Smart Contracts and Web 3.0: The Evolution of Law?

May 2, 2017

Think about the last time you lent money to a friend. You
probably did it because you trust them. What about your savings? Are they in
the bank or in a stocks ISA? Why there? Probably because banks and markets are
large and stable, they offer guarantees and so you trust them. 

Now when was the last time you invested in an Argentinian
winemaker? An Indian start-up? What about the Filipino engineer who needs to
patent his idea? These might all be great investments, but most people aren’t
involved because they don’t know or trust the participants and the costs of
enforcing on a bad deal are disproportionate. You’re no expert in Argentine
law, so who knows if you’ll get your money back. If you do make these sorts of
investments, you’ll probably do so through many layers of banks, investments
companies and consultants – each taking their cut along the way. 

What if this could change? It can change and it is changing.
The last three years have seen a rapidly emerging technology that creates a
secure trustless world network running “smart contracts”. What does
that mean? Put it this way: while Uber made everyone a taxi driver, eBay made
everyone an auctioneer, and AirBnb made everyone a hotelier … blockchain
technology lets anyone build a legal system. In this article we explore how the
blockchain works, and what it means for lawyers.

The Origins of the Blockchain

Understanding the blockchain requires a bit of a history
lesson, which starts in the heady early days of the internet. Back in 1999,
bored college students figured out that they could compress the music on their
CDs and share it (illegally) with others over the internet. They would upload
their files to Napster’s servers, who would send them on to anyone who asked.
The weakness is obvious in hindsight: Napster was the single point of failure
and soon enough, the lawyers came knocking.

Except the lawyers didn’t come soon enough. At its peak,
Napster had over 80 million users, now all disgruntled and looking for an
alternative. By 2002, Bram Cohen at Buffalo University had invented BitTorrent:
a completely decentralised file-sharing system with no single point of failure.
Whilst the early versions still required a central tracker (not to store the
files, but to say who had them), it was soon followed by Distributed Hash Table
(DHT) technology which sent that look-up information to hundreds of peers
across the network, obliterating any single point of failure. It worked: today
BitTorrent has over 250 million worldwide users and accounts for about 25% of
all internet traffic.

Decentralisation gave people ideas. Napster’s servers
worked like banks, storing everyone’s money and keeping track of who owned
what. The 2008 crash taught us that banks are not impregnable. What if we could
decentralise the money system?

In January 2009, an anonymous software engineer using the
pseudonym Satoshi Nakamoto released the first build of a new product called Bitcoin
Core. Bitcoin brought together decentralised technology with cryptographic
techniques that had already developed elsewhere.

Cryptography in the blockchain

Cryptography has been around for thousands of years, but
until recently suffered from a singular problem: the sender and the receiver
had to agree on a secret key before any messages could be sent. This required a
meeting, or at the very least some form of non-secure communication before
secure communication could begin. That was a problem.

The solution came in the 1970s. Public-key cryptography
uses pairs of keys – a public
and a private one – to encrypt and decrypt messages. A public key can be widely
promulgated. Anyone with your public key can run it through a special algorithm
to produce an encrypted message that can only be decrypted with the
corresponding private key (which the recipient keeps secret). For the first time,
it was possible to send an encrypted message to a person you had never met.

Around the same time, cryptographers needed a way to
check that a message had not been corrupted or tampered with during
transmission. Various cryptographic ‘hash functions’ were invented to map a set
of data of arbitrary size (the input) to a string of fixed size (the hash). The
best hash functions would radically alter the hash given very minor variations
in the input. Given the message and the hash, it was possible to verify whether
the message received was exactly the same as the message sent.

The blocks were beginning to fall into place. It was
possible to send encrypted messages to anyone along a coordinated but
decentralised network.

So what is the blockchain?

Blockchain is best described in the context of an example.
Bitcoin is one implementation of blockchain technology, and a good example.

It works like this. Anyone can join the Bitcoin network by
creating for themselves a public/private key pair and connecting to local nodes
through a Bitcoin client. Let’s assume Alice has done this, and she has five
Bitcoins. She wants to send one to Bob. Alice broadcasts a message to the
decentralised network, with Bob’s public key and the amount she wants to send.
She hashes the message and signs it with her private key so that everyone knows
that it is her making the broadcast. The nodes around her check a global ledger
to make sure that she’s got enough Bitcoins, and they check her signature is
valid. When enough nodes validate the transaction, the ledger is amended to
reduce Alice’s account to four Bitcoins, and increase Bob’s account by one. The
ledger is public, and everyone holds a copy.

