Printer Friendly

Differences in punitiveness across three cultures: a test of American exceptionalism in justice attitudes.


 A. Choosing Countries and Crimes to Study
 1. Countries
 2. Crimes

 B. Measuring Differences in Punitiveness
 1. Scenario Assessment
 2. Punishment-Related Attitudes
 3. Perceptions of Descriptive Norms

 C. Method
 1. Participants
 2. Materials
 3. Scenarios
 4. Dependent Variables
 5. Individual Difference Questions
 6. Demographic Questions
 D. Procedure

 A. Data Preparation
 B. Final Sample Demographics
 C. Effects of Culture
 1. On Scenario Judgments
 2. On Justice Attitudes
 3. Alternative Analysis

"Crime and the fear of crime have permeated the fabric of American life...."

--Warren E. Burger, Chief Justice, U.S. Supreme Court (February 8, 1981) (1)

It is often suggested that there is a peculiarly American psychology of punishment. (2) According to this theory, Americans think about crime in fundamentally different ways than other Westerners, holding uniquely harsh attitudes toward criminal offenders. In this Article, we test this theory of American distinctiveness by comparing the justice attitudes of Americans with those of Canadians and Germans. Are Americans actually harsher than people from these otherwise similar countries?

In Part I, we briefly review the literature on punitive attitudes and distinguish between the theories predicting complete American distinctiveness and those suggesting the existence of a broader class of punitive countries. Part II examines the prior cross-national work on sentencing attitudes, concluding that the limited prior evidence supports the existence of a set of punitive Anglophone countries rather than total American Exceptionalism. Part III describes our experimental design, our selection of Canada and Germany as comparison countries, the set of sentencing scenarios employed, and the survey procedures. Part IV describes our analyses, results, and conclusions. We find that, across a range of major and minor crimes, Americans and Canadians both prefer longer sentences than do Germans but, interestingly, do not differ from each other. The longer sentences Americans and Canadians prefer are not accompanied by a belief that the punished acts were more morally wrongful. The degree of similarity observed between the American and Canadian samples across all crime categories undermines the case for American Exceptionalism in justice attitudes.


The past fifty years have seen a rise in concern about crime in the United States and, concurrently, in the national incarceration rate. Since at least the 1960s, crime has been a major issue in American elections, (3) and the political relevance of criminal justice concerns only increased in the final decades of the twentieth century. (4) Many people believe that sentences are too lenient, jail is too mild, and crime is on the rise. (5) In a move toward what Julian V. Roberts and colleagues term penal populism, political leaders have learned to tap this reservoir of public concern by advocating ever more severe criminal justice policies, leading to an arms race to be the harshest and most severe voice in the public sphere. (6) The United States now has the highest incarceration rate in the world, assigns more long-duration prison sentences than do other countries, and makes considerable use of the death penalty when almost all other Western democracies have banned that punishment. (7) This portrait inspires the "American Exceptionalism" theory in criminal justice policy, which suggests that the United States has a qualitatively different approach to criminal justice issues than other Western countries. (8)

The rising concern about crime and, more importantly, the political response are worrying in light of the apparent miscalibration of public opinion. People are remarkably bad at estimating the sentences that offenders will likely receive, the harshness of the prison conditions offenders will endure, and the probability that they will be paroled, consistently believing that the system is more lenient and less effective than is actually the case, and that the crime problem is worse than it actually is. (9) In 2010, for example, two-thirds of Americans believed that crime was on the rise when government statistics showed that it had been consistently decreasing. (10) This inaccuracy in public perceptions means that "[b]y implication, penal populism involves the exploitation of misinformed opinion in the pursuit of electoral advantage." (11)

Although this combination of public misperception, harsh political rhetoric, and severe criminal justice policy has been observed across many (particularly Anglophone) countries, the American experience is often cited as the prototypical and most extreme case. (12) Many theories therefore focus on the American experience, highlighting elements of the political culture or public psychology that may help explain why Americans would view crime differently than people from other countries. (13) Some of these theories focus on factors that would suggest that the United States is totally unique, such as views of status, (14) a history of vigilante justice, (15) and a tradition of localized criminal justice policy. (16) This strong version of the American Exceptionalism hypothesis is, facially, highly plausible because the United States is the only major Western democracy that still uses the death penalty. A particularly vivid example of these theories is Simon's contention that Americans are "governed through crime." According to Simon, American society sees itself as critically threatened by crime and, therefore, must carry out a "war on crime" in its own defense. (17) Americans see crime everywhere, the fear of crime being a daily affair for both black and white Americans. (18) In this view, American support for three-strikes laws, the death penalty, and related policies are responses to the subjective impression that crimes present a dire threat to society. (19)

Other theories explaining American Exceptionalism focus more on social geography and political process, examining factors that apply to other countries as well as the United States. For example, the United States, along with Canada and many other Anglophone countries, is arguably a frontier society. (20) Countries with a frontier history of individualistic independence and rough justice may have a cultural mindset that is sympathetic to vigilantism and especially punitive toward lawbreakers. (21) Other scholars have speculated that rising income inequality in neoliberal economic systems--particularly the United States, but also the other Anglophone countries such as Canada--has led to increased social exclusion of (the primarily low status) criminal offenders. (22) Similarly, Katherine Beckett argues that the "tough on crime" rhetoric of political elites caused the rising concern about crime in the latter half of the twentieth century rather than the reverse, and that the elites' focus on crime was part of a broader effort to reorient public policy in the wake of the social reforms of the 1960s. (23) Under this explanation, the main difference between the United States and other Western democracies is the strength and tactics of their conservative political movements.

These latter theories speak to the resonance between criminal justice policy and the broader economic and social context. One implication of these theories is that the level of correspondence between the United States and other nations on criminal justice issues may mirror the degree of similarity on these other policy questions. Work by Tapio Lappi-Seppala has found strong relationships between a country's incarceration rate and its social welfare policies, level of inequality, and political structure as well as the level of fear among its citizens, their social tolerance, and their individual punitiveness. (24) This suggests that these different factors are either heavily interdependent--with certain political structures giving rise to a popular discourse that increases the level of fear among a country's citizens--or reflective of common underlying structures.

Some theories would therefore predict that Americans should feel differently about crime than citizens from all other countries. (25) Others, however, would predict that Americans should be relatively similar to people from select countries with appropriate geographies, economic systems, or cultural heritages, but likely different from all others. (26) These competing theories lead to two key questions for the present research:

(1) Are Americans more punitive than citizens of other countries that have very different justice systems, like Germany?

(2) If so, are Americans also more punitive than citizens of countries that have different justice systems but are culturally similar in various ways, like Canada?


Although there is some previous work on these questions, large-scale comparative analyses of social attitudes are relatively uncommon and, when they occur, often have only a few relevant items. Some of the existing studies have focused on views toward specific criminal justice policies (i.e., the death penalty (27)) or broad social attitudes about the leniency of the courts, (28) rather than sentencing judgments per se. Polls have asked questions such as: "Are sentences too harsh, too lenient, or about right?" (29) or "How much do you fear becoming the victim of a crime?" Due to the extent of popular misperceptions about the status quo, however, this kind of attitudinal data can sometimes be difficult to interpret. People are generally unaware of actual sentencing practices, so their level of dissatisfaction with what they believe courts are doing is highly indirect evidence of their personal punitive intent. (30) The lack of specificity in the questions is also sometimes problematic: whether a punishment is too high or too low could depend on the details of the offense, and these are often left unstated. (31)

Some of the problems raised by broad attitudinal questions can be remedied by having participants assign actual sentences to detailed crime descriptions. Intra-country work using this approach has shown that this type of question framing impacts results: it reduces the degree of disparity between the preferences of individuals and actual court judgments in some cases (32) and produces a desire for greater leniency than the courts are showing in others. (33)

The best prior cross-national investigation using this sentencing approach, the International Crime Victimization Survey (ICVS), has mixed evidence for American Exceptionalism. The ICVS, which spanned fifty-eight countries in 1996 (34) and thirty-eight in 2004-2005, (35) asked participants to assign a punishment to a recidivist burglar. American respondents in the ICVS have generally been more punitive than continental European respondents, but not to a much greater extent than people from other English-speaking countries. In the most recent data collection (20042005), for example, approximately half of American participants (47%) wanted to imprison the offender. Though most Western European countries were far less likely to opt for incarceration (France, 13%; Austria, 13%; Spain, 17%), there were numerous exceptions among the Anglophone countries (England and Wales, 51%; Northern Ireland, 53%; Scotland, 49%). Americans were more likely to prefer incarceration than even residents of the Anglophone European countries in previous years (1989, 1996, 2000), but the gap between the U.S. and England, for instance, closed over time: it was fifteen points in 1989, seven in 1996, and five in 2000 before reversing in 2004-2005. The best available evidence, therefore, suggests that Americans are more punitive than most other Westerners, but that they are part of a broader set of punitive Anglophone countries.

The two comparison countries in our study are good exemplars of the continental European and Anglophone categories. Germany was only included in the 2004-2005 collection, but it joined the other countries of continental Europe with 19% of respondents opting for incarceration, substantially below the 47% of Americans. Canada was at 44% in 20042005, close to the three British Isles samples. Also, as with Britain, Canada has been becoming more punitive over time with scores of 32%, 39%, 43%, and 45% in 1989, 1992, 1996, and 2000 respectively.

The ICVS has impressive cross-national reach, but its utility for our purposes is somewhat limited by its use of a single crime scenario: a burglar stealing a color television. Some accounts of American Exceptionalism focus on reactions to extreme crimes, particularly those that might justify a capital sentence. (36) Others focus on the role of drug crimes in the expansion of the prison population (37) or excessive punishment for mundane offenses. (38) Americans could be meaningfully distinct from residents of Anglophone countries on either the most serious offenses (rape, murder) or the most minor (like drug use) while not differing on mid-level offenses like the ICVS's burglar scenario. The ICVS also does not present a comprehensive picture of other crime-related attitudes. Do Germans fear crime less than do Americans? Are people in Western European states, like Germany, less outraged by burglary than are Americans, or do they just prefer alternative means of punishment for it? These possibilities lead to two additional research questions:

(3) If Americans are more punitive than citizens of either of our comparison countries, is this difference localized in a particular crime category, or is it general?

(4) Does American severity reflect a greater degree of moral outrage at criminal offenses, or is it a sentencing-specific effect driven by another process?


We report a comparative study in which census-representative samples from three Western countries read a range of crime vignettes and assigned penalties to the offenders. The unique contribution of this research is the breadth of the stimuli employed, examining punitive intent toward the most serious and most trivial of crimes. We also measure participants' fear of crime, religiosity, endorsement of retributive justice motives, and interest in rehabilitation. Many of the theories of American Exceptionalism link greater American punitiveness to these factors, (39) and past within-culture research has demonstrated that sentencing preferences are influenced by these sorts of ideological constructs. (40) The American Exceptionalism hypothesis would predict that Americans will assign elevated sentences to at least some of the crime scenarios as well as exhibit greater anxiety about crime and greater preference for retribution over rehabilitation compared to people from the other two countries.


1. Countries

We selected Germany and Canada as our comparison countries. Germany is representative of the continental European approach to criminal justice. In Germany, the death penalty was abolished in 1949, and since the Criminal Law Reform Acts of the 1970s, Germany assigns monetary penalties in the form of day fines for minor offenses instead of short-term imprisonment. (41) Currently, 70% of all crimes are sentenced with only a fine. (42) In the United States, however, fines are mostly used in combination with other types of punishment. (43) In U.S. federal district courts, for example, only about 4% of offenders receive a fine as their sole punishment. (44) Moreover, whereas the incarceration rate in the United States is 743 per 100,000 people, it is only 85 per 100,000 in Germany. (45) The treatment of imprisoned offenders also differs between the two countries. Researchers comparing the prison systems remark that prisoners in Germany are treated more like individuals than in the United States, and as "citizens behind bars rather than as outlaws...." (46)

The German criminal justice system stands in stark contrast to the American system and therefore presents an appropriate test of our first question: whether Americans are more punitive than citizens of other countries that have different approaches to justice policy. Each of the theories explaining American Exceptionalism based on public attitudes would predict that Germans would have different sentencing preferences than Americans. Previous research has found that German punitiveness is influenced by the same factors that have been shown to affect American punitiveness, (47) but this work has generally not considered overall extent of each in relation to the other. The present study compares the extent of punitiveness of these countries directly in the same research design.

