Nobody is in charge of Ethereum. It suffers from a chronic lack of governance, a lack of structure. And as a result, it is in crisis.
That’s one theory, anyway.
“Ethereum governance has failed. We are a de facto technocracy, where a small group of technocrats, the core devs, have final say over what goes into the protocol,” declared former Ethereum core developer Lane Rettig recently.
“But the challenges we face today are increasingly non-technical. Core devs don’t want to make these decisions because they feel unqualified, fear legal liability, are conflict avoidant, and prefer just to write code.”
Not all agree.
Gavin Wood, co-founder of Ethereum told Cointelegraph Magazine: “The idea that ‘no one is in charge’ of either Bitcoin or Ethereum is fallacious. Vitalik [Buterin] is to all intents and purposes ‘in charge’ of Ethereum. The Ethereum Foundation (EF) controls its trademark and he controls the EF.”
The MIT Technology Review said much the same in late 2018: “Everyone knows that for all Ethereum’s ambitions to be decentralized, [Vitalik] Buterin is still its north star. When difficult times have arisen in the past, the community has leaned heavily on him to guide them.”
One recalls, too, the 2016 DAO hack, following which the key response was made by “a small group of people advocating successfully for the hard fork,” as law professor Michele Neitz recounted. She described the members of this group as Ethereum’s “agents of influence”.
From these recent utterances, one might assume decentralization is failing. Originally a political term suggesting the dispersion of powers, as from a central government to regional or local governments, decentralization has taken on a new meaning and importance in the Crypto Age. According to MIT Technology Review, it is “the principle, which any cryptocurrency community strives for, that no one entity or group should be in control.”
Who rules Bitcoin?
Maybe Bitcoin (BTC), the first decentralized blockchain project, offers clarity. “No one owns the bitcoin system,” said Bitcoin Core’s lead ‘maintainer’ Wladimir van der Laan in a 2016 blog in which he explained why he had to remove BTC luminary Gavin Andresen’s ‘commit access’ privileges in accordance with the “principle of least privilege.”
“No one controls Bitcoin,” stated Jameson Lopp in an influential blog titled Who Controls Bitcoin Core.
But Bitcoin has its skeptics. Its “governance [actually] consists in a form of domination based on charismatic authority, largely founded on presumed technical expertise, ” Primavera De Filippi and Benjamin Loveluck asserted in an oft-cited paper.
Moreover, it almost belies credibility that any organization can exist over time without its internal ‘influencers.’ Bitcoin is basically governed by Bitcoin Core, its software client used to access the Bitcoin network, and within Bitcoin Core there are only a handful of individuals with the ability to merge code into the master branch, so-called “maintainers.” At present there are five maintainers — holders of the PGP keys that can sign merge commits.
‘Lead maintainer’ appears to be an esteemed position. Satoshi Nakamoto, Bitcoin’s creator, was the first lead maintainer. Gavin Andresen was the second. Van der Laan is only the third. Bitcoin has had hundreds of core developers in its twelve-year history, but over that period it has had only about a dozen ‘maintainers’ including leads. Aren’t these ‘agents of influence’?
Not according to Lopp, co-founder and CTO of Casa, who told us that:
“While there are a handful of GitHub “maintainer” accounts at the organization level that have the ability to merge code into the master branch, this is more of a janitorial function than a position of power.”
“The question of who controls the ability to merge code changes into Bitcoin Core’s GitHub repository tends to come up on a recurring basis,” noted Lopp, who argued that the question itself is a red herring “that stems from an authoritarian perspective — this model does not apply to Bitcoin.”
After all, “Anyone is free to propose code changes to improve the software by opening a pull request against the master branch on bitcoin/bitcoin.”
But is it really that simple? One could argue whether Gavin Andresen deserved to lose his commit access privileges in 2016 or otherwise (“Gavin hadn’t done anything as a maintainer for a year or so, and before that he already was hardly active for a long time,” according to van der Laan), but assuming he deserved to be terminated, someone had to do it — and van der Laan wrote his blog, presumably, to justify why he had committed the deed.
Human politics have not been eliminated
“The [Bitcoin] development team is not autocratic,” commented Vili Lehdonvirta, Associate Professor and Senior Research Fellow at the University of Oxford. Moreover, other parties are also influential in how Bitcoin’s rules are set, including mining pools. “The point is that Bitcoin has not in any sense eliminated human politics; humans are still very much in charge of setting the rules that the network enforces.”
In a recent interview, Cointelegraph Magazine asked Lopp about the criticism that Bitcoin’s actually “consists in a form of domination based on charismatic authority.” He answered:
“During scaling debates we often saw these different types of arguments clashing with each other. In general I’d disagree with the characterization of debates as “domination” unless it’s being used to describe someone dominating a particular argument because the opposing side has done a poor job of defending their position with logic. At the end of the day, no one can force node operators to change the software they are running.”
But surely maintainers serve more than a “janitorial function”? Else, why would Gavin Andresen be upset about losing his “commit access” privilege?
“Gavin Andresen appeared to not consider his role to be a janitorial function,” answered Lopp. “There were points in time in which he acted more like a benevolent dictator. Unfortunately for him, it turned out that the Bitcoin Core organization did not want a benevolent dictator. I’m actually not so sure that Gavin was upset about losing commit access; he had not used it in over a year at the time it was revoked. The drama about it being revoked seemed to mostly come from other people outside of Bitcoin Core.”
With Bitcoin, Ethereum, and other decentralized blockchain projects, there seems to be this paradigm that “no one is in control,” that all decisions are made by consensus — and there seems to be a real reluctance that all these projects might have their internal ‘influencer.’ “For a very long time these politics were not explicitly recognized,” said Lehdonvirta, and many people don’t recognize them, preferring instead the idea that Bitcoin is purely ‘math-based money’ and that all the developers are doing is purely apolitical plumbing work.”
Is it fair to say, then, that Bitcoin may be less decentralized than most people — inside and without — believe?
Daniel Resas, associated partner at German law firm Schnittker Möllmann Partners, where he heads the firm’s digital assets & blockchain practice group, suggested that: “The answer to that question heavily depends on your definition of ‘decentralization’. But it’s probably fair to say that the majority of users do not critically question (obvious) signs of centralized elements relating to the operation of the Bitcoin network, namely in the context of decision-making and transaction validation.”
He continued: “As a matter of fact, the protocol’s simplicity by design, which is also reflected in the absence of a formalized protocol governance regime, leads to centralized elements which are likely to be the only efficient way to come to reliable decisions in the context of major protocol updates which have a significant impact on the stakeholders crucial to the network’s stability. In other words, sometimes, especially in challenging situations such as the ‘inflation bug’ case in 2018, a smaller group is likely to be the more efficient ‘governing body’, even if this obviously constitutes a massive element of centralized network governance.”
Dominance by miners?
And then there is the matter of miners. It’s widely recognized that Bitcoin’s mining function is heavily concentrated. Together, just four entities — i.e., mining pools — control more than 50 percent of Bitcoin’s mining/voting power, according to Bitcoinera. These miners are responsible for validating all transactions given Bitcoin’s proof of work consensus protocol. It’s similar for Ethereum, the second largest cryptocurrency by market share — just three entities account for more than 50 percent of mining/voting power. (This could change when Ethereum moves to a proof of stake consensus mechanism.) Aren’t these mining pools agents of influence?
“The economic incentive schemes [i.e., proof of work] designed to run the network in a reliable manner are prone to the economics of scale, leading to a centralization of mining power,” said Resas. “Yet it’s probably too early to say whether such concentration of hash power in the hands of a few mining pools is an actual concern for the network’s resilience.”
Charles Hoskinson, CEO of IOHK and founder of Cardano, told us that: “Due to Bitcoin’s use of proof of work, it relies on mining pools to keep the ledger updated and maintained. Unfortunately, as the user base has grown, so has the required computing power to become a miner. The resultant ‘arms race’ has resulted in a handful of mega-sized mining operations — with the largest eight mining almost 90% of all Bitcoin.
“In a truly decentralized project,” continued Hoskinson, “there should be nobody who is ‘more equal than others’ — including in terms of the ability to participate in consensus or to mine blocks. Proof of stake protocols are uniquely placed here, as they avoid the ‘hardware arms race’ of proof of work systems.”
A benign dictatorship?
What about Ethereum? Co-founder Gavin Wood, as noted above, has no doubt about who is in charge: Vitalik Buterin. He expanded on this:
“If he [Buterin] decides something regarding the mainnet protocol, then it’s unlikely that it won’t happen. If a protocol change is needed that he doesn’t appear to agree with, then it’s unlikely it will gain sufficient momentum to be successful. In reality, Ethereum is an opaque and largely benign dictatorship adulterated by a good dose of (mostly self-inflicted) chaos.”
But is this necessarily a bad thing?
“I see the role that prominent ‘network ambassadors’ like Vitalik Buterin or Joe Lubin or influential core dev groups play not as a problem per se from a governance perspective,” answered Resas. “Early-stage networks do rely on visionary thought-leadership. In order to act in an efficient manner, even the most extensive degree of decentralized decision-making relies on core stakeholders sparking and sounding ideas beneficial to the network and its users.”
Influence only becomes a serious problem if and when conflicts of interest arise: Where the decisions or power of ‘influencers’ threaten the interest of the majority of stakeholders. Said Resas: “Transparency can be a very effective means of addressing such possible conflicts of interest. However, there may indeed be situations where, for various reasons, such transparency can only be achieved afterwards — for example, as the Bitcoin core developers did after fixing the ‘inflation bug.’”
Avoiding the governance issue
If one concedes that blockchain doesn’t have a dictator problem — or that at least that its dictators are benign — what about the opposite, it suffers from lack of governance, a vacuum, with no decision-making authority — or, in Rettig’s words, “the tyranny of structurelessness.”
Said Resas: “Absolutely. I couldn’t agree more. To be more precise, there is no vacuum in the first place. The absence of formalized governance schemes in Bitcoin and Ethereum just paved the way for informal decision-making processes. Such informal structures might have been criticized for their lack of transparency in the past, but they still represent a form of governance, just an informal and intransparent one.”
University of Oxford’s Lehdonvirta isn’t quite so sanguine. What Resas calls informal governance, Lehdonvirta views more like a cop out. When we asked him why, he explained that “The project that carries the Bitcoin name today has a thing about not making any changes to the protocol [….] The idea that governance can be avoided by just having a chain split every time someone disagrees with a choice doesn’t seem very viable in the real world.” By way of example, he added: “If tokens are meant to represent something outside their own little bubble, such as claims to land titles, then having multiple parallel records that conflict with each other is obviously a non-starter.”
Resas, by contrast, gives both Bitcoin and Ethereum passing grades for how they managed things in the past. “Both networks have proven that they are able to master very challenging situations from a governance perspective. It is an interesting thought experiment to ask how systems relying on explicit governance schemes, maybe even ones completely running on-chain, would have dealt with similar challenges.”
What’s the point of a blockchain, anyway?
There’s another thought to consider: something described as the blockchain paradox. As posited by Lehdonvirta, it holds that once you learn how to govern a blockchain, you don’t need a blockchain anymore. That is, when a network has finally developed a trusted governing inner core — benign ‘agents of influence,’ let us call them — then it might as well deploy a distributed database. Blockchains are only necessary when you have a trust problem, i.e., when trust is in short supply.
Has the paradox been resolved in the four years or so since Lehdonvirta first began discussing it? “I’m not sure if anyone has learned how to govern a blockchain project successfully, but many if not all of the more commercially successful projects (used for applications outside the crypto trading bubble) were arguably distributed databases to begin with. The central bank digital currency projects that are underway are tightly governed and don’t appear to be using any real blockchain technology, even if that was part of their original inspiration,” he told us.
If Bitcoin and Ethereum are still wanting governance-wise, what about more recent blockchain projects like, say, Polkadot? “The prevailing approach to governance among those who take it seriously seems to be to try to engineer the hell out of it. I have skimmed many white papers that describe extremely convoluted governance arrangements,” said Lehdonvirta. “The more complexity they add, the more things can and will go wrong.”
Wood told us: “Going forward with Polkadot, it seemed clear to me that to avoid this dichotomy, we needed a way of allowing the protocol to transparently benefit from good leadership without being duly compromised by the according errors. The governance structures of Polkadot are an initial attempt to do this; the protocol’s direction is not set by a single ‘spiritual leader,’ but by its assembled stakeholders, weighted toward those who have locked themselves in for the long run.”
Governance problems arise, in Lehdonvirta’s view, because these projects are often run by capable engineers rather than competent political scientists. They have trouble “acknowledging that things like truth, power, wealth, and identity can’t ultimately be disentangled from a messy human context that gives them their content. And that many of the goals they seek are probably best served via community and institution building rather than engineering.”
Heeding the stakeholders
Lopp, for one, isn’t ready to give up on engineered solutions. “The idea that we can remove politics from various aspects of our lives is incredibly appealing to me — by removing the impact of politics we can create more reliable platforms because we don’t have to worry about the rules being changed on a whim without our consent,” he said.
“Indeed, I believe Bitcoin is pioneering a new form of governance that inverts traditional hierarchical bureaucracies.”
“The network needs some systematic way of listening to them [i.e., stakeholders],” Wood told us. “Networks, where it is the developers who decide when and to how much they listen to the inclinations of the stakeholders, are doomed to eventual failure as the leadership will eventually make a misjudgment.”
Centralization is an inherently political force: Can we really wring politics out of our lives, as Lopp proposes, or is Bitcoin just an organization like any other, with influencers and power brokers who are essentially its political class?
Before Bitcoin’s second decade concludes, perhaps we’ll have some answers.