Every 10 minutes, certain participants in the network
(called ‘miners’) collect all validated transactions into a block. In order for
a block to be accepted into the network, the miner must create proof-of-work.
Proof-of-work regulates the supply of Bitcoins, which is essential to preserve
value in any money system. Proof-of-work is achieved by solving a complex
cryptographic problem, designed to be hard to find but easy to verify. The
proof-of-work requires miners to find a number called the ‘proof’ (or ‘nonce’),
such that when the block content is hashed along with the proof, the result is
numerically smaller than the network’s current difficulty target. Every couple
of weeks, the difficulty target is automatically adjusted to keep the mining time
to about ten minutes. Miners compete to solve each block, and the winning miner
is rewarded with 12.5 Bitcoins (currently about £15,000).

A solved Bitcoin block contains four things: a timestamp,
a hash representing all transactions in that block, the proof found by the
miner and – importantly – the hash of the previous
block, thereby forming a chain.

The chain is important because it prevents attacks on the
network, and attacks must be expected on any money system. If an attacker
controlled enough nodes, they could authorise a fraudulent transaction and
publish it to the ledger. There would then be two ledgers: one with the
fraudulent transaction, and another created by honest users who invalidated the
fraudulent transaction because it didn’t follow the rules. For the fraudster’s
ledger to be accepted, he would need to solve the block faster than the honest
users, so would need to control more than half of the computing power in the
whole network for at least ten minutes. Even if he managed that, he would need
to keep solving blocks faster than anyone else, every ten minutes, to keep
ahead of all the other nodes that contradict his blockchain history. To
historically alter the blockchain is even harder – the attacker would need to
fork the blockchain by solving each and every cryptographic challenge in the
network for as far back as he wanted to go – requiring orders of magnitude more
computing power than the rest of the network put together. There comes a point
where controlling that much computing power just isn’t worth the reward.

Bitcoin has been wildly successful. It is accepted by
PayPal and many online retailers, as it offers zero transaction fees when
compared to the 2-3% levied by credit card companies. It is increasingly
popular in China and other jurisdictions that strictly control their national
currencies. At the time of writing, the value of the world Bitcoin supply is
about £18.1 billion.


Whilst Bitcoin is important, it is the underlying
blockchain technology that is the real prize. Just as decentralisation got
people thinking about blockchains, so blockchains have people thinking about
other kinds of trustless networks. Contracts are the obvious candidate.

The state creates money to support commerce. It imposes
two rules on using money, which may be so obvious that we do not even recognise
them as rules. They are: (1) you can’t spend more than you have; and (2) you don’t
still have money after you’ve spent it.

The Bitcoin network applies these rules programmatically
when it validates a transaction.

But what if we extended the platform to execute any rules we wanted?

When two or more people write down private rules for
their conduct, we call that a contract. We put our trust in contracts because
we know that the court system can step in when contracts are broken. But
courts, as we know, are slow, expensive and occasionally unpredictable. They
might work well nationally, but who is going to start an international
arbitration over a £50 debt? Smart contracts provide the answer.

Smart contracts are a way to reduce obligations to
executable code and have it executed by a cryptographically secure worldwide
network. Let’s say a bank enters into a smart contract car loan. Whilst the
loan is outstanding, the borrower can drive the car but the bank retains the
right to stop the borrower selling it whilst the loan is outstanding. If the
borrower defaults, the contract rescinds access to the car and grants control back
to the bank. If the loan is repaid, the bank’s rights to the car are deleted
and the borrower assumes complete control.

The ability to reduce contracts to code has existed for
decades, but has never gained traction because we have never before had a
secure trustless network, outside the control of either contracting party,
which we know will execute the contract in accordance with its terms.

In July 2015, a young Russian programmer named Vitalik
Buterin designed a blockchain-based system called the Ethereum Virtual Machine
(EVM). It has generated an enormous level of excitement in the technology
industry, and its currency, ‘Ether’, is already second in value to Bitcoin
worldwide. The EVM is a Turing-complete computer capable of executing scripts
on an international network of public nodes. It is similar to Bitcoin but
extended to run any kind of contract, effectively as a cryptographically-secure
world computer.

How does it do this? Well, the Bitcoin network and the
EVM network both have a ledger that records which Accounts hold currency (Bitcoins
on the Bitcoin network; Ether on the EVM). In addition to Accounts, however,
the Ethereum network also holds Contracts (as compiled code) and records the
machine state of each Contract on the network. Users pay tiny amounts of money
(called ‘Gas’) to have the network run cycles of their contract and move money
around the network. The amounts really are tiny, especially when compared to the
2-3% fees charged by credit cards: the average transaction today costs about
half a penny, irrespective of value.

The big advantage of Ethereum – and smart contracts – is
that they are automatically executable. If you have a stock option that is not
honoured, you have to go to court, secure an injunction and maybe even call in
the bailiffs. With Ethereum, that option automatically executes on the network
when its conditions are met, moving money between accounts without user input. 

Ethereum and Real-life Law

Does this mean the end of lawyers? In short, no. Law is
flexible; it requires interpretation and judgement but is corruptible and
sometimes uncertain. Machine code is rigid, inflexible and absolute. There are
roles for both solutions.

Forming the junction between life and code does not
necessarily require human judgement, but often it will. In the car loan example
above, the question of whether a loan has been paid is a binary one, capable of
being rationalised by a machine. By contrast, a relatively simple contract to
paint a house might require the subjective evaluation of a human where the
quality of the workmanship is in dispute.

Smart contracts might deal with subjectivity by
incorporating call-out functions. If the painter is not paid, the owner might
be required to raise and specify the dispute within a certain time, failing
which the painter is paid automatically. Once the dispute is raised, the smart
contract code might then define its parameters (‘Is this workmanship
adequate?’) and transmit that question to a third-party arbitrator. The jurisdiction
and scope of the dispute is pre-defined, reducing the potential for satellite
litigation. A bid/offer system might allow the owner to offer less than the
contract value, putting pressure on the painter to accept something less than
the price in an effort to avoid the cost of an arbitrator’s intervention
(similar to CPR Part 36 in England). By making the losing party liable for the
arbitrator’s costs, we can disincentivise the raising of unmeritorious

Ultimately, the subjectivity call-out functions could
themselves be contracted out to the network. Let’s say an aggrieved party
submits evidence of their grievance to 100 human ‘judges’ across the network,
who vote on the outcome. The ‘judges’ can themselves gain trust and respect from
the network by consistently voting in line with (what transpires to be) the
consensus, such that vote weight can be adjusted in favour of those who have
demonstrated competence and impartiality in the past.

Not everything can be reduced to written evidence and
there will always be a role for inspection, cross-examination and advocacy.
Smart contracts are, and probably only ever will be, a way to reduce basic and
binary disputes to a simpler, cheaper and more certain means of dispute

What’s the point?

Smart contracts reduce transaction costs and improve
trust in those transactions. By implementing what would, in effect, be a global
legal system for private law, Ethereum allows individuals to cut out the
middleman and invest directly in places they might otherwise ignore.

To return to my opening examples, why wouldn’t you invest
in that Filipino engineer if you knew (with cryptographic certainty) that you’d
get your money back if his patent was rejected? Why not buy equity in that
winemaker if you could be sure that you’d be automatically paid a share of his
profit? Low-cost, high-trust transacting unlocks the economic potential of new
markets and new parts of old ones.

Decentralisation also has social utility. A decentralised
information network cannot easily be censored. A decentralised money system takes
control away from governments (see, eg, Bitcoin’s current popularity in China).
Right now, a decentralised microblogging platform called Eth-Tweet prevents anyone
but the original poster removing their post.

Whilst this talk of cryptography and blockchains might
sound very abstract, there are already real-world blockchain applications.  New start-up creates Wi-Fi connected
locks for bikes, lockers and apartments – designed to interface directly with
smart contracts. WeiFund is a Ethereum-based crowdfunding platform, which
creates individual smart contracts between backers and pitchers. KYC-Chain is
positioning itself as a trusted gatekeeper for consensus-based, and KYC
regulation compliant, digital identities.

Blockchain technology has yet to reveal its full
potential. Right now, it is in its ascendency. The next few years will be an
interesting time for lawyers and inventors. 

Peter Brogden is a technology barrister at Keating Chambers in London.