When Does Transparency Backfire? Putting Jeremy Bentham's Theory of General Prevention to the Experimental Test

• Date: 01 December 2019
• Author: Christoph Engel
• DOI: http://doi.org/10.1111/jels.12231
• Published date: 01 December 2019

Journal of Empirical Legal Studies

Volume 16, Issue 4, 881–908, December 2019

When Does Transparency Backfire? Putting

Jeremy Bentham’s Theory of General

Prevention to the Experimental Test

Christoph Engel*

Jeremy Bentham brought enlightenment to criminal policy. He argued that the primary

purpose of criminal sanctions should be deterring future crime. To that end he advocated

complete transparency. This article investigates Bentham’s intuition in a public goods lab

experiment by manipulating how much information on punishment experienced by

others is available to would-be offenders. Compared with the tone that Jeremy Bentham

set, the result is unexpected: when would-be offenders learn about punishment of others

at the individual level, they contribute much less to the public project. This is due to an

inevitable side effect. Information about punishment is only meaningful together with

information about the infraction.

I. Research Question

Punishment is meant to hurt. As a child of the Enlightenment, Jeremy Bentham was

abhorred by the concomitant excess of vulgar sentiment. He campaigned for a more noble

perspective. In line with his forward-looking, utilitarian thinking, the purpose of punish-

ment should not be retribution, but deterrence. The sufferings of the convict should be

turned into a service for the community. In his own words (Bentham 1830:Bk. V, Ch. III):

Hence the prevention of offenses divides itself into two branches: Particular prevention, which

applies to the delinquent himself; and general prevention, which is applicable to all the mem-

bers of the community without exception.

General prevention is effected by the denunciation of punishment, and by its application,

which, according to the common expression, serves for an example ….

*Address correspondence to Prof. Dr. Christoph Engel, Max-Planck-Institute for Research on Collective Goods,

Kurt-Schumacher-Straße 10, D 53113 Bonn, Germany; email: engel@coll.mpg.de.

This article started as a joint project with Bernd Irlenbusch. Due to other obligations, he has withdrawn from the

project. Research assistance by Karsten Lorenz and Lilia Zhurakhovska and helpful comments by Markus Englerth,

Brian Galle, Nuno Garoupa, Gregory Klass, Robert MacCoun, Andreas Nicklisch, Joshua Teitelbaum, two anony-

mous referees, and audiences at the Conference on Empirical Legal Studies and at Georgetown University Law

School are gratefully acknowledged.

This is an open access article under the terms of the Creative Commons Attribution-NonCommercial License,

which permits use, distribution and reproduction in any medium, provided the original work is properly cited and

is not used for commercial purposes.

881

General prevention ought to be the chief end of punishment, as it is its real justification. …

The punishment inflicted on the individual becomes a source of security to all. … That punish-

ment, which, considered in itself, appeared base and repugnant to all generous sentiments, is

elevated to the first rank of benefits, when it is regarded not as an act of wrath or of vengeance

against a guilty or unfortunate individual who has given way to mischievous inclinations, but as

an indispensable sacrifice to the common safety.

This shift in perspective has a clear policy implication(Bentham 1830:Bk. V, Ch. III):

The punishment suffered by the offender presents to every one an example of what he himself

will have to suffer if he is guilty of the same offense.

Others tempted to break the law are held back by learning in which ways the crimi-

nal system reacts. Full transparency is what turns the punishment inflicted on the

offender into a tool for governing society.

Bentham’s intuition would be difficult to test in the field. One cannot easily con-

trol how much information bystanders receive. It would be even less possible to make

sure that the amount of information is the only difference between two groups. This is

why a lab experiment is helpful, despite the fact that what one studies is only analogous

to what one wants to understand. A lab experiment gives the researcher full control over

the institutional setting. One can unambiguously see whether and to which degree partic-

ipants react to the manipulated amount of information.

Experiments must be tuned to the research question. If the research aims at testing

a claim about mental process, the inputs of this process should be manipulated. Yet ulti-

mately the law is not interested in mental process per se. It wants to know in which ways

widespread behavioral effects matter for the choice between alternative institutions. It is

therefore a legitimate approach in experimental law and economics to test the presence

or absence of an institutional intervention. This article is written in this tradition.

Were one interested in mental process, one might want to separately manipulate

feedback about the choices made by other participants, and feedback about the punish-

ment these choices attract. Yet just giving the population more fine-grained information

about the type and degree of crime would clearly be counterproductive. There is no plau-

sible reason why this intervention would reduce the incidence of crime. By contrast, in

line with Jeremy Bentham’s thinking, informing them about individual reactions of the

criminal system may strengthen deterrence. Yet information about individual punishment

is pointless if it is not accompanied by information about the individual crime. It is this

inevitable side effect that is the topic of the present article.

One may also position the present approach on the continuum between a lab

experiment that isolates some mental process and a randomized control trial in the field.

In a randomized control trial it is taken for granted that the intervention is more aggre-

gated. One accepts the loss in precision in the interest of greater external validity. In the

same spirit, a lab experiment testing a meaningful institutional intervention trades a

slight loss in identification (one only sees the effect of the institution, not individual ele-

ments of which the institution is composed) for a gain in external validity (the isolated

elements of the institution would not be meaningful from a policy perspective).

882 Engel