Law & Psychiatry

Behavioral Genetics and

the Punishment of Crime
Paul S. Appelbaum, M.D.

B ehavioral genetics would seem

to have little impact on the daily
work of most mental health profes-
crime, however, are much more so-
phisticated and are already stirring
considerable interest in the legal lit-
interactions of the participants’ ge-
netic endowments with their envi-
ronmental circumstances. Specifical-
sionals, attorneys, judges, and others erature. ly, the researchers were interested in
involved in the criminal justice sys- A 1993 report in the respected how experiences of maltreatment
tem. Arcane research findings, often journal Science described a Dutch between the ages of three and 11
not replicated, that suggest some kindred in which several males ex- years affected the later antisocial
link between genes and the propen- hibited a syndrome of borderline propensities of participants with ei-
sity to commit criminal acts are cur- mental retardation and “abnormal ther a high or low MAOA activity.
rently of dubious relevance to a sys- behavior, including disturbed regu- Using four separate measures of
tem that struggles simply to provide lation of impulsive aggression” (3). antisocial behavior, including convic-
basic justice to a complex mix of They were found to have a complete tions for violent crime, the research
criminal defendants. Recent re- absence of activity of the enzyme team found that each measure was
search findings, however, suggest monoamine oxidase A (MAOA), significantly increased in the group
that behavioral genetics may be the which breaks down many of the that had both low MAOA activity
next frontier for the world of crimi- brain’s key neurotransmitters. Ge- and a history of severe maltreatment.
nal justice, and mental health profes- netic analysis revealed that the af- In contrast, for participants with
sionals are likely to play a critical role fected men carried a mutation on the high levels of MAOA, no significant
in helping the courts make sense of X chromosome in the gene that increase was found in any of the an-
the new data. codes for MAOA. As intriguing as tisocial measures, even when they
Of course, claims that hereditary this finding was, it seemed to be of had experienced the same level of
factors play a causal role in the gen- limited import for a general under- maltreatment. The overall impact of
esis of criminal behavior are hardly standing of genetic factors affecting this gene-environment interaction
new. Late 19th- and early 20th-cen- violence, because complete absence can be judged from the fact that the
tury theories of crime relied heavily of MAOA activity was clearly rare. 12 percent of the cohort that had
on now-discredited beliefs that of- Just under a decade later, however, both low MAOA and maltreatment
fenders who manifested hereditary Science published another paper accounted for 44 percent of the co-
degeneration, which was believed to that extended the scope of the hort’s convictions for violent crime.
be apparent by their physiognomy MAOA findings. This time the data Looked at somewhat differently, 85
and low intelligence, were responsi- came from a major longitudinal percent of the males with both risk
ble for a majority of criminal acts (1). study of a birth cohort of 1,037 chil- factors developed some form of anti-
In the 1960s and 1970s men who dren in Dunedin, New Zealand. The social behavior.
carried an extra Y chromosome—the participants had first been assessed It did not take long for the
so-called XYY syndrome—were at three years of age and had just Dunedin data to be noticed by the
thought to be at increased risk of vi- completed their ninth follow-up at legal community and for the poten-
olence, another subsequently dis- the age of 26 years (4). Because mu- tial impact of these findings on the
proven contention (2). The latest tations that lead to an absence of criminal process to become the ob-
findings on a connection between MAOA activity were unlikely to be ject of intense speculation (5,6). The
genetic predispositions and violent present, the researchers examined elegance of the Dunedin data
442 males in the group for differ- notwithstanding, it is worth under-
ences in the promoter region of the scoring that the results will, as the
Dr. Appelbaum, who is editor of this col-
umn, is A. F. Zeleznik distinguished pro-
gene, which determines how strong- authors themselves noted, require
fessor and chair in the department of psy- ly the gene is expressed (that is, how replication before being accepted as
chiatry at the University of Massachusetts much MAOA is produced). And established fact. However, whether
Medical School, 55 Lake Avenue North, rather than merely looking at the ef- or not the Dunedin findings stand up
Worcester, Massachusetts 01655 (e-mail, fect of MAOA levels on criminal be- to scrutiny, it seems likely that with
