RegTech: Knowledge, Belief and Suspicion and the trap of Wilful Blindness.

RegTech: Knowledge, Belief and Suspicion and the trap of Wilful Blindness.

Nigel Morris-Cotterill

I first wrote about wilful blindness in 1996 in my book "How not to be a money launderer" and I said then that it was destined to be the central point on which all counter-money laundering laws would hang. I wasn't wrong then and I'm not wrong now.

To understand it, we'll take a quick romp through knowledge, belief and suspicion and then why machines cannot and for the foreseeable future will not be able to identify suspicion.

Blog / cast:

To understand it, we'll take a quick romp through knowledge, belief and suspicion and then why machines cannot and for the foreseeable future will not be able to identify suspicion.

There are many areas of life where we act on knowledge, belief and suspicion.

There are many jobs that depend upon the ability to differentiate between them and to give appropriate weight to them.

For example, if we were to ask those working in the media whether they are journalists or reporters, many would say they are journalists but, in truth, many are reporters that is to say that they report what they see or hear or what another person tells them what that other person has seen or heard. Or they republish, perhaps with minor style changes, press releases.

Reporting: Mrs Jane Smith died yesterday when she was hit by a bus in Market Street.

Journalism: Mrs Jane Smith, 41, was killed yesterday when she ran in front of a bus in Market Street. Passers-by said that Mrs Smith’s pet miniature poodle had jumped out of the bag she was being carried in and ran into the road. Mrs Smith and her dog were well known to both shoppers and market traders. She lived nearby and went to the market daily, Mr Peter Potatohead, a greengrocer, told this newspaper. Several passengers on the bus were injured when the driver, Mr Gordon Scott braked hard to try to avoid hitting Mrs Smith. Mr Scott, 54, is a veteran of more than 30 years driving City buses. ″I will be retiring soon and I have never had any serious accident until today. It is very upsetting but there was nothing I could do as she ran out just feet in front of my bus.″ A police spokesman said ″we have reviewed the facts of the accident and have taken witness statements. It is obvious that the driver of the vehicle did everything he could to avoid hitting the victim and it is clear that the victim ran into the road with other things on her mind and simply did not think about, see or hear the bus. We are not pursuing any other lines of inquiry.″

Both versions contain facts. But there is also opinion.

The newspaper will, or should, check the facts.

Yet, when a newspaper receives a press release, if it chooses to publish it, it will review it for libel or criminal purpose and perhaps for house style (spellings, certain phrases, etc.) but it will not be checked for the accuracy of fact. This is because the newspaper ″reports″ the press release. It does not write about it in the sense of journalism.

We are not here talking about so-called citizen journalists, bloggers (yes, I see the irony) and other so-called social media writers. We are talking about established, reputable media which is, at least in theory, supposed to present a balanced view of the stories they present or, if they don’t, to present them as opinion and declare that is what they are.

Media entertains guests, which it very often refers to as ″experts.″ TV programmes all too often describe the person they feature as ″the best.″ Or as ″world leading.″ That description is almost always hyperbole and very often wrong. The journalist presenting the story is trying to create a halo effect in which he looks good because he has someone ultra-brilliant to talk to.

So, why do we have this in relation to financial crime? It’s because there are the same central problems.

We all have to understand knowledge, belief and suspicion.

Knowledge comes from evidence and irrefutable fact. That’s easy, then. Because a fact is a fact, right?

Well, no. It’s not right. What about changing medical opinion? In the mid 20th Century, medical opinion was that moderate smoking of cigarettes (which did not have filters, remember) was good for you. But by the end of the 1950s, there was a small but growing body of opinion that it was bad. Soon, the fact that smoking is good for you became the opposite of a fact.

But, interestingly, it is a discredited fact that tried to come back more than half-a century later when CoVid-19 was found to infect via the nasal passages, there was a body of opinion that said that smoking would prevent the virus getting into the nasal passages. It was, of course, rapidly debunked, as were claims that the virus could be cured by drinking bleach and other improbable or downright dangerous suggestions.
The serious point, however, is that there were people who were convinced that such claims were true, that they were fact, that the facts we so strong as to be the basis not of belief but of knowledge.

We can look around the internet and find extremists, fraudsters and crackpots advancing all manner of theories and trying to establish them as ″fact″ and, if they are fact, that anyone who does not accept them as such, and therefore knowing them to be true, is dangerous.

Religions are referred to as ″belief systems″ precisely because we cannot prove the existence of God. Indeed, the strength of religions is that they require ″faith.″ This in an acceptance of something which cannot be proved, as if it were true.

Belief comes from information and demonstrable fact. The facts do not demonstrate the single most important thing: the existence of God. The facts demonstrate a tendency to establish some non-human force. It can be the appearance of the face of a deity in a potato, a statue that appears to cry, even something so beautiful that it is hard to conceive of it coming from the mind, much less the hand, of man.

However, while it may shore up belief, it is not proof, it is not demonstrable fact, it is not evidence.

