Tag Archives: James Surowiecki

Logical Fallacies II – Evolved To Be Illogical?

7 Sep

We’re all guilty of flawed thinking because our brains evolved to win others round to our point of view – whether or not our reasoning is logical.
– Dan Jones

Continuing from the previous post on this topic, in a recent New Scientist article ‘The Argumentative Ape”, Dan Jones describes some cognitive biases and argues that they may in fact have evolutionary advantages even though they mean we often make logical mistakes as a result.

He describes the example of  “confirmation bias – the mind’s tendency to pick and choose information to support our preconceptions, while ignoring a wealth of evidence to the contrary“.

He says that we are much better at spotting the flaws in someone elses arguments, but blind to the faults in our own. Though there are ways around this, for example I’ve seen other research that shows if we try to justify something in a second language they we will be less biased in our argument.

This ability to argue back and forth may have been crucial to humanity’s success – allowing us to come to extraordinary solutions as a group that we could never reach alone.

He says this helps us to:

– discern whether to trust someone

– helps us toward more critical thought

– give us an ability to convince others of our point of view

Consider the confirmation bias. It is surprisingly pervasive, playing a large part in the way we consider the behaviour of different politicians, for instance, so that we will rack up evidence in favour of our chosen candidate while ignoring their competitor’s virtues. Yet people rarely have any awareness that they are not being objective. Such a bias looks like a definite bug if we evolved to solve problems: you are not going to get the best solution by considering evidence in such a partisan way.

But if we evolved to be argumentative apes, then the confirmation bias takes on a much more functional role. You won’t waste time searching out evidence that doesn’t support your case, and you’ll home in on evidence that does

In addition to confirmation bias and the framing and attraction effects, [there are] many other seemingly irrational biases that might be explained by our argumentative past, including the sunk-cost fallacy – our reluctance to cut our losses and abandon a project even when it would be more rational to move on.

Jonathan Haidt says “we are simply trying to justify our gut reactions and persuade others to believe our judgments, rather than attempting to come to the most just conclusion. He says ‘Moral argumentation is not a search for moral truth, but a tool for moral persuasion’.

And this fits with what Nietzsche argues in Beyond Good And Evil (reviewed by me here) that philosophers learn not to search for the truth, but to argue more convincingly for their own views.

The idea that we evolved to argue and persuade, sometimes at the expense of the truth, may seem to offer a pessimistic view of human reasoning. But there may also be a very definite benefit to our argumentative minds – one that has proved essential to our species’ success. This means that when people get together to debate and argue against each other, they can counterbalance the biased reasoning that each individual brings to the table.

Studies show that “a group’s performance bears little relation to the average or maximum intelligence of the individuals in the group. Instead, collective intelligence is determined by the way the group argues.

This is clearly argued in James Surowiecki’s book The Wisdom Of Crowds (reviewed by me here).

The tradition of debate can be seen throughout history and cultures. The art of rhetoric is venerated in ancient Greece and Confucian China and is still popular today from parliament to university debating societies.

David Pavett gives a good approach to avoiding the pitfalls of this bias saying: “that to make a case for something, the most important thing you have to do is to consider the strongest possible evidence and arguments against it. If one does not do that then one’s argument is essentially worthless – however much the piling up of evidence that is consistent with one’s conclusion may delight those who are keen to see attacks on those they oppose. A scientist or engineer who relied on this technique would not keep his or her job for long“.


The Wisdom Of Crowds by James Surowiecki

13 Aug

Surowiecki presents a nice approach to trusting “the crowd”, based on some interesting research.

His main proposition is that a group can come to a wiser decision than an individual if three conditions hold: diversity, independence and decentralization. He breaks such decisions into three categories: cognition, coordination and cooperation problems, giving examples of each.

– Diversity means that people have different experience, knowledge, and approaches and so add more little bits of wisdom to the crowd.

– Independence means that each member makes their own decision rather than being influenced by the decision of others.

– Decentralisation is talking about the idea that knowledge is spread so that those closer to a problem have the knowledge to solve it.

Then, aggregating the individual decisions gives an overall wise choice. Note that just because the group comes to a wise conclusion it doesn’t mean that the individuals within the group will. In fact, for a group to perform better it needs opposing ideas so that the consensus of the group is challenged and so strengthened or replaced.

The chapter on working in small teams covers this and other issues of influence within a group. I found this practical: e.g. the idea that when a group meets to come to a decision the first to speak will usually frame the possible solutions and thus limit possible outcomes. Also the one with most airtime is more likely to have a larger influence on the final decision.

He focuses on the financial markets and although a lot of what he said was conjecture, his conclusions backed up why around 80% of fund managers under perform the index they aim to beat.

His discussion about how capitalism is based on trust was a nice new paradigm. He looked at the example of how the Quakers did so well – they were trustworthy in pricing, quality of goods and upholding agreements. Funnily they did better than everyone else and eventually their competitors had to follow their business model, bringing us closer to the system we have today.

While Surowiecki presents a convincing case, the book is so filled with non-sequiteurs and a baseless view that the US is somehow fundamentally different to every other society it’s not always easy to take him seriously. Phrases along the lines of “it is clear that” or “you can’t disagree with the view” didn’t help.

That said, I valued what I learnt and will use the ideas in the future.

%d bloggers like this: