The Deterrent Theory

The value of the punishment must not be less in any case than what is sufficient to outweigh that of the profit of the offence.

Read those together and the deterrent theory is not a theory of severity at all. Bentham's rule that punishment must not be needless is an argument for the lightest effective sentence, and his rule that it must not be unprofitable is an argument against long imprisonment for offences that do little harm. An answer that presents deterrence as a theory of harshness has misread its founder.

The conditions deterrence needs

Deterrence makes a factual claim, and the claim depends on four conditions being satisfied.

The person must know the penalty. Deterrence through a penalty nobody knows about is impossible.

The person must believe they will be caught. This is the condition that fails most often, and it is why certainty matters more than severity. See [Does Punishment Deter? The Evidence].

The person must be calculating. Offences committed in rage, in intoxication, under compulsion or by a person who does not expect to survive the act are not deterrable by any penalty.

The person must have an alternative. A person who steals because they have no other way to eat is not weighing a cost against a gain in the way the model assumes.

Each failed condition removes a class of offences from the theory's reach, and between them they remove a great many.

What Indian law does with it

The Sanhita's maxima are deterrent in design. Very high maxima for offences the legislature regards as serious, with the actual sentence left to the court.

Mandatory minimum sentences are pure general deterrence. They remove the court's discretion precisely so that the message is certain, and they are the sharpest form of the theory in Indian law.

Enhanced punishment for a repeat offender, section 13 of the Bharatiya Nyaya Sanhita 2023, is special deterrence in statutory form: the ordinary sentence did not deter this person, so the next one is heavier.

The 2013 and 2018 amendments to the law of sexual offences are the clearest recent example of a legislature acting on general deterrence, and they are dealt with in [Deterrence and Heinous Crimes Against Women].

A worked example

A city has a problem with drunken driving. The legislature raises the maximum sentence from two years to five.

What the deterrent theory predicts. Some potential offenders, learning of the change, will decide the risk is not worth it and the offence rate will fall.

What the four conditions ask. Do drivers know the maximum has changed? Do they believe they will be stopped and tested? Are they calculating at the time, given that the offence is by definition committed by an intoxicated person? Do they have an alternative way home?

What Bentham would say. If the same reduction could be achieved by more testing at a lower cost in human suffering, then the increased penalty is needless and therefore unjustified on his own principles.

What actually reduces the offence. The evidence in every jurisdiction points to enforcement, that is to the perceived probability of being stopped, rather than to the size of the penalty.

That example gives a complete answer to a question on the deterrent theory in about a hundred and fifty words, and it demonstrates the criticism without merely asserting it.

The criticisms

It uses the offender as a means. The offender is punished not for what they deserve but in order to influence other people. Kant's objection is that a person must never be treated merely as a means to the welfare of another, and this is the standing philosophical charge against deterrence.

It has no upper limit of its own. If a heavier sentence deters more, the theory alone gives no reason to stop. The limit has to come from somewhere else, which is why every workable system imports retributive proportion as a constraint.

It punishes the detectable rather than the culpable. Where detection is difficult, deterrence requires a heavier penalty to compensate, so the offender who is unlucky enough to be caught bears the weight of everyone who was not.

Its factual claim is weak where it matters most. The evidence is discussed in the next chapter.

Quick revision

• Deterrence justifies punishment by its effect on future conduct through fear.
• General deterrence works on the public by example; special or individual deterrence works on the offender by experience.
• Bentham: all punishment is mischief and is justified only to exclude a greater evil; it must not be groundless, inefficacious, unprofitable or needless; and it must at least outweigh the profit of the offence.
• Bentham's limits make deterrence an argument for the lightest effective sentence, not for severity.
• Four conditions: knowledge of the penalty, belief in detection, a calculating offender, and an available alternative.
• In Indian law: high statutory maxima, mandatory minimums, and section 13 of the Bharatiya Nyaya Sanhita 2023 for repeat offenders.
• Criticisms: uses the offender as a means; has no internal upper limit; punishes the detectable rather than the culpable; and its central factual claim is weakly supported.

Test yourself

1. Distinguish general from special deterrence and say why the distinction matters. General deterrence works on the public at large, who are deterred by the example of the punishment inflicted on the offender. Special or individual deterrence works on the offender, who is deterred by having undergone it. The distinction matters because the two can require different sentences: a sentence heavy enough to make an example may be heavier than this offender needs, and one tailored to this offender may be too light to serve as an example.

2. State Bentham's four cases in which punishment ought not to be inflicted. Where it is groundless, no mischief having been done; where it must be inefficacious, because it cannot act to prevent the mischief; where it is unprofitable, the mischief it produces being greater than the mischief it prevents; and where it is needless, the mischief being preventable at a cheaper rate.

3. Why is it wrong to present the deterrent theory as a theory of severity? Because its own founder's limits point the other way. Bentham held that all punishment is in itself an evil, justified only so far as it excludes a greater evil, and that it must not be needless where the same result could be achieved more cheaply. Deterrence therefore argues for the lightest sentence that will work, and the arguments for severity come from elsewhere.

4. Name the four conditions deterrence requires, and say which fails most often. Knowledge of the penalty; belief that detection is likely; a calculating offender; and an available alternative course of action. The second fails most often: in most jurisdictions the perceived probability of detection is low, which is why the evidence consistently shows that certainty matters more than severity.