Bryan Caplan  

Dude, Where's My Theory of Everything?

PRINT
Collective Bargaining "Rights"... Mubarak, Freddie, and Fannie...
Almost all traits run in families.  But why?  People have literally debated the question for thousands of years.  Is the cause nature/heredity/genes, nurture/upbringing/parenting, or some mixture of the two? 

Until a few decades ago, the debaters basically just chased their own tails.  And then... enlightenment happened.  Social scientists finally discovered a Rosetta Stone to disentangle nature from nurture.  Or to be precise, they discovered two Rosetta Stones.  The first was the twin study: comparing identical to fraternal twins.  The second was the adoption study: comparing adoptees to their adopted families - and occasionally their biological families as well.

Since then, researchers have used these Rosetta Stones to decipher a massive list of mysteries.  As I recount in my forthcoming book, twin and adoption researchers studied human health, intelligence, happiness, success, character, values, appreciation, and more.  Their answers are beyond surprising.  With a few important exceptions, they learned that nature handily wins its ancient cage match with nurture, especially in the long-run.  Traits run in families primarily due to heredity, not upbringing.  The mighty effects that people ascribe to parenting are largely imaginary.

Faced with these achievements, you'd expect almost any social scientist to be impressed, even awed.  But not Tyler Cowen.  His reaction, instead, is to complain that twin and adoption methods don't contribute more.*  Who cares if you've solved the ancient nature/nurture debate?   He wants a Theory of Everything.

I submit that this is both unreasonable and ungrateful.  Behavioral geneticists don't have a Theory of Everything.  No one does.  But behavioral geneticists have answered many important, age-old questions.  How many other sub-disciplines in social science can say the same?

Now I'll reply to Tyler point-by-point.  Tyler's in blockquotes, I'm not.

"Culture" and "genes" are two major factors determining individual outcomes, toss in parenting, and if you wish call parenting and culture two parts of "environment."  It is obvious that culture matters a great deal, and this comes from knowledge which existed prior to rigorous behavioral genetic studies.

... "The culture word" may be overused and abused, but still the power of culture is evident.

If "culture" just means "everything besides heredity and upbringing," then Tyler's clearly right.  Identical twins raised together are hardly ever literally identical, therefore other stuff matters.  A lot.  But if you define culture more falsifiably, things get complicated very quickly.  Sure, there are traits like accent that clearly stem from humans' tendency to copy each other.  And yes, you can't be "bookish" unless your society has books.  Nevertheless, many allegedly cultural traits could easily be genetic, and we don't yet have a Rosetta Stone to disentangle the two.

Tyler continues:

If twin adoption studies seem to show that parenting does not matter much, I think:

1. Matter for what and for whom?  Parenting matters a lot for language and religion and obedience and also one's sense of "how the world works," and those factors matter to parents even if they don't always matter to researchers and economists.  The word "matters" is going to carry real weight here; in my admittedly extreme pluralist view, "doesn't affect adult income" does not translate into "does not matter."...

Sigh.  In my book, which Tyler not only read but blurbed, I cover the twin and adoption evidence not just for income, but for an entire Parental Wish List: health, intelligence, happiness, success, character, values, and appreciation.  Religion falls under "values," and the punchline is that parenting has a big but superficial effect.  Parents strongly affect what you say your religion is, but have little long-run effect on your intrinsic religiosity or observance.  I don't discuss language, but it's pretty clear how a twin or adoption study would play out: You can make your kid semi-fluent in another language with a lot of effort.

2. We already know that culture matters a great deal in shaping what kind of adults children become, but often individual families cannot much affect the broader culture a child is raised in.  It's sometimes the individual family which is impotent, not the surrounding culture as a whole.

Plausible.  I've made such arguments myself.  But twin and adoption methods are poorly designed to test such claims, and it isn't reasonable to expect them to.

3. Most parents are deep conformists.  There isn't always a lot of cross-sectional variation in adoption studies.  Even if most parenting strategies don't matter (if only because they are not varying much), if a child is raised in an Orthodox Jewish family, or in a strict American-Chinese family, or among the Amish, that probably matters, even adjusting for genes. 

The adoption studies can be showing that a) most parents don't so much shape a child's culture at the margin, or b) that environment doesn't much matter in light of the power of genes.

Twin and adoption studies measure the effects of the kinds of parenting that people in the First World frequently use.  I say this repeatedly in my book.  If you want to do social policy or weigh whether to join the Amish, it's an important limitation.  If you want to answer the kinds of questions that most parents in the First World are actually asking, it's not.  And if you want to call the vast majority of Western parents a bunch of "conformists" and claim that their parenting is all basically the same, give credit where credit is due.  On the surface, parenting styles seem to vary widely.  The only reason anyone would conclude that these diverse approaches are roughly equivalent is by reasoning backwards from their effects.  And the only reason anyone would conclude that these effects are small is twin and adoption evidence itself.

* Tyler singles out "twin adoption studies," which is normally a synonym for "separated twin studies."  But his critique applies to ordinary twin and adoption studies as well.


Comments and Sharing





COMMENTS (20 to date)
Biopolitical writes:

Adoption studies and other evidence support your view that parenting style has little effect on offspring outcomes. And you argue that this is an important reason for parents to be more relaxed about child rearing. These are your really interesting arguments.

You also embrace the idea that differences among people are mainly due to genetic differences. In this case, it's already known that adoption and twin studies are of little help because of confounding factors. But this is just a boring topic that does not affect your argument about parenting and having more kids. Your argument could just be that parents can do little to change their children's genes or environments in order to mold their behavior in predictable directions, that they should give up genetically or environmentally engineering their kids, that they should instead relax and enjoy family life.

Your reliance on genetic as opposed to environmental influences is irrelevant for your advocacy of having more children, but it is relevant for your desire of breeding a clone of yourself. Your clone could end up resembling you much less than you think.

RAD writes:

Bryan, I have trouble with your blurb about language:

I don't discuss language, but it's pretty clear how a twin or adoption study would play out: You can make your kid semi-fluent in another language with a lot of effort
It is easy to make a child fluent in a specific language: move to a region that speaks that language before the child reaches a certain age. The hard part is fluency in a language that is not native to the region.

The language case is one that Pinker uses to demonstrate the truthfulness of Harris' claims. Children acquire language through their peers rather than their parents or teachers. Culture obviously has a role but, in my opinion, it somehow works through the peer group. For language, there is some magic formula that determines whether children develop a Pidgin language or adopt the cultural language. Language may be a more innate than personality but I think the peer group should be assumed to be responsible for all malleable traits (i.e. non-genetic traits) unless there is strong evidence of a different mechanism.

What confuses me most is why you are not shouting "Peers Peers Peers!!!!" from the rooftops.

OneEyedMan writes:

I thought the hardest line churches were the ones with the best retention and participation rates. Is that really all genetic?

Tracy W writes:

OneEyedMan - I'm confused. As far as I can tell, Caplan is entirely open to the idea that which church you attend is driven predominantly by cultural factors, and, possibly, free-will. Why would you think that attending church would be entirely a matter of genetics?

Tom West writes:

Oddly enough, I find the idea that parents have little effect on their child's later success, etc. to be a profound reason *not* to have children.

The reality is that a child's failure is going to break a parent's heart and massively decrease happiness.

Now to find out that parents have very little influence at all on the child's outcome? My recollection is that one of the most effective paths to unhappiness is to have a great deal invested in an outcome upon which you have no control.

Finch writes:

> Now to find out that parents have very little
> influence at all on the child's outcome?

Huh? Parents have a large influence on the outcome. They just have that influence through heredity, not parenting.

Introspection will tell you a fair amount about your children's likely outcomes.

Finch writes:

> The reality is that a child's failure is going
> to break a parent's heart and massively
> decrease happiness.

I think there's potentially something to this. I've wondered before, in this forum, whether there are more important sources of parental unhappiness than those Bryan identifies.

Nagging my oldest kid to finish his homework doesn't bring me unhappiness. Seeing him on a ventilator in the pediatric ICU a few years back (everything's fine now, thanks), that made me unhappy. How dominated are parental happiness statistics by severe adverse events?

I guess I'm not convinced that being a laid-back parent is so much better for your happiness than being an uptight one.

Dan Weber writes:

Oddly enough, I find the idea that parents have little effect on their child's later success, etc. to be a profound reason *not* to have children.

Bryan's thesis isn't that parents have no effect. It's that parents' behavior has no effect.

Chinese parent's child writes:

[Comment removed for rudeness. Email the webmaster@econlib.org to request restoring your comment privileges. A valid email address is required to post comments on EconLog.--Econlib Ed.]

Dan Weber writes:

I can't believe I've never thought to ask this before, but, how does home-schooling fit in with the "laid-back parents" model?

Matt writes:

I'm only superficially knowledgeable about this area (if that) but I'm curious if the twin and adoption studies have taken into account more recent discoveries such as the short/long allele for serotonin? Or another study (sorry, no cite) that showed that the "mothering style" of a mouse affected the expression of certain traits and genes in those young mice she raised (even those not related to her).

There seems to be a growing, nascent body of evidence suggesting a very direct link between gene expression and environment in at least some very limited set of traits/genes.

And Bryan, you fail to address the final 'graph quoted by TC (regarding smoking rates) in his original post. Could you respond to the issue, and offer a possible explanation for the change in smoking rates among age cohorts within, or in spite of the framework of heredity dominance?

Tom West writes:

They just have that influence through heredity, not parenting.

It's that parents' behavior has no effect.

I should have been more clear. It's control (or the illusion of control) that's important to happiness. Parent's genes may be important, but who is going to be happy to wait 18 years to see if the genes you gave your children panned out :-)?

I'm not prepared to make a stand as to whether Bryan is correct or not. But I do think his supposition makes being a parent *less* attractive.

Anton Tykhyy writes:

Bryan, have you noticed that you are dividing by zero here? You say almost in one breath that "Traits run in families primarily due to heredity" (relying on twin&adoption studies for this judgment) and that twin&adoption studies aren't good at distinguishing the effects of culture from the effects of heredity, to say nothing of pre-natal influences.

Finch writes:

> I should have been more clear. It's control (or
> the illusion of control) that's important to
> happiness. Parent's genes may be important, but
> who is going to be happy to wait 18 years to see
> if the genes you gave your children panned out
> :-)?

I still don't understand. How is this different for behavior? You _also_ have to wait 18 years to see if your parenting behaviors panned out.

Short-term genetic and behavioral effects aside. You seem to be thinking about outcomes here, not, say, household discipline.

Tracy W writes:

@Matt: twin and adoption studies can't distinguish between direct genetic effects and genetic-effects that operate via influence on the environment (eg, a pair of identical twins are separated at birth, both of them grow up tall, so they're a bit better at basketball, so they enjoy playing it more, so they separately persuade their parents to install a hoop, so they practice more, so they get picked for the school team, so they practice a lot more, and then wind up in adulthood as great basketball players).

So the "short/long allele for serotonin" gets bundled up into those results, along with everything else.

As for the mice study - interesting, but not necessarily applicable to humans. The evolutionary point of learning is to be able to adapt to environments that your genes can't know about ahead of time. Humans have evolved living in large social groups, so why wouldn't evolution take advantage of that to collect information from multiple people about what the world is like? It's long been known that in a risky situation a forecast that averages input from a lot of different noisy sources will probably be more accurate than the best one of the noisy sources (see Armstrong: Long-Range Forecasting).

Tom West writes:

You _also_ have to wait 18 years to see if your parenting behaviors panned out.

Yes, but in the meanwhile you're actually *doing* something that you think will have a direct bearing on the outcome...

It's the difference between someone being told "At the end of the year, you have a 10% chance of dying" and "if you work at it, you have a 90% chance of beating this thing". They mean the same thing (in fact the second is worse), but I'd guarantee most people would be much happier being told the second... Knowing that you are an active participant in your fate is important to human happiness.

Michael Vassar writes:

To me, parenting styles don't seem on the surface to vary widely, excluding cultural outliers like Amy Chua.

Daniel Carroll writes:

While I haven't read Bryan's book (I'll put it on my Kindle wish list), I am familiar with the adoption and twin studies independent of what he wrote. My understanding is that the conclusion of those studies is that "it's complicated", not that it is "all genetics."

Environment controls outcomes for certain traits (Type 1), genetics controls outcomes for other traits (Type 2), and genetics modified by environment for still other traits (Type 3). Introversion versus extroversion is believed to be strongly Type 2, while attitude, self image, beliefs, open mindedness, and critical reasoning abilities are mostly Type 1 or 3. Disabilities are usually attributed to Type 2.

Further there is a third category of influences - epigenetics, or the physical/hormonal environment inside the womb. Or for that matter, the physical environment in early childhood.

Other unrelated studies show that certain parenting styles are more effective than others, except that methods should be adapted to the genetic predisposition of the child. The extremes of authoritarian and permissiveness is typically not effective, while a mix of the two is. A passive child can do will with a permissive parent, while an assertive or aggressive child needs more structure. And that combined with active involvement in the child's life - absentee parenting is not effective.

Tracy W writes:

@Michael Vassar:

Parenting styles vary. Eg:
- Does Mum have a paid job
- How much time do the kids spend in day care
- How much TV can kids watch
- Do the parents use corporal punishment
- Do the parents play with their kids, or ignore them
- How many kids do the parents have? You can't use the same parenting techniques on 3 kids, or 10 kids, as you can on one (joke about parents: parents' first baby drops their dummy on the floor, dummy whisked away and sterilised in boiling water; parent's second baby drops dummy on floor, dummy wiped down with a clean tissue; third baby drops dummy on floor, dummy picked up and shoved back in again.)
- How many parents does the kid have? If Dad's dead or disappeared before the kids' birth, that's a hell of a lotta different parenting style to a Dad like Bill Crosby.
- Are the parents living together or divorced?

Tracy W writes:

@Daniel Carroll - can you name some of these studies? And do they control for the fact that normally children are raised by their genetic parents? And how do they control for peer effects, eg let's say that a child is adopted by a family where the parents later turn out to be criminals. The child grows into an adult with a criminal record. Was it because of their adoptive parents' rearing, or a genetic coincidence, or because their adoptive parents lived in a high-crime area and the child picked up the criminal behaviour from their peers, the same way that their adoptive parents did in turn? (I have a NZ accent and my parents have NZ accents, but I very likely got that through my parents' decision to raise me in NZ, like they themselves were raised, and if they had chosen to raise me in another country, I probably would have the accent of that country.)

The extremes of authoritarian and permissiveness is typically not effective, while a mix of the two is.

Or alternatively, parents with good children have a good deal of flexibility in how they can approach their kids, parents with defiant children find themselves faced with the two options of either being full-on authoritarian or giving up.

And that combined with active involvement in the child's life - absentee parenting is not effective.

Yes, I'm with you on that. Small children cannot be expected to find food for themselves. Parents matter. The question is what do parents matter for.

Comments for this entry have been closed
Return to top