EVALUATION OF LEGAL REFORM:
LESSONS FROM CRIMINOLOGY
"Now, what I want is, Facts. Teach these boys and girls nothing but Facts … In this life, we want nothing but Facts, sir; nothing but Facts!"
Charles Dickens, Hard Times
Several lines of research converge in the topic of institutions, legal reform and their effects on economic development. These can be mutually supportive and complementary provided that the scope and limitations of each is clearly understood.
The penal area is only one part of the legal system, but it is useful to analyze what is known about its arena of action for several reasons. The first is that there is a long experience of both substantive as well as procedural reforms. The second reason is that, as one of the oldest and most universal areas of law, it has been investigated from several different perspectives. Criminal justice is the legal branch that historically has been the most analyzed and debated, and in which the data have been the most thoroughly scrutinized. The third is that economists have made significant empirical contributions to the field but, at the same time, their policy recommendations in this area have aroused negative reactions, or frank indifference. Thus, some reflections on the economics of crime would appear to be in order to identify the possible contributions and limitations of the economic approach, currently quite dominant in the evaluation of legal reform.
This chapter has two objectives. First, it argues that the design and evaluation of legal reform require a specific, expert, empirical, local and multidisciplinary approach. None of the disciplines currently involved in the task has adequate conceptual tools and methodologies to tackle it alone. The responsibility for analyzing criminal behavior, and for suggesting and evaluating policies to control it has fallen to the field of criminology, an auxiliary discipline of criminal law from which several lessons can be drawn.
The second objective, necessarily exploratory in nature, is to analyze these practical, theoretical and normative lessons offered by criminology for the design of legal reforms and the evaluation of their impact.
The remainder of this chapter is divided into three sections. The first focuses on how criminology has approached the issue of information gathering. The second refers to some theoretical insights that emerge from the evolution and current status of the conceptual debate on crime. In the third section some very simple methodological guidelines are suggested to evaluate the impact of legal reforms.
LESSONS FOR INFORMATION GATHERING
There is No Substitute for Direct Observation
The first lesson from criminology would seem to be the reading of Thomas Gradgrind, the Dickens character obsessed by Facts: to understand and evaluate any institution it is necessary to observe it, record and measure what happens around it. In addition, consistency of the theories and the data that are used to test them is required.
One surprising feature of current work on institutions and development is the abyss between theory, which continues to be micro, and the information that is frequently employed to verify hypotheses, aggregated opinion indexes. In open contrast to the importance accorded to institutional matters, the efforts made to use direct observation, to specify the relevant actors and behaviors, their incidence, and their effect on the economic environment are unimpressive. Given the determining role of the rules of the game, the amount spent to measure prices, employment, investment, consumption, fiscal accounts, the balance of payments, national accounts is entirely disproportionate to the small amount invested in observing, recording and measuring the institutional behavior of economic actors.
The priority given to the quantification of the aggregated impact of such poorly identified, observed and measured behaviors is also strange. In other areas, the seemingly more logical sequence was the inverse: before talking about the economic impact of a phenomenon, such as inflation, the fiscal deficit, or the exchange rate, the more elementary matter of measuring its magnitude was already resolved. There are innumerable examples of economic, social and demographic variables that are today well measured without anyone having yet indicated concern as to their impact on GDP growth or foreign investment.
Criminal behavior could be included in this category of phenomena that are observed, recorded, documented and measured so as to understand it or change it long before its economic impact awakens any interest. A thousand years ago, that is several centuries before Adam Smith, or the concern about poverty, William The Conqueror began to set up a strong criminal justice system. In particular he showed much concern for reducing lethal violence. Some legal reforms introduced both by William and his successors not only laid the groundwork for the institutions most highly valued by economists but also provided an exceptional base of information on homicides for Europe in the Middle Ages, with standards of quality and coverage that have still not been matched by several contemporary societies. It would not be risky to associate the English king's rush to have detailed information on all violent deaths with the also early monopoly on criminal justice. In the 13th century, the English sovereign was not content with the opinion of his advisors on the issue of public order. He regularly received information on the homicides that were taking place in each region, those committed in villages and in the countryside, data on the victim, and, when available, data on the perpetrator.
The most convincing and lasting evidence that William was a monarch interested in evaluating the performance of his administration and of the justice system through the meticulous recording of what happened in his territory is the monumental and detailed public record of people, animals, properties, and even litigation known as the Domesday Book, prepared at the close of the 11th century. In this work, a mixture of census and judicial inquiry, all the controversies related to land and land ownership during the previous twenty years were recorded. The objective was to contrast living conditions before and after the arrival of the Normans. In modern jargon, William sought to evaluate the impact of a sweeping legal reform, the Norman Conquest. He wanted “to know everything that men could tell him about this new kingdom, its inhabitants, its wealth, its provincial customs, its traditions, and its tax-paying capacity." 
Although less ancient than the Domesday Book, or some key criminal justice institutions from that epoch that still exist, such as the coroner, criminology, from its origins, has shown a marked inclination toward primary information and direct observation. Without entering into a detailed account of the information gathering procedures that have historically been used to understand crime since the discipline was born in the 19th century, the most relevant advances in methods can be summarized. Despite the limited development both of statistics as well as sampling and survey design, the eagerness to observe, record and systematize primary evidence received special attention from that date forward.
The so-called phrenologists tried to find relationships between certain physical features and the moral qualities of individuals. They hoped to use them to identify the characteristics of murderers, thieves, or prostitutes. César Lombroso (1835-1909) examined almost 400 skulls of criminals and compared them to those of people without criminal backgrounds, and in particular with hundreds of soldiers killed at the battle of Solferino. His conclusion was that there existed an abnormally developed occipital region in the criminal skull, closer to that of mammals than civilized man. In The English Convict (1913), Goring and Pearson studied close to 3,000 prisoners in detail and isolated 96 features used to compare them with a control group. Their findings contradicted those of Lombroso and showed that there were no notable morphological differences between prisoners and the general population, but there were differences in matters such as weight and height: the criminals were smaller.  A Harvard anthropologist, E.A. Hooton, continued that line of research based on the analysis of criminals and a control group, adding up to a total of 17,000 individuals in eight states. His conclusions were similar to those of Lombroso. Several authors disputed the findings of Hooton, criticizing his racist tone and the validity of his control group. Borman, another American, concentrated on malfunctions of the endocrine glands, more frequent among Sing Sing prisoners than in a control group. Working in a New Jersey detention center, other specialists using the same method obtained different results. Later, attention was devoted to the possibility that the prisoners' unbalanced diet was affecting the results.
One of the most influential works in prison policy, John Howard's State of the Prisons (1777), was the result of a meticulous investigation carried out in England and several places in Europe. The English philanthropist visited prisons, speaking to inmates, and recording the living conditions in these centers in detail. Howard succeeded in drawing attention to the general status of the prisons. His influence was due not so much to having pointed out the injustice, and the unhealthy and deplorable prison conditions, which had already been done, as to the detailed and systematic description of these problems.
Other empirical studies on criminality were based on official sources, the police, courts, and prisons. As of 1775, the jurisdiction of Paris began to publicize the data on sentences for crimes and offenses, which led a judge to suggest that urban expansion was the main reason for the growth in the numbers. Crime statistics began to be published in France in 1826. The "Franco-Belgian School of Social Ecology" used this type of information to try to explain delinquency in the environment beyond the individual. It was able to draw up maps on the geography of crime, and, in some cases, identify several of the economic, social and demographic characteristics of the perpetrators.
In 1833 geographer A.M. Guerry pointed to an association between crime and geography, climate and seasons in his Essai de la statistique morale de la France. Cartography became a method for comparative research on crime and related political, social, demographic and economic factors of the regions under study. The Belgian Quételet (1796-1874) also used statistics to analyze crime by regions. He observed that southern Europe had more homicides during the summer months, whereas the incidence of robbery was greater in the northern part of the continent and during the cold weather. Anticipating the interest of Lombroso in the physiological characteristics of criminals, in 1871 he formalized the method of anthropometry, which measured the different parts of the human body. A few years later, the French judicial police applied this type of measuring procedure regularly.
In his era, Quételet himself pointed out the main problem with official statistics: only the offenses reported to the authorities were included. “Our observations refer only to a certain number of offenses that have been reported and judged, out of a total number of offenses committed that is unknown,” he observed. In 1840, H. A. Frégier took the same tack, pointing out that the official figures lacked "the offenses and crimes that were not reported and therefore are not pursued.”
In the United States, Austin Porterfield published in 1946 the first study based on a self-report survey as Youth in Trouble. The author analyzed the judicial records of somewhat more than 2,000 juvenile offenders and identified close to 50 infractions for which they were convicted. He then surveyed 300 university students to see if they had committed these offenses. He found that each of the students was guilty of at least one of the infractions that, although less frequent, were as serious as the ones charged to the offenders. However, very few of them had been held accountable in the justice system.
At the end of the 1950s, Rowell Huesmann and his colleagues began a longitudinal study intended to establish the continuity of aggressive behaviors, following a group of men and women for 22 years. For more than four decades, Harvard University professors Sheldon and Eleanor Glueck studied criminal behavior and aggression. For their best known work, Unraveling Juvenile Delinquency (1950), they selected a sample of 500 delinquent youths between the ages of 10 and 17 in the Boston area and a matched control group of 500 non-delinquents. The Gluecks collected extensive information on the 1000 youths' social, psychological, biological characteristics, family life, school performance, and work experience between 1939 and 1948. They interviewed family members, teachers, employers and neighbors as well as social services workers and officials of the legal system concerning their offenses. The original group was followed up when they were 25 and 32 years old concerning major life events: marriages, divorces, changes of residence or employment, children, and military service. Detailed information was collected on work habits, income, periods of unemployment and they also asked about community activities and the use of free time. The Gluecks studied both the late initiation of criminal activities in the control group as well as desistence from crime by the juvenile offenders.
For different reasons the influence of the Gluecks in their day was not great. However, their main findings, purely inductive, such as the key role of the family in the initiation of criminal careers, or the association between age and criminal conduct, or the stability in the tendency to engage in crime have not only been corroborated by later work, but today occupy an important place on the research agenda of criminology. The richness of their databases allowed them to be reprocessed and analyzed with new statistical methods and theory 40 years later.
In the Pittsburgh Youth Study, begun in 1987, Rolf Loeber and his colleagues engaged in a longitudinal study of an initial sample of more than 1500 youths to document the development of aggressive behaviors from infancy to adulthood, the risk factors associated with these conducts, and the help they received. Similar studies were conducted in Denver and Rochester.
Since the 1960s, the validity of criminal data from police sources has been questioned in the United States. Concern grew that a significant number of criminal attacks were not becoming known to the authorities The possibility of adapting household surveys to construct crime indicators less vulnerable to the arbitrary nature attributed to official statistics was explored: direct surveys of citizens could be used as statistics on criminality. The first research grants awarded by what would become National Institute of Justice included allocations for fieldwork with victims of crime, and users of the justice system.
Victimization surveys have been consolidating themselves at the international level as an essential tool for the analysis of crime, independent of police statistics. In essence, this is an extension of household surveys done on a representative sample of the population that is interviewed about their experiences related to crime and on their perceptions of the penal justice system and security. Most of the industrialized countries now use this type of instrument for rates of victimization, variations by population segments, percentage of crime reported, the reasons for not reporting a crime, the measures taken to protect oneself, and perceived insecurity and response from the authorities. The ICVS (International Crime Victimization Survey) is a project that involves 67 countries spanning the globe, and is coordinated by the United Nations. The standardization of the surveys on self-reporting of infractions has been slower.
The Project on Human Development in Chicago Neighborhoods, which was a multidisciplinary effort to analyze the association between urban disorder, collective efficacy, and crime, used three sources of information, including a "systematic social observation of 23,000 blocks on the streets of 80 neighborhood clusters selected to maximize ethnic and racial variations and socioeconomic strata".  Videotaping techniques captured aspects of the micro-environment related to risks to social health (i.e., garbage in the street, public drunkenness, unsafe dwellings).
One pattern stands out in the evolution of data collection methods in criminology and that is proximity to the object of study: the criminal and his environment. It is understood that the data based on later steps in the penal process offer contaminated information on the criminal and his behavior, because such data also depend on the reactions of third parties and different law enforcement bodies such as police, prosecutors, judges and juries, whose behavior merits separate study.
Operators and People from Various Disciplines Are Essential
Several factors contributed to the early eagerness of criminology to collect and record evidence from the source. First is the direct contact with operators in the penal justice system. It was possible to capitalize on the detailed logs that some of them kept of their activities and their willingness to share their experiences or to seek systematically for solutions to work-related problems. Police inspectors, those responsible for prisons and penitentiaries, judges, and even philanthropists interested in the subject began their reflections on criminal acts based on data and descriptions of their perpetrators and the circumstances in which they were committed. Paris police prefect M. Dubois set up the first board of health in 1802, which was tasked with examining adulterated drinks, unhealthy workshops, and soon after, prison visits, among other things. Between 1815 and 1828, the board produced more than 4,000 reports. One report, by Frégier -a division chief of the Sena police prefecture- on the dangerous classes of the population in the large cities was published in the mid-19th century and received a prize from the Academy of Moral and Political Sciences.
It was thanks to his position as a military doctor that Lombroso had access to the skulls of soldiers, which he used to compare to his sample of criminals' crania. Rafael Garófalo, another representative of the Italian positivist school, was not only an academic but also a practicing criminal lawyer, prosecutor and magistrate confronting the daily problems of the penal justice system. He was in a position to discuss the proposals of Lombroso or Ferri in conceptual and theoretical terms but also to evaluate the practical implications of their work.
Shaw and McKay, two influential criminologists of the Chicago school, did not hold university positions. They were doing research in a juvenile treatment center. They mapped the spatial distribution of delinquency and its evolution in Chicago, drawing on detailed research into the court records of a cohort of minors over several decades.
At the end of the 1950s a European, Denis Szabo, set up shop in Montreal with the aim of doing research and teaching criminology. Szabo opted for the pragmatic alternative of solving concrete problems and participating in the debate on urban administration rather than the discussion of the big philosophical issues. His initial approach was based on collecting citizens' impressions of laws, regulations and their enforcement. He was interested in evaluating the justice system, the police and the legislative bodies. Through surveys, seminars, conference papers and the systematic dissemination of the results of his research, he steadily gained influence in the administration of prisons and the police. Despite some institutional reticence, the criminologists of his school gained a reputation for being the ones who know what is going on. The results of his research influenced areas such as the education of police and prison guards, modernization of the psychiatric wings, court management, and legal aid. They began to discuss methods to tackle organized crime and a permanent committee was created on legislative reform.
The specialty known as profiling, an interdisciplinary technique that draws on knowledge of fields as diverse as psychology, psychiatry, forensic medicine and ballistics, began with the efforts of psychiatrist James A. Brussel to assist the New York Police Department to find a serial bomber in the 1950s. The technique requires a sophisticated mix of observation, inference, deduction and ability to make predictions. Brussel followed the criminological procedure of searching for clues to understand the motivations and predict the next action of the perpetrator through a detailed analysis of his modus operandi. The method is to “...study how to understand the why, basing it on the experience of a multitude of cases.”
The role that doctors -- and in particular public health physicians, forensic doctors and psychiatrists -- have played at different times in the development of criminology should also be mentioned. Parent-Duchatelet (1790-1836) was a doctor who worked with the Paris board of health at the beginning of the 19th century. He made tremendous efforts to collect direct information for the activities of this board, speaking with those affected, observing their living and working conditions, verifying their health status and family histories. He expressed the view that "even more than in medicine, books do not replace practice and the ones that exist on these subjects are frequently less able to clarify than to incite to error.” Paul Dubuisson (1847-1908) was licensed to practice law and also a medical doctor. He directed the Sainte-Anne psychiatric hospital and worked as a forensic expert. The Archives d’Anthropologie Criminelle and the criminological school of Lyon were developed by Alexandre Lacassagne (1843-1924), a professor of forensic medicine who contributed to work on the justice system based on his forensic reports and the assistance of numerous medical students who worked under him.
Lacassagne examined why the aptitudes of a good forensic doctor are useful for the study of crime. First, there is a close connection with the law and criminal justice system, and the forensic doctor must be up to date with any reform to the laws. On the other hand, forensic medicine, based in observation and experimentation, provides justice with elements that help to make decisions. Thanks to forensic sciences, evidence provided through testimony, on which criminal investigation had been based, was replaced by technical evidence -- "mute witnesses, doubtless, but which the expert can reveal and that contribute a set of decisive and irrefutable proofs.”
In Great Britain, Davie (2005) shows that the systematic study of criminals coincided with the appointment of doctors to the prisons during the mid-19th century. Several physicians began a series of detailed empirical investigations into the mental and physical characteristics of the inmates.
Although criminality is not the main interest of psychiatry, this discipline has included it in its mandate since the alienists, or asylum superintendents, cooperated with the criminal justice system. They had to describe the different states of madness and from that, differentiate normal from abnormal subjects in order to establish criminal responsibility. They also supervised criminals who were receiving psychiatric treatment in asylums. This obliged them to write meticulous reports for the courts, which frequently constituted genuine case studies. Many psychiatry students learned their specialty by visiting the insane asylums. The determination of responsibility by psychiatrists was an intellectual task that served as fertile terrain for reflection on the causes of crime. The psychiatrists took on not only a consideration of criminal law but also the social and family background of the mentally ill, whose violent behavior reinforced the proximity to the criminal world. Between the years 1887 and 1914, it was common to read contributions from French and Italian psychiatrists in the Archives d’Anthropologie Criminelle.
The contributions of public health doctors were no less important to the development of criminology. This was one of the areas in which the step was taken to move from the study of individual cases to analysis using statistics. In addition, the direct contact with a large number of people afflicted by some illness or epidemic placed the public health physicians in the position of being intermediaries between the community and the authorities. They were the ones on the ground, who perceived the demand for state services.
Debuyst et al. (1995) suggest that criminology is a complex discipline that presents important challenges in three dimensions. First, it must articulate scientific knowledge with an ethical reflection. It also must keep in mind the contributions of different branches of knowledge, including law. Finally, the interaction between theory and practice is essential to be able to formulate relevant and realistic policy.
Several elements facilitated the development of criminology as an interdisciplinary undertaking. First, the professions that forged it also had a tradition anchored in several areas of specialization. Consequently, it has been possible to replicate, analyze, and, especially, criticize and improve the accumulated information from several different perspectives. Second, the empirical and positivist approach, based on finding the correlates of crime -- today known as risk factors -- led to the need to explain them. This was generally done by grouping them in large categories, establishing membership according to the criteria of some discipline that was then involved in the business of explaining crime. Gottfredson and Hirschi (1990) point that the table of contents of Crime: Its Causes and Remedies (Lombroso, 1899) looks quite similar to a university catalog, with sections like geology, anthropology, demography, education, economics, religion, genetics, and politics.
No less important is the existence of a common language that can be read, understood and analyzed by specialists in other areas and even by the average citizen. The possibility that its conclusions could form part of a public debate -- not necessarily intellectual or technical -- depends on this point. Several of the forensic physicians at the dawn of criminology were in turn jurists and, since they drafted documents for the courts or participated in parliamentary debates, they had to have the capacity to argue convincingly. What mattered was not so much the ability to reveal universal truths on human nature but possessing the rhetorical skill to affect decisions on concrete cases. The great influence of Lombroso, his capacity to communicate ideas, was probably related to the fact that he had studied literature, linguistics, and archeology before becoming a doctor.
It is likely that this peculiarity -- that the writings can be read by non-specialists -- has something to do with the fact that novelists have often participated in the formulation of theories of crime. Victor Hugo and Fyodor Dostoevsky, among others, have taken an interest in human misery, moral dilemmas, and the suffering that can lead to extreme behaviors. The romantic hero à la Robin Hood, who struggles against tyranny, has been a recurring literary figure. Mark Twain's Huckleberry Finn continues to inspire scholars of youth gangs in Central America. On the other hand, several works on criminals achieved mass dissemination and influence perhaps because they were very well written, like a novel. Examples of this type could include the study of Bernaldo de Quirós on Andalusian gangs in the 19th century or No nacimos pa' semilla [Born To Die in Medellin, 1990] by Alonso Salazar on young hired gunmen in a Colombian city.
Judicial Information is Not Sufficient
From the beginning, criminologists working with judicial statistics were aware that this information represented only a fraction of a phenomenon that had an unknown magnitude. The comparison among data from different sources has made it clear that this is not a simple problem of numerical representativeness. Judicial information is almost always a biased sample of the universe of criminal acts, the seriousness of the bias being proportional to the differential access of certain groups to the justice system. This was clear for Quételet, who, in the 19th century, pointed out not only the existence of an unknown magnitude of crime but also noted that this varied according to the incident. Murders could be expected to show a high correspondence between actual and reported cases, whereas matters such as the official statistics for burglary were affected not only by reports, but also the capacity of the authorities to identify the perpetrators.
The first statistical studies using judicial statistics in the United States indicated that crime was concentrated disproportionately in the most impoverished and marginal urban areas and that those convicted by the justice system were above all low-income people or members of ethnic minorities. In the early 1930s some sociologists such as Merton and Sutherland corroborated Quételet's intuition, showing that this type of information was not the most appropriate for use in diagnostic assessment since it did not show the scope of hidden delinquency, or the dark side of criminality.
The 1930s research Can Delinquency Be Measured? by Sophia Robinson estimated that the number of juvenile delinquents doubled when some nongovernmental agencies were considered and not just the juvenile courts. Moreover, it reported that the social characteristics of perpetrators or questions such as their race or religion depended a great deal on the place where the minors were sent. The author concluded that the judicial information was not only inadequate, it was misleading. Similar conclusions were drawn in later studies.
Regular surveys of victimization in the United States arose out of the growing skepticism with the figures compiled in the Uniform Crime Report (UCR), which had used data supplied by local police departments since 1929. The FBI, which published the UCR, took no responsibility for its reliability. The apogee of this distrust came in the mid-1960s when the New York Times published an expert's opinion that “the F.B.I.’s figures are not worth the paper they are printed on.” The questions had to do not only with underreporting but, above all, with the partiality of the data. The police had a clear conflict of interest and concern was expressed that they could be “cooking the books.” It was recognized that officials at different levels could exercise their discretion in recording events. The fact that reforms in the police departments were frequently accompanied by recounts in the practice of "killing crime" on the books raised suspicion that poor records did not originate only in the failure to report crime but also could be linked to tampering in the hopes of getting more resources. The main results of the pilot surveys that were done before adopting the National Crime Survey (NCS) confirmed those fears. Not only were attacks quite a bit higher than the existing statistics suggested but, even controlling for cases that appeared as unreported in the surveys, higher rates were obtained than those in police data. The conclusion was that “non reporting by the police may account for more of the dark figure than non reporting to the police”. 
The Census Bureau and the agencies in charge of doing the surveys of victimization did not have conflicts of interest when it came to recording crime and, in practice, they eliminated the fear that the figures were being manipulated. When Crime in the United States 1973 was published, the survey was immediately cited and gained acceptance in the media, academia, and public debate.
Another demonstration of criminology's skepticism with official figures and indeed all sources of information has been the custom of comparing the magnitudes and tendencies from different sources, not necessarily to invalidate any one of them but rather because these discrepancies deserve to be explained and can shed light on the behavior of the agencies responsible for the records. Almost a century ago, physician-lawyer Lacassagne, who was accustomed to collecting evidence and verifying it with sometimes contradictory sources in his role as a legal expert, showed how any datum is a simple indication that must be subjected to scrutiny in light of all available information. His early observation on the underreporting in legal statistics arose precisely from comparing them with data from a different source. Lacassagne was concerned about the information on poisonings and abortions. So, he did a detailed analysis of the poisons used in legal cases and compared them to the non-lethal cases that came to the hospitals. He did a similar exercise using incidents categorized as sudden death. The doctor also analyzed the credibility of the official figures on abortion: he established a relation between births and, for some hospitals, compared it to cases of women who came into care for abortion-related complications. Lacassagne concluded that "if we examine closely how these crimes are committed, and the means employed, we realize that a fair number of them are not known.... there seems to have been a lessening of legal criminality, or at least of the crimes prosecuted, but an increase of apparent criminality.” He found that the incidence of delinquency had increased among minors, especially the offenses of robbery and homicide, and that, therefore, the average age of delinquents must have dropped. He corroborated this observation with the ages of members of three youth gangs operating in Paris. He pointed out appreciable differences in the evolution of the more violent criminal patterns in Paris and Lyon. He contrasted the credibility of legal figures of the latter with the results of the autopsies performed by forensic doctors.
One of the main obsessions of the Gluecks in tracking their cohort of delinquent boys and the control group was to collect from multiple sources of information. They surveyed not only young people but also their parents and teachers about reports of infractions in addition to references found in the official records of detentions. Even in the case of basic information on the family, they attempted to interview various people.
In light of the persistent inconsistencies between self-reporting and official figures on delinquency, at the end of the 1970s efforts began in the United States to examine the source of such incoherence. In 1979 a comparative study got underway that examined statistics from three sources -- official data, victimization surveys and self-reports -- as well as the characteristics of the delinquents that could be derived from each of these. The study found that there was greater similarity between victimization surveys and official statistics on the question of the perpetrators' characteristics than there was between these and the self-reports. The possibility was raised that the self-reports were capturing a different set of behaviors than the other two sources. Later it was observed that part of the problem could have originated in the fact that a small number of youths committed a disproportionate number of offenses. In the surveys, those frequent-offenders could be underrepresented.
In Colombia, during the three five-year intervals between 1970 and 1986, the legal statistics on open homicide cases captured relatively well the general tendency of the police information, which was, in turn, consistent with data from the Forensic Medicine Institute. Beginning in 1987 and as a result of changes in penal procedures, the legal figures on homicide grew distant from the evolution of the respective reports. By the mid-1990s, the difference between the sources was one to three. During the same period, it is estimated that the discrepancies in Peru between official statistics on deaths and the data calculated on the basis of the censuses were on the order of 50 percent. For Lima, the difference between the number of homicides reported by the Ministry of Health and the figure on reports is six to one. Meanwhile, in El Salvador, there is a difference of 3 to 1 between homicide data reported by Prosecutor General and by the Police, which in turn shows discrepancies of more than 60 percent with the data calculated on the basis of census information. It is useful to point out that an important part of the effort that has been made in Latin America to collate and analyze critically the crime information from different sources has been promoted, as in Europe in the 19th century, by professionals in the fields of public health and forensic medicine.
Aggregate Information Is Not Always Appropriate
One might think that the emphasis on direct observation, on detailed local information, and on concrete situations is the result of the influence of forensic doctors in early criminology. In truth, it dates from even earlier times. Although since Roman times the notion of law has been inseparable from general and impersonal rules or legal dispositions, originally, law was based on concrete precedent. Common law or case law based on concrete judicial decisions existed prior to legal interpretation and the formulation of legal theory. Even in countries with a tradition of Roman law, certain fields such as administrative law are of a clear praetorian derivation, in other words, forged by judges.
Several Western criminal institutions arose not from conceptual reflection but rather as an answer to concrete problems. The coroner was established by William the Conqueror for tax purposes, to investigate violent deaths. The Inquisition was instituted by Pope Innocent III to repress crimes by the clergy, such as concubinage, simony and heresy. The Public Ministry arose in France out of the procurators named by the king or feudal lord to defend him. Another illustrative example is the jury system. Adam Smith was aware of the importance of this institution not only to ensure the independence of the judicial system from the other branches of power but also to maintain some degree of citizens’ control over judicial decisions. Although initially the jury existed in various European countries, it survived only in England, thanks to a constant effort to defend its decision-making capacity, which was based on a precise knowledge of the facts. This was only possible because the decisions and analysis of the facts were based on a concrete, local context: “Nothing can be more careful and exact than the English law in ascertaining the impartiality of the jurors. They must be taken from the county where the persons live, from the neighborhood of the land if it be a dispute of property, and so in other cases.”
The first effort to consolidate a large number of crimes using statistics was done in a detailed manner, grouping only homogeneous incidents that corresponded to a certain type of crime. From the start, in 1825, the Compte Géneral de la Statistique Criminelle considered ten types of incidents among crimes against persons: homicide, murder, parricide, assault and aggravated assault, assault and injury against family members, infanticide, abortion, poisoning, rape and attacks on decency. A similar number of incidents were classified as crimes against property: falsification of currency, falsification of commercial documents, whether private or on public registries, robbery on public highways, theft by a domestic employee, other aggravated theft, abuse of confidence, bankruptcy and arson.
The Uniform Crime Reports (UCR) were classified into seven criminal categories: homicide, rape, robbery, aggravated assault, burglary, larceny, and motor vehicle theft.
The ICVS distinguishes between more than fifteen categories of incidents. In crimes against property, there is a distinction between house burglary, vehicle theft, theft of objects inside the vehicle, and personal objects, with or without violence. In terms of sexual assault, there is a distinction between rape and harassment.
Surveys of victim reports of crimes look at thirteen types of crimes against property and six against persons, in addition to vandalism, premeditated arson, drug and alcohol consumption, and other infractions.
In contrast to this trend, and in line with Gary Becker’s seminal work at the end of the seventies, the economics of crime not only argues that it is possible to incorporate all crimes committed into one global magnitude of crime, but proposes as an ultimate goal of criminal policy the minimization of the social costs of this variable. Isaac Ehrlich (1996) talks specifically about the notion of a market for crime and summarizes the underlying assumptions to aggregate the various types of crimes.
The influence that these ambitious proposals for economic aggregation have had in mundane areas such as criminology and criminal justice has been minimal. However, the economic approach to legal analysis has left an unfortunate legacy, which is less explicit than in the criminal sphere, and that is the belief that individual conflicts and legal incidents may be consolidated and incorporated into one sole instance that can be measured in monetary units. It is not an exaggeration to identify in Gary Becker’s work this final and untimely trend in economics toward aggregation and monetization of two elements that have always been separate and outside the realm of market relations: legal relationships and judicial decisions.
Posner (1992)  identifies three major stages in the historical development of Law and Economics (L&E) in the United States. In the first stage, during the seventies, work focused on the analysis of anti-trust cases. The legal information contained in these files was used by economists concerned about monopolies to study and analyze current business practices and to induce a theory of firm behavior in non-competitive settings. In terms of legislation, this school of thought was highly influential in the anti-trust field. The second stage of L&E began with the work of Ronald Coase and Guido Calabresi, who set up an analytical framework to study the allocation of property rights. Once again, this involved analyzing legal information from an economic viewpoint. The focus of attention was still the firm. In both the anti-trust and Coase stages, the proposed policy objective, economic efficiency, was clear and unquestionable. Moreover, it could be adjusted without any difficulty to the North America legal framework and was shared by legal scholars. The third major influence in L&E was the publication, near the end of the seventies, of Gary Becker’s work on criminal behavior, mentioned above. This stage differed in two important ways from previous stages. On the one hand, the central object of study ceased to be the firm and began to include the individual. Even if the economic goal sought by the firm - maximum profit - is universally accepted, individuals do not maximize profits but rather they maximize utility. Thus, in terms of human behavior, L&E had to take a different stance and adopt debatable assumptions. Dealing with individual utility and not with benefits is problematic for two reasons ; first, because the utility function for individuals includes elements other than wealth, and second, because the utility functions differ from person to person and strictly speaking, interpersonal comparisons should be avoided. If they are to be made, they are risky. In this context, tools such as cost-benefit analysis that may work to analyze changes in wealth generated by policies for profit-maximizing business may not be appropriate, or universally accepted, for the purpose of analyzing the effects on individual utility.