appelbap@ummhc.org). havior, the investigators explored the continued advances in behavioral ge-
PSYCHIATRIC SERVICES ♦ http://ps.psychiatryonline.org ♦ January 2005 Vol. 56 No. 1 25
netics we will at some point be able genetic variables such as low MAOA, ment of an attorney, noting that evi-
to identify genes that combine with combined with relevant environ- dence that he had not been told of
environmental stimuli to predispose mental factors like severe childhood his “genetic predisposition to addic-
persons to increased rates of crimi- maltreatment? The very nature of tion” was mitigating—although what
nality, including violence. genetic evidence will likely make it role this comment really played in
One important question for the difficult to meet traditional stan- the Court’s decision is unclear (10).
criminal justice system, therefore, dards for exculpation. One can imag- Should genetic propensities miti-
concerns the applicability of the ine that most relevant genetic data gate punishment for criminal behav-
MAOA data and similar findings for will demonstrate an increased pre- ior? The answer to that question de-
determining criminal responsibility. disposition among people with a par- pends in part on whether one be-
Anglo-American law has created cat- ticular combination of genetic and lieves that factors reducing but not
egories that excuse defendants from experiential factors to engage in vio- eliminating one’s capacity for self-
culpability when their capacity to lent and other criminal behavior but control ought to reduce moral re-
choose their behaviors is significant- will not be able to establish a defini- sponsibility for one’s behavior. Moral
ly impaired. The insanity defense tive causal link between the genetic philosophers differ on this question,
and the defense of automatism are defect and the defendant’s act. Ge- with some requiring an impairment
two well-known examples. If mental netic data will probably be inade- of normal deliberative processes or
disorders that impair the under- quate to meet criteria for an insanity at least a repudiation of the desires
standing or appreciation of wrong- defense and may not even be suffi- that the person finds difficult to con-
fulness or the ability to control be- cient to satisfy more lenient ap- trol before they would accept an ar-
havior negate culpability, why should proaches, such as the proposed de- gument for mitigation (6). In the end
genetic determinants (for example, fense of genetic determinism. Ge- utilitarian considerations are likely to
low MAOA activity) with similar netic data may even run afoul of determine how the courts deal with
consequences not have the same ef- standards for admissibility of evi- this issue. Unlike most mental ill-
fect? Indeed, one author has already dence, because guidelines such as nesses, genetic propensities for
proposed a defense of “genetic de- the Federal Rules of Evidence ex- criminality are currently not treat-
terminism” that would work in this clude testimony aimed solely at able. Hence, no clear means exists of
way (7). demonstrating that the defendant reducing the risk presented by these
This would not be the first time has a propensity to behave in a par- defendants. That leaves defense at-
that courts have had to grapple with ticular way. Successful defenses torneys with the rather unpersuasive
such claims. In the 1970s several de- based on genetic factors are not very argument that because their clients
fendants attempted to introduce evi- likely. are more likely than most people to
dence of their XYY status in an effort Even if genetic predispositions are offend—and presumably to reof-
to negate their culpability. Courts not exculpatory, they might be seen fend—as a result of their genetic en-
uniformly rejected the defense, but as mitigating and be taken into ac- dowment, they should be sentenced
the grounds for their decisions var- count at sentencing. Indeed, a con- more leniently. Until effective inter-
ied. Their rulings were based alter- victed murderer in Georgia tried to ventions emerge for any condition
natively on the conclusions that only take advantage of the potential miti- that serves as the basis for a genetic
a showing of legal insanity would ex- gating effect of genetic data soon af- claim for leniency, such arguments
cuse the defendant, that XYY syn- ter the Dutch study first document- are not likely to make much headway
drome did not constitute a requisite ed a link between the absence of in the courts.
“mental disease or defect,” that evi- MAOA and violence (9). Although Although I have focused on the
dence demonstrating a causal link the Georgia courts declined to au- use of genetic information for miti-
between XYY and violence was not thorize the prisoner to be tested for gating punishment, it is worth noting
sufficiently certain to be admitted, MAOA activity at state expense to that its impact at sentencing might
and that even if XYY predisposed a establish a basis for appealing his run in the opposite direction as well.
defendant to violence, it was not pos- death sentence—probably because Whether prosecutors can introduce
sible to show that any particular be- he offered no basis to believe that he genetic evidence of a predisposition
havior was linked conclusively to the might have the rare condition re- to criminality as an aggravating fac-
genetic abnormality (8). Taken as a ported in the Netherlands—many tor that should lead to more severe
whole the XYY decisions reflect the more such requests can be anticipat- punishment is an open question. In a
deeply inbred skepticism of the ed in the wake of the dissemination small number of states, such as
criminal law for nonculpability de- of the Dunedin data. Oddly, at pres- Texas, which require prosecutors to
fenses. Substantial hurdles must be ent no reported cases exist in which establish that defendants are likely
overcome to convince any court that genetic data of any sort have been to commit future acts of violence be-
someone who committed a pro- cited as the basis for mitigation at fore the jury can impose a death sen-
scribed act should not be punished criminal sentencing. The most wide- tence, one can imagine prosecution
for it. ly discussed case is of questionable efforts to introduce MAOA levels in
What then is the future of noncul- relevance; the Supreme Court of support of that contention. The pow-
pability defenses that are based on California overturned the disbar- er of the state to test a defendant’s
26 PSYCHIATRIC SERVICES ♦ http://ps.psychiatryonline.org ♦ January 2005 Vol. 56 No. 1
genetic profile without his or her References 5. Stone RD: The cloudy crystal ball: genet-
consent is another issue that may ics, child abuse, and the perils of predict-
1. Lombroso C: Crime: Its Causes and ing behavior. Vanderbilt Law Review
well be subject to future litigation. Remedies. Boston, Little, Brown, 1911 56:1557– 1590, 2003
We stand, in all likelihood, at the 2. Wikin HA, Mednick SA, Schulsinger F, et 6. Wasserman D: Is there value in identifying
threshold of an era in which we will al: Criminality, aggression, and intelli- individual genetic predispositions to vio-
see progressive growth in our knowl- gence among XYY and XXY men, in Bioso- lence? Journal of Law, Medicine, and Eth-
cial Bases of Criminal Behavior. Edited by ics 32:24–33, 2004
edge of the genetic bases of behav- Mednick SA, Christiansen KO. New York,
ior. Genes that alone or in combina- Gardner Press, 1977 7. Johnson M: Genetic technology and its
impact on culpability for criminal actions.
tion with environmental influences 3. Brunner HG, Nelen M, Breakefield XO, Cleveland State Law Review 46:443–470,
put persons at high risk of violence et al: Abnormal behavior associated with a 1998
and other crimes will be identified. point mutation in the structural gene for
8. Coffey MP: The genetic defense: excuse
monoamine oxidase A. Science 262:578–
Faced with that prospect, it would 580, 1993
or explanation? William and Mary Law
behoove us to think through now Review 35:353–399, 1993
how we believe our criminal justice 4. Caspi A, McClay J, Moffitt TE, et al: Role 9. Mobley v State, 455 SE 2d 61 (Ga 1995)
of genotype in the cycle of violence by
system should be responding to the maltreated children. Science 297:851– 10. Baker v State Bar of California, 781 P2d
inevitable dilemmas that will arise. ♦ 854, 2002 1344 (Calif 1989)

Submissions Invited for Multimedia Reviews Column

Multimedia Reviews is a quarterly column focusing on innovative applications of
multimedia technologies and programs in clinical, education, and research set-
tings. The column’s editor is Ian E. Alger, M.D., clinical professor of psychiatry
at New York–Presbyterian Hospital of Weill Medical College of Cornell Univer-
sity in New York City.
Traditional audiovisual programs are being joined with rapidly evolving virtu-
al-reality computer programs and with digital video technologies, which bring
leading-edge concepts and applications to education, research, and clinical prac-
tice in exciting and challenging ways. For the column, Dr. Alger welcomes re-
views of teaching, training, and therapy programs presented on film, video, au-
dio, virtual reality, and combinations of these media. Reviews should be no more
than 1,600 words and should be submitted directly to Dr. Alger.
For more information about the column or to propose a submission, please
contact Dr. Alger by e-mail at ianalger@aol.com or by mail at 500 East 77th
Street, Suite 132, New York, New York 10162.

PSYCHIATRIC SERVICES ♦ http://ps.psychiatryonline.org ♦ January 2005 Vol. 56 No. 1 27