We see the same approach in relation to, for example, environmental issues. The world has sold its soul to electric vehicles but they move and mutate the environmental issues of fossil-fuelled vehicles. We know that windmills are not very efficient and that they have to be millions of them to replace fossil fuel plants. But by the time we reach that point, we will be on Generation X of the windmills; there is no viable plan for removing old windmills set in the ocean floor nor for the disposal or re-use of the whole thing. How far is this into the future? It isn’t: it became a problem when the first generation began to fail several years ago. Be it knowledge or belief, it is clear that it is not the complete solution we are being sold.

Solar panels are not the solution, either. Those that are entirely silicon and metal can, in fact, be very close to totally recycled. But in only April 2020, it was being posited that ″landfill disposal of solar panels likely safe for humans.″ That was ″″ Is that a fact? A month later PV magazine said ″there’s limited interest in recycling. But given the presence of heavy metals, such as lead and tin, if waste is managed poorly, we’re on track for another recycling crisis.″ Is that a fact? Are they inconsistent? If they are inconsistent are they opinions? What value can we place on opinions?

And this is before the environmental disasters of mining the metals needed for batteries and the problems of disposal of batteries that have reached their end of life. The e-waste crisis that much was made of in the 2000s has been replaced with a dead-battery crisis that is receiving almost no publicity. Group think says electric is best but only by those seeking to abolish fossil fuel and, perhaps, nuclear power plants. Stand back and look at the problems and a different perspective appears.

This returns to the question, can we identify that which is evidence, information or intelligence? And to what extent is opinion the driving force?
Counter-money laundering and anti-terrorist financing laws are designed to prevent those offences, and those for tax evasion and the prevention of bribery and corruption, and regulations made under them. They all speak of ″knowledge″ and ″belief.″

Courts require facts to be proved by evidence. Facts are not proved by information or opinion, other than opinion of genuine experts and even then they take second place to physical evidence.

But it’s when we get to ″suspicion″ that things become very woolly.

Suspicion is not proved.

Suspicion may be founded on facts but if those facts, combined, amount to certainty, that’s knowledge.

But often, suspicion is based on ″intelligence″ which is about as far away from evidence as one can get. Intelligence is unsubstantiated; it is not corroborated; it is not far removed from gossip.

This is where it gets silly. Silly but very serious.

A court may decide ″as a fact″ that you were suspicious because the judge or jury takes a different view of the intelligence, information or evidence in your hands. Laws talk about ″had reasonable cause to believe″ or ″had reasonable cause to suspect.″

This is the first step in a concept known as ″wilful blindness.″

Wilful blindness is first found in English law in the late 19th Century which, elsewhere, I have described as ″the golden age of English jurisprudence.″ And it was. Many of the concepts that we find all over the world, not only in Common Law systems, originate from the senior English courts of that period.

Wilful blindness isn’t complicated but prosecutors and judges find it difficult to explain to jurors and in this short piece, I’m going to skim over it too. But it’s vitally important: we find it all over the world in financial crime laws, often phrased in ways that make it appear as if the draftsman was trying to disassociate the concept from its origins (yes, Australia, I’m talking about you, amongst others) .

To summarise: it’s where a person who is required to make an informed decision about something deliberately, recklessly or negligently fails to ensure he has the relevant information to enable him to do so.

A person who doesn’t ask what’s in the package someone has given him to hand carry on an aeroplane can’t escape liability: he should have asked. If he didn’t ask he
a) elected not to or
b) didn’t think about it or
c) chose to run the risk when he was at the check-in desk and the counter-clerk asked ″has anyone asked you to carry anything for them?″

So where does this leave us?

The fact is that a Court can find that there was Knowledge, Belief or Suspicion if a reasonable person with the defendant’s characteristics (language, education, background, experience and the proven facts, information or intelligence available to him) would have been suspicious.

The second is that this is always, by definition, second-guessing. It’s a decision reached ex post facto. Worse, the pressures on a jury are very different to the pressures on a person working in a bank, etc. and having to decide if something is suspicious. And the information is presented to them in a very different way to how it arrived, if it did arrive, before the defendant in the course of his routine job.

The third is that we may think we only know what we know but we often know, or have information or intelligence relating to it, that which we don’t know we know.

The fourth is that some people think that knowledge and belief are the same; they are not. Others think that suspicion requires proof: it does not.
Suspicion is an emotional reaction to that which is before us. It is a state of mind It’s like or not like; trust or not trust. It is vague. It is a value judgement, often reached in a matter of moments.

This is why machines cannot reach suspicion. Machines cannot develop a state of mind. They best they can hope for is that someone has risk-scored every possible relevant circumstance and set value that, if exactly those circumstances were present at that time and in that order, a human would consider to be suspicious. That’s not going to happen, no matter what the technologists say. Why? Because machines can’t feel. They cannot think. They cannot make value judgements.

Can a machine be ″wilfully blind?″ The answer to that is ″no″ and that’s why the answer to ″can a machine define suspicion″ also has to be ″no.″
And that’s a fact.

For now.


The topics in this blog are expanded upon, with many other approaches to understanding suspicion, in my book ″Understanding Suspicion in Financial Crime.″ which, in May 2021 will go out of print and will be revised, updated and added to as an e-learning course at .