Asian Criminology (2014) 9:189–203
DOI 10.1007/s11417-014-9185-7
Punishment and Security
Marcel Alexander Niggli & Stefan Maeder
Received: 31 May 2013 / Accepted: 28 January 2014 / Published online: 12 February 2014
# Springer Science+Business Media Dordrecht 2014
Abstract Policymakers and the public widely believe that punishment can reduce crime and
therefore improve security. Empirical data, however, prove the preventive effect of punishment
to be very limited at best. Especially, the assumption that imposing longer sentences will
reduce crime rates seems erroneous. In our opinion, this misconception is due to a confusion of
time perspectives: Criminal law necessarily looks back to the past, as it reacts to a deed.
Security, on the other hand, means preventing dangers, and its focus must be the future. Hence,
time orientation of criminal law and the logic of protection clash. Criminal law cannot provide
the security wished for. The same mistake is repeated in the prevalent theories of punishment,
notably the relative or unified theories of punishment. Security cannot be achieved or fostered
through harsher punishment or punishment threats. On the other hand, one means of decreasing crime is increasing the number of policemen.
Keywords Criminal law . Prevention . Security . Theory of punishment . Police
Criminal law and the criminal justice system most often are linked to the concept of security.
Punishment, it is argued, is legitimate because it protects us from evil. Accordingly, criminal
policy is focused on crime control and protection of society (Queloz 2010, p. 95). The logic of
protection and defence pervades criminal law to such an extent that we find it almost
everywhere, be it with reference to the criminal justice system as a whole or with reference
to specific norms. The link between criminal law and security seems so obvious that it does not
even need to be argued or explained. However, that relationship is less clear than it might seem
or outright dubious.1
The problem stems from the fact that protection and criminal law have completely different
perspectives. Protection against attacks or offences necessarily looks to the future. Nothing can
be protected from harm or damage already done. It can only be repaired or healed. The
metaphor of protection, hence, is a prospective one and necessarily so. Criminal law, on the
other hand, necessarily has a retrospective orientation.
1
The following reflections are based on arguments developed in Niggli and Maeder (2010, 2011, 2012a, 2012b).
M. A. Niggli (*) : S. Maeder
Faculty of Law, University of Freiburg, Beauregard 11, 1700 Freiburg, Switzerland
e-mail: marcel.niggli@unifr.ch
URL: www.unifr.ch/ius/niggli
S. Maeder
e-mail: stefan.maeder@unifr.ch
190
Asian Criminology (2014) 9:189–203
Time Perspectives
Criminal law looks back to the past; it looks back to a deed. This is a logical and necessary
consequence of principles stemming already from the age of enlightenment or more directly
from the German legal scholar Paul Johann Anselm von Feuerbach (1775–1833), the great
reformer and criminalist and his principle of nulla poena sine lege (or more complete: nullum
crimen, nulla poena sine praevia lege poenali), meaning that there can be no punishment
without a prior and precise rule stating so. Feuerbach came up with these principles because, as
an intellectual disciple of Immanuel Kant (1724–1804), he was unsatisfied with the situation
that the courts made up (or at least changed) the charges in the course of the ongoing trial.
Feuerbach proposed what is nowadays universally accepted, namely that we need to define
beforehand what one is allowed to do and what not. Excluding strict liability, to blame
someone for breaking the law implies that the one to be blamed had the possibility to behave
correctly. This, in turn, presupposes that he was able to know what is expected of him, i.e. that
there was a norm saying so which served as information and criterion.
Of course, the consequences of such a position are clearer and simpler for civil law
countries and their tradition of codifying the law than for common law countries, since in
common law jurisdictions, there does not always exist a statutory definition of every crime. On
the whole, however, we might state that, by and large, Feuerbach’s principles are valid
throughout the world. Take for example ex post facto laws which are forbidden almost
everywhere, although, here again, countries belonging to the codification tradition are, generally speaking, much stricter on the subject than are common law countries. However, the still
existing differences between the different legal traditions are alleviated by the fact that the US
constitution (art. I, sect. 9, clause 3) as well as the European Convention on Human Rights (art.
7) prohibit ex post facto laws. Regulations like article 1 of the Swiss Penal Code (or the almost
identical § 1 of the German or again mostly similar art. 111-3 of the French Penal Code) can
serve as examples.2 Such regulations conform perfectly to the ideas and concepts of
Feuerbach. However, they have huge implications for the criminal justice system. We have
already hinted at the necessity to lay down precise regulations of do’s and don’ts and the
prohibition of ex post facto laws, but there is more to it.
Since penalty may be imposed only “for an act”, this means that it can only be imposed
“after” the fact, i.e. after the act has been committed. Just thinking or dreaming of crimes does
not make someone punishable. Punishment becomes possible only if and when these thoughts
find consequence in an action (the crime is attempted at least or someone else is instigated to do
so). Before that very moment, there is no possibility of punishment, no direction towards future
possible development, and no possibility of preventive punishment. As long as a potential
perpetrator does not start to act against the law, he is safe under the law and even protected by it.
Accordingly, and other than in the field of ethics, it remains indifferent why exactly someone
observes the law. That person is qualified as a law-abiding citizen regardless of motives and
convictions for that conformity. Whether it is cowardice, inability, and weakness or—contrary
to all this—respect of the law, all of these motives are treated alike by the law.
But there is another reason yet why punishment always looks back to the past: the principle
of guilt. If we leave aside strict liability, punishment normally requires a fault and this fault can
be defined with regard alternatively to the perpetrator himself or his behaviour, his deed. If we
do not focus on the perpetrator (as the national-socialist laws of the 1930s for instance did) but
on his deed, we are limited to reactions to deeds. But in re-acting to perpetrations, punishment
2
Art. 1 Swiss Penal Code (No sanction without law): A penalty or measure may be imposed only for an act
expressly sanctioned by law.
Asian Criminology (2014) 9:189–203
191
obviously comes too late. The damage is done and cannot be made undone (see also Jescheck/
Weigend 1996, p. 4). All actual victims of crime have not been protected by criminal law;
otherwise, they would not have become victims.
