Exploring Decentralization: Blockchain Technology and Complex Coordination
“Our model of the cosmos must be as inexhaustible as the cosmos. A complexity that includes not only duration but creation, not only being but becoming, not only geometry but ethics. It is not the answer we are after, but only how to ask the question.”
—Ursula K. Le Guin, The Dispossessed1
The late science fiction author Ursula Le Guin describes a dynamic framework of knowledge that prioritizes reflection and self-organization. The irony of her beautiful description, of course, is that a model is not supposed to be as complex, as “inexhaustible,” as the system it describes. It’s supposed to summarize, streamline, and help us describe and manage the messy, complex, dynamic, real world. In the words of political economist Elinor Ostrom, describing the temptation to oversimplify and idealize: “The power of a theory is exactly proportional to the diversity of situations it can explain.”2 Reductive descriptions, or models, are essential for building our human understanding of the universe, and we continue to build tools to help us refine them in hopes of arriving at a more perfect description, at something like an answer.
The concept of the Singularity imagines human intelligence and knowledge narrowing to, as the term itself suggests, a single point. The standard model of particle physics (while incomplete, it is the most complete model we have for describing the observed physical cosmos) suggests that the entire physical cosmos was once a single point, but what came before, we cannot say. The math breaks down. And what comes next, it stands to reason, may be up to the machines to discover without us. If the Singularity indeed connotes a narrowing, in which human minds and actions begin to seem obsolete (or worse, the subjects of new oppressions), then we have ample reason to be cautious, and indeed pessimistic. But artificial intelligence, as Ito observes, might also augment the human capacity for reason, establishing a symbiosis that enriches the entire ecosystem.3
Computers and digital models are a long way from processing comprehensively the messiness of the “real world,” but the end goal of using computers for pursuing knowledge is only one way of looking at the potential outcomes of digital tools. And they are still just tools. Alongside their capacity for computation is their role in enhancing coordination, or the ability of humans to communicate, transact, and investigate productively, together. Peer-to-peer digital tools such as blockchain technology offer an opportunity for an opening, a dispersion of power and information, and profound possibilities for collaboration on as-yet-unseen scales.
Peer-to-peer networks are participatory systems that resist control by a single or outside power. The participants establish agreed-on rules that evolve as need or complexity arises. “Feedback systems that enable both flourishing and regulation,” to borrow Ito’s language, is an apt description of many early ecosystem participants’ hopes for blockchains. By distributing power and value across global systems, the exchange of information and value can become more efficient, equitable, and open: more collaborative. Blockchain technology arguably could make the digital universe look more like a complex adaptive system of the kinds found in nature, if it becomes the best version of itself. This is a big if, and it will be many years in the future, if it comes to pass at all. This essay will explore the possibilities for blockchain to facilitate for the first time complex coordination between strangers on a global scale, a human-machine ecosystem that emerges to bring about abundance without excess, multiplicity without superfluity, complexity without chaos.
Of course, blockchains will not be a cure-all for society’s ills. No single tool or technology ever can be. But their adoption on a broad societal scale could potentially counter some of Ito’s concerns about the reductionist direction of Silicon Valley tech-evangelism and outsized faith in artificial intelligence. First, the financial monopolization by Silicon Valley of all the data and value contained in the Web 2.0 landscape can begin to redistribute toward individual users in Web 3.0.4 Second, blockchain-facilitated social decentralization can potentially redistribute and re-democratize patterns of human participation and cooperation. Third, blockchains are not controlled by a central authority, but by the entire network of participants, who establish the rules for participation themselves and can elect to evolve the system according to consensus; this makes them censorship-resistant and inherently more elastic than most other decision-making mechanisms for large groups of people.
More importantly, blockchain-supported technologies can potentially facilitate decentralized coordination and alignment of human incentives on a scale that only top-down, command-and-control structures previously could. Decentralization is the process of dispersing functions and power away from a central location or authority. In a decentralized architecture, it is difficult if not impossible to discern a particular center. The World Wide Web was originally developed as a decentralized platform. Blockchain technologies such as Bitcoin and Ethereum are examples of decentralized architectures and systems.
The challenge of coordinating groups of humans and getting them to behave in productive, peaceable ways has been the central story of civilization. A common claim about the potential for blockchain technology to facilitate social decentralization is that it could move power from centers—major metropolises, governments, large hierarchical organizations and companies—to the edges. Decentralization is also a social challenge: Everyone alive on Earth today has lived under the paradigm of hierarchy and top-down command and control, so we tend to default to them as organizational modes. The temptation to return to these familiar modes of coordination is great at times, and the transition to a less centralized social paradigm with natively digital tools will need to be a conscious one, made many times over, by the participants in the network. They are not working toward a singular vision or goal, an endpoint of optimization—a Singularity—as much as they are searching for productive pathways and ways to transact freely.
Conway’s Law states that “organizations which design systems... are constrained to produce designs which are copies of the communication structures of these organizations.”5 The structure and style of a group of programmers building a software tool will have a far greater influence on the final product than the assumptions they make about how their intended users will use it. By this logic, a network of people trying to build software tools to facilitate decentralization ought to be decentralized themselves: diverse, interacting randomly, coalescing around projects, conducting experiments, cultivating or abandoning them in a fluid state of co-relation. The “participant design” Ito proposes, “design of systems as and by participants,” is, in fact, the core ethos of a decentralized ecosystem.
An emergent ecosystem arises from self-organizing group behaviors. The cognitive challenge of accepting the fundamental impossibility of designing or mapping such a system, of allowing feedback loops their space to loop, and certain objectives sometimes to fail, is significant. Our brains are inclined to simplify or abstract hyper-complex systems for the sake of coherence. They want to detect signals in the noise, even if they are often false signals. But complex adaptive systems are inherently capable of self-regulation and constant evolution.
Unbreaking the Internet
News of data breaches and leaks—recently including the Equifax hack and Facebook/Cambridge Analytica fiasco—and increasing anxiety about the role of social media in digital culture portent a rising pessimism toward Web 2.0 technologies. While for half of the world the Internet has undoubtedly expanded awareness and information access and accelerated the pace of much social change, mistrust and a demand for regulation have begun to appear. But the centers of power in Web 2.0—Facebook, Google, Amazon, Apple, not to mention private Internet providers—still hold the cards, since they own almost all the data.
The Internet and World Wide Web were conceived as decentralized technologies. Web 1.0 was the first version of the dial-up web, essentially a virtual library consisting of static web pages. Web 2.0 is the web of applications. It is much faster, more participatory, and more interconnected; it is interactive, social, and mobile. It has become the Internet of data behemoths like Google, Facebook, and Amazon. Digital technologies on Web 2.0 have focused on the immediacy of commerce and communication. The focus of many vocations has shifted accordingly, and the pace of work has accelerated. Digital villages have emerged around common interests or motivations, irrespective of geographic location. News disseminates rapidly and gets editorialized nearly as rapidly. We can conjure a live video feed, get an answer to any question, pay someone, or otherwise make something happen nearly at the speed of thought.
This surplus of connectivity and two-way, participative information exchange has not brought with it broad expansion of economic opportunity, however. While Web 2.0 allows people to communicate and exchange information on a global scale, it does not allow us to coordinate at scale. Feedback loops burst forth and propagate so far as to stop intersecting with one another: infinite tunnels of echo chambers. Although the rapid exchange of information on the Internet has enabled the formation of various communities, and some open-source platforms like Linux, Github, and Wikipedia have flourished, the Internet is not democratic. Most data is siloed, and the people generating it by communicating, browsing, or shopping will never see nor learn from it, much less profit from it themselves. The dominant Web 2.0 platforms are centralized services that exert control over user-generated content; corporations store and monetize user-generated data and personal information. Information lives in silos, on private servers that can’t interoperate efficiently and are vulnerable to breaches.
Furthermore, the Web is decades old and still has not reached everyone on Earth. Far from it: Less than half of the world’s population has good Internet access. The “digital divide,” or the differential access and ability to use information and communications technologies between individuals, communities, and countries and the socioeconomic and political inequalities that result, is vast. What or whom has the Internet liberated? Common tech-sector buzzwords like disruption and innovation have all but lost their meaning when most of the allegedly disruptive and innovative platforms have upheld oppressive and unfair systems and created wealth disparities on par with the Gilded Age.
Technologies do not have inherent meaning; people imbue them with meaning and purpose by using them. No technology can incite broad societal change unless people gather around it, use it to collaborate, make decisions, and create new systems. As in the human body and the biosphere, selective specialization and adaptation continuously evolve into a more complex system, not into total chaos. Decentralization at scale, across institutions and societies, has never been possible the way it is now. We have a new way to get things done.
