node and “zoom” into it, revealing smaller and smaller branches. These are the sub-DAOs, the guilds, the channels. Singular units of micro-collaboration that can both organically and synthetically sprout within the confines of the DAO architecture, still governed by the token and executed by the Core. Within this model, the established geometry no longer works. Instead, the “mesh” of a DAO is actually a fractal geometry of endless scope. A DAO can grow and shrink and copy and mirror itself in perfect unison endlessly and at any arbitrary scale. Again, this is not new thought: Rowan Yeoman modeled this DAO architecture in June of 2022 after recognizing that hierarchical delegative structures become incomprehensible when viewed on a flat plane.

Even still, the hyperstructure model can apply here, as (and this is where I may start to become contentious in certain schools of thought) I believe the hyperstructure can scale infinitely well within the fractal one way, that is, down the pipeline from the Core. Though we have understood within the industry of DAO architecture that DAOs are less entities and more flows due to the decentralized, permissionless nature of their resources, the fact remains that at some level an entity remains in the form of immutable governance (smart contracts and the constitutional tokens that reign over their interactions). It is this entity that the hyperstructure essentially is. It therefore exists as a governance to any further sub-DAO that may form. Each sub-DAO acts in the good of the parent, utilizing the same token for ownership and governance within the broader structure. Now of course this sub-DAO could form their own token or token derivative and begin governing in some way independent of that of the parent structure but at that point, would we include them within the parent hyperstructure of the DAO or would they become their own separate structure? I argue the latter, that DAOs exist and sub-DAOs exist, but we do not yet have an adequate term for what is beyond that. If a sub-DAO splits off, it simply becomes its own DAO in terms of governance. But what if it is still inexorably linked to the parent DAO?

Imagine the following. Suppose a sub-DAO, a node, breaks away from the Core, deploys new smart contracts, and sets up new governance. However, in the course of setting up this new governance, a key requirement was that the new token model would be based upon the original ownership of the original parent token. All original holders that split would be able to redeem their current tokens for a distribution of the new governance token. What’s more, just to make it interesting, this new token is pegged to the old one, or some other mechanism of linking them one might imagine. The point being, is this a unique DAO? Is it too linked to the parent? Is it its own hyperstructure? Perhaps, however I don’t believe the model of the hyperstructure works to describe the relationship between these two entities.

This is the DAO of DAOs: As DAOs are collaborative Cores surrounded by nodes, and also fractals of similar systems, they also scale oppositely where DAOs will inevitably interact with other DAOs through resource sharing (as DAOs are naturally permissionless, broadly speaking). This completes the fractal model, but also indicates a fabric, where the resources of one DAO, governed by one token ecosystem, may also flow to the ecosystem of a completely different DAO. This is the concept of metagovernance and is, in my opinion, above the realm of hyperstructures. Metagovernance details the doctrines and contracts that govern the interactions between DAO resources, and metastructures are the architectures behind them. These structures exist as a means of extending the collaborative nature of the DAO to the interoperation of permissionless resources between DAOs.

Metagovernance already has a proper definition, then, however it remains strikingly unclear how it would evolve or what it would look like. I believe this is an important realization because ultimately metagovernance can’t be ignored and so a structure must in place to manage it. This is derived from an existential premise of necessity: I believe that DAOs are necessary. If we are to live in a global community (enabled by web2 and the early invention of networked communication), and if we are to accept that AI will become a driving economical force (enabled by the development of the transformer model of computing), then we must also accept that permissionless work is the only viable competitive ground for employment going forward. DAOs enable this, but we remember that DAOs are not strictly speaking entities, but rather a descriptor of resource flow, a mental model for explaining the schema behind a protocol. Because DAOs exist as a permissionless flow within many legal structures and frameworks, then it is reasonable to assume there must be a way of governing these flows to ensure fairness and stability of the underlying system.

If a hyperstructure of a DAO can be described through the model of mathematical fractals, then the metastructure of the collective conglomeration of DAOs could perhaps best be described through the model of a Matrioshka brain. In a Matrioshka model, a megastructure is created to output compounding energy from the combined output of internal Dyson spheres. Within this model, the defined points of the spheres are less important than the output of the whole megastructure with the internal resource sharing being both individually necessary while also belying this complexity with the visage of a unified entity. The output of each internal sphere is extremely important work as it drives the output of the whole and without this output the Matrioshka structure would fail to meet its threshold, however this output is also meaningless on its own. The model adds an aspect of collaborative dimensionality as a fourth dimension to the DAO fabric, and I think it fits quite nicely as a way of thinking about DAOs as a collective: Each DAO is independently a fractal hyperstructure, infinitely deep, however it exists in a soup of resource sharing to create singular outputs in the form of an overarching collective: The Metastructure.

Metastructures are an evolution over hyperstructures, however they are not a replacement for the theory. As such, they should abide by their own unique properties and while some of these are shared with other structure types, others are not. Fundamentally, metastructures differ in that governance cannot exist within a vacuum. The Core of every DAO is a necessary evil in the current state of DAO tooling. Ideally, a DAO is quite literally just a smart contract that governs the actions of resource flows through stakeholder alignment. The realities are that off-chain executive action is still required if for no other reason than to build tooling, enact policy, and direct legal action (thus the headless foundation legal framework we’ve seen used to great success in some case studies like ENS). In a metastructure, however, this Core doesn’t exist because it simply can’t. A single off-chain executor could not impartially govern the actions of many different governed hyperstructures effectively without recreating the hierarchy problems that DAOs originally sought to solve for. But as they are a collective brain encompassing the DAO fabric, this is unbothersome for our modeling.

Now that we have at least some model to guide the mind on this thought experiment, let’s begin to more clearly focus on what a metastructure would look like from the lens of metagovernance. In its simplest primitive, metagovernance is really just the field of how multiple DAOs might work together to generate cohesive results. As an example, let us fling ourselves a bit into the future and imagine for a moment that several nation states are governed as DAOs, with a massive and carefully written smart contract at the very center that executes policy automatically based upon whatever governance framework the DAO has adopted (this isn’t important and that is precisely the point as the reader should eventually see). In this world, where nation states are now hyperstructural DAOs, we encounter some very simple logical problems that are very real and easy for anyone to ask: How does international trade occur? How are treaties signed? How are partnerships formed? How are territories defined? And let’s not forget to zoom in. Since this is a world of DAOs, legal structures are vastly different. Remember, DAOs are both fractal and a flow, and since hyperstructures imply permissionlessness, it stands to reason that in this world resources could move freely between these nation states and the various sub-DAOs within them. Someone could very hypothetically do work for one DAO in one nation and hold tokenized governance in another DAO in another nation, and could do so without ever once deanonymizing themselves. Such is, of course, the power of web3 and decentralized on-chain identity, but the point is: How would any of this work? Since these are the questions to solve for within the model of a Matrioshka brain encompassing set of fractals, we can infer some things about metastructures:

1. Unified Emergence - Like the Matrioshka brain harnessing the collective energy of nested spheres to achieve a singular, monumental output, the metastructure derives its essence from the unified emergence of interconnected DAOs. This principle asserts that the collective whole transcends the capabilities of individual components, enabling feats unattainable in isolation.

2. Recursive Interdependence - In the layers of a Matrioshka brain, each shell relies on the inner ones for sustained energy flow, creating a symbiotic relationship. Similarly, metagovernance is founded on the recursive interdependence of DAOs, where each node both influences and is influenced by others, embodying the notion that governance is inherently interconnected across all scales.

3. Decentralized Cohesion - The Matrioshka brain operates without a central core dictating its function; instead, cohesion arises from the harmonious alignment of all layers. Metagovernance must embrace decentralized cohesion, ensuring that collective order emerges naturally from the distributed interactions of autonomous DAOs, adhering to the first principle of decentralization.

4. Fractal Scalability - Reflecting the fractal nature of both DAOs and the Matrioshka brain, where patterns repeat at every scale, metagovernance should inherently support scalability. This principle maintains that governance structures remain effective and coherent, regardless of the metastructure’s expansion or the complexity of its constituent parts.

5. Adaptive Resilience - The layered design of a Matrioshka brain provides resilience against disruptions in any single sphere. Metagovernance must prioritize adaptive resilience, enabling the metastructure to absorb and adapt to changes or adversities within individual DAOs without compromising the integrity or functionality of the whole.

6. Symbiotic Resource Flow - Just as energy seamlessly transfers between the nested spheres of a Matrioshka brain, metagovernance relies on the symbiotic flow of resources between DAOs. This principle is rooted in the understanding that collaborative sharing amplifies the collective capacity and efficiency of the metastructure.

7. Intrinsic Neutrality - The Matrioshka brain does not favor any particular layer; each contributes according to its nature and position. Metagovernance must uphold intrinsic neutrality, ensuring that no DAO is privileged over another, and that the metastructure remains impartial, fostering an environment of trustless trust.

8. Sovereign Autonomy within Unity - While each sphere in the Matrioshka brain is integral to the whole, it maintains its distinct existence. Similarly, metagovernance acknowledges the sovereign autonomy of individual DAOs within the unity of the metastructure. This principle respects the unique identities and needs of each DAO while harmonizing them within the larger collective.

9. Ethical Pluralism - The Matrioshka brain functions without imposing a singular ethical directive upon its layers. Metagovernance should embrace ethical pluralism, recognizing and accommodating the diverse moral frameworks of participating DAOs. This principle asserts that cohesion does not necessitate ethical homogeneity but thrives on respectful coexistence of varied values.

10. Anticipatory Hostility Awareness - In the vast expanse of the Matrioshka brain, safeguards against internal anomalies are implicit. Metagovernance must be designed with hostility awareness, presuming potential malicious intent as a first principle. This proactive stance ensures the metastructure’s resilience by integrating defense mechanisms against systemic risks without relying on optimistic assumptions.

11. Permissionless Evolution - The Matrioshka brain’s architecture allows for the natural addition or removal of layers without disrupting its overall function. Metagovernance must facilitate permissionless evolution, where DAOs can join or depart the metastructure freely, upholding the decentralized ethos while ensuring that such changes do not destabilize the collective consensus.

12. Transparent Interoperability - Transparency in energy transfer between the spheres of a Matrioshka brain is essential for its operation. Metagovernance relies on transparent interoperability, where the interactions between DAOs are open and verifiable, rooted in the principle that clarity fosters trust and effective collaboration across the network.

13. Immutable Foundations with Dynamic Adaptation - The core laws governing the Matrioshka brain are immutable, yet the structure adapts dynamically to optimize function. Metagovernance should be built upon unchanging foundational principles—such as decentralization and autonomy—while allowing for dynamic adaptation in governance practices to meet the evolving needs of the metastructure.

14. Collective Intelligence Amplification - The computational prowess of a Matrioshka brain arises from the cumulative contribution of all its layers. Metagovernance seeks to amplify collective intelligence, leveraging the diverse insights and innovations of each DAO to enhance the problem-solving capacity and evolutionary potential of the entire metastructure.

15. First-Principle Integrity - At its core, the Matrioshka brain adheres to fundamental physical laws that govern its existence. Metagovernance must be grounded in first-principle integrity, ensuring that all governance models and decisions are derived from basic, self-evident truths that maintain coherence and alignment throughout the metastructure.

These attributes allow us begin to envision a framework for how metagovernance might emerge, but before that vision can be fully realized we must first consider the evolution of DAOs from isolated entities into interconnected nodes within a vast metastructural network. This metamorphosis is not instantaneous but a gradual process of progressive decentralization. It is through this lens that we can begin to architect a framework capable of supporting autonomous DAOs operating seamlessly within a global metastructure.

At the outset, DAOs emerge as singular hyperstructures, each self-contained and operating under its own governance protocols. However, as these organizations recognize the limitations of isolation—such as resource scarcity, innovation bottlenecks, and systemic vulnerabilities—they naturally gravitate towards collaboration. This is the first phase of progressive decentralization: interconnection through mutual benefit. This is observable organically within the current sphere of the DAO ecosystem and is a natural property of emergence within DAO theory (I briefly touched on this subject during my work on the GET Protocol DAO architecture). In this phase, DAOs begin to establish bilateral relationships, sharing resources and expertise to enhance their individual capacities. These initial connections are governed by smart contracts that enforce agreed-upon terms, embodying the principle of symbiotic resource flow. Importantly, these contracts are designed to be adaptive, allowing for modifications as the needs of the DAOs evolve, while also retaining their immutability on chain. This point is a crux to the theory of DAOs and metagovernance as a whole: The immutability of the contract is what builds trust in the system, the “smartness” of the system is what enables automation and unlocks AI governance, and the ability to enact changes through redeployments is what maintains human value within the system and unlocks the new worker economy. As the number of interconnected DAOs increases, a network effect takes hold. The relationships between DAOs become more complex, necessitating a shift from bilateral agreements to a more systematized framework of governance. Here, the metastructure begins to take shape, guided by our first principles.

The metastructure operates on the foundation of decentralized cohesion, where order and governance emerge from the collective interactions of DAOs rather than from a central authority. This is an important philosophical distinction that enables world governance to coalesce around artificial intelligence while still maintaining individual sovereignty amongst stakeholders (in this case, nations, states, cities, conglomerates, and individuals). Smart contracts and AI play pivotal roles in this system, automating governance processes and enabling fractal scalability. Each DAO retains its sovereign autonomy within unity, contributing to the metastructure while preserving its unique identity and objectives. In this envisioned future, AI agents govern DAOs autonomously, making decisions based on predefined protocols encoded within smart contracts. Individuals interact with these AI agents not by direct control but through governance mechanisms that influence AI behavior. This aligns with the principle of collective intelligence amplification, where human insight guides AI decision-making to achieve optimal outcomes. The AI agents are programmed to adhere to metastructural principles, ensuring that their actions contribute positively to the metastructure’s goals. This is likely going to be expanded upon organically over the lifespan of the various nodes within this fabric as synthetic data facilitates the natural evolution of these agents. They facilitate transparent interoperability by maintaining open records of transactions and decisions, all recorded immutably on-chain. This transparency is crucial for building and maintaining a state of “trustless trust” across the network and, ultimately, completes the metaphor of the Matrioshka structure.

Of course, for the metastructure to function effectively, there must be standardization of protocols and interfaces. This doesn’t imply uniformity but rather compatibility. By adopting interoperable standards, DAOs can communicate and transact seamlessly, respecting the principle of intrinsic neutrality. When viewed through the lens of the Matrioshka brain, we can clearly see what the implications of this system are and where it leads: The end goal of metagovernance should always be that of human flourishing and it thereby removes the focus on the needs of individual nations and realigns incentive drivers to a common good. This is achievable because of two reasons. First, the use of AI as the executor implies an automated commonality that adheres to purely logical paradigms for success and eliminates human and nationalistic biases in the process. Second, human governance where stakeholders are commonly aligned via effective consensus models allows humans to always guide the system toward individual and nationalistic needs for their respective communities without influencing the common good. Balance is achievable through this system via the bicameral nature of it, creating room for disagreement and unique alignment needs while still reaching flat consensus.

With all this in mind and given the dynamic nature of both technology and societal needs, the metastructure must employ adaptive governance mechanisms. These mechanisms are designed to respond to changes proactively, guided by anticipatory hostility awareness. By assuming the potential for malicious intent, the metastructure incorporates robust security measures from the outset, enhancing adaptive resilience. Decision-making processes within the metastructure leverage decentralized consensus algorithms that are scalable and efficient. AI agents play a role here as well, analyzing vast amounts of data to provide insights that inform collective decisions. The use of AI in governance must be transparent and accountable, ensuring that the metastructure remains aligned with its foundational principles, again building on itself recursively via synthetic data that will graduate the autonomous nature of the metagovernmental structure.

A critical aspect of the metastructure is its embrace of ethical pluralism. In a global network of DAOs, cultural diversity is not just inevitable but desirable. The metastructure must, therefore, facilitate an environment where differing ethical frameworks can coexist without conflict. This is achieved through governance protocols that prioritize mutual respect and non-interference, allowing DAOs to operate according to their values while contributing to the collective good. Ultimately, this comes down to the influence exerted within individual DAO frameworks as they operate as nodes within the greater metagovernance structure; individual DAOs need not worry about their unique impact within the metagovernance framework as that is naturally accounted for within the consensus mechanism. DAOs, even at a meta level, are natural consensus-extraction machines. As long as consensus is reached within the node, it can also be reached at higher and higher levels of abstraction while still preserving the aligned goals of all original stakeholders involved. If this sounds utopic rest assured it is not: DAOs build consensus naturally and handle conflict elegantly when designed according to proven frameworks and voting curvatures. One might imagine a form of optimistic governance structure applied at the meta level that is able to deftly handle these kinds of issues with ease, though this will obviously need to be hammered out in extreme detail.

It is this natural gravitation toward incentive alignment that allows the metagovernance structure to become sustainable in the long term. Aligned incentives ensure that DAOs are motivated to act in ways that benefit both themselves and the metastructure. Tokenomics play a significant role here, with value flows designed to reward contributions to the network. This financial element of the theory of DAOs is partly what drives the magic that makes all this feasible in the real world: DAOs naturally create a new kind of economy driven by a new kind of value hard linked back not to production of goods, but rather to the attention given to governance and is what drives the future of work in an AI age. By attributing value to attention a metagovernance can ensure sustainability through incentive alignment amongst the nodes, even when disagreements arise (I personally find holocratic models to be especially adept at this as they create a market out of the process of governance rather than the pre- and post-result and are especially resilient at edge cases against bad actors). These economic models must be adaptable to accommodate the diverse needs and contexts of participating DAOs. By embedding economic sustainability into the metastructure, we ensure its longevity and relevance in a rapidly changing world.

That said, in any complex system conflicts are inevitable. The metastructure must, therefore, include mechanisms for conflict resolution that are fair, transparent, and efficient. Smart contracts can automate certain dispute resolution processes, but human judgment may still be necessary for more nuanced cases. This is a fundamental principle within the theory of optimistic governance, a proven theory used to great effect within Aragon and the Optimism Collective, that essentially assumes a positive outcome for efficiency, but allows human interventions to occur prior to execution and thereby further rewards attention with value. I am not completely convinced that an optimistic framework is the pure ideal, but elements of it should always be considered in general as it is the most cutting edge theory of governance that avoids gridlock within a DAO while still synergizing both human and autonomous agents (I have also done work on experimental models such as probabilistic quadratic voting curvatures that I believe are also viable, perhaps more so, to this end so the options are there and very real, but more research must be conducted in this space).

As the metastructure matures, the concept of nation-states governed as DAOs becomes not just plausible but practical. In this scenario, traditional governmental functions are encoded into smart contracts, and AI agents manage day-to-day operations. Citizens interact with their government through decentralized platforms, exercising governance rights via tokenized systems. To each individual stakeholder (corporate, individual, or otherwise), the model will always appear to be flat, a fabric of nodes with some minor exposure to other nodes as needed and the metagovernance never fully exposed or in view unless specifically sought out. Zoom out and you would see the three-dimensional fractal nature of the Matrioshka model. Metagovernance provides the framework for these nation-state DAOs to interact with one another, facilitating international relations, trade, and cooperation. Sovereign autonomy within unity becomes a geopolitical reality, where each nation-state DAO maintains its independence while contributing to global stability and progress. It is therefore possible in my mind to achieve a global form of governance that is “for the people” without negatively impacting any one singular nation’s own sovereignty.

Metagovernance will exist and the structures that make it up will likely form organically to a certain extent. This is a given as more and more entities progressively decentralized and come on chain and just as the Matrioshka Brain need not centralize the whole of the structure to output (instead, operating on the aligned incentives of the unique nodes within the sphere), so too will metagovernance need no central authority or a Core of planners to produce. Instead, the viability of such a system will rely upon solid, well structured DAO design at the node level. Standards must be developed for structuring DAO frameworks that are drop-in, yet scalable and resilient. I will likely revisit this topic in a future article, but for now, it is important to start thinking about the meta-level impact of DAO development and what that looks like at the highest levels of abstraction so as to not be blindsided by its emergence. It is my hope that this article will provide a foundation for how to think about metagovernance and how to model it appropriately in ways that make sense going forward, not necessarily to provide an exacting how-to guide as I believe metagovernance is best left to organic emergence, not authoritative planning which is a dangerous, deeply dangerous, route to go down.

In the meantime, gm friend. The world of the future will be better. Let’s build towards that starting now.