1. Behavioral Genetics: Crime and Violence
At some level all behaviors are genetically determined since our genes do influence our humanness and any behavior is an expression of being human. However, in recent years a more ambitious research program has attempted to demonstrate that discrete genes are responsible for complex behaviors (violence, crime, mental illness, personality). This research program is of some importance, since it may provide insight into the causes (and therefore possible treatments) of conditions such as schizophrenia, alcoholism or drug addiction. While the field of behavioral genetics represents a rather reductionist approach to human behavior, it has been an attractive area of study and is currently one of the dominant paradigms in modern behavioral science. However, this field has had its share of embarrassing errors and it has a number of critics who question its validity.
We will study the attempt to apply the techniques of this field to criminal behavior as a case study to examine this approach. We start with a book chapter that will give you a broad introduction to thinking about criminal behavior. We then turn to two articles written 18 years apart (1984 and 2002) to compare two very different types of studies. As you read these two articles, you may want to think about a number of questions, including:
Is the “behavior” being studied a single behavior or a mixture of many behaviors (if we take violence as an example, are we guilty of lumping together the ax murderer, the guy who gets in a drunken fight, the husband who abuses a spouse, but of neglecting to include the politician who orders a slaughter of innocents?).
Similarly, are we in danger of not recognizing that a single “behavior” may have multiple causes? (If schizophrenia is really several different diseases with similar symptoms it will be very difficult to identify “the” gene for it.)
What behaviors are we interested in proving genetic, and what behaviors are we less interested in proving genetic? Does this tell us something about our society?
What are the full implications of deciding a given behavior is genetically determined?
Steven Rose dealt with this in the 1995 Nature article we read.
Rose pointed out that in an imperfect world if the cause of pain and inequality lies outside ourselves, then we should use the social sciences to understand it and the political realm to address it. But, if the origins of our shortcomings can be traced to our biology, then the fault ultimately lies in ourselves. Of course, Rose points out that the real fallacy is setting up a dichotomy which forces us to seek unitary explanations for complex and changing phenomena. Rose suggests that while proponents of behavioral genetics give lip service to complexity, their basic assumption rests on an analysis that argues that if there is violence in the cities, it is because people have “violent” or “criminal genes,” and if alcoholism is a social problem, it is because alcoholics have “alcoholic genes.”
Rose argues that this is “based on a faulty sequence of analysis whose steps include several elements, which I briefly summarize again here because I want you to keep these in mind as we do our readings.
“Reification converts a dynamic process into a static phenomenon.” For example, violence, rather than describing an action or activity between persons, becomes instead a character – aggression, that is studied separately from its social context.
When several different reified characters are lumped together to create a single category, Rose calls this agglomeration. Thus, criminal activity becomes the term used to describe murder, arson, burglary, failure to pay child support or parking tickets, drunk and disorderly conduct, or even workers participating in an illegal strike, or protestors complaining about social injustice. This wide range of activity, any one of which can result in an individual obtaining a police record, is assumed to manifest some underlying character defect caused by some genetic mechanism.
This refers to the fact that reified and agglomerated characters are assigned numerical values that are assumed to have some reality and are then applied uncritically. Do three criminal reports really constitute “high crime” tendency? Assigning quantitative values to very imprecise measures is one device to give those measures more validity than they warrant.
Statistical Normality refers to the assumption the distribution of some behavioral trait in any population must take a normal Gaussian form (bell shaped curve). Once we accept the validity of a bell shaped curve it defines a concept of normality statistically and leads to a second assumption that to lie one standard deviation above or below the mean is to be abnormal. But this may be a totally arbitrary exercise.
Rose argues that during aggression animals or people may indeed show dramatic changes in neurotransmitters or brain responses, but that it is unwise to assume causality rather than consequence or correlation. One of the major errors in neurogenetic determinism is to use single gene diseases with uniform consequences (Huntington’s Disease, Lesch-Nyhan syndrome) as a model and then extend this model to traits that are likely to be both multigenic and to show strong gene-gene and gene-environment interactions.
Some Other Thoughts
Critically examine the claim of “increased frequency”
Let us say a negative behavior is exhibited by 2% of the population and in studying allele frequencies at a specific gene you find that individuals who have this particular allele show a 3x greater frequency of exhibiting the behavior.
What does this tell you?
It tells you that in this population the frequency would be 6%. But this still means that 94% of the population does not exhibit the behavior. So, is the increased frequency a problem or not? This leads to the second “error.”
Treat people as individuals and not as members of a population.
A fundamental principle in our society is to treat people as individuals and not as members of a group. It is important that each individual be treated as fairly as possible and not as a member of some larger group. But the most fundamental conclusion from behavior genetics is often that knowledge about group averages will somehow help us make decisions about individuals.
2. Partial Bibliography
Many books have been written on this topic in recent years, both popular and technical. Following is a small sampling.
2.1. Books which promote the behavioral genetic paradigm
Behavioral Genetics (4th Edition) by R.Plomen W.H. Freeman pub (2001)
This is the standard textbook of the topic.
The Blank Slate The Modern Denial of Human Nature by Steven Pinker
Pinker, a Harvard professor of Psychology argues that all human behavior is primarily genetically fixed and tries to discredit the “blank slate” idea that behavior is mainly affected by experience. Reviews suggest that he is a convincing writer, but that he sets up a “straw-man” of the blank slate ideal that few people would still subscribe to.
Genetics and Criminal Behavior Wasserman & Wachbrott
Human Genetics for the Social Sciences (on-line)
This is essentially a full textbook on line with many chapters on classical chapters and on evolutionary psychology and behavioral genetics
Behavioral Genetics Association (A page of links from the BGA home page)
Behavioral and Brain Science Journal
2.2. Books that take a more critical approach to behavioral genetics
Behavioral Genetics: The Clash of Culture and Biology R.A. Carson and M.A. Rothstein, eds) Johns Hopkins Univ. Press (1999)
The Dependent Gene: The Fallacy of Nature vs Nurture Moore, D.S. W.H Freeman, Publ (2002)
Phenotypic Plasticity: Beyond Nature and Nurture Massimo, P. Johns Hopkins Univ. Press 2001
Not in Our Genes: Biology, Ideology and Human Nature by R.C. Lewontin, S. Rose and L.J. Kamin Pantheon books (1984)
More of an attack on biological determinism than a real analysis of behavioral genetics, this book does set out many of the basic arguments.
2.3. Books Written for a more general audience
Abraham Lincoln's DNA and Other Adventures in Genetics by P. Reilly Cold Spring Harbor Laboratory Press (2000)