Punishment and Deterrence

Suggested Readings


In early American history there was a "debate" about prisons between the so-called "Pennsylvania" and "Auburn" plans:

  • Philadelphia or "separation" system : contemplation, work, reform – "a setting in which man’s natural grace could emerge"
  • Auburn or New York system: discipline, quiet, correction – "a setting in which his natural wickedness could at least be curbed and bent to the needs of society."

The former is associated with the Eastern State Penitentiary, built in 1829 as bold experiment in prison reform based in the controversial idea of changing behavior through "confinement in solitude with labor."1

The latter approach, developed in the early 1800s, involved work in groups during the day, solitary confinement at night. Prisoners wore striped uniforms, maintained silence, and were subject to military-like discipline (e.g., marching in step).


Both approaches had the goal of changing the future behavior of the criminal, but the two systems represented distinct theories about deviance and social control. Both involved solitary confinement (which was an innovation compared to mass imprisonment)2

Ideologically, the Philadelphia approach resonates with Quaker sensibilities and the belief that rehabilitation lies in the re-emergence of "inner goodness." The name of the institution — penitentiary — conveys this quasi -religious approach: in a monastery-like atmosphere one can repent and renew. The Auburn system, by contrast is rooted in the idea that rehabilitation is about re-imposing social discipline on the criminal by teaching them personal discipline and respect for work, property, and other people.[2]

At issue here is the question of which approach works better. In other words, we are looking at social control here as an independent variable and some general notion of rehabilitation as the dependent variable.

Although this formulation makes changing the person the primary goal — and hence we'd put it in the category of "therapeutic" social control — it is fair to say that this is an intermediate goal. The bottom line for society here is to use social control to reduce the likelihood of subsequent offending.

Topic Lends Itself to Easy Confusion

There are so many motivations for responding to rule violations (we are avoiding the word "punishment" for now since we've said it is but one of many types of response) that confusion reigns when we try to answer the question "why and how to commit social control?".

The offender must "pay the price." The victim should be made whole. The offender must be changed so s/he will not do it again. Others must be reminded not to do this. The conditions that give rise to such behavior must be dealt with.

But we also have meta-norms: the "punishment must fit the crime"; no cruel and unusual punishment; and finally, the social control should be effective. But what do we mean by effective?

We have spoken at length of the functions for groups of identifying and responding to breaches. It reminds the group of where the boundaries are. It unites the group against rule-breakers. It makes conformers feel more connected to the idea of the group. It exercises the collective consciousness. It generates solidarity by giving people an opportunity to interact with a focus on group rules and common beliefs.

It is easy to glide right over these as by now obvious to our sociological sensibilities, but let's look a little more closely at question of being reminded where the boundaries are. What we mean here is a collective sense of what is OK and what is not OK in our group. We like to quote Durkheim saying "its does not offend the CC because it is a crime, it is a crime because it offends the CC." Another way to hear this is that the deviant act causes damage to the social fabric, and that social control is, in part, about repairing that damage.

What does it take?

But we have also talked about "deviants" as "patrolling the boundary" and that their detection, apprehension, and punishment serves as a warning to the rest of us.

And we have talked about the collective consciousness and perhaps you've found yourself wondering where it comes from and answering that when we see people punished for particular behaviors it leaves a trace, so to speak, in everyone's mind, reminding us of what we believe.

And remember that in Garfinkel's degradation ceremony the denouncer speaks on our behalf. "In the name of all the non-adulterers here I condemn this adulterer!"

And we have the history of punishment which includes a trend toward increasing scope of individualized punishment. Specialized treatment for juveniles, women, mentally ill, sex offenders, vagrants, alcoholics, drug addicts, etc. : "Punishing the criminal, not the crime."

There seem to be a number of different things going on here. Can we sort them out?

What we'll use: already we have "spread out" social control along a dimension of styles and amounts.
To which we'll add: a taxonomy of offenses (we've already begun this in the distinction we drew between mala prohibita and mala per se)

Basic Definitions

  • The specific deterrent effect of punishment is the reduction in likelihood that a punished person will re-offend. This amounts to "learning a lesson."
  • The general deterrent effect of punishment is the reduction in likelihood that people other than the punished person will not offend. This amounts to "sending a message."

Research Question

Do specific and general deterrent effects exist? In other words, do we get a payoff for what we invest in punishment? Or should we simply be satisfied with the "making them pay" aspect of punishment?

The more specific research question : under what conditions (what kind of offenses, what kind of offenders) does what kind of punishment (severity, certainty, celerity (swiftness)) have what kind of deterrent effect?

Why Do We Care?

In general, societies do have an interest in maximizing conformity to their rules.

Public policies are invented, designed, advocated, and implemented on the basis of explicit or implicit theories about human behavior, about what works to channel human behavior. If we are interested in particular outcomes (reducing domestic violence, say, or traffic fatalities) then we want to know what policies are likely to be most efficacious.

If the main justification for doing something is the benefit it provides to society, conccience demands that we verify that those benefits are real.
What is the debate about general prevention? Why would the quoted doctor say "I shudder when I think what this essentially fictitious concept has cost us, in terms of thousands upon thousands of wasted, bitter man-years of imprisonment…."?

In other words, when we "throw the book at" someone, it might be the case that the only good that comes out of it is some retributive satisfaction, the feeling that we've taken revenge, or at least "done something." Given the high costs (to society and to the punished persons) we might want, as a society, to know the facts so we can assess whether it is what we want to do.

Why do we care about the theory/reality of prevention? It affects how we make laws, how we apply laws, how we think about crime and criminals.

Note belief in specificity: "…a whole system of institutions…combat crime at its source … in a manner adapted to each age group and category of crime." (from 1902, p. 5)

The Conventional Theory of Deterrence

At the core of conventional thinking about specific deterrence is the utilitarian idea of making it "not worth it" to violate a rule. This assumes that there is some benefit to breaking the rule and if one could simply make it more expensive then people would not do it.

In thought deriving from Adam Smith and Jeremy Bentham we have the idea of "matching the punishment to the crime" — taken to its extreme we try to really fine tune the system so that we only "spend" as much punishment as will actually pay off to reduce the incidence (the cost) of a given offence.

As an example, let's imagine some behavior X of which a society or group disapproves. Suppose there is no organized effort to sanction the behavior when it occurs but it is widely known in the culure that folks do not approve of it. What rates of incidence can we expect?

Now let's think about designing a policy for intervening.



Words and Phrases

  • jurist – here refers to legal scholars as well as judges.
  • general prevention/deterrence – effect that punishment has on reducing likelihood of other people violating norms in the future
  • specific prevention/deterrence – effect that punishment has on reducing the likelihood that the punished person will violate norms in the future

General Prevention

Andeneus. 1974. Punishment and Deterrence[1]

Recall back in Erikson[2], we talked about society needing to draw the lines and let people know where the lines are drawn. We also need to think about how it is that we get people to pay attention to the lines.

Setup for later reading of Foucault and soon reading of Stone

General prevention = ability of law and enforcement to make citizens law abiding

How to accomplish this? Public awareness, morality tales, threats, punishment, etc.

Variations: certainty of detection. Certainty of punishments. Severity of punishments.

Goal is to produce conscious and unconscious inhibitions

  • Fear of punishment
  • Habits
  • Respect for authority
  • Moral linking
  1. 1) Functions of punishment
    1. a) Specific prevention
      1. i) Deterrence
      2. ii) Reformation
      3. iii) Incapacitation
    2. b) General prevention (in other words, what's in "sending a message"?)
      1. i) Deterrence
      2. ii) Increase in moral inhibitions
      3. iii) Increase in habitual law abiding conduct
      4. Latter two most "effective" because they don't ultimately depend on surveillance, detection and punishment, thus "socio-pedagogical influence" of punishment.
      5. i) Societal reaction as propaganda in favor of respect for norms. A means of expressing social disapproval. A means of combatting demoralizing consequences of witnessing norm violations.
      6. ii) This can be conscious or unconscious.
    3. (1) Feuerbach — effectiveness of punishment depends on risk of punishment x punishment > expected benefit
    4. (2) Conscious: affect the deliberation of the potential norm violator
    5. (3) Unconscious: move violations away from consideration


Purely deterrent function of punishment — it is clear that it depends not simply on the risk of being punished, but also on the nature and magnitude of the punishment. How important this factor is depends on the characteristic motivation for the crime, and on many other circumstances. (24)


  1. Magnitude/certainty of punishment has more deterrence effect the more the norm violation involves careful weighing of pros and cons, less when it involves sudden impulses or overpowering emotions.
  2. Higher the moral condemnation attached to deed, the less influence the magnitude of the punishment on deterrence effect.
  3. Increases in severity of punishment affect deterrence more quickly than moral inhibition.
  4. Question: can increasing punishment increase level of moral inhibition? Where would this be relevant? Date rape, for example?

Other variables : commitment, premeditation.

Definition and Logic

  1. Knowledge of rule.
  2. Knowledge of magnitude of punishment
  3. Knowledge of likelihood of punishment

Studying Deterrence

The logic of the preventative effect of punishment is, at first glance, simple: if you know you will be punished, you'll think twice about committing a crime (or violating a norm). By now, we hope, you'll object that it's a little more complex than that.

We know, for example, that there are multiple possible mechanisms by which behavior changes. The other day we spoke of several:

  1. Habit
  2. Respect for authority
  3. Moral repulsion
  4. Rational deterrence

We also noted that there is an "incarceration" or, more generally, an "incapacitation" effect (as when taking away your keys prevents you from driving while intoxicated or locking you prevents you from committing a crime).


"I think the reason to support the death penalty is because it saves other people’s lives," and further that "It's the only reason to be for it." By contrast, earlier that year, Attorney General Janet Reno stated, "I have inquired for most of my adult life about studies that might show that the death penalty is a deterrent, and I have not seen any research that would substantiate that point." [4]

The view that the death penalty deters is still the product of belief, not evidence. The reason for this is simple: over the past half century the U.S. has not experimented enough with capital punishment policy to permit strong conclusions. [4]

"1975, when University of Buffalo Professor Isaac Ehrlich published an influential article asserting that during the 1950s and '60s, each execution "…saved eight innocent lives" by deterring murder"[5]

Logic: influenced by Gary Becker theory of crime "would-be murderers would choose between illegal and legal behavior based on the threat of execution" [5].

Lots of attention, became a meme. But study itself announced limitations and criticism of methods and conclusions was fast and sustained. "In 1978, an expert panel appointed by the National Academy of Sciences strongly criticized Ehrlich's work and rejected its conclusions."[5] The argument went on for years without reaching a resolution.

Debate is back in the new century. Lots of research on both sides. Estimates of deterrent effects even bigger than Ehrlich's in the 70s.

What do we make of this recent work? A few bullet points from an essay by J. Fagan of Columbia Law School [5]:

  1. the new deterrence studies fall short of social science standards for causal findings
  2. studies fail to account for incarceration rates and life sentences (that may drive down crime rates via deterrence or incapacitation)
    1. studies that do so show prison effects that dwarf the deterrent effects of execution
  3. studes fail to account for social factors such as drug epidemics associated with fluctuations in the murder rate over time.
  4. studies lump all homicides together. don't look separately at those eligible for death penalty (should be most sensitive to deterrence effect)
    1. one that did showed no significant changes over time in homicides in the face of variation in the execution rate.
  5. most studies lump all forms of murder together, claiming that all are equally deterrable;
    1. one did not found domestic homicides more deterrable, a claim that contradicts decades work showing spontaneity and unpredictability in killings between intimate partners
  6. statistical models are often flawed.
    1. simple corrections for missing data produce estimates of deterrent effect no different from chance.
    2. alternate statistical models produces results that show that changes in homicide rates are statistically unrelated to any measure of capital punishment.
  7. causal model in the deterrence story unpersuasive. Little empirical evidence fits the complex causal story of awareness and calculation.

