At what point does reaching for an ideal leave one at odds with reality? And can such delusion be justified, in some cases, insofar as it may drive progress?

The ‘D’ in ‘DAO’ stands for ‘decentralized,’ and yet the vast majority of groups calling themselves ‘DAOs’ are not. This isn’t a judgement, but more of an observation. Most of these groups, across their variety of missions, arguably would not even benefit from decentralization.

One could understand the quality of ‘decentralized’ as relatively less centralized than traditional organizations, or as lacking a center altogether. This semantic ambiguity might be part of the problem, in terms of how these groups understand and present themselves. That said, the deeper part of the problem is arguably the entrenched and categorical fixation on decentralization as an end in itself, irrespective of how useful it is in the context of a given enterprise.

In this piece, I will attempt to diagnose this widespread fixation, prescribe some potential ways forward, and explain why we may want to limit our usage of the term ‘DAO’ in the industry, in light of active legislative developments.

Decentralization as a Horizon

One could argue that the appeal of organizational decentralization is a consequence of popular dissatisfaction with the corporate and political (meatspace) status quo, where executive corruption and unaccountability can leave the majority of people effectively without a voice, but with the bill. Here, the advent of DAOs can be seen as consistent with other technological trends, such as blockchain in general, distributed computing more generally, and the open-source movement at large.

Alas, we may generalize beyond even technological movements, as we consider the appeal of decentralization. The utopian vision of communism can be articulated as a completely horizontal society, wherein some ideal relations of production entail no extraction of surplus value from laborers - wherein even the potential of such extraction is precluded.


Well I’m no historian, so take this with a grain of salt. Not only have we never reached such a societal state at scale (unless some example is lost from our history, or unless I’m simply unaware of it), but arguably there was never even a proper roadmap laid out. The closest thing we got to a roadmap, in hindsight, was the revolutionary vanguard leading the proletariat to “seize the means of production” and install a socialist state, only to have it devolve into a dictatorship of the proletariat which fails to self-abolish and usher in a utopian regime approximating the visions of Marx, and before him of More, and before them of probably many others.

Of course, the DAO movement, being rooted in libertarian technology which takes for granted the free market conditions of a liberal economy, is arguably at odds with one of the essential steps toward communism, namely the abolition of the private sector. That said, the values placed on community and collectivism do live on in the DAO space, and even manage to co-exist with values of individual liberty and sovereignty.

This is all to say that the appeal of decentralization, even as some virtually unreachable horizon which nonetheless motivates apes and apparatchiks alike, is nothing new, and indeed drove much of our most explosive history - that is, until our history ended.

Return to Reality

There is a strong argument to be made that our collective insistence upon striving for decentralization as an end in itself, despite its current operational inefficiency, will force the development of technology which enables decentralization to become more efficient. That said, there is also the current reality to reckon with, namely that “decentralization” usually means unproductive arguments on message boards - in other words, as has been joked, Dudes Arguing Online.

This reckoning is two-fold. Firstly, we should acknowledge that, regardless of what we should aim for, the current reality is that most DAOs are centralized, whether that takes the form of a multisig, admin access to a protocol, or control of a Twitter account. Secondly, we should acknowledge that decentralization isn’t worth aiming for in every situation, indiscriminate of circumstance, and that sometimes centralized initiative is the pragmatic route.

How can we preserve the promise of decentralization-enabling technologies, without sacrificing operational efficiency in the meantime? A new mental model, and maybe some new terms, can help.

Instead of using the term ‘DAO’ as the umbrella term for organizations with tokenized membership, multisig treasuries, or other web3 mechanisms, let us consider using a more general term, such as ‘on-chain organization.’ Some on-chain organizations may embrace various degrees of centralized governance, while others may have a use for, or an interest in, a decentralized governance model. DAOs would fall under the latter half of this umbrella.

This is more than a terminological shift, however. It is a recognition that decentralized governance is not categorically a best practice, and that we shouldn’t all keep pretending it is. That said, we can also recognize that the long line of dead vaporDAOs occasioned a great deal of learning. The “Cambrian explosion” of decentralized governance experiments has led some, with a fresh take on first principles, to the conclusion that centralization isn’t always a bad thing.

One could view this developmental trajectory (a departure from the centralization of orthodox business management, into the on-chain frontier to battle-test decentralized governance models, and back toward the recognition that centralized organizational practices do have value after all) as an integration of the opposite and a dialectical return to a more informed middle way.

Words Matter

Within the operational contexts of DAOs, much of this may reduce to immaterial pedantry, but in light of active policy developments, such as those from our federal legislature here in the US, these distinctions can have hegemonic consequences.

There was a provision on DAOs in the Lummis-Gillibrand bill, linked to above, which reads as follows:

Of course, this definition is nested in an understandably circuitous index of interdependent definitions for novel terms and technologies, and believe it or not, our actions in the industry, and our public research efforts, are informing these definitions.

One could argue that the best way to regulate on-chain organizations is to address the various technological mechanisms individually, as a growing suite of tools and practices for organizations of all entity types which operate on-chain, rather than trying to cover this whole field of organizational practices with the single term ‘DAO’ - a term which is a veritable misnomer most of the time, no less.

Some states have introduced entity types explicitly intended for DAOs, such as DAO LLCs and LAOs (Limited Autonomous Organizations). This is one approach within the larger challenge of reconciling the ideals of decentralization not only with the imperatives of operational efficiency, but also with the imperatives of compliance within regulatory regimes which require certain levels of organizational centralization in order to limit personal liability.

In other words, by and large, the main tradeoff for decentralized organizational structure is the absence of personal liability protection. Now, not only am I not a historian, but I’m also not a lawyer. That was historical insight, but this is not legal advice.

This is all to say that our continued generalized use of the term ‘DAO,’ as overfit as it is to only a subset of the variety of on-chain organizations to which it indiscriminately refers, risks playing a determining role in the development of federal policy which is similarly overfit to a subset of on-chain organizations.

In Summation

The DAO space is a cultural inheritor of a profound lineage of movements and ideologies across human history, and constitutes a unique culture occupied by people who are disillusioned with the world of today, and are thus trying to build a better one.

As we build, we leave a wake of case studies and precedents which inform the legislative efforts defining the law of this frontier, and gradually settling the wilderness. Sub-optimal policy here may have far-reaching liability ramifications for those in this space, an effect which may, accidentally or intentionally, grind this movement to a halt.

If most DAOs aren’t decentralized, and wouldn’t even benefit from actually being decentralized, then our ecosystem as a whole is sending delusional and misguided messages to legislators who need clear signals to pass healthy policy. We need healthy policy if we want the ethos driving this technology to realize its potential.

Its up to us whether our space continues to be labeled by a misnomer, whether this technology’s vapid speculative applications continue to outshine its innovative public goods funding applications, and whether it all ends up being swallowed by the portfolios of profit-maximizing asset management firms. We have an opportunity to dispel our delusions, having learned from the delusions of communism, and to build a better kind of capitalist democracy, where individual liberty and communitarian solidarity may enjoy a perhaps unprecedented parity.