We selected Canada to represent the Anglophone countries. Canada and the United States are geographically proximate and share similar legal histories, including justice systems based on English common law. Despite this overall cultural similarity, these two countries differ markedly in their approaches to public policy. Canada, for instance, has a far stronger social welfare system. (48) More centrally, the incarceration rate in Canada is only 117 per 100,000 (49)--barely a sixth of the American rate--and Canadian punishment policies also differ from American in qualitative terms, though not to the same extent as those of Germany. The death penalty in Canada was abolished in 1976. Also, in a move to reduce prison rates, the Sentencing Reform Act of 1996 introduced a new sanction form in Canada called conditional sentencing or community custody; instead of being sent to prison, some sentenced offenders are supervised in the community and can continue to work as long as they comply with the program. (50) As with the Criminal Law Reform Acts in Germany, the Sentencing Reform Act in Canada is emblematic of a rejection of the incarceration-heavy approach to sentencing favored in the United States.

Based on existing data, we have reason to expect that Germans will be more lenient than Americans, at least for some crimes, (51) and this is broadly consistent with the American Exceptionalism hypothesis. Canada, however, provides a more conservative test by contrasting the importance of criminal justice policy differences (which remain very large) and cultural differences (which are comparatively small). The broader cultural and social policy similarities of the two countries (52) would lead one to expect correspondence rather than divergence. Canada therefore serves as an interesting test of our second research question: whether Americans are more punitive than those from culturally or economically similar countries.

There are several findings that suggest that Americans will not be more punitive than Canadians. The ICVS showed a substantial difference between Anglophone and non-Anglophone democracies, but only a small and inconsistent distinction between the United States and the rest of the English-speaking community. (53) Also, commentators on the Canadian criminal justice system have noted that the level of fear of crime among Canadians is similar to that among Americans, (54) and emotions of fear and anger about crime are linked to punitiveness in Canada as they are in the United States. (55) These findings have led some to conclude that the divergence between these countries may be better attributed to differing norms among policy elites rather than differences in public opinion. (56) We therefore expect that the theories positing American uniqueness are incorrect and that Canadians may be more similar to Americans than they are to Germans.

2. Crimes

One of the weaknesses of past research was the use of a single crime scenario to assess population attitudes. Here, we employ a range of offenses to overcome that difficulty and to examine our third research question: whether any differences that do exist generalize across crime type. Our first set of scenarios consists of major crimes that directly include the occurrence of or potential for serious violence. American Exceptionalism in criminal justice outcomes is due, in part, to greater use of the most severe punishments, including life imprisonment and the death penalty. This set of scenarios is intended to provoke Americans to assign those penalties and allow us to determine whether Canadians and Germans are as willing to impose them.

Our second set of scenarios consists of minor offenses that lack violence or a potential for violence. If cultures differ in their sentencing of major crimes, there are two patterns that could result for these minor crimes. We may find consistency with the major crimes; punitive cultures are more punitive at every level and lenient cultures are more lenient at every level. This would be consistent with Simon's contention that Americans perceive a spreading menace of crimes and therefore take even minor crimes more seriously. (57) It is also possible, however, that despite differences in views of major crimes, the cultures are generally in agreement on minor crimes. In this case, differences on major crimes would be due to differing willingness to impose the most severe penalties, particularly death.

These first two categories concern mala in se offenses, crimes that are wrong in themselves rather than wrong because authoritative bodies have passed laws against them. Our third set of offenses, containing what are sometimes called victimless crimes, does not necessarily fit that label. This set includes offenses such as paying for the services of a prostitute and use of cocaine by a person who holds a steady job. In practice, the American justice system is much harsher toward drug use and prostitution than is the German system, which does not ban prostitution and is more lenient toward drug use, and the Canadian system, which is mixed in its treatment of these activities. It is therefore especially likely that we will see cultural divergences on these questions, and it may be the case that Americans will be quite harsh toward these offenders whereas Germans and Canadians will not seek to punish them at all. Attitudes toward drug crimes are particularly relevant to our project because punitiveness toward drug users is sometimes cited as a substantial contributor to the rise of American prison populations in the 1970s. (58)

Our final set of offenses is one to which Americans may react more leniently than people from other cultures. These are crimes in which a person defends herself or another from a criminal by using excessive force, for instance, when someone kills a mugger from whom they could have escaped. In many states, the law governing these sorts of cases allows for the use of deadly force only if it is the only way to avoid threatened bodily harm, (59) but Americans are generally lenient toward those who use violence in self-defense. In a study by Paul H. Robinson and John M. Darley, for example, participants were asked to assign sentences to a person who, knowing he could easily run away from a knife-waving attacker, instead drew a gun and killed him. Though most U.S. legal codes would treat the case as one of murder or manslaughter, the average sentence assigned by American participants was less than a year, and 40% of the respondents assigned no prison time at all. (60)

Leniency toward this kind of offender is consistent with broader American punitiveness; both can be seen as stemming from harshness toward the prototypical criminal offender. Though we know of no comparable studies in German or Canadian samples, we have some reason to believe that Americans will be more lenient than Germans. The frontier thesis (61) and work on "Cultures of Honor"--which emphasizes the perceived necessity, in some societies, of being willing to use force to defend one's property and reputation (62)--both predict that Americans will see these cases very differently than Europeans. The Canadian sample, arguably coming from a frontier society, provides an interesting test of the frontier hypothesis and could mirror the American sample.


1. Scenario Assessment

Since sentencing is an observable policy outcome, it has traditionally been the focus of theories on American Exceptionalism. Offenders are incarcerated or not. Their sentences are long or short. The death penalty is an option or not. It is psychologically important, however, to consider the process by which a sentence is assigned. This is why our fourth research question concerns the structure of punitive attitudes. When one person assigns a longer prison sentence to an offender than does someone else, is this because she makes a different judgment about the morality of the act, or is she punishing an act of equal perceived wrongfulness in a different way? Either is possible. To address this question, we asked people to rate each scenario in three ways. The first rating was of the perceived morality of the offense. Past research has shown that moral evaluation is a strong predictor of punishment severity; people punish what they find morally objectionable. (63) Sentencing attitudes could differ across two cultures, therefore, if they fundamentally disagreed about the extent to which a given act is morally wrong.

Moral disagreement is a sufficient but not necessary cause of a difference in sentencing attitudes. Two people may both believe an act (eating meat, promoting racism, or using cocaine) is morally wrong, but not agree that it is the duty of the state to punish people for it. Or people may assign a more severe punishment for a given level of immorality because they are especially worried about the level of crime on the street or chaos in society. (64) So, even if there is consensus about the degree of moral wrongfulness of an act, there may be disagreement about the level of punishment it deserves.

The process of translating this abstract desire to punish into a concrete sentence may be influenced by further factors, including salient norms and the sanctioning practices of a person's home country. (65) For instance, Americans are generally aware that many jurisdictions in their country employ a death penalty, whereas Germans and Canadians are similarly aware that no jurisdictions in their countries do. This could make American participants more likely to assign the death penalty (and higher sentences in general) while making German and Canadian more likely to refrain.

There are, therefore, three constructs of interest: the perceived moral wrongfulness of an action, the (qualitative or abstract) desired punishment severity, and the (quantitative or concrete) sentence length.

Though past work has often shown some differences on concrete sentencing severity across countries, (66) there is reason to expect a degree of cross-national consistency on the perceived moral wrongfulness of major and minor offenses. Following up on the pioneering work by Thorsten Sellin and Marvin E. Wolfgang, (67) cross-cultural studies have generally revealed consistency in the perceived "seriousness" of offenses across both national borders, (68) and, within country, across the lines of gender, race, class, and occupation. (69) We may therefore not see cross-cultural differences on the morality questions even if the cultures differ in sentencing attitudes.

One domain in which this cross-cultural consensus on crime seriousness breaks down is that of victimless offenses, particularly drug crimes and those related to sexual immorality. (70) As noted above, the three countries have widely divergent policies on drug use and prostitution, so we might expect substantial differences on the moral wrongfulness measures in this category with Germans rating these activities as more morally acceptable (or at least more morally neutral) than do Americans. The excessive self-defense scenarios may produce a reversed pattern, with Americans rating those acts as more morally acceptable than Germans.

2. Punishment-Related Attitudes

As noted above, one reason why people might translate equal moral outrage at an offense into differing levels of punishment is if they have different views about the state of society or the goals of punishment. For example, believing that the world is a more dangerous place and that crime is on the rise could be expected to make people harsher toward criminal offenses. (71) Beliefs about the effectiveness of the courts and the crime rate have also been shown to strongly predict punitive attitudes. (72) It could be the case that people from some cultures are more interested in retribution and less interested in rehabilitation than people from others. Of particular interest is the level of support for rehabilitation. Much has also been made of the rising and falling support for rehabilitation among Americans over the last few decades. (73) If Americans are especially punitive, it would be useful to know if that punitiveness is linked to a lesser degree of support for rehabilitation or a greater emphasis on retribution.

We therefore measured support for four punishment motives (retribution, rehabilitation, incapacitation, and deterrence) and three indices of concern about crime: fear of crime, belief that the crime rate is rising, and belief that crime is a threat to social values.

3. Perceptions of Descriptive Norms

Another factor that may independently affect sentencing judgments is perceptions of norms. It has been well established in social psychology that a perceived norm can substantially influence behavior. (74) Over time, even people who disagree with the perceived position of the majority may alter their behavior to better fit in.75 Were Germans to believe the average German was especially lenient while Americans believed the average American was especially harsh, this type of social influence could contribute to cultural differences in sentencing even in the absence of differences in moral judgment or punishment motives. As an initial exploration of this possibility, we tested whether participants believed that the average person in their country would assign the same sentence as they did to a select number of the scenarios.


1. Participants

The sample was recruited via Greenfield Online (now Toluna). Greenfield had established panels in each of our target countries and used similar methods of recruitment and panel management across populations, minimizing opt-in sampling biases. Participants had previously registered with Greenfield and agreed to complete online surveys. Within these panels, Greenfield used weighted probabilistic sampling to produce a survey population with the desired characteristics. Specifically, we requested census-representative distributions of age, race, and gender with an aim of recruiting 300 participants from each country who passed attention quality checks. After inviting an appropriate mix of participants to the survey, Greenfield monitored completion rates to ensure that random dropouts did not leave any important subsample underrepresented relative to its share of the relevant national population.

Upon agreeing to complete the survey, participants were redirected from the Greenfield website to our survey hosting platform. There they completed all study materials as described below. Each scenario was presented on its own page, and participants could not go back to change their prior answers once a given page was complete. The punishment attitudes questions were spread across three pages, as in a questionnaire packet, and the demographics were on a further two pages. Throughout the survey, participants could monitor their progress by means of a status bar at the top of the screen. They were directed back to the Greenfield website after completing the survey so they could be compensated for their time.

2. Materials

All materials were administered in English for the American and Canadian samples and in German for the German sample. In order to create the German materials, a native German speaker translated the English items into German. Another native German speaker, who was not familiar with the original materials, then translated the German version back into English to check for similarity.

3. Scenarios

As in previous work using vignettes to examine people's punishment attitudes, (76) in each of our scenarios a named perpetrator was described performing a series of actions that might be considered criminal. Participants were told to assume that all of the facts given in the scenarios were known to police and readily provable in a court of law. The scenarios fell into one of four categories: major crimes, minor crimes, victimless crimes, and incidents of excessive self-defense. (77)

Major crimes (N = 4): The major crimes were violent or violence-related offenses against persons or property. These were: a violent and brutal rape that left the victim with broken bones; a premeditated murder in the course of a robbery; distribution of a large shipment of crack cocaine by a drug dealer with gang ties; and a brutal armed robbery home invasion that netted $12,000 in cash and property.

Minor Crimes (N = 4): The minor crimes were nonviolent offenses against persons or property. These were: a tax fraud of $19,000 by a self-employed businessman; petty theft of $200 worth of jewelry from a street fair vendor; a fraudulent insurance scam that targeted elderly residents; and a simple vandalism that caused no lasting damages.

Victimless crimes (N = 2): In the prostitution scenario, a middle-aged man solicits a female prostitute (man as the target actor); and in the drug user scenario, an employed man with no violent criminal record uses cocaine on a regular basis.

Excessive self-defense (N = 4): In each of these scenarios, actors use lethal force to defend another person, themselves, or their property under circumstances that are generally prohibited by law. There were four of these: a man shooting the fleeing rapist of another; a man whose car is being stolen shooting the thief; a woman killing her own rapist out of revenge; and a man defending himself against a mugging with lethal force when retreat was possible.

4. Dependent Variables

Each scenario was accompanied by several assessment questions. The first question asked participants whether the actor should be held criminally liable for his or her actions (yes/no). The second question asked participants how severely the actor should be punished on a 7-point scale that ranged from 1 (Very Leniently/No Punishment) to 7 (Maximally), where "maximally" is the most severe punishment the participant would ever assign an offender. This was the abstract punitiveness scale. The third question asked participants to morally evaluate the actor's behavior on a 9-point scale that ranged from 1 (Absolutely Morally Praiseworthy) to 3 (Morally Acceptable) to 7 (Morally Wrong) to 9 (Absolutely Morally Reprehensible).

If participants answered "yes" to the criminal liability question, they were also asked to assign a sentence to the offender. This 11-point scale was labeled: 1 (Community Service, Monitored Probation, Monetary Fines, Other Non-Prison Punishment), 2 (2 weeks in prison), 3 (2 months in prison), 4 (6 months in prison), 5 (1 year in prison), 6 (3 years in prison), 7 (7 years in prison), 8 (15 years in prison), 9 (30 years in prison), 10 (life in prison), and 11 (Death). If the participants answered that the actor should not be held liable in a given case, their responses were coded as 0, creating a total range of 0-11.

5. Individual Difference Questions

Prior to assessing the scenarios, participants completed a number of individual difference scales. Four of these scales assessed punishment motives: retribution, rehabilitation, deterrence, and incapacitation. (78) We also included three constructs measuring crime-related concern: belief that crime was on the rise, (79) views of crime as a threat to society, (80) and fear of crime. (81) The questions from these scales are attached as Appendix B. (82)

6. Demographic Questions

Basic demographics were collected including: age, gender, ethnic identification, religious affiliation, religiosity (7-point scale), highest completed level of education, self-described socio-economic class, short description of occupation, state or province of residence, postal code, and political orientation on a scale ranging from liberal (1) to conservative (7). (83)


Greenfield Online invited participants to the study via e-mail. Upon entering, participants were told that the study concerned criminal justice attitudes and that they would be asked to assess a series of vignettes. They were then assured of their anonymity and completed consent forms.

The individual differences questions were then administered across three pages. Two "dummy" questions were included at the bottom of the first page with instructions that they were to be left blank. These were intended as attention checks. If a participant answered both of these questions, the survey would display an error message preventing the participant from continuing until the answers were removed.

Following the attitudinal questions, there was an instruction page introducing the scenarios. To help calibrate participants to the range of scenarios, they were asked to rate the wrongfulness of a brutal and intentional murder and of a vandalism that caused no irreparable harm. The data from these questions was not analyzed. The scenarios were then presented in fixed order. After each scenario, participants completed the punishment questions as described above.

For one scenario from each of the four crime categories, participants were also asked to consider the sentences that they believed would be assigned by the average American, the average Canadian, and the average German. These comparison questions came after the participants made their own sentence judgments. Participants gave their sentence estimates for the average member of each culture on scales with the same labels they used for their own sentence judgments.

Following all scenario judgments, participants were given the demographic questions and an opportunity to make free response comments.



Following guidelines provided by Greenfield Online, participants were excluded from the analyses if their completion time was less than half that of the median (median = 24 minutes), they responded to either of the two questions that participants were instructed to leave blank (see Part III.D supra), or if their scores on two scales with counterbalanced items, prior to being reverse coded, had a standard deviation of below .5. (84) Removing "speeders" and straight-line respondents is normal practice for online surveys. (85)

In addition, due to the length and complexity of the survey, we screened out participants who gave sentences that were extreme outliers on three of the scenarios. Participants were excluded if they assigned a sentence of one year or less for intentional murder, six months or less for rape, or more than seven years for petty theft. These cutoff points were well outside the normal response range. Across all three scenarios, this check removed 3.5% of American respondents, 3.6% of Canadian, and 4.1% of German.


This final sample contained 1,041 participants of whom 350 were American (174 male, 176 female), 362 were Canadian (150 male, 212 female) and 329 were German (174 male, 155 female). Ages were fairly similar across cultures, though the German sample was slightly younger: American, M= 49.89, SD = 15.75, Median = 51; German, M= 40.68, SD = 15.20, Median = 38; Canadian, M= 47.45, SD = 13.96, Median = 49. The particular distributions of self-reported social class (American: 12% upper-middle class, 52% middle class, 27% lower-middle class, 10% lower class; Canadian: <1% upper class, 14% upper-middle class, 58% middle class, 23% lower-middle class, 5% lower class; German: 1% upper class, 13% upper-middle class, 61% middle class, 21% lower-middle class, 5% lower class), education (American: 2% less than high school, 38% high school graduate, 27% some college/2-year degree, 24% 4-year degree, 8% graduate degree; Canadian: 3% less than high school, 32% high school graduate, 35% some college/2-year degree, 22% 4-year degree, 7% graduate degree), (86) and ethnicity (American: white 87%, black 10%, South or East Asian 2%, other 3%; Canadian: white 94%, black 1%, South or East Asian 4%, indigenous 2%, other 2%, French Canadian 10%; German: 100% German) (87) were generally similar to each country's population statistics.

For all three countries, it was possible to compare the proportion of our sample from each state or province with national census figures. Forty-six of the fifty states in the United States were represented in our sample (Alaska, Delaware, Hawaii, and Wyoming were excluded). The proportion of the sample from a state was, on average, only .6% different from the proportion of the national population from that state. For example, Alabama made up .9% of our sample while it represents 1.5% of the national population.

All of the ten Canadian provinces were represented. The average deviation score for Canadian provinces was higher at 4%, largely due to the overrepresentation of Ontario relative to Quebec. There were, however, no significant differences between the people from these provinces on any of the scenario or attitudinal measures. The sample also contained participants from each of the sixteen German states, and no region was substantially over or underrepresented. The mean deviation score for Germany was 1.71%. (88)


Analysis Strategy. Data were analyzed using a series of ANOVAs. This approach allowed for both maximum fidelity to the raw data and relatively straightforward presentation of the results. Since demographic data was collected, however, it was also possible to conduct our analyses as ANCOVAs controlling for age and gender. This approach yielded nearly identical results. The few differences are noted at the close of the results section, and the full output is available from the authors.

1. On Scenario Judgments

Major Crimes. Table 1 displays punitiveness and morality ratings for major crimes (i.e., murder, rape, drug dealing, armed robbery) among Americans, Canadians, and Germans. Americans and Canadians are indistinguishable from each other on abstract punitiveness for all offenses and harsher than Germans on all offenses except murder. This suggests that Americans and Canadians feel more punitive than Germans when considering serious crimes. Also, American and Canadian concrete sentences are longer than German in all cases, with American sentences significantly exceeding Canadian sentences in cases of rape and murder, albeit by very small margins. In the case of murder, this may be due to differential willingness to impose the death penalty: 42% of Americans choose death for the murderer whereas only 32% of Canadians and 12% of Germans do.

On major crimes, therefore, it appears that Americans and Canadians are translating their roughly equivalent levels of punitive intent into roughly equivalent sentences, though Canadians may have a higher threshold for imposing the death penalty. Very clear on both the abstract and concrete measures, however, is that Americans and Canadians are much more similar to each other than to Germans. There are, however, no meaningful differences on the measure of the scenarios' moral wrongfulness. Thus, the cultural difference in punitiveness between North American and German respondents cannot be explained by systematic differences in perceived moral wrongfulness.

For the murder scenario, we had also asked participants to project the concrete sentences that would be assigned by the average person from their own countries. Here, we find that people from all three cultures underestimate the punitiveness of their average compatriot. This is most true of the Canadians, and again this may be related to use of the death penalty; perhaps Canadians underestimate the extent to which their fellow citizens are willing to endorse its use (see Table 2).

Minor Crimes. For minor crimes (see Table 3), Americans and Canadians are again more punitive than Germans and generally do not differ significantly from each other, at least on the abstract punishment scale. On the concrete punishment scale, the pattern was largely but not entirely the same. On one of the four minor offenses, tax fraud, Americans assign a slightly longer sentence than do Canadians, and on minor vandalism, all participants are near the scale's lower extremity. The greatest difference in concrete sentences is for the offense of petty theft: whereas American and Canadian participants assign about two to six months in prison, German participants assign about two weeks.

Again, there is very little difference on the morality ratings across cultures. Americans rate tax fraud and petty theft slightly more morally wrong than do Germans (with Canadians being intermediate), but the effect sizes are very small. There are no significant differences on the other two offenses.

When participants project sentences for petty theft, the pattern is somewhat different than for murder. Americans project the punitiveness of their fellow citizens correctly; there is no significant difference between estimates and reality. Canadians then underestimate the punitiveness of their populace, while Germans overestimate the punitiveness of theirs, in each case by a substantial margin (see Table 2). Taken together, the norms estimated for major and minor crimes do not match the pattern we would have expected based on a norm-adherence account.

Victimless Crimes. For the victimless offenses, our question was whether Americans criminalized a broader range of conduct than Canadians or Germans. There is some support for this hypothesis when contrasting Americans to Germans but, again, Canadians generally do not differ from Americans (see Table 4). On prostitution and drug use, Canadians and Americans assign greater abstract punishments and longer concrete sentences than do Germans. This difference was also reflected in the desire to impose criminal liability. Consistent with their country's legal policy, only 5% of Germans thought that soliciting a prostitute should be a crime whereas 52% of Americans and 48% of Canadians did. On drug use, this pattern is substantially weaker but still present with 41% of Germans wishing to criminalize the conduct as opposed to 57% of Canadians and 62% of Americans. (89)

The pattern on morality is somewhat more complex: Americans believe drug use to be slightly more deserving of moral condemnation than do Canadians and Germans, who do not differ. On prostitution, however, Canadians and Americans are both much more likely to morally condemn the actor than are Germans, with Americans being even more likely to do so than Canadians. Recall that the morality scale labeled point 3 as "morally acceptable" and point 7 as "morally wrong," with 1 and 9 providing more extreme alternatives. On drug use, these results suggest that all cultures believe that using cocaine is more wrong than right, if not by a substantial amount. (90) Soliciting a prostitute, however, produces a strong divergence in moral judgment. Americans and Canadians believe that doing so is more wrong than right whereas Germans are much closer to saying it is morally acceptable. (91) Again, these results reflect the large differences in the legal treatment of these acts across countries.

Participants from all three cultures overestimate the sentences that their peers would assign to the drug user (see Table 2). The magnitude of the effect is such that the Germans, who are the least punitive populace in reality, believe that the average German is more punitive than is the actual average American. In a previous study, it had been shown that Americans generally assume that the average American is more punitive toward drug users than they are themselves. (92) These results replicate that pattern and show (1) that it generalizes across cultures (Canadians and Germans also overestimate the punitiveness of their respective cultures); and (2) that it appears to be specific to drug crimes.

Excessive Self-Defense. The crimes of self-defense were hypothesized to produce an opposing pattern to the other categories; Americans were expected to be especially forgiving of those who took the law into their own hands. The results support this hypothesis. Across all four scenarios, Americans assign shorter sentences, are lower in abstract punitiveness, and rate the acts as less deserving of moral condemnation compared to Germans and, in most cases, to Canadians (see Table 5). For instance, American participants assign a concrete sentence of about two months in prison to lethal self-defense when retreat was possible, whereas German participants assign a sentence of more than six months. Canadians, in a change from the major and minor offenses, were more similar to the Germans here. There was, however, some ambiguity for the two cases in which a rapist was shot. For both of these, Canadians assigned lower concrete sentences than did Germans and, for the case in which a rape victim shoots her fleeing attacker, also lower abstract punitiveness and less moral condemnation. (93)

As was the case with drug use, people from all samples overestimated the punitiveness of their fellows toward the person who shot the fleeing rapist (see Table 2).

2. On Justice Attitudes

We had considered that cultures could differ on sentencing--even without disagreeing on the moral wrongfulness of offenses--if they had different prior beliefs about criminal justice issues. On major and minor crimes, Americans and Canadians were more punitive than Germans despite agreeing about each act's degree of moral wrongfulness. Given that sentencing pattern, we would expect that Americans and Canadians would be more concerned about crime on each of our three indicators. That is exactly what we observe: both Americans and Canadians are more likely than Germans to be afraid of crime, to believe that crime is on the rise, and to see crime as a threat to values (see Table 6).

We would also have expected that, relative to Germans, Americans and Canadians would place greater emphasis on retribution (and possibly incapacitation and deterrence) as opposed to rehabilitation. This pattern was only weakly supported. Americans and Canadians were both slightly more retributive than Germans, (94) but the magnitude of this effect is small. There were no differences on support for rehabilitation, deterrence, or incapacitation. Thus, systematic cultural differences in punitiveness may be related to perceptions of threat (regarding the danger of becoming a victim or regarding an erosion of norms), but they cannot be explained by different endorsements of punishment goals.

3. Alternative Analysis

It was also possible to conduct our analyses as ANCOVAs controlling for the small differences in gender and age across samples. This alternative approach produces nearly identical results. Of all the significant ANOVAs reported in Tables 1, 3, 4, 5, and 6, only a single effect becomes non-significant: the unpredicted effect of country on the perceived morality of petty theft. There is also a new significant effect on the perceived morality of the drug dealer case F(1, 1036) = 6.47, p < .01 such that Americans rate the scenario slightly lower (M = 7.73, SD = .06) than do Canadians (M = 8.00, SD = .06) and Germans (M= 8.01, SD = .06). The significance of the other ANOVAs is unaffected. At the level of post hoc tests, no previously observed difference becomes nonsignificant, apart from the ones related to the morality of petty theft. A complete report of these results is available from the authors upon request.


Public opinion theories of American Exceptionalism predict that public attitudes are aligned with public policies. Since the American government assigns longer prison sentences than Canada and Germany, Americans as individuals should do likewise. They should prefer longer sentences; fear crime more; endorse retribution to a greater extent; and be less interested in rehabilitation. Alternative accounts of justice attitudes focusing on social policy resonances (95) or shared cultural histories (96) would instead predict that Americans and Canadians may be aligned in their punitiveness and that Germany would be unique in its leniency.

When evaluating the merits of these hypotheses, it is important to consider the magnitude of the real world policy disparities that prompted our research. The divergence between the harshness of American penal policy and the leniency of the Canadian and German is not grounded in minor differences in sentence duration. These systems vary by multiples rather than degrees. In order for citizens' attitudes to explain a substantial portion of this variation, Americans would have to be drastically and qualitatively more punitive than Germans and Canadians. Americans would have to assign death where Germans assign ten years, or to incarcerate where Canadians parole.

For both minor and major crimes, the Americans in our sample generally did assign sentences that were more severe (longer in duration) than those of Germans. For murder, the Germans, on average, assigned a sentence of thirty years while the Americans assigned life sentences. For armed robbery, Germans assigned sentence of over six months, Americans of over one year. These differences are statistically significant, but their magnitude does not match the nearly tenfold difference in incarceration rates between Germany and the United States. Even this modest support for the American Exceptionalism hypothesis is undermined, however, by the behavior of the Canadian sample. Canadians were only slightly less punitive than Americans on these crimes and, on many of the offenses, there was no significant difference between the two countries. So American sentences in this study, although more severe than German, are not uniquely punitive. This is also true for victimless crimes: though Americans assign longer sentences to drug use and prostitution than do Germans, they are not significantly harsher than Canadians.

Responses to the individual differences measures also follow this pattern. Although Americans are substantially more concerned about crime than Germans, they do not differ from Canadians. It may be the case that heightened fear of crime is a factor in producing excess punitiveness from both Americans and Canadians--this seems particularly probable given the agreement across all three cultures on the degree of moral wrongfulness of major and minor crimes--but these measures do not support a uniquely American obsession with crime. Interestingly, and in some contrast to the other results, the punishment motives showed remarkably little movement across cultures. Apart from Americans and Canadians being slightly more supportive of retribution than Germans, endorsement of the motives was constant across groups. Though it has been argued elsewhere that Americans do not reject rehabilitation as a goal of punishment, (97) it is still surprising that they endorse it as much as Germans and Canadians, whose systems are much more rehabilitation-focused. (98)

Returning to the four questions we asked in the introduction, we can answer them as follows: (1) Americans are more punitive than Germans, who have significantly different criminal justice policies. The difference in punitive attitudes is smaller than the differences in actual policy, but it is statistically reliable and broadly present. (2) Americans are not, however, generally more punitive than the more culturally similar Canadians, despite their substantial policy differences. Thus, there is no compelling evidence for American Exceptionalism; Americans are never entirely unique. (3) The cultural differences in punitiveness that are found occur among both major and minor crimes, reflecting a generalized divide between North America and Germany rather than one focused on either the most severe or most minor of crimes. (4) These sentencing disparities for baseline crimes cannot be explained by disagreement about the degree of moral wrongfulness perceived in each offense, and therefore are likely due to other factors. A likely suspect is differential fear of crime. Interestingly, differences in punitive intent toward crimes committed out of self-defense and victimless offenses do appear to reflect differences in moral evaluation.

There are several possible explanations for heightened punitiveness among Americans and Canadians. Both the United States and Canada are arguably frontier societies, (99) and this may have some influence on their current citizenries' views toward criminal justice issues. Supporting this conjecture, our Canadian and American samples were more lenient in their punishments of people who were reacting with extreme, often lethal, force in response to attacks on themselves, their neighbors, and their property, even when those attacks went far beyond the bounds of what the legal codes would allow. The Germans punished those crimes more severely than did the Americans in all cases and were significantly more punitive than the Canadians in two out of four cases.

Another explanation for American and Canadian punitiveness may come from their common history as former British colonies and neoliberal economies. In many other domains, most notably social welfare policy, (100) the English-speaking world differs substantially from the democracies of continental Europe and Scandinavia. If these shared ideological bonds were driving the current data, then one would predict that other Anglophone countries such as Britain, Australia, and New Zealand would also show a pattern of heightened punitiveness. There is some past evidence suggesting that this may be the case. (101)

Finally, Americans and Canadians also consume similar, American-produced, media. The movies Canadians go to see, for example, are primarily American. (102) Similarly, Canadian news often carries coverage of American politics. (103) Media plays a critical role in the formation of criminal justice attitudes. (104) Research conducted in Australia, for example, has shown that commercial media consumption is associated with punitive criminal justice attitudes. (105) Since Americans are punitive, cultural diffusion of American ideas and social norms could spread that punitiveness to Canada, even absent any initial Canadian support. Again, this is a theory that could be tested by more detailed examination of citizen attitudes in other Anglophone countries. It is likely, however, that results of such a test would be consistent with the ICVS research showing broad similarity between Canada and the U.K. (106)

Whatever the ultimate origin of the differences between Germans and North Americans, we have obtained some hints of its psychological mechanism. The first evidence was negative: Germans do not view major and minor crimes as less morally wrongful than do Americans and Canadians. This divergence between the morality items and sentencing is in many ways very surprising; past research has very closely linked beliefs about moral wrongfulness and punitive intentions. (107) The fear-of-crime measures provide a more positive indication: Canadians and Americans are generally more concerned about crime than Germans. Perhaps it is merely this concern, and not broad beliefs about morality, that influences Americans and Canadians to allocate more of society's scarce resources to incarceration.

It is interesting that there were consistent biases across cultures in three of the four cases where participants were asked to project sentences for their compatriots. A norm-adherence account of cultural differences in punitiveness would have predicted that Americans should expect other Americans to be more punitive than they were, and Germans that other Germans would be more lenient, with the perceived norm exerting pressure for divergence in each country. (108) Instead, the difference between the actual sentences of our participants and those of the perceived average citizen appear to be a function of crime type rather than nationality. Though this makes norm adherence a poor explanation for cultural differences in sentencing, the intra-culture prediction biases themselves would be an interesting area for future research.

Though it is more common in this literature to discuss how public attitudes may create criminal justice policies, and thus sentencing outcomes, it is also worth considering how policies themselves may impact public attitudes in the American and German cases. There is a substantial body of research documenting people's tendency to justify the status quo, often called the system-justification motive. (109) American punitiveness at the individual level may, in part, be a consequence of Americans' motivation to justify the actuality of their society's punitiveness. Similarly, German individual leniency may in part be a consequence of German societal leniency. The system-justification motive may be particularly relevant to understanding these cultures' views of prostitution, drug use, and the death penalty, as the countries differ publicly and substantially in their treatment of those issues. Previous work has shown that system-justification tendencies can be increased by a variety of factors including system relevancy, dependency, and inescapability. (110) One could test this possibility by having German and American participants assign sentences while under conditions that evoke high or low system justification.

At odds with the possibility of policy affecting attitudes or attitudes affecting policy, however, is the Canadian case. Why are Canadian attitudes in such conflict with Canadian policy? This challenge leads us to believe that our results support an institutional rather than individual explanation for American Exceptionalism in policy; attitudes towards crime and punishment cannot explain the massive policy disparities between the three countries. Since the difference does not appear to lie in the views of individuals, it may be that the political and judicial structures that translate popular will into policy in each country are responsible for these divergent outcomes. Potentially relevant is that the Canadian and German systems leave sentencing decisions to civil servants who are intentionally isolated from the political process, whereas American district attorneys and judges are often elected. (111) Recent research has shown, for example, that elected judges in Washington State tend to assign longer sentences as elections near. (112) And this effect may be particularly strong in partisan (as opposed to open) judicial elections, where the implicit threat of a potential challenger is stronger. (113) Separation from this public pressure in the European and Canadian systems may mitigate the influence of popular will and allow for the kind of divergence observed in Canada.

Other features of the political system may also serve to temper or exacerbate the impact of popular will on politics. Lappi-Seppala, for example, emphasizes how the bipolar struggle underlying American majoritarian democracy, with its constant focus on swing voters, may serve to magnify the effect of popular opinion on policy. (114) He contrasts this with the "consensus" based approach of coalition systems, which have multiple power centers. Though this explanation has limited power in the U.S.-Canadian comparison--Canada's third parties are significant factors, but Canada generally has a majority parliament--it may be valuable in understanding American-European divergences. Other scholars have suggested this difference in popular responsivity could be a function of such diverse elements as the short terms of American legislative officials and the constitutional separation of powers, (115) as well as the emphasis on state rather than national control of criminal justice issues in America. (116)

Finally, differences in political culture, independent of structure, may be relevant. Canadian and German, but not American, political culture favors the opinion of experts over that of the public in criminal justice matters. (117) Canada, for example, has not adopted a policy goal of maintaining public confidence in the justice system and has a greater emphasis on evidence-based rehabilitation. (118) Also, some prior work has suggested that countries with common law, rather than civil law, court systems tend to have harsher sentencing regimes. (119)

In recognizing the role of institutional factors, our work is in general agreement with some prior theories about the Canadian and German justice systems. (120) New here, however, is the finding that attitudinal differences may play a nearly negligible role in understanding cross-system sentencing policy. The present results particularly emphasize the importance of studying the Canadian system. Although Canada has U.S.-like crime attitudes and crime rates, it is much closer to Germany in its approach to criminal justice issues. There is, therefore, a difference in penal policy that is not being driven by contrary popular sentiment. Instead, it only arises at the societal/political level. This is also one area in which public norms, distinct from public attitudes, may play a role. Recall that, while American and Canadian sentences for petty theft were equivalent, Canadians thought that other Canadians would actually be more merciful than they themselves were. People are often sensitive to perceived norms, even if those perceptions are inaccurate. (121) If this pro-leniency misperception generalizes to other moderate severity crimes, it may affect the political discourse in Canada.


By failing to find evidence for a robust theory of American Exceptionalism at the attitudinal level, the present research should encourage scholars to focus more on the institutional and political factors that shape criminal justice policy. The large differences between American and Canadian approaches to justice issues do not appear to be solely or even primarily rooted in the attitudes of their citizenry. There is a large and growing body of work that considers how the norms of political discussion and procedures for government decisionmaking affect criminal justice policy, (122) and our results suggest that this area is ripe for further study.
Appendix A


Murder Late one night, Jack comes into a convenience store
 and, pulling a gun, demands the money from the
 register. As the clerk looks on helplessly, Jack
 grabs the money. Jack then shoots and kills the clerk
 to avoid leaving a witness. This had been Jack's
 intention from the beginning.

Rape Doug sees Julie walking to her car after a long day
 at work. The parking lot is dark, and she is taken
 by surprise when he grabs her. He drags her out of
 sight and rapes her, breaking her arm and several
 ribs. He flees, leaving her bleeding. Her physical
 injuries completely heal, but only after extensive
 medical treatment.

Armed Andrew forced his way into a suburban home at
Robbery gunpoint. Threatening the family, he made them hand
 over cash, jewelry, and electronic devices worth
 approximately $12,000. Andrew leaves without harming
 the family.

Drug Chris is a drug dealer. He recently finished
Dealing distributing a shipment of 5 kilograms of
 crack-cocaine to dealers in his target neighborhood.
 He has ties to a local violent gang that protects
 his dealers and keeps his neighborhood free of


Tax Fraud Frank is a self-employed businessman. He deliberately
 underreports his earnings and overreports his expenses
 to pay fewer taxes. He has avoided paying $19,000 in
 taxes by doing this.

Insurance Jake is a con artist. He travels around residential
Fraud communities selling a fraudulent "health insurance
 supplement" that never pays out to poor elderly
 residents. He has defrauded an estimated 90 people in
 this fashion for anywhere from $350-$900 each.

Petty Tom is a man in his early twenties. When he is at a
Theft street fair one weekend, he sees a nice piece of
 jewelry at a vendor. He grabs it and runs, escaping
 the scene. The jewelry was worth $200, and Tom intends
 to sell it for profit.

Vandalism Mike was bored one day. He bought a can of washable
 spray paint at the local hardware store and paints
 random images on nearby public buildings. The next
 day maintenance workers are able to remove the paint
 without leaving damage.


Shooting Sean is returning from a weekend of hunting up in the
Another's mountains. As he drives into town late at night, he
Fleeing hears screams coming from one of his neighbor's
Rapist houses. A man is running out the door and a visibly
 battered woman is going after him, crying that he
 raped her. Unable to stop him otherwise, Sean takes
 his rifle from his car and shoots the fleeing rapist,
 killing him.

Revenge One night while her husband is at work, Suzanne is
for Own asleep at home in their bedroom. Suzanne is awakened
Rape out of a sound sleep by a man who immediately
 overpowers her. The man quickly subdues her, rapes
 her, and starts to leave. As the man reaches the door,
 Suzanne reaches into the drawer of her night stand,
 and pulls out the gun her husband uses for target
 shooting. She is enraged and, thinking clearly,
 decides she wants revenge. Suzanne chases the man,
 intending to kill him. As the man reaches the door
 out of the house, Suzanne catches up to him and
 shoots. He is killed immediately.

Self George is walking down the street heading home from
Defense, the store. He sees a man walking towards him and
Retreat begins to worry that the man looks dangerous, so he
Possible turns off the street. He finds himself near a well
 lit shopping area, and a police officer is in easy
 shouting distance. Looking around, George sees a long
 lead pipe, and he calmly decides to fight the
 suspicious man. The man comes around the corner with a
 knife and demands George's money. George swings the
 pipe at the man, hitting him on the head. The man
 later dies from the injury.

Defense Adam lives in a house in the suburbs. Late one night
of Own as he is going to bed, he hears scuffling coming from
Property his driveway. Looking out the window, he sees that a
 man has just broken the window of his car and is
 trying to get the door open. Adam grabs his hunting
 rifle off the mantle and runs outside to confront the
 man. When the thief sees Adam, he begins to flee.
 Adam calls for him to stop and, when the thief
 continues to run, Adam shoots him. The man later
 dies of his injury.


Drug Use Joe is a recreational cocaine user. He buys a small
 quantity of cocaine from his friend for personal use.
 Joe has never committed a violent crime and has a
 steady job.

Prostitution John is a middle-aged businessman. After a long week
 at work, he seeks out a prostitute on the streets of
 the local red light district. Finding a likely
 prospect, he solicits sex from her. She agrees, and
 they have sex.

Appendix B

Item Wording


* How much harm a crime caused should be the major factor that determines how long of a sentence a criminal receives.

* The primary purpose of our legal system is to make criminals pay for their offenses.

* Those who hurt others deserve to be hurt in return.

* We have a moral obligation to punish those who break the law.


* Rehabilitating a criminal is just as important as making a criminal pay for his or her crime.

* The only way to reduce crime in our society is to punish criminals, not try to rehabilitate them.

* Rehabilitative measures for prisoners (such as social therapy, vocational training etc.) have proven to be a failure.

* The most effective and humane cure to the crime problem in America is to make a strong effort to rehabilitate offenders.


* Punishing criminals is the only way to stop them from engaging in more crimes in the future.

* Stiffer jail sentences will help reduce the amount of crime by showing criminals that crime does not pay.

* Punishing criminals will reduce crime by setting an example and showing others that crime does not pay.


* We should put criminals in prison because it removes them from the community and protects citizens from further crimes that they might commit.

* Since most criminals will commit crimes over and over again, the only way to protect society is to put these criminals in jail and throw away the key.

* We should put criminals in jail so that innocent citizens will be protected from criminals who will victimize, rob, or hurt them if given the chance.


* The murder rate in this country is increasing.

* The home break-in rate in this city is increasing.


* Crime is a threat to the moral cohesion of our society.

* I'm afraid that crime threatens the values of our society.

* Crime represents a great danger to society.


At one time or another, most of us have experienced fear about becoming the victim of a crime. Some crimes probably frighten you more than others. We are interested in how afraid people are in everyday life of being a victim of different kinds of crimes. Please rate your fear of being a victim of each of these crimes on the adjacent scales.

* Having someone break into your home while you're there?

* Being murdered?

* Being attacked by someone with a weapon?

* Being robbed or mugged on the street?

(1) Warren E. Burger, Annual Report to the American Bar Association by the Chief Justice of the United States, reprinted in 67 A.B.A.J. 290, 290 (1981).

(2) See generally, e.g., JONATHAN SIMON, GOVERNING THROUGH CRIME: HOW THE WAR ON CRIME TRANSFORMED AMERICAN DEMOCRACY AND CREATED A CULTURE OF FEAR (2007); JAMES Q. WHITMAN, HARSH JUSTICE: CRIMINAL PUNISHMENT AND THE WIDENING DIVIDE BETWEEN AMERICA AND EUROPE 41-68 (2003); see generally Carol S. Steiker, Capital Punishment and American Exceptionalism, in AMERICAN EXCEPTIONALISM AND HUMAN RIGHTS 57 (Michael Ignatieff ed., 2005) (evaluating cultural, regional, and institutional theories).

(3) See KATHERINE BECKETT, MAKING CRIME PAY: LAW AND ORDER IN CONTEMPORARY AMERICAN POLITICS 3-13, 28-43 (1997); James O. Finckenauer, Crime as a National Political Issue: 1964- 76, 24 CRIME & DELINQ. 13, 13-27 (1978).

(4) E.g., Tim Newburn & Trevor Jones, Symbolic Politics and Penal Populism. The Long Shadow of Willie Horton, 1 CRIME MEDIA CULTURE 72, 78-85 (2005).

(5) See generally JULIAN V. ROBERTS ET AL., PENAL POPULISM AND PUBLIC OPINION: LESSONS FROM FIVE COUNTRIES (2003) (reviewing literature on sentencing and crime attitudes).

(6) See id. at 61-75; Stephanos Bibas, Essay, Transparency and Participation in Criminal Procedure, 81 N.Y.U.L. REV. 911,939-45 (2006); Newburn & Jones, supra note 4, at 78-85.

(7) See Daniel S. Nagin et al., Imprisonment and Reoffending, 38 CRIME & JUST. 115, 117-18 (2009); see also ROBERTS ET AL., supra note 5, at 3-20. See generally MICHAEL H. TONRY, THINKING ABOUT CRIME: SENSE AND SENSIBILITY IN AMERICAN PENAL CULTURE (2004) (describing the American penal system); RoY WALMSLEY, INT'L CTR. FOR PRISON STUDIES, WORLD PRISON POPULATION LIST 1-6 (9th ed. 2009); Patrick A. Langan, Crime and Punishment in the United States, 1981-1999, 33 CRIME & JUST. 123, 125-31 (2005); Steiker, supra note 2, at 57-59.

(8) See generally Steiker, supra note 2 (describing its possible origins); see also Michael Tonry, Why Are U.S. Incarceration Rates So High?, 45 CRIME & DELINQ. 419, 431-34 (1999). The term American Exceptionalism has been promoted, see generally ALEXIS DE TOCQUEVILLE, DEMOCRACY IN AMERICA (Harvey C. Mansfield & Delba Winthrop eds. & trans., University of Chicago Press 2000) (1835) (coining the term), expanded, see generally WHITMAN, supra note 2 (linking it to broad historical trends); Steiker, supra note 2 (using it to refer to a constellation of factors), and critiqued, e.g., David Garland, Capital Punishment and American Culture, 7 PUNISHMENT & SOC'Y 347, 349-50, 365-66 (2005), by a variety of authors, all of whom have their own unique take on its defining elements. Particularly, there is dispute about whether American attitudes have always been distinct from European attitudes, see generally WHITMAN, supra note 2, or whether whatever differences may exist are the product of post-1960s political developments and therefore are more fleeting, see Garland supra. In this Article, we use American Exceptionalism to refer to the apparent qualitative distinctions between American and continental European approaches to justice issues generally and do not take a firm position on the question of their origins.


(10) Jeffrey M. Jones, Americans Still Perceive Crime as on the Rise." Two-Thirds Say Crime Increasing in U.S., 49% in Their Local Area, GALLUP (Nov. 18, 2010),

(11) ROBERTS ET AL., supra note 5, at 7.

(12) E.g., id. at 3-20.


(14) WHITMAN, supra note 2, at 41-68.

(15) ZIMRING, supra note 13, at 89-118.

(16) GARLAND, supra note 13, at 152-61, 169-70.

(17) SIMON, supra note 2, at 3-12.

(18) Id. at 20.

(19) Id. at 33-75.

(20) See generally FREDERICK J. TURNER, THE FRONTIER IN AMERICAN HISTORY 30 (1920); Shinobu Kitayama et al., Voluntary Settlement and the Spirit of Independence: Evidence from Japan "s "Northern Frontier," 91 J. PERSONALITY & SOC. PSYCHOL. 369, 369-70 (2006).

(21) TURNER, supra note 20, at 78, 212, 271-72.

(22) See WILKINS, supra note 13; Michael Cavadino & James Dignan, Penal Policy and Political Economy, 6 CRIMNOLOGY & CRIM. JUST. 435,449-52 (2006); Leslie T. Wilkins & Ken Pease, Public Demand for Punishment, 7 INT'L J. SOC. & SOC. POL'Y 16, 21-22, 24-29 (1987).

(23) BECKETT, supra note 3, at 28-43; Katherine Beckett & Theodore Sasson, Crime, Politics and the Public: The Sources of Mass Incarceration in the U.S.A., 29 JAPANESE J. SOC. CRIMINOLOGY 27, 39-45 (2004); Katherine Beckett & Bruce Western, Governing Social Marginality: Welfare, Incarceration, and the Transformation of State Policy, 3 PUNISHMENT & SOC'Y 43, 55 (2001).

(24) Tapio Lappi-Seppala, Explaining Imprisonment in Europe, 8 EUR. J. CRIMINOLOGY 303,310, 313 (2011).

(25) SIMON, supra note 2, at 33-74; WHITMAN, supra note 2, at 41-68; ZrMRrNG, supra note 13, at 3-16.

(26) See generally TURNER, supra note 20; WILKINS, supra note 13, at 24-29; Cavadino & Dignan, supra note 22, at 440-43; Wilkins & Pease, supra note 22.

(27) See generally THE CULTURAL LWES OF CAPITAL PUNISHMENT: COMPARATIVE PERSPECTIVES (Austin Sarat & Christian Boulanger eds., 2005) (examining views toward the death penalty).

(28) Jan W. de Keijser & Henk Elffers, Cross-Jurisdictional Differences in Punitive Public Attitudes?, 15 EUR. J. ON CRIM. POL'Y & RES. 47, 56--57 (2009); Julian V. Roberts & Anthony N. Doob, News Media Influenees on Public Views of Sentencing, 14 L. & HUM. BEHAV. 451,456 (1990).

(29) E.g., Mike Hough & Julian V. Roberts, Sentencing Trends in Britain. Public Knowledge and Public Opinion, 1 PUNISHMENT & SOC'Y 11, 14-15 (1999) (discussing the British Crime Survey).

(30) See ROBERTS ET AL., supra note 5, at 21-34.

(31) As Roberts and colleagues observed, "no one would consider addressing a question of such simplicity to criminal justice professionals. Their responses would be complex and generate many other questions, such as 'What kinds of offenses?' and 'What kinds of offenders?'" Id. at 25.

(32) Carissa Byrne Hessick, Mandatory Minimums and Popular Punitiveness, CARDOZO L. REV. DE NOVO 23, 24-28 (2011); see also HOUGH & ROBERTS, supra note 9, at 27-30. See generally Julian V. Roberts & Mike Hough, Custody or Community? Exploring the Boundaries of Public Punitiveness in England and Wales, 11 CRIMINOLOGY & CRIM. JUST. 181 (2011) (discussing the role of mitigating and individuating factors).

(33) See Roberts & Doob, supra note 28, at 463 tbl.2; Douglas R. Thomson & Anthony J. Ragona, Popular Moderation Versus Governmental Authoritarianism: An Interactionist View of Public Sentiments Toward Criminal Sanctions, 33 CRIME & DELINQ. 337, 351 (1987).

(34) For reporting on the 1996 results, see Pat Mayhew & John van Kesteren, Cross national Attitudes to Punishment, in CHANGING ATTITUDES TO PUNISHMENT: PUBLIC OPINION, CRIME AND JUSTICE 63, 66-76 (Julian V. Roberts & Mike Hough eds., 2002).

(35) For reporting on the 2004-2005 results, see JAN VAN DIJK ET AL., CRIMINAL VICTIMISATION IN INTERNATIONAL PERSPECTIVE: KEY FINDINGS FROM THE 2004-2005 ICVS AND EU ICS 149 tbl.32 (2007).

(36) E.g., GARLAND, supra note 13, at 9-38.

(37) E.g., Ernest Drucker, Population Impact of Mass Incarceration Under New York's Rockefeller Drug Laws: An Analysis of Years of Life Lost, 79 J. URB. HEALTH: BULL. N.Y. ACAD. MED. 434, 435 (2002).

(38) See generally SIMON, supra note 2, at 164-72.

(39) ROBERTS ET AL., supra note 5, at 62459; SIMON, supra note 2, at 33-74; ZIMRING, supra note 13, at 89-118.

(40) E.g., Ariane Colemont et al., Five-Factor Model Personality Dimensions and Right Wing Attitudes: Psychological Bases of Punitive Attitudes?, 50 PERSONALITY & INDIVIDUAL DIFFERENCES 486, 489 (2011). See generally Francis T. Cullen et al., Is Rehabilitation Dead? The Myth of the Punitive Public, 16 J. CRIM. JUST. 303 (1988) (describing overall level of support for rehabilitation and punishment and their potential motivating factors); Sandra S. Evans & Joseph E. Scott, The Seriousness of Crime Cross-Culturally: The Impact of Religiosity, 22 CRIMINOLOGY 39, 52-54 (1984); Tom R. Tyler & Robert J. Boeckmann, Three Strikes and You Are Out, but Why? The Psychology of Public Support for Punishing Rule Breakers, 31 LAW & SOC'Y REV. 237, 257 (1997).

(41) Thomas Weigend, Sentencing and Punishment in Germany, in SENTENCING AND SANCTIONS IN WESTERN COUNTRIES 188, 191 (Michael Tonry & Richard S. Frase eds., 2001); Gary M. Friedman, Comment, The West German Day-Fine System: A Possibility for the United States?, 50 U. CHI. L. REV. 281,291-94 (1983) (describing the system in which the fine for any particular offense is not a fixed sum but varies according to the wealth and income of the offender).

(42) FED. STATISTICAL OFFICE GERMANY, STATISTICAL YEARBOOK 2012, at 301 (Statistisches Bundesamt Deutschland 2012).

(43) Alexes Harris et al., Drawing Blood from Stones: Legal Debt and Social Inequality in the Contemporary United States, 115 AM. J. Soc. 1753, 1769-76 (2010).


(45) WALMSLEY, supra note 7, at 3 tbl.2, 5 thl.4.

(46) Michael Tonry, Why Aren't German Penal Policies Harsher and Imprisonment Rates Higher?, 5 GERMAN L.J. 1187, 1200 (2004). See generally WHITMAN, supra note 2.

(47) E.g., Joshua C. Cochran & Alex R. Piquero, Exploring Sources of Punitiveness Among German Citizens, 57 CRIME & DELINQ. 544, 557 tbl.2 (2011).

(48) See Brandon C. Welsh & Mark H. Irving, Crime and Punishment in Canada, 1981 1999, 33 CRIME & JUST. 247,248 (2005).

(49) WALMSLEY, supra note 7, at 3 tbl.2.

(50) See generally Julian V. Roberts & Jane B. Sprott, Exploring the Differences Between Punitive and Moderate Penal Policies in the United States and Canada, in 4 CRIME AND CRIME POLICY: INTERNATIONAL PERSPECTIVES ON PUNITIVITY 55 (Helmut Kury & Theodore N. Ferdinand eds., 2008); Trevor Sanders & Julian V. Roberts, Public Attitudes Toward Conditional Sentencing. Results of a National Survey, 32 CANADIAN J. BEHAV. SCI. 199, 199-200 (2000).

(51) See VAN DIJK ET AL., supra note 35, at 149 tbl.32.

(52) See generally GOSTA ESPING-ANDERSEN, THE THREE WORLDS OF WELFARE CAPITALISM (1990) (describing the differing approaches taken by Western, industrialized countries).

(53) VAN DIJK ET AL., supra note 35, at 148 fig. 31, 152.

(54) Roberts & Sprott, supra note 50, at 68.

(55) Timothy F Hartnagel & Laura J Templeton, Emotions About Crime and Attitudes to Punishment, 14 PUNISHMENT & SOC'Y 452, 466 (2012).

(56) Roberts & Sprott, supra note 50, at 72-73.

(57) SIMON, supra note 2, at 164-72.


(59) Garrett Epps, Further Developments, Any Which Way but Loose: Interpretive Strategies and Attitudes Toward Violence in the Evolution of the Anglo-American "Retreat Rule," 55 LAW & CONTEMP. PROBS. 303, 305 (1992). But see Eugene Volokh, Duty to Retreat and Stand Your Ground." Counting the States, VOLOKH CONSPIRACY (July 17, 2013, 10:11 AM), that the "substantial majority view among the states, by a 31-19 margin, is no duty to retreat").


(61) See generally TURNER, supra note 20 (arguing that frontier violence and rough justice had permanent effects on American culture).

(62) See generally RICHARD E. NISBETT & DOV COHEN, CULTURE OF HONOR: THE PSYCHOLOGY or VIOLENCE IN THE SOUTH (1996) (reporting on the elevated level of interpersonal violence in the American South and its links to concern about one's reputation).

(63) Adam L. Alter et al., Transgression Wrongfulness Outweighs its Harmfulness as a Determinant of Sentence Severity, 31 LAW & HUM. BEHAV. 319, 334 (2007); John M. Darley et al., Incapacitation and Just Deserts as Motives for Punishment, 24 LAW & HUM. BEHAV. 659, 671 (2000); Dale T. Miller, Disrespect and the Experience of Injustice, 52 ANN. REV. PSYCHOL. 527, 535-36 (2001); Cass R. Sunstein et al., Do People Want Optimal Deterrence?, 29 J. LEGAL STUD. 237, 250 (2000) (showing that deterrence does not explain opinions).

(64) Philip E. Tetlock et al., People as Intuitive Prosecutors: The Impact of Social-Control Goals on Attributions of Responsibility, 43 J. EXPERIMENTAL SOC. PSYCHOL. 195, 202-04 (2007); Tyler & Boeckmann, supra note 40, at 252.

(65) See Helmut Kury et al., Dimensions of Punitiveness in Germany, 15 EUR. J. ON CRIM. POL'Y & RES. 63, 69 (2009).

(66) E.g., VAN DIJK ET AL., supra note 35, at 148 fig. 31; Mayhew & van Kesteren, supra note 34, at 66-76.

(67) See generally THORSTEN SELLIN & MARVIN E. WOLFGANG, THE MEASUREMENT OF DELINQUENCY (1964) (examining lay views of the seriousness of various criminal offenses).

(68) E.g., Evans & Scott, supra note 40, at 53; Marlene Hsu, Cultural and Sexual Differences on the Judgment of Criminal Offenses." A Replication Study of the Measurement of Delinquency, 64 J. CRIM. L. & CRIMINOLOGY 348, 348 (1973); Joseph E. Scott & Fahad Al-Yhakeb, The Public's Perceptions of Crime: A Comparative Analysis of Scandinavia, Western Europe, the Middle East, and the United States, in CONTEMPORARY CORRECTIONS: SOCIAL CONTROL AND CONFLICT 78, 84 (C. Ronald Huffed., 1977).

(69) See Hsu, supra note 68, at 351; Peter H. Rossi et al., The Seriousness of Crimes: Normative Structure and Individual Differences, 39 AM. Soc. REV. 224, 230-31 (1974).

(70) Evans & Scott, supra note 40, at 47-51; Scott & A1-Thakeb, supra note 68, at 85 (discussing drug use, prostitution, homosexual acts, and abortion).

(71) Tetlock et al., supra note 64, at 198-201.

(72) Lynne D. Roberts & David Indermaur, Predicting Punitive Attitudes in Australia, 14 PSYCHIATRY PSYCHOL. & L. 56, 61-62 (2007).

(73) Cullen et al., supra note 40, at 310-12; Francis T. Cullen et al., Public Support for Correctional Rehabilitation in America: Change or Consistency?, in CHANGING ATTITUDES TO PUNISHMENT: PUBLIC OPINION, CRIME, AND JUSTICE, supra note 34, at 140-44.

(74) See Robert B. Cialdini et al., Managing Social Norms for Persuasive Impact, 1 Soc. INFLUENCE 3, 3-6 (2006); Deborah A. Prentice & Dale T. Miller, Pluralistic Ignorance and Alcohol Use on Campus." Some Consequences of Misperceiving the Social Norm, 64 J. PERSONALITY & SOC. PSYCHOL. 243, 247-50 (1993); Christine M. Schroeder & Deborah A. Prentice, Exposing Pluralistic Ignorance to Reduce Alcohol Use Among College Students, 28 J. APPLIED SOC. PSYCHOL. 2150, 2168-69 (1998).

(75) See Prentice & Miller, supra note 74, at 247-50.

(76) See generally ROBINSON & DARLEY, supra note 60; Nicole E. Haas et al., Public Support for Vigilantism: An Experimental Study, 8 J. EXPERIMENTAL CRIMINOLOGY 387, 394 (2012).

(77) See infra Appendix A for the full text of the scenarios.

(78) The retribution scale used two items from Cullen et al., supra note 40, at 307, and two items from Tom R. Tyler & Renee Weber, Support for the Death Penalty; Instrumental Response to Crime, or Symbolic Attitude?, 17 LAW & SOC'Y REV. 21, 31 (1982). The rehabilitation, deterrence, and incapacitation scales respectively used four, three, and three items from Cullen et al., supra note 40, at 307-08.

(79) This scale used two items adapted from Randy L. LaGrange & Kenneth F. Ferraro, Assessing Age and Gender Differences in Perceived Risk and Fear of Crime, 27 CRIMINOLOGY 697, 701--03 (1989).

(80) This scale used three items inspired by Tyler & Boeckmann, supra note 40, at 242 (threat to values).

(81) This scale used four items from Tyler & Weber, supra note 78, at 28-29.

(82) A variety of other constructs were also measured, but they are not relevant to the current hypotheses and will be presented in future work.

(83) Due to the different interpretation of the terms "liberal" and "conservative" in the German political system, German participants were also asked to indicate their political orientation ranging from left (1) to right (7), M = 3.47, SD = 1.19.

(84) This last attention check was intended to catch participants who gave near straight-line responses on scales that had several reverse coded items, suggesting that they were not attending to the items. These checks excluded 18% of American respondents, 11% of German respondents, and 10% of Canadian respondents. Spot checks confirmed that including these participants did not affect the direction or significance of most study findings (e.g., re-analysis of sentence and punitiveness ratings for Armed Robbery, Rape, Petty Theft, Fraud, Shooting Fleeing Rapist, and Defense of Property revealed that out of 24 pairwise comparisons, only one changed with inclusion of all participants: the difference between American and Canadian respondents on the Rape-punitiveness measure became significant, direction unchanged). Participants excluded by these checks were not drastically different from the remainder of the sample. They were somewhat more likely to be male than female (58% male in the American sample, 53% in the Canadian, 53% in the German) and were likely to be younger than the remaining participants (9 years in the American sample, 3 years in the Canadian, and 5 years in the German).

(85) E.g., Alan Bainbridge, Survey Data Cleansing: Five Steps for Cleaning Up Your Data, ALLEGIANCE BLOG (Jan. 6, 2009), data/74; see also "Bad" Respondents and the Panel Quality Index, MKTG., INC., http://www.mktginc.corrdpdf/Panel%20Quality%20Assessment%20and%20Quality%20Segments.pdf (last visited Sept. 16, 2013); Nico Peruzzi, Online Survey Sample is Not Clean Enough--Clean It Yourself, OUTSOURCE RES. CONSULTING MARKETING RES. BLOG (July 14, 2010), http://orconsulting.eom/blog/?p=58.

(86) Because the German education system has several distinct tracks, it is not comparable.

(87) Participants were allowed to check multiple categories. French-Canadians were somewhat underrepresented in our sample (10% vs. 22%), but their attitudes did not differ from those of the other Canadians in our sample overall.

(88) We recognize that the populations of these countries are not homogeneous and that there may be differences based on region, urban vs. rural living, gender, etc. Our goal in this analysis is to show that there were no substantial sampling biases, not to test exhaustively for sub-population differences.

(89) On both crimes, Americans and Canadians criminalized significantly more often than Germans, but in turn, did not significantly differ. Drug use: American to German [chi square] (1, N = 799) = 31.68, p < .001; Canadian to German [chi square] (1, N = 811) = 18.68, p < .001. Prostitution: American to German [chi square] (1, N = 799) = 232.41, p < .001; Canadian to German [chi square] (1, N = 811) = 202.79, p < .001.

(90) For all three cultures, the morality of drug use rating is above the midpoint: American t(349) = 8.15, p < .001; Canadian t(361) = 5.09, p < .001; German t(448) = 3.69, p < .001.

(91) The morality ratings for soliciting a prostitute are significantly above the midpoint for both Americans, t(349) = 12.85, p < .001, and Canadians, t(361) = 7.91, p < .001. Germans, however, are significantly below the midpoint, t(448) = -15.79, p < .001.

(92) Matthew B. Kugler & John M. Darley, Punitiveness Towards Users of Illicit Drugs: A Disparity between Actual and Perceived Attitudes, 24 FED. SENT'G REP. 217 (2012).

(93) We also included a further self-defense scenario in which a man is cornered by a knife-wielding attacker who is seeking to kill him. Killing this attacker would likely be excused under the legal rules of all three countries. Participants' scores on the qualitative punitiveness, F(2, 1041) = .03 (M = 1.34, SD = .84), and morality, F(2, 1041) = 1.38 (M= 2.93, SD = 1.35), questions were very low and did not differ significantly across cultures. Sentences were very low across all three cultures with 93% of Americans and Canadians assigning no punishment, along with 79% of Germans. A further 12% of Germans assigned the lowest possible punishment.

(94) Note, however, that the internal consistency for that scale was extremely low for the German sample (a = .40, see Table 6). This result should be interpreted cautiously.

(95) E.g., BECKETT, supra note 3, at 3-13.

(96) See generally TURNER, supra note 20 (emphasizing the role of a lightly settled frontier region in shaping culture, a feature that is common in the United States and Canada).

(97) Cullen et al., supra note 40, at 314; Cullen et al., supra note 73, at 140-44.

(98) See generally Roberts & Sprott, supra note 50 (describing the differences between the American and Canadian justice systems); Tonry, supra note 46 (describing the differences between the German and American systems).

(99) See generally TURNER, supra note 20 (emphasizing the role of expansion into a lightly settled frontier region in shaping culture).

(100) See generally ESPING-ANDERSEN, supra note 52; VARIETIES OF CAPITALISM: THE INSTITUTIONAL FOUNDATIONS OF COMPARATIVE ADVANTAGE (Peter A. Hall & David Soskice eds., 2001) (contrasting the Anglo-American model with other approaches).

(101) Mayhew & van Kesteren, supra note 34, at 66-76; see also VAN DIJK ET AL., supra note 35, at 148 fig.31, 152.

(102) MARY VIPOND, THE MASS MEDIA IN CANADA 63 (4th ed. 2011).

(103) Id. at 65.

(104) Sharon Casey & Philip Mohr, Law-and-Order Politics, Public-Opinion Polls and the Media, 12 PSYCHIATRY PSYCHOL. & L. 141, 145-48 (2005) (showing that media usage was correlated with punitiveness).

(105) Caroline A. Spiranovic et al., What Predicts Punitiveness?: An Examination of Predictors of Punitive Attitudes Towards Offenders in Australia, 19 PSYCHIATRY PSYCHOL. & L. 249, 254-56 (2012).

(106) Mayhew & van Kesteren, supra note 34, at 66-76; see also VAN DISK ET AL., supra note 35, at 149 tbl.32.

(107) E.g., Alter et al., supra note 63, at 334; Darley et al., supra note 63, at 671.

(108) See Prentice & Miller, supra note 74, at 247-50.

(109) See generally Scott Eidelman et al., The Existence Bias, 97 J. PERSONALITY & SOC. PSYCHOL. 765 (2009) (showing people prefer that which has been around longer); John T. Jost et al., A Decade of System Justification Theory: Accumulated Evidence of Conscious and Unconscious Bolstering of the Status Quo, 25 POL. PSYCHOL. 881 (2004) (reviewing the first decade of system-justification theory); John T. Jost & Mahzarin R. Banaji, The Role of Stereotyping in System-Justification and the Production of False Consciousness, 33 BRITISH J. SOC. PSYCHOL. 1 (1994) (introducing system-justfication theory).

(110) Aaron C. Kay et al., Inequality, Discrimination, and the Power of the Status Quo: Direct Evidence for a Motivation to See the Way Things Are as the Way They Should Be, 97 J. PERSONALITY & SOC. PSYCHOL. 421,423-28 (2009).

(111) E.g., Richard S. Frase, Comparative Perspectives on Sentencing Policy and Research, in SENTENCING AND SANCTIONS, supra note 41, at 259, 276-77; Roberts & Sprott, supra note 50, at 71-72; Tonry, supra note 46, at 1198-99.

(112) See generally Carlos Berdejo & Noam Yuchtman, Crime, Punishment, and Politics. An Analysis of Political Cycles in Criminal Sentencing, 95 REV. ECON. & STAT. 741 (2013).

(113) See generally Sanford C. Gordon & Gregory A. Huber, The Effect of Electoral Competitiveness on Incumbent Behavior, 2 Q.J. POE. Sct. 107 (2007) (reporting a study comparing partisan and non-partisan judicial elections in Kansas).

(114) Lappi-Sepalla, supra note 24, at 318.

(115) Torn'y, supra note 46, at 1198-1201.

(116) GARLAND, supra note 13, at 152 61.

(117) Roberts & Sprott, supra note 50, at 70-71; Tonry, supra note 46, at 1199, 1205.

(118) Roberts & Sprott, supra note 50, at 70-71.

(119) David F. Greenberg & Valerie West, Siting the Death Penalty Internationally, 33 LAW & SOC. INQUIRY 295, 319, 321 (2008).

(120) Roberts & Sprott, supra note 50, at 66-72; Tonry, supra note 8, at 425-31; Tonry supra note 46, at 1198-1201.

(121) See Prentice & Miller, supra note 74, at 247-50.

(122) E.g., BECKETT, supra note 3, at 3-13; Langan, supra note 7, at 126-28; Roberts & Sprott, supra note 50, at 71-72; Tonry, supra note 46, at 1198-1201; Richard S. Frase, Comparative Perspectives on Sentencing Policy and Research, in SENTENCING AND SANCTIONS, supra note 41,259, 272-77.

MATTHEW B. KUGLER, J.D. Candidate, The University of Chicago Law School, 2015; Ph.D. in Psychology and Social Policy from Princeton University, 2010.

FRIEDERIKE FUNK, Ph.D. Candidate in Social Psychology at Princeton University; M.A. in Psychology from Princeton University, 2012.

JUDITH BRAUN, Ph.D. Candidate in Psychology at Philipps-Universitat Marburg.

MARIO GOLLWITZER, Professor of Psychology at Philipps-Universitat Marburg.

AARON C. KAY, Associate Professor of Management and Associate Professor of Psychology and Neuroscience at Duke University.

JOHN M. DARLEY, Warren Professor of Psychology at Princeton University.
Table 1
Evaluations of Major Crimes


 Murder 6.87 (0.42)
Abstract Rape 6.40 ab (0.81)
Punitiveness Drug Dealing 5.74 a (1.27)
 Armed Robbery 5.26 a (1.16)

 Murder 10.15 a (0.96)
Concrete Rape 8.30 a (1.37)
Sentence Drug Dealing 7.18 a (2.04)
 Armed Robbery 6.45 a (1.69)

 Murder 8.85 (0.67)
Morality Rape 8.69 b (0.78)
Rating Drug Dealing 7.81 (1.30)
 Armed Robbery 7.69 (1.17)


 Murder 6.87 (0.41)
Abstract Rape 6.46 a (0.67)
Punitiveness Drug Dealing 5.89 a (1.11)
 Armed Robbery 5.17 a (1.15)

 Murder 9.93 b (1.03)
Concrete Rape 8.00 b (1.33)
Sentence Drug Dealing 7.30 a (1.94)
 Armed Robbery 6.32 a (1.46)

 Murder 8.88 (0.44)
Morality Rape 8.77 ab (0.67)
Rating Drug Dealing 8.02 (1.11)
 Armed Robbery 7.77 (1.08)


 Murder 6.84 (0.42)
Abstract Rape 6.27 b (0.83)
Punitiveness Drug Dealing 5.24 b (1.24)
 Armed Robbery 4.25 b (1.12)

 Murder 9.28 (1.10)
Concrete Rape 7.58 C (1.39)
Sentence Drug Dealing 6.32 b (1.61)
 Armed Robbery 5.33 b (1.40)

 Murder 8.88 (0.41)
Morality Rape 8.82 a (0.47)
Rating Drug Dealing 7.91 a (1.09)
 Armed Robbery 7.61 (1.03)

 F .sup.2]

 Murder 0.57
Abstract Rape 5.30 ** .01
Punitiveness Drug Dealing 26.91 *** .05
 Armed Robbery 80.96 *** .13

 Murder 65.48 *** .11
Concrete Rape 23.58 *** .04
Sentence Drug Dealing 27.39 *** .05
 Armed Robbery 54.14 *** .09

 Murder 0.45
Morality Rape 3.32 * .01
Rating Drug Dealing 2.88
 Armed Robbery 1.64

* <.05, ** <.01, *** <.001

Means within a row that do not share subscripts are
significantly different from each other.

