<%ParlWebsiteContext.IncludeResources()%> <%ParlWebsiteContext.RenderHeader()%>

Legislative Policy and “Two-Sided” Crimes:
Some elements of a pluridimensional theory of the criminal law
(Drugs, Prostitution, etc.)

Expert’s Report
prepared for the
Special Committee of the Senate of Canada on Illegal Drugs

Alvaro P. Pires
Canada Research Chair in Bijuralism and Criminal Justice
University of Ottawa


À Sandro,

I dedicate this paper to the memory of my friend and former professor, Alessandro Baratta, of the Institut für Rechts-und-Sozialphilosophie of the Université de la Sarre, Sarrebruck. He left us, "in his usual natural way", while I was writing.




I would like to take this opportunity to thank a number of my colleagues for the suggestions they offered or the critical comments the made during the course of this research project: Line Beauchesne (Criminology, University of Ottawa), Daniel Dos Santos (Criminology, University of Ottawa), Louk Hulsman (Faculty of Law, University of Rotterdam), André Jodouin, (Faculty of Law, University of Ottawa), Dan Kaminski (Criminology and Criminal Law, Université Catholique de Louvain), Colette Parent (Criminology, University of Ottawa), Jacques Laplante (Criminology, University of Ottawa). 

I would also like to thank all the members of my team, the researchers at the CIRCEM (Interdisciplinary Research Center on Citizenship and Minorities, University of Ottawa), who played an active part in this project and who provided, in addition to continuous technical support, a host of intellectual contributions at all stages of the project, through their comments, suggestions, etc.: Richard Dubé (criminologist, doctoral student), Margarida Garcia (lawyer and criminologist, doctoral student), Gérald Pelletier (historian, researcher) and Nicole Sommers (criminologist, master’s student).




The objective of this study is to produce an interdisciplinary discussion paper that will be general in scope, on the appropriateness of using, and role of, the criminal law in public policy making.  We shall first look specifically at the problem of illegal drugs, or, more particularly, what are called “soft” drugs.  In trying to achieve our dual objective, the approach that is taken here will focus on an analytical question that we will use to propose a bridge between the criminal law and public policy.  

That question may be stated as follows: Are there distinctions that should be made in the criminal law among different types of prohibitions?[1] In other words, can we say that the prohibitions against killing, assault and theft, on the one hand, and the prohibitions relating to drugs, sexual services or homosexual activities, on the other hand, are of the same “order”?  Are all of the behavioural norms (the “charges) in the criminal law the same, or are there different types of norms that have come to be regarded as “crimes”?  We hope that the meaning of these questions will become clearer as this paper proceeds. 

This approach to the discussion may be regarded as interdisciplinary in nature, because it is based on knowledge that comes to us from the sociology of the (criminal) law, from the philosophy of the criminal law, from criminology, and from legal knowledge itself (theory of the law, literature regarding the criminal law, etc.).  We will of course also be referring to knowledge that originates in other branches of learning.  

We are going to try to show that it is possible to identify certain theoretical and empirical criteria that can contribute to public policy, and can provide us with a recommendation that looks like this: “if we have a problem, or a law, that has these particular characteristics, it is important that we pay attention to these particular things.”  Those criteria are flexible enough that they can take into account the (internal) diversity that is unique to each sub-group of prohibited behaviours identified. 

In addition, as we shall see later, the points addressed here are still exploratory in nature: despite the considerable theoretical contributions that have been made, se still do not have systematic, well-established criteria that we can use to observe criminal behaviours and identify differences among them.  That is why the title of this paper refers to elements of a pluridimensional theory (one using diverse criteria) of the criminal law (of behaviour). 




When we wish to discuss the role of the criminal law in public policy making, and more specifically the role of the criminal law in public policy relating to illegal drugs, a number of closely related questions come to mind.  Are there criteria that can be used in discussing whether or not the criminal law should be used to solve a social problem?  Is there an equal or sufficiently sound basis for all of the crimes that we create using the law?  What actions may we criminalize (prohibit by a criminal law), and what situations, on the contrary, should we decriminalize?[2]  What should we think about when we consider creating or reforming a criminal law?  What are the rational conditions that must be present in order to “criminalize” something regarded as a “problem”, and can those conducts be criminalized using different styles (for example, without allowing the use of imprisonment)?  If a behaviour is to be decriminalized, must the behaviour be accepted, or may it be one that we still disapprove morally?  

            This study is an invitation to discuss these problems having regard to the experience of the western criminal law system with the creation of laws, and to a certain part of the accumulated knowledge that seems to be relevant in assisting us to understand that legal experience.  Since the 1960s, the study of humanities and of the law has demonstrated a growing concern with these questions.  

            We know from history that all crimes do not have the same sound basis, even at the time they were created (consider, for example, crimes of witchcraft).  But we also know from historical and contemporary basis that all crimes created by legislatures are always “justified” (that is, that there are reasons that make them acceptable). 

To understand this paradox – a crime whose existence has been justified by a legislature, but which at the same time has only a dubious basis – it is worth recalling one of the theses of Vilfredo Pareto (1848-1923), a classical sociologist from the turn of the 20th century.[3]  He contended that a human being was perhaps (also) a rational being, but was above all a reasoning being, and at times an irrational being. Pareto meant by this that all human beings [TRANSLATION] “want to give the appearance of logic to conduct that is not in essence logical” (Aron, 1967, p. 422).  He then said that human beings [TRANSLATION] “have a very pronounced tendency to apply a logical veneer to their actions” and that they very often “cite some reason to justify their actions” (Pareto, Treatise, § 154, quoted by Aron, 1967, p. 412).  Although that thesis has undoubtedly contributed to Pareto’s great unpopularity, and even if we do not like the self-portrait that portrays us as a reasoning being, we can learn from this lesson and distinguish between the justification and the correctness of the justification.  When we create a law, we justify it.  It is difficult to understand how a legislature could pass a law and at the same time say that it is not justified, or that it is going to produce “will produce sadness in the Realm”!  The theoretical problem is then a matter of determining how we can distinguish among the various kinds of justifications we offer, and how we can discover whether there are – or are not – justifications for which there are at least prima facie sounder bases than for others.  

The challenge then is to try to discuss the criminal law and the soundness of the basis of crimes so that we have a guide for our legal and political decisions.  We therefore must identify criteria that enable us to better see and understand what we have done (in the past), what we are doing (in the present) and what we are preparing to do (extended present or near future).  For all those purposes, we must make an effort to be able to “take the long view”, including while we are still right in the middle of the problem. 

            That is the ambitious goal of this theoretical essay.  The thoughts presented here must also be seen in the context of a series of earlier contributions[4] to the discussion of criminal laws and the decisions of the courts.  Hart (1963) and Schur (1965), in particular, tried to identify two broad groups of crimes, one of which did not have a sufficiently sound basis.  In Hart’s view, the crimes for which there was no sound basis are “crimes that cause no harm” (such as homosexuality, sodomy, living off the avails of prostitution, etc.).  In Schur’s view, the ones that cause difficulties were the “victimless crimes” (homosexuality, abortion, drugs).  Whatever problems can be identified with these two approaches,[5] we are going to use their hypothesis that there are in fact two broad groups of radically different prohibited behaviours that are criminalized.  Unlike those approaches, we are going to propose a set of criteria that are designed, inter alia, to identify and construct those two broad groups, and also to determine what intermediate forms of prohibited behaviour that are that fall between those two extremes.


The distinction between the two broad types of prohibited behaviours 

It is often said today in the philosophy of science that in order to be able to observe the physical and social world around us, we must make distinctions between tangible and intangible things.[6]  In other words, we must identify or propose differences between this and that in order to be able to observe and talk about a particular thing.[7]  And then we must indicate – or focus our attention on – one of the two sides of the distinction we have made.  In commonplace terms, let us say that we want to observe a chair in a furnished room; we must actively distinguish it from its environment, that is, the other objects (tables, other types of chairs, etc.) that are also present.  As paradoxical as it might seem, if we do not make the “this chair / other objects” distinction, our environment does not do it for us, even where it “offers” us the opportunity to do it.  We are then not attentively observing the chair that is in the room.  

If that assertion is correct, it means that, at least in part, the type of observation we make depends on the type of distinction we propose.  There are certainly some distinctions that are – or that may be determined to be – better than others.  From the standpoint of knowledge, therefore, we might say that the quality and efficacy of the observations we make depends (in large part) on the quality of the distinctions we make.  The mathematician George Spencer Brown (1979) formulated a directive for the starting point of any observation: “draw a distinction”.[8]  We are going to follow that directive, and “draw” a distinction between two things. 

            What we are going to do is propose a distinction so that we can observe certain phenomena that occur in relation to criminal legislation.  This exercise is a good example to show how the absence of distinctions between things prevents us from seeing certain differences that nonetheless “exist”.  When stated in a more positive form, this exercise should show us how the distinction between things sometimes leads us to see aspects of reality that were not (entirely) visible before.


            The criminal law has a tendency to regard all crimes (criminally prohibited behaviours) as “crimes against Society”, “actions that endanger the prerequisites for Society to exist”, or “crimes against the common standards of a particular society’s morality”.  When we observe in this manner, we at first have the impression that there are no fundamental distinctions to be made among prohibited behaviours.  The criminal law does not stop there, however: the criminal law itself proposes some (non-fundamental) distinctions so that, for example, we can see that there are some prohibited behaviours that are more serious than others, and that they do not all relate to the same values or “goods” (for instance, there are crimes against the person, crimes against property, and so on).


What is the impression that we take away from all this?  On the one hand, that all crimes are the same (against society), and on the other hand, that the only important differences are those proposed by the criminal law.  This is like walking into a living room, with the homeowner as our guide, and that homeowner informing us immediately that all the objects in the room are “furniture that enables the living room to exist”.  And right away, he or she distinguishes the chairs from the tables, and so on.  If you stay there, you are no longer capable of observing anything else, and you will believe that it is all one harmonious unit.  What if there were a fundamental difference that divided the furniture into two broad groups – for example, two completely different styles of furniture?  That difference would escape your notice.


Let us now assume that you suggest that the homeowner make a distinction between two styles, and you say to him: “Sir, you have furniture here that belongs to two incompatible styles.”  He replies along these lines: “That’s not important, because the furniture has in common that it is in the living room to cause the living room to exist.”  There are two aspects to that answer, one of which is stranger than the other.[9]  The first and strangest aspect is that you know perfectly well that what the two styles of furniture “has in common” is that they are in the same living room.  So that does not solve the problem.  On the contrary – in your mind, part of the problem may be the very fact that these two styles of furniture are there in the same living room.  The second aspect of the answer is not illogical, but it is neither sufficient nor persuasive.  The homeowner must be able to show you that if certain pieces of furniture were removed, or arranged some other way in the house (or even in that room), this would cause the living room to cease existing as a living room; it would kill the living room.  You don’t want to kill the living room; quite the opposite – you want to make it more coherent and more pleasant.  But you don’t see how the living room could die or disappear if it were decorated differently.


In order to observe differences other than the differences proposed by the criminal law itself, we must propose and test some new distinctions – distinctions that are not those proposed by the law itself, and that do not assume from the outset that the law is a harmonious unit.  Our starting point will be a question.  Can we distinguish, in criminal laws, two radically different broad groups (or types) of prohibited behaviours that have been made crimes?  In other words, are there prohibitions in the criminal law that are not the same as the others – that are not like murder, sexual and physical assault, theft, fraud, wilful damage to someone else’s property, and so on?[10]


We will answer “yes”, and we hope to be able to show, using seven criteria,[11] that there are two fundamental broad types of prohibited behaviours in criminal laws.  As we shall see later, in order to fall into the second group, a prohibited behaviour need not necessarily meet all seven criteria.  Table 1 sets out the cleavage question and the answer we have reached (theoretical final hypothesis), and provides a few examples of crimes in each group to assist in understanding.



Table 1: Cleavage question, theoretical hypothesis and examples of both groups


Cleavage question:

Are there (at least) two different broad groups or types of prohibited behaviours in criminal laws ?


Answer (theoretical final hypothesis):


  1. YES, there are two broad groups that are radically different from each other

  2. Group 2 prohibited behaviours have many points in common with one another, but there are also important differences (which will not be addressed here)


  1. There are other groups (“grey areas”) between these two extremes (which will also not be considered here)


Examples of Group 1


Examples of Group 2

- Murder

- Sexual or physical assault

- Theft

- Fraud

- Wilful destruction of someone else’s property

- Etc.

- Illegal drugs

- Gambling

- Prostitution

- Abortion

- Homosexuality

- Sodomy

- Vagrancy and begging

- Attempted suicide

- Etc.


The answer to this question leads to another set to questions that are practical in nature, and relate to the political system, the legal system and the moral system in our society.  These questions include the following.  What do we do with these two groups of prohibited behaviours, and with each of them in particular?  Should we continue to consider all of these prohibited behaviours as crimes against society and treat them all exactly the same way?  Is it morally acceptable to preserve the present situation?  What are the alternatives, in terms of legislative policy and (criminal) law reform?


We are going to baptise these two broad types (or groups) as follows: “standard prohibited behaviours” and “two-sided (or self-referential) prohibited behaviours”.  There really are no perfect labels for them, so we must (at least provisionally) be content with the ones that seem to be most satisfactory, or least unsatisfactory.


            The label standard prohibited behaviours (group 1) comes from the first meaning of the word “standard” in ordinary usage.  That meaning (representative, analogous) is perfect for our purposes.  Standard refers to “agreed properties for manufactured goods” or “the ordinary procedure”, or “a document specifying agreed properties for manufactured goods”.[12]  This is precisely the idea we want to focus on here: standard prohibited behaviours are those which conform to certain standards for manufacturing the criminal law.  But, like any manufacturing plant, the criminal law also, from time to time, produces products that do not comply with those standards, and that reflect varying degrees of defectiveness.  Those standards may contain “latent defects”.[13]  In that sense, a prohibited behaviour may be of very long standing, or very old (like attempted suicide, sodomy, witchcraft, etc.) and yet not be “standard”, that is, it may not comply with certain standards that are usually considered to be capable of ensuring a minimum level of quality in what is produced.  That is why we prefer the expression “standard” to “conventional”.


We pondered the appropriate label for the second group at length.  At the outset, the idea of calling them two-sided prohibited behaviours came from the observation that in the case of these prohibited behaviours, the criminal law confuses the perpetrator and the direct victim – or, more precisely, the observation that in these cases the criminal law was not capable of clearly identifying a perpetrator and a concrete, distinct victim of the crime.  The situation is the same as picking up a coin from the ground: you necessarily pick up both sides at once.  In this sense, the label “two-edged prohibited behaviours” also seemed acceptable.  We were then attracted by the idea of self-referential prohibited behaviour, the reason for which will become apparent later.  All these images remain, but as the analysis progressed, it became obvious that the expression “two-sided” was more heuristic and could take on other meanings, for each criterion. For example, it also calls up the image of “two-sided cloth” – fabric that is woven in such a way that it can be used on both sides.  That image suits the second criterion.  And so we chose the label that seems to be justified by a set of criteria.[14]


Of course, these two broad types of prohibited behaviours are not the only ones possible.  But the hypothesis we are working with is that they are the two most extreme cases on a continuum on which other prohibited behaviours, or less opposite sub-groups, may be found.  Those intermediate prohibited behaviours or sub-groups comprise “grey areas” that share at least one of the seven (theoretical and empirical) criteria indicated.  It is very important that these grey areas be identified, from the standpoint of both public policy and the legal system, because they can be used to rethink the appropriateness and, where applicable, the appropriate forms of intervention to be used in those cases.  We need not explore the details of those grey areas here; all that is needed is that the theoretical indicators (criteria) used for that purpose be characterized, and the two groups of prohibited behaviours identified which are, in our hypothesis, the most radically opposite: standard and two-sided prohibited behaviours.


These seven criteria for distinguishing prohibited behaviours take the form of a comparison between the two groups.  The criteria used to identify the two-sided prohibited behaviours all, individually, indicate different types of serious difficulties when it comes time to decide whether a behaviour must be regarded as a crime by a criminal law (law making process) or be the subject of a finding of criminal guilt by a court (decision making by courts).  Even more acutely, these criteria point to the problem of using incarceration both in the law and in sentencing.


In addition, these criteria point to the possibility of special difficulties in the more general area of law enforcement by the police and political policies.  In the specific case of two-sided prohibited behaviours – to which a number of these criteria apply – mere criminalization of these prohibited behaviours accompanied by a sentence of imprisonment could result in formidable social and economic costs (the cure may be worse than the disease: it may exacerbate it, and produce a chain of other diseases).


            Table 2 summarizes all of these characteristics, but we recommend that the reader read it only after reading the description of each characteristic.  The reason is pedagogical: the table proposes a distinction that is so unfamiliar that we are afraid that it will not really say much on its own.


To illustrate the distinction between the two groups of prohibited behaviours, we have given eight examples of two-sided prohibited behaviours (or groups of two-sided prohibited behaviours), while trying to have as much diversity as possible among them.  They are: illegal drugs, gambling, prostitution, abortion, homosexuality, sodomy, vagrancy or begging, and attempted suicide.  These examples come with no claim to being exhaustive.  Moreover, we would repeat that there may be a large number of other situations or prohibited behaviours to which at least one of these seven criteria apply.


From a methodological standpoint, the choice of the examples of two-sided prohibited behaviours was motivated by the need to visualize both their commonalities and their differences, or internal contrasts.  The advantage of this is that we get both a more complex and a more differentiated picture of the group.  However, we cannot describe the points of difference systematically, or explain all the commonalities, for each example.


Because our primary objective is to clarify this category, from both the theoretical and practical standpoints, it is of little relevance whether all these examples are still regarded as crimes today.  The important thing is that they enable us to elucidate the difficulties posed by the group and to answer certain theoretical questions that may arise from this distinction, and that we are able to anticipate.


If we borrow an image from the French sociologist Raymond Boudon (1986, p. 81), we can take a tour of a table that will compare two fundamental types of prohibited behaviours using seven criteria.[15]



Table 2: Comparison between standard prohibited behaviours and two-sided prohibited behaviours


Type 1 prohibited behaviours

(standard prohibited behaviours or crimes)


Type 2 prohibited behaviours

( two-sided prohibited behaviours or crimes)

Examples: Homicide, sexual and physical assault, theft, fraud, destruction of property


Examples: Drugs, gambling, prostitution, abortion, homosexuality, sodomy, vagrancy and begging, attempted suicide


(1) Deviance or conflictual interaction: this takes the form, whether direct or potential, of a conflict between two social actors or an infringement by one on the other.


(1) Deviance by exchange or personal relationship: it takes the form of a personal relationship between individuals or a fair, consensual and direct exchange of services or goods.

(2) Capacity of the law for discernment: the law may see a direct victim and distinguish that victim from the deviant.  (The legal enactment that may punish the deviant is not usually used to punish the victim).


(2) Incapacity of the law for discernment (or blindness of the law): the law may not see (or find) a direct victim and the enactment may punish human misery, lifestyle choices or morality in itself, or victims of their own behaviour.


(3) Relationship to someone else and unavoidably negative character of the act: the actor may be persuaded to acknowledge the negative consequences of his/her conduct.


(3) Self-referentiality: possibility that the act may be a matter of preference or “not necessarily negative” in itself.

(4) Limitation on natural liberty: the question of the rights or liberty of the deviants to engage in the prohibited action does not arise per se.


(4) Restriction of liberties imposed by the law: the issue of respect for the relative autonomy of deviants vis-à-vis the criminal law may arise.


(5) Weak cognitive dependence / component: whether or not there is a sound basis for the criminal prohibited behaviour does not depend on specific knowledge and is not directly dependent on such knowledge; in that sense, it is “self-evident” that there is a sound basis for the norm.


(5) Strong cognitive dependence / component: knowledge (or reasoning) whether or not there is a sound basis for the criminal prohibited behaviour is guaranteed by or depends directly on knowledge (or reasoning): it is “not self-evident” that there is a sound basis for the norm.


(6) Predominance of reactive formal intervention (after a “complaint”): victims file complaints.


(6) Predominance of proactive formal intervention: the justice system acts without receiving a complaint (at the initiative of formal agencies of social control).


(7) The criminal law as not producing effects that counteract the prohibition: the problems created by applying the criminal law may not compromise the norm prohibiting the behaviour itself.  Problems may be mitigated or avoided by changing our manner of reacting.


