Blockchain? Probably Not.

Most of what you are being told about the future of blockchain in the legal profession is nonsense.

Don’t get me wrong. Blockchain is very cool. And I am the furthest thing from a luddite you will find in the legal profession. I’m a part-time LLM student at the University of Alberta Faculty of Law and Department of Computing Science studying the automation of legal reasoning. I’m an ABA Innovation Fellow for 2018/2019, writing open-source software to automate analogy to prior cases. I am an advocate for the adoption of technology in law.

So I’m enthusiastic about blockchain being adopted for the things that blockchain can actually do. But there is a lot of information out there, and most of it is wrong. Like really, really wrong.

I was invited to write something for Slaw.ca designed to help. So here’s my attempt at providing you, dear reader, with a blockchain BS detector.

. . .

Blockchain is the technology that powers digital currencies. That is the fundamental problem that it was designed to solve. The problem the developers of cryptocurrency faced was this: how do we let people who don’t know each other and don’t trust each other trade a single currency with anyone else, anywhere in the world, with no middle-man or government involved?

That’s a big problem. You need a digital currency that can’t be forged or stolen. That means there has to be a single authoritative record of who has how much of it. But to avoid central authorities, everyone needs a copy, and it needs to be independently verifiable in each place. Each person looking at the record needs to be able to satisfy themselves, without referring to any other authority, that the record is correct, and current. And it needs to be universally available, with no possibility that anyone can prevent anyone else from viewing it. Plus the currency has to come from somewhere.

. . .

Here is the shorthand of how blockchain solves that problem.

In order to record a new set of transactions (a “block”) on a blockchain, someone has to do something hard with their computer. The difficulty adjusts until on average it doesn’t happen for a set amount of time, regardless of how many people are trying. It is an encryption task that involves the current data, and the data being added. It is hard to do, but easy to undo. So when it happens, everyone else can quickly check to see that it is correct, and confirm it, and pass it along to everyone else. The contents can still be read, all the way back to the first transaction. The encryption allows you to verify, without asking anyone else, that the data is accurate. After a short while, it’s essentially impossibly complicated to change the data.

In order to motivate people to volunteer to do the hard work (for their computer) of encrypting the new data as it arrives, an amount of crypto currency comes into existence and is given directly to the person who successfully encrypts the new data. This is called “mining.”

A public blockchain can be set up for a specific purpose, like BitCoin, or it can be created for general use, like the Ethereum blockchain. Whether or not cryptocurrency is what you use blockchain for, the mining system process is what makes it possible to run a public blockchain.

A private blockchain can be entirely inside a single organization, or among a limited number of actors. In a private blockchain you can more or less ignore the tokens that are used as a reward for mining. Some purists object to the use of the term blockchain to describe these private networks, preferring to call them distributed verifiable ledgers, or some other similar term.

. . .

Blockchain is designed to solve the problem of verifiability of data in the absence of trust. It makes a number of sacrifices in order to achieve that objective.

What are the sacrifices? Well, for starters, you have to use a blockchain. A new technology has costs associated with learning and implementing it. Blockchain is no different.

For another, if you are using a public blockchain, you cannot undo mistakes. Let’s say you accidentally post the unencrypted business plan for next year to the chain. It’s there, publicly available, forever. That is a risk that doesn’t exist with most other data repositories. Immutability, as it is called, is a double-edged sword.

On public blockchains, there are small, but perhaps not insignificant risks with regard to mining. The mining system adjusts its own difficulty in order to ensure it is always economic for people to start mining the cryptocurrency, as long as it has any value. But that system is limited in how quickly it can respond to changes in the value of the currency, or the cost of electricity. If changes happen faster than the chain can react, you might have temporary slow-downs in the processing speed of the chain.

Even if there is no slow-down in the blockchain network, the speed at which data can be added to the blockchain is set by the system, and can’t be changed after the blockchain is deployed. You will never be able to speed up the period of time between when data is added, and when it is verified and shared.

There is also an environmental concern about using public blockchains. Because blockchain mining operates as a competitive system, blockchain is hugely energy inefficient. The average BitCoin transaction involves the use of 300kWh of electricity. Around 80% of all mining happens in China, where the electrical system is largely coal-powered. The total amount of electricity being used on BitCoin mining is on the scale of a small country of a few million people, and growing.

. . .

But the question is not “does blockchain have costs and risks?” The question is “does blockchain have costs and risks that outweigh the benefits?” So what are the benefits of blockchain?

Blockchain allows for a single, shared, independently verified database of information that can be added to by anyone, but cannot be changed or restricted, all without requiring the parties on the network to trust each other or anyone else.

More specifically, blockchain distributes data widely without having a central copy, and provides the ability to verify the data without having to refer to a central version. You will note that the word “central” keeps appearing. We have ways of sharing data. We also have ways of verifying data. We also have ways of doing these things online. What blockchain adds to the toolkit is the ability to do these things without having to trust a central authority.

. . .

Whether or not blockchain is the right solution for a given problem depends on more than just what blockchain can do, and what blockchain costs. If something else can do 95% of what blockchain can do, and costs 5% of what blockchain costs, blockchain may still not be the best choice.

So to understand blockchain’s value, you have to understand its alternatives.

To understand where blockchain lies relative to its alternatives, you need to think about the difference between making something less likely and harmful, and making the same thing impossible.

. . .

Having your house burn down is a bad thing. It is a bad thing that happens, sometimes. We have a number of strategies that we use to mitigate that risk. There are building codes, fire extinguishers, conveniently-located fire stations, 911 systems, fire insurance, and even the courts that hear claims for negligence and police that investigate arson. But all of these things are mitigation strategies. Rather than making it impossible for your home to burn down, they make it less likely, and less important.

No one would suggest that the next time you build a house, your house ought to be completely fireproof. Well, they might, but you would likely scoff, and ignore them. Because making a house out of fireproof material would be insanely expensive in order to obtain only a marginal additional mitigation of the risk of fire. And there would be other downsides beside the cost.

Suggesting that someone should use blockchain for something is like suggesting that their house should be fireproof. Blockchain does not merely mitigate the risk that data shared among parties will be changed, or deleted, or access to it will be restricted. Instead, blockchain makes those things impossible.

So nothing is going to be built on blockchain that doesn’t fall into one of these two categories:

1. Stuff that cannot be built any other way, because only a system that does not require the users to trust anyone will work.
2. Stuff where eliminating the requirement of trust is not strictly necessary, but the value of using blockchain is higher than the value of using the best alternative.

(There is course a third miscellaneous category, which includes scams, experiments, and other adoptions of blockchain that are not sustainable.)

Cryptocurrency is an example of category 1 above. There may be others. But the vast majority of problems will fit in category 2, if at all.

. . .

Blockchain solves a trust problem by eliminating the need for trust where it is absent. Traditionally, we solve those problems by adding trust in a third party where trust between two parties is absent. People don’t trust your cheque, they trust your bank. They don’t trust your web server, they trust the SSL certificate issuer who says that you are who you say you are. They don’t trust you that you own the Manhattan Bridge, they ask the land titles registry.

Blockchain does a lot of things. It distributes data, it verifies data, it prevents outages, and prevents hacking. Prior to the advent of blockchain, we solve those sorts of problem using data security techniques which can be categorized as “backups and stuff.”

You can see that these are both mitigation strategies. Backups and stuff don’t necessarily prevent data manipulation (though companies like Amazon and Google seem to do just fine using “backups and stuff.”). They just make them less likely, and mitigate the damage if they happen. Similarly, trusted third parties can also act badly, which is why the legal trust safety industry exists.

It’s not perfect. It’s like fire insurance, not fireproofing. But the combination of trusted third parties and backups and stuff has been good enough for all of human civilization until now. Blockchain might be better, but only if it provides more of what we want, or avoids more of what we don’t, and does so efficiently.

. . .

So when I am trying to figure out whether or not blockchain is a good solution for something, I ask myself this question:

“Who is going to implement blockchain for this purpose, and for that person, does blockchain seem like it has a better cost/benefit value than trusted third parties and backups and stuff?”

. . .

Let’s try applying that test to some of the proclaimed benefits of blockchain in the legal profession. For starters, let’s talk about land titles registries.

Yes, blockchain could be used to provide more efficient access to land titles records. But more efficient access to land titles records could also be provided by any other number of online database technologies that are better understood, and don’t have the double-edged sword of immutability. The only thing that blockchain would have over those technologies is that you would no longer need to trust the land titles registry.

I have heard a lot of complaints about the land titles registry in my career, but distrust of the land titles registrar is not one of them. But even if buyers and sellers are tired of the land titles registry’s constant lying, that is data controlled by the government, and it will not be put on the blockchain unless the government perceives some benefit to go along with the loss of control that would entail. What does the government gain from a technology that no longer requires you to trust the government?

