You can’t democratise trust

August 2nd, 2016 by Peter Evans-Greenwood | Centre for the Edge

There's a lot of talk around the interwebs on using technology to democratise trust. Blockchains, in particular, are supposed to create some new fangled type of distributed or networked trust, transforming society in the process.

Hillary Clinton has proposed blockchains as a solution to the lack of trust in public institutions. Blockchains might be able to heal the rift Brexit created, according to Don Tapscott. There's even talk of new trust layers, liberating trust from central institutions, making it frictionless.

The problem is that you can't democratise trust. Nor is trust fungible.

A fairly conventional definition of 'trust' would be "a firm belief in the reliability, truth, or ability of someone or something." Consequently trust is a subjective experience, it's a measure of our confidence that some actor will behave as we expect. Trust is something we construct internally in response to observing a consistent pattern of behaviour.

We trust our car as, based on past experience, we don't expect it to fail and leave us stranded in the dark. We trust an institution — firm or government — as we've personally found its actions understandable and consistent. We trust 'Honest Ed' as he hasn't lied in the past (or, at least, we're not aware of him lying).

This means that trust is contingent and can't be assumed. Trust is easily lost if an actor behaves against expectations, and can naturally decline over time if our relationship with the actor isn't actively maintained. Someone we trusted yesterday might, though action or inaction today, lose our trust tomorrow.

So what does technology have to do with trust? Not as much as we think.

As one of my colleagues pointed out, we could create a register on a public blockchain (or any publicly accessible register really) containing stats of successful (and unsuccessful) operations by every surgeon in the country, and do nothing to increase trust. Does this mean that you could use the register to pick a surgeon that you trusted to do a bad job?

What technology can do is provide us with tools to improve our management of risk.

Consider the emergence of money.

We're often told that money was a response to the problem of the double coincidence of wants. I have a fish and want a bicycle, you have a bicycle and want a hat, but we cannot barter as our wants don't coincide. This has it all wrong though. Barter is something that we do with people we don't know. Most exchanges, back in the day, were with people we did know, facilitated via debt, bilateral obligations. Money was of minor interest in many communities until the Industrial Revolution when the mass market was created, and when colonial powers used money to force communities into their economies.

Money is a technology for managing the risk inherent in exchanging value with a counter party that we don't know, or don't trust, by providing us with an intermediary – money – that we do trust. We trust commodity money as we believe that its value won't collapse, the commodity being generally useful and therefor desirable. If it's fait money then we're trusting the government to honour its debts. Indeed, one of the reasons governments regulate money is to stabilise its value, making it more trustworthy. (Which is also the reason why the recent moves by the Ethereum team to roll back the ledger are so bad for the Ether.)

Technology is a tool to manage risk. We can use technology to ensure that we have better and more current information, and thereby mitigate risk, typically by reducing the opportunities that a bad actor can play one stakeholder off against by making it harder for them to repudiate what they said or to say different things to different people. “The cheque’s in the mail”, and all that. This is the benefit of using a distributed ledger to facilitate trade finance, or any other multi-party business process, breaking information out of silos to ensure that all parties are fully informed. Or we can use an intermediary both parties trust to facilitate an exchange, with money one possible intermediary. Any broadly accepted form of money provides this benefit, including a cryptocurrency assuming that both parties trust it.

Where we're using a very broad definition of technology, along the lines of “tools and practices deliberately employed as natural (rather than supernatural) means for attaining clearly identifiable ends". 

Trust is different to technology. We can't democratise trust. Trust is a subjective measure of risk. It’s something we construct internally when we observe a consistent pattern of behaviour. We can't create new kinds of trust. Trust is not a fungible factor that we can manipulate and transfer.

The idea that technology enables us to democratise or distribute trust is attractive as it promises a quick fix to the challenging problem in rebuilding trust in a society where many think our institutions are irreparably broken. There are no quick technological fixes though. If we want to rebuild trust then we need to solve the hard social problems, and create the stable, consistent and transparent institutions (be they distributed or centralised) that all of us can trust.


blog comments powered by Disqus