(7) The criminal law as producing effects that counteract the prohibition: the problems created by criminalizing the behaviour compromise the criminal law itself.  The act of creating a criminal law may be seen as counter to the stated goal of the prohibition, as being counter-productive or as being internally inconsistent in terms of values.  This may be referred to as an “error” or “aberration” in relation to the legislative policy.



  Criterion 1: “deviance by exchange” or personal relationship 

This criterion is taken directly from Schur (1965, p. 170). It is the key element in his definition of the concept of “victimless crime”.  We have abandoned that concept, but by using this criterion we have adopted his fundamental idea.  Broadly, Schur distinguishes between two situations: conflictual (interaction or) deviance and deviance by exchange.  In the first, the norm deals with a conflict between two parties; in the second, the norm deals with a consensual interaction, relations or relationship between two individuals.


In the case of a conflictual deviant behaviour, we can talk about infringing on part of someone else, and ultimately a problem of resolving conflict between two social actors, that is, between two parties: an assailant and a victim.  In the second case, on the other hand, we are dealing with a situation involving a direct, fair exchange (Schur, 1965, p. 170) between two persons.  Schur notes that this is a relationship involving a well-organized, fair exchange of goods or services – that are otherwise prohibited by law – between two parties.[16]  We will not misrepresent him if we include in this emotional or sexual personal relations between individuals, such as, for example, homosexual relations.[17]


Generally speaking, this criterion refers to direct, fair and well-organized personal or consensual relations between two parties.  In the case of the sale of drugs or sexual services, it is, properly speaking, an exchange of goods and services, respectively.  That means that there is no conflict between the parties; it is the norm that is external to the interaction that we use in order to talk about deviance and that lays the foundation for a potential second-level conflict, that is, a conflict with the justice system.


(a) The concept of “victim” in the criminal law

Note that the criminal law uses the word victim to have at least three meanings, each of which corresponds to a different level of abstraction.  Table 3 sets out those meanings in descending order, from the most abstract (level 3) to the most concrete (level 1):


Table 3: Three levels and three meanings of the concept of “victim” in the criminal law


3rd Level (most abstract)

Deviant vs. society:

The victim is society or the group’s morality


2nd Level (intermediate)

Deviant(s) vs. third party observers:

The victim is represented by the immediate entourage (family etc.), the neighbourhood or the nearby surrounding community


1st Level (most concrete)

Deviant vs. direct victim:

The victim means the person who is directly and immediately offended by a conduct prohibited by the law (in the event of death, the victim also includes that person’s family and friends)



At both the most technical and very heuristic level (3rd level), the only victim in the eyes of the criminal law is society, understood as an indivisible whole.  It is in that sense that the criminal law says “every crime is a crime against society”.  This is the modern version of the old legal saying that “every crime is a crime against the King”.[18]  This usage takes the following form: Deviant vs. Society.  In this case, a person who was deliberately killed by another person (or the family of the dead person) is not the victim, in the eyes of the criminal law: the victim is Society.  If there has been a theft, the person who has been stolen from has the status of witness.


            But also, at the other extreme, the criminal law uses the word “victim” in the everyday sense (1st level): the victim is distinguished from the perpetrator; the perpetrator has made a victim.  If your house has been burgled (not by you in order to collect the insurance), you are called the victim by the police, the courts, etc.  The law may allow you to tell your story to the court; you may be eligible for a crime victims compensation program; there may be a victims’ assistance service at your trial to explain to you what procedure to follow, etc.  This usage takes the following form: Deviant vs. (direct) Victim.


            And, even more vaguely and usually not found in norms, there is another meaning of the word “victim”.  It is used, in particular (but not exclusively), to refer to a nuisance situation that does not directly affect the parties involved, but rather affects their entourage or “third party observers”.  A typical example is soliciting on the street for the purpose of selling sexual services.  The neighbours may feel disturbed and there may even be economic losses resulting from a drop in real estate values.  In this case, it might be better to talk about nuisance (or economic loss, where applicable), to distinguish it from a wrongful act in the strict sense in conflictual direct relationships.  This usage takes the form: Deviant(s) / third party observers.


            We cannot observe prohibited behaviours using the three distinctions at the same time.  We can move from one to another, but each time we do that a part of the situation disappears from sight: if we observe from the Deviant vs. Society standpoint, we abandon the opportunity to observe whether or not there is a concrete, distinct or deviant victim that is not Society.  Conversely, if we observe from the Deviant vs. Victim standpoint, Society as a whole becomes invisible.  But there is a noteworthy difference between these two extreme levels.  Observation at the first level may lead us to conclude that there is no direct victim.  On the other hand, any behaviour that disturbs someone or breaks some law (including the civil law) may, in theory, be regarded as an offence against society.  This level of analysis requires that we exclude nothing, and that is why it is not used to discriminate between what should and what should not be dealt with by the criminal law.  One of the great legal thinkers of 19th century Germany correctly identified the problem that this level of abstraction presents, when he wrote: [TRANSLATION] “If contracts of sale are not performed, if loans are not repaid, Society is threatened just as if the public were killing or stealing from one another” (von Jhering, 1877, p. 319). In other words, the civil law also “protects” society.


(b) Crimes where there are no conflictual social interactions


Schur (1965) observed the conflict exclusively at the first level, and then said that some forms of deviance are characterized by a direct, fair exchange relationship between the parties.  He then noted a curious fact that often goes unnoticed, precisely because we move quickly from level 1 to level 3 without realizing the mental operation we are performing.  He saw that, at that level, there were “victimless crimes”, that is, crimes where there was no conflict and no direct harm between two parties.


For example, let us consider the prohibitions on trafficking in drugs.  This activity does not involve the same style of conflictual social relationship that we find in standard prohibited behaviours.  Here, we have a number of typical cases: a mere user, a user who resells, a user who traffics and a trafficker who does not use.  But in all these cases, the relationship between the parties is consensual at the time of the purchase and sale, and may be distinguished from the relationship between a thief and the victim of theft.  To visualize this, let us imagine for a moment that there is no prohibition on drugs.  In this case, we note that, from an economic standpoint, this activity cannot be distinguished from ordinary industrial and commercial activities: production, marketing, sale, investment of profits in the financial market, etc.  The action is the same as for producing cars, sausages, wine, cigarettes or firearms.


One of the great American sociologists, Robert Merton (1957, p.79), pointed out that “in strictly economic terms, there is no relevant difference between the provision of licit and of illicit goods and services”.[19]  If the consumer a victim of some other fraud on the part of the vendor (the car sold has a hidden defect, etc.), he or she cannot be seen as a victim; rather, the consumer is making a consensual choice.  Certainly, individuals’ choices may – in the longer or shorter term – jeopardize their health, affect the lives of the people around them, etc.  All of this may, in some cases, justify information programs, prevention programs, treatment programs and legal regulation programs.  But here again, those people are not the “victims” in the same sense as an individual who directly assaults, or kills, his or her victims.


Consumers who are looking for drugs does not feel “assaulted” by the vendor as they would feel if they were assaulted in the street by a person who pointed a revolver at them and demanded that they hand over their wallet.  The vendor offers, perhaps even insists, but does not coerce.  The “victim”, so to speak, participates willingly in the process that is the subject of the prohibition.


We shall disregard the error identified by Hunt (1993, p. 100) which is made by some criminologists: “crimes of the powerful” against financial institutions, or crimes such as tax evasion, are not, in Schur’s view, “victimless crimes”.  That is a complete alteration – although a different one – of the meaning of the concept.  At the first level, that relationship is neither fair nor a mere exchange relationship; rather, it is a relationship of expropriation or infringement.  “Diverting” money that belongs to someone else always produces a victim somewhere, so there is a victim at the first level, even if we do not know it (as in the case of a fraud between two individuals that goes unnoticed).


Criterion 2: incapacity of the law for discernment (or blindness of the law)


The second criterion that differentiates between the two kinds of prohibited behaviours relates to the other side of the preceding problem. Instead of stressing what happens in relation to the deviance or behaviour, we shall see, in this case, what happens in relation to the criminal law.


(a) Capacity of the criminal law for discernment


We shall first examine what happens with standard prohibited behaviours, and freely say that in those cases, the criminal law is capable of observingin turn – the three levels of victims that we identified in Table 3: it can see that there is a direct victim of the action (level 1), or observe the conflict in relation to level two or three.


Note that in the purest or most pristine situations, the law has all the conditions that it needs in order to see clearly a direct victim and distinguish that victim from the deviant, using the actual definition of the crime (the enactment).  These pristine situations occur when the enactments (laws) in the criminal law, which create the crimes (e.g., in the case of homicide, sexual and physical assault, theft, fraud, wilful damage to property, etc.), urge the legal system to clearly see the difference between the perpetrator and the victim.  In short, they occur in all cases where the charge is capable of guaranteeing a radical separation between two independent actors, where one can be depicted as the perpetrator of the prohibited behaviour, and the other as the victim of that behaviour.  Thus a person who kills another person is not phenomenologically “included” in the person who is killed; a person who steals something is not included in the person from whom it is stolen; and so on.


Thus, if one person steals from another, the criminal justice system can discern, at the first level, that the person who has been stolen from is the “victim” and the person who stole is the perpetrator of the act, or the deviant.  These two social roles are both perfectly clear and distinct and external to the law itself, and recognized as separate in the law by the definition of theft – to the point that in this case the criminal justice system cannot make this kind of error: “punish the victim for being stolen from.”[20]  The enactment that creates the crime of theft and prohibits someone from stealing itself makes it possible to observe the conflictual deviance, because it separates out the perpetrator of the crime and leaves the victim aside.  Another example would be that a woman who is a victim of sexual assault may not be imprisoned for sexual assault, even though she may experience other problems along the way.  We shall come back to this.


(b) The example of a “grey area”


We should note that certain situations (which are not two-sided) fall into a grey area and in that case the separation between the perpetrator and the victim is not always pristine, and sometimes calls for special alertness so as not to be misled.  This happens in situations where the enactment (the charge) separates a phenomenological reality that is not as clear as the enactment seems to suggest.  We are thinking of the prohibited behaviour of “criminal negligence”, or of certain situations or behaviours that fall within the concept of “duty offences” proposed by the German legal thinker Claus Roxin (1970/2000, p.34).[21]  Consider a random situation that recently occurred in Canada.  A woman who was responsible for caring for children had a car accident that resulted in the death of several children, including one of her own.  There had been a violation of road safety rules, and the media announced that the legal system would be (on its own initiative) charging the “perpetrator” of the crime.  The problem is plain: the “criminal negligence” enactment (or an equivalent provision) makes it possible to name a perpetrator and to separate the perpetrator from the victim (in this case, the children).  But the situation was not pristine: the perpetrator of the crime was “included” in the result of her action, and she suffered from that action from all standpoints.  Not only was she in the car that crashed and a witness to the death of the children for whom she was caring, but she also lost a child.  This was when the community reacted against the (temporary) blindness of the legal system and prevented a charge from being laid.  A charge of that nature may be seen from outside the criminal justice system as an absurd and inconceivable decision, because the system is setting out to “over-penalize” a person who has already suffered the consequences of the act.  In these cases, we can say that the consequences themselves affirm the norm, because they comprise what the classical philosophers called sanction by reality.  Nonetheless, when seen from the standpoint of sentencing theory and some other related considerations, the charge is self-evident and can seems to be justified.  For all these cases, we must do more investigation to determine what is going on.[22]


From the theoretical standpoint we are adopting here, this situation does not correspond to the pristine or pure version of a standard prohibited behaviour.  According to that theoretical criterion of standard prohibited behaviours, the person who commits the criminal action is not the person – or one of the persons – who also suffers the consequences of the action.  We can also state this criterion more paradoxically and provocatively: in the pristine standard prohibited behaviours, the victim of a prohibited action is not charged because of the action itself.  The woman driving the car was also a victim of the accident which is, to all intents and purposes, defined by the law as criminal negligence.  The reason she may be charged is that the charge does not allow us to see her as also having been a victim.


(c) Aberration of the criminal law in the case of standard prohibited behaviour


Of course, even in the case of standard prohibited behaviours, the criminal law must also change levels and observe the victim at level three and say, for example, “the theft from X is ‘in reality’ (i.e. from the standpoint of the system) a crime against society”.  In that hypothesis, X – who was the victim – becomes a witness or a mere first-level victim who is superseded and overshadowed by a third-level victim (Society).  In the competition between level 1 and level 3 victims, the latter prevails, in theory, because carries more weight in the legal culture and the normative structure of the law.  For example, if the first victim wants to forgive, but the second victim (Society) does not, the crime will not be dealt with by mutual agreement.


            When the criminal justice system observes the phenomenal world through clouds (level 3), some problems may occur in terms of standard prohibited behaviours where there is a clear enactment – where the criminal justice system is, in theory, capable of plainly seeing and discerning the victim of the perpetrator of the act.  For example, we have seen, in Canada, a court sentence a woman who was a rape victim to a term of imprisonment for contempt of court, because she refused to testify in court against her assailant.  In that case, the court observed the rape from the Deviant vs. Society standpoint, and consequently lost sight of the phenomenal reality: it “did not see” that there was a concrete victim sitting on the witness bench.  It then acted “in good conscience” and, in the name of protecting the public, sent the rape victim to prison.


This becomes cognitively possible because the concept of “protection of society” produces, from a psychological and socio-psychological standpoint, a blind spot concerning the internal functioning of the system and the protection methods used.  To paraphrase George H. Mead (1917), the system does not clearly see what it is doing behind the walls that were built to protect: protection has become what is important.[23]  It is very difficult for the human mind to want, at the same time, to protect Society “at all costs” and to think about how that protection is achieved and at what cost.


(d) Incapacity of the criminal law for discernment (or blindness of the law) in the case of two-sided prohibited behaviour


            And now what about two-sided prohibited behaviours, those in which there is precisely, in phenomenal terms, no conflictual deviance?


            We will begin to answer this question with one of the clearest examples: a casual soft drug user who smokes a marijuana cigarette at home.  There is not a sociologist in the world, except one who is hallucinating, who is capable of observing any conflictual deviance in this case in phenomenal terms (criterion 1).  Neither, therefore, can the criminal justice system.  In this case, it cannot produce one of the two mirror effects: make things that are not there appear.  There is no victim who is independent of the perpetrator, in phenomenal terms (first level).  The legal system must therefore find another reason to justify intervening in a manner that may send this person to prison.


Certainly, there is still a last resort if we are looking for a victim in phenomenal terms.  We can use our imaginations, and project ourselves into the future to find the possibility of some kind of risk to the user himself or herself: the risk of becoming addicted; the risk that he or she will later become a hard drug user, and then an addict; the risk that he or she will bring up his or her children badly; the risk that he or she will become violent; etc.  But that artifice cannot be used to circumvent the second criterion, because the person who commits the action charged is also the person who suffers the consequences of that action.  In other words: the victim (to all intents and purposes) of the prohibited action (the joint smoker) is charged solely by reason of that action.[24]


In short, where drug use is controlled and causes no harm, there is only a perpetrator in relation to himself or herself.  That is, there is no direct and immediate victim, not even the individual himself or herself.  The most that it is, in some cases, is like engaging in risky sports.  In the worst cases, “perpetrator and victim” are one and the same person.  If the criminal law intervenes to deal with the user, it is punishing either a perpetrator who has no victim, or both the perpetrator and the victim as represented by the same individual. 


But protecting people from themselves, as laudable and justified as that may be, is still not enough to sentence someone to a (severe) term of imprisonment, let alone on the basis of hypothetical risks that do not, in large numbers of cases, ever materialize.  Treatment outside the criminal justice system and not controlled by that system (Baratta, 1990, pp. 169-170) would be another matter, even if it is still without sufficient justification in this case.  And the difficulty increases when we make a mental comparison between the risk of marijuana and the risks of using alcohol or tobacco, which are permitted drugs or substances, or when we make a comparison with other risky activities in modern life, including occupational and recreational activities (extreme sports, etc.).


(e) Role of abstraction in the undue justification of criminalization (a cognitive trap)


To justify criminalization, the criminal law then goes on to observe other things, at other levels of abstraction: precisely the place where our mind is more detached from empirical reality, to “see” another type of conflict, to see another type of problem, and avoid uncomfortable comparisons.  When the system cannot summon up a direct, actual victim in its mirror – as no one could, in these circumstances – it can only pivot and separate from the phenomenal surface.  And so it creates for itself a diversion effect, in both senses of the word: it moves toward other “goals” and other levels of abstraction in order to go and “see” somewhere else.  The goals in this case play a crucial (and negative) cognitive role, in that they change the type of reasoning used and provide another direction in which the mind can go.  We leave the realm of analytic reasoning (“reasons to understand or know”) and embark on a kind of consequentialist reasoning (“reasons to do”) which, in this case, we are often unskilled in or fail to examine properly.  Consider that the system then provides for itself a goal that has high value, at the second or third level.  For example, the highly valued goal of “protecting children”.  For obvious reasons, once it has that goal the system no longer sees what is going on in phenomenal terms, and finds it very difficult to see what it is doing.  It has gone somewhere else.  The law then becomes blind in practical and phenomenal terms.


The change of the “goal” and of the level of observation produces a cognitive trap for the criminal justice system.[25]  In the case we are examining, that trap results in our failing to consider the principal problem (drugs) because our mind is occupied with what is, in the circumstances, a secondary and even imaginary consideration.  That secondary consideration (“protecting children”) may even suggest to us that we must continue to make one mistake (in terms of solving the principal problem) in order to achieve the second goal!  As the mathematician Gabriel Stolzenberg points out (1988, p. 279), it is a characteristic of cognitive traps that they cannot be seen by someone in the trap for what it is.  Let us consider prostitution, for example.  Suppose that you believe that prostitution “in itself” – even between consenting adults in private and in whatever form it takes – causes harm to children, sex workers, etc.  This is not unlike a person who is convinced that all violence on television also harms children, journalists themselves, etc.  What then happens if you agree to keep prostitution illegal (to “protect children”), despite all the problems that doing so causes?  You forget to give serious consideration to the principal problem (should we be sending sex workers and their customers to prison?) and you become convinced that although it is a mistake to do so, and it is contrary to certain human rights, the mistake is justified and is necessary (the only method) in order to produce happiness in children.  That is a cognitive trap.


(f) The criminal law in relation to human misery and lifestyle choices


            Let us return to our question: what about enactments in respect of two-sided crimes?  Because those crimes have no direct victim in phenomenal terms, the criminal law does not clearly see whom it is punishing.  It then runs the risk of punishing human misery: the prostitute with a background fraught with misery, exploitation and violence, the drug addict who needs help rather than punishment, young (casual, habitual or dependent) users of various drugs (soft and hard), etc.  And, in phenomenal terms, it punishes precisely the young people who are protected, at the second or third level of abstraction, by the punishment.  The system has surpassed itself in the paradox it has constructed in portraying the drug addict: it punishes the victim of the victim’s own act.  To borrow the words of the Belgian criminologist Dan Kaminski (1990, p. 180), [TRANSLATION] “charging people for using narcotics makes a criminal of the drug addict”!  Because the system does not identify either the perpetrator or the victim, in empirical reality, the enactments in the criminal law criminalize and punish both at once.


            That is why, in deciding to criminalize and severely punish two-sided prohibited behaviours, the criminal justice system thereby agrees to run the serious risk of severely punishing  human misery, lifestyle options (these being sometimes the most tragic for individuals, as in the case of abortion) or sexual identity, and of punishing the victims of their own behaviour.  And it does this in the name of “morality itself” (Hart, 1963; Packer, 1968), of protection from an imaginary risk or protection of another legal goal or good that is valued at the third level of abstraction, that is, quite remote from empirical reality and with no direct observable relation with all the facets of the problem under consideration.


If, for example, the criminal law prohibits the sale of sexual services between consenting adults, it is not capable of distinguishing, as in the case of standard prohibited behaviours, between the perpetrator and the victim for the simple reason that both are the “perpetrators” of a consensual exchange.  Between them, there is no perpetrator or victim at the time of the exchange.  The nuisance problem that may arise from soliciting to sell sexual services on the street can be addressed by a special regulatory law.  It is improper – or in a sense quite different (second level) – to say that “the victims are the people in the vicinity”.  Certainly this does not in any way prevent the criminal law from prohibiting certain forms of exploitation of human misery, such as pressuring or forcing children into selling sexual services.  In that case, it is protecting the child from exploitation with a lower risk of simply punishing human misery.