Capital Punishment and Politics

Seeking an End to an Execution Law They Once Championed
Published: April 6, 2012
Two men behind Proposition 7, a tough death penalty initiative passed in California in 1978, now see it as one of the biggest mistakes they ever made. [((bibcite ballotpedia_prop7))]

In 1978 California voters passed Proposition 7 re-instating and broadening the death penalty by a 70 to 30% margin. In 2012 there is a ballot measure to repeal. About half of the state's voters favor life in prison over death for someone convicted of first degree murder but the number supporting the death penalty is still in the 70% range. [9]

"Changes and expands categories of first degree murder for which penalties of death or confinement without possibility of parole may be imposed. Changes minimum sentence for first degree murder from life to 25 years to life. Increases penalty for second degree murder. Prohibits parole of convicted murderers before service of 25 or 15 year terms, subject to good-time credit. During punishment stage of cases in which death penalty is authorized: permits consideration of all felony convictions of defendant; requires court to impanel new jury if first jury is unable to reach a unanimous verdict on punishment. Financial impact: Indeterminable future increase in state costs." [((bibcite ballotpedia_prop7))]

What are some of the arguments?

  • Killing is wrong, no matter what.
  • The system is imperfect, allowing execution of people later found to be innocent.
  • The method used is cruel and either inhuman or unconstitutional.
  • Capital punishment does (or does not) deter crime.
  • Some crimes so horrific that there is a need for collective expression of condemnation.
  • System costs too much. Even if there is a deterrent effect, it is not efficient use of public resources.


Main Logical Approaches

  1. Trends over time
  2. Natural Experiments
  3. Spatial Comparisons
  1. Surveys and Interviews


Trends and Temporal Comparisons
Changes in social control over time vs. changes in behavior.
Challenges: causality (which comes first) and isolating effects of changes
Sometimes we get a natural experiment. Example: Great Britain Road Safety Act of 1967[11] which reduced allowable blood alcohol level for drivers to 0.08 per cent with stiff penalties for first offenses. For first few months and years the drop off in DUI casualties was significant but the effect diminished with time.[((bibcite lawlibdeterrence ))] Ross's interpretation was that initial publicity led people to overestimate the likelihood of detection but that over time people developed a more realistic estimate. By contrast, a Finnish reform a decade later involved a liberalizing of the penalties but a sharp increase in detection via random road testing. They saw a big drop in incidence of DUI.


When laws and social control practices vary across geographic jurisdictions, comparisons can be made between offense rates in the hope of zeroing in on social control practices as the explanatory variable for differences in behaviors. Perhaps the most common use of this technique is in research that attempts to ascertain the presence or absence of a deterrent effect in capital punishment by comparing laws and crime rates across jurisdictions.

Classic Studies

Drinking and Driving


Killoran, Canning, Doyle, & Sheppard 2010. "Review of effectiveness of laws limiting blood alcohol concentration levels to reduce alcohol-related road injuries and deaths"


Domestic Violence

"Early studies showed mandatory arrest to be the most effective policy in deterring batterers from future violence. Sherman and Berk (1984a, 1984b) were the first to study mandatory arrest, with numerous studies to follow."http://www.musc.edu/vawprevention/policy/mandarrest.shtml policy's success is tied to whether an offender is "good risk" or "bad risk." (Berk et al., 1992)

Sherman, Schmidt, and Rogan (1992) suggest jurisdictions replace mandatory arrest policies with mandatory action or police action chosen from a list of possibilities.

"Mandatory Prosecution"

"Few studies have examined mandatory prosecution policies. In fact, Ford and Regoli (1993) conducted the only randomized study of no-drop prosecution. They found that the type of prosecution strategy used (drop-permitted versus no-drop) has a significant effect on the future behavior of the batterer. Victims who chose to file charges against the perpetrator under a drop-permitted policy were less likely to experience future violence than were victims whose batterers were prosecuted without their input. However, the opposite was true for victims who chose to drop charges against their batterers; they were more likely to experience abuse again than those dealt with under mandatory prosecution."

"Ford and Regoli hypothesize that the preventative impact in drop-permitted cases comes from a victim's personal empowerment. They suggest this power derives from women using prosecution as a bargaining chip with their partners, allying with law enforcement, and being provided with a voice in determining sanctions."

"The Effect of Empowerment on Recidivism"

"Sherman and Berk (1984a) briefly addressed victim empowerment in their examination of mandatory arrest (although the replication studies failed to do so). They found a relationship between police concern and batterer recidivism. When batterers were arrested, victims experienced repeat abuse in 26 percent of the cases. When batterers were arrested and the victim perceived the police as concerned and willing to listen, the repeat abuse rate dropped to nine percent. Sherman and Berk hypothesized that the rate of recidivism dropped with police concern because victims felt empowered by the interaction."

