Saturday, July 30, 2011

Looking for Insight in All the Wrong Places: What can science really tell us about human nature?

In a review in the New York Review of Books entitled, "Fooled by Science," biologist H. Allen Orr nicely dismantles David Brooks' new book, The Social Animal: The Hidden Sources of Love, Character, and Achievement. The book treats many of the issues involving science and human beings that Brooks has written about recently in magazine and newspaper articles.

The gist of the articles (several of which we have written about here) has had generally to do with what the glandular and chemical events going on when we think, choose, and emote say about the thinking, choosing and emoting--which, Brooks seems to think, is quite a lot:
The Social Animal is an attempt to write an accessible treatment of a set of weighty topics, many of which require Brooks to stretch in a distinctly scientific direction. The book, which was excerpted earlier this year in The New Yorker, focuses on big and somewhat diffuse questions: What has science revealed about human nature? What are the sources of character? And why are some people happy and successful while others aren’t?
The lack of Brooks' own expertise in actual science is matched only by his enthusiasm for making declarations about what its findings are. Brooks talks a lot about human nature, but it is never clear exactly what he means by it. Maybe this is clarified in the book. The traditional definition, of course, is metaphysical, rather than scientific in nature, and so it is rather hard to say exactly what science could say about it. And then there's the problem that much of modern scientistic thought (in which Brooks seem to share) is anti-metaphysical.

It is, in fact, a curious thing that the less confidence modern materialist scientists have in the actual existence of a human nature, the more they claim to be able to tell us about how it is effected by the physiology of human beings.

The first problem Orr spots is the simple fact of Brooks being in over his head:
... Brooks never seems fully comfortable with all this science. He often appears ill at ease in a world of technical journals, disagreements among experts, and statistical measures of uncertainty. A working scientist knows, for example, that some findings are more secure than others, often because the former derive from studies that involved many subjects and the latter from studies that involved few. 
Brooks doesn’t seem to grasp this difference. To Brooks, science is science. It’s all equally sound and can be taken at face value. His lack of expertise also presumably accounts for his occasional reliance on popular scientific journalism. Thus we’re treated to conclusions from Malcolm Gladwell’s Outliers and Jonah Lehrer’s Proust Was a Neuroscientist, among others. Since these writers are also nonscientists, Brooks’s analysis sometimes leaves us two steps removed from the actual scientist and his facts, facts that are often accompanied in the scientific literature by caveats or exceptions. 
While Brooks concedes his lack of scientific savvy, it nonetheless leads him into several difficulties. For one thing, his arguments sometimes simply don’t make sense.
And of course not making sense is a bad problem to have when you're talking about a subject in which you're not well versed. But in addition to not making sense, it appears that Brooks has some problem simply being consistent:
Brooks also sometimes champions both of two opposing scientific views, apparently without appreciating the resulting absurdities. On several occasions, for example, he praises emergentism, the view that a whole (say, an organism) is greater than the sum of its parts. Emergentism is often taken as opposed to reductionism, the view that we can understand a whole by understanding its parts. (“Divide and conquer; the devil is in the details. Therefore, for decades we have been forced to see the world through its constituents.”) But The Social Animal veers erratically between Brooks’s endorsement of emergentism and his recitation of major accomplishments of reductionist science. Indeed the science that Brooks reports is mostly reductionist. There may not be a flat contradiction here but there is at least a serious tension and it’s one to which Brooks seems oblivious.
Brooks couches his discussion in a narrative in which two of the characters are Julia and Rob. Orr gives us a sample of the way Brooks executes his scientific analysis:
As Julia and Rob semi-embraced, they silently took in each other’s pheromones. Their cortisol levels dropped. 
Later in their relationship, Rob and Julia would taste each other’s saliva and then collect genetic information. 
When parents do achieve this attunement with their kids, then a rush of oxytocin floods through their brains. 
But the caudate nucleus and the VTA [ventral tegmental area] are also parts of something else, the reward system of the mind. They produce powerful chemicals like dopamine, which can lead to focused attention, exploratory longings, and strong, frantic desire. Norepinephrine, a chemical derived from dopamine, can stimulate feelings of exhilaration, energy, sleeplessness, and loss of appetite. Phenylethylamine is a natural amphetamine that produces feelings of sexual excitement and emotional uplift.
Doesn't a passage like that just clarify everything? What is it about chemical descriptions of romantic encounters that always makes you want to reach for an antiemetic? But Orr points to the real problem here, which is that all of this impressive-sounding scientific jargon doesn't really tell us much:
What of our view of humanity changes if, when parents achieve an “attunement with their kids,” the molecule that “floods through their brains” is schmoxytocin, not oxytocin? The salient fact is that some molecule or some part of the brain underlies various aspects of consciousness or unconsciousness. But this is hardly news. As the philosopher Jerry Fodor once quipped, it’s been clear for a while now that mental processes occur north of the neck. The rest is a sort of biological bookkeeping that, while significant to the specialist, seems to provide the popular writer only with a long list of factoids. It’s not that these facts are wrong or unconnected to the higher-level phenomena—lust, emotional uplift, or insight—that Brooks discusses. They’re just superfluous.
Why would an analysis of the activity and composition of chemicals during a particular human activity tell us any more about human nature than an analysis of the movements my fingers as I type this post and the composition of the pixels on this screen would tell us about the meaning of this post?

Finally, Orr questions what all this has to do with politics and policy, a connection Brooks makes in the book:
In the end, The Social Animal presents a lot of science and it presents a laudable goal of increasing human happiness and improving public policy. But it spends next to no time plausibly explaining how the former is supposed to lead to the latter.
Supposedly scientific explanations of human life and nature never seem to yield what they promise. In fact, more often than not, they turn out not to be explanations at all, but only descriptions of the mechanical and chemical conditions that happen to accompany other important things.

If I went to see a production of Shakespeare's Hamlet, I would discover little of the meaning of the play by studying how the producers built the set and what they used to design the costumes. I would discover nothing of lasting importance in George Eliot's Silas Marner by analyzing how the printer copied and bound the book I read it from. And I would find out nothing of any real significance about the meaning of the movie "Casablanca" by spending my time figuring out how the image was projected onto the screen.

But this same procedure is used by people like Brooks in their attempt to better understand human nature.

Go figure.


Andrew said...

When my friend Tim wanted to teach his girl friend Sue how to drive a stick he began by diagramming and explaining the transmission to her.

That's what I compare teaching a teacher about neuroscience to. And it's analogous to all this fancy talk.

It's the eternal problem of confusing cause and effect with correlation.

Singring said...

'It's the eternal problem of confusing cause and effect with correlation.'

Maybe it is. A very good point.

So then what is your evidence that human behaviour or 'nature' is due to anything other than natural processes in the brain, for example?