The criminal law is quite aware that the perpetrator/victim relationship in the case of two-sided prohibited behaviours is not the same as the relationship it is accustomed to: it knows that it is dealing with a “victim” of something (an organized prostitution ring, an illegal drug market, etc.) and it defines the prohibited behaviours in such a way as to address the whole thing.  Nonetheless, the paradox is this: it knows that there is a victim of something, but at the same time it does not see the victim, or sees the victim too quickly for its attention to be caught.  When the criminal justice system has noticed that when it comes to the sale of sexual services it is criminalizing “exclusively” the seller and not the buyer, it has redressed this “inequality” by punishing both.[26]  By declaring war on “prostitution” as a whole (third level of abstraction), it appears to be both fairer and more justified.  The war on prostitution as a whole can also be seen as a goal to protect children (“reason for doing”).  It is only when it observes in the abstract, disregarding the relationship between the individuals (because it is not conflictual), that it can cross over into criminalizing the victim without appearing cruel, and without realizing it.  Paradoxically, however, that is the point at which it makes the most mistakes.[27]


Criterion 3: self-referentiality 

The third criterion for distinguishing between the two types of prohibited behaviours refers to the difficult problem of individual autonomy and the self-referential nature of certain prohibited behaviours.  We shall simply make a few comments that will provide an overview of this issue.


            In standard prohibited behaviours, despite the different ways of dealing with the situation and of self-persuasion, the act takes on – or may take on – a bilateral meaning, that is, for the perpetrator and the victim.  In addition, the problematic or harmful nature of the act to someone else is more likely not to pass entirely unnoticed by the person who commits the act than in the case of two-sided prohibited behaviours.


Certainly, an individual may assign preferences to illegal behaviours which, in his or her eyes, are more acceptable than others (tax fraud, etc.),[28] develop mental “neutralization mechanisms” (Matza, 1969) to justify his or her conduct, etc.  But in candid discourse, it is virtually impossible for the individual to avoid recognizing that he or she has (deliberately or not) caused harm to someone else.  The “social negativity” (Baratta, 1983) of the act is factual and immediate to the extent that the person who commits the act is able to see himself or herself as the immediate and proximate “cause” of the harm, regardless of the problem of acceptance of full subjective responsibility, particularly if the harm is not a contingent and remote risk, but the result of an action that the individual was trying to commit (intentional homicide, assault, theft).  Even in the legal depiction of an attempt, for example, a virtually inescapable bit of reality[29] leading to the harm has already started to materialize.


When we turn to two-sided prohibited behaviours, we cannot but recognize that this does not seem to happen in entirely the same manner.  The individual may see himself or herself as making a decision relating to the individual personally, to his or her private or professional life, and not be obliged to see that act as causing a problem for other people.   Depending on the circumstances – we are excluding suicide here – the individual may even claim that the act does him or her no more harm than other legal options available in social life (smoking tobacco, using alcohol, engaging in a risky sport in order to experience strong sensations, etc.). 


(a) Absence of malice toward someone else


We may even go so far as to say that malice in the relationship to someone else is not present – or cannot be inferred from the behaviour – and does not seem to be an immediate fact in the experience of the individual himself or herself.  It takes a good deal of imagination, not to say a touch of loss of contact with reality, to see smoking a joint while listening to music as an interference with someone else or a “crime against Society”.  The individual who commits the act is undoubtedly perfectly aware of the norm that prohibits the conduct.  But that is a different question: the individual knows that the act is prohibited, but that has nothing to do with any “malicious intent” in respect of anyone else, or even with recognition of any “immorality” in the act.  The individual cannot – with good reason – even accept a mistaken pedagogical message stating that this will destroy his or her health.[30]  There would be more reason to fear parachute jumping than smoking a joint, or even to fear smoking tobacco.  The norm may therefore seem to the individual to be entirely unreasonable, or not sufficiently reasonable to justify the scope and punitive form it takes.  Because there is no conflictual deviance, or obvious contempt for anyone else, or even necessarily for one’s self, the decision to commit the act may take on a strong self-referential dimension for the individual: it may be the individual’s way of balancing the individual’s own being with his or her existence.  The relationship to a large number of such behaviours then has a tendency to fall, de facto, into the realm of “lifestyle options that hurt no one else”, or that are less harmful to anyone else than other risks or problems of existence that every individual increasingly has to deal with.


(b) The limits of legal paternalism by the criminal law


For a majority of the situations that are criminalized by two-sided prohibited behaviours,  there is not necessarily any intention to do one’s self harm.  This was very shrewdly pointed out by a Belgian journalist, Philippe Toussaint (1983), who specializes in legal and judicial problems:[31]

[TRANSLATION] I have always and tragically remembered what was said one day, in an off the record way at a hearing, by an eminent magistrate in the prosecutor’s office during the trial of a number of drug users.  He leaned toward one of the accused and, rather as if he wanted to point out that one of his bootlaces had come undone, said:

- Psst!  Sir!  Sir!

The man raised his head, surprised, and the prosecutor said:

- You are going to die!

Precisely as if he were offering some information.

                The justification for punishment is then sufficient: it consists in protecting these unfortunates from themselves; at its most extreme, it is assistance to a person who is in danger of dying.

                And yet we are well aware that this is not true.  We are dealing with someone who absolutely does not want to die. … (Toussaint, 1983, p. 291; emphasis added).


Toussaint added that he is not sure that our duty to intervene in such cases is as clear as the duty that we have in the more extreme cases where we are dealing with a priest who is preparing to light himself on fire.  He then raises the difficult problem of the limits of legal paternalism and, by implication, the methods used (the criminal law and incarceration) to “protect” individuals from themselves.


In a book published in the early 1960s, before enough thought had been given to the drug problem, H. L. A. Hart, was undoubtedly a little too hasty to justify legal paternalism[32] in the criminal law.  In Hart’s defence, it must be said that this question was marginal to his study and that he was not specifically concerned, in it, with examining the limits of that justification and the (morally and legally) acceptable forms of allowing room for paternalism in the law.


Hart was not wrong to argue that at least a certain degree and certain form of paternalism in the law seems to be unavoidable, and legitimate, in order to protect individuals from themselves on some occasions.  That is a given.  For example, we cannot authorize the practice of assisted suicide in response to the slightest depression, when the individual says that he or she has lost interest in living.  Nor, wherever possible, are we going to allow people to kill themselves in the public square to express their support for a political cause.  As well, we may legitimately prohibit duelling and consensual physical assaults that take place outside the context of certain properly regulated sporting activities, etc. 


However, the method by which we intervene to protect individuals from themselves must not still not be the same.  We might think that the criminal law prohibits consensual physical assaults (duelling, fighting, etc.) because there is a violent relationship with someone else.  But the same is not true of individuals who wish to commit suicide, self-mutilate or simply use drugs to deal with life or enjoy themselves, without having any intention of doing themselves harm.  It must also be recalled that some illegal drugs do less harm than some legal drugs (such as alcohol and tobacco).  Intervention by the criminal law in those circumstances is even less justified and justifiable if it uses incarceration and, above all, long sentences.  In this case, the paradox reaches a paroxysm: the outside “protection” may cause the individual more harm than his or her own behaviour.


In addition to the problem of the limits of paternalism, there is the problem of the illogic in legislative policy, a point that has been vigorously argued by Caballero (1992, p.18): the criminal law has decriminalized attempted suicide, and except for exceptional situations (military justice, for example) no longer criminalizes self-mutilation.  And [TRANSLATION] “generally speaking, risk behaviours designed to produce sensations are tolerated by the law” (ibidem).  He goes on to say:

Although some behaviours are extremely dangerous (car racing, hang-gliding, bungee jumping, etc.), they are not prohibited, just as there is no prohibition on drinking alcohol, smoking tobacco and taking tranquilizers, etc.  Alcoholism, nicotine addiction and drug dependency cause about 100,000 deaths ever year in France.  (Caballero, 1992, pp. 18‑19).


We see the problem: how far can we go to protect individuals from themselves?  Can we prohibit individuals – by a criminal law – from doing themselves harm when engaging in an activity in which they are not trying and do not want to do themselves harm, but rather to do themselves good?  What about if a majority of the people who engage in these practices do not do themselves harm?  And what about if a large portion of them harm they may do themselves is more the result of prohibitionist policies that restrict individual and societal control over the quality of the product and the circulation of accurate information about its use?  Can we completely prohibit a practice – and punish all those who engage in it (whether or not they need to be protected from themselves) – because some individuals lose their self-control?  Should we, for example, prohibit taking the risk of driving on the roads because some individuals drive over the speed limit and kill themselves?  Should we completely prohibit taking the risk of skiing or swimming elsewhere than at supervised clubs and beaches because some people represent a danger to themselves?  And what then about alcohol use, etc.?


The need to accept a certain degree and form of legal paternalism is thus not a sufficient reason to automatically authorise the use of the criminal law.  The absence of any conflictual relationship, malice or intention to do harm, the possibility that the behaviour can be seen as ordinary management of one’s own existence, etc. tend to disqualify criminal intervention, and particularly the use of degrading ceremonies and punishments (Garfinkel, 1956):

The work of the denunciation effects the recasting of the objective character of the perceived other: The other person becomes in the eyes of his condemners literally a different and new person … .  The former identity stands as accidental: the new identity is the ‘basic reality’. What is now is what, “after all”, he was all along (Garfinkel, 1956, pp. 421-422).


While today we reject the use of the criminal law and incarceration in the case of attempted suicide, even though we accept legal paternalism, that paternalism cannot ever lead to punitive intervention against individuals who are not a danger to themselves in the same sense as individuals who actually intend to do themselves harm.  The excessive legal paternalism – which comes with no limits in terms of the appropriate methods – is too similar to totalitarian philosophies.[33]


As W. de Jong (1983: 283)[34] also pointed out, using an illegal drug, like doing illegal work, is not regarded by the heroin user as a barrier, even though heroin is a hard drug.  We can imagine the difficulty of seeing soft drug use as a “problem”. 


(c) The individual’s responsible autonomy


When an individual must make a decision concerning his or her existence, and the decision is by definition not an assault on himself or herself or anyone else, it is difficult for the individual to see and accept that the criminal justice system might, in certain situations, see the risk that he or she is taking (to avoid another) as a danger to everyone, including the individual.


It is, of course, up to the law to level the playing field, to establish conditions and impose certain limits.  It may even be desirable not to do with drugs what we have done with gambling.  In that case, we have gone from one extreme to another: from criminal offence to permission to engage in the highly publicized “active business” of lotteries and casinos.  Some experts have already come up with more ingenious methods, based on the idea of a “passive business”, [35] in which the quality and sale of the product would be controlled and advertising prohibited.


In a society that stresses the individuation process (Beck, 1986) and increasingly demands individual responsibility, people instead have to learn to be autonomous.  As Foerster wrote (1981, p. 67), autonomous individuals are those who “regulate their own regulation”.  He does add that [TRANSLATION] “autonomy involves responsibility: if I alone am to decide how I act, then I am responsible for my acts.”  Teaching and demanding responsibility is different from a sort of clairvoyant paternalism that determines a prior what is good for everyone and imposes severe punishment to enforce that vision.


Of course, there is nothing to prevent us from using the criminal law or some form of regulatory law to prohibit an aspect of the situation that may take a particularly problematic form, such as driving while intoxicated.  At the outside, we can prohibit certain conduct without using the criminal law, or imprisonment, to effect the prohibition.


(d) Individual, society and the principle of optimizing fundamental rights


As a final point, we would note, as did Charles Taylor (1985), that valuing responsible autonomy must not be confused with an atomist or individualist moral doctrine or policy.  By that, we mean a doctrine [TRANSLATION] “that attributes priority value to the human being as an end in relation to the communities to which he or she belongs” (Abbagnano, 1999, pp. 554-555). We must try in some way to find another method of observing both individual/society relationships and the fundamental rights that exist between them, a way that is not defined exclusively in the form of a confrontation or radical choice opposing individual vs. society.


One of the most unfortunate effects of the individual vs. society opposition is that it does not contribute to the development of a style of intervention: that is, the manner in which the legal system as a whole, and the criminal law in particular (style of procedure and types of sanctions), are used for intervening.  Two closely related problems are involved here.


The first is the tendency to place an absolute value on one of the principles (the principle of the protection of society) at the expense of the other (the principle of individual liberty adopted), instead of viewing the problem as a requirement for optimising the actualization of both basic principles.[36]  As Alexy (1986, chap. 3, I, 2), accurately noted, [TRANSLATION] “the principles are norms that require that something be done to the fullest extent possible. … Consequently, the principles are orders of optimization … .”  Accordingly:

It is easy to argue against the validity of absolute principles in a legal system that recognizes fundamental rights.  The principles may refer to collective goods or individual rights.  When a principle refers to collective goods and is absolute, the norms of fundamental rights cannot establish any legal limit.  Consequently, until the absolute principle is found, there can be no fundamental rights (Alexy, 1986, chap. 3, I, 7.2).


            It is important to properly understand the implications of Alexy’s argument: if, as in the case of drugs and the sale of sexual services, we are citing an absolute principle of “protection of society” to completely circumscribe and severely punish all conducts relating to those situations (see criterion 4), there are no more fundamental rights in that area.  From the legal standpoint, that means that the constitutional norm of fundamental rights has been violated because no limit has been placed on the erosion of the rights: in proper fundamental rights theory, the actualization of the two principles must be optimized – even when one of them is given ad hoc priority – and one of the extremes must not be completely obliterated as if it did not exist.  There is no democratic society that can justify doing that, as there is no democratic society that can justify accepting discrimination on the basis of race or sex, at least in theory.


As Alexy pointed out a little later (1986, chap. 3, III,,

The more the legal intervention affects elementary expressions of the liberty of human action, the more carefully we must weigh the reasons presented in favour of the basis for the intervention against the individual’s fundamental right of liberty.


The second problem lies in the virtually knee-jerk use of the criminal law, degrading “processual” ceremonies and incarceration because of the close and culturally overlapping “enemy of everyone” representation expressed by the criminal law in the 18th and 19th centuries (that persists today).


For instance, if we decide that Society “is right” to impose a limit, we cease to think more carefully or seriously about the form of legal intervention, or even the style of intervention, that is appropriate in the criminal law (what procedure, what sanction).  Punishment by imprisonment becomes automatic.  In other words, the criminal law loses its sense of boundaries, and even of any moral ideal relating to its own improvement (Pires, 2001).


We must instead construct what Taylor (1985) calls a “principle of belonging” that looks at the importance of individual rights through the importance of the community, and vice versa: each depends on the other.  We should even venture to say that society takes some sort of precedence, if, like Aristotle, we agree that the individual is a “social animal”, or, like the French sociologist Émile Durkheim, that society makes the individual, and not the other way around.  The responsible autonomy of the individual is then no longer confused with individual self-sufficiency.  But on the other side, if society takes precedence, in order for it not to become a form of totalitarianism, and for it to be able to foster the modern individuation process, that precedence must be seen as being for the purpose of creating and protecting a space for its members’ responsible autonomization: the creation of the responsible individual.  And it must not for any reason and in any case take the form of an absolute principle that would derail the rule of law.


As we can see, there are several representations of the potential conflict between individual and society, but some of those representations have now become outmoded and unacceptable in a proper theory of the law.  One of the oldest representations (18th century), and one of the most problematic, is the one that sees the legal world as divided by an odd papal bull, fashioned after the Treaty of Tordesillas.[37]  On the left side of the parallel, we find an area reserved for individual rights.  On the right side, we have the world of Society, which is completely protected by the navy of the criminal law, standing ready to move the border represented by the parallel absolutely, in the name of the safety of the Society.  We have to put an end to this kind of “cold war” between individual and society, which is fought with the criminal law and severe punishment.  The principle of belonging is a more subtle and delicate principle for both sides of the coin: it is a principle of taking care of other people, of “one’s self as another” (Ricoeur, 1990), and of the existence of each and every person.  We cannot move the line without seriously considering other people at the same time.


That is why the first concept of “defending society” that was expressed by the criminal law in the 18th and 19th centuries no longer makes sense in the context of a principle of belonging: we are no longer defending against each other.  We support the mutual obligation to take care of each other, rather than the right of self-defence.[38]  From that standpoint, the criminal law must once again change its identifying self-portrait: it must not be seen as still having to be used, and when it is used, it must not necessarily be used in its most punitive form (i.e., imprisonment).  It must both modify its methods of intervening and find another way to design the portion of the protection of the social relationship that may be assigned to it.


(e) The concept of risk: “rationality of the risk” (Luhmann) and “risk society” (Beck)


We shall conclude on this point with some parenthetical comments on the related problem of risk.  Today, we are learning, at our own expense, that there is no such thing as a non-risky decision.  That is one of the reasons why risk is not a sufficient condition to justify intervention by the criminal law.


It is a mistake to see the decision to smoke tobacco, or smoke a joint, as involving a risk of disease exclusively.  Not only does it also provide a degree of pleasure, but it may also enable the individual to avoid other even more certain risks (depression, boredom, etc.).  In one of his maxims, Cardinal Richelieu said, a bad thing that can happen only rarely must be presumed never to happen – especially if, in order to avoid it, one risks many other bad things that are inevitable and have much more serious consequences (quoted by Luhmann, 1993, p. 12).


From the etymological standpoint, the concept of risk is closely related to the concept of decision.  The word “risk” appeared in the transitional period from the late Middle Ages to the beginning of the modern era (Luhmann, 1993, p. 9) [TRANSLATION] “with the awareness that unanticipated [and undesirable] results may flow from our own activities or decisions” (Giddens, 1990: 30).[39]  We can say that [TRANSLATION] “risks relate to the harms that are possible – but have not yet materialized or are relatively improbable – that flow from a decision; that is, harms that may be brought about by that decision and that would not materialize if a different decision were made” (Luhmann, 1992/1997, p. 133; emphasis added).  Luhmann wrote that “we do not talk about risk unless the consequences may be attributed to the decisions” (ibidem). That led to the mistaken idea that it is possible to avoid all risks and achieve safety if we decide something else.  However, Luhmann goes on to say, the problem is that [TRANSLATION] “any decision may produce unintended consequences”; all decisions are therefore risky, although not in terms of the same consequences.  We can only decide based on the type, extent and probability of the harm (ibidem) that we will accept as a risk and as a danger, because “exposing one’s self to danger is a risk” (Luhmann, 1992/1997, p. 134).


First, we have a question of fact: we take risks every instant that we decide, whether or not we are aware of them.  If we consciously take the risks into consideration before acting, there is also a question of rationality (of the risk), or rational risk management.[40]  This happens when we represent certain negative contingent consequences of our decisions to ourselves, and we assume them, while trying to control the dangers that the decisions produce for us or for other people (“rational decision taking the risks into account”).  Here we see the relationship between the concepts of decision and risk.  Note that the risk that we are necessarily going to take necessarily means a certain (objective) danger for us.  Staying shut up in the house in order to avoid getting mugged will amount to an objective danger of boredom or a nervous breakdown.  Some of the risks that we take can also present as an objective danger for other people.[41]  If we agree to the plan to construct a nuclear plant in a community, there is an objective danger of an accident for the people in that community, and perhaps for others as well.  But not building it involves other risks, and may prevent us from securing certain benefits (more energy, more jobs, etc.).


The paradox is that our culture values the “risk to be taken” by decision-makers (“nothing ventured, nothing gained”, “the sky’s the limit”, etc.) and at the same time has developed a strong intolerance for danger caused by another person.[42]  But we cannot – and must not – think of using the criminal law to avoid risks: it can only assert and support certain normative expectations that are such as will assist in regulating conduct.


Three researchers at the University of Toronto (Lazarou, Pomeranz, Corey, 1998) have estimated that correctly prescribed legal medications kill, on average, 100,000 people per year in North America.  Although for methodological reasons that figure was cut back by one half or two thirds, it nonetheless illustrates the enormous losses of human life that go undetected by any monitoring system, including the legal system.  No one thinks that this danger should be avoided by prohibiting medical prescriptions – the risky decisions made by physicians – or denying the “right to use” medications.  Why?  Because we do not see how that solution could be preferable to the solution of taking risks responsibly.  Knowing that this problem exists, we will try to find other solutions, such as better quality control for the products, etc.  Nor (fortunately) do we consider assigning criminal responsibility to physicians for taking the risk of writing a correct prescription, knowing that even correctly prescribed medications can cause death.


            We often observe the world with the risk / safety distinction in mind.  But we generally know that there is no such thing as absolute safety (Luhmann, 1993, p. 20 et 28). There are other ways of observing that seem to better reflect things, having regard to the increasing complexity of our world.  To get that alternative view, safety and benefits are assumed and sought in any decision made, but they comprise more than a backdrop.  No decision will produce safety (and secure certain benefits) without risk, and it is by taking risks that we avoid certain dangers (while creating others).  In that sense, the concept of risk contrasts, rather, with the concept of danger (Luhmann, 1993, p. 21).


We can say that we also often observe the world with the “risk / danger” distinction in mind.  Note that we cannot think risk and danger at the same time and that these two ways of seeing do not produce the same effects on us.  When we think “danger”, our decision-making universe shrinks, because we have a tendency to stop and withdraw from action.  On the other hand, when we think “risk”, our field of action broadens and our margin of decision-making freedom expands.  We have a tendency to “distance ourselves” from unwanted results, and to decide on the basis of our estimate of our ability to control the risks (whether that estimate is entirely imaginary, or based on the usually accepted parameters).  One of the advantages of observing the world with the risk / danger distinction in mind rather than the risk / safety distinction is that it becomes absolutely clear to us that we cannot avoid risks when we are deciding in a particular way (Luhmann, 1993, p. 28). We can then take the drama out of the risks and better understand the complexity of other people’s decisions (which does not mean that we must necessarily accept them).


            This way of reasoning (risk / danger) is distinct from simple consequentialist reasoning based on the costs and benefits of a final result (final cost / benefit).  Consider, for example, extreme sports.  If the athlete thought in terms of classical consequentialist reasoning (cost / benefit of the possible final result), he or she would not set off on an objectively dangerous ski slope, because a few minutes of pleasure do not compensate for death or serious injury.  If, on the other hand, the athlete reasons in terms of risk / danger, his or her margin of “freedom” expands, because he or she will not stop when faced with the mere possibility of death or accident; the athlete will assess his or her skills, chances and desire, or not, to do it.  If the athlete sees the slope as a danger to him or her, the tendency will be not to do it; otherwise, the athlete will take his or her chances.  Risk reasoning offers different individuals more flexibility: where one will stop, the other will go ahead.  If the individual puts the emphasis on the risk, he or she will tend to forget the dangers; if, on the contrary, the individual puts the emphasis on the danger, he or she will forget the benefits that the risky decision might produce (Luhmann, 1993, p. 24). Of course, this applies to the decisions to smoke tobacco or a soft drug, to use alcohol or eat more or less fatty food, etc.  If we do not understand this, we have a tendency to see those decisions – like the decision to participate in a risky sport – as purely irrational, when, on the contrary, they are made within the framework of the dominant reasoning of our time: we live in a risk-oriented society (Luhmann, 1993, p. 28).


            Although the word “risk” started to become widespread in the 16th century, Ulrich Beck (1986)[43] was right to reserve the concept of “risk society” to refer to a “developmental phase of modern society”.  He said that this new phase came about in two stages.  The first stage began after the 1950s, with the establishment of the post-war “military-industrial” complex; the second began after the 1970s (taking Germany as the empirical reference point), with public discourse focusing on the (new) risks.  We could say that risk reasoning is the older form.  We tend to regard it as having become the dominant way of thinking as early as the second half of the 19th century.  Starting at that point, we can speak of a “risk-oriented society” in the sense used by Luhmann, but there was undoubtedly a major qualitative leap after the 1960s and 70s.  In other words, we can say that the scale of “risk reasoning” (Luhmann) changed with the advent of the “risk society” (Beck).


In order to better see the difference between these forms of thinking for action (final consequentialist / risk) we must note the following.  In risk reasoning, we are, in the relevant cases, making two assessments: one prior to and one subsequent to the harm, where harm occurs (Luhmann, 1992/1997, p.134).  In the subsequent assessment, we review our initial decision to see whether, despite the harm, the risks accepted were reasonable or not.  If we determine that they were reasonable, we regret the harm but we do not regret the decision to accept it.  We considered, in advance, the possibility that the unwanted result would occur and we maintain the conviction that this possibility was minimal or controllable.  So the risk seems, from the outset, to be a contingency, and the materialization of the danger appears as an unfortunate event that does not cast doubt on the rational decision to undertake the action.  We can tell ourselves and others that at the time we made that decision, we did indeed take into account the undesirable effect, and we weighed it carefully.  The decision was and is still “good”, despite the materialization of the “bad contingency”.  Thus an accident in a nuclear plant built after a risk calculation does not necessarily cast doubt on the decision, in the view of the people who made it.  Someone who sees the decision from the outside may regard the construction of the plant as an unacceptable danger, but while the decision-makers regret the result as bad luck, they do not regret their decision.  On the other hand, if our retroactive assessment of the decision is that we determine it to have been “risky” or “too risky”, we may experience what is called, in administration, “postdecisional regret” (Luhmann, 1992/1997, p. 134).


From a cultural standpoint, Luhmann seems to say that we have moved from a society that places the emphasis on danger to a society that places the emphasis on risk.  In other words, there has been a shift from a “fatalistic” society with a worldview based on predetermination, fate, and on reasoning that does not expand our field of action, to a society in which risk reasoning is becoming ever more prevalent.  We could add that in the 18th and 19th centuries we saw the dominance of reasoning that focused on the value assigned to final consequences as the criterion for rational decision-making in some areas of cultural life, and risk reasoning came to be particularly valued after the 1960s.  Table 4 summarizes these two ways of observing risk and the world.

Criterion 4: (fundamental) entrenched liberties 

There is another important criterion that divides the waters between standard prohibited behaviours and two-sided prohibited behaviours.  It relates to the concept of liberty as an entrenched (and not merely protected) legal good

Let us first – perhaps somewhat brutally – visualize this difference.  When someone steals something, the person cannot claim any “constitutional right” to the freedom to steal, even gently and in moderation.  Such a claim would quickly be rejected as absurd.  And it would be even more absurd to imagine a scenario in which an individual came before the court to demand that it order her neighbour to leave her house key in the lock so that the petitioner could exercise some sort of “freedom to steal” as guaranteed by the Charters.  But in societies that have fundamental rights, that kind of argument is not absurd when it comes to two-sided prohibited behaviours.  It is not difficult to imagine a person asking the court to recognize his or her entrenched liberty under the Charters to smoke a joint at home, the right to perform or obtain an abortion, etc.  So what is going on between these two groups of prohibited behaviours?  Why would this seem absurd in one case, and not the other? 

Given how difficult it is to express this idea, we are going to introduce an image to get to the heart of the question, but it must not be taken entirely literally.  Let us say, to begin with, that theft, homicide, fraud, wilful destruction of property, etc., neither take root in nor grow out of a kind of liberty that is entrenched or established by the law, but rather are, so to speak, a sort of flip side of rights; these behaviours are prohibited by a law.  Imagine two “layers of law” superimposed on each other.  In the first layer, you have a right and a corresponding prohibited behaviour.  For example, the right to property and (among many other things) the prohibition on stealing.  These would form the first layer.  Then, you have a second layer where you grant a right of a slightly different type, because it has the status of fundamental rights.  Those liberties and rights are more likely to have to negotiate their position and scope with other rights conferred on other individuals or on the state itself.  For example, the liberty to enjoy privacy and to control one’s own body, the liberty to do anything that does not harm someone else, etc.  In this second later, certain liberties are corresponding “rights/liberties”, that correspond to at least a certain obligation, to all intents and purposes, on the part of the state not to interfere or to interfere as little as possible.[44]  It is to those “rights/liberties”, in the strongest sense of the term, that we are referring when we talk about entrenched liberties. 

Why should this image not be taken too literally?  Because, for example, we can see the right to property as a limitation on the exercise of the right to come and go, or the right to take possession of something that belongs to no one.  In this respect, there is always another right that relates to each right (in complex societies such as ours).  But this fact in no way cancels out the theoretical and empirical value of the distinction.  We do not see a gang of thieves (or professional hit-men) creating a credible charter of rights for themselves, or going before the courts, while sex workers have created just such a charter and two-sided prohibited behaviours are often challenged in the courts based on entrenched rights and liberties.  We are therefore touching on something fundamental here. 

In the case of standard prohibited behaviours, we can say, as the early philosophers did, that the norm introduces a limit on individuals’ natural liberty, that is, on a liberty such as would exist in a hypothetical “state of nature” or in an imaginary society “without law” (which, empirically, does not exist).  For example, in such a society, there would have been no norm prohibiting killing or stealing.  Those norms then appear as a limit on the natural freedom to act.  The image that is of interest to us here is that of a law that is created in a land without law, a no man’s land.  The law does not restrict any prior right, because there are as yet no rights: it limits what Kant would call the “wild freedom” to act.[45]  The norm “do not steal” does not limit some entrenched liberty to steal, but rather protects the entrenched right to have property.  In the classical depiction, the law prohibiting theft simply limits the wild freedom (which would pre-date the law, which would institute “real freedom” from the standpoint of the law: legal freedom).[46]   

On the other hand, in the context of two-sided prohibited behaviours, we are dealing with a restriction on a liberty that is entrenched by fundamental rights, that is, a liberty that is established as such by the legal system itself and that may engender societal debate.  We then get the following questions: should we use the criminal law to limit those (two-sided) rights, or should we simply, where necessary and within clear limits, use another form of law?  And if criminal intervention is accepted, should we allow the use of imprisonment for behaviour the legal value of which is, at least in theory, recognized as such? 

For the moment, we need only point out that two-sided prohibited behaviours present a much more serious problem in terms of the limitation of liberty than do standard prohibited behaviours.  In this case, we are touching on a liberty that is loudly proclaimed by the Charters and that is the strong product of a modern democratic system of law.  We can ask whether the law that limits that entrenched liberty should not be used more sparingly, and be less punitive.[47]

Criterion 5: the strong cognitive dependence / component 

Before we make more systematic observations, we will start with an initial, rougher approximation to clarify this criterion.  

When we consider the question of drugs, we find a host of ultra-specialized pharmacological studies trying to determine the real harmfulness of substances and comparing their effects with the effects of other substances (the effect of alcohol or cigarettes, for example, with the effect of marijuana).  We also see a tendency to consider and address the overall problem of drugs as if all things were equal, and not to recognize cognitively that there are very diverse internal situations within this problem.  And we see quite a lot of stereotypes which relate to, among other things, the nature of the individuals who supposedly “take drugs”, the relationship between crime and drugs, etc. [48] 

Something similar happens when it comes to prostitution.  It is impossible to justify punishing the sex trade criminally if we refer exclusively to what sex workers and their customers do.  What occurs between the parties is a consensual deviance, not a conflictual deviance: the “harm” that some people try to see in it does not stem directly from the social interaction or relationship, as in the case of theft or assault.  We also observe quite a lot of social stereotypes that relate not only to sex workers or to prostitution itself, but also to the “reasons” to continue putting both sex workers and their customers in prison.[49] 

One last example.  When we start to look at certain situations that are criminalized as possibly arising from a “disease” or falling within the purview of the health system, we often develop a less punitive approach to the situation.  This has been observed for a long time in relation to abortion, drugs, suicide, etc. (Schur, 1965, pp. 178-179). As well, as beggars and vagrants [TRANSLATION] “have shifted into the sphere of assistance”, and have “gradually – but definitely – left the sphere of punishment” (Damon, 2002, p. 126).  That means that informed or serious knowledge of the phenomenon (among other things) may change our attitude not only to the criminal law, but also to an entirely different form of coercive legal intervention.  

If we now turn to standard prohibited behaviours, this situation, in which knowledge seems to play a critical role in providing a basis for, or radically altering, how we position ourselves in relation to criminalization of the behaviour by the actual law (or in relation to the overall legitimacy of the criminal intervention) seems not to exist.  And when it exists, it does not present in entirely the same way.  

Let us first take the case of homicide.  We do not see many studies comparing one form of killing with another in order to determine which of the two is the most harmful for the victim and which produces no harm.  Certainly there are studies that, for example, compare deaths from accidents on the job with deaths by standard criminal homicide.  Those studies show that the first form causes more losses than the second.  They show that the losses are, on some occasions, due to failure by employers to comply with safety standards.  They also indicate that the criminal justice system is replaced by other forms of legal intervention and compensation, excluding the use of the crime of criminal negligence.  But those studies do not support decriminalizing standard homicide, or contend that homicide is not harmful or is in the nature of an “individual choice” that must be respected.  

There are undoubtedly a set of stereotypes about “serial murderers”, etc.  They affect how we position ourselves in relation to these cases, and how we think of the sentences.  If we see these murderers as “sick”, we are also likely to give them psychiatric treatment.  But that in no way affects the fact that the nature of the deviance (killing) continues to be seen as wrongdoing and as a clearly negative or deplorable situation.  And the legitimacy of legal intervention to deal with this situation is also clear, even if it is to authorize forcible confinement in a psychiatric hospital for a long period. 

The depiction of theft (or fraud) is different from homicide because, first, the nature of the damage is qualitatively different and, second, it may be broken down into situations that range from a virtually insignificant wrongdoing to much more significant cases.  It is then easy to support social regulation by mutual agreement, or purely spontaneous resolutions, in certain cases where that is possible (return the merchandise, give the merchant a sum of money in a shoplifting case, etc.)  We may also, in some situations, want to implement another form of less “stigmatized” legal regulation than the criminal court (for example, a small claims court, etc.).  And we may even want to amend the punishments provided in the Criminal Code so that, in a large number of cases, the use of imprisonment is prohibited or made particularly difficult.[50]  But we still think that this behaviour is unacceptable and aggravating, even in its minor manifestations.  It is undoubtedly easier to accept our children smoking tobacco or drinking alcohol – even though this may disturb us – than to see them engaging in small insignificant thefts here and there.  We see in theft at least a form of incivility to which we should react, one way or another, and even if the style of the reaction changes.  When we are dealing with theft, we always want to affirm the normative expectation that “you don’t do that”.


We would therefore think it ridiculous to propose that the “occupation of shoplifting” be regulated, for example.  But note that we would not find it ridiculous or absurd – in any event, not in the same sense of the word – to propose to make prostitution an acceptable and regulated occupation.[51]  And we see no particularly compelling reasons to go and ask someone to study whether theft is harmful to one’s health, or regrettable, or harmful in terms of social relationships.  We are quite able to see that – despite all its differences from homicide – theft is still very different from two-sided prohibited behaviours from the standpoint of its relationship to knowledge.


In short, in standard prohibited behaviours, as is commonly said, “we do not ask science to tell us whether the sky is blue”.  In other words, we do not call on knowledge to tell us whether harm has been done or whether there is a conflict between two parties, and the development of credible knowledge is less likely to involve the actual content of the norm, that is, the moral message that this kind of norm expresses.  Certainly knowledge may question and ultimately cast doubt on the social and institutional management of a norm that has moral content (standard prohibited behaviour), and even the type of normative regulation chosen (criminal law / another form of regulation).  It is precisely this point in respect of which a dividing line is drawn to separate standard prohibited behaviours from prohibited behaviours (and certain other prohibited behaviours that are not two-sided but are also heavily dependent on the real or mistaken knowledge that we have about them).


(a) The role of cognitive mistakes in the bases of some criminal laws


We shall now make some more systematic observations.  In his analysis of prohibited behaviours relating to drugs, homosexuality and abortion, Schur (1965) made a very astute empirical observation.  He said that something abnormal occurs in connection with these situations (which we are calling two-sided prohibited behaviours and which he conceptualized as “victimless crimes”).  Unfortunately, he did not place enough value on that observation from a theoretical standpoint.[52]  We are going to try to do this here – in an exploratory and provisional way – still using Schur’s starting point, which is “key” in getting an overall understanding of the problem.  In a sense, it was his starting point that we introduced and developed in our own words in the introduction to this criterion.  

Schur (1965, p. 175) presented this empirical observation in the negative, as we observe it directly in the situations he studied: these three cases of prohibited behaviours, he wrote, are based at least on crucial [cognitive] mistakes as to the nature of the deviant behaviour.[53]  Schur had no difficulty recognizing that stereotyped ideas are certainly not the only support for existing reactions to these problems, but he added that the predominance of mistaken ideas is staggering, and it is particularly interesting to note that in the three cases information about the relatively harmless aspects of the deviance has not received wide attention. (ibidem).  

In order to move this empirical observation gradually to the level of a theoretical statement, we will keep it, for the moment, in its negative formulation, but change it to the following: there do seem to be prohibited behaviours for which criminalization (creation of a criminal law) depends or is based on crucial mistakes as to the actual nature of the deviance.


(b) Three theoretical statements of the basis of criminal laws 

What happens in the case of certain two-sided prohibited behaviours (and probably also in the case of other norms that depend on knowledge)?  Is it possible that some behavioural norms have – or may have – a kind of particular dependence on a bad or good rational, informed or scientific knowledge in order to retain their status as criminal prohibited behaviours? And if so, how is it expressed? 

We are going to break our hypothesis down into three separate theoretical statements.  For the moment – and for the purposes of this paper – we are going to limit the validity of those statements entirely to the interior of the legal system of western societies, starting in the second half of the 18th century:

  1. Some behavioural norms (homicide, theft, etc.) have a weak cognitive dependence or do not need a strong cognitive (rational, informed) component (even where it seems to exist) to provide a sound basis for the creation of a criminal prohibited behaviour.  That is the case for homicide, sexual and physical assault, theft, fraud, wilful destruction of property, etc. (pure or pristine standard prohibited behaviours).  The fact that these prohibited behaviours are rooted in conflictual relationships (stabilized by the social system) is sufficient to make them less dependent on knowledge when it comes time to criminalize them.  To depict them, we may say that the roots of the sound basis for these prohibited behaviours are found in the conflictual deviance in relation to concrete social relationships.  Those “roots” are, so to speak, “pointing downward”.
  2. Other behavioural norms are fundamentally dependent on a strong cognitive (rational, informed) component – whether mistaken or true – in order to be selected and retained as criminal prohibited behaviours that result in deviants being severely punished.  The fate of the sound basis for, or justification for criminalizing, these behaviours, and for applying heavy sentences, depends on a rational or informed knowledge relating to the “reasons or purposes for laying charges” and depends heavily on it.  A change in or challenge to the knowledge is a sufficient condition to fundamental undermine the sound basis of the norm and justify either repealing it or using other less punitive forms of legal regulation.  In one sense, the concrete social relationships between individuals do not need to change in order for this to occur.  To depict this, we can say that the roots of these prohibited behaviours are found in “serious” knowledge,[54] in terms of the relevant knowledge (cultural level).  Those roots are, so to speak, “pointing upward”.
  3. Two-sided prohibited behaviours are, as a general rule, behavioural standards with a strong cognitive dependence or component (drugs, gambling, prostitution, homosexuality, sodomy, vagrancy and begging).  However, keep in mind that there are behavioural norms with a strong cognitive dependence that are not two-sided prohibited behaviours as they have been characterized here[55] and that there may be two-sided prohibited behaviours that do not necessarily have a strong cognitive dependence (such as abortion).


Table 5 illustrates the reversal of direction of the bases for standard prohibited behaviours and two-sided prohibited behaviours.


This theoretical criterion is fundamental.  First, it indicates the presence of two different kinds of bases in the behavioural norms (in the criminal law): norms that are based on – or have as their sound basis – a conflictual interaction, and norms that do not involve this type of interaction and instead are based on the state of serious knowledge (strong cognitive dependency).  This result is based on and similar to the one obtained by Hart (1963), but extends its theoretical reach.  The basis of certain norms in criminal law is not provided (or taken away) solely by pure morality, but also by other sorts of knowledge (religious, scientific, etc.).  Second, this criterion may suggest that there are two kinds of behavioural norms: those whose origin and internal validity, and the length of time they last, are heavily dependent on the state of knowledge regardless of the concrete social interactions between individuals, and those in which all of this depends more on concrete social interactions and the way we think about the regulation of the conflicts observed. The durability and stability of the norms would then vary because of their type of basis.


(c) The two extremes of criminalized prohibited behaviours


We would recall that this criterion, involving a theoretical distinction, is, as such, independent of two-sided crimes, even though it helps in identifying them empirically.  However, the two groups of prohibited behaviours – standard and two-sided – are still the two extremes of the continuum.  Table 6 represents the theoretical hypothesis of the demarcation between two sorts of behavioural norms, based on the role played by knowledge and the hypothesis of the creation of the two radically different sample cases.  The rectangle with the question marks indicates the place where norms or situations that fall into the grey area are, that is, the ones that are neither standard nor two-sided.  Of course, for this purpose, the continuum hypothesis must be maintained.  The oblique dotted arrow from weak cognitive norms to two-sided prohibited behaviours shows that in exceptional cases, a two-sided prohibited behaviour may come to have a norm with a weak or weaker cognitive component.