Hence, time orientation of criminal law and the logic of protection clash. Invoking protection
as a goal of criminal law necessarily means invoking its preventive effect. Punishment then
does not serve as protection against wrongs already done but against future evils. Nonetheless,
even if we assume the existence of preventive effects of punishment, any reaction of the
criminal justice system necessarily depends on at least one concrete already realised perpetration. Let us imagine for a moment that we could identify future murderers already when they are
still children. Would it be acceptable to lock them up before they commit any of their future
deeds? Certainly not under the title of punishment. As long as no offence has yet been realised,
no legitimacy of any possible punishment exists. But then, of course, the situation would be
totally different from the point of view of administrative law. In such a perspective, preventive
measures are completely normal and acceptable. However, by focussing on prevention, we
would look to the future and necessarily lose “punishment” as a goal of our intervention.
In the past 25 years, criminal law has increasingly come under pressure from the public.
One could be tempted to ascribe this to the shift of focus away from freedom towards security.
And of course, we acknowledge such a shift. But then, given the different outlooks and time
perspectives of security policy and criminal justice, we would not expect this to have
immediate effects on expectations towards the criminal justice system, since the link between
criminal law and security is not an obvious one. Rather, these expectations have been fostered
and cultivated by politicians all over the world, making promises of increasing security by
changing criminal law, i.e. by increasing penalties, as if an increase of penalties could directly
and immediately lower crime levels and thereby increase security.
It is this unfortunate confusion of perspectives (prospective focus of security policies;
retrospective focus of criminal law) which is at the root of the actual malaise we witness.
The expectations towards criminal law do not, however, stem only from outsider’s (especially
politicians) misunderstandings. Rather, the confusion is fostered by criminal law itself, mainly
by the theory of punishment and the theory of protected legal assets (Rechtsgutstheorie).
Theory of Punishment
Usually, we divide theories of punishment into absolute and relative theories, following the
distinction of Protagoras3 and Seneca.4 As it is obvious from the statements, the difference is
essentially one of time perspectives. Absolute theories look to the past (quia peccatum), and
relative ones look to the future (ne peccetur). The difference, hence, reminds us of the
difference between the punishment and prevention perspectives.
Absolute Theories of Punishment
Absolute theories of punishment (revenge, retribution or atonement) look back to the past, to
the criminal deed, and aim at balancing the harm done. Their legitimacy rests on the idea of
3
Plato, Protagoras (324a and b), translated by W.R.M. Lamb, Cambridge, MA 1967: No one punishes a
wrongdoer from the mere contemplation [324b] or on account of his wrongdoing, unless one takes unreasoning
vengeance like a wild beast. But he who undertakes to punish with reason does not avenge himself for the past
offence, since he cannot make what was done as though it had not come to pass; he looks rather to the future and
aims at preventing that particular person and others who see him punished from doing wrong again.
4
Seneca, De ira, Book 1/19: Nam, ut Plato ait, nemo prudens punit, quia peccatum est, sed ne peccetur.
192
Asian Criminology (2014) 9:189–203
justice, of just desert, and punishment correlates with guilt. Deliberately doing harm to a
perpetrator compensates harm done by him and thereby tries to re-establish the balance of
justice. Punishment, therefore, should neither be harsher nor lighter than the perpetrator’s guilt.
Punishment does not serve any social purpose or goal apart from the re-establishment of
equivalence. It does not aim at prevention of any kind; it is in fact “absolutus”, i.e. detached
from concrete goals, but serves the purpose of justice exclusively (Trechsel and Noll 2004, p.
16; Seelmann 2009, 21 et seq.; Riklin 2007, § 5 N 27 and 30 et seq.; Stratenwerth 2005, § 2 N
10; Bommer 2007, Vor Art. 19 N 46; Roxin 2006, § 3 N 2 and 7; Kienapfel and Höpfel 2005,
Z 2 N 3 and 5; Fuchs 2004, Chapter 2 N 2 et seq.).
Retribution cannot heal wounds or repair damages, and of course, the infringement of law
will not be remedied either. Retribution can only re-establish the claim of law for integrity; it
can reinforce the expectation of law-abiding behaviour. Such a pure retribution perspective,
detached from all social aims, is not frequent nowadays and rather delegitimised in today’s
Europe (Bommer 2007, Vor Art. 19 N 47 and N 60; in detail Trechsel and Noll 2004, 20
et seq.; Riklin 2007, § 5 N 28; Stratenwerth 2005, § 2 N 10; Roxin 2006, § 3 N 8; Fuchs 2004,
Chapter 2 N 17).
Relative Theories of Punishment
Relative theories of punishment do not focus on what has happened but rather on
what should not happen again. Essentially, they are prevention theories (Trechsel and
Noll 2004, p. 17; Riklin 2007, § 5 N 34). Punishment is linked to its social benefit. It
is not a goal in its own right but rather a means to an end (Seelmann 2009, p. 22;
Riklin 2007, § 5 N 34 f.; Roxin 2006, § 3 N 11; Fuchs 2004, Chapter 2 N 5). Its
legitimacy stems from its social benefit.
Two distinctions are usually made: (1) between a focus on the individual lawbreaker versus
a focus on all potential perpetrators (that is the difference between general and special
[sometimes also “specific”] prevention) and (2) between negative (deterring) effects on the
one hand and positive (educating) effects on the other. The theories are called negative and
positive prevention theories, respectively. The combination of these two distinctions gives us
the traditional matrix of prevention theories (Table 1).
General prevention addresses everybody and aims in its negative variant at deterrence.
Everybody is considered a potential lawbreaker who can be deterred by the threat of punishment. In its positive variant, punishment addresses everybody as law-abiding citizens who are
confirmed in their common beliefs in the law. Punishment acts as a factor of integration.
Punishment of lawbreakers stabilises the law-abiding community and makes clear to everybody that values incorporated in norms still are valid (Trechsel and Noll 2004, p. 17; Seelmann
2009, p. 22 f.; Stratenwerth 2005, § 2 N 20; Riklin 2007, § 5 N 14 et seq.; Roxin 2006, § 3 N
21 and 25 et seq.; Kienapfel and Höpfel 2005, Z 2 N 6; Fuchs 2004, Chapter 2 N 14 f. and N
20 et seq.). Special prevention theories on the other hand aim at the individual perpetrator.
Punishment serves as a means of deterring the perpetrator (in its negative variant) or as a
means of education (in its positive variant). In the latter sense, lawbreakers shall be bettered,
re-socialised and educated.
Table 1 Types of prevention
Negative
Positive
General
Deterrence
Integration
Special
Deterrence
Re-socialisation
theories
Asian Criminology (2014) 9:189–203
193
It is sufficiently known that all these theories have their flaws and most of them, if applied
in a pure form, lead to unacceptable results: An orientation towards negative general prevention (deterrence) demands punishments as draconian and terrible as possible, even for petty
offences. If deterrence is the goal, then more and harsher punishment should be more effective
and hence better. This concept, of course, is not only disproportionate (Riklin 2007, § 5 N 44:
“Justice of terror”; Roxin 2006, § 3 N 32). It also infringes on human dignity, since a
perpetrator is treated as a mere object (serving common purposes). In a pure special prevention
perspective, perpetrators with minimal risks of recidivism should not be punished (not even for
very serious crimes), e.g. national-socialist war crimes as well as other crimes with a low risk
of recidivism as homicides, although the general public might not accept such a policy
(Stratenwerth 2005, § 2 N 18; Riklin 2007, § 5 N 42; Roxin 2006, § 3 N 19). Conversely,
even the pettiest offences would be punished very harshly, if such seemed necessary to prevent
further delinquency. This, however, would contradict our actual understanding of personal
freedom (Roxin 2006, § 3 N 17).
We can leave aside positive general prevention theory in this overview for two reasons: (1)
it is most difficult to test empirically, and (2) it does claim a possible preventive effect of the
criminal justice system as a whole but not of individual norms or sanctions. Now, we shall not
doubt that the criminal justice system as a whole offers preventive effects. But, we could not
do without punishment even if such an effect did not exist, as a simple reflection shows: If
calling a rule “valid” should mean anything at all, then behaviour conform to the rule must
have different consequences than rule-breaking behaviour. If it does not matter whether we
conform to a rule or not, if there is no difference at all, then such a “rule” cannot be
distinguished from a mere wish (cognitive expectations are not distinguishable from normative
ones). Hence, consequences to rule-breaking are a necessary part of the rule’s definition itself.
The main problem with every prevention theory consists in the fact that such a theory is
necessarily detached from the principle of guilt and therefore detached from individual justice.
Since the theory does not focus on the past, on an offence already realised, but inversely on
expectations, on prospective prevention of potential future offences, it cannot be linked to a
specific offence. Rather, it must use either the perpetrator or society’s need for security as a
starting point. So, whenever a relative theory of punishment is to remain linked to justice, and
thus to a perpetrator’s guilt, such guilt can present only an upper limit to a possible punishment,
but never its cause and ground (Stratenwerth 2005, § 2 N 18; Riklin 2007, § 5 N 28 and 45).
Unified Theories of Punishment
The mentioned deficits of absolute and relative theories of punishment have led to so-called
unified theories of punishment. These theories try to unite ideas of retribution with ideas of
prevention in different variants. We can distinguish retributive unified theories of punishment
(which include absolute as well as relative positions; Roxin 2006, § 3 N 33 et seq.) from
preventive unified theories of punishment (which embrace only different relative theories;
Roxin 2006, § 3 N 37 et seq.; see also Kienapfel and Höpfel 2005, Z 2 N 8 f.).
A unification would read like this: Society shall threaten to punish, and the penalty shall be
pronounced and executed with the aim to protect society against further offences (prevention),
but all this shall be done in a way adequate to the wrongdoers’ guilt (retribution), so that
prevention is realised in a just way (Bommer 2007, Vor Art. 19 N 59, Riklin 2007, § 5 N 47).
Guilt is a necessary but not sufficient condition of punishment and merely serves as a delimiter.
The actual and final reasons for punishment consist in public interests of prevention that can
only be realised through punishment (punishment as ultima ratio; cf. Stratenwerth 2005, § 2 N
28). This is typical of that position that confuses punishment’s ground (the reason why
194
Asian Criminology (2014) 9:189–203
punishment exists, why it is necessary) with its reason (the aim punishment seeks to realise).
The result is a general confusion: Punishment as a reaction to individual wrong (retrospection,
retribution, justice) delimited by general common interests (prospective, preventive) is not
distinguishable from punishment as a means for common social aims (prospective, preventive)
delimited by guilt (retribution, retrospection, justice). In continental Europe, unified theories of
punishment are dominant in the criminal justice discourse (Roxin 2006, § 3 N 33 et seq.; for an
advocate of a unified theory in China [with all the mentioned problems] see Wang 2010). The
main problem with these theories, however, remains that even their exclusively preventionoriented variants cannot dispense with retribution and guilt (at least as a delimiter of penalty;
Roxin 2006, § 3 N 51 et seq.; Fuchs 2004, Chapter 2 N 23 et seq.).
Punishment, in this view, is simultaneously retrospective and prospective. It looks simultaneously backward and forward. Accordingly, the concept of guilt has a double function: it
simultaneously justifies and delimits punishment. All this, however, rests on the assumption
that punishment is credited with a preventive effect.
Is There Any Preventive Effect at all?
Prevalent criminal law theory in continental Europe understands criminal law as aiming at
prospective–preventive protection of socially important goods and values. It is construed from
a protection perspective. Doing so, however, we enter the field of empiricism, for if the
proposition that criminal law protects real-life goods (and not only metaphysical ideas) holds
true, then, of course, we should be able to measure that empirically. Alas, empiric, criminological results are rather disappointing in the sense that a preventive effect (negative as well as
positive) seems to be very limited and harsher punishment does not help our need for security
(In detail Kunz 2008, § 26 N 1 et seq.; Durlauf and Nagin 2011). However, researching the
effects of punishment remains rather difficult. Let us take a simple example.
Table 2 shows the sentencing rates for selected crimes and offences in 1984 and 2010 in
Switzerland. Now, irrespective of all the discussions about the validity of the different types of
crime statistics, what could we possibly conclude from such a table for a possible preventive
effect of criminal law? Do 7,238 convictions for theft mean that the corresponding norm in the
Swiss Penal Code (art. 139) works well and even better than 30 years ago? Do only eight
convictions for murder in 2010 (with a reduction of 27 % over time) prove the efficiency of the
relevant article 112 or do they rather show that the provision has become unnecessary?5 Of
course, we all know that statistics (of whatever kind) cannot answer the question whether a
specific behaviour should be punishable or not. No statistic at all can ever tell us whether a
norm was effective or not because this would suppose that we have data of an identical
situation minus the norm in question. Moreover, statistics cannot really explain why crime rates
go up or down because already the question whether crime has increased or decreased is tough
to answer (Kunz 2008, § 19 N 11 f.), let alone taking into account changes in prosecution
behaviour or differences in the sentencing behaviour of courts.
If we turn to the question whether punishment has a deterrent effect, we first have to
distinguish between effects of the criminal justice system as a whole (see, e.g. Suhling and
5
Fascinatingly enough, the Swiss government in fact seems to argue like this: For example, art. 278 Swiss Penal
Code (Interference with military services) shall be abolished because its practical weight is too small since there
were only 124 convictions from 1960 to 2007 (see Erläuternder Bericht zum Bundesgesetz über die Harmonisierung
der Strafrahmen im Strafgesetzbuch, im Militärstrafgesetz und im Nebenstrafrecht, 41 [www.ejpd.admin.ch/
content/dam/data/sicherheit/gesetzgebung/strafrahmenharmonisierung/vn-ber-d.pdf, April 7 2012]).
Asian Criminology (2014) 9:189–203
195
Table 2 Sentences for crimes and offences 1984/2010 in Switzerland
Sentences against adults for crimes and offences 1984 and 2010
Absolutea
1984
Per 100,000 inhabitantsb
2010
1984
2010
Theft
9,283
7,238
143.79
91.97
−36.04 %
Sexual acts with children
Rape
449
69
319
112
6.95
1.07
4.05
1.42
−41.72 %
33.15 %
Negligent injury
831
1,050
12.87
13.34
3.65 %
Simple bodily harm (assault)
724
2,593
11.21
32.95
193.79 %
Grievous bodily harm (assault)
32
132
0.50
1.68
238.38 %
Unintentional killing (negligence)
425
140
6.58
1.78
−72.98 %
Intentional killing (wilful)
41
65
0.64
0.83
30.05 %
Murder
9
8
0.14
0.10
−27.08 %
Gross violation of circulation regulations
3,253
25,605
50.39
325.34
545.68 %
a
Data can be found at: www.bfs.admin.ch/bfs/portal/de/index/the-men/19/22/lexi.Document.21939.xls, April 7, 2012
b
Permanent resident population at the end of the year. Data can be found at: www.bfs.admin.ch/bfs/portal/de/
index/themen/01/02/blank/data/01.Document.67175.xls, April 7, 2012
Greve 2009, p. 420 f.; Robinson and Darley 2004, p. 197) and the question whether specific
changes in the system can have a deterrent effect (see the overviews of von Hirsch et al. 1999
and Nagin 1998). As already mentioned, the results are disappointing. What is important, so it
seems, is primarily the risk of being sanctioned. Harshness of punishment or the type of
punishment is clearly less important in comparison to these factors (Kunz 2008, § 25 N 8
et seq.; Kleiman and Kilmer 2009, p. 14234; Robinson and Darley 2004, p. 174 and 199 f.;
Kaiser 1997, p. 81 ff. Darley 2005, p. 194 f.; Cesaroni and Bala 2008–2009 [regarding adolescents]). Above all, the psychological process of decision making seems much more complicated than usually proposed by the classical deterrence model (for the importance of individual
factors, see Piquero et al. 2011; for the importance of social norms, see Wenzel 2004). Rewards
to be gained by non-compliant behaviour are particularly important (Baker and Piquero 2010;
Ward et al. 2006; Mulder 2008). One very simple argument might serve as an example:
Excluding the aggravated death penalties we know from the Middle Ages, it is probably safe
to assume that death penalty is the harshest threat of punishment one can face nowadays.
Evidently, though, if the goals pursued are worth it, people risk their lives. Death penalty
was very common in the Middle Ages and the Early Modernity: The Constitutio Criminalis
Carolina of 1532 (Peinliche Halsgerichtsordnung Kaiser Karls V., “CCC”) threatens rapists,
robbers, insurgents, rioters, abortionists, and manslaughters with death penalty by the sword
(art. 119, 126–129, 133 and 137 CCC), arsonists with death through fire (art. 125 CCC) and
burglars with the gallows (art. 159 and 162 CCC; Radbruch 2001, 320 f.). It is not the lack of
punishment threats which is significant of the early ages, but the lack of police.6 And research
on the effect of capital punishment does not permit definitive conclusions whether death
penalty has any effect at all on crime rates (in detail Nagin et al. 2012; likewise
6
For a nice and short summary of the state of things, see Mortimer (2008, p. 216); compare to that the situation
in the sixteenth century: Mortimer (2012, p. 292) and following. Remember, that Henry Fielding, famous author
of the notorious “Tom Jones”, together with his brother, was the founder of the first police troupe of London, and
this is in 1749!
196
Asian Criminology (2014) 9:189–203
Radelet and Lacock 2008–2009; Levitt 2004, p. 175 f.). Of course, the lack of evidence for a
deterrent effect does not prove its inexistence. But then, if we take philosophy of science
seriously, every hypothesis brought forward (such as, e.g. capital punishment having a
deterrent effect) must be regarded as false if no evidence can be brought forward to prove it.
The lack of evidence does not speak in favour of any one hypothesis, but to its contrary.
Hence, if its proof fails, so does preventive theory.
Or take another example from Switzerland, well-suited because this country (with its 26
cantons, all of which handle prosecution differently) is like a small laboratory for research on
the effects of punishment. Driving under the influence of alcohol can be punished with a fine
or with (conditional or unconditional) imprisonment. The different cantons have very different
sentencing habits. Figure 1 shows recidivism of perpetrators with their first conviction between
1987 and 2003 and no prior conviction. Fines (green colour) represent only 10 % of all
reactions in the canton of Uri (UR) but more than 50 % in Schwyz (SZ), although the two
cantons are located very close to each other. The same holds true for the suspended prison term
(yellow colour) ranging from over 80 % to under 40 %. Unsuspended prison term (red colour)
is the least common sanction. As it becomes clear from the black bullets in the lower section,
specific recidivism rates are around 5 % within 3 years of the conviction. These rates are not
influenced by the type of sanction handed down.
The finding that recidivism is independent of the type of sanction holds also true if we look
at repeated recidivists. Figure 2 shows convictions for drunk driving between 1987 and 2003
in Switzerland (according to canton of sentence). Only sentences for drivers with at least one
prior specific conviction for drunk driving are shown. The sanctions are harsher in Fig. 1 and
red (unconditional imprisonment, i.e. unsuspended prison term) dominates, while green (fine)
has almost vanished. Still, the proportion of yellow (suspended prison term) varies greatly
from under 5 % in Schaffhausen (SH) to over 50 % in Geneva (GE). And yet, specific
recidivism rates 3 years after conviction are not influenced by the primary sanction at all.
Again, these findings imply that if there is a preventive effect at all, it is not dependent on
the type and harshness of sanction but rather on the apprehension risk or the general public
rejection of the behaviour. The same trends, by the way, are found with other offences.7
Different criminal sanctions seem to be mutually exchangeable to a large extent without
loss of preventive effect (Kunz 2008, § 26 N 24.). A suspended prison term prevents
perpetrators of future offences just as well as an unsuspended one or a fine. Therefore, if there
is deterrence at all, obviously, it does not depend on the type or harshness of sanctions.
This, by the way, can be corroborated by a further fact. If we look at the incarceration rates in
federal and state prisons8 (per 100’000 inhabitants) in the USA, they present themselves as
follows:
Prison inmates per 100,000 inhabitants in the USA9
Year
1880
1904
1923
1950
1972
1980
1985
1990
1995
2000
2005
2007
2010
61
69
74
118
93
139
200
292
411
478
491
506
497
7
Federal Office of Statistics, media information 19.5.2009, www.bfs.admin.ch/bfs/portal/de/index/news/
medienmitteilungen.Document.120361.pdf.
8
If counting in jail inmates as well, the US’s incarceration rate has surpassed 750 per 100,000 in 2008 (cf.
Durlauf and Nagin 2011, p. 13).
9
Source for the years up to 1972: Stuntz (2011, p. 33); for the following years: Sourcebook of criminal justice
statistics online, Table 6.23.2010: rate (per 100,000 resident population) of sentenced prisoners under jurisdiction
of state and federal correctional authorities on December 31 (see www.albany.edu/sourcebook/pdf/t6292010.pdf,
April 7, 2012).
Asian Criminology (2014) 9:189–203
197
Fig. 1 Recidivism after conviction for drunk driving (1987–2003)—different sanctions and cantons—No prior
conviction. See www.bfs.admin.ch/bfs/portal/de/index/themen/19/04/03/02/02.html, unfortunately no longer
online
For comparison, look at the numbers in Switzerland for the last 12 years:
Prison inmates in Switzerland per 100,000 inhabitants10
Year
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
82
79
71
68
71
81
83
79
76
76
79
79
78
While in Switzerland, prison population remained constant over the past decade, and in the
USA, that number increased. One would guess that a growing number of prison inmates
correlate to a diminishing number of offences, but this holds not true. Rather, criminality
started to decrease in 1991, more than 20 years after the prison population exploded. Harsher
criminal laws cannot be the cause of the decrease which otherwise would have started earlier
and stronger (Stuntz 2011, p. 36 f.; with the same conclusions Darley 2005; dissenting Levitt
2004, p. 177 et seq., nonetheless admitting: “The real puzzle in my opinion, therefore, is not
why crime fell in the 1990s, but why it did not start falling sooner.”, p. 186).
Summing it up, although we commonly link criminal law with security and protection, the
assumption that criminal law can protect specific goods seems rather at odds with our findings.
10
Source: Federal Office of Statistics, see www.bfs.admin.ch/bfs/portal/de/index/themen/19/03/05/key/
ueberblick/wichtigsten_zahlen.Document.128203.xls, April 7, 2012 (including prisoners in remand and
preventive detention).
198
Asian Criminology (2014) 9:189–203
Fig. 2 Recidivism after conviction for drunk driving (1987–2003)—different sanctions and cantons—At least
one specific prior conviction. See www.bfs.admin.ch/bfs/portal/de/index/themen/19/04/03/02/02.html,
unfortunately no longer online
But if the preventive effect of criminal law is so limited,11 why then do we create this link so
naturally?
Why the Link Between Criminal Law and Security?
The answer to that question is twofold, institutional and conceptual. On the conceptual level,
we have to distinguish the general public values we foster from the concrete, specific actual
objects which can be aggressed (mostly physical), and we distinguish Rechtsgut (legal asset) as
a general value we are holding from the Angriffsobjekt (object of attack). For example, human
life constitutes such a legal value (legal asset), but as such, it cannot be aggressed directly, only
in aggressing one or more human beings. So, we distinguish the idea of life from its
concretisations. If criminal law protects anything at all, it protects these ideas and these values
(as realised in the criminal provisions) and it does so by making sure that anyone who
11
Kaiser (1996, § 31 N 29 et seq., N 34): “keine […] Anhaltspunkte für beachtlich abschreckende Wirkungen
von Strafen ergeben, weder bezogen auf Schwere noch auf Wahrscheinlichkeit. Gesetzlicher Strafrahmen,
Sanktionsart, Sanktionswahl und richterliche Strafhöhenbemessung haben danach relativ geringes Gewicht für
die Befolgung von Gesetzen, wenn überhaupt” ([…]no evidence for a significant deterrent effect of punishment,
neither in regard to severity nor certainty. Statutory range of punishment, type of sanction, choice of sanction and
judicial sentencing procedure have little influence on legal compliance, if any). See also Durlauf and Nagin
(2011) and Gottschalk (2011, p. 123 et seq.).
Asian Criminology (2014) 9:189–203
199
disrespects that idea or value shall face a common public reaction (which normally would be
punishment). In this sense, punishment states “we do not back off”, and “we hold strong to our
values and convictions”. Criminal law, then, stabilises (like law in general) specific expectations (contra-factually) as Luhmann would say (1987, p. 40 et seq.). However, as explained
before, (criminal) law cannot protect concrete objects from attacks. In the realm of real
physical life, it is not norms and laws that protect concrete, specific objects against an attack,
but people or technical devices. Objects and people are protected by the police (or other
security forces), not by laws.
It is quite easy to grasp that: If I want to prevent someone from entering a door, I will not
primarily take refuge to a norm of prohibition, but simply lock it. This is obvious in cases
where there is no consciousness to be addressed, no human actor: If I want to prevent things
from falling down, I do not prohibit them to do so, but secure them. Of course, norms can help
to convince or to explain my provisions. A sign over the door saying “No entrance” would
help to explain why it is locked. But if it is important to me that no one enters it, I would lock it
up.
Institutionally, the link between criminal law and security consists in the police. It is the
police who are responsible for the implementation of criminal norms. It is the police who are
enacting and executing them and the police whom we call to protect us if we feel threatened or
aggressed. Because the police are the criminal justice system’s most visible part, it frequently is
identified with it. And, of course, with regard to police law we find realised all the expectations
that we found incompatible with criminal law. Police law is part of administrative law, and as
such, there is no problem with its prospective orientation. Police law can pursue prevention
and hazard control, and it should do so and frequently achieves this goal, too. Nonetheless,
police law and criminal law are fundamentally different and the link between them is basically
institutional only.
Criminal Law Concepts Fostering Expectations of Security
It would not be fair, though, to trace the link between criminal law and security to exogenous
factors alone. Developments in nineteenth century criminal law theory have contributed at
least as much. In continental criminal law especially, Franz Von Liszt (1851–1919) contributed
to the confusion or mingling of retrospective criminal law and prospective police law. The
cascading goals of criminal law, developed in his Marburg programme, are arguably his bestknown argument (Von Liszt 2002 [1882/1883], p. 42):
1. Rehabilitation (reformation) of offenders who can be rehabilitated (reformed)
2. Deterrence of offenders who do not need to be rehabilitated (reformed)
3. Incapacitation of offenders who cannot be rehabilitated (reformed)
Such a programme incorporates police duties (understood as defence against dangers and
risks) into criminal law. But this, of course, contradicts classical ideas of punishment and
retribution, as proposed by Feuerbach. In common law countries, developments are surprisingly similar. Almost the same year Oliver Wendell Holmes (1881, p. 46) writes: “For the most
part, the purpose of the criminal law is only to induce external conformity to rule. All law is
directed to conditions of things manifest to the senses. And whether it brings those conditions
to pass immediately by the use of force, as when it protects a house from a mob by soldiers, or
appropriates private property to public use, or hangs a man in pursuance of a judicial sentence,
or whether it brings them about mediately through men’s fears, its object is equally an external
200
Asian Criminology (2014) 9:189–203
result”. Criminal law’s aim shifts towards the protection against risks that could be caused by
the perpetrator. This does not conform to classical concepts of punishment or retribution as is
obvious in Von Liszt’s third postulate: incapacitation of offenders who cannot be reformed (see
for the “Schulenstreit” [clash of schools]: Dubber 2005). The focus is on future events
exclusively. The actual offence becomes a symptom, an indicator of the risks that emanate
from the offender. Criminal justice reacts to that risk (the offender’s potential future dangerousness); his actual offence merely triggers the governmental reaction (to his dangerousness).
Blocking a road due to a high probability of an avalanche and incarcerating someone due to
the risk of his committing an offence are structurally the same. We simply do not need the
concept of guilt to justify our measures. Very clearly Holmes (1881, p. 46) writes: “Public
policy sacrifices the individual to the general good”. So, it seems quite obvious that criminal
law itself has set important causes to mix retrospective and prospective orientations,
unfortunately.
How Can Security Be Achieved or Fostered?
If we ask how, then, security can be achieved or at least be fostered, the answer does not and
cannot refer to criminal law, punishment or punishment threats. One answer12 is through the
police, through policemen.
Let us have another look at the USA: In the 1990s, New York City witnessed the most
important decrease of criminality of all major cities in the USA. In the same decade, prison
population in the State of New York showed a growth rate below average and a further
decrease, later on. What had happened? In the year of 1990, New York City counted 367
policemen per 100,000 inhabitants. This rate was increased to 508 policemen per 100,000
inhabitants in 1997, the biggest increase of all major cities in the USA (Stuntz 2011, p. 287
et seq.). States that, like New York, increased the number of policemen above average and in
which prison population’s increase rates were below average saw an average drop in violent
crimes of 31 %. The respective decrease in other states was at 2 %. So, more policemen in the
streets seem to correlate with decreases in criminality and prison populations (Stuntz 2011, p.
288; see also Kleiman and Kilmer 2009, p. 14230; Robinson and Darley 2004, 202 et seq.;
Levitt 2004, p. 176 f.; Durlauf and Nagin 2011, p. 26 with additional references).
The reason for this seems to be the interplay between certainty and severity of punishment
in producing deterrence. The empirical support for the deterrent effect of certainty is far
stronger than for severity (Durlauf and Nagin 2011, p. 16 et seq.), and with more police on
the streets, the perceived risk of apprehension or the perceived certainty of punishment
increases (cf. Durlauf and Nagin 2011, pp. 31–33). That way, would-be offenders seem to
be deterred by the perceived high risk of the intended crime (Durlauf and Nagin 2011, p. 34).
Of course, the number of policemen is only one possible explanation for the crime drop in
the USA and a partial one at that. Other explanations, as the increased prison population, have
already been discussed above. Still others refer to the strong economy of the 1990s, changing
demographics, gun control laws, laws allowing the carrying of concealed weapons and the
increased use of capital punishment. These, however, do not appear to have influenced crime
significantly (Levitt 2004) or—more importantly—they simply do not seem to be controllable:
12
Some authors are of course skeptical of “piecemeal measures”, and rightly so (e.g. Currie 2010, 1985;
Gottschalk 2011). However, in this paper, we specifically address the links (or lack thereof) between criminal
law, police and security. Therefore, arguments that address the structure of society and its relevance for violence
and/or crime must be left out.
Asian Criminology (2014) 9:189–203
201
For most European countries, death penalty is simply not an option, and changing demographics or a strong economy (as desirable as they are) is not simple policy move a
government can decide upon and just implement, as the recent financial crisis demonstrates.
Policing strategies are still hotly debated: Levitt (2004, p. 172 f.), for example, imputes little
effect to them, although he points out the lack of hard evidence and the limited data.13 Stuntz,
on the other hand, endorses community policing (2011, p. 291 et seq.).
Eventually, Donohue and Levitt (2001) explain the crime drop with the legalisation of
abortion. But even if this is true, it can only account for a part of it (Levitt 2004, p. 181 et seq.).
Moreover, once abortion is legalised, it is no longer a policy option in preventing crime even
further.
Finally, conditions that are neither the result of crime policy changes, nor economy, nor
demographics might heavily influence crime rate (Stuntz 2011, p. 278 f.), yet such conditions
are beyond our control. Hence, with regard to what can actually be done by decision-makers,
Stuntz (2011, p. 288) concludes: Putting more police officers on city streets belongs on a very
short list of policy moves that should reduce both crime and the number of prisoners (equally
Durlauf and Nagin 2011; Sherman and Weisburd 1995).
Summing it all up, it seems safe to conclude that concrete technical provisions are much
more effective in prevention and protection of concrete goods than criminal law provisions or
punishment threats. Hence, if security is our goal, we should not tamper with criminal law or
punishment threats but concentrate on policing (quite similar: Darley 2005, p. 202 et seq.).
But then, the question remains whether we are ready to pay the price for such an
undertaking. Financial considerations sure enough seem to be the explanation for the increasing use of criminal law for security: punishment is believed to be more cost effective than
regulation and pre-emptive control of compliance. Shavell (2012), for example, argues that
there is a fundamental enforcement cost advantage of the negligence rule over regulation, since
the use of the negligence rule is only triggered by harm. However, this ignores what we know
about the preventive effects of punishment. As we have tried to show, punishment cannot
deliver the security its advocates promise. Security remains a costly commodity, and its price is
to be paid not only in the currency of money but also in that of liberty.
References
Baker, T., & Piquero, A. R. (2010). Assessing the perceived benefits—criminal offending relationship. Journal of
Criminal Justice, 38, 981–987.
Bommer, F. (2007). Vor Art. 19. In: Niggli, M.A., Wiprächtiger H. (Eds.), Strafrecht, Basler Kommentar. 2. ed.,
Basel: Helbing Lichtenhahn.
Braga, A. A. (2001). The effects of hot spots policing on crime. Annals of the American Academy of Political and
Social Science, Vol. 587, What Works in Preventing Crime? Systematic Reviews of Experimental and QuasiExperimental Research, 104-125.
Cesaroni, C., Bala, N. (2008-2009). Deterrence as a principle of youth sentencing: no effect on youth, but a
significant effect on judges. 34 Queen’s Law Journal, 447-481.
Currie, E. (1985). Confronting crime. New York: Pantheon Books.
Currie, E. (2010). On being right, but unhappy. Criminology & Public Policy, 9(1), 1–10.
Darley, J. M. (2005). On the unlikely prospect of reducing crime rates by increasing the severity of prison
sentences. Journal of Law and Policy, 13, 189–208.
13
For evidence of a deterrent effect of police patrols in crime “hot spots” cf. Sherman and Weisburd (1995) and
Braga (2001); an only modest and very short-lived deterrent effect of police raids on crack houses show
(Sherman and Rogan 1995). Problem-oriented policing approaches seem to have statistically significant but
nevertheless relatively modest positive effects on crime and disorder problems (Weisburd et al. 2010).
202
Asian Criminology (2014) 9:189–203
Donohue, J. J., III, & Levitt, S. D. (2001). The impact of legalized abortion on crime. The Quarterly Journal of
Economics, CXVI(2), 379–420.
Dubber, M. D. (2005). Theories of crime and punishment in German criminal law. Buffalo Legal Studies
Research Paper Series, No. 2005-02 (download at: http://ssrn.com/abstract=829226).
Durlauf, S. N., & Nagin, D. S. (2011). Imprisonment and crime. Can both be reduced? Criminology & Public
Policy, 10(1), 13–54.
Fuchs, H. (2004). Österreichisches Strafrecht, Allgemeiner Teil, Grundlagen und Lehre von der Straftat (6th ed.).
Wien: Springer.
Gottschalk, M. (2011). Extraordinary sentences and the proposed police surge. Criminology & Public Policy,
10(1), 123–136.
Holmes, O. W. (1881). The common law. Boston: Little.
Jescheck, H.-H., & Weigend, T. (1996). Lehrbuch des Strafrechts, Allgemeiner Teil (5th ed.). Berlin: Duncker & Humblot.
Kaiser, G. (1996). Kriminologie (3rd ed.). Heidelberg: C. F. Müller.
Kaiser, G. (1997). Kriminologie. Eine Einführung in die Grundlagen (10th ed.). Heidelberg: Müller.
Kienapfel, D., & Höpfel, F. (2005). Grundriss des Strafrechts, Allgemeiner Teil (11th ed.). Wien: Manz.
Kleiman, M., & Kilmer, B. (2009). The dynamics of deterrence. Proceedings of the National Academy of
Sciences in the United States of America, 106(34), 14230–14235.
Kunz, K.-L. (2008). Kriminologie. 5. ed., Bern et al.: Haupt.
Levitt, S. D. (2004). Understanding why crime fell in the 1990s: four factors that explain the decline and six that
do not. Journal of Economic Perspectives, 18(1), 163–190.
Luhmann, N. (1987). Rechtssoziologie (3rd ed.). Opladen: Westdeutscher Verlag.
Mortimer, I. (2008). The time traveller’s guide to medieval England: a handbook for visitors to the fourteenth
century. London: The Bodley Head.
Mortimer, I. (2012). The time traveller’s guide to Elizabethan England: a handbook for visitors to the sixteenth
century. London: The Bodley Head.