The second characteristic of the L&E era that began with Becker has to do with its public policy proposal, i.e., economic efficiency as an objective of criminal law, and which in contrast to what had occurred up to that point, did not arise from prior knowledge and analysis of legislation or jurisprudence. For the first time, economists were suggesting objectives for the legal system, in areas where no in-depth prior investigation had been done nor had they even consulted with legal scholars. Becker’s proposal seems to have persisted and has been progressively adopted without the previous task of analyzing the objectives, explicit or implicit, of criminal law. The recommendation seems to have extended beyond the limits of criminal law to include the complete legal framework and indeed, the justice system.
Thus, without a proper debate, either conceptual or practical, L&E adopted the metaphor of prices and markets for legal behavior. With some strong assumptions one could accept that, at a micro level and in a way analogous to price, a legal sanction is an incentive to individual behavior. However, the notion that all institutions and rules of the game can, just like the GDP, be aggregated together in a weighted total of transactions with a monetary equivalent, is more than science fiction. Besides the lack of technical viability, it involves a conceptual effort that is of no interest to non-economists.
Even for experimental L&E, the inadequacy of the assumption that various legal matters can be compared or aggregated, or that judicial decisions can be treated like a price system, begins to be noticeable. Cass R. Sunstein, for example, shows how, even taking into account a set of shared social norms and with relative agreement on ordering the seriousness of the incidents, punitive damage awards in monetary terms have little to do with prices, since they are totally unpredictable, inconsistent and erratic. Moreover, they differ when analyzed individually or as a group.
It is not surprising that most of the aggregate indicators economists use to measure institutional or judicial quality and seek correlations between economic variables are based on perceptions and opinions, and not on observable matters that can be measured.
All of the evidence available on criminal issues indicates an enormous heterogeneity between different societies, as well as over the course of time, with respect to the incidence of different types of crimes. For this reason, among others, a basic lesson from criminology is to focus the analysis on a few properly typified incidents, examined separately. Extending this lesson to other branches of law seems very reasonable.
The legacy of utilitarianism that has filtered into the area of legal and judicial reform through economics by placing the emphasis on the common, aggregate impact on economic performance is inappropriate for various reasons. First, it takes operators and practitioners out of the debate, because they are more interested in measuring concrete matters than in macroeconomic aggregates. Secondly, it complicates the dialogue with other fields that are also interested in legal reform, but that are not such fans of Jeremy Bentham (for example, legal scholars, public health professionals, anthropologists, psychologists, psychiatrists, historians, and geographers). Consequently, it makes it difficult to make progress and concur in the identification, definition, and measurement of the phenomenon under study. This, in turn, negatively affects the possibility of verifying hypothesis and developing theory. In the following section, it will be shown how progress in the understanding of crime was feasible by a focused and inductive approach.
Finally, the eagerness to aggregate data makes the area of legal reform more difficult and complex and encourages interminable doctrinal discussions on the specification of policy objectives. Part of the difficulty that exists currently in the evaluation of the impact of legal reforms might be due to this unfortunate propensity for focusing on aggregates, which is an evil imported from economics. For these reasons, a basic proposal to achieve progress on the evaluation of the impact of legal reforms, either in terms of their description or explanation, is to move away from utilitarianism and concentrate on measuring, explaining, theorizing and setting policy objectives for a limited set of concrete and well-defined incidents. The principle of Occam’s razor should also be applied to normative issues: given a choice of two policy objectives it is preferable to choose the simplest to formulate, measure and evaluate.
LESSONS FOR THEORY
Good Theories Can Emerge from a Process of Induction
From the brief history of how criminology has confronted the problem of collecting information comes a defined pattern: ideas, inspired by systematic observation are formulated as hypotheses susceptible of being verified using a different type of evidence. And, when they are empirically rejected, they are reformulated to recommence the procedure of subjecting them to examination with new data.
Di Bella (1998) speaks of “from stealing to the thief” to characterize the way criminology has chosen to develop theory. The inductive approach has been based on detailed observation of certain specific events in order to make inferential hypotheses on the behavior of actors involved in those incidents. It could not be other way because, as mentioned, if anything has characterized criminal law, it has been its eagerness to precisely categorize conducts worthy of punishment. In addition, the acceptance or rejection of a behavior, its prohibition in the criminal code, has always had a large local component. Mediterranean societies, for example, can be divided between those in which stealing is formally outlawed an others in which it is permitted under certain conditions, even promoted. Cattle rustling is considered an offense depending on who commits it (an enemy, a friend, a neighbor) and the use to which the cattle will be put. Stealing from the rich was permitted but it was totally forbidden to steal from a poor person.
Criminology's classical period, which was characterized by deductive postulates, is considered a pre-scientific stage. The reorientation toward practical questions which took place more than a century ago arose as a response to a not very fruitful period characterized by exclusively dogmatic reflections which, at the end of the day, did not contribute much to the explication and understanding of criminal phenomena. "For us, the experimental (inductive) method is the key to all knowledge; for them, science needs only pen and paper, and the rest comes forth from a brain stuffed more or less abundantly with knowledge from books...For us, science requires spending a great deal of time, examining the facts one by one, evaluating them, reducing them to a common denominator, and extracting from them the central idea."
Work from influential theorists such as Gabriel Tarde or Emile Durkheim is based on the correlations identified in empirical studies like those of Quételet and emerged to challenge their principal findings, for example the so-called Thermal Law, or the effect of climate on crime.
In the United States, the fieldwork done by Shaw and McKay was essential to the theoretical developments proposed by Edwin H. Sutherland. The notion of social disorganization, suggested by the former to describe those neighborhoods where controls had weakened and criminals challenged conventional institutions, was rendered more sophisticated in the concept of differential social organization as expounded by the latter.
Victimization surveys contributed to knowledge on both victims and aggressors. They provided information at the level of incidents that could later be grouped in different ways. They also yielded, for the first time, estimates of unreported crime. It was possible to obtain new classifications of crime: "stranger crime," "between acquaintances," "workplace crime." The new information base facilitated the development of theory. The consequences of crime were studied for the first time, and the responses of victims were no longer assumed but became the object of analysis. A great deal of attention was centered on why victims turn to the authorities or do not do so. It was found that previous experiences with the police affected the tendency to report later crimes. Researchers also began to investigate reactions other than turning to the police. Emphasis was placed on the daily activities of the victims. Ever greater sophistication in empirically verifying theory was possible because the surveys were collecting information both on victims as well as those who were not attacked. It was observed that a small number of people accounted for a significant percentage of the victimization. The possibility that a first victimization left the person more exposed to later risk was floated, and efforts were centered on protecting victims to prevent crime. The surveys also shed light on the characteristics of perpetrators. Many of these profiles had not been captured by the police.
The empirical discovery in the early 1970s that less than 10 percent of the individuals were responsible for more than 50 percent of the criminal attacks was crucial in reorienting research, and theory, toward the study of criminal careers. These habitual offenders' lives were examined from infancy. Through systematic study of longitudinal samples of youths in some American cities, Rolf Loeber and his colleagues worked in an eclectic and inductive manner to suggest the existence of three pathways to juvenile delinquency.
A more recent observation is that a few families in different communities account for a high proportion of delinquents. This has challenged the idea of social setting as the only determinant of criminal behavior. The traditional hypothesis of socialization has been reexamined in the light of new empirical findings: i) inheritable features of parents change the environment they provide to their children; ii) individual genetic structures affect the environments they encounter; and iii) environmental influences do not seem to explain all the similarities among members of a family.
Beyond Individualism and Holism
From its beginnings, with the debates among the classical school, the positivists, criminal anthropology and sociologists interested in understanding crime, a tension has existed in criminology between a purely individualist focus on the human being and the holistic or exclusively social.
The distinctive mark of the classical school was its emphasis on the individual criminal, guided by his own interests, capable of evaluating the costs and benefits of his actions, and acting consciously. Under these premises, the response to crime had to be adapted to the offense and not to the physical or social characteristics of the offender. Bentham, Beccaria, Weber, and today's economics of crime can be considered representatives of this line of thought. Although the first emphasized the mechanisms of political sanction - the penalties established in the law - they never ignored other categories of sanction. The distinction among criminal acts, or sins, or negligence, or failure to comply was not the most relevant distinction in the individual behavioral theory put forward by Bentham. Behavior is governed by pleasure and pain in all cases and what changes is the nature of pain, that is, the sanctions. If these are political, they lead to a definition of crime, if religious, those of sin, if social to sanctions of nonconformity or bad manners, if physical to recklessness. In fact, Bentham thought that the actions of neighbors and the community were individuals' most important sources of pleasure and pain. The economics of crime has centered on political, legal or state sanctions, sidelining those that are moral, religious, or social in nature. The field of sociology, however, considers the latter most pertinent.
The first positivists, like the phrenologists, sought scientific proof that crime was an outcome of an individual's physical features and characteristics. So perpetrators of crime should be eliminated, locked up indefinitely, or cured through medical intervention. Later, as new correlates of crime, or risk factors, were discovered in other areas, the array of interventions that could alter criminal behavior expanded.
In the first half of the 19th century a series of movements were born that sought definitive remedies for the evils of industrialization. Marx and Engels viewed crime as a type of revolt against exploitation. Scotsman Robert Owen (1771-1858) argued that private property was the true cause of criminality, which would cease to exist in a socialist world. At the beginning of the 20th century in the United States, crime, which was of particular interest to sociologists, began to be considered as a social product. It was essential to understand the social roots of crime, and the processes of education and socialization of individuals, in particular the most disadvantaged, had to be transformed if it were to be prevented.
Thus, different theories on crime had different consequences for criminal policy. In the area of legal reform, it is not hard to imagine that the emphasis given in economic literature to the judicial response as the most appropriate way to diminish transaction costs has its roots in the individualistic inclination of the discipline: the challenge is how to change incentives.
At the moment, it can be said that criminology has increasing consensus on the need to take a distance from the extreme positions of the individualism/holism dichotomy, accepting that each of them has some relevance. “While each of these two extreme social philosophies has a grain of truth, both miss the central truth, namely, that every individual, even the hermit, belongs at once to several social systems, such as family, network of friends and acquaintances, firm, club, gang, school, religious congregation, political party … Every social action elicits some reactions that propagate along several networks … Since crimes involve at least two persons, they are social facts. And all social facts involve people embedded in social networks that are in turn included in society at large.
Gottfredson and Hirschi, in A General Theory of Crime (1990), show that there is no reason why the individualism of the classical school necessarily excludes the recognition of social influences on behavior. Both the observation that crime is the result of certain characteristics of the actor as well as the statement that the environment affects the incidence of crime have been empirically supported; therefore, there is no reason for them to be incompatible. Effort should focus then on seeking a theory that is consistent with the two approaches. Although they are quite critical of radical empiricism for its conceptual deficiencies, their work is itself an explicit recognition of the corpus of accumulated information and evidence. It was this stock of data what kept the knowledge of crime from being subsumed into a single discipline and a rigid conceptual framework.
Thanks to their empirical inclinations the positivists accepted the importance of social factors. Lombroso himself softened his position in the different editions of his work, and at the end he recognized the influence of religious, social and cultural elements. The stigmatization to which his work has been subjected for being, supposedly, limited to biological determinism, drew attention away from his principal contribution, which was anticipating the notion of multi-causality of crime. “Every crime has its origins in a multiplicity of causes, often intertwined and confused, each of which we must, in obedience to the necessities of thought and speech, investigate singly. This multiplicity is generally the rule with human phenomena, to which one can almost never assign a single cause unrelated to others”.
Only the explanatory frameworks with scant empirical inclination, such as Marxism, critical criminology, and currently the theoretical economics of crime continue to be locked into a more dogmatic position based on a sharp and unequivocal distinction between individualism and holism.
The main consequence of superseding the theoretical framework of individualism has been the introduction of the idea of crime prevention, acting on the environmental and social causes and not exclusively on the individual. As holism is left behind, in turn, the idea that any legal reform requires a prior in-depth change of the social structure has been discarded.
Softening, almost abandoning a sterile debate between two different ways of seeing human nature is criminology's current intermediate and pragmatic position. It provides a pertinent reading for those involved in the analysis of legal reforms, where tension between individualism and holism is still apparent among professionals involved in the task of analyzing the relationship between institutions and development. Bussani and Mattei (1995) show, for example, the impasse that has been reached at a conceptual level between L&E and legal anthropology, or between the neoclassical economists and those with Marxist training interested in development. In Latin America, the conceptual disagreement between economists, the intellectual heirs of Bentham and Smith, and the lawyers, who are closer to the holism that characterized Durkheim or Marx, are evident, and in the short run, appear insurmountable.
It is not altogether risky to predict that, like criminology, the convergence toward an intermediate position, relevant to operators and for the evaluation of performance, will only be feasible on the basis of empirical work that is shared, debated, and critiqued by different areas of specialization and not on the basis of an imposition of a theoretical vision of the world.
Prevent or Punish
César Beccaria maintained that it is better to prevent a crime that to have to punish it. He notes, moreover, that "another means of avoiding crime is by rewarding virtue... if the prizes given by academe to the discoverers of worthwhile truths have multiplied the news and the good books, why couldn't awards distributed by the benevolent hand of a ruler thus multiply virtuous actions in the same fashion? The coin of honor is always inexhaustible and fruitful in the hands of the wise dispenser.”
In theorizing on crime it has always been recognized that punishment is not the only way to enforce criminal law. In the literature on legal and judicial reform, however, the alternative of prevention as a mechanism to achieve the rule of law remains conspicuous by its absence.
In criminology the political option for one or another alternative has been the subject of a long-running debate. The prevention/sanction debate shares characteristics of the conflict between classical sociology's holistic worldview and the rational choice approach. The place of public intervention in the first perspective should emphasize change in the economic and social conditions that push certain social actors into delinquency. Assuming rational choice, on the other hand, the response should be to send a dissuasive message through the application of sanctions to the one who has decided to commit an offense, in order to change the incentives that affect his choice.
Although there are still different currents in criminology that assign a distinctive role to sanctions, the dichotomy is blurring. Sherman et al. (1996) suggest that the concept of prevention can be understand as a result, which can be attained through the use of different means or instruments, one of which would be penalties. They propose to define prevention as “any policy which causes a lower number of crimes to occur in the future than would have occurred without that policy.” Thus, certain sanctions would have a preventive effect while other might not, or they might offer a stimulus to new crime. Something similar can be said of other interventions.
It is surprising to see that an idea that is so venerable, simple, suggestive and even rooted in conventional wisdom such as "an ounce of prevention is worth a pound of cure" has still not found its place in the reflections on legal and judicial reform. This oversight could be an additional indication of the influence of economics in this field, and the abyss that separates theory and practice. In the area of accidents, for example, in which L&E centers the analysis on the judicial response, tort law, there is increasing agreement among several disciplines that it is more effective to encourage careful behavior through measures of prevention. Psychologists have developed a set of guides to create habits of cautiousness. The same opinion is gaining ground in the area of industrial safety. There is evidence that some companies have succeeded in altering habits around workplace safety through the use of routines.
No pedagogue or parent would be surprised at the idea that honesty, keeping one's word and meeting one's obligations, not cheating, and paying debts on time are attitudes that can be inculcated from infancy, without waiting for judicial incentives.
Taking the preventive dimension into account, the relationship between institutions and economic performance becomes more complex. On one hand, the argument that, historically, greater economic opportunities have enabled the improvement of institutional supply is supported. In non-judicial prevention of conflicts the causal relation between economic and institutional is clear: through investment of resources, for example in education, health or family support, economic agents can be induced to adopt patterns of behavior that imply low transaction costs. The time horizon relevant for the analysis also expands. The impact of certain conflicts on economic development is not perceptible perhaps at the moment, but they are determining factors in the long run, if, for example, they affect access to markets for the poorest or investment in human capital. The so-called youth problems – dropping out of school, drug consumption, adolescent pregnancy, single motherhood -- enter in this category of matters with little or no interference in the day to day economic environment but fundamental to development in the long run.
In the area of prevention, it is also surprising to see the scant attention paid to family structure and conflicts in the economic literature on legal and judicial reform, for, as criminologists know, they are determinants of problematic behaviors. It can be suspected that they affect several areas acknowledged to be critical for economic performance, such as property rights. Tanzi (1995) is convincing when he points out the links between family structure and corruption.
The affirmation that family arrangements are a critical determinant of institutions is more than conjecture. The study of cohorts of juvenile offenders shows that what we are as adults vis-à-vis criminal law is no more than the outcome of a cumulative individual sequence of small incidents over the life course that the immediate environment accepted and consolidated, or rejected and corrected. In this process of socialization, or civilization, the role of the family is crucial. Although the extension of this scheme to non-penal areas of the institutions has not been done, there are reasons to think that it is pertinent. It is sufficient to read a paragraph that describes in cursory fashion the critical place of the family in the development of self control, something that reduces the probability of problematic behavior in adolescents, thinking that, perhaps, through this same route opportunistic, rash, or dishonest behaviors in adult business relationships could be influenced. “The ability and willingness to delay immediate gratification for some larger purpose may be assumed to be a consequence of training. Much parental action is in fact geared toward suppression of impulsive behavior, toward making the child consider the long-range consequences of acts. Consistent sensitivity to the needs and feelings of others may also be assumed to be a consequence of training. Indeed, much parental behavior is directed toward teaching the child about the rights and feelings of others, and how these rights and feelings ought to constrain the child’s behavior”.
The acceptance and legitimacy of the institutions is not only a matter of rational adults who one day meeting in open council, or after a round of spontaneous negotiation, design, demand, or adopt some rules of the game. Institutions require families that accept, legitimate and transmit them. And not all family structures are equally receptive to the different institutional arrangements.
The key question that criminology poses for other legal areas is if, as occurs with criminal offenses, drug use, and even some accidents, there are certain early behaviors that can predict uncooperative, opportunistic, and irresponsible behaviors of economic agents, that could be prevented without having to wait for a civil court judge to rectify them with an always delayed and costly sentence.
EVALUATION OF LEGAL REFORM
The evaluation of legal rules by legal scholars
For the evaluation of a legal rule, Norberto Bobbio (1992) sets out three basic criteria: the justice of the rule, its validity and its efficacy. The first criteria, used to determine if it is a fair or unfair rule, focuses on the extent to which it meets the goals of an organization; it has to do with the coherence of the rule in terms of what should be done. The second exercise consists in comparing it with what is accepted as the set of valid legal rules; to determine if it is part of the valid legal order. Lastly, the matter of the efficacy of a legal rule is related to its impact on the behavior of its intended recipients and, in particular, is concerned with their observance of it.
Bobbio emphasizes the fact that these three criteria are independent of each other and that they define three major areas of legal thought: the theory of justice, which deals with defining the values that are behind the law, the general theory of law, which is concerned about the validity and coherence of the legal order, and finally, sociology of law, which analyzes the application of the legal rules and their effect on individual behavior. In addition, Bobbio points out the dangers of reductionism, understood as the tendency to ignore one of the value criteria, or to incorporate it into the other two.
An important lesson from criminology is that it has placed itself squarely in the realm of sociology of law, which is concerned with efficacy and based on empirical study. Criminal law is taken as given and straying into the field of legal theory or the theory of law is avoided. L&E, in contrast, is presented as a more ambitious endeavor. It is not limited to analyzing the behavior of individuals under the law, but rather states that, going beyond its area of interest, the legal system and legal reform should be mechanisms to foster economic efficiency.
One might suspect that it has been this misinformed imposition of legal criteria that has most negatively affected the acceptance of economics in the criminal field, even obscuring the valuable empirical contributions made. It is also easy to argue that economists rush to study aggregates, that always problematic tendency of adding apples and oranges according to the imperfect similes of the marketplace, which is a hazard of the profession inherited from utilitarian doctrine, and which has imposed useless obstacles to gathering information and the joint development of theories together with other fields, especially the legal field.
Thus, a requisite to achieve progress in the complex task of evaluating the impact of legal reform is to overcome these “numerous hidden biases in economic reasoning, the smuggled-in value premises and the frequent naturalistic fallacies”. The first step in this direction consists of simplifying, focusing, and breaking down the theories and the data to compare them, as well as the policy objectives and the criteria to evaluate the reforms.
Data, Theory, Policy Proposals, and Disaggregated Evaluation
Since criminology has been recognized from its beginnings as a discipline supporting criminal law, the branch of law that is obsessed with typifying the behaviors that earn coercive sanctions, it has turned its attention to the analysis of concrete and precise conflicts with the law, both to gather information and to study those who commit these acts. The principle of evaluating criminal law reform has normally been based on the analysis of its impact on the precise and specific matters contemplated in the reform, and not on aggregate criteria that are ambitious but often vague and difficult to measure.
Identifying the conflicts or incidents – criminal or not – that are most relevant in a society, the frequency of their occurrence, the proportion that are resolved through the court system, or their impact on economic growth and poverty, is a matter that is more empirical than conceptual. The evidence on crime suggests there are great differences between societies, regions, and time periods. In other words, it is not reasonable to suppose that the basket of incidents that concerns law and justice or affects development is similar among various countries.
A micro-focus, based on conflicts or incidents, makes it more viable to replicate the conditions that allowed the development of criminology. On one hand, with a mentality that breaks down the information, one can achieve progress in terms of the basic problem of gathering information that can be analyzed, critically studied and improved by other disciplines. On the other hand, it becomes possible to formulate a set of hypotheses or micro-theories that, tested with the data, not only allow a better understanding but also can guide legal reform and evaluate it. Third, daily problems of practitioners can be solved, which also contributes to improving information systems and perfecting explanations.
Some concrete examples of legal reform serve to illustrate how an approach that is practical, modest, detailed, and open to various disciplines and currents of thought can be more viable and fruitful than a more ambitious focus centered on large economic aggregates.
Bolstered by the influence of conservationist trends, the National Systems of Protected Areas (SNAP) have grown rapidly. In the Amazon region, the area has increased 30% over the past decade, and in Bolivia it has nearly doubled. By definition, this is an area that involves legal reform of the property regime and is local, yet far-reaching. The SNAPs seek the preservation of certain species in specific geographic spaces, by means of issuing rigorous laws. There is often antagonism and conflict between its proponents, the economic interests of those who engage in the “predatory exploitation of natural resources”, and even local populations, who are the would-be beneficiaries of these programs since they live off those resources.
How can one evaluate the impact of this type of legal reform? Faced with this task, a criminologist could take the SNAP as given and merely evaluate its efficacy; he or she could conduct a detailed follow-up of real issues susceptible to measurement, such as the areas and numbers of protected species; keep a regular record of the breaches and if possible, interview the offenders and enforcers of the SNAP. This information could be used later to draw up a set of hypotheses on the behavior of these actors, which eventually would allow the researcher to suggest adjustments in the mechanisms used for sanction or prevention. This is a viable task, with the advantage of attracting various disciplines – anthropologists, geographers, biologists – to support the work and with the potential to generate a shared knowledge base.
Another project falling in the area of conservation is the Coral Reef Rehabilitation and Management Program (COREMAP) in Indonesia, a project with clear, concrete and measurable objectives. The evaluation of the efficacy of the component of legal reform of this project also appears to be a task suited to criminology, which could use it as data and would not seek to imbue it with more ambitious objectives. With a detailed and focused approach, it could generate information and some rudimentary theories on legal offenders, knowledge that would also encourage progress in the design of this type of reform.
Projects for the administration, record keeping, and legalization of property, and land registration also have an important aspect of legal reform. Using methodology similar to that used in the Domesday Book a thousand years ago, a group of anthropologists, geographers, and agricultural specialists could count and measure lots, interview landowners, and document the processes of legalization and any conflicts. It is unlikely that they will be concerned about the impact of the reform on growth or the investment climate over the next decade.
Even for projects where the explicit objective is to improve efficiency in a sector – such as the energy sector – and that contemplate changes in the legal regime, evaluation of the efficacy of the reform has an enormous element of gathering and recording basic information. Restructuring and privatization programs for public corporations, for example, railways, which involve changes to the property regime, also need evidence from the specific area that supplies the good or service that is being privatized in order to evaluate them; in this case, the transport of passengers or cargo, and the rates before and after the reform.
For all of these projects, one might fear that an economic approach will not be satisfied with evaluating the efficacy of the reforms: it will try, above all, to question the justice of them. At the micro level, an economist might try to refine the analysis, for example, specifying an optimal system of fines for offenders or mechanisms to maximize the flow of information in order to ensure the parties involved in conflicts can arrive at a spontaneous agreement à la Coase. At the macro level, the analysis will focus on whether it is the best alternative to maximize economic growth. It does not seem necessary to go into the field to gather basic information for either of these tasks.
The relationship that exists between the tendency toward macro evaluations and the difficulty of carrying it out is also obvious in projects where the objectives are designed in a global, aggregate fashion, consistent with the L&E approach, such as “improving the legal framework for credit and increasing the efficiency of the judicial system” or “strengthen the government institutions/agencies which have a central role to play in the reform effort, by creating an effective and efficient enabling environment in support of private sector development.”
Even at the micro level, the influence of L&E persists and unnecessarily complicates the impact of legal reform, thereby making the dialogue with other disciplines less fluid, in particular, with legal scholars. A project in Ecuador serves to illustrate this observation. A previous study by the World Bank found that there were insufficient legal resources in this country for women who were victims of domestic violence or who were seeking child support. A reform took into account this observation in its design and legal aid centers were created. After the project was completed, a detailed evaluation, with field work, interviews, and focus groups, all done with the assistance of local NGOs, showed that “women who used the legal aid centers are better off legally, economically, and subjectively, as reflected in qualitative and quantitative measures … Participation in the legal clinics increases the probability of receiving child support payments, decreases the incidence of domestic violence after separation, and is associated with a more positive outlook toward the judicial system”. In addition, these achievements were consistent with specific objectives of the World Bank: “reduction of poverty, empowerment of women, and the promotion of education.”
Despite what is stated above, an economic analysis of this case leads to unusual considerations. “While the obvious and direct questions are an important means of testing whether the clinics had any impact at all, they are not the right questions to ask if one is interested in finding which legal reforms have the greatest economic impact, or if one wishes to compare legal reform with other potential uses of scarce development support resources. The right questions to ask are whether the legal aid clinics in Ecuador increased the incomes of the women they did not serve, whether domestic violence against these women was reduced, and whether the children of these women received more or better education. In other words, what really matters in assessing impact is measuring the spillover effects on the incomes of people whose economic status was improved because the clinics changed the expectations of poor women and their former domestic partners.” Although, as expected, the calculation of the spillover effects could not be done, in any event, the recommendation is that “in the future, legal and judicial reform projects should be undertaken … to promote economic development by capitalizing on the law’s potential leverage or spillover effects on the incentives of economic agents”.
Thus, the detailed and rigorous work of empirical evaluation is marred by a normative postulate: the law, including in the area of domestic violence or child support, must look for economic efficiency. It is prudent to recall that this utilitarian legacy is not always shared by the operators who work with abused women or by family judges or by the NGOs that support their work, or by the anthropologists and sociologists who study the problem, or the jurists who draft legal reforms, or the majority of the politicians who debate them. And that the simple statement multiplies exponentially the difficulties of accumulating information, knowledge and experience.
In the area of criminal policy, the disaggregated, or micro focus adopted by criminology without big normative pretensions has lowered or avoided conceptual and ideological confrontations among disciplines since the objectives of the reforms or the evaluation criteria are based on less debatable, real incidents. It has been easier to gain agreement among the different professions, with the judges, with the auxiliary operators of the judicial system or with informal substitutes on concrete objectives of reform (such as improving the health of people in detention or lowering the number of recidivists, or sexual assaults in Managua, or youth homicides in Boston or Belo Horizonte) than on more ambitious agendas.
Analogously, it is more viable for legal reform to establish concrete goals (such as reducing the number and deadliness of accidents in the workplace, or the nonpayment of mortgages, or the proportion of toys in the market that are dangerous to child health, or the number of entrepreneurs that cannot support the weight of regulation) than aggregate and imprecise questions, such as protecting property rights, improving efficiency, investment climate, social capital or institutional environment. These targeted efforts do not exclude the possibility that an interested party could calculate its impact on some economic variable or that the result could later be used to argue for a subsequent legislative reform.
In fact, an approach centered on the analysis of targeted and homogeneous matters has been adopted in several evaluations of the judiciary based on fieldwork. A survey of lower court judges was used to do an international comparison of the time taken in business cases. A study done in the federal courts of Mexico examined the shortcut procedures for debt collection, the Juicio Ejecutivo Mercantil (Commercial Executive Judgment). A series of studies done in Brazil also focused on debt collection. In another study of courts and users in Argentina, one of the observations is, precisely “the need to base any specific reform on more detailed diagnostics … a focus on real events … [which] might alter the on-going, largely theoretical debates over reform needs.”
Several aspects, which also characterize what is done in criminology, are notable in these works. First, they have been useful to verify or contest entrenched ideas that do not meet the test of data and that, without this effort, would have inspired inadequate legal reforms. Second, they have defined their own dependent variable, in terms that are both concrete and measurable. Third, they have generated their own data and have made the methodology for collecting it explicit so that the procedures are replicable. Fourth, the report that summarizes the results can be read and understand by the average citizen who is not a specialist and, in particular, by those operators or users of the institutions under study. Fifth, they have not concerned themselves with demonstrating the links between legal matters and macroeconomic variables.
This type of detailed, slow, hand-crafted and doubtless tortuous effort to collect primary information through the analysis of judicial cases or by conducting surveys to test and verify simple ideas is the most promising alternative to accumulate knowledge on judicial institutions.
Among the practitioners who can be considered to have more exposure to the economic focus for the evaluation of legal reform -- the officials of the multilateral agencies who work on the topic -- the opinion is frequently heard that the link between judicial performance and aggregate economic variables is very tenuous. Some state explicitly that the relationship is so ethereal and difficult to detect that, in practice, including it in the agenda contributes nothing. In a pragmatic attitude that harks back to the comments of 19th century prison doctors who tried to make decisions in the absence of adequate information or theory, they consider that it is important to try to solve some daily problems found in the field. Among other examples, they assume this would include certain detailed relations and micro- theories, for example on the architectural quality of the courts and the performance of judges, that perhaps for their modesty and scope, have drawn insufficient interest from academia, where the focus is on large general proposals that are difficult to convert into testable hypotheses that could be used by those working with existing institutions.
Even when the explicit objective of a legal reform is to contribute to economic development, to stimulate investment, or to reduce poverty, the evaluation of that reform should go through a microanalysis of its effect on the actions of certain agents and, only then, consider the impact of this behavioral change on economic variables. A case-specific –criminal, civil, economic and administrative- analysis of courts and judges in China shows that “each type of case has different political implications and economic consequences” . The issue of what happens after any given incident, as criminology shows, is analytically distinct; it does not always involve the same actors; and it requires information that is as detailed and specific as that needed to understand the first step of the sequence. As stated, the examination of the consequences of crime was more than a century behind the study of its causes, and it was necessary to have information not only on perpetrators and their conduct but also the reactions of victims and their control group.
On the subject of explanations, hypotheses, or micro-theory, the work of the European geographers in the 19th century showed that the different demographic, economic, social or cultural elements that affect the probability of a criminal incident have a significant local component. Contemporary scholarship on crime has corroborated this early observation. Non-criminal incidents would seem to have the same characteristic of high regional heterogeneity, even within a country.
From the standpoint of the actors involved in the conflicts, including those that come before the justice system, there also seem to be not only local peculiarities but differentiated reactions to the legal environment. Based on judicial case studies in Venezuela, Pérez Perdomo (1996) categorizes litigants into four types: the functional, the instrumental, the expressive, and the duty-bound. Only the first two fit into the model of the rational actor. Galanter (1974) distinguishes two types of litigants: the frequent who turn to the courts time and time again, even having several cases simultaneously, and the occasional, who use the courts sporadically.
According to Rodríguez (2001), there are appreciable differences between frequent and occasional litigants. The greater experience of litigation allows the former to better evaluate the progress and perspectives of the cases. They are even able to establish “the terms of the legal relations that later gave rise to the case; so, for example, companies employing staff write the terms of the contracts, and financial entities set the clauses governing the loans they grant.”
Only through a focus that is targeted to concrete behaviors and the identification of the agents responsible for the actions can a clear idea be gained of the users or plaintiffs, real and potential, of the legal system. This stage of analysis is indispensable to an understanding of their reactions to legal reforms and, through this, an evaluation of the consequences of their actions on economic development.
Aggregate performance indicators for the legal system (efficiency, congestion, impunity, or backlog) and the possibility of making international or longitudinal comparisons have been useful as a first diagnostic stage, to identify possible bottlenecks and test conventional wisdom. However, their utilization as the only indicators for monitoring and evaluation of the legal system and of legal reforms could be unsatisfactory for several reasons. The first, mentioned several times, is that the justice system represents only a portion of the conflicts that need to be understood. The second reason is that the statistics can be manipulated by the entities that supply the information necessary to calculate these indicators, especially when these are used to evaluate the entities that produce them, a clear conflict of interests. The same skepticism that police statistics provoked in the mid-twentieth century and that led to the development of the victimization surveys is applicable to judicial statistics today.
In addition, something that has become clear for criminologists, the further a statistical record of an incident is from the incident it hopes to understand, the more contaminated it will be by the behavior of other agents and variables in the environment. In a detailed analysis of debt collection cases in Brazil, it was found that the same real case or conflict could lead to the opening of several processes, which were recorded cumulatively in the legal statistics: appeals, execution of sentences and even requests not to pay court costs were recorded separately, giving the erroneous impression that the courts were overcrowded.
The third reason is that, even in the absence of direct tampering with the statistics, performance indicators can introduce inappropriate or even perverse incentives. The agencies under supervision can dedicate themselves to fulfilling the criteria suggested by the indicators to the detriment of the quality of justice. This observation is more than theoretical speculation. For criminal matters, for example, there is a great deal of variability involved in the efforts to clarify different types of offenses, with the aggravating factor that the effort is usually proportional to the seriousness of the incident, and, therefore, to the priority that should be assigned to it. It is much more difficult for a criminal court judge, or a prosecutor, to deal with a homicide than a violations of author's rights to intellectual property or a slander or libel suit. If the performance criterion that has been established is the proportion of cases that are resolved, the judicial system subjected to this parameter of evaluation will dedicate itself to resolving the largest number of trivial cases possible to the detriment of the most complex and serious. Something like this has been documented in Colombia. As an outcome of a criminal system reform, performance in criminal investigation began to be assessed based on the relationship between the number of indictments presented and the cases that went to trial, and there was a drastic drop-off in the number of indictments: the indicators improved to the detriment of the capacity to investigate crimes, and it was proportional to their seriousness.
It is necessary for the indicators to be grounded, that they are tied to a real, observable conflict or incident that can be measured by an agency outside the judicial system in order to be able to evaluate its performance, defined as its capacity to affect the probability that the incident occurs.
Finally, to concentrate on incidents that get into the legal system leaves out the analysis of the alternatives of prevention, a point that does not receive attention in the literature on judicial reform. Clarity on the possible options in prevention also requires reference to disaggregated and concrete: the technology on what can be done to prevent traffic accidents is quite different from the one required to prevent customers from defaulting on mortgages.
A recent set of recommendations to improve judicial and legal reform design and evaluation can easily be interpreted as a call to follow the path of criminology in terms of direct observation and data gathering beyond the courts. “It is difficult to examine the record of successes and failures in legal and judicial reform projects because we have little baseline data against which to assess progress ... We need to know what happens to cases that are simply not pursued. In effect, we need to know about the incentives that draw people away from the courts. If we want to make courts more efficient, we nee to know about the efficiency of the informal competition ... Donor agencies should support empirical research on the world’s legal systems”. 
Beccaria, Cesar (1764, 1994). De los delitos y de las penas. Madrid: Altaza
Bobbio, Norberto (1992). Teoría General del Derecho. Madrid: Debate
Bunge, Mario (2007). “A systemic perspective on crime” en Wikström and Sampson (2007) pp. 8 a 30.
Buscaglia, Edgardo and Maria Dakolias (1999). Comparative International Study of Court Performance Indicators. A Descriptive and Analytical Account. The World Bank - Legal and Judicial Reform Unit - Legal Department
Bussani, M &, U. Mattei (1997) "Making the Other Path Efficient: Economic Analysis and Tort Law in Less Developed Countries." en Buscaglia, Edgardo, William Ratliff, and Robert Cooter (1997). Law and Economics of Development Greenwich, Conn.: JAI Press. http://www.jus.unitn.it/cardozo/Review/Torts/Mattei1.html
Cantor, D & JP Lynch (2000) “Self-report surveys as measures of crime and criminal victimization” - Criminal Justice. http://www.ncjrs.gov/criminal_justice2000/vol_4/04c.pdf
CIJUS (1997) "Percepciones Ciudadanas sobre la Justicia - Informe final de Investigación", Mimeo, Bogotá : Ministerio de Justicia, Universidad de los Andes
Coffin, Jean-Christophe (2006). “La place de la psychiatrie dans les archives d’anthropologie criminelle”. Crimonocorpus: Dossier Thématique Nº 1- http://www.criminocorpus.cnrs.fr/article69.html
Cooter, Robert & Thomas Ullen (1998). Derecho y Economía. México, Fondo de Cultura Económica.
Davie, Neil (2005) “L’expertise médicale et légale dans l’émergence de la criminologie en Grande-Bretagne (1880-1918)”. Crimonocorpus: Dossier Thématique Nº 1- http://www.criminocorpus.cnrs.fr/article.php3?id_article=33
Debuyst, Christian, Françoise Dignefe, Jean-Michel Labadie, Alvaro Pires (1995). Histoire des avoirs sur le crime & la peine. Bruxelles: DeBoeck Université
di Bella, Maria Pia (1998). Vols et sanctions en Mediterranée. Ámsterdam OPA, Ordres Sociaux, EAC
Douglas, David (1964). William The Conqueror. The Norman Impact upon England. Berkeley: University of California Press
Drobak, John (2006) Ed. Norms and the law. Cambridge University Press
Ehrlich, Isaac (1996), ‘Crime, Punishment and the Market for Offenses’, Journal of Economic Perspectives, 10, pp. 43-67.
Fu, Hualing (2003). “Putting China’s Judiciary into Perspective: Is It Independent, Competent, and Fair?” in Jensen & Heller (2003) pp. 193-219
Galanter, Marc (1974), “Why the ‘Haves’ come out ahead: Speculations on the limit of legal change”. Law and Society Review 95.
García- Pablos, Antonio (1996). Criminología. Una introducción a sus fundamentos básicos para Juristas. 3a Ed. Valencia: Tirant Lo Blanch
Given, James B (1977). Society and Homicide in Thirteen-Century England. Stanford: Stanford U. Press
Gottfredson M.R. & Hirschi, T (1990). A general theory of crime. Stanford: Stanford University Press.
Hawkins, J. David (1996). Delinquency and Crime. Current Theories. Cambridge Criminology Series, Cambrideg University Press
Hovenkamp H (1995) "Law and Economics in the U.S. : a brief historical survey" en Cambridge Journal of Economics Vol 19 pa g 331-352
Jensen, Erik (2003). “The Rule of Law and Judicial Reform: The Political Economy of Diverse Institutional Patterns and Reformers’ Responses” in Jensen & Heller (2003) pp. 336-381
Jensen, Erik and Thomas C. Heller (2003) Eds. Beyond common knowledge. Empirical Approaches to the Rule of Law. Stanford University Press
Junger-Tas Josine, Terlouw J.G. and Klein Mlacolm (1994). Delinquent Behavior among Young People in the westren World. Amsterdam: Kugler
Kornhauser, Lewis (2000) “El nuevo Análisis Económico del Derecho: Las normas jurídicas como incentivos” en Roemer A (2000)
Lacassagne, Alexandre (1913). “ Les transformations du droit pénal et les progrès de la médecine légale, de 1810 à 1912”, Archives d’anthropologie criminelle, 1913, p. 321. http://www.criminocorpus.cnrs.fr/ebibliotheque/ice/
Laingui, André y Arlette Lebigre (1979). Histoire de droit pénal. II La procédure criminelle. Paris: Synthèse CUJAS
Leca, Antoine (1998). La Genèse du droit. Essai d´Introduction historique au Droit. Aix-en-Provence. PUF
Lilly, Robert, Francis T. Cullen & Richard A. Ball (2007). Criminological Theory. Context and Consequences. SAGE Publications
Llorente Mª Victoria & Mauricio Rubio (2003). Comp. Elementos para una criminología local. Bogotá: Alcaldía Mayor- Ediciones Uniandes
Loeber, Rolf (1996). “Developmental Continuity, Change, and Pathways in Male Juvenile Problem Behaviors and Delinquency” en Hawkins (1996) pags 1 a 27
Londoño, Juan Luis, Alejandro Gaviria & Rodrigo Guerrero Ed. (2000). Asalto al desarrollo. Violencia en América Latina. Washington: BID
Négrier-Dormont, Lygia (1992). Criminologie. Paris: Litec
Négrier-Dormant, Lygia (2007). “Dans la tête du tueur”. Cerveau et psycho, Nº 19 Jan-Fev, pp. 30-35
Macfarlane, Alan (2000). “From predation to production” in The Riddle of the Modern World: Of Liberty, Wealth and Equality. London: Macmillan. http://www.alanmacfarlane.com/TEXTS/S4.PDF
Owen, Bruce & Jorge Portillo (2003). “Legal Reform, Externalities and Economic Development: Measuring the Impact of Legal Aid on Poor Women in Ecuador”. Stanford, SIEPR Discussion Paper No. 02 02-32
Pérez Perdomo, Rogelio (1996) "De la Justicia y Otros Demonios" en Boza Mª Eugenia y Rogelio Pérez Perdomo (1996). Seguridad Jurídica y Competitividad. IESA, Caracas.
Posner, Richard (1992) Economic Analysis of Law 2d Ed. Boston: Little, Brown and Company
Reiling Dory, Linn Hammergren & Adrian Di Giovanni (2007) “Justice Sector Assessment Handbook” – World Bank Publications and Research on Law and Justice Institutions. http://go.worldbank.org/PJHSQT4RA0
Renneville, Marc (2006). “La criminologie perdue d’Alexandre Lacassagne (1843-1924)”. Criminocorpus: Dossier Thématique Nº1 - http://www.criminocorpus.cnrs.fr/article143.html
Robison, Sophia M. (1966). “A Critical View of the Uniform Crime Reports”.
Michigan Law Review, Vol. 64, No. 6, Apr., pp. 1031-1054
Rodríguez, César (2001). “La Justicia Civil y de Familia” in Sousa Santos & García (2001)
Roemer A (2000) Ed Derecho y Economía: una revisión de la literatura. México: Itam, FCE
Rubio, Mauricio (1999). Crimen e impunidad. Precisiones sobre la violencia. Bogotá: Tercer Mundo – Uniandes
Rubio, Mauricio (2007). Economía Jurídica. Una introducción al análisis del derecho iberoamericano. Bogotá: Universidad Externado de Colombia
Sampson, Robert and John H. Laub (1995). Crime in the making. Pathways and turning points through life. Cambridge: Harvard University Press
Sampson, Robert (2003). “Desorden urbano, crimen y eficacia colectiva en el barrio” en Llorente & Rubio (2003) pp. 39 a 51
Sherman , Lawrence W, Denise Gottfredson, Doris MacKenzie, John Eck, Peter Reuter, Shawn Bushway (1996) “PREVENTING CRIME: WHAT WORKS, WHAT DOESN'T, WHAT'S PROMISING” Report to the US Congress, National Institute of Justice, Department of Criminology and Criminal Justice University of Maryland. http://www.ncjrs.gov/works/
Sousa Santos, Boaventura de & Mauricio García (2001) Eds. El caleidoscopio de las justicias en Colombia. Bogotá: Siglo del Hombre Editores
Streeten, Paul (2002). “What’s wrong with contemporary economics?”. Interdisciplinary Science Review, Vol. 27, Nº 1 pp. 13-24
Sunstein, Cass R. (2006). “Damages, Norms and Punishment” en Drobak (2006) pp. 35 a 56
Tanzi Vito (1995) "Corruption : arm´s-length relationships and markets" in Fiorentini, Gianluca & Sam Peltzman (eds) (1995). The Economics of Organised Crime. Cambridge MA: CPER - Cambridge U Press.
Thornberry, Terence y Marvon Krohn (2000). “The self report Method for measuring delinquency and crime” en D. Duffee (2000) Ed. Measurement and Analysis of Crime and Justice (pp.33-83). (Criminal Justice 2000, Vol.4.) Washington, D.C.: U.S. National Institute of Justice.
Wikström, Per-Olof and Robert J. Sampson, Eds. (2006). The Explanation of Crime. Context, Mechanisms and Development. Cambridge University Press
World Bank (2001). “Dominican Republic. Statistical Review of the Justice Sector”. PREM Unit, Report Nº 23203-DR
World Bank (2002). “México. The Juicio Ejecutivo Mercantil in the Federal District Courts of Mexico” PREM Unit, Report Nº 22635-ME
World Bank (2003). “An análisis of Court Users and Uses in two Latin American Countries”. PREM Unit, Report Nº 26966
World Bank (2004). “Brazil. Making Justice Count. Measuring and Improving Judicial Performance in Brazil”. PREM Unit, Report Nº 32789-BR
 Given (1977) p. 6.
 Douglas (1964), p. 353.
 Négrier-Dormant (1992) p. 50.
 Lilly et al. (2007) p. 24.
 Négrier-Dormont (1992) p. 54.
 Debuyst et al. (1995) p. 141.
 Debuyst et al. (1995) p. 142.
 Négrier-Dormont (1992) p. 66.
 Quételet (1835) quoted in Debuyst et al. (1995) p. 146.
 Quoted in Debuyst et al. (1995) p. 161.
 Thornberry and Krohn (2000).
 Sampson and Laub (1995) pp. 26 - 31.
 Sampson and Laub (1995) is the result of this effort.
 Cantor and Lynch (2000).
 Methodology in http://www.unicri.it/wwd/analysis/icvs/methodology.php
 List in http://www.unicri.it/wwd/analysis/icvs/data.php
 In 1988, experts on this type of survey from 15 countries met and agreed to begin a comparative project based on surveys with standardized questions. See Junger-Tas et al. (1994)
 Sampson (2003) p. 42
 For example The Jack- Roller, A Delinquent Boy's Own Story (1930), or The Natural History of a Delinquent Career (1931). Lilly et al. (2007), p. 39.
 Négrier-Dormant (1992) pp. 146 - 147.
 Négrier-Dormant (2007) p. 32.
 Quoted in Duyst et al. p. 156.
 Renneville (2004).
 Lacassagne (1913), p. 347.
 Coffin (2006), Debuyst e al. (1995) p. 215.
 de Quirós, Constancio Bernaldo (1913, 1992). Bandolerismo y delincuencia subversiva en la Baja Andalucía. Sevilla: Renacimiento.
 Debuyst et al (1995) p. 146.
 Thornberry and Krohn (2000).
 Thornberry and Krohn (2000).
 Robinson (1966) , p. 1031.
 “Cantor and Lynch (2000) p. 102
 Lacassagne (1913) p. 327.
 Sampson and Laub (1995) p. 35.
 Thornberry and Krohn (2000).
 Rubio (1999).
 Londoño et al. (2000).
 Leca (2001) p. 22.
 Laingui and Lebrige (1979) pp. 49, 58.
 Quote of Smith’Jurisprudence in Macfarlane (2000) p. 13.
 Lacassagne (1913) p. 325.
 Junger-Tas et al. (1994) p. 62.
 Becker, Gary (1968) "Crime and Punishment : An Economic Approach" Journal of Political Economy 76, Nº 2: 169-217.
 Becker (1968) or Cooter & Ulen (1998) pp. 549-550.
 Also Hovenkamp (1995).
 Hovenkamp (1995).
 “The effort to map moral judgments to dollars is an exercise in scaling without a modulus … Even when norms are widely shared, the unbounded dollar scale is a recipe for arbitrariness and unpredictability.” Sunstein (2006) p. 38.
 di Bella (1998).
 Debuyst et al. (1995) p. 35.
 Ferri, E, Polémica in defesa dellla Scuola Criminale Positiva, 1886. Quoted in García-Pablos (1996) p. 22.
 Négrier-Dormant (1992) p. 67.
 Lilly et al. (2007) p. 41.
 Cantor and Lynch (2000).
 Loeber (1996).
 Moffit and Caspi (2006) p. 110.
 Gottfredson and Hirschi (1990) p. 7.
 Debuyst et al. (1995).
 Bunge (2007), p. 9.
 Crime: Its Causes and Remedies (1899) as quoted in Gottfredson and Hirshi (1990), p. 48.
 Rubio (2007).
 Beccaria (1794,1994) p. 110.
 Shermann et al (1996) p. 2-2.
 Garry Martin and Joseph Pear (1988), cited by Larry L Behrens (1997) “Reducing occupational deaths and injuries: The implementation of positive reinforcement concepts to shape employee safety behavior.” Westmar University http://betacomm.com/posreinf.html
 Behrens, Larry L (1997) “Reducing occupational deaths and injuries: The implementation of positive reinforcement concepts to shape employee safety behavior.” Westmar University http://betacomm.com/posreinf.html
 Moreno and Associates (1997) The behavioral safety process. Internet, http:// safetyonlinelnet/bsp/, home.html.
 Gottfredson and Hirschi (1990), pp. 96- 97.
 Rubio (2007).
 Streeten (2002) p. 14
 See http://www.iabin.info/
 Pasquis Richard, & Pierre Usselmann (2003) “Los Sistemas Nacionales de Áreas Protegidas (SNAP) entre necesidades de conservacióń de la bio y socio-diversidad e imperativos de desarrollo: elementos para el desarrollo sostenible”. Lyonia 4 (2): pp. 187-220.
 http://www.indo.ausaid.gov.au/completedprojects/coralreefrehabilitation.html and the World Bank - COREMAP - Indonesia PO36048.
 See for example, “Bosnia and Herzegovina - Emergency Electric Power Reconstruction Project” P044395, P045483 - http://lnweb18.worldbank.org/oed
 World Bank - Bulgaria, “Technical Assistance for Economic Reform Project” – POO8307
 World Bank (2006). Developing a Methodology to Measure the Impact of Law and Justice. Mimeo.
 World Bank (2000). “Millennium Development Goals” cited by Owen and Portillo (2003).
 Owen and Portillo (2003) p. 8. Emphasis in the original.
 Owen and Portillo (2003) Abstract.
 Buscaglia and Dakolias (1999).
 World Bank (2002).
 World Bank (2004).
 World Bank (2003) p. iii.
 Reiling et al. (2007) discuss the difficulties of different methodologies.
 Fu (2003) p. 194
 For Colombia, see CIJUS (1997).
 Rodríguez (2001) p. 553.
 “There is now sufficient evidence of a certain amount of tampering with court statistics to drive the point home. For example, interviews with one Latin American judicial council revealed that judges submitted differing sets of statistics on caseloads to different offices. One set, including backlog, went to the office that determined whether additional judges were needed. Another set, without backlog, went to the office that evaluated the judges’ own productivity.” Reiling et al. (2007) p. 51.
 World Bank (2004) p vi.
 Rubio (1999).
 Jensen (2003) p. 361