(d) The process of creating a criminal law, decriminalization and reform of the criminal style


Generally, and to date, in the process of creating a criminal law, no attention is given to the differences that there may be in terms of the soundness of the bases of behavioural norms.  Accordingly, the two broad types of prohibited behaviours (standard and two-sided) may present, at the beginning of a legislative criminalization process (creation of a criminal law) as if they were obvious and of the same nature.  We then do not see that some norms are more fragile than others.  That is, we cannot see that some proposals for the creation of a criminal law do not – as in the case of standard prohibited behaviours – meet the minimum requirements for criminalizing behaviour.  Two-sided situations do not prima facie lend themselves to the criminal law, let alone to the use of incarceration. 


On the other hand, we must keep in mind that there is a wide variety of options when we are considering decriminalizing an action that does not meet the necessary conditions for criminalization.  The following is a general outline of the options for decriminalization and for reforming the style of intervention by the criminal law:[56]


(1) We could, for example, completely eliminate the normative character we had assigned to a situation, a behaviour, or certain aspects of a situation, that is, accept the situation from now on as a cognitive expectation (“there are people who do this and people who do that”).  Certainly, there is nothing to prevent anyone from having pedagogical or personal preferences about certain choices.  We can provide advice, make recommendations, give out information, etc.


(2) We can also retain the norm as a mere moral norm for ourselves.  For example, an individual may decide: “it is wrong to have an abortion, but it is not appropriate to criminalize abortion for this reason or that reason.”  Here, the behavioural norm (or some aspects of the situation) is removed from the legal system but not from the moral system.  Parents do not need a criminal code to rear their children, something we too often forget.  And health professionals do not need one either to treat or assist their patients.  This does not mean that we do not need the legal system itself to regulate their professional practice in some respects.


(3) We can also retain the norm in the legal system, but choose another legal sub-system (not the criminal law) to assert it.  For example, the civil law or a special coercive or administrative law under which imprisonment was not available, and with different procedural rules, could be used.  We would then change the legal status of the norm, as well as the manner in which it was imposed.  That happens when we have learned something new or previously unknown concerning the nature of the deviance or the functioning and social costs of the criminal justice system.  In those cases, our earlier way of asserting the norm no longer seems to us to be justified, acceptable or consistent with other human and moral values that we hold dear.


(4) We may want to keep the norm (or the regulation of certain aspects of the problem) in the criminal justice subsystem, but change the way in which the criminal law deals with the problem.  In that scenario, what will happen is not decriminalization, but internal reform of the criminal law.  In other words, we want to stay with the criminal law, but change the way in which it intervenes.  We can then change the criminal procedure or the available sanctions, or create separate courts (as we have done for young people) to intervene in a clearly different manner (with another philosophy).  We can also simply eliminate the use of imprisonment (at least for certain cases).


            The important point to remember is that all two-sided prohibited behaviours to which this criterion applies exhibit certain specific problems.  (i) They all have a more precarious, more ideological or more fragile endogenous basis because they are not rooted in a concrete, conflictual deviance and because the norms are not sufficiently detached from certain forms of (purely moral or religious) knowledge or are not sufficiently unaffected by knowledge of facts.  (ii) They are therefore more subject to a process of selection from the available knowledge and to the actual value of the knowledge that we select or that is available to us in respect of them at a particular point in time.  That means that a critical and serious examination of the knowledge is of crucial importance.  (iii) They are, to all intents and purposes, more polemical and subject to public debate at a particular point in time, and more likely to be based on major cultural or cognitive misapprehensions.  All of that means that in a democratic society characterized by a process of responsible individuation, criminalizing these behaviours is always surprising and adventurous, particularly if incarceration, with high maximum sentences, is allowed.  We should consider our options carefully before taking this route (or when we come to rethink the matter).


(e) The question of consensus


            To conclude our remarks on this criterion, note that the fundamental distinction here is between the dependence and independence of the norms in relation to certain specific knowledge, and not between consensus and dissent (or “dissensus”).  We might say, doubtless exaggerating somewhat, that the distinction is between a “stand-alone norm” and a norm that is heavily dependent on our adoption of some specific knowledge.


From a theoretical standpoint (and “reasons to know”), the consensus / “dissensus” distinction is a mere ancillary and contingent criterion: a sort of surface criterion.  Why?  Because when we are in a “period of consensus” and we share that consensus, we cannot observe the endogenous fragility of the behavioural norms that have a strong cognitive dependence or component using this distinction.  The “dissensus” may not exist at a particular point, or it may e precarious and marginal, or strong but ignored or even rejected by the political or legal systems.  That greatly limits the theoretical efficacy of the criterion: the consensus / “dissensus” distinction does not allow us to see the problem with the norms until after there is “dissensus”, and there again everything will depend on how we assess it or whether or not we take it seriously.


“Dissensus” takes more time to become apparent and to organize, as well as a lot of energy in terms of social mobilization.  It is also highly dependent on the capacity of the opposition forces to “break down or break through stereotypes” and the false knowledge that is taken as true.  It must also be accepted by the social subsystem that controls the regulation of the problem, or by more than one subsystem at once (for example, the political and legal systems).  “Dissensus” may make the problem visible, motivate change, etc., but it does not pinpoint the theoretical and factual source of the problem.  On the other hand, using the conflictual deviance / serious knowledge distinction, we can identify, from the standpoint of knowledge, the fragility of the basis for the norm, even if we share the belief expressed in the knowledge that supports the norm.  And we can anticipate that the transformation of the knowledge may call the behavioural norm into question.  The significant benefit of this is that it provides an advance invitation to greater wisdom in the legal or social regulation proposed.  It is better to anticipate catastrophes when it comes to violating fundamental rights using the criminal law and imprisonment.[57]  That is why we place less weight on consensus than it is generally given.


Criterion 6: proactive formal intervention


In referring to “victimless crimes”, Schur (1965, p. 171) clearly identified an empirical peculiarity of those crimes that applies to two-sided prohibited behaviours. That peculiarity, which is also of tremendous theoretical and pragmatic significance, is the predominantly proactive aspect in the functioning of the criminal justice system.


What does this mean?  We know today that a large majority of “standard crimes” that come to the attention of the police and provide crime statistics information are reported by way of complaints.[58]  Only a very small number of standard prohibited behaviours are detected in flagrante delicto or by an unreported proactive prior investigation by an individual in the “public” social role or by an agency that is not directly involved in enforcing the law (some cases of corruption that are investigated de officio by special police squads, etc.).  On the other hand, a majority of two-sided crimes (drugs, prostitution, abortion, vagrancy, etc.) are detected at the initiative of the police or local criminal policy programs.


For “victimless” crimes, properly so-called or so regarded by those who commit them, the police have do their own work.  Peter Conrad and Joseph Schneider (1980, p. 120) used the concept of “crime without a complaint” to refer more specifically and unequivocally to this characteristic.  In the case of two-sided crimes, if the authorities want to catch the perpetrators, they must act on their own initiative: they must go fishing, or wait for the rare cases when a third party complains (as happens with street prostitution).  Drug users and their suppliers, sex workers with their customers, beggars who cause disturbance only by their actual presence and who aren’t looking for a room in a prison to spend the winter, women whose doctors perform their abortions – none of these people contribute significantly to the compilation of criminal statistics by filing complaints. 


In the case of certain administrative prohibited behaviours, or certain specific kinds of proceedings, proactive action is therefore a necessity and generally produces more positive effects than the problems or counter-indications.  Consider, for example, highway traffic patrols or environmental protection against polluting activities.  Proactive action may also be very important in relation to certain proceedings that fall into the category of standard prohibited behaviours, as in the case of smuggling, certain underground corruption activities, a nation’s internal security, etc.  But in the case of two-sided crimes, proactive action takes on another meaning and produces a set of disturbing problems in the functioning of the system itself.  There is a strong risk of cost/benefit relationship becoming reversed in this case.  From the standpoint of public policy, we must then carefully ask whether the type of intervention we are engaged in, in these cases, justifies the damage and social costs it also causes.


Nonetheless, we should apply the “proactive action” criterion in a broad sense, to include political pressure on the national or international scene, the creation of specialized enforcement agencies to deal with these “prohibited behaviours without complaints”, etc.  This would seem to be particularly important if we are to discern the scope of the thing in the matter of drugs.  This kind of “proactive action” by the political system in relation to a crime without a complaint or a direct victim seems to be likely to have some novel effects on the criminal justice system.


(a) The theoretical meaning of proactive action for two-sided prohibited behaviours


What is the theoretical and sociological meaning of proactive action in relation to two-sided crimes?  It plainly illustrates the fact that there is no direct conflictual deviance between the perpetrator of the act and a victim: the proactive nature of the formal control mechanism functions, in these cases, as a sort of indirect empirical indicator of the absence of any perception of conflictual deviance between the people directly involved.  No victim, no complaint.  This sixth criterion for double-sided crimes strengthens (by another route) the first criterion for these prohibited behaviours.


Of course, this indicator does not tell us the same thing about all prohibited behaviours.  In corruption cases (standard prohibited behaviour), for example, as we noted earlier, proactive action by the police often simply makes up for the fact the ignorance of the direct victim, who does not see that he or she has been outwitted.  But in the case of two-sided prohibited behaviours, this indicator tells us that the exchanges between people do not result in a perception of victimization.    The theoretical value of the indicator depends on the subject to which it relates.


Nor does this mean that there are never complaints in the case of double-sided crimes.  We all know that some complaints do get made in these cases.  However, they often come from a third party (a neighbour, a family member, an eye-witness, etc.) and sometimes reflect a degree of human misery, or people’s lack of resources or even a disconcerting naivety in relation to the least harmful way of solving a problem.  One example would be the kind of tragic story in which a desperate parent (generally from a disadvantaged background), acting in all goodwill, calls the police to solve his or her child’s drug problem.  A lot of us would never have thought of doing that, and this perhaps indicates that there is indeed a serious problem with intervention by the criminal law intervenes (at least in its current form).


(b) Some social costs of two-sided prohibited behaviours for the justice system and society


Some studies in the social sciences have stressed the fact that proactive action relating to certain two-sided crimes (drugs, gambling, prostitution) produces a number of perverse effects in the functioning of the legal system itself, and even in other social subsystems.  Three negative effects that relate to the system itself are given particular attention: the problem of corruption, the problem of forms of police intervention that border on the illegal (Baratta, 1990, p.167; Hulsman and Ransbeek, 1983, p. 275) and the problem of sentencing by the courts and its effects on the penitentiary system.  We would also like to draw attention to a fourth aspect which is less pointed out less directly, but which emerges in scattered form from empirical research done to assess drug law enforcement policies: a kind of legal inequality that is actively produced by random and variable expressions of political will.  We will briefly address these four points.


Schur (1965, p. 171) suggests:

Another apparent consequence of privacy and lack of a complainant (combined with public ambivalence about the law) is the invitation to police corruption.


This effect, police corruption, may be observed empirically, to widely varying degrees, in a number of western countries.  Sometimes it may spread like wildfire through the business world, the courts and elected politicians.  In addition, there is the concern prompted by the new police powers, which focus specifically on drug law enforcement,[59] which in turn justifies the new powers.


That is not all: pressure for severe punishment of some two-sided prohibited behaviours may destabilize the criminal justice system, both in terms of policies on when charges will be laid and sentencing and on the cognitive level (theories, legal justifications, assessment and decision-making criteria, etc.).  Here we are referring to a kind of subtle distortion, a loss of quality in the professional practice of the law, and a loss of rigour in legal reasoning that is very difficult to assess and demonstrate.  These cognitive problems may also arise elsewhere in the system among other actors who, at various levels, must then invent all kinds of justifications in order to accommodate to situations that are logically untenable.  For example, we have seen a huge rise in the severity of sentences since the 1970s, and that is in part attributed to the extreme pressure for disproportionate punishment for drug offences and the fact that severity has become commonplace, or that this punishment has had a domino effect on the sentences handed down for other offences.[60]  We expect that a judge who starts to increase sentences for drugs would also think about raising sentences for other offences.  And the legal good of “liberty” has begun to be eroded.  As Charras (2002, p. 106) recently observed, if we look at statutory maximum sentences, [TRANSLATION] “in the western world, there is virtually nothing other than terrorism, treason and first-degree murder that is punished more severely”.


As a final point, another problem with two-sided crimes is that they are the source of social inequality that is actively produced by the proactive policies implemented by both the political system and the criminal justice system.  What do we mean by this?


The legal system and the political system must necessarily interface at a number of points.  Those interfaces take concrete form, first, in the criminal policy guidelines given to the police by way of directives that come from one government department or another, depending on the political organization of the country in question.  The police also have sufficient autonomy (which is a positive thing in itself) to adopt their own policies.  And second, the political system may try to control the legal system by allocating resources, thereby creating one program or another, targeted more or less appropriately in different cases.


The problem of two-sided prohibited behaviours is both that they are the most sensitive to these law enforcement policies, since enforcement depends largely on proactive action by the system, and that they are where enforcement is most likely to (proactively) penalize, or, conversely, relieve, human misery.  A policy that the executive crafts “wisely” will then create “good legal inequality” in the sense of moderation (as compared to times or places where no such policy existed): it enables a government to suspend criminal intervention that produces pointless suffering, or to create alternative intervention programs that are less – or more – punitive, to deal with a particular problem or for the benefit of a particular group of individuals.  On the other hand, a “zero tolerance” policy, or “cleaning up the city”, often results in the legal system (with very little capacity for resistance) being induced to engage in the active production of a new “bad legal inequality”, one that is likely to have the worst effects on the least harmful prohibited behaviours, and the ones most closely associated with human misery.


As a final point, from the standpoint of reforming law and practice, empirical studies have repeatedly shown that without a clear new legislative policy, exhortations to law enforcement institutions to use their discretion not only produce striking differences in how things are dealt with, but most importantly have insecure and unstable results (Charras, 2002, p. 109).


Criterion 7: the criminal law as producing effects that counteract the prohibition


Here, we will consider the last criterion for distinguishing these types of prohibited behaviours; it is not easy to formulate, and it is still being studied.  We must therefore proceed with caution, having regard to the incomplete and exploratory nature of this distinction.


To assist in understanding it, let us suppose that we are considering using the criminal law to prohibit a behaviour.  The question that the legislature must ask is then this: “Is there a risk that using the criminal law to prohibit this behaviour will create social problems no matter how the law is enforced?”  We may then have two situations, in empirical terms.  In the first, the norm that we want to impose on other people and on ourselves will get the green light prima facie.  That means that the problematic effects that this behavioural norm may create will depend mainly on the norms that apply in respect of procedure, punishment, etc.; in short, the way in which we react to enforcement of the law.[61]  In the second situation, the norm that we want to create will be facing a yellow light.  That means that we anticipate that it might, by itself, create problematic effects that cannot be remedied using procedure, sentencing or enforcement methods alone: the norm is an integral part of the problem.


Table 7, which will be explained in more detail later, illustrates these two situations.  It shows that the enforcement of any criminal law creates problems, but that in the type 2 situation, those problems concern or relate more to the actual existence of the prohibition.  In the latter hypothesis, the effects produced by criminalization of the action will place the idea of creating a criminal law norm to regulate this behaviour squarely in issue.[62]  At least two broad types of problematic effects may occur here: they may be counter-productive, that is, criminalization may counteract the goal of more effective, more rational, more human and less expensive control of the behaviour; they may produce a contradiction in terms of values, including for the legal system itself.  In this case, criminalization creates new empirical situations that result in the system producing effects that it was not predisposed to create, effects in terms of increasing human misery, exacerbating a social situation that already disadvantaged a group of individuals, etc.  Of course it is not necessary in this case that criminalization create the new situation out of whole cloth: it is enough that it institutionalize a problematic social situation.  These negative effects must nonetheless arise directly from the criminalization as a whole (from the existence of the behavioural standard in the criminal law and from the action by the system) and not merely from the form in which the norm is enforced.


Table 7: Tentative formulation of the criterion:

Criminal law as not producing / producing effects that counteract the prohibition


Scenario 1:


The criminal law as not producing effects that counteract the prohibition


Scenario 2:


The criminal law as producing effects that counteract the prohibition


Norm prohibiting the conduct


(1) “Killing is prohibited”

(2) “Stealing is prohibited”


[splitting of norm and enforcement]

Norm prohibiting the conduct


(1) “Abortion is prohibited”

 (2) “Using drugs is prohibited”



Problems associated with the forms in which the law is enforced


Problems associated with the existence of the prohibition and with the forms in which the law is enforced



Pragmatic aspects and problems associated with the forms in which the law is enforced




Sui generis pragmatic aspects and problems:

The law may be counter-productive and/or produce new empirical situations that are contradictory in terms of values




(a) The first scenario


In the first scenario, the problems that may arise when the law is enforced do not place the idea of creating a criminal law to assert the norm in issue.  For example, let us assume an imaginary society that is creating the criminal law norm “killing is prohibited” for the first time.  To simplify the demonstration, imagine as well that the justice system in that society is very enforcement-oriented and equipped with extraordinary and formidable social control mechanisms.  The sentence provided is ultra-exemplary: torture followed by public execution.  In addition, the entire population is compelled to witness the ceremony, by means of a highly-developed electronic system: people must position themselves in front of a television screen with a bracelet that identifies them individually, etc.  As well, the police and the justice system are 100% efficient: one week after each murder, the perpetrators are arrested, and they are sentenced and executed about a month later.  Despite this, during the thirty years that follow the making of this law, the number of murders continues to climb every year.  Some people will then say that all of this cruelty is pointless and the punishment is disproportionate or downright inhuman and unacceptable under any circumstances.  Other people will take a position opposite to these statements, and argue that were it not for all this the number of murders would be even higher, and we must think up more horrible forms of torture, etc.

Note that in this debate no one has directly called the norm “killing is prohibited” into question.  There are some criticisms that would have very little chance of being heard, or would be regarded as bizarre.  For example: “we must decriminalize homicide because the law is not cutting the number of murders”; “the creation of a criminal law prohibiting homicide is causing the number of homicides to rise”; “we have to legalize some forms of homicide in order to avert the creation of a black market in murders that will jeopardize the purpose of the law to prohibit murders”; etc.


What are we observing here?  First, there is a very clear split between the problem of criminalization de jure and the problem of enforcing the norm (including sentencing).  Second, the split is such that the norm may apply to us and to other people regardless of the empirical results it produces and our pragmatic expectations.  The empirical results do not persuade us to reject criminalization, that is, the idea of imposing the norm on other people.  And if our cognitive expectations regarding crime are not met (crime continues to rise) the norm is also not called into question: we can attribute the norm’s lack of success to problems associated with the enforcement of the norm, or to an external causality (changes in society, etc.).  Third, that split means that the norm (“killing is prohibited’) may be isolated and protected from problems associated with the manner in which it is asserted and enforced.  We may then deal separately, and completely freely, with the practical question of enforcement without placing the norm in issue.  The undue suffering caused by enforcement of the norm can be minimized or avoided simply by changing the procedure, the sanctions or the overall form of our reactions to the deviance.  For example, we may retain the criminalization (de jure) of the behaviour and institute other forms of dispute resolution in order to decriminalize it de facto, that is, by treating it differently.  The ability to separate the norm from the way it is enforced allows us to “value” the norm a particular way: it has value by itself and we can protect it from the negative effects caused by the way we react by modifying our way of reacting. 


In short, in the first scenario, the decision to create or retain a criminal law (criminalization de jure) is not affected by either the failure to achieve our pragmatic expectations or the observation of the problems created by enforcing the law.  Even the conviction that the norm has failed in practice[63] does not persuade us to abandon it.[64]  We can then say that the criminal law does not in this case create special problems that make criminalization illogical, or problems that result in our acknowledging that we should not impose it on other people.  In other words, in the first scenario the problems do not result in our directly acknowledging, for example, that “decriminalizing the behaviour de jure is preferable to criminalizing it”, or “it is better to authorize the behaviour and establish forms of legal limits other than outright prohibiting it by a criminal law”.


(b) The second scenario


In the second scenario, the opposite occurs: the problems are in fact created by the complete criminalization of the behaviour and not only by the law enforcement.  The behavioural norm is an integral part of the problem.  It is then not sufficient to change our method of reacting: we must actually tolerate and accept the behaviour in order to solve the problems caused by criminalization.  When we tolerate an action de facto that corresponds to the second scenario, that tolerance means more than simply accepting (or tolerating) another way of reacting or solving the problem: it relates to the actual existence of the law-breaking behaviour.  From the standpoint of a principle that a moderate reaction is to be preferred, de facto decriminalization will be regarded, in the case of two-sided prohibited behaviours, as being preferable to actual criminalization, although it is still insufficient: mere de facto decriminalization, in the case of these prohibited behavoiurs, is a precarious and inequitable measure that is likely to lead to arbitrary exceptions.


Take the example of abortion.  Criminalizing abortion creates a lot of special (a particular type) and novel practical problems.  For instance, some women will be prompted by a variety of circumstances (pregnancy resulting from rape, cultural pressure, etc.) to want to terminate their pregnancies and to use the services available on the black market.  Women who come from disadvantaged backgrounds are particularly often victims of situations that are horrible in human terms.  Their decisions to abort may even be made despite their own primary wishes, and be due to force of circumstance.  The prohibition on abortion then automatically creates discrimination, with significant human consequences: women who have fewer financial resources (or less relationship capital) cannot obtain high-quality services on the black market, that is, services that involve a lower risk to their own physical security.  That is what we mean when we talk about a law that “produces new empirical situations that are contradictory in terms of values” (Table 7).


Let us hope that these comments will be sufficient to be able to visualize the specific nature of the problems directly created by criminalization in this case (or by prohibition, period).  Recognition of these facts may result in our being morally both against abortion (individual morality) and against criminalizing abortion (social ethics).  We will then go on to call for positive measures so that a woman is not prompted to have an abortion: birth control programs, policies to provide families with financial aid, etc.  At the most extreme, we will disapprove personally of any decision to seek an abortion, but we can also recognize that the hardships involved in that decision must not be imposed on everyone by a criminal law.  In the circumstances, the fact that the law will never completely prevent abortions from being performed is a factor against the law itself, because that means that the special and novel consequences of the law will be just as perennial.  First, the law will go on producing a sort of social inequality, and problems that are disastrous and intolerable, for some women, on a regular basis, and second, it will not prevent abortion.  When those two aspects co-exist, they bring another dimension to the observation of the ineffectiveness of the law to counteract the undesirable phenomenon: in these circumstances, that observation may be more easily cited as a reason to question the law itself.


When the legal system enforces a criminal law, it is predisposed – in certain circumstances (which may be more or less exigent, depending on the point in time and the norms of the system) – to agree that some sanctions will “offend” the values the system protects.  For example, if the system is going to incarcerate a hired killer, it is predisposed to agree that a sentence of imprisonment “offends” the legal value liberty.  In this case, paradoxically, there is no contradiction in terms of values, or the contradiction exists but does not have at all the same meaning as in the case of prohibiting abortion that we have just cited.  Let us now see where this takes us: we are also predisposed to acknowledge the need to incarcerate a hired killer in order to prevent the killer from engaging in his or her trade, but we may feel very uneasy about prohibiting abortion when the prohibition has tragic impacts on women’s personal security (particularly women from disadvantaged backgrounds) and when we also find imprisoning a woman for obtaining an abortion objectionable.  And what happens is that we become predisposed to criticize the law itself for its ineffectiveness in combating the phenomenon.  


Abortion has often experienced a sort of de facto decriminalization,[65] but in this case that does not simply mean choosing another method of social control of abortion; it means tolerance of abortion itself.  Compare that with theft.  It is unlikely that we will see tolerance of theft develop.  De facto decriminalization of theft would mean putting another way of dealing with theft in place.  Of course, we can simply give a warning, express disapproval of the behaviour in words rather than deeds, forgive it in certain circumstances, etc., but we will not go so far as to question the norm that “no one must steal”, and we will continue to hold that the norm must be valid for everyone.


What do we observe in this second situation, then?  First, that there is not as clean a break between the norm and how the norm is enforced.  The two things seem to go hand in hand, and the problem created by the prohibition in itself cannot be satisfactorily resolved simply by changing how we enforce the law.  Second, in these cases, there may be two kinds of special, strong pragmatic expectations: a positive pragmatic expectation of certain concrete results, [66] or – and mainly – negative expectations that are both cognitive and normative, that we do not intend the law, by its very existence, to produce certain infringements of other core values.[67]  Third, our attachment to that norm, as a norm of the criminal law, is not independent of certain effects that it should not produce.


In short, there is no split, in the second scenario, between criminalization de jure and the enforcement of the criminal law because enforcement will generate effects that are counter-productive or that create a contradiction in terms of values.   Those values may prompt us to acknowledge that we should not use the legal system to impose that sort of norm on others, let alone by using criminalization.  Of course, we can keep the norm for ourselves, apply it to matters under our own control, within the family, etc., but not generalize the prohibition to other people.  In this case, we will say that “criminalization is a mistake”, that the “law (not just enforcement of the law) is a cure that is worse than the disease”, and that “the desire to criminalize produces a result that is contrary to the goal of the law itself”, etc.  We will now say that what we have here is a special dimension, or a pragmatic problem sui generis.


(c) Two additional comments on the criterion


First, let us say that this criterion enables us to distinguish some two-sided prohibited behaviours from standard prohibited behaviours, and it also has the special characteristic of dividing two-sided prohibited behaviours internally into sub-groups.  This means that it functions in two directions: toward the outside of the group of two-sided prohibited behaviours, to separate them from standard prohibited behaviours, and toward the inside, to identify distinct kinds of empirical situations among two-sided prohibited behaviours.  For the moment, it seems clear to us that this criterion applies to the cases of drugs, gambling, the sale of sexual services and abortion.  Nonetheless, we shall focus on illegal drugs here.


Next, we must be aware that the only way that we can apply this criterion with certainty is retroactively: it is by using the experience of disappointment in respect of the enforcement of the law that we can see what we are going to attribute the problematic effects to.  But it is sometimes possible to anticipate the sui generis effects mentally, by making comparisons to other similar situations that have already been experienced, or by simple conjecture – for example, by looking at what happened with the complete prohibition on the use of alcohol in certain countries, we are able to anticipate problems of the same type if we prohibit drugs.


(d) Two indicators of the sui generis pragmatic dimension in the case of drugs


            Two empirical indicators of the existence of a sui generis pragmatic dimension may be found in learned discourse relating to illegal drugs.  The first lies in the fact that we have a tendency to assess criminal intervention the same way that we assess any other social program designed to effectively change a situation.  We want to know whether the promised effects have actually been achieved, whether other effects (desirable or undesirable) are being produced, and whether the methods adopted are relevant and effective or directly contrary to the intended goal.  We do not assess the performance of all agencies involved in enforcement activities relating to standard crimes in entirely the same way.  The second indicator consists precisely in the existence of a challenge to the criminal behavioural norm based on an analysis of the real nature of the problem, and also on an assessment of legislative policy that criminalizes certain drugs.  This is what a committee of the New York Academy of Medicine had to say on the question of drugs in 1955:

There should be a change in attitude toward the addict. He is a sick person, not a criminal. … The Academy believes that the most effective way to eradicate drug addiction is to take the profit out of the illicit drug traffic.  (quoted by Schur, 1965, p. 156)


First, the first two sentences in this quotation show that the authors have changed the type of expectation: they have gone from a normative expectation (criminal law) to a cognitive expectation (science).  Second, in the context of that cognitive expectation, they want to effectively solve what is perceived as a practical health problem (“eradicate drug addiction”).  Note that this is learned discourse, which finds its sources in the health care system.  This is also an old discourse, which may explain why the authors still thought that the behaviour could be eradicated.  Third, we must also see that the problem is limited to drug addiction, and the solution advocated relies solely on the criminal law.  The focus is more on caring for individuals than on criminalizing drugs.  In this case, the preferred social technology is not likely to produce the same mistakes and suffering as if it had upheld criminalization of the behaviour.  And fourth, the report points out that at that time criminalization was already producing a counter-productive effect in terms of solving the problem.


Let us now look at another message, nearly fifty years later, which is generally to the same effect but places more emphasis on some of the negative effects of the legislative policy of criminalizing drugs:

[TRANSLATION] Thus it has been pointed out that the politics that have been followed for the last century had failed to stem the development of drug addiction, that anti-drug laws were more toxic than the substances themselves and endangered public liberties, and that the prohibition contributed to the success of the underground economy, or even that it was counter-productive in terms of health care. (Charras, 2002, p. 108)


In that second passage, it is clear that research is assessing the policy of criminalizing drugs and finding it to be counter-productive, as a cure that is worse than the disease, and that criminalization per se is not justified.  Criminalization is found (even more explicitly than in the first quotation) to have been a failure, in two senses: it has produced new and undesirable parallel effects, and it has not counteracted the phenomenon.  The repetition of this kind of pragmatic comment suggests that there is something specific in the case of drugs, and that we should understand that case differently from the way we understand standard prohibited behaviours.


When it comes to homicide, theft, fraud, etc., we would be very surprised and disconcerted to hear learned discourse directly questioning the value of the criminal law norm that prohibits the behaviour and that continues to affirm that people must not kill, steal, commit fraud, etc.  We can see this difference both in the narrower meaning placed on the concept of “failure” here and on the reaction to that failure.  The failure – or unsuccessful portion – of the standards is attributed exclusively to enforcement of the law or to transformations in social life, and not to the behavioural norm itself.  Accordingly, we do not use the ongoing incidence, or any rise in the occurrence, of homicide, theft or fraud to question the norm that says that these behaviours are unacceptable and prohibited.  The heuristic content of the norm as “valid for everyone” stands intact, even if enforcement agencies are not able to reduce the incidence of the phenomenon.  We still hold to the norm, despite its failure or regardless of the actual success or failure of the formal social control.  And most importantly, we seldom suggest that the norm that prohibits the behaviour is, itself, at least in part, directly responsible for the worsening of the situation.  Thus in the criminal law we have behavioural laws (i) that we do not make in order to learn from reality”[68] either in terms of their on-going lack of success in respect of behaviours or in terms of their general social consequences, and (ii) that are not directly called into question by problems relating to enforcement.


If we examine the quotation from Charras supra, we see that the assessments may lead to calling the actual criminalization of drugs into question.  They indicate that criminalization has worsened the situation in fact.  In other words, they say that the means are contrary to the end. Those norms are accompanied by strong cognitive expectations, and they have the capacity both to be “counter-productive” (as a result of their interaction with other social variables) and also to create a “contradiction” from the standpoint of core values.[69]  We shall now see how criminology has clearly identified some problems that are directly associated with criminalization.


(e) Two types of problems associated with drugs


Criminology has created two new concepts to address sui generis pragmatic problems.  Hulsman and Ransbeek (1987, p. 272) clearly distinguish between two categories of problems: the primary problems of drugs – which [TRANSLATION] “are associated with the drug use and arise for the user personally and for the people around him or her” – and the secondary problems of drug policy, which affect not only the user and the people around him or her (stigmatization, etc.), but also the justice system, the drug market and society as a whole.  By this, they mean that criminalizing drugs not only fails to solve the primary problems associated with drug use appropriately, but also creates other problems that are then added onto the primary problems and in some respects form a symbiotic relationship with them.


In a similar vein, there are the equivalent concepts of the primary effects and secondary effects of drugs advanced by Baratta (1990).  On this point, he wrote:

[TRANSLATION] Based on a significant research approach, we mean, by “secondary effects” of drugs, the effects due to criminalization.  The “primary effects”, on the other hand, are the effects associated with the natural properties of psychotropic substances, regardless of whether the use of the substances is criminalized.  (Baratta, 1990, pp. 163‑164)


On the question of standard prohibited behaviours, when we use the same concept of secondary effects of criminalization, we think less of the criminal law as such than of the way in which it is enforced (promoting stigmatization, etc.).  We can then separate the examination of the value of the norm from the examination of the appropriate way of reacting.  First, the behavioural norm is in this way “protected”, and second, we can consider reacting in the form of alternatives without compromising the norm, because in any event, all alternatives will have the effect of affirming and stabilizing the norm.[70]  Of course, that does not mean that all reactions are equally acceptable or reasonable according to other criteria, but simply that all reactions, from the most punitive to the most moderate, have the capacity to affirm the norm.[71]


On the question of illegal drugs (and other similar situations), prohibition by the behavioural norm is regarded as a direct source of secondary problems that are observed and considered to be unacceptable.  At this point, the norm cannot be separated from enforcement of the norm.  Criminalization can be seen as a “mistake”.  We must therefore be especially vigilant in these cases and we must, to the extent possible, avoid a legislative policy that uses imprisonment.



            In conclusion, we are first going to do as concise a review as possible of the principle results of this discussion paper, and make a general recommendation that follows from them.  Then, in closing, we will offer a few comments on the role of the “goals” that we want to achieve by applying a public policy involving the criminal law.  The basis for doing this is that the criminal law is often used, in legislative or judicial policy, as a way of “producing happiness” or even (less commonly) of “reducing suffering” on the part of the virtual victims, and this creates a set of major paradoxes.


(a) Summary of results and general recommendation


The following are some results, ideas and guiding principles that come out of this discussion paper regarding the contribution that philosophy and sociology make to the criminal law and also the contribution of criminology and the theory of the law itself to a legislative policy on (soft) drugs.


1.         Prohibited behaviours involving soft drugs belong to what we call the group of “two-sided crimes” and are distinct in a number of aspects from the standard prohibited behaviours (homicide, assault, theft, fraud, etc.) which characterize the criminal law.  In general, two-sided prohibited behaviours:

a.       criminalize behaviours that are fair and well-organized between individuals (criterion 1),

b.      in large measure punish human misery, that is, individuals who suffer the effects of their own actions, or who are trying to find a better balance or less suffering in what has often been a less fortunate life (criterion 2),

c.       violate a parameter that may be seen today as a matter of individual preference and not necessarily or equally negative for everyone; they are tainted by a paternalistic approach that may be regarded as extreme, with no clear limits (imperium paternale) and incompatible with the principles of the contemporary (criminal) law and with the construction of individual autonomy, an approach that compels individuals “to be happy in a certain way” (Kant,1793), even when they do not want to do themselves harm and their lifestyle choice does not harm “the freedom of everyone can coexist together with the freedom of everyone in accordance with a universal law” (Kant, 1793) (criterion 3),

d.      excessively restrict the legal liberties instituted by the charters of human rights, in the sense that they do not comply with the principle of optimizing all fundamental rights which requires that balance between “principle of the collective welfare” be weighed against the “principle of individual welfare” must always seek to optimize both of the opposing extremes in establishing their mutual legal limits (Alexy, 1986), and not completely cancel out one of the extremes in favour of the other (criterion 4),

e.      often have a fragile and shifting philosophical and legal basis, in so far as that basis depends on our belief at a particular point in time in certain types of knowledge (political, religious, moral, scientific, professional, , etc.), a basis that is not rooted in the actual nature of the concrete social relationships between individuals (criterion 5),

f.         open the door too wide to institutionally-created social inequality – and to major social costs for the legal and political systems themselves – in so far as these prohibited behaviours are “crimes without complaints” (Conrad and Schneider, 1980) that are very sensitive to the local policies of each organization and the resources allocated to less coercive alternatives (criterion 6),

g.      produce special perverse effects that call into question the actual meaning of a criminal law prohibition, because these criminal laws produce new empirical problems that are counter-productive (impossibility of regulating certain harmful products on the market, stigmatization and punishment of persons who need help, etc.) or that produce a contradiction in relation to other values (punishing an action as something harmful that is the least harmful for some individuals, or even not harmful at all, or is a less risky form of pleasure than some other legally created actions, etc.).


Principal recommendation:         


For all these reasons, two-sided prohibited behaviours are not suited, in principle, to being made the subject of a criminal law prohibition and, very specifically, should not allow for sentences of incarceration (let alone a severe maximum sentence).


Any criminalization of these behaviours (or other form of coercive restriction) must be strictly circumscribed to as to apply to a special situation in which the aim is to point out a tangible and direct risk to someone else (for example, driving while intoxicated) or to affirm the principle that underaged children are not to be exploited (for example, forcing them to sell sexual services).  In all these cases, there should be a range of sanctions, preferable not including incarceration, and the prohibited behaviours should not be subject to minimum sentences.  In addition, we should resist devaluing liberty and avoid inflationary maximum sentences.  Lastly, where certain behaviours are prohibited, it is advised that we adopt extraordinary measures to affirm the value of the rights of the individuals involved in those behaviours to be reintegrated into society (for example, when it comes to criminal records and access to employment and publication of their names in the media, particularly after sentencing, etc.).


2.         With respect to the criminal law, we should consider having a sort of principle of the primacy of national fundamental rights over international agreements.[72]  That principle would hold that requests for reform or for legislative policy (creation and enforcement of criminal laws) set out in international agreements would not take precedence over national legislative approaches that apply fundamental rights norms to make the criminal law more moderate and, increasingly, an instrument of “last resort” (Law Reform Commission of Canada, 1976). In fact, national fundamental rights today often exist in tandem with international charters of fundamental rights that set out equivalent norms.  This principle of the primacy of national fundamental rights has already been recognized, to some extent, in international agreements themselves, when the parties stipulate that the agreement must be applied “subject to the constitutional provisions” of each signatory.


3.         The decriminalization of soft drugs does not operate to create a sort of “right to drugs”, as is sometimes said in the legalization/prohibition debate.  In fact, there are at least two methods of de jure decriminalization but neither of them produces entirely what that expression sometimes suggests.


a)       Decriminalization de jure, in the weakest meaning of the expression, consists simply of replacing the criminal law with a special non-criminal law[73] under which incarceration cannot be used and a criminal record is not created, and that often provides for different procedural rules.  In this method of decriminalization de jure, the behaviour continues to be regarded by the legal system as a prohibited behaviour, but it is no longer a criminal prohibited behaviour.  There is a transfer from one system of regulation to another.  In short, decriminalization in this case simply means that, for various reasons, the state and the legal system have decided that the phenomenon should not be dealt with as a criminal prohibited behaviour.

b)      Decriminalization de jure, in the strongest meaning of the expression, recognizes and applies a sort of established liberty in the strict sense, that is, an area from which the criminal law withdraws with a corresponding obligation of non-interference (at least) on the part of the state.


Even in the second scenario, there is no “right to drugs”, properly so called, in the sense of a right that individuals may invoke as against their teachers, parents, friends and family, etc., let alone an automatic right to a positive service in the nature of “receiving drugs”.  When the corresponding obligation of non-interference relates exclusively to the state, other social institutions may create norms to regulate conduct.  For example, the state may not interfere with the behaviour of smoking tobacco in public and private places, but a business may prohibit its employees from smoking on the institution’s premises.  Note that the state may still prohibit the company from discriminating against tobacco smokers in its hiring policy, on the ground that this would be an interference with individual liberty.  What we have here is a sort of “right/liberty””[74] (se emphasize the word liberty) accompanied by this reinforcement of a corresponding obligation of state non-interference, or even a degree of legal protection against improper private interference, because the “right/liberty” is supported by fundamental rights (freedoms established by those norms).


Carbonnier (1963), a French sociologist of law, labels these established spaces of liberty (among other things) “non-right” spaces, because they are often created by a “self-limitation on the right”.  The expression “non-right” is useful here for two reasons.  The first is that it allows us to visualize the fact that in this case decriminalization gives a liberty more than a right, which is a not insignificant point in the debate about drugs, which often takes a highly emotional turn.  The second is that it points to the absence of a state right of interference, which is a consequence of this legal liberty.  Of course, the only disadvantage is that it does not clearly indicate that this non-right area is, in reality, a sort of “reinforced right/liberty”, that is, a parameter for liberty that is reinforced as a liberty by fundamental rights.[75]  In this case, decriminalization means that the state and the legal system have realized that the criminalization process went too far and that the fundamental law norms (or optimization of those norms)[76] called for the creation of this reinforced perimeter of liberty.


This in no way prevents there being other forms of normativity (moral norms, pedagogical norms, etc.) within those areas.  With respect to certain aspects of the problem, there may also be withdrawal by the criminal law accompanied by coercive legal regulation that does not use incarceration, for example when the aim is to control, by law, the production and sale of a good that was previously a criminal offence.


(b) The cognitive trap of “goals” in using the criminal law


We know that the concept of “goal” (end outcome) is closely connected with a teleological action that tells us “what must be done”.  As such, however, it fails to elucidate (dark side) the question of how to do what must e done when more than one option is available.  Curiously, the concept of goal is also closely connected with justifications: “the why (we believe that) a particular thing must be done”.  As such, the end outcomes comprise a virtually unlimited inventory of choices of justifications, in that they may be stated post factum, that is, retroactively, to validate the means actually adopted.  We shall see that end outcomes present a set of problems in the criminal law, precisely because they are all laudable.


There are at least two positions we can adopt in relation to a reform of criminal law norms.  In the first, we want to use criminal law norms as methods to accomplish valuable outcomes, or to produce happiness.  In that case, the criminal law is regarded mainly as an instrument for obtaining certain results (end outcomes) in the longer or shorter term.  For example, we create a norm to protect health, to reduce, traffic accidents, to denounce a behaviour, etc.  Everything happens as if we were not looking – or not looking enough – at the actual “quality” of the institution that is used as the means.  This happens because we have our gaze fixed on the goals.


If we take the second position, we try to reform the criminal law in order to construct a criminal law that is better without regard to a specific outcome other than the one of having a law that is more respectful of liberties and of all of our human values (including forgiveness, solidarity, reconciliation, respect for individual autonomy and the dignity of all human beings, etc.).  The criminal law is really regarded not only as a remedy, but also and at the same time as a problem.  From that perspective, criminal law norms are considered less to avoid suffering by certain individuals in future or to protect a particular good (protecting children, health, etc.) than to reduce suffering by everyone, gradually, in the present – and to the extent possible – by implementing a reform that also regards the criminal law as a goal.  In short, there is less difference (or discrepancy) between short-term goals (means)[77] and long-term goals (outcomes) in this case.


To the first position, the criminal law is simply a means; to the second, the law is a means that has the value of a goal.  In the first case, the expressions “battle (or war) for” and “battle (or war) against” take on a meaning similar to the military meaning; in the second, those expressions take on a meaning similar to the idea of “war against poverty”.  In the first case, the value of the goal does little to control the disadvantages of the means; in the second, because the means are more certain results than the desired result, there is no haste to accept them if they run the risk of producing suffering. 


We are going to draw on some of the thoughts of Karl Popper (1945/1962, chap. 9);[78] in a work whose title is significant, The Open Society and its Enemies, he distinguished between two sorts of social engineering.  However, we are going to adapt his comments freely to our objectives, since what we have in our sights is not entirely the same as what he had in his.  Popper was, at that time, concerned about projects that, in his view, suggested a kind of totalitarian or dogmatic intention, particularly – but not exclusively – if their ultimate goal was a thorough-going transformation of society, as in the case of Marxism.  However, as some sociologists have noted,[79] Popper’s ideas are much broader in scope, and may be applied to ordinary projects and trends for reform in democratic societies.  This is the aspect of them that interests us here, provided that we can adapt them to the very specific problem of the criminal law.


Because Popper had Marxism particularly in mind, he called those two social engineering approaches utopian engineering[80] and piecemeal engineering.  We will avoid assigning labels with a pejorative label, because the appropriateness of these approaches may change completely from one field to another, or one type of object to another.  Nonetheless, the first model of social engineering (what Popper calls “utopian engineering”) is the one that presents the most problems when the criminal law is involved.  Popper’s second model, however, if it is left unchanged, is not a real alternative in the criminal law.  For that reason, we will tweak it somewhat here.  But first, we will look at how Popper presents these two models:

The Utopian approach may be described as follows. Any rational action must have a certain aim. It is rational in the same degree as it pursues its aim consciously and consistently, and as it determines its means according to this end. …

[A]nother approach to social engineering, namely, that of piecemeal engineering … is an approach which I think to be methodologically sound. The politician who adopts this method may or may not have a blueprint of society before his mind … . But he will be aware that perfection, if at all attainable, is far distant, and that every generation of men, … have a claim; perhaps not so much a claim to be made happy, for there are no institutional means of making a man happy, but a claim not to be made unhappy, where it can be avoided. They have a claim to be given all possible help, if they suffer. The piecemeal engineer will, accordingly, adopt the method of searching for, and fighting against, the greatest and most urgent evils of society, rather than searching for, and fighting for, its greatest ultimate good. This difference is far from being merely verbal. (Popper, 1945/1962, pp. 157-158)


            The comments that follow comparing the two models have in mind the situation of the criminal law in particular, where the means adopted necessarily (with a high degree of certainty) cause suffering.[81]


            Of course, in making this distinction between the two models, Popper is not opposed to the idea of promoting good, and his comments must not be taken as meaning that.  However, he fears that the model that consists of choosing an abstract good leads [depending on what it relates to] more easily to violence, excess, and an appeal to emotions in order to achieve that ideal.  That means that this approach can be carried away by the beauty of the ideal, and pay insufficient attention to the coercive, punitive or violent means used to achieve it.  The greater the good seems to us – or the more worthy the “goals” chosen seem to us (for example, “protecting our children”, protecting human dignity) – the greater the risk that the ideal will obscure the problem of the means chosen.  It may then lead us to accept violent forms of social engineering as reasonable and proportionate.


In the criminal law, means that cause suffering would perhaps be regarded as proportionate and acceptable in relation to the Great Ideal to be achieved, but acceptance of those means presents at least two difficulties: (1) the negative results that will be produced by the means (criminal law) are always more certain than the anticipated positive results;[82] (ii) the means would not necessarily be proportionate to the real empirical harm caused by the individualized behaviour itself.  For example, sentencing someone to incarceration who engaged in a fair sale of a soft drug is not proportionate to that behaviour itself, only to the supposed broader harm caused by the practice (the harm of Trafficking) or the value of the abstract good (protecting Health).  In addition, those means would not be designed to reduce everyone’s concrete, immediate suffering, but rather to prevent the future, contingent suffering of certain individuals.  For example, the utopia of a “drugless society” may cloud our minds when it comes to the means to be used, and cause us to choose the criminal law for imposing that ideal on everyone.  Instead of investing energy in cases where drugs cause suffering, with the aim of relieving that suffering, we cause suffering in the believe that we can prevent it, or achieve a worthy goal in future.


Note that in the first model (punishment to protect or prevent), we must believe that there is a causal link between the criminal law (means) and the prevention or reduction of drug use (goal), or between the punishment and the behaviour (deterrent effect).  If that causal link is not as certain or effective as we believe, we will produce suffering that is morally unjustified and ineffective in practice.  Another characteristic of the first model is that the means are not very open to negotiation or compromise (Popper, 1945/1962, pp. 159).


            On the other hand, social engineering designed to reduce suffering (model II) does not presuppose what is good for everyone: people who are suffering will not completely lose the power to make decisions in respect of their own situations.  In addition, by trying to reduce suffering, we have more chance of doing less harm if our projects do not produce the anticipated results.  The process is less authoritarian and less categorical.  It is also much more cautious and rigorous in terms of how the means are chosen, because it wants to reduce suffering. It is not cognitively preoccupied with the need to find a worthy goal that is likely to justify the suffering that will be caused by the means adopted. 


            We must be particularly vigilant in the use of the criminal law.  Why?  Because, as a German sociologist, Fritz Sack (1968, p. 469), has pointed out, it is a characteristic of the criminal justice system, unlike other social subsystems, that it distributes a “negative good” (negatives Gut), punishment, as opposed to “positive goods” (positiven Gütern), such as social assistance, education, health care, etc.  The Law Reform Commission of Canada (1975, p. 14) expressed exactly that idea when it described the criminal justice system as a system that distributes “anti-welfare”.  As well, sentencing theories that are still popular assign excessive value to suffering as a means of intervention.  The criminal law itself identifies heavily with punishment.  For those reasons, it is dangerous to consider only the criminal law as a means to achieve certain ideals; it is better to regard it, at least in large part, as an objective in itself.


In various sentencing theories, the criminal law has taken on a sort of “quasi-medical justification”, as it was described by Luhmann (1997, p. 283), which was formulated in a 1576 work by Pierre Ayreault dealing with order and formality in judicial procedure: [TRANSLATION] “disease is healed by disease” (quoted in French by Luhmann, ibidem).  In that context, we assign a sort of “necessity” status to suffering to achieve the good (Pires, 1998).  As well, that law believes that it is the punishment that expresses the value we place on certain norms, and so if we assign more value to a norm, we must increase the (maximum) punishment, to express that value.  For example, if you like domestic animals more now than people did in the first half of the 20th century, you should increase the maximum sentence for prohibited behaviours involving cruelty to animals so that the sentence can express your greater love both for animals and for the norm.  The paradox is obvious: in order to express our love for animals, we must express more hatred for – or greater indifference to the suffering of – the offender.  The criminal law thus tells us to assess our own value – and the value that it assigns to the norms – based on the suffering that it imposes on the deviant.  In a policy to control drugs, if you believe that controlling drugs “is worth a lot”, you are predisposed to accept heavy sentences.  In short, in the old culture of the criminal law that is still with us, the more value it assigns to something, the more likely it is to regard it as appropriate to punish violation more severely (Pires, 1995; 2001).


That is why the “goals” that involve producing happiness or protecting a legal good in particular present formidable problems when we want to accomplish them by using the criminal law.  For example, the more value we assign to the goal of preserving health, the more we slide toward an imperceptible justification of severe sentences, and at the same time we lose a kind of critical scepticism that is particularly needed when the instrument of intervention is the criminal law.

Table 8: Two models of social engineering that are suited to the use of the criminal law in public policy

Social engineering model I:

Seeks to achieve the greatest good for the majority

or to protect a “legal good”


(The criminal law as a means)

Social engineering model II:

Seeks to reduce individual suffering without losing sight of

suffering caused by the method of intervention and

loss of individual autonomy

(The criminal law as the goal)


- Any rational action (type I) must have a goal consisting of the production of happiness or protection of a legal good




- The action is rational because it consciously aims to achieve a conscious goal




- The means are chosen based on the goal; openness to sacrificing other people to achieve an end


- A distinction must be made between the final goal and intermediate goals (means to the final goal)


- Any rational action (type II) must have a goal, achievement of which does not depend in the long term on a means (immediate result) that produces suffering or the causal link of which to the effect is uncertain.


- The action is rational because it aims to reduce suffering and the use of means that cause intermediate suffering; it does not value means of control that rely on punishment to achieve the goal


- The means are chosen based on their capacity to produce the least suffering (in intensity and duration)


- A reduction in suffering, including suffering caused by short-term means, must be required

Characteristics of the method

- Dangerous method that may lead to intolerable increase in human suffering and misery


- Application of the method may easily be postponed while waiting for better conditions


- Method that easily leads to violence and extreme means (immediate results) that produce suffering


- Method is open to appeals to emotion to mobilize and achieve the ideal; tendency to consider the end (long-term or ideal results) as more important than the means (short-term results)


- Method not open to negotiation and tends to impose means

Characteristics of the method

- Reasonable method for building a more humane society bit by bit


- Method that can always be applied immediately



- Only method for improving things that has in fact been successful everywhere (at all times and in all places)


- Method suited to filtering out emotion, closed to mobilization of emotion to achieve an ideal, and closed to comparisons between value of the goal and acceptability of the means


- Method open to “negotiation” and to choice and revision of means

Questions regarding control:

1. Do the means seem to us to be appropriate for achieving the desired end?



2. Does the value of the desired end justify the negative or coercive nature of the methods advocated or adopted?




3. Are the means chosen or adopted useful for other purposes?

Questions regarding control:

1. How certain are we, really, that the means will actually get us to the expected end?  Should we rely dogmatically on our conviction, given the (punitive) means being considered?


2. If we agree that the means are capable of producing the effect, can we compare the suffering caused by the means chosen with the suffering caused if those (punitive) means are not adopted?


3. What new disturbances might the means cause?  What will be the overall result (including the result of the means) of the proposed action?


ABBAGNANO, V.N. (1999), Dicionário de Filosofia, Sao Paulo, Martins Fontes. 

AGRA, C. (1993), Dizer a droga ouvir as drogas, Porto, Radicário .

AGRA, C. (ed.) (1998), Entre droga e crime. Actores, espaços, trajectorias, Viseu,Tipografia Guerra.


Alexy, R. (1986), Teoría de los Derechos Fundamentales, Madrid, Centro de Estudios Políticos y constitutionales, 2001.

ALEXY, R. (2001), Teoria de los derechos fondamentales, Madrid, Sociedad Anónima de Fotocomposicion.

ARON, R. (1967), Les étapes de la pensée sociologique, Paris, Gallimard, Bibliothèque des sciences humaines. 


ATI DION, G. (1999), Les pratiques policières et judiciaires dans les affaires de possession de cannabis et autres drogues, de 1995 à 1998: portrait statistique, Gouvernement du Québec, Comité permanent de lutte à la toxicomanie.


BARRATA, A. (1990), “Une politique rationnelle des drogues? Dimensions sociologiques du prohibitionnisme actuel”, Déviance et Société, vol. XIV, no 2, pp.157-178


BATESON, G. (1972), Steps to an Ecology of Mind, New York, Ballantine Books.


BEAUCHESNE, L. (2002), Comprendre le désastre de la prohibition (Tome 1), manuscrit inédit, sous presse.


BECK, U. (1986), Risk Society. Towards a new modernity, London, Sage.


BENTHAM, J. (1970), Of Laws in General, London, University of London & The Athlone Press,      (orig. 1782, edited by H. L. A. Hart).


BERTRAND, M.-A. (1992), “La situation en Amérique du Nord”, in Francis Caballero (ed.), Drogues et Droits de l’Homme, Paris, Les Empêcheurs de tourner en rond, pp.111-130.


BERMAN, H. J. (1983), Law and Revolution. The Formation of the Western Legal Tradition, Cambridge, Mass., Harvard University Press.


BOUDON, R. (1986), L’idéologie, Paris, Fayard.


BROCHU, S. (1997), “Drogues et criminalité: point de vue critique sur les idées véhiculées”, Déviance et Société, vol. 21, no 3.


BROWN, G. S. (1979), Laws of Form, London, Allen & Unwin.


CABALLERO, F. (ed.) (1992), “Drogues et Droits de l’Homme en France”, Drogues et Droits de l’Homme, Paris, Les Empêcheurs de tourner en rond.


CARBONNIER, J. (1963), “L’hypothèse du non-droit”, in J. Carbonnier, Flexible droit, Paris, Librairie Générale de Droit et Jurisprudence, 1988.


CHARRAS, I. (2002), “L’incrimination de l’usage et du trafic des stupéfiants”, in [sic],  pp. 101-110


CONRAD, P. & SCHNEIDER, J. W. (1980), Deviance and Medicaliztion, Missouri, Mosby.


COUNCIL OF EUROPE (1980), Report on Decriminalisation, Council of Europe Publishing.


DAMON, J. (2002), “Vagabondage et mendicité: délits périmés, contrôle persistant”, in L. Mucchielli and P. Robert, Crime et Sécurité: l’état des savoirs, Paris, La découverte, pp.120-129


DE JONG, W. (1983), “Pour une véritable assistance aux consommateurs”, Déviance et société, vol. VII, no. 3, pp.281-286. 


DREYFUS, H. & RABINOW, P. (1982), Michel Foucault: parcours philosophique, Paris Gallimard.


FAUGERON & KOKOREFF, (2002), “’’Il n’y a pas de société sans drogues’: un processus de normalisation?”, dans Claude Faugeron et Michel Kokoreff (ed.), Société avec drogues: enjeux et limites, Ramonville Saint-Agne, Érès, pp. 7-31.


FOUCAULT, M. (1975), Surveiller et Punir, Paris, Gallimard.


GARFINKEL, (1956), “Conditions of Successful Degradation Ceremonies”, American Journal of Sociology, 61, pp.420-424.

GARRAUD, R (1901), Précis de droit pénal, Paris, Larose et Focel.

GEMME, R. & PAYEMENT, N. (1992), “Une modification législative au Canada et ses effets sur le phénomène social de la prostitution”, in La prostitution, quarante ans après la convention de New-York, Brussels, Bruylant, pp. 153-185.


GIDDENS, A. (1990), Les conséquences de la modernité, Paris, Harmattan.


HABERMAS, J. (1997), Droit et démocratie. Entre fait et normes, Paris, Gallimard


HART, H. L. A. (1963), Law, liberty and morality, Standford, Standford University Press. 


HART, H.L.A. (1982), Essays on Bentham: jurisprudence and political theory, Oxford, Clarendon Press.


HOFFE, O.(1993), Principe du droit: Éthique, théorie juridique et philosophie sociale, Paris, Cerf.

HULSMAN, L & van RANSBEEK, H. (1983), “Évaluation critique de la politique des drogues”, Déviance et société, vol. VII, no. 3, pp.271-280.

HUNT, A. (1993), Explorations in law and society: toward a constitutive theory of law, London, New-York, Routledge.

KAMINSKI, D. (1990), “Toxicomanie, le mot qui rend malade”, Déviance et société, vol. VII, no. 3, pp. 179-196.

KANT, E. (1793), Sur l’expression courante: Il se peut que ce soit juste en théorie, mais en pratique cela ne vaut rien, Paris, J. Vrin.


LAW REFORM COMMISSION OF CANADA (1974), The Principles of Sentencing and Dispositions, (Working Paper 3), Ottawa.


LAW REFORM COMMISSION OF CANADA (1976), Our Criminal Law. Report, Ottawa, Minister of Supply and Services Canada.

LUHMANN, N. (1990), La differenziazione del diritto, Bologna, Società editrice il Mulino.

LUHMANN, N. (1993), Risk: A sociological theory, New York, Aldine De Gruyter.


LUHMANN, N. (1997), Das Recht der Gesellschaft, Frankfurt am Mein, SuhrKamp.


LYRA, R. & VICENTE, L. (1969), Postilas de direito penal, Brasilia, Coordenada. 

MATZA,D. (1964), Delinquency and Drift, New York, Wiley. 


MEAD, G. H. (1917), “The Psychology of Punitive Justice”, American Journal of Sociology, vol 23, no. 5, pp. 577-602.


MERTON, R. (1957), Social theory and social structure, Glencoe, Illinois, Free Press.


MOREL, C. (2002), Les décisions absurdes: sociologie des erreurs radicales et persistantes, Paris, Gallimard.


MOURRE, M. (1998), Le petit Mourre. Dictionnaire d’Histoire, Paris, Larousse.


PACKER, H.L. (1968), The limits of the criminal sanction, Standford, Standford University Press


PARENT, C. (2001), “Les identités sexuelles et les travailleuses de l’industrie du sexe à l’aube du nouveau millénaire”, Sociologie et Sociétés, Vol. 33, no. 1.

PERRAULT, M. & L. CARDINAL (1996), “Le droit au choix et le choix du Droit: l’expérience de l’Ontario Coalition for Abortion Clinics et le jugement Morgentaler”, Canadian Journal of Political Science, vol. XXXIX, no. 2.  

PIRES, A. P. (1995), “Quelques obstacles à une mutation du droit criminel”, Revue générale de droit, 26 (1), pp. 133-154.

PIRES, A. P. (1998), “La formation de la rationalité pénale moderne”, in C. Debuyst, F. Digneffe & A. P. Pires Les savoirs sur le crime et la peine. La rationalité pénale et la naissance de la criminologie, (Vol. 2, pp. 1-220), Brussels, De Boeck Université.

PIRES, A.P. (2001) “La rationalité pénale moderne, la société du risque et la juridicisation de l’opinion publique”, Sociologie et Sociétés, vol. XXXIII, no. 1.

PIRES, A. P. (2001), “La ligne Maginot en droit pénal: la protection contre le crime versus la protection contre le Prince”,  Revue de Droit Pénal et de Criminologie, 81( 2), pp. 145-170.

POPPER, K. R. (1963), The open Society and its Enemies, Volume 1, New York, Harper.

RICOEUR, P. (1990), Soi-même comme un autre, Paris, Le Seuil.


ROXIN, C. (1970/2000), Política criminal e sistema jurídico-penal [Kriminalpolitik und Strafrechtssytem], Rio de Janeiro, Renovar.


SCHUR, E. M. (1965), Law and Society, New York, Random House. 


SCHUR, E. M. (1965), Crimes without victims, New Jersey, Prentice-Hall. 


STOLZENBERG, G. (1988), “Une enquête sur le fondement des mathématiques peut-elle nous apprendre quelque chose sur l’esprit?”, in Paul Watzlawick, L’invention de la réalité: contributions au constructivisme, Paris, Seuil, pp. 277-327.


TAYLOR, C. (1985), “What’s Wrong with Negative Liberty”, in C. Taylor, Philosophy and the human sciences, Cambridge, Cambridge University.


THON, A. (1878), Rechtsnorm und subjektives Recht, Weimar.

TOUSSAINT, P. (1983), “L’utopie de la drogue”, Déviance et société, vol. VII, no. 3, pp.287-292. 

UNSWORTH, C.R. (1989), “Witchcraft beliefs, and criminal procedure in early modern England”, Legal record and historical reality, Thomas Watkin (ed.), London, Hambleton Press, pp. 71-98.


von JHERING, R (1877), L’évolution du droit (Zweck im Recht), Paris, A. Marescq.

von FOERSTER, H. (1981), “La construction de la réalité” in Paul Watzlawick, L’invention de la réalité: contributions au constructivisme, Paris, Seuil, pp. 45-69.

WEBER, M. (1969), Economia y Sociedad, Mexique, Fondo de Cultura Económica, 2 vols.

[1]  Compare, for example, Hart (1963, pp. 53, 57, 59-60 and 83) with Schur (1965, pp. v, 6, 6, 169-171).

[2] The concept of decriminalization is used here to refer to the operation of taking away the “jurisdiction” of the criminal system to impose sanctions for a specific behaviour (for example, homosexuality).  There are two general types of decriminalization.  Legal or de jure decriminalization involves an amendment of a statute (or mandatory norm) that is part of the system; factual or de facto decriminalization is the gradual reduction of the system’s reaction to specific behaviours (for example, the purely political or organizational decision not to charge individuals who smoke a marijuana cigarette, without altering the jurisdiction of the criminal justice system) (Council of Europe, 1980, pp. 13-14).  We must be careful, however: when we take away the jurisdiction of the criminal justice system (de jure decriminalization), that does not mean that we are necessarily taking away the jurisdiction of the legal system as a whole.  Other, non-criminal laws can regulate and control behaviour that has been decriminalized.  In this paper, the expression decriminalization will be used to refer to de jure decriminalization, except when we are referring expressly to de facto decriminalization.

[3] The ideas presented here are found in his Treatise on General Sociology (Florence, 1916).  A lovely, clear presentation of his ideas is given by the French sociologist Raymond Aron (1967, pp. 407-496).  Here, we have provided a free rendering of some of the thoughts highlighted by Aron.

[4] Here we must make particular note of the contributions by the English legal philosopher Herbert L. A. Hart (1963) and the American writer on the sociology of law deviance Edwin M. Schur (1965). 

[5] We cannot discuss these in detail here.

[6] There are quite a number of references on this point.  Here, we would adopt the comments made by one of the most illustrious German sociologists of the second half of the 20th century, Niklas Luhmann (1993; 1990, and specifically chap. 2).

[7] Gregory Bateson (1972), one of the most illustrious information theoreticians, has proposed a classical formulation based on this idea of difference.  In his view, the decisive aspect of information is that it produces “a difference that makes a difference”.

[8] Luhmann refers to this fundamental intuition on the part of Spencer Brown in many of his systems theory studies, adapting it to his perspective and to the humanities.

[9] For an in-depth discussion of the criminal law on this point, see Hart (1963, pp. 48-52).  This passage is modelled in some respects on Hart’s discussion.

[10] Compare this question, for example, with the similar efforts by Hart (1963, pp. 53, 59-60 and 83) and Schur (1965, pp. v, 6, 169-171), also to distinguish between two broad groups of prohibited behaviours.  The distinctions that the three of us propose are not the same, and they do not lead us to observe entirely the same things, but we are trying to climb the same mountain from different sides.  Here, we will benefit from some of the equipment organized by them.

[11] Toward the end of this paper, we have identified an eighth criterion that could not be incorporated here.  It does not alter the results presented.

[12] See the definition of thi word in the Oxford Concise Dictionary.

[13] In legal terminology, a latent defect is a defect in a thing that is often hidden or barely visible, that may be a reason to vary the terms of, or rescind, a contract of sale or lease.  The justification for this is that the defect presents a major, if not absolute or radical, problem.

[14] Note that the label “two-sided prohibited behaviours” does not derive from the criterion of causing no harm (Hart, 1963) or the criterion of the absence of a real, distinct victim (Schur, 1965). Whether or not harm is done is neither a central nor a direct criterion, in our view.  In fact, it applies to only a few prohibited behaviours in this group.  The criterion of victimlessness has become an improper way of referring to things – first, because the issue is less that there is no victim than that there is no conflictual interaction between two social actors (see Schur, 1965, pp. 170-171), and second, because it can also be said that the issue less that there is no victim than that the perpetrator and victim have been telescoped, or become a blind spot in the criminal law, which is unable to observe the two figures (perpetrator/victim) that it can, on the other hand, observe in standard prohibited behaviours.  In reality, Schur’s concept of “victimless crimes” burst out and gave rise to a theory (of the criminal law) that is under construction, and is based on a host of other distinctive criteria.  That set of criteria includes two of the key criteria stated by Schur (1965) and also others.  In short, two-sided prohibited behaviours are neither necessarily those behaviours that cause no harm nor, properly speaking, victimless. 


[15] As will become clear later, having regard to the criteria adopted, the distinction proposed here has nothing to do with the old distinction between mala in se and mala prohibita.

[16] Schur (1965, p. 170) also refers to these situations as “exchange crimes”.

[17] Certainly, in the case of homosexual relations, the emphasis that Schur placed on the exchange of goods and services is less appropriate.  His overall conceptualization, however, expressly encompasses such relations.

[18] As Berman pointed out (1983, 181), between the 10th and 12th centuries, the image of the law-breaker was one of an enemy of his or her victim.  Between the 13th and 18th centuries, the law-breaker was seen as attacking the sovereign in addition to his or her immediate victim (Foucault, 1975, p. 51; Garraud, 1901, pp. 32-33).  The image of the law-breaker and the crime was transformed again at the end of the 19th century, when the law-breaker ceased to be seen as a personal enemy of the King and became the enemy of everyone, whom it was to everyone’s benefit to prosecute (Foucault, 1975, p. 44).  On these transformations, see also Pires (1998).

[19] Schur (1965, p. 134) also cited the comment by Robert Merton.

[20] Criminal law theory sometimes uses the expression “active subject” to refer to the perpetrator of the crime and “passive subject” to refer to the direct victim.  In the case of the same of sexual services between a sex worker and a customer, neither of them is the “passive subject”.  No such subject exists, at the first level.

[21] Roxin contrasts “duty offences” – which include offences by omission – with “action offences”.  In the latter, the behaviours and situations are more clearly defined in the criminal law itself.  We cannot discuss this in greater depth here.

[22] It would not be impossible for us to find behaviours or situations here that correspond to grey areas between the two types of prohibited behaviours.

[23] On this reading of Mead and other related questions, see our study (Pires, 2001).

[24] We may ask how that is possible.  The legal scholar, sociologist and legal philosopher Alessandro Baratta (1990, p. 169) provides the answer: it is because, in practical terms, the system for treatment and assistance is incorporated into the criminal justice system; legislation concerning drugs then produce a sort of contradiction in relation to the stated aim (ibidem). What they do in this case is to justify intervening to protect the perpetrator/victim from himself or herself, but, in reality, they are prepared to punish that person.

[25] In the humanities, studies dealing with disasters and bad decisions (whether occasional or persistent) have contributed to the development of our knowledge about various kinds of cognitive traps.  For an excellent overview of that literature, and an original approach, see Morel (2002).

[26] On the 1985 change in the law in Canada, see Gemme and Payement (1992).

[27] Abortion is an extreme and controversial case in terms of this criterion.  We will have something to say about it in the point dealing with the list of control for two-sided prohibited behaviours at the end of this section.

[28] See Brochu (1997, p. 305) where we find a series of factors that are relevant for this purpose.

[29] We have borrowed this expression from the Brazilian legal writer and philosopher Roberto Lyra Jr. (1969, p. 19) in his examination of the elements of criminal offences in the abstract.

[30] We know today that pedagogical messages about drugs, because of the blinkered and unilateral way in which they portray that experience, have lost their credibility with a young public that has [TRANSLATION] “differing experiences concerning the message conveyed to them” (Baratta, 1990, p. 170).

[31] Philippe Toussaint is a journalist in a Brussels newspaper specializing in justice issues, Le Journal des Procès, and took part in a debate on drug policy organized by the international scientific journal Déviance et Société (Brussels, Geneva, Paris, 1983, vol. 7, no. 3).

[32] Unfortunately, that justification was taken up, a little too hastily in turn, by the drug policy of the early 1970s.  On this subject, see the excellent examination by Beauchesne (2002).

[33] For a critique of moralism and paternalism in respect of drugs, see Beauchesne (2002).

[34] Wounter de Jong was treasurer of the Rotterdam Junkie-Bund, which is a union of junkies made up essentially of heroin users in the Rotterdam region; he provided direct information about the drug scene in a debate on drug policy organized by the international scientific journal Déviance et Société (Brussels, Geneva, Paris, 1983, vol. 7, no. 3).

[35] The concept of “passive business” was proposed by Francis Caballero to refer to the need for regulation of the production, sale and use of drugs, while preventing incitement to buy and use, through advertising.  He is director of the Centre d’étude du Droit de la Drogue (CEDD) and a professor at the Université de Paris X.

[36] On this point, the comments of the German legal thinker and philosopher Robert Alexy (1986, chap. 3) on fundamental rights seem to us to be inescapable (whether or not one agrees, moreover, with the idea that fundamental rights should be seen through the concept of “principles”).

[37]  This was a treaty signed in 1494 between the Catholic Kings of Spain and the King of Portugal drawing the line between their future overseas possessions.  [TRANSLATION] “All lands discovered to the east of that line should belong to Portugal, and to the west, to Spain.”  (Le Petit Mourre, 1998, pp .1143-1144).

[38] See, on this point, the similar comments by Muller (1995) concerning the principle of non-violence in relation to institutional practices and philosophies.

[39] Luhmann (1993, p. 10) points out that when the word appeared, the language already had words for danger, undertaking, chance, luck, courage, fear, adventure, etc.  We may therefore assume, he says, that this new word came about to [TRANSLATION] “refer to a problem situation that was not expressed sufficiently precisely by the vocabulary available”.  That term became widespread starting in 1500 (ibidem).

[40] This expression is also cited by Luhmann (1992/1997, p. 135), but we are using it in a broader sense.

[41] We owe the distinction between the concepts of risk and danger mainly to the German sociologist Niklas Luhmann (1993).  We are entirely in his debt for the analysis that follows, but we have taken liberties with it.

[42] We very probably owe this idea to Luhmann, but we were unable to locate the exact citation in time.

[43] Ulrich Beck is a contemporary German sociologist whose name has been associated with the concept of “risk society” which he advanced to characterize our contemporary societies.  His 1986 text is regarded as a best seller and has been translated into several languages.  We have used this concept to examine the current environment of the criminal law (Pires, 2001b).

[44] Bentham (1970) and Hart (1982) offer a number of options for a more in-depth examination of this point, which cannot be undertaken here.  The expression “rights/liberties” is taken from Hart (1982).

[45] See Pires (1998b, p. 149-155).

[46] We probably owe the most sophisticated representation of this image to Kant himself (Hoffe, 1993, pp. 217-220).

[47] It should be recalled that constitutional provisions take precedence over the recommendations in international conventions, and that those conventions usually expressly recognize that fact (see Ati Dion, 1999).  Failure to recognize the precedence of fundamental rights in national laws that implement the conventions may also be seen as a kind of violation of the conventions.  Those conventions recognize the need to respect fundamental rights, and they are also indebted to the international human rights charters.

[48] This aspect of the problem, and the exaggeration of the risks, have also been identified repeatedly in the specialized literature dealing with drugs.  Among many others, see Agra (1993; 1998), Beauchesne (2002), Bertrand (1992), Brochu (1997), Charras, (2002), and Faugeron and Kokoreff (2002).  The last study clearly states – as its title indicates – the [TRANSLATION] “merits of recognizing that there can be no society without drugs” (Faugeron and Kokoreff, 2002, p. 14).

[49] For an excellent analysis of the cultural, scientific or simply cognitive aspect of the criminalization of prostitution, see Parent (2001).  The researcher also shows how knowledge about sexual identity and about sex workers has been a barrier to the reform of the criminal law that has been anticipated for several decades.

[50] On all these points, see the prestigious Council of Europe Report (1980) on Decriminalization, which was extremely favourably received both in scientific and legal circles and among elected representatives.

[51] Sex workers themselves make this proposal publicly and, in conjunction with public authorities and researchers, develop charters of rights (see the World Charter for Prostitutes’ Rights, International Committee for Prostitutes’ Rights, Amsterdam, February 1985).  They have found – and are increasingly finding – support in scientific analyses (Schur, 1965, p. 179; Parent, 2001).  We find nothing like this in relation to theft or the other typical forms of standard prohibited behaviours.

[52] In a chapter dealing with the meaning of empirical data to scientific theory, Robert Merton (1957) specifically drew attention to the fact that astute observation of an unforeseeable, anomalous or strategic fact is beyond a doubt one of the steps in initiating, reformulating, deviating from or clarifying a theory.  But in order for the fact to have a significant theoretical impact (or an impact on the theory), it has to be transformed into something more.  Merton was referring here more to what the (theoretical) observer adds to the fact than to the fact itself.  His analysis seems to us to be similar to the analysis that Karl Popper calls a “bold conjecture”, despite the opposite impression given.  As Popper (1979, p. 81) wrote, “The method of science is the method of bold conjectures and ingenious and severe attempts to refute them.”  We thank our colleague and friend Claude Lamontagne, a specialist in the study of perception and artificial intelligence in the University of Ottawa Department of Psychology, for generously preparing a short discussion paper for us on this aspect of Popper’s thought.  

[53] Our position deviates from Schur’s in relation to abortion.  The cognitive mistake here is not of the same order, and the issues surrounding this prohibited behaviour are very diverse.  This gives rise to a number of possibilities for envisioning and constructing the problem.  This reservation in respect of Schur’s position is not important to our understanding of the criterion.

[54] We have taken this idea from Dreyfus and Rabinow (1982, p. 76), adapting it to our thesis.  The serious knowledge is assigned a certain authoritativeness or potential credibility by reason of the status of its source or of the form (scientific, philosophical, legal, religious, political) it takes.

[55] This is probably the case for prohibited behaviours relating to tax fraud and witchcraft, although the latter example is outside the historical period examined.  Unsworth (1989) draws particular attention to the importance of the cognitive dimension in the case of witchcraft.  Apart from the specific mechanisms brought to bear by the common law, the strong cognitive dimension is also valid for the Latino-Germanic legal tradition.

[56] This general outline may lead to more complex scenarios when we take other variants into account.

[57] This would have required a more in-depth analysis of the connections between the principle by which modern democracies function and the criminal law.  Some of the problems in those connections were very shrewdly identified by Max Weber (1969), although that does not mean that his approach to the question was satisfactory.

[58] Without going into the details of the statistical variations from one place to another, when we combine all of the crimes in the Criminal Code and all offences under the federal Narcotic Control Act, we estimate, roughly, that fewer than 100% of crimes are known to the police because of proactive intervention (no complaint).  At least 90% of known crimes come from complaints.  A majority of two-sided crimes, however, are known as a result of proactive action.

[59] The specialized literature refers repeatedly to the question of police corruption and to other social costs of legislative drug policy for the criminal justice system (Hulsman and Ransbeek, 1983, p. 275;  Baratta, 1990, pp. 167-168).  One journalist who specializes in judicial matters says that enforcement efforts against drug trafficking had been, and was still in Belgium in the early 1980s, [TRANSLATION] “a major source of destabilization within the police forces” (Toussaint, 1983, p.288-289).

[60] To our knowledge, this problem is mainly – if not exclusively – caused by drug enforcement, given the intensity of that activity and of political efforts to “inflate sentences” out of all proportion.  Even if we consider other two-sided prohibited behaviours that involve networks and a relatively organized market (gambling, abortion, sexual services, etc.), we have no empirical evidence to show any similarly significant impact on sentences.

[61]  Of course, that does not necessarily mean that we should create a crime; it simply means that we have passed the test imposed by this criterion.

[62] This criterion is found in some empirical observations made by Schur (1965, pp. 156-158; 163-164), but those observations were not distinguished from a set of empirical observations that are common to all criminal laws.  Consequently, the theoretical specificity of the observations that interest us here was not expanded on.


[63] That is, the conviction that the norm and enforcement of the norm are not succeeding in counteracting the undesirable phenomenon.

[64] We have drawn on and adapted some of what Luhmann (1985) said.

[65] De facto decriminalization may also be effected by the courts themselves, which sometimes develop legal arguments that produce this result.  On this point, see Perrault and Cardinal (1996).

[66] This seems to happen very specifically in the case of drugs, since that phenomenon may easily be perceived by looking at the cognitive expectations of a solution to the health problem, rather than at the normative expectations of a solution to the problem of law-breaking.

[67]  It is in this sense that the recommendation by the Law Reform Commission of Canada (1976) must be understood; the Commission said that, before criminalizing, the enforcement measures necessary for using criminal law against the act will not themselves seriously contravene our fundamental values.

[68] This expression is taken from Luhmann (1985) and we have drawn on his discussion of normative and cognitive expectations here, but have adapted it considerably.

[69] The Law Reform Commission of Canada (1974, p. 34) proposed four questions for determining whether an action should be criminalized or not.  The action should not be made a crime unless these four questions or criteria can be answered “yes”.  According to the Commission, we should ask: (1) “Does the act seriously harm other people?”  (2) “Does it in some way so seriously contravene our fundamental values as to be harmful to society?”  (3) “Are we confident that the enforcement measures necessary for using criminal law against the act will not themselves seriously contravene our fundamental values?”  (4) “Given that we can answer ‘yes’ to the above three questions, are we satisfied that criminal law can make a significant contribution to dealing with the problem?”  Drugs would very certainly fail the first, third and fourth tests.  The second is more ambiguous, but it would still seem to be difficult to argue that drugs (and especially soft drugs) violate our fundamental values (and so seriously).


[70] On this point, Luhmann (1985) clearly demonstrated that there are different mechanisms for affirming norms that are, or may be, functionally equivalent.

[71] We cannot address this issue in detail here.  We will say that we must distinguish between a system of regulation that is completely inoperative or negligible from a system that seriously addresses the problem, although in a moderate way.

[72] This principle was recently recognized in the criminal law by Portugal’s new official drug law reform policy (Government of Portugal, 2001).  See also Ati Dion (1999).

[73] Legal terminology is uncertain here, and varies from country to country.  Sometimes it refers to “administrative penal law”, or “penal law” as opposed to “criminal law”, or simply to “social law”, etc.

[74] As we said earlier, this expression is taken from Hart (1973/1982).

[75] In these remarks, we also benefit indirectly from the comments made by Hart (1973/1982).

[76] There is a debate between Habermas (1997, pp. 276-290) and Alexy (1986/2001) concerning the status of fundamental rights norms.  However, it seems to us that their differences would not change our view here.

[77] As pointed out by, inter alia, Karl Popper (1963, chap. 9, note 6), we often call the immediate results “means” and the secondary results “outcomes”, “goals”, etc.  In that case, the action consists of producing an initial result which is the means, in order to produce a secondary result (outcome, goal).

[78] Karl R. Popper is a renowned philosopher and epistemologist who was born in Vienna in 1902.  He wrote a series of works on modern science, but his œuvre also includes essays on social morality that have been translated into several languages.

[79] On this pint, see the collection edited by Podgorecki, Alexander and Shields (1996). The first article by Adam Podgorecki provides an excellent overview of this issue.

[80] By theoretical choice, we prefer the term utopist to utopian. The reason is that Popper give the term “utopia” a purely pejorative meaning, which we do not do.  Rather, we follow a different sociological perspective, which distinguishes between utopianism (negative sense) and utopia (positive sense).  Nonetheless, but for this terminological reservation, there is no substantive divergence from this passage from Popper.

[81] Table 8 summarizes these two models.  It appears at the end of this conclusion because we have not explained all of the elements in the text.  It may still be useful.  It draws on what Popper said (1963), but has been freely adapted for our purposes.

[82] See the comments by Popper (1945/1962, chap. 9) concerning ends and means in one of the footnotes in that chapter.

Top of document