Be mindful of results, not the method

David King warns that we should modify society, not childrens' brains. This is a response to a recent Radio 4 documentary on "the criminal mind", which discussed recent evidence for biological underpinnings of some forms of antisocial behaviour and the possibility of reducing it using vitamins, drugs or early interventions. Dr King quite rightly points out that the image given by the program tends to oversimplify things and promote a reductionistic view of the causes of crime. But he also appears to contend that complex social problems cannot be solved through biological interventions. In this he is likely wrong.

Contraceptives are a good example of how a limited technological
solution can have profound and generally positive social effects. While
unwanted children, overpopulation, spread of STDs and women being unable to control their reproduction are largely social
problems with causes outside medicine, contraceptives ameliorate these
problems significantly – and better than non-technological solutions such
as teaching fidelity and abstinence. Thanks to this, other aspects of these problems can now be addressed with more ease.

President Thabo Mbeki of South Africa
regarded AIDS as more linked to poverty than the virus: poverty
contributes to bad nutrition, stress, illness, lack of social networks
and many behaviors that promote the illness. Without the poverty the
disease would be far less dangerous. Mbeki seems to have gone much further in his denialism,
but for this post lets just keep to the sensible part of the argument. The link to
social factors is true and important – but clearly there is a
biological factor that, if it could be fixed, could abolish at least
the AIDS problem if not poverty. An AIDS vaccine (or just ART) would also be
immensely useful not just in fixing the illness but also in relieving
the significant economic and social burdens imposed by the illness on society.
Sometimes dealing with a partial cause can make the whole tangle easier
to deal with.

The causes of antisocial behavior are complex, involving everything from genetic factors, impoverished upbringing, learning in a social environment, group psychology, economy, formal and informal institutions and individual decisions. There are probably no "real" and final causes of crime in the same
sense as there are no "real" causes of this week's weather – there are
some important factors such as season, but it is also influenced by a
myriad other things. Crime is by definition social (otherwise it could not be antisocial) and involves very complex feedback between different factors.  But it seems absurd to not recognize that some of these factors are biological and amendable to influence. That they do not address the whole problem is irrelevant; purely social interventions can similarly be accused of not meeting the biological side: removing the economic or social constraints on a person does not help his impulse control.

Dr King points out:

The programme also revealed that simply giving offenders vitamins and
other dietary supplements cut their rate of offending by 26%. The key
point is that we need to make sure that all children get the adequate
diet, not to target "high-risk individuals". And we need to do that
because it is a child's right, not because we are trying to reduce
crime.

Again, true and laudable – assuming that vitamins will benefit everyone. This also seems to imply that we should aim at enhancing everybody's cognition rather than focus at "high-risk" individuals. It could also be that only some are benefited, in which case it could still be a good thing but perhaps not cost effective. But that is an empirical matter that needs to be studied. This is true for most interventions, whether biological, psychological or social – there are plenty of examples of inefficient social interventions, and some forms of interventions currently done for individuals at risk might actually have good effects far outside that group (e.g. working memory training for ADHD).

Dr King also warns against how social biases could make medical interventions biased and dysfunctional. But clearly that is true for many social interventions too: Mbeki's social explanation model biased him against medical explanations and resulted in enormous damage. Interventions that declare certain people to be abormal can harm by stigmatizing them, but interventions refusing to look at individual differences can harm by not giving them what they individually need. There are far too many theory-driven attempts to fix the ills of society and far too few data-driven (not to mention the lack of political decisions based on real data).

Indeed, we should not jump to embrace seductive but simplistic solutions. But equally, we should not discount the power of imperfect biological solutions.

Further reading:

Daniel Sarewitz &
Richard Nelson, Three Rules for Technological fixes, Nature 456 871-872 (18 December 2008) (freely accessible article on the main points of the essay)

  • Facebook
  • Twitter
  • Reddit

Comments are closed.

Authors

Subscribe Via Email

Affiliations