I'm safely back from my honeymoon, and I was catching up on the Scientific American articles when I found one that quite disturbed me. I don't usually use this blog as a forum for thoughts about things that aren't bacteria, but this is something I found important, particularly as I've spent most of the holiday reading Mary Midgely books.

The article is by Michael Shermer, and you can read it here. It's about human deception and deception-deception (the process by which we deceive ourselves into believing our deceptions). Called "the lies we tell ourselves" I'll do a quick summary here:

It starts off the main argument with the quote: "A selfish gene model of evolution dictates that we should maximise our reproductive success through cunning and deceit". It then points out that due to game theory we are aware that everyone else is also using cunning and deceit, which means the best way to go is to "feign transparency and honesty and lure them into complacency before you defect and grab the spoils". He finishes off with the idea that this is where morality comes from: "It is not enough to fake being a good person ... you actually have to be a good person by believing it yourself and acting accordingly."

So that is how human behaviour works. If you're a cunning, sneaky, nasty person it's because that's how your genes tell you to be. If you're a good, honest and truthful person it's because you've successfully managed to buy your own con.

Is this way of thinking justified?


Starting at the beginning then with that wonderful "selfish-gene model of evolution". The "selfish gene" was a metaphor used by Dawkins to explain gene-based evolution. Genes are not literally selfish any more than rocks are. And selfish was just one word, "opportunistic" might have been a better one, because genes don't work alone. Many of them need other genes, or entire gene clusters in order to function. They need proteins, and the study of protein evolution and epigenetics is an exciting subject in its own right. There's been some interesting work as well into lipid evolution and how the composition of the cell membrane when cells divide can determine their fate. No gene is an island.

And even if 'selfish' is a useful metaphor to explain genetic behavour, why on earth is it a sensible idea to abstract that up to human behaviour? Sure our genes help to determine our behavour, but so do our proteins, our neurons, our cells, our social surroundings and a whole host of other factors. Individual cells in the human body are not selfish, they are in fact highly cooperative and communist. Each cell must obey orders exactly, and if it doesn't it must commit suicide instantly. There are some cells that break away into an individualist life of freedom but these are cancer cells.

Why must the selfish-gene model predict human behavour, why not the communist-cell one?

In fact, why not go further down? Why not look at the way atoms, or quarks behave, and then say that humans must behave like that!

Two glances around in any human society will tell you that humans are manifestly not selfish individuals all waiting for a change to "defect and grab the spoils". Human society doesn't work like that. If you break down society, people don't just scatter to the selfish winds, they form new little societies to survive within. Look at the internet - a great anarchic gathering of people from all societies, with no rules thrust upon them, and what are the most popular sites (disregarding pornography)? Social networks, social forums and online communities. People like being social, they like being with others. Sure they exhibit selfish behaviour within those societies, but they also show behaviour which is loving, altruistic, angry, excited, and a whole range of emotions that the "selfish-gene" model does not abstract too. There is no reason to arbitrarily decide that any conventionally "Good" emotion is a deception-deception.

Human societies evolve by human cooperation. By the sharing of knowledge and resources, by the protecting of those more vulnerable, and the slow and shaky development of general morals. These morals are decisions made by the society (or occasionally by the one tyrant in charge of the society, but nothing is perfect) about what behaviours are acceptable. Looking at society this way isn't it just as justifiable that cooperation and sharing are the "natural" human behaviours? That people who cheat are somehow deceiving themselves into believing that they don't need society, and have deceived themselves so well that they believe it?

These aren't identical to the way our genes behave because people are not genes. Behaviours are emergent properties of humanity, not dictated attributes of our component parts. People are largely made up of water, yet no-one suggests that lying down and sort of sloshing around is natural human behaviour.

If you want to study human morality, you really need to start asking philosophers. That's what they're there for. Historians, anthropologists, even literature students and theologians are equipped with the understanding and the tools used to discuss human society, emotions and behaviour. This is an area that scientists can find interesting, and even contribute too, but in studies of behaviour and morality science is simply not the major player.

I'm sure there are great ways to build a secular civil society. But basing your foundations on the unjustified abstraction of a dodgy metaphor is not a good way to go about it.