CAN WE DETER CRIME?

If you can't do the time, don't do the crime.

Why do we punish criminals? Big Questions like this have an inbuilt tendency to keep a person awake at night, especially if the person in question happens to be a criminologist. Today’s big question raises a number of other Big Questions, too...but don’t worry, I shall detain you for only a few minutes.

Some people prefer to avoid Big Questions entirely. Not you, though. Brave, inquiring readers of the Crime & Psychology Substack have never been known to fear Big Questions. Indeed, they actively seek them out.

Let’s imagine for a moment that the Ancient Greek philosopher, Socrates, drops in on your nan. Being a good-hearted soul, she doubtless welcomes him with a cup of tea and some of that Battenburg cake that no one really likes. Polite Socrates pretends to enjoy the cake but attempts to distract himself by bringing the conversation around to a topic he finds absorbing. ‘Why do we punish criminals?’ he asks your nan. ‘Why, simply because they’re criminals,’ she naïvely replies and segues seamlessly onto the subject of her bunions. Socrates, as we can imagine, grows frustrated, and not just by the fact that there’s still a lot more Battenburg left. He finds your nan’s answer unsatisfactory. It fails to address such issues as whether we are right or wrong to punish criminals, what we might hope to achieve by doing so, and what obstacles might lie in our way. This is what he wants to know about.

But here at Crime & Psychology we do try. And we never talk about bunions.

A complete answer must start by acknowledging that there are several different justifications for punishing criminals. Different Schools of criminology, and different legal systems, prioritise one or the other. Your nan might not have this information at her fingertips, but why should she? She has other things on her mind. We’ll start with a little background.

Most western legal systems reflect the Enlightenment thinking of the era in which they were formulated. The Enlightenment’s most influential criminologist was the Italian, Cesare Beccaria (1738-94). His work exemplified the approach known as Classical Criminology.

Here are Beccaria’s basic ideas (they are straightforward enough, but you’ll see that plenty of consequences spring from them): Human beings are driven by reason. You and I, not to mention and the criminal population, are capable of rational thought. Not only that, but we are capable of acting upon it – in other words, we possess free will. When we are busy deciding whether or not to commit a crime, then, we do so on the basis of rational calculation. Two principles underlie this calculation: maximising pleasure and minimising pain. This view is known as hedonism.                                                                             

The legal system must punish the crime rather than the criminal. This important point is worth a short paragraph:

Each individual citizen was in a contractual relationship with the state. The citizen’s part was to give up certain freedoms in order to maximise the common good. Meanwhile, the state’s part was to enforce the laws. These laws must be written down, so that a citizen who is considering crime can consult a law book, check what the punishment might be, and make a rational decision. Certain dieters do a similar thing: they use a calorie-checking app to look up the content of the foodstuff in front of them. They then claim to make a rational decision about whether or not to have that extra slice. (In this case, rather than Battenburg, which literally everyone is able to resist, let’s assume it’s red velvet cake. I’ve noticed that it has addictive properties. Certain otherwise decent people seem particularly vulnerable to whatever spell they cast on the stuff, back in the cake factory.)

It is vital to establish a mechanical, cause-and-effect relationship between a given crime and its punishment. Every similar crime must be punished in a similar way. The recidivist must be punished identically to the first-time offender. Only in that way could citizens make a well-informed, rational decision about crime.

The purpose of the law is deterrence. Punishment should be daunting enough to make a rational person decide that committing a crime isn’t worth it. Life will be better, hedonistically speaking, if they choose the straight and narrow path over the primrose one.

Laws should be limited. Any punishment greater than what it takes to deter a rational person is simple cruelty. Punishment must be certain – no one must think they can get away with it – and it must be swift – because, as we all know, delays make punishment less effective.

Let’s summarise the Classical position in as few words as possible: Everyone is a rational decision maker. Everyone can choose between obeying the law or committing crime. They do so on the basis of hedonism. The purpose of law is to help them resist. We call this deterrence.

Classical ideas in general, and Beccaria’s in particular, proved enormously influential. They underlay new legal codes in revolutionary France and the United States, among others. They have come to form the essential vocabulary that you and I use even today when we talk about the law.

In the nineteenth century, another approach to criminology appeared. It had vastly different assumptions and implications. This was Positivist Criminology. Like Classical Criminology, it reflected the era in which it was created. Remember that the nineteenth century is sometimes called the ‘century of science’.

The biggest shadow in Positivist Criminology was cast by another Italian, Cesare Lombroso, whom I wrote about in this newsletter. Lombroso, a scientist by training, took a scientific approach to his subject. Here’s what I mean by that: Scientists look for the causes of things. Indeed, some would say that ‘discovering the causes of things’ is what science is all about. Lombroso talked about the causes of crime.

In recent decades, positivist ideas have become much more subtle and sophisticated than anything Lombroso could have dreamt of. For instance, there was Hans Eysenck’s book, Crime and Personality. Eysenck claimed that criminality was intimately linked to personality traits, which in turn were a predictable outcome of the interaction between nature and nurture. Some time later came a book called Crime and Human Nature, by James Wilson and Richard Herrnstein. The authors took a ’biosocial’ approach to explaining who might become and criminal and who might avoid the fate. Recently, so-called ‘biopsychosocial’ approaches have become even more complex and clever. Yet the almost unspoken assumption behind all of them is one that Lombroso would have recognised: crime has causes.

Incidentally, if you would like to read about Eysenck’s model, you can check it out here.

‘Crime has causes’ – this is a big problem for Classical criminologists. A really big one. If crime has causes, it’s unclear that we can deter people from committing it. After all, if you let go of a stone you’re holding, the stone will fall towards the centre of the Earth. It will have no choice in the matter because it is obedient to the laws of gravitation. They cause it to behave in a certain way. A criminal may have no more choice than the stone: he or she is equally obedient to the laws that govern human behaviour.

To put it another way: human beings, like everything else in the universe, obey scientific laws. Their behaviour has causes. That being the case, it is not obvious that they possess free will. In its absence, they cannot truly decide whether or not to commit a crime. They may well know what the law says and know that punishment is coming, but that knowledge may make no difference.

If crime has causes, deterrence shouldn’t work.

Or, at any rate, it should have minimal effect. Admittedly, some philosophers’ conceptions of free will do make certain allowance for causation, but this is one of the Big Questions that we are going to do no more than acknowledge as we pass by. Otherwise we’ll never get finished.

Positivist criminologists tend to be determinists. In other words, they believe human behaviour is subject to natural laws and that to escape their influence is impossible.

What about the law? Well, its purpose is to accomplish one of two things: either incapacitate the criminal so that they are no longer a threat (sometimes called ‘social defence’) or rehabilitate them. The latter should follows on neatly from positivist principles. If we can use science to discover the causes of crime, after all, surely we can use it to eradicate those causes.

A Positivist Criminologist might ask this question: If a person is a born criminal, and is therefore doomed to a life of crime, can you and I truly claim that they ‘choose’ their crimes freely and might therefore be deterred by the threat of punishment?

That’s a sciency question in itself. It deals with causes. In order to answer questions about causes, scientists turn to laboratory experiments. This is where positivist criminology runs into problems. But how to conceive of such an experiment? What would it even look like? What ethics committee would allow us to run an experiment which may turn the participants into criminals? Clearly, we must fall back on other, less convincing, methods.

There is no shortage of anecdotal evidence to indicate that deterrence does not work. Take capital punishment for a moment.

What did Beccaria think? Well, he was no great fan. Beccaria felt that capital punishment was not only indefensibly violent, but it was not even effective. It was justified only under the most extreme circumstances. Even then, Beccaria suspected that life imprisonment would actually be more feared than death, and therefore would be a more effective deterrent.

Of course, capital punishment is reserved for the very worst crimes. The idea seems to be that a potential murderer will consult the law books, sit for a while with chin in hand, pondering options, and then decide that, no, the murder he has been contemