Who is responsible for making the decision on how to make decisions? Researcher Kei Kreutler analyzes the decentralized, consensus-driven decision processes implemented in blockchain technologies from a more general perspective of governance. Such an approach allows for a more nuanced negotiation of agency, power, and stakes in decision processes for both technical and social organization.
One neuroatypical Roman emperor-to-be had a steep onboarding process for his court astrologer. Tiberius Julius Caesar would walk prospective astrologers along the coastline, drawing out their forecasts for pivotal events: a river flooding, a coup on the current emperor. As they wound their way to the highest cliff, he would ask them to look upon their own astrological birth chart and foresee their death, only to be thrown into the rocks below moments later.
It wasn’t until the astrologer Thrasyllus—who would go on to successfully predict the current emperor’s ousting and the rise of Tiberius—looked up from his own chart to say, “It looks like I’m in grave danger right now,” that he filled the court position.
While we look back condescendingly on the centuries when stars acted as movable spokes for power, the credibility we attribute to our own maps of the future may stand on equally influenced ground.
The Bitcoin whitepaper claimed blockchain to be a general solution for Byzantine fault tolerance in computer science, which takes its name from the game theoretical Byzantine Generals’ Problem.
The Byzantine Generals’ Problem has to do with agreement. The illustrative story tells of a group of commanding generals who encircle a city and have to formulate an attack plan. There are two options: attack or retreat, and the generals need to come to a consensus decision in order to make the plan effective. Further complicating matters, having surrounded the city, the generals are geographically distant from one another and to communicate their decision to the others, they have to rely on a messenger. A messenger who might take the fastest route through the city center to the other side might be efficient, but there would be no way of knowing if their message had been tampered with: Were they bribed at the gate? What if the messenger never arrives?
In computer science, “Byzantine fault tolerance” describes the dependability of a system to reach accurate consensus with potentially unreliable actors. The Bitcoin blockchain addresses this issue by assuming that the computational power required from a malicious actor to attack the network would be too great and too expensive to be able to override the other actors on the network.
Rather than seeing technical, incentivized solutions to social problems, we could alternatively read governance issues as archetypal—mutable and immutable, recurring, and taking intractable forms.
In the current debate around decentralized platforms, the question, Who is responsible for making the decision on how to make decisions?, comes into focus.
Nick Szabo, who happens to be the originator of the term “smart contracts” and the second likeliest suspect behind the pseudonymous Satoshi Nakamoto, wrote:
While the arguments around governance abound, the question of what’s at stake in the discussion remains hazy. Depending on who you ask, it’s anything from the future of all asset exchange or organizational design to simply the fates of a small group of developers and a few entangled bystanders arguing over software updates. Either way, “governance” and “game theory” are terms that can be easily deployed to obfuscate a task at hand. In computer security, there’s a concept of “attack surface,” the total possible attack vectors in a technical system—that is, hackable entrance points for data entry and extraction. It’s best practice to keep the attack surface as small as possible. A simple metaphor would be how many doors with faulty locks one house has.
Szabo writes that where in security there is an attack surface, in questions of governance, there is an “argument surface”: the total optionality in a given system or organization. That is, each next step or action that can be enacted in more than one way opens the space for deliberation and argumentation, and, well, perhaps it would be better to limit the argument surface, too, by off-loading as much as we can to a technical system, in order to keep acting. For example, if the amount of transactions processed in every “block” had to be decided by the community, it would open a vast possibility for argument and contention. Instead, what is referred to as the “block size” is technically predetermined by the protocol.
There is, however, something like a third way: an inexact science for navigating the Tower as tragedy of the commons, as the community-owned resources lie misused all around you, while the assembly meeting goes two hours over schedule. The importance of what Szabo highlights in any governance framework’s “argument surface” is something evidenced by the backlash against Ehrsam’s post calling for urgent action on blockchain governance: who can claim the ability to make the first decisions and who can continue to custom frame the governance issues. While perhaps lack of governance has seen the radical politics of early Bitcoin become stalemated into creating privately distributed ledgers for improving banks’ paperwork, there is a tendency of governance creep in any fledgling organization or stewarded resource, in which the frequency of discussions on governance processes has an oddly distinct relationship to the effectiveness of putting said governance in place and avoiding external capture.
When all basic decisions start becoming decisions of governance, when all questions are questions of agency to make a decision, a curious epiphenomenon occurs. It’s here that totalizing jurisdiction approaches the asymptote of impossibility of action.
The Byzantine Generalization Problem describes when all issues in an organization or technical system are framed as governance issues. Any question that’s brought before an organization can become a question of who has the agency to provide the solution. Even if it’s fixing the broken plumbing in a housing co-op, when framed as a governance issue, the debate on who should take care of it could last for weeks as the pipes leak and the toilets fail to drain. The generalization of governance can be applied to every challenge an organization faces, especially when there is no clear—or technically embedded—solution at hand. When the complexity of the organization grows, this “generalization problem” applies exponentially, with considerably less clarity on the necessity of action than in the example above.
A DDoS attack on governance, the Byzantine Generalization Problem is the opposite means to the same end of the tyranny of structurelessness. In consensus culture, often the most indefatigable wins, and power is implicitly consolidated by those who provoke interrogations of process. Perhaps a social formula lurks that can be translated fully into equation—the seat atop the precious bell curve in the pseudo-chart above—that would lead to transparency, clarity, and appropriate calibration of deliberation on governance systems. Everyone has their opinions on how to solve it regardless, ad nauseam and ad infinitum.
That said, as we learn from programs that sponsored the use of consensus, the Byzantine Generalization Problem can be used as a strategy to delay decision-making when necessary and to tactically, obliquely question the right to power. Ambiguous, unattended responsibilities are the “attack surface” and way in.
Another strategy could be to start with a given end goal and seek to map the context and required actions for it to be reached. Such a theory of change relies on an adequate mental model of the environment—with the corollary that the greater the complexity of the environment, the more obfuscated its model’s underlying assumptions are—and importantly, it relies on formulating a multiplicity of understandings of how change may occur in the environment. While governance, incentives, and protocols may be the most explicit frameworks to guide organizational action, there are always other tools available and already at play.
A theory of change can metabolize its environment through time and narrative, incorporating strategies with ulterior rhythms and cloaked intentions. The most efficient path from A to B may be the most hidden and illogical one, based on learning acquired from missteps and faulty data. In the pace-laying diagram above, we can conceive of the different rhythms of change—from the slow yet impactful arc of culture and the informative flippancy of fashion—and formulate narrative strategies and technical impositions intersecting with each.
Aesthetic narratives and false juxtapositions propel change equally as, if not more than, governance protocols do. Every narrative can be framed in every light. Ethereum, a blockchain platform and protocol that launched in 2015, has seen its own innumerable iterations and adaptations since. Each Ethereum logo has its own part to play, or something. Trusted governance and authenticity must be considered a question not solely of explicit or defined processes but also of subtler forms of manipulation.
It’s become critical to realize that there has always been fake news, and moreover, that exposing the most accurate information will not in any way necessitate it being acted upon. Fake news—what previous generations may have confined to an understanding of propaganda—is the inevitable result of both massive media industries and small-scale social gossip that travels from one town to the next. Misinformation often comes in gradients, and its propulsion can be as similarly effective as diverse aesthetic and implicit narratives. The Truth is always preceded by a bounty on its name.
Each time you point a finger at falsity without offering a significant, desirable counter narrative, politically or aesthetically, you grant the central authority of truth-making a wider warrant to dismiss all facts.