Centralization Will Not Last: How Decentralization is Reshaping Technology and Governance

Original Title: “The Center Will Not Hold: How Decentralization is Reshaping Technology and Governance”

Authors: Bruno Lulinski & David Kerr, The Defiant

Compiled by: Runsheng & Gu Yu, Chain Catcher

Decentralization is a concept historically used to describe local governance structures, referring to the distribution of planning and decision-making responsibilities among its members rather than being held by a central authority.

Although the terms "centralization" and "decentralization" were coined during significant political upheavals in Europe in the 19th century, these concepts have existed since the dawn of society. Despite the clear benefits of decentralization in terms of inclusivity, representation, and individual freedom, human history has primarily been a study of centralized authority, as it offers advantages in decision-making efficiency and effectively governs large geographical areas.

However, recent technological advancements have allowed for a more effective utilization of decentralization principles, as many historical limitations no longer apply, and powerful governance mechanisms of decentralization can now be leveraged.

Mechanisms and Choices

In particular, blockchain has become an opportunity to develop new systems that can more effectively embrace the benefits of decentralization, serving as both a mechanism for member-representative governance models and a choice for users to access and develop application functionalities in a decentralized economy.

While the historical definition of the term still serves as a comparative measure against centralization, the term "decentralization" itself has become synonymous with certain characteristics of blockchain technology and serves as an analytical point for determining its technological, legal, economic, and political functions. Although there is still no unified definition of decentralization, its usage in blockchain typically reflects certain characteristics traceable back to the original Bitcoin white paper.

While the term decentralization is not explicitly stated in the white paper, the ability to construct organizations and processes "without a central authority" in the absence of "trusted third parties" for transactions or other interactions has established an independent concept used within the industry. The application development of decentralized technology, regulatory actions, legal analyses, economic potential, and political theory provides additional context for the significance of decentralization for public blockchains and applications built on these blockchains.

The purpose of this article is to help clarify why people in the broader crypto community organize around the concept of decentralization. While a unified comprehensive definition would undoubtedly be clearer than the current contextual definitions, the reality is that decentralization within blockchain is an evolving concept that must currently be revisited based on circumstances to determine how foundational processes adapt to its current applications.

In this summary paper, we will explain the different components of decentralization, the benefits of decentralized systems, provide examples of how different projects handle the decentralization process, and offer good-faith critiques of decentralization. This article will serve as a TL;DR, summarizing the concept of decentralization within the current usage scope while providing relevant resources for those interested in delving deeper into specific areas.

"Blockchain is politically decentralized (no one controls them), architecturally decentralized (no infrastructure central failure points), but logically centralized (there is a commonly agreed state, and the system behaves like a single computer)."------“The Meaning of Decentralization,” by Vitalik Buterin (co-founder of Ethereum)

This article does not attempt to provide an all-encompassing definition of the term "decentralization," nor does it aim to reach an objective measure of project decentralization. Ethereum co-founder Vitalik Buterin wrote an article titled “The Meaning of Decentralization,” emphasizing the difficulty of accurately defining the term.

Attempts to measure decentralization include: Miles Jennings' paper on the principles and models of decentralization, Balaji Srinivasan's post on the Nakamoto coefficient; a paper from Beijing Jiaotong University using various metrics to measure decentralization; and Ketsal's post on open standards for measuring blockchain decentralization. Given the ever-changing nature of the space, as the subject continues to evolve, we will publish updated versions of this document.
"Aiming for decentralization as the ultimate goal often means targeting a vague and potentially shifting target."------“Measuring Centralization in Blockchain Networks,” by Josh Garcia and Jenny Lueng

Decentralization Standards for Layer 1 Blockchains

The core value proposition of many blockchains, including Ethereum, is to serve as a trustless infrastructure where developers can build immutable decentralized applications. While other blockchains are striving for incremental decentralization, Ethereum's first-mover advantage and its status as the first widely adopted smart contract platform—i.e., a blockchain that natively supports smart contracts, allowing for the construction of various composable applications on top of it—make it a natural benchmark for decentralization in Layer 1 blockchains.

