👮 Censorship Resistance¶

It's critical to speak truth to power; history is proof of this. However, history also shows how the existence of any kind of power structure influences what truths can be thought in the first place. How can we free ourselves from this constraint? One suggestion we will continue to make throughout the rest of Kernel is: programmable incentives.

When the incentives which define the structure of power in society can be programmed by anyone, anywhere; censorship resistance becomes an engineering problem, not an ideological one. This is a major claim, and we hope you’ll take it into dialogue and test it thoroughly with others as you go through this chapter.

That said, the fact that censorship resistance is an engineering problem in the context of blockchains does not absolve us from the need to take an ethical stance on why it matters.

One of the core functions of healthy cognition is what John Vervaeke calls “opponent processing”: we need to cultivate ecologies of practices in order to both refine our ability to meditate upon or contemplate within the world, as well as engage with it via effective relationship and action. The concept of “opponent processing” is not only applicable to individual cognition: but the distributed cognition we call “governance” and “culture”. John points out that democracy is based on this idea, opponents debating perspectives based on preferences which are representative of groups of people. However, modern politics interprets these debates as adversarial–someone has to win and winning means convincing the other of the rightness of my view–whereas opponent processing is not really meant to be such a finite game.

Opponent processing does not end in win or loss conclusions: the idea is that meeting opponents honestly, listening openly, and engaging truly is an infinite game the ongoing outcome of which is “transframing”: transformation, transcendence, and reframing. Dialogue with those who differ from us has the potential to fundamentally change our perception of our self and the world.

If we are not allowed to engage honestly with one another, in immediate and direct contact, then we cannot find our way to the collective wisdom which is the necessary result of transframing when it occurs between large groups of people.

What is fascinating here is that blockchains fuse money and speech, meaning that expression and (economic) action occur simultaneously. This opens up the field for transframing further. The speech act is one of the most potent ways we have to transfer meaning. The fact that such acts now literally carry economic value means that others can more clearly see the specific value we attach to our signs. This does not mean it is a silver bullet–such a convergence could be used in very reductive ways too–but, if we cultivate a listening society, we can shift our freedom to speak away from the formal and towards the substantive.

Economic proof¶

Recall that trust is only possible once you have encoded what it means to cheat. Having stated our ethical argument above, we're now interested in debating the evils of censorship. We're interested in building clear and complete threat models which allow us to understand all possible benefits for potential censors.

The engineering required for this has profound second-order effects which extend beyond computer science and into economics, psychology, sociology, and culture. This is most simply demonstrated by the fact that even if you have the computational power required to revert Bitcoin's history, there is still no benefit in doing so because any economic value you stand to receive will be eclipsed by the loss of trust (psychology), and therefore, value in the network as a whole (sociology and culture).

Censorship can be more insidious than overt reversions of history though. So, we need to develop even clearer cost models. Proof of Work (the consensus mechanism used in the Bitcoin blockchain as well as in Ethereum 1.0) actually fails to ensure that censorship is not profitable, since if you censor a block you can (i) take all of its transactions for yourself, and (ii) in the long run take its block reward.

Quasi Turing-complete object models (i.e. Ethereum) already provide interesting means of making censorship costly outside the actual consensus mechanism. For instance, we prevent the Halting Problem and denial of service attacks by assigning a "gas" cost to each operation and limiting the amount of gas per block to ensure all programs terminate. Much Ethereum 2.0 research is about both higher fault tolerance and stronger censorship resistance based on more complete threat models and cost analysis. Such research and implementation indicate why economic engineering is orders of magnitude more effective than legal lip-service.

Higher Stakes¶

In Understanding Ethereum we discussed how there are, for instance, no technical limits on what can be put in the

data

field of an Ethereum transaction, though there are economic ones: it costs 68 units of gas per byte. This feature crops up over and over, and takes on new dimensions in a full Proof of Stake architecture (a concept that will be discussed later in this module). The only technical limits to the data you can add are the amount of gas per block that the network can currently process. Of course, in order to have an effect on shared state, the data must be meaningfully encoded and validly signed, where “meaningful” is determined by shared rules that define what counts as malicious, and valid signatures adhere to the current elliptic curve cryptography used across the network.

Bitcoin uses mathematics in the form of elliptic curve cryptography to route around the need for human regulation and thereby ensure some degree of censorship resistance. Ethereum did this too. However, Ethereum 2.0 uses a different kind of mathematics - game theory - in addition to cryptography to ensure not just censorship resistance, but to prove objectively that censorship is asymmetrically expensive for those who would attempt it.

We'll be using value created on a public network to secure that same public network. This is the kind of profound economic feedback loop which lays the foundation for what emerges from network-centric protocols for money; what we have called digital gardens in which time flows differently.

Listen¶Further References¶