Censorship Resistance

Censorship resistance is not a switch that is simply on or off: it is an emergent result of how much redundancy and independent participation a network has. A blockchain earns the property by spreading transaction propagation and block production across thousands of unaffiliated full nodes and miners or validators, so there is no company, server, or jurisdiction that a regulator or attacker can pressure to stop a specific payment.

Mechanically, a broadcast transaction sits in each node's local mempool until some miner or validator, anywhere in the world, chooses to include it in a block. Because anyone can operate a miner or validator without approval, excluding one participant from the process does nothing: the transaction simply waits for the next willing block producer, which on a healthy network is usually only a matter of minutes.

The property has been tested directly. When the US Treasury sanctioned the Tornado Cash mixing contract on Ethereum in 2022, a large share of validators and relays began filtering related transactions from their blocks. Total censorship never occurred, since non-filtering validators kept including the transactions, but researchers found the network's resistance was more fragile than assumed. Ethereum developers have since proposed inclusion-list mechanisms, targeted for a future upgrade, to make selective filtering harder to sustain.

Real-world censorship resistance therefore has limits worth noting: filtering at the relay or mining-pool level can delay, though rarely permanently block, a transaction; geographic concentration of nodes or validators in a few countries weakens the guarantee; and centralized exchanges or fiat on-ramps remain separate chokepoints outside the blockchain itself. Despite these caveats, censorship resistance remains a defining reason public blockchains are used for payments, savings, and applications in regions with unreliable banking or restrictive capital controls.