Thursday, November 01, 2007

Nature and Nurture Reprised

I have taken some flack for allegedly being a genetic determinist. Regular reader Jack Perry has written a post on his blog, Cantanima, about his son. Jack was wondering if, as a parent, he can do anything about the way his eleven-year-old has decided not to believe in God. Was this, he wonders, all determined by genes?

Let me first lay down what the science appears to tell us. Many human traits have a heritability of roughly 50%. This includes personality, religious proclivity and political allegiance. Fifty percent means half – so about half of who you are is genetic. We know this because identical twins have identical genes and we can test them to find out how different they are from each other. We find they share about half their characteristics (actually, it is more complicated than that but works out to about 50% overall). Non-identical twins have about a 25% correlation to each other, as do siblings who are not twins and children to their parents. This is because we take half our genes from each of our parents. If you share half your genes with someone then you will have a quarter of your heritable traits in common with them.

I think most of us can live with our genes having a half share in ourselves. St Augustine identified this long ago and called the propensity we get from our genes to behave other than we would like ‘original sin’. He also realised it was inherited from our parents. Clever guy, St Augustine.

The shocking fact is not that we are half made from our genetic natures. It is rather than we can find no room for nurture. If genes are half the story, we naturally assume that the other half must be our upbringing and environment. The trouble is, we have no evidence for this at all.

Over the years, scientists have conducted loads of twin studies. They take identical twins who have been separated at birth. As adults, these twins have exactly the same amount in common with each other as when both twins have been brought up by their birth parents. In other studies, it has been found that an adopted child has nothing in common with their adopted parents but the usual 25% correlation to their birth parents, even if they have never met them. This seems to mean that parenting does not have an effect on the traits contributed to by our genes. And, if parenting has no effect, it is hard to believe other environmental factors do either. For instance, in Freakonomics we learn about work in the Chicago schools system, where places are allocated at random, shows no correlation between pupils’ performance and the school they go to, once schools reach certain minimum standards.

That leaves the 50% of ourselves, for which our genes do not appear to have responsibility, unexplained. Personally, I am quite pleased that we are left with this gap. If scientists had said we are half determined by our genes and half by our environment then there would be no room left for self determination. I would suggest that the other half is who we want to be. We do have the freedom to decide. As even Richard Dawkins admits on the last page of The Selfish Gene, we can defy our DNA. Jack may not be able to make his son more likely to return to God. Instead, it is the boy himself who will decide where he wants to go and what he wants to believe in. Again, the theologians of old were right and science has just caught up.

Click here to read the first chapter of God's Philosophers: How the Medieval World Laid the Foundations of Modern Science absolutely free.


Martinbg said...
This comment has been removed by the author.
Martinbg said...

Hi again. I said something about this the last time I commented, but my main point then was about correlation in general. But nature and nurture interest me even more than correlation, so:

The observation that a certain trait has a heritability estimate of 50% does not mean that "about half of who you are is genetic."

Heritability estimates does not directly describe indivduals, only populations. More precisely, it says how much of the variance in the population, that can be described by genetic variance. Put in another way, it says something about to what extent our different scores on the trait is connected to differences in our genes.

In reality, genes and environment (where environment means: Eeverything that is not genes) are not different influences converging on an organism: They are different parts of the same process.

What genes do, is that they affect the probability that we will interact with the environment in a certain way. (Eg, that we will transform certain nutrients from our mothers into arms and legs, or that people will treat us in a way that makes us more or less trustful.) The heritability estimate says something about how strong effect a certain set of genes for a certain trait has on a certain population.

If environmental conditions change, heritability might very well change. (And have been known to do so. To take height as an example: This trait has stronger heritability in populations where there is less malnutrition, because then, malnutrition has less effect on how tall people become.) So genes are in no way deterministic on their own, no matter how strongly they influence their organisms under certain conditions.

Genes are useless without some proper kind of environment. Environment is inorganic without the influence of genes. An actual organism is 100% of both.

PS: Of course, the moment I had posted this, I noticed that it had errors. I deleted the original comment, and repost it with some of those errors corrected. There may be more uncaught ones.

jack perry said...

Argh. Of all the my deep thoughts that someone would draw attention to...

Jack may not be able to make his son more likely to return to God.

...and that is a very hard thing to accept.

I've read something similar about language. Children may learn the rudiments of language from their parents, but their accents, dialects, and idioms they take from their peers. I'm observing that in my son, too.

That we can defy our DNA is a very interesting remark which I don't recall your making before. Very interesting!

Anonymous said...

It's my opinion that this...

Children may learn the rudiments of language from their parents, but their accents, dialects, and idioms they take from their peers. really the problem you are facing, and it need not be so.

I'd further argue that it is in large part what is skewing the results of science being able to track the effect of parenting.

In modern western society, we've done something very odd (and seemingly unnatural for the human species).

We have transformed our society from one in which culture is built to support childhood attachments towards the adults in their life, to a culture in which children are encouraged...even forced... to orient their attachment towards peers.

Attachment Theory is a field in psychology that is well established and tested and they know with a great deal of certainty that human beings in general (and children in particular) take their cues, their accents, etc. from those who they are most attached to.

Of course, when you build a society which focuses on peer orientation for children, as a competitor to adult orientation for children's attachment, the result is that they will take their cues from their peers.

I highly recommend Dr. Gordon Neufeld's book Hold On to Your Kids where he makes a very compelling case that this is what has occurred in modern Western society.

We have a hard time seeing it without help because most of us have already grown up with it as the norm.

I also recommend the Wiki article on Attachment Theory as a primer. Like most Wiki articles, it's not perfect, but it's a good starting point.

This is something that James' (who I respect tremendously) post (and most likely the research)are missing, but which is absolutely critical to understanding the influence a parent does or does not have on their child.

Bottom line is that you DO have much more influence in your son's life than might be imagined, but only if you get his attachment re-oriented away from his peers to at least some degree, and back towards the adults in his life.

The person/people who hold a child's heart and attachment, also hold the power to most greatly influence that child.

In our culture, at the age of your son...that's unfortunately almost always immature peers rather the mature adults.

James said...


I agree that heritability is only meaningful within a population and that I greatly simplified the treatment.

Diet is the big one as far as environment is concerned. As far as I am aware, it is the only environmental factor where measurable effects have been detected. Twin and adoption studies also tend to take place within cultures so the effect of living in Japan against Australia has not been measured.


I'm about to do some reading on language and will report back.

Best wishes


Andrew Criddle said...

Hi James

Although the direct influence of parental religion on the religion of their childrens may be unclear, it seems obvious that religious belief and practice is influenced by environment in the broad sense.

The major differences between different times and different places in religious belief and practice are overwhelmingly caused by local environment (in the broad sense) rather than by genetic differences.

Nate said...

I'm curious about nature versus nurture debates, mostly because they are not keeping up with science. Epigenetics has been demonstrating that nature follows nurture. In a nutshell, nurture can affect how the genes do their work. The nature versus nurture dichotomy is at least in some ways false. Unless we are hard core materialists who think that all of nurture is nature anyway, this gives us more to think about.