Coinbase engineer Yuga Cohler even stated that Ethereum's upcoming transition to a proof-of-stake consensus mechanism, if successful, will "prove the viability of decentralization as a principle of social organization."

Applications built on Layer 1 blockchains inherit some decentralization properties of the underlying layer, but merely being built on a decentralized layer does not necessarily mean that the applications themselves are decentralized.

While applications inherit the immutability and censorship resistance of the underlying layer by default, each application will make design trade-offs that affect the degree of decentralization of the individual application. In other words, the decentralized infrastructure layer provides a foundation for both decentralized and centralized applications to work together, with each application and its corresponding community weighing trade-offs to achieve their desired state of decentralization.

Ethereum's degree of decentralization has not been without criticism. Potential centralization challenges for Ethereum often revolve around the centralization of liquid staking derivatives and the majority client risk, both of which focus on potential unexpected centralization failure points that may arise in the network's future.

"The simpler, the better"

The details of these two issues are beyond the scope of this article but have been discussed in detail elsewhere. For more information on liquid staking centralization, see Ethereum researcher Danny Ryan's post on the risks of liquid staking derivatives and the research from decentralized staking provider Lido. For more information on majority client risk, see Ethereum researcher Dankrad Feist's article and the section on client diversity on Ethereum.org.

The Ethereum blockchain can be considered a "sufficiently decentralized" infrastructure, serving as a basis for discussing decentralization. The Bitcoin blockchain certainly qualifies as "sufficiently decentralized" as well, but by design, Bitcoin has fewer embedded functionalities in its protocol compared to Ethereum's more flexible smart contract platform.

The Bitcoin community largely embodies the spirit of "the simpler, the better," as Bitcoin itself is a manifestation of decentralization, with less vulnerability compared to more complex protocols. This article will not delve into the details of this argument, other than to acknowledge that Bitcoin qualifies as "sufficiently decentralized" and that Bitcoin has been building the Lightning Network to support the construction of more decentralized applications on top of the network.

Ethereum has a more robust application ecosystem from which to discuss different levels of application decentralization, but Bitcoin's achievement as the first decentralized blockchain has paved the way for the entire industry.

For more foundational background on Ethereum and the overall blockchain, co-author Bruno Lulinski has written “A Simple Guide to Ethereum,” which introduces blockchain, DeFi, NFTs, the decision-making process of the Ethereum community, and Ethereum's future scaling solutions.

Different Components of Decentralization

Decentralization is crucial to the value proposition of several parts of the broader crypto ecosystem, making it worthwhile to examine the meaning of the term relative to its application areas. The elements of decentralization are both discrete (i.e., is the project decentralized in that specific area?) and interrelated (i.e., how does the decentralization of one component affect the decentralization achieved in another area of the project?).

Because decentralization in one component's context implies something different from decentralization in the context of different components (while sharing foundational activities that influence other categories), projects need to consider how each component can operate as intended.

The components of decentralization can be categorized into three main categories, involving the effectiveness of decentralized systems along three axes: technological, economic, and legal decentralization.

As Miles Jennings states in his in-depth discussion of decentralization principles and models, "The effectiveness of these decentralized web3 systems will depend on their security, economics, and equality of information"—each system corresponds to one of the three components listed.

• Technological Decentralization—The global permissionless infrastructure layer and the applications built upon it require a trusted decentralized technological foundation. The underlying blockchain provides an execution layer for the on-chain components of various applications, but the applications themselves still need to achieve their own technological decentralization in the form of permissionless clients to interact with the foundational smart contracts, user-owned data (and data portability), and decentralized governance of smart contracts by a broad range of participants in the form of decentralized autonomous organizations (DAOs).

Questions to consider when thinking about technological decentralization: How are these systems designed? How are upgrades conducted if needed? What processes support the execution of upgrades (i.e., Compound's 48-hour time lock)? What blockchain underpins the application, and what trade-offs does that blockchain impose on the application? Can users easily "exit" the system, i.e., can users leave the system and use (or build) different methods to interact with the core protocol?

From the perspective of determining the decentralization of the blockchain itself—how many clients are there, and what is the distribution among clients used for miners/validators? How do individual participants verify the authenticity of a given blockchain, and how difficult is it for individuals to participate in that verification process? There are many ways to consider the technological decentralization of a blockchain.

Ultimately, technological decentralization is a necessary foundation for achieving economic and legal decentralization.

• Economic Decentralization—Public blockchains create opportunities to reimagine the economic interactions between application developers and the users and stakeholders of that application. In the "traditional" pre-blockchain world, companies were incentivized to view their users as sources of value extraction, primarily in the form of user-generated content or corresponding data from end-users, engaging in behind-the-scenes transactions between the company and willing advertisers.

Blockchains allow for systems that do not rely on central leadership, enabling a balance of incentives among application developers, application contributors, and application users. These economically decentralized structures are essentially the next generation of open-source software communities but with embedded transparent economics. In an economically decentralized ecosystem, contributors can participate in the value creation of the application while being rewarded for their contributions.

Questions to consider when thinking about economic decentralization: How are the underlying tokens of the application designed and distributed? How are airdrops structured, and what considerations did early project developers make to prevent the concentration of ownership of most project tokens? How are early investors and project contributors compensated, and what do the token lock mechanisms look like for each party? How is DAO funding allocated, i.e., how are funds distributed to initiatives and/or working groups aimed at advancing the project's mission?

• Legal Decentralization—In addition to technological mechanisms and the economic incentives of decentralization, there are regulatory and legitimacy issues, including taxation, liability, ownership, intellectual property, reporting, and privacy. While U.S. securities law is an important area of analysis for determining how decentralized systems can utilize digital assets, it is not the only legal area affected by the decentralization provided by public blockchains. Although decentralization exists within the current legal framework—most notably in the form of general partnerships—an important question is how the default rules established for participation and liability can be fairly applied to decentralized systems that are vastly different from the activities that generate existing laws.

Beyond the superficial similarities with existing rules and laws, the decentralized activities enabled by blockchain represent a significant shift in concepts such as fairness, ownership, and control. These differences highlight the distinct relationships and responsibilities compared to the activities that generate current laws and regulations, creating significant uncertainty in how to handle decentralized organizations on the blockchain.

Since these activities can generate taxable events and act in ways that trigger lawsuits, taxation and liability will quickly become equally prominent issues alongside securities law when considering legal decentralization.

Early projects require some form of central leadership and planning to determine project objectives and provide critical activation energy. These teams may retain some influence in projects that are about to decentralize, but the level of retained influence may significantly affect whether the project is viewed as decentralized from the perspective of regulators and other government entities.

Degree of Influence

Securities regulation primarily stems from the intent to prevent information asymmetry among market participants. While there is no clear standard for the concept of legal decentralization, the level of influence retained by early project contributors during the decentralization process, as well as the transparency of information among participants, will be key in determining whether a project is legally decentralized.

Many have written high-quality overviews of the topic of decentralization from the perspective of securities regulation:

• Scott Kupor wrote an article on key regulatory issues regarding cryptocurrencies and described the considerations for determining whether a token is a security;
• David Kerr and Miles Jennings drafted a legal framework for decentralized organizations;
• Stephen Wink and Shaun Musuka described the pursuit of full decentralization as a clear path to avoiding securities regulation.

Questions to consider when thinking about legal decentralization—how much influence do early project teams have, and where does that influence come from? Is their influence derived from the significant voting power retained in so-called decentralized organizations, or from their voice in the community decision-making process? How much influence do early investors have? Can community members be held accountable by other community members? Does the project rely on the efforts of a central entity? Do different stakeholders possess asymmetrical information based on organizational structure design? For more information on legal decentralization, see the section below on securities law.

Benefits of Decentralized Systems

As mentioned above, the term "decentralization" is itself a reflection of the term "centralization." Examining the attributes that decentralized systems may possess, such as censorship resistance and distributed decision-making, can make this concept easier to visualize.

Censorship Resistance

Censorship resistance refers to the inability of any single governing body to unilaterally make decisions that restrict the network behavior of other participants. Historically, coordination among humans has relied on a certain degree of trust.

A goods transaction between two people requires trust that both parties will indeed deliver the goods to each other, while reaching some ceasefire or treaty between conflicting nations necessitates trust that the other party will continue to adhere to the agreed-upon treaty.

Immutable code deployed on decentralized public blockchains lays the groundwork for innovations that resist censorship and protect privacy. These censorship-resistant systems are not entirely immune to enforcement, but they are a necessary complement to the institutions and platforms (world governments, social media platforms, etc.) we rely on outside of crypto. If the infrastructure layer (the blockchain itself) is not decentralized, a powerful government can easily shut it down—simply by finding the party responsible for the network and coercing them.

As evidenced by China's Bitcoin ban, decentralization makes this difficult, as sufficiently decentralized systems require coordination beyond the reach of most governments; just months after China's Bitcoin mining ban was issued, several underground mining operations emerged in China to fill the void left by the ban.

Individuals also use NFTs to preserve information in the face of authoritarian governments—however, these NFTs still require individuals to remain anonymous to avoid direct coercion from the government. Even democratic governments like Canada have recently expressed a willingness to exercise censorship by compelling financial institutions to conduct financial scrutiny on certain citizens.

Power and Influence

When cross-border remittance companies initially restricted remittances to Ukraine (later easing these transfer limits to Ukraine), countries like Ukraine effectively leveraged the censorship-resistant qualities of public blockchains to fund their defense efforts.

Tech giants like Apple, Facebook, and Google have expanded into immense power and influence, allowing them (whether they want to or not) to engage in public debates about interactions on their platforms (as well as the processes guiding their often-controversial decisions). Governments frequently use Twitter to communicate directly with their constituents, providing a great example of the benefits of decentralization—In 2018, Twitter revoked access to various APIs used by independent developers to build applications on Twitter.

A decentralized system would have censorship resistance against decisions made by platforms like Twitter. In fact, the transparent and impeccable rules of participation are a form of censorship resistance that blockchain-based applications naturally inherit, as the code deployed to public blockchains is open-source by default.

Even though former Twitter CEO Jack Dorsey has committed to adopting open protocols and long-term neutrality (as he later lamented), the commitment to censorship resistance on Twitter will always fall short over a sufficiently long timeframe—this is a natural conclusion of game theory. The concepts of open-source code and allowing users to own their private data fundamentally oppose the business models of companies built on closed protocols, which rely on end-user data to generate financial returns for their shareholders.

Resilience Against Attacks and Decoupling

Vitalik Buterin believes that decentralized systems are more resilient to attacks and less likely to experience accidental failures than centralized systems.

Crucially, the cost of attacking decentralized systems is often higher because there are no sensitive central failure points for attackers to target—attackers cannot simply infiltrate the Ethereum Foundation and press a big red "HALT" button (because there is no big red button), and attackers cannot coerce Buterin into shutting down the blockchain (because Buterin does not have that centralized power, despite being the original founder and rule-maker).

At the application layer, trust assumptions, key management, and security practices may differ, meaning that the level of resilience against attacks varies. The resilience of Layer 1 blockchains against attacks will still provide application developers with a trusted, permissionless infrastructure.

Decentralized networks also tend to lead to redundant systems, resulting in stronger security. Tim Beiko, one of the chief coordinators of the Ethereum developer community, recently referred to this benefit of redundant systems as "uncorrelated failure modes" on Farcaster. The overall idea is that multiple solutions (i.e., different client implementations, different approaches to specific problems, or simply different schools of thought) can reduce the likelihood of catastrophic failure across the entire stack.

(Due to the difficulty of quantifying decentralization, Beiko technically uses "decoupling" and "uncorrelated failure modes" as substitutes for the term "decentralization." We use "uncorrelated failure modes" as a benchmark for sufficient decentralization. Thus, from our perspective, decentralized systems must inherently possess uncorrelated failure modes, but in any case, the sentiment remains the same.)

An example of correlated risks leading to catastrophic failure is the 2008 global financial crisis, where risks stemming from mortgage-backed securities, credit default swaps, and bubble lending practices were improperly underwritten by rating agencies. As homeowners defaulted, this highly intricate web of risk led to correlated failures, resulting in lender defaults, which led to counterparty defaults, causing severe damage.

Underwriting systemic correlated risks is challenging, and in complex interrelated systems, it can lead to disasters. Decoupling systemic risks through the open boundaries of blockchain can help mitigate these risks and reduce the attack surface.

"If you had asked an average person in 2007: 'If it turns out that investors misjudged the super-senior risks of synthetic collateralized debt obligations built from subprime mortgages, how would that affect your life?' that person would have said 'I don't know what you're talking about, but I can't imagine how that group of words would affect me.' But it did."------Money Stuff, May 12, 2022, Matt Levine.

Transparent Incentives and Distributed Decision-Making

While shareholders of publicly traded companies may successfully request that the board include shareholder proposals in the annual proxy, the board has a certain tolerance for proposals that must be included in discussions, and many large tech companies (Facebook, Snap, and Google) have dual-class share structures that grant insiders absolute majority voting power, denying stakeholders any meaningful outcomes.

While decentralized systems are not merely tools for distributed decision-making, they do achieve governance transparency, potentially enhancing the fairness and effectiveness of critical decision-making processes. Good-faith critiques of distributed decision-making systems include the inevitable concentration of attention required by early project teams, the tragedy of the commons, and voter apathy, all of which can manifest in these types of horizontally distributed decision-making processes.

The governance of these distributed decision-making systems is itself a complex topic (which will be discussed in future collective articles from DAO research). However, supporters of decentralized systems argue that the transparent record of actions is one of the main benefits of decentralized organizations.

Critiques of Decentralization

While occasionally misunderstood, the decentralized economy built on public blockchains does not replace all centralized entities. Instead, it is a structural extension that allows decentralized and centralized organizations to interact in ways that were previously technologically infeasible or impractical. Nevertheless, within the context of public blockchains, there are still some good-faith critiques of decentralization, which will be discussed below.

Only Completely Horizontal Structures Can Be Considered Decentralized

For some, the idea of decentralization means a complete absence of hierarchy, a total lack of structure, and no guidance or leadership. Some critics of decentralization argue that any influence exerted by a few participants in a decentralized system proves that the system is centralized, or that any structural attempts can maintain centralization.

This argument has been used to claim that Ethereum is not decentralized. But as described in this article, decentralization takes many forms and must be distinguished through specific reference frameworks to differentiate between different types of decentralized systems.

Kevin Owocki of Gitcoin discussed this issue in a Twitter post, pointing out that decentralization can refer to "decentralized governance through permissionless tokens," rather than "a chaotic loose network of individuals."

In the book "Against Capture," Spengrah writes: "Confusing decentralization with permissionlessness is one of the most common mistakes in the DAO space." Spengrah discusses the concept of anti-capture, which is a framework for designing human networks to resist governance capture by bad actors. For projects that cannot be simplified to purely non-human programmatic functions, anti-capture governance is a more reasonable decision-making goal.

Consumers Lack Sufficient Recourse

One concern regarding decentralization is that decentralized entities do not provide recourse for consumers. The critique is that many (or most?) consumers will not care about self-custody of assets, and many would be happy to claim that "decentralization" is a virtue of web3… until their bored apes are stolen. How will assets be recovered if users lose them due to malicious actors or fat-finger errors? In a decentralized world, how does this process work?

In March 1933, as the United States was recovering from the Great Depression of the late 1920s, President Franklin Roosevelt addressed the banking crisis and the emergence of the Federal Deposit Insurance Corporation, which provided insurance for bank depositors up to $250,000. His first sentence illustrated the situation at the time:

"I want to talk to the American people about banking—relatively few understand the banking mechanism, but more specifically, I want to talk to the vast majority who use banks to deposit and withdraw checks."

User Demand

At that time (and still today), many people do not understand the inner workings of the banking system. In the near future, many will still not understand or care about the mechanisms of the underlying blockchain, but there will be solutions to meet consumer needs (in this case, wanting to own bored apes or participate in DAOs) without consumers needing to understand the specific mechanisms of the system. Through an iterative process, society learns and improves. The same goes for cryptocurrencies.

Insurance will become more robust, and either users or the protocol itself (or both) will provide programmatic insurance for these types of damages. The user experience of wallets will improve (for excellent wallet user experiences, see Argent or Rainbow), reducing the burden of self-custody while still retaining the benefits of self-custody.

In a world of social recovery wallets, concerns about losing a 24-word secret recovery phrase will become irrelevant, and solutions will continually emerge to meet those who wish to enjoy the benefits of decentralized systems while still ensuring reasonable forms of potential assistance in difficult situations. Most importantly, centralized organizations will coordinate with decentralized organizations to provide these solutions, preserving the advantages of decentralized systems while also creating user experiences that meet the needs of their individual user bases.

Centralized Entities Prove Centralization

Centralized entities existing in the crypto world—such as the centralized exchange Celsius—are often used to argue that the crypto world lacks true decentralization. This critique rightly targets those projects that claim decentralization as a selling point to attract users to an obviously centralized (by any definition) project, including many recent disasters in the crypto space (such as Luna and Celsius). These should be criticized.

However, as detailed in this article, decentralized systems are not merely systems that are completely horizontally distributed; rather, several individual components must be considered when judging the level of decentralization of a project. Crucially, the critique that "centralized entities within the system prove the system's centralization" often overlooks the concept of data portability. Polygon's Chief Information Security Officer Mudit Gupta refers to data portability as "the ability to decentralize."

Centralized systems can exist and create value for end-users by making it easier to interact with permissionless blockchains, but ultimately, blockchains enable users to exit using their own data. If the centralized NFT marketplace OpenSea decides to censor a subset of NFTs sold on its platform (by not displaying them on the OpenSea user interface), or if OpenSea decides to start charging users higher fees, users can simply stop using OpenSea and move to another NFT marketplace.

OpenSea does not actually hold users' NFTs—OpenSea is merely a venue for showcasing and trading (more importantly, a venue with significant liquidity that makes the market more efficient and price discovery better, but still just a venue).

(By the way, OpenSea recently launched Seaport, a decentralized protocol supporting its marketplace.)

Traditional internet companies do not provide users with the flexibility of data portability because they have no incentive to do so, but blockchain-based applications must embed data portability into their operations. While centralized crypto companies can enhance efficiency (e.g., organizing liquidity, providing customer support, and standardizing user interfaces), the ability for users to exit the system can check any centralized entity's power over the system and ultimately over its users.

Ineffective Governance and Potential Oligarchic Rule

Criticism that decentralization will lead to ineffective governance may be the most accurate critique of the ecosystem at present. As of early 2022, governance in decentralized organizations has largely been ineffective—participation is low, and as most decentralized organizations tend to do, pure token voting has various embedded issues that may produce oligarchic regimes even more than before.

The governance paper from the DAO Research Collective will address many of these issues, and Fred Ehrsam of Coinbase/Paradigm wrote a prescient article in 2017 about blockchain-based governance systems, highlighting some of the benefits and challenges of on-chain governance, as well as future approaches. Ultimately, whether decentralized governance can be as effective (or more effective) than traditional centralized governance systems remains to be seen.

"Only by establishing technical systems that provide various mechanisms to check the concentration of power, while building social ideologies that continuously seek failure modes for these mechanisms, can we hope to succeed in the face of previous failures of decentralization attempts." “Liberation Through Radical Power Decentralization,” Vitalik Buterin and Glen Weyl

(Bruno Lulinski is the author of “A Simple Guide to Ethereum,” aimed at understanding the different components of the Ethereum ecosystem. David Kerr is the research director of the DAO Research Collective and the chief advisor of Cowrie LLC.)