Building N-Sided Networks
If hierarchy has been the optimal way to regulate human behavior to this point, decentralization at scale offers an unprecedented alternative. Ant colonies, slime molds, flocks of birds, and schools of fish exhibit emergent and self-organizing behavior in which interactions between individuals give rise to spontaneous order. If power or information is concentrated in one area, order is less likely to emerge freely from interactions dispersed across a system. Self-organizing, distributed systems are nimbler and capable of self-repair at points of local failure.
Individuals in such a system are freer to pursue whatever they deem productive or promising. Blockchain technology can facilitate the alignment of incentives so that potentially profitable behaviors for individuals can simultaneously benefit the broader network. More and shorter feedback loops mean faster progress and also some degree of self-regulation to recognize malicious actors, unfavorable consequences, or undue centralization of power.
Enter Web 3.0. Whereas Web 2.0 is a two-sided client-server architecture, with a business hosting an application and users (customers and advertisers), Web 3.0 decentralizes that architecture on open platforms. Built on a back-end of peer-to-peer, decentralized network of nodes (computers), the applications run on decentralized storage systems rather than centralized servers. These nodes together comprise the so-called “World Computer,” capable of automatically executing smart contract programs on an open, peer-to-peer, Turing-complete, global system. The two-way interactions of Web 2.0 can evolve n-sided markets of shared value and exchange across a complex network—and a decentralized one.
Ethereum is an example of a distributed computing platform and operating system that can run any decentralized application; after Bitcoin, it is the most widely used example of blockchain technology.6 Ethereum founder Vitalik Buterin defines software decentralization along three axes.7 Architectural decentralization is how many computers a system is made up of, and how many it can tolerate breaking down at one time. Political decentralization is the number of individuals or organizations that control the computers that make up a system. Logical decentralization describes how the interface and data structure behaves: as a monolith, or a swarm whose constituent parts could begin to operate independently if it were cut in half, for example. As he puts it, “Blockchains are politically decentralized (no one controls them) and architecturally decentralized (no infrastructural central point of failure) but they are logically centralized (there is one commonly agreed state and the system behaves like a single computer).” Decentralized systems are more fault-tolerant, attack-resistant, and collusion-resistant than centralized systems.
The decentralized architecture of blockchain platforms is global, and therefore can tolerate localized shutdowns or attacks; it is censorship-resistant, because there is no one from whom permission is required to join the peer-to-peer network, as long as one operates according to the protocol; it is open-source, so its maintenance and integrity are shared across a network of engineers; and it is distributed, so there is no central server nor administrator from whom a large amount of value or information might be stolen.
Blockchains are effectively just databases, or distributed ledgers of recorded blocks of data, representing a consensus of participants. All of the information on-chain is visible to anyone, anywhere in the world, all the time. (But it is also tamper-proof: Even the most powerful nation-states cannot rewrite the data stored in the blockchain database.) With smart contracts, or programs that enable trusted transactions and carry out automated agreements, the Ethereum blockchain is fully programmable. Hashing algorithms conceal and encrypt data. Public key cryptography allows private data to move freely and visibly on public networks without revealing the encrypted information. No participant or group of actors can interfere with the data or smart contract, even if up to 50 percent of actors on the network are malicious, so all participants can trust its shared infrastructure as a shared source of truth.
Peer-to-peer networks can facilitate more meaningful, direct human interactions, maintained by its participants according to an open agreement and without the need for costly intermediaries. The word “trustless” is common in describing the nature of blockchain networks, but the word is misleading. Outside of their immediate social communities and contexts, many people have outsourced trust to institutions: credit scores, background checks, and state-issued identifications tell us about someone with whom we might do business, and we might trust contracts and the legal system to uphold our rights if someone violates an agreement we’ve made. Natively digital versions of identity and reputation can support the quick and easy exchange of funds or information that blockchain technology makes possible, and trust can be established between strangers much more quickly and seamlessly.
All of these features of blockchain-supported networks of collaborators are not just examples of participant design, but also avoid many now-familiar Silicon Valley pitfalls of oversimplification, reduction, siloed data ownership, and unchecked growth. The ideal outcome is a move away from data monopolies and monoculture, and toward a digital culture of inclusion, active participation, and more productive coordination.
Blockchain technology is still nascent. Satoshi Nakamoto released the Bitcoin white paper in 2008, and Vitalik Buterin released the Ethereum white paper in 2014. A lot has to happen before the full potential of blockchain-facilitated decentralization can be realized. The blockchain ecosystem today is far from diverse and has yet to achieve anything close to mass adoption. Much of the wealth contained in cryptocurrencies is held by a very small fraction of the networks’ actors: 97 percent of Bitcoin, for example, is held by 4 percent of investors.8 For the participants building the protocols and features to enable public blockchains to scale, it is essential to cultivate diversity in the ecosystem and to build solutions and applications with their potential users instead of for them. As many people as possible need to have access to the fundamental building blocks so they can construct local solutions according to their particular social contexts. The redistribution of money and power could just as likely give rise to a new but still unjust society, especially if the majority of influence in the ecosystem remains with a small few. But we have the opportunity to coordinate in entirely new ways, to open up new pathways and connections, and to build a potentially more diverse and resilient kind of system.
The potentially diverse applications of blockchain technology go far beyond financial and enterprise or business solutions, though it’s reasonable to expect that an initial wave of interest and investment from those sectors will generate a greater momentum of development. Borderless value exchange and self-sovereign digital identity potentially can empower people currently excluded from the global financial system. Some of the core features of blockchains, such as faster and lower-friction payments, provenance tracking, shared asset ownership, and direct peer-to-peer transfer of value can just as easily benefit musicians, authors, artists, and storytellers as they can banks and Fortune 500 companies. Cabals of cultural arbiters will have less power to dominate the funds that make creative work possible. Far beyond the scope of cryptocurrencies, a host of fascinating peer-to-peer projects experimenting with new creative models are underway, putting metadata, intellectual property, publishing, fact-checking, and much more into blockchain-intermediated digital systems.
While most of this essay has focused on ways in which peer-to-peer digital technologies can counteract the reductionist direction of machine dependence and enable greater human participation and autonomy in digital spaces, it is worth noting that there is substantial interest in the space of how blockchain and AI can interact. One obvious way is that decentralized, encrypted storage of the reams of data on which AIs compute can vastly improve security in those systems by removing them from the control of a single owner, particularly when it becomes possible for AI to interact with data without needing to decrypt it. Blockchain can also make previously isolated silos of data and user interactions interoperable, with reputation and credibility-building systems to aggregate that data across the Web. There are efforts to enable different AI systems to communicate securely with one another, comparing calculations and possibly reducing the risk of errors. Effectively, blockchain solutions can introduce checks on AI systems and the companies controlling them, even potentially making them more collaborative and interoperable.
The tone of cultural anxiety around digital technologies in general and artificial intelligence in particular suggests that the notions of linear progress and exponential growth require reexamination. The networks of interlocking complex systems that compose our societies and planet are not problems to be solved, nor experiments to complete. In his study of scientific revolutions, historian and physicist Thomas Kuhn suggested that a paradigm shift begins with hard science and makes its way through the social sciences and outward to culture.9 He also posits that we can’t observe a paradigm shift until it has already taken place. Ito’s opposite suggestion that the path to a new culture of flourishing will begin not with the hard sciences but with the arts reverses Kuhn’s theory, which after all does not leave much room for the idea of participant design. This revolution, when it takes place, will not be purely scientific. The observer is part of the system and the essential element of changing it. The mysterious clues of quantum (as opposed to classical) models suggest that everything has an opposite, that a spin somewhere will always result in a counter-spin somewhere else, in a great cosmic murmuration of dynamic signals.
The point of the “Singularity,” if reframed with these ideas in mind, would not be a civilization reaching the peak of comprehension, proving that the cosmos is ultimately reducible, but to unlock sufficient information to get to the other side of our current explanations, to come up with a richer description of its many interacting systems. The patterns of nature show us that there is always another layer unfolding recursively toward infinite and infinitesimal scales. Our civilization is not a pyramid, perhaps, but part of a fractal pattern.
Distributed power, opportunity for local action, emergent self-organization embracing complexity and diversity: once hard to achieve at scale, now possibly within reach. What is abundance without excess? Broader distribution of value and information across networks and geographies. What is multiplicity without superfluity? Greater efficiency and transparent information sharing across human societies, reducing friction and waste, from food production to resource consumption to intellectual capital. What is complexity without chaos? The opposite of reduction: richly productive and cooperative human efforts, at societal scale, supported by the new tools of decentralized technology.