An sidebar on doing reverse citation research.

Murder and Capital Punishment

Deterrence in the Context of International Relations

See http://en.wikipedia.org/wiki/Deterrence_theory "deterrence means using the threat of military action to compel an adversary to do something, or to prevent them from doing something, that another state desires."

Berejikian, Jeffery D. 2002. A Cognitive Theory of Deterrence," Journal of Peace Research, vol. 39, no. 2, 2002, pp. 165–183
Prospect theory is an empirical model of choice that stands as the leading alternative to rationality for explaining decisions under conditions of risk. While many still defend the assumption of rationality as an appropriate starting point for the construction of international relations theory – deterrence theory especially – there is growing support for models of international politics grounded in the actual capacities of real-world decisionmakers. This article accepts that standard depictions of deterrence incentives capture much of the essential character of deterrent relationships. However, it substitutes cognitive assumptions in place of traditional rational choice. Using prospect theory, the article reconsiders three typical deterrence games. The new model of military deterrence put forth unearths a set of conditions that are required for successful deterrence and uncovers a set of causes for deterrence failures that run counter to conventional understanding.

Deterring Plagiarism

In this University of Toronto document we read of these strategies:
Make Assignments an Integral Part of Learning in the Course
Demonstrate Your Expectations
Look at the Process as well as the Product

See problem 65.


  • Callahan, David. 2004. The Cheating Culture : Why More Americans Are Doing Wrong to Get Ahead. New York: Harcourt.
  • From review by John Leland in NYTBR February 8, 2004:
  • "From the adults around the Little League star Danny Altamonte who lied about the boy's age, to the looters at Enron, WorldCom, and Tyco, cheating scandals have filled the news. Though Callahan concedes in a postscript that no one really knows whether such cheating is on the rise — no small concession, given his project — he makes the case that high payoffs for winners and weak punishments for offenders have combined to encourage dishonesty among students, doctors, lawyers, taxpayers, auto mechanics, job applicants, stock analysts, file sharers, and corporate executives. In a society obsessessed with the bottom line, he writes, 'The message isn't just that the world is unfair and the rich can get away with murder; it's that people who cut corners get ahead.' Recent scandals involving mutual fund chiselers and steroid use among baseball players seem timed for the sequel.
  • "…Callahan's proposed remedies — an $8.50 minimum wage, affordable healthcare, increased financial aid for college, ethics curricula in professional schools, a tougher S.E.C. — are bigger than the problems they're meant to fix, however seriously you take Little League cheating."
  • Note focus on incentives and relative magnitudes of temptation, risk, and punishment. Only the "ethics curricula" lean in the direction of norms.


1. Andenaes, J. 1974. Punishment and Deterrence
: ballotpedia_prop7 : n.d. California Proposition 7, The Death Penalty Act of 1978" in Ballotpedia.
2. Erikson, K T. 1967. Wayward Puritans.
3. Donohue, J, Wolfers J, 2005. "Uses and Abuses of Statistical Evidence in the Death Penalty Debate," Stanford Law Review. 58, 787 (2005).
4. Donohue, John and Justin Wolfers. 2006. The Death Penalty: No Evidence for Deterrence," The Economists' Voice, April 2006
7. Foucault, M. 1977. Discipline and Punish.
8. Net Industries. 2012. "Deterrence - Methods Of Research," in Law Library - American Law and Legal Information
9. Nagourney, Adam. 2012. "Seeking an End to an Execution Law They Once Championed," New York Times April 6, 2012 (accessed on 9 April 2012).
10. Ross, H. Laurence Law, "Science, and Accidents: The British Road Safety Act of 1967." The Journal of Legal Studies Vol. 2, No. 1 (Jan., 1973), pp. 1-78 (JSTOR)
11. Ross, H L 1984. "Social Control Through Deterrence: Drinking-and-Driving Laws," Annual Review of Sociology, Vol. 10. (1984), pp. 21-35 (JSTOR)
12. Sherman, L. and R Berk. 1984 "The Specific Deterrent Effects of Arrest for Domestic Assault," American Sociological Review 49:261-271 (JSTOR)