Mulder, L. B. (2008). The difference between punishments and rewards in fostering moral concerns in social
decision making. Journal of Experimental Social Psychology, 44, 1436–1443.
Nagin, D. S. (1998). Criminal deterrence research at the outset of the twenty-first century. Crime and Justice, 23, 1–42.
Nagin, D. S., Pepper, J. V. (eds.)/Committee on Deterrence and the Death Penalty/Committee on Law and
Justice/Division on Behavioral and Social Sciences and Education/National Research Council (2012),
Deterrence and the death penalty. Washington D.C.: The National Academies Press.
Niggli, M. A., & Maeder, S. (2010). Was schützt eigentlich Strafrecht (und schützt es überhaupt etwas)? In F.
Matscher, P. Pernthaler, & A. Raffeiner (Eds.), Ein Leben für Recht und Gerechtigkeit. Festschrift für Hans
R. Klecatsky zum 90. Geburtstag (pp. 479–500). Wien: Neuer Wissenschaftlicher Verlag.
Niggli, M. A., & Maeder, S. (2011). Philosophie des Schweizerischen Strafvollzugs. Eine Success-Story unter
politischem Druck. In N. Queloz, U. Luginbühl, A. Senn, & S. Magri (Eds.), Druck der Öffentlichkeit auf die
Gefängnisse: Sicherheit um jeden Preis? (pp. 173–208). Bern: Kriminalität, Justiz und Sanktionen.
Niggli, M. A., & Maeder, S. (2012a). Die funktionale Stellung der Schuld in absoluten und relativen
Straftheorien. In N. Queloz, M. A. Niggli, & C. Riedo (Eds.), Droit pénal et diversités culturelles.
Mélanges en l'honneur de José Hurtado Pozo (pp. 297–314). Zürich: Schulthess.
Niggli, M. A., Maeder, S. (2012b). Sicherheit als Ziel des Strafrechts? In: Brägger B., Steiner S., Vuille J. (eds.),
Bedrohte Sicherheit oder bedrohende Sicherheit? (pp. 3-33). Reihe Kriminologie, Vol. 29, Bern: Stämpfli.
Piquero, A. R., Paternoster, R., Pogarsky, G., & Loughran, T. (2011). Elaborating the individual difference
component in deterrence theory. Annual Review of Law and Social Science, 7, 335–360.
Queloz, N. (2010). Politique criminelle, politique sociale, politique économique et … politique politicienne. In C.
Riedo, G. Fiolka, & D. R. Gfeller (Eds.), Liber amicorum für Marcel Alexander Niggli, Von Lemuren, Igeln
und anderen strafrechtlichen Themen (pp. 85–100). Basel: Helbing Lichtenhahn.
Radbruch, G. (2001). Die Peinliche Halsgerichtsordnung Kaiser Karls V. von 1532 (Carolina), herausgegeben
und erläutert von Dr. Gustav Radbruch, Professor an der Universität Kiel. In: Gustav Radbruch
Gesamtausgabe, Bd. 11, Strafrechtsgeschichte (pp. 255-336). Heidelberg: C. F. Müller.
Radelet, M. L., Lacock, T. L. (2008-2009). Do executions lower homicide rates? The views of the leading
criminologists. 99 Journal of Criminal Law & Criminology, 489–508.
Riklin, F. (2007). Schweizerisches Strafrecht, Allgemeiner Teil I, Verbrechenslehre (3rd ed.). Zürich: Schulthess.
Robinson, P. H., & Darley, J. M. (2004). Does criminal law deter? A behavioural science investigation. Oxford
Journal of Legal Studies, 24(2), 173–205.
Roxin, C. (2006). Strafrecht Allgemeiner Teil, Bd. I. 4. ed., Munich: C. H. Beck.
Seelmann, K. (2009). Strafrecht, Allgemeiner Teil (4th ed.). Basel: Helbing Lichtenhahn.
Shavell, S. (2012). A fundamental enforcement cost advantage of the negligence rule over regulation. Harvard
Law, Economics, and Business Discussion Paper No. 731. http://ssrn.com/abstract=2144553 or doi:10.2139/
ssrn.2144553. Accessed 23 January 2014.
Asian Criminology (2014) 9:189–203
203
Sherman, L. W., & Rogan, D. P. (1995). Deterrent effects of police raids on crack houses: a randomized,
controlled experiment. Justice Quarterly, 12(4), 755–781.
Sherman, L. W., & Weisburd, D. (1995). General deterrent effects of police patrol in crime “hot spots”: a
randomized, controlled trial. Justice Quarterly, 12(4), 625–648.
Stratenwerth, G. (2005). Schweizerisches Strafrecht, Allgemeiner Teil I: Die Straftat (3rd ed.). Bern: Stämpfli.
Stuntz, W. J. (2011). The collapse of American criminal justice. Cambridge: The Belknap Press of Harvard
University Press.
Suhling, S., & Greve, W. (2009). The consequences of punishment. In M. E. Oswald, S. Bieneck, & J. HupfeldHeinemann (Eds.), Social psychology of punishment of crime (pp. 405–426). Chichester: Wiley.
Trechsel, S., & Noll, P. (2004). Schweizerisches Strafrecht, Allgemeiner Teil I (6th ed.). Zürich: Schulthess.
Von Hirsch, A., Bottoms, A. E., Burney, E., & Wikström, P.-O. (1999). Criminal deterrence and sentence
severity. Oxford: Hart.
Von Liszt, F. (2002 [1882/1883]). Der Zweckgedanke im Strafrecht [1882/1883]. Baden-Baden: Nomos.
Wang, S. (2010). Rethinking the purpose of criminal punishment. Peking University Journal of Legal Studies, 2,
82–89.
Ward, D. A., Stafford, M. C., & Gray, L. N. (2006). Rational choice, deterrence, and theoretical integration.
Journal of Applied Social Psychology, 36, 571–585.
Weisburd, D., et al. (2010). Is problem-oriented policing effective in reducing crime and disorder? Findings from
a Campbell systematic review. Criminology & Public Policy, 9(1), 139–172.
Wenzel, M. (2004). The social side of sanctions: personal and social norms as moderators of deterrence. Law and
Human Behavior, 28(5), 547–567.