Arvind Narayanan, an assistant professor of computer science at Princeton, wrapped up his Bitcoin and Cryptocurrency Technologies course with a warning: not everything that can be decentralized should
Arvind Narayanan, an assistant professor of computer science at Princeton, wrapped up his Bitcoin and Cryptocurrency Technologies course with a warning: not everything that can be decentralized should be.
In his final lecture, Narayanan examined how blockchain technology might reshape various sectors over the coming decades. He acknowledged the technology's potential but pushed back against what he called a kind of missionary zeal in the crypto industry. His framing of the problem was direct: "What I mean when I say the future of bitcoin is a set of ways in which blockchain technology can be used, as proposed, to decentralize a variety of things: stocks, bonds, even property (whatever that means — we'll see). So, in other words, people have looked at bitcoin and went, 'Hmm. We managed to decentralize currency. That actually worked. Let's decentralize everything.'"
Why would anyone want to decentralize everything?
That gap between technical possibility and actual utility is where Narayanan places his focus. Many blockchain projects chase decentralization as an end in itself, he argued, rather than building something people need. The concern appeared elsewhere. At the Inside Bitcoins conference in New York, the Bitcoin 2.0 panel raised similar questions about whether removing intermediaries improves how users experience a service.
Narayanan framed the tension around cryptography and trust. The technology exists where parties don't start out trusting each other, he explained. That's not the same as saying trust should vanish from the world. He laid it out: "I feel that there's often this confusion between two things. One is the fact that cryptography is often used in contexts where, unfortunately, there's not much trust between entities — so the lack of trust is a starting point and cryptography is a solution. This often becomes confused with, 'Oh, now we have this hammer of cryptography. Let's try to use this to move to a world where nobody has to trust anyone anymore. Trust minimization is not the goal. Lack of trust is not the model we're hoping to move to. It is, instead, our unfortunate starting point.'"
Narayanan used a car sale to test his ideas. The blockchain could handle this transaction without a middleman. But the example revealed obstacles.
Security poses the first problem. People struggle with private key management. Someone who loses a key to a car held as smart property on a blockchain ends up with a worthless asset. Narayanan saw security as more than a technical matter. It's a human problem.
Multisig addresses allow multiple parties to share control without a central custodian. But funds stay locked during the mediation period. This creates an inefficiency that never existed before.
Blockchain ledgers can't enforce anything in the physical world. A dispute arises, and there's no institution to call for help. Narayanan acknowledged that decentralization might speed up transactions when everything goes right. When something breaks, it offers no remedy.
New tokens can represent real assets. Colored coins could track a car's ownership. But someone still has to prove that the token corresponds to an actual vehicle. The issuer must be trusted.
Dispute mediation is specialized work. Property rights demand expertise and involvement with the court system. Moving that onto the blockchain means participants lose access to that knowledge. They get to choose their own mediators, but something gets sacrificed.
Narayanan revisited his example to sum up: "Let me be very clear. I deliberately picked this car security and car sale example as sort of an extreme way of illustrating why we really need these human institutions."
The second part of this analysis covers where blockchain and decentralization do add value.