Blockchain would also increase the quality of the land titles service because you will be able to verify the state of title at the time of registration in an automatic way. But independent verification is not the problem that blockchain solves. Blockchain solves the problem of independent verification without reliance on trust. If you still trust the government to be in control of the data (and even if you don’t, I’m pretty sure the government does), then you can achieve the same thing without blockchain by using verified database technologies like AWS QLDB.

So no; blockchain is never going to be used for land titles. That’s nonsense.

. . .

Next, let’s talk digital identities. The claim you hear is that people will all have their own identifying information on the blockchain, and have third-parties add their relevant authoritative data to it. So perhaps your university would add your degrees to the chain, and your government would add your driver’s license. Everyone will just be able to give the keys to anyone who needs identity information, and that information can be verifiably downloaded from the blockchain all in one place.

So this anticipated use of blockchain would involve the person handing you their blockchain ID number, and you would go to the blockchain and download something that says “I have verified that the person who looks like this has the following name,” for example.

That’s great. Except who put the information on the blockchain, and why do you trust them?

Wait, no. I don’t care why you trust them. I only care that you trust them.

If you trust them, then you don’t need the information they provide to be on the blockchain to have confidence that it has not been altered. Just verify that it came straight from them, which can be done with basic digital signatures and a trusted certificate authority. There is no reason to use blockchain for this. Nonsense.

What about the digital identity certificates that underlie encryption technology themselves being put on the blockchain? Extremely bad consequences can arise from losing access to them, particularly once they are more widely relied on for things like banking. But the certificate is only valuable if either you, or a third-party, has verified the person’s identity. If it’s you, just keep the certificates yourself. You usually trust yourself.

If it’s a trusted third party who is confirming that a given website is owned by a given person, and you trust them to tell you the truth about that, why would you not also trust them not to use backups and stuff appropriately so as to avoid losing access to the data? That’s what we do now. As far as I’m aware, there has not been a spate of security certificates suddenly becoming useless because the certificate authority was hacked, or went out of business. It solves a problem that doesn’t exist.

. . .

Smart Contracts. Oh, man. Where do we start?

First, a smart contract is to blockchain as a boat is to Hudson’s Bay. You might find one there, but it’s not the only place, or even the most likely.

Smart contracts and blockchain are not the same thing. You can do blockchain without smart contracts, and you usually do. And you can do smart contracts without blockchain, and you usually will. Too often people conflate smart contracts with blockchain because of the technologies that are out there to make them cooperate with one another.

The Accord project, for example, is seeking to create a standard computer language for expressing contracts. One of the hopes is that you would be able to use that language to generate software code that lives and runs on the Ethereum blockchain, called a distributed application, or a “dApp”, which automatically executes the requirements of the contract.

But not all self-executing smart contracts will be on blockchain. Nor do all smart contracts need to be self-executing. Encoding a contract can be done for the purposes of analysis, or as a means of building applications to help people execute the contract accurately, or for contract management, or for any number of purposes which have nothing to do with trustless execution.

Blockchain aside, smart contracts are a future area of law. There are problems to sort out. One major issue is how to integrate the court system with smart contracts to retain the best parts of both. There are questions about what is the contract, the code, or the written agreement to that code was supposed to model? If the code is the contract, did you have a lawyer who could read that code, and advise you on it? If not, what does it mean to have independent legal advice about a contract? What happens when you need to sue, but you can’t explain the contract code to the judge?

Smart contracts really will impact the future of the legal profession. And a part of that will involve understanding how smart contracts were implemented over the blockchain, if they were implemented over the blockchain. That is not nonsense. But it will also be possible to have a thriving smart contracts career that has nothing to do with blockchain. Saying that a smart contracts lawyer needs to know blockchain is nonsense.

. . .

Whether or not blockchain is the right solution depends primarily on one factor: Trust. Whether or not blockchain is the right solution depends mostly on how much the absence of trust is the problem, and it cannot be solved with third parties and backups and stuff.

Blockchain is not the next industrial revolution. It is not on the scale of the printing press and the Internet. You and your clients are not going to be using the blockchain for everything. It is not AI. It is not the same thing as a smart contract. Courts and law firms will find little use for it. Governments and businesses will find use for it primarily when dealing with people in other jurisdictions, or untrusted parties.

Take a deep breath. Relax. Everything is going to be OK.
_______________

Jason Morris is a lawyer at Round Table Law in Sherwood Park, Alberta, and an LLM student in Computational Law at the University of Alberta.