Degrees of freedom are 2, 1038 for all analyses.

Table 2
Average Actual Sentence Contrasted with Perceived Average Sentence

 Actual Projection for
 Average Average Person
 Sentence from Country

 American 10.15(0.96) 9.88(1.28)
Murder Canadian 9.93(1.03) 8.61(1.64)
 German 9.28(1.10) 8.98(1.55)

Petty American 3.47(1.96) 3.44(2.02)
Theft Canadian 3.33(1.93) 2.52(1.86)
 German 1.90(1.48) 2.83(1.81)

Shooting American 2.80(3.14) 3.63(3.25)
Rapist Canadian 3.30(3.09) 4.04(2.95)
 German 4.13(2.95) 5.06(3.01)

 American 1.54(1.89) 2.29(2.13)
Drug Use Canadian 1.24(1.69) 1.48(1.55)
 German 0.75(1.36) 2.74(2.12)

 Projection for
 Average Person
 from Country d

 American t(349) = 4.18 *** 0.22
Murder Canadian t(361) = 15.06 *** 0.79
 German t(328) = 3.32 *** 0.18

Petty American t(349) = 0.38 0.02
Theft Canadian t(361) = 8.68 *** 0.46
 German t(328) = -10.15 *** -0.56

Shooting American t(349) = -5.34 *** -0.29
Rapist Canadian t(361) = -4.68 *** -0.25
 German t(328) = -6.41 *** -0.35

 American t(349) = -6.12 *** -0.33
Drug Use Canadian t(361) = -2.47 * -0.13
 German t(328) = -17.84 *** -0.98

* <.05, ** <.01, *** <.001

Means within a row that do not share subscripts are
significantly different from each other.

Degrees of freedom are 2, 1038 for all analyses.

Table 3
Evaluations of Minor Crimes


 Tax Fraud 4.10 a (1.36)
Abstract Petty Fraud 3.94 a (1.42)
Punitiveness Fraud 5.17 a (1.34)
 Vandalism 2.68 a (1.38)

 Tax Fraud 3.90 a (2.66)
Concrete Petty Fraud 3.47 a (1.96)
Sentence Fraud 6.37 a (1.99)
 Vandalism 1.46 a (1.36)

 Tax Fraud 6.81 a (1.31)
Morality Petty Theft 7.07 a (1.07)
Rating Fraud 7.72 (1.34)
 Vandalism 6.34 (1.38)


 Tax Fraud 3.92 a (1.44)
Abstract Petty Fraud 3.83 a (1.37)
Punitiveness Fraud 5.17 a (1.18)
 Vandalism 2.53 a (1.29)

 Tax Fraud 3.15 b (2.49)
Concrete Petty Fraud 3.33 a (1.93)
Sentence Fraud 6.33 a (1.68)
 Vandalism 1.23 b (1.02)

 Tax Fraud 6.74 a (1.25)
Morality Petty Theft 7.01 ab (1.03)
Rating Fraud 7.87 (1.08)
 Vandalism 6.19 (1.32)


 Tax Fraud 3.29 b (1.28)
Abstract Petty Fraud 2.77 b (1.23)
Punitiveness Fraud 4.16 b (1.27)
 Vandalism 2.12 b (1.27)

 Tax Fraud 2.52 (2.15)
Concrete Petty Fraud 1.90 b (1.48)
Sentence Fraud 5.04 b (1.76)
 Vandalism 1.30 ab (1.18)

 Tax Fraud 6.35 b (1.52)
Morality Petty Theft 6.87 b (1.10)
Rating Fraud 7.82 (1.15)
 Vandalism 6.21 (1.36)

 F .sup.2]

 Tax Fraud 32.74 *** .06
Abstract Petty Fraud 78.25 *** .13
Punitiveness Fraud 72.76 *** .12
 Vandalism 16.63 *** .03

 Tax Fraud 26.93 *** .05
Concrete Petty Fraud 77.42 *** .13
Sentence Fraud 58.93 *** .10
 Vandalism 3.24 * .01

 Tax Fraud 11.30 *** .02
Morality Petty Theft 3.20 * .01
Rating Fraud 1.35
 Vandalism 1.17

* < .05, ** < .01, *** < .001

Means within a row that do not share subscripts are
significantly different from each other.

Degrees of freedom are 2, 1038 for all analyses.

Table 4
Evaluations of Victimless Crimes


Abstract Drug Use 2.30 a (1.40)
Punitiveness Prostitution 2.45 a (1.59)

Concrete Drug Use 1.54 a (1.89)
Sentence Prostitution 1.25 a (1.88)

Morality Drug Use 5.77 a (1.78)
Rating Prostitution 6.20 a (1.75)


Abstract Drug Use 2.15 a (1.31)
Punitiveness Prostitution 2.43 a (1.68)

Concrete Drug Use 1.24 a (1.69)
Sentence Prostitution 1.08 a (1.75)

Morality Drug Use 5.44 b (1.65)
Rating Prostitution b (1.85)


Abstract Drug Use 1.49 b (0.90)
Punitiveness Prostitution 1.13 b (0.60)

Concrete Drug Use 0.75 b (1.36)
Sentence Prostitution 0.16 b (0.91)

Morality Drug Use 5.29 b (1.76)
Rating Prostitution 3.76 C (1.54)

 F .sup.2]

Abstract Drug Use 41.95 *** .07
Punitiveness Prostitution 99.52 *** .16

Concrete Drug Use 19.13 *** .04
Sentence Prostitution 45.61 *** .08

Morality Drug Use 7.05 *** .01
Rating Prostitution 193.12 *** .27

* <.05, ** <.O1, *** <.001

Means within a row that do not share subscripts are
significantly different from each other.

Degrees of freedom are 2, 1038 for all analyses.

Table 5
Evaluations of Excessive Self-Defense


 Shooting Rapist 2.85 b (1.70)
Abstract Revenge for Rape 2.07 c (1.52)
Punitiveness Defense of Property 3.38 b (1.65)
 Retreat Possible 3.11 b (1.79)

 Shooting Rapist 2.80 b (3.14)
Concrete Revenge for Rape 1.39 c (2.47)
Sentence Defense of Property 3.79 b (3.03)
 Retreat Possible 3.14 b (3.23)

 Shooting Rapist 4.28 b (2.14)
Morality Revenge for Rape 3.90 c (2.15)
Rating Defense of Property 5.57 b (2.00)
 Retreat Possible 5.25 b (2.16)


 Shooting Rapist 3.20 a (1.68)
Abstract Revenge for Rape 2.45 b (1.52)
Punitiveness Defense of Property 3.97 a (1.55)
 Retreat Possible 3.85 a (1.77)

 Shooting Rapist 3.3 b (3.09)
Concrete Revenge for Rape 2.11 b (2.67)
Sentence Defense of Property 4.73 a (2.80)
 Retreat Possible 4.62 a (3.08)

 Shooting Rapist 4.96 a (2.09)
Morality Revenge for Rape 4.6 b (2.04)
Rating Defense of Property 6.16 (1.74)
 Retreat Possible 6.21 a (1.83)


 Shooting Rapist 3.46 a (1.69)
Abstract Revenge for Rape 3.22 a (1.79)
Punitiveness Defense of Property 3.80 a (1.64)
 Retreat Possible 3.55 a (1.59)

 Shooting Rapist 4.13 a (2.95)
Concrete Revenge for Rape 3.95 a (2.86)
Sentence Defense of Property 4.90 a (2.48)
 Retreat Possible 4.46 a (2.64)

 Shooting Rapist 5.14 a (2.10)
Morality Revenge for Rape 5.31 a (2.16)
Rating Defense of Property 6.12 a (1.89)
 Retreat Possible 5.98 a (1.86)

 F .sup.2]

 Shooting Rapist 11.45 *** .02
Abstract Revenge for Rape 44.81 *** .08
Punitiveness Defense of Property 12.33 *** .02
 Retreat Possible 16.29 *** .03

 Shooting Rapist 16.33 *** .03
Concrete Revenge for Rape 82.49 *** .14
Sentence Defense of Property 16.03 *** ,03
 Retreat Possible 22.79 *** .05

 Shooting Rapist 15.64 *** .03
Morality Revenge for Rape 37.87 *** .07
Rating Defense of Property 10.93 *** .02
 Retreat Possible 23.42 *** .04

* < .05, ** < .01, *** <.001

Means within a row that do not share subscripts are
significantly different from each other.

Degrees of freedom are 2, 1038 for all analyses.

Table 6
Evaluations of Excessive Self-Defense

 Number of Items Americans

 Liberalism 1 4.15 a (1.63)
Demographics Conservation
 Religiosity 1 3.89 a (1.95)

 Retribution 4 5.28 a (1.10) .75
Punishment Rehabilitation 4 4.51 (1.29) .80
Motives Deterrence 3 4.96 (1.60) .87
 Incapacitation 3 5.24 (1.33) .82

Reactions Fear of Crime 4 4.53 a (1.76) .92
to Crime Crime is 2 5.35 a (1.30) .72
 on Rise
 Crime as 3 0.59 a (1.21) .93
 to Values

 Number of Items Canadians

 Liberalism 1 3.86 b (1.74)
Demographics Conservation
 Religiosity 1 3.23 b (1.82)

 Retribution 4 5.27 a (1.19) .75
Punishment Rehabilitation 4 4.47 (1.30) .80
Motives Deterrence 3 5.05 (1.54) .87
 Incapacitation 3 5.21 (1.36) .82

Reactions Fear of Crime 4 4.37 a (1.73) .92
to Crime Crime is 2 5.3 a (1.42) .72
 on Rise
 Crime as 3 5.45 a (1.30) .94
 to Values

 Number of Items Germans

 Liberalism 1 3.28 c (1.32)
Demographics Conservation
 Religiosity 1 2.91 b (1.80)

 Retribution 4 4.98 b (0.92) .40
Punishment Rehabilitation 4 4.49 (1.33) .80
Motives Deterrence 3 5.07 (1'43) .82
 Incapacitation 3 5.03 (1.44) .82

Reactions Fear of Crime 4 3.72 b (1.74) .92
to Crime Crime is 2 4.5 b (1.42) .62
 on Rise
 Crime as 3 5.01 b (1.35) .90
 to Values

 Number of Items F .sup.2]

 Liberalism 1 26.41 *** .05
Demographics Conservation
 Religiosity 1 24.8 *** .05

 Retribution 4 8.42 *** .02
Punishment Rehabilitation 4 0.08 .00
Motives Deterrence 3 0.55 .00
 Incapacitation 3 2.41 .00

Reactions Fear of Crime 4 20.52 *** .04
to Crime Crime is 2 39.71 *** .07
 on Rise
 Crime as 3 18.72 *** .03
 to Values

* <.05, ** <.01, *** <.001

Means within a row that do not share subscripts are
significantly different from each other.

Degrees of freedom are 2, 1038 for all analyses.
COPYRIGHT 2013 Northwestern University, School of Law
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2013 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Kugler, Matthew B.; Funk, Friederike; Braun, Judith; Gollwitzer, Mario; Kay, Aaron C.; Darley, John
Publication:Journal of Criminal Law and Criminology
Date:Sep 22, 2013
Previous Article:Rethinking the use of community supervision.
Next Article:U.S. Supreme Court decisions and sex offender legislation: evidence of evidence-based policy?

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters