On Stalin, Child Abuse, and Crime

Freakonomics makes the case that good parenting doesn’t necessarily produce good children. But what’s the effect of bad parenting — especially child abuse?

Martin Amis offered some evidence on that subject in a talk at the New Yorker Festival this weekend, in the form of a “charming anecdote about Stalin.”

In 1937, Stalin liquidated most of his census board for reporting an undesirably low population for the Soviet Union. Afterwards, he visited his mother, whom he hadn’t seen in years.

“What are you now?” she asked.

“Remember the Tsar?” he asked. She did. “Well, I’m something like that.”

“So, Mom,” Stalin asked, “why did you let Dad beat me so much when I was young?”

She shrugged. “We must have done the right thing — look how far you’ve come!”

To the extent that Stalinist Russia was a criminal enterprise, she was probably right. At least one study, by the economists Janet Currie and Erdal Tekin, finds that childhood maltreatment (including both abuse and neglect) nearly doubles the chance that a given individual will commit criminal acts.

According to Currie and Tekin, one reason for this conclusion is that abused children tend to engage in crime earlier than the non-abused. Currie and Tekin explain:

Starting to engage in criminal behavior early may increase illegal human capital by raising experience in criminal activities, and decrease human capital in legitimate activities, such as schooling or being in the labor market.

Beaten young though Stalin may have been, he did manage to enroll in a theological seminary before he engaged with the labor market, so to speak.


I question the relevancy of this article. The concept that parental abuse is strongly connected to crime can hardly be construed as novel. Parental abuse is a factor in causing crime. Nothing new there. Crime leads to dictatorships such as Stalin's. I'm still waiting for the revelation - have you stopped halfway into the article?


I agree. A new study was needed to draw this correlation? I expected more from you.


Why does it always have to be a revelation? Can't it just be interesting without revelating some mysterious hidden treasure of truth?

I think it's very interesting, entertaining, and informative, especially the Stalin's Mom joke.

Miguel V.

Is the article trying to compare very unusual individuals like Stalin with the average street criminal??

In my experience, jokes involving Stalin and/or the Soviet Union are always much funnier than other jokes. Is there a freakonomic explanation to this, perhaps?


Reminds me of Roald Dahl's article on the birth of Adolf

Ian G

If we want to talk about tyrants (and criminals) it most likely goes two ways: abused children and those overly endulged probably equally go on to careers as despots (in civics or business) given the opportunity.

It is probably just the case that anyone who rises to unquestioned power will be able to exhibit their worst traits (however developed) without consequence.

I'd like to see the math behind this study, I wonder how you assign abuse a numerical value (is it really quantitative?)


The revelation comes from the why, not the what. We all know messed up kids are more likely to commit crime, but because it starts early and they develop their criminal skills is something noteworthy. Instead of just being criminals, it is a ramp up thing.

Also, I don't see how anyone can expect huge revelations everytime they read this blog when it produces about 3-4 posts a day. And this poster isn't Steven or Stephen, so you can't expect their level of expertise.

Its still interesting though.


I think there's another relationship in here... The violent parents that beat their kids, give them also their genes.... So it's more likely that if you are a violent person, your kids will be too... Sounds logic? It's not only the "nurturing" but also the genetics... The idea of that just being violent with a kid will make him violent it's too simple.

Also, there should be a link between too much testosterone as a male and tendency to violence (guys with too much testosterone may look more attractive and more successful inseminating females, but they usually are not good parents)... The reduction in general of world's violence (spanning several centuries), also should have a direct relationship with males that are becoming better parents and have more "femal-ish" physical features... What do you think?


And yet, a large enough number of abused children grow up to be such nurturing parents that the old "cycle of abuse" myth about an abused child growing up to become abused has been thoroughly trounced on many occaisions.

Yes, abuse increases the odds that someone may abuse someone else. Yes, abuse increases the odds that a victim of abuse will commit crimes.

Smoking also increases the odds of lung cancer. It doesn't guarantee it.

Really, I think I have to hold with the others on this. This article has no real value here and only perpeturates a stereotype.


I love this blog for the interesting articles... but doesn't it seem to attract blow-hards like moths to a light?

Dr. Troy Camplin

My question is this: how are "good" parents defined? As those that don't abuse? How, then, are we defining "abuse"? Are we including all forms of spanking? What about other forms of discipline? If we include parents who do not discipline their children as good parents, can we really be surprised that we don't see a correlation between "good parenting" and good children? Perhaps we should rather look at the good outcomes and ask what kinds of parents those people had and see if we can see any commonalities. We should then, perhaps, label those kinds of parents who typically turn out good children as good parents.

Michelle Wetzler


Great point about inherited personality. Abusers are criminals so we shouldn't be suprised that they produce more criminals.

Personality inheritance is discussed at length in the book "Born That Way: Genes, Behavior, Personality". I highly recommend it!

I know Freakonomics discussed the effects (or lack of) from reading to one's children or having books in the house, but I don't remember if it was Freakonomics or Born That Way (or both) that described how complicated the relationship between environment and personality is due to the fact that the environment (nuture) is a reflection of the parent's personality (nature). In either case, I highly recommend Born that Way if you have an interest in personality inheritence. The twin studies discussed in the book are fascinating!


Is there a Freakonomics explanation for why blog posts can be so banal?

anita chatterjee

like some , I felt there was more to come in the article...
Crime and bad parenting... not always related, joke on stalin...poor taste


I tend to think some of the posters here are overly hard on Ryan. He may not be a Stephen/Steven, but that's not tacit admission to abuse him.

Yes, child abuse is an interesting phenomenon, and repercussions of this common practice are still not entirely known. However, here's the dilemma of causality- did the abuse cause the criminal, or did the environment that allowed the abuse to occur create the criminal?

Some parents abuse their children because there is tacit approval in the other spouse, who approves or ignores the problem, or even participates. So too, society will not lay blame to parents. And in many cases, you can't blame Jeffry Dahmer's parents for raising a psychopath- some people simply change after leaving the nest.

Remember that statistics simply follow trends in the population- it is not useful at all to predict individuals. In a statistical analysis, Stalin is not in the bell curve. Instead, he is an outlier (a data point that is not part of a natural data scatter).




The perception that violence is declining in the modern world is highly subjective.

It is now contrary to Western norms to mutilate or torture animals publicly, as it used to be, and yet animals are subject to severe cruelty in factory farms (and no, I'm not a vegetarian).

There are fewer wars between states, but more wars inside states.

Modern norms stringently condemn torture, and yet one only has to read the past week's news to know that it still occurs.

These are also highly subjective observations, and of course not 'evidence.' But personally, I feel that the amount of violence in the world may not have decreased, it may just have moved further from my Western daily life.


The problem is partly a statistical one, due to a fairly low "base rate" of the problematic behavior. For example, if 1 out of 100 people overall becomes a criminal, but 10 out of 100 abused children become criminals, one can conclude that abused children are ten times as likely to become criminals, even though 90 percent of abused children don't become criminals. In fact, rather than "identifying with the agressor", many abused children are not criminals or abusive at all, and work hard at being different than their parents.

Scott W

The quote reminds me that Martin Amis' 'Koba the Dread' is worth reading. To say Stalin was a horrific person is pretty obvious, but the book manages to give you a good idea of the incredible pettiness and paranoia of the man.

Sorry, I'll come back and whine about the quality of the blog post when I've had some breakfast.



I admit that upon reaching the end of this entry I thought a browser error had occurred and I was missing several paragraphs. However, as with many other entries on this blog, the quality of your links far outweighs the pithy commentary.