Consciousness, Science, and the Law
We could seek an empirical answer to the question of what makes lawmakers decide to criminalize a behavior. But doing so is not necessary. This issue does not need to be examined at the level of individual lawmakers’ decision-making processes. Moreover, thinking of the problem in this way would simply reinforce the individual-level biases that already structure the discipline. It ought to be clear that this is not the appropriate approach. The real issue is whether the criminal law provides lawmakers with the rules they need to decide if an act of deviance ought to be treated as a crime. If we examine the content of the criminal law, it is clear that the criminal law does not say, “Here is the nature of those forms of behaviors that ought to be treated as crimes.” The law is not so direct. It leaves lawmakers extraordinary discretion to decide which acts of deviance will be defined as crime in the criminal law. Rather than provide this kind of direction, the criminal law simply provides a list of the decisions lawmakers have rendered. What the criminal law gives the lawmaker by virtue of the lawmaker’s structural location in the law-making process is the power to make law and to decide the content of the law. It gives them nothing else.
The scientific issue here that law does not raise is whether crimes are behaviors that are similar by their very nature. The criminal law does not require lawmakers to define an act of deviance as an act that should be regulated by the criminal law due to its nature. In fact, the criminal law does not require that any act of deviance come under the purview of the criminal law. In this sense, the criminal law is not a tool for determining which things ought to be crime; it is, as we have noted, simply the collection of deviant behaviors that lawmakers have already decided are crimes.
The problem that confronts us here, then, is the difference between deviance and crime, that, as far as the criminal law is concerned, is simply the product of politically creating and applying a label to some forms of deviance. As criminologists, we must be aware that this political process lacks validity. Since the difference between deviance and crime is nothing more than a politically constructed label, using that label as the basis for the scientific study of the causes of crime presents serious problems. If crime is a political construction that lacks scientific basis or rationale, criminologists must take that fact into account when creating explanations of crime. They must, if they choose to use the criminal law as the demarcation between law and deviance, include in their hypotheses some way of addressing the influence of lawmakers on the concept, crime.
As an example of this problem, we can select any act of deviance that is sometimes, but not always, defined as a crime. Let us call this act W. W is a deviant act, and we can ascertain this through various measures such as surveys of public attitudes toward this act. In some places or at some times, act W is a crime—that is to say, sometimes W is an act of deviance, and sometimes it is a crime. To differentiate these two instances of W, let us call “W,” the act when it is considered deviance and “W ” the act when it is considered crime. The
c
problem, however, is that Wd = Wc not as a concept but as an indicator of the same behavior. Conceptually, however, they are being treated as if they were different.
Enter the criminologist. The orthodox criminologist accepts that a crime is a behavior defined by the criminal law. They are concerned, therefore, only with Wc—or the locations and times in which W is defined as a crime. Should the criminologist who wishes to understand the causes of a particular behavior only focus on Wc? Or should they also examine the causes of Wd, which is the same but is not a crime in all jurisdictions and times?
If the criminologist decides only to focus on Wc, what validity problems emerge? First, Wc is itself being “caused” by the behavior of lawmakers in certain places and certain time periods who view this behavior as criminal and set the political process of law making into motion. Why they make that selection, or the factors that influence that selection, affects the distribution of Wc. If the goal of criminology is to explain crime and why it varies, then we must know something about the factors that produce crime and the variations in the legal definition of crime. We need to control for those effects in analytic models if we are ever going to be able to appropriately test explanations of crime. We can describe this set of conditions as a mathematical set of relations:
- 1. Orthodox criminology accepts the legal definition of crime in the criminal law (CL) as its concept (CC) and measure of crime (CM). In other words, CL = CC = CM.
- 2. Methodologically speaking, CM is the “dependent” or “outcome” variable in analyses designed to help establish the causes of crime. So CM = Y, where Y is the dependent variable as commonly identified in statistical studies.
- 3. Orthodox criminology ignores the fact that law varies over (VT) time and (VP) place, which affects the measurement of crime. Since no empirical studies of this reality have been tested, the variability in CM due to VT and VP is an unknown function, which we shall identify as u.
- 4. Because the criminal-law definition of crime varies and that variability is related to things that affect the structure of law, the variability in Y cannot be entirely explained as a function of independent variables that represent individual-level characteristics. We can define these characteristics as the combination of the effects, or as the sum of the effects of all the selected independent variables, regardless of the exact nature of those variables.
- 5. Empirically, the typical orthodox explanation of crime can be depicted as follows: Y = a + bXt + bX2 + . . . bXn + e, where e is the error in prediction that results from omitted explanations of Y.
These variables X{ and X2 may take a variety of different forms and shapes and can include interactions between variables.
6. Because, however, the criminal law varies both conceptually and in its measure, some portion of Y must be explained not by variation in individual-l evel characteristics but rather by the structural influences that also shape Y. As a result, the correct empirical model should be Y = a + bXt +bX2 + . . . bXn + e + u.
In short, this model suggests that one way to correct for the problem in the attempt to test microlevel explanations of crime based on the criminal-law definition is to control for the structural factors that influence the measurement of Y. This may be a fair mathematical representation of what ought to be done, but it is not something that we know can be done, since there is insufficient empirical literature identifying the factors that affect the political construction of Y that simultaneously predicts the causes of crime outside of macrolevel research. In other words, u is an unknown. Thus, in order to move forward using this model, criminologists would first need to conduct research that identifies those structural factors u.
Because the variability in crime across individuals, and across those individuals in time and place, is bound up with structural forces that shape the definition of crime, and an unknown portion of the variability in crime is due to these forces, the specification of the microlevel model explaining the causes of crime will be affected by two different problems. The first is that the model will always be incorrectly specified when the causes of the law are omitted—that is, the law clearly reflects the distribution of crime across individuals, but its effect is excluded in microlevel assessments of the causes of crime. One could argue that one way to address this problem is to assess the effect of macrolevel processes that shape law and then to estimate the effect of the causes of crime on the remaining unexplained variance. Second, however, one could argue that the necessary correction is unknown. Consequently, most studies that use the legal definition of crime will produce poor models of crime at the individual level and likely yield invalid results. As a result, we argue that there are consequences for not paying attention to the distinction between the nature of crime, deviance, and the law.
Conceptually, deviance and crime are related to one another by criminologists through the law. Criminologists ignore that it is the nature of law to change and that therefore the legal measure of crime, while it might indeed be the standard by which deviance is turned into crime, is not an objective concept or measure of crime, since its content fluctuates. That fluctuation is created by lawmakers who, responding to different conditions over time and place, construct the law differently and without rules to guide them. As a result, the criminal law is neither an objective nor a scientific concept. The fluctuation in crime is, as we have argued, a function of conditions outside of microlevel criminological measures when they attempt to predict the variability in crime across individuals. Because those factors that affect the political construction of the law are not controlled for in microlevel research, the equations specified in microlevel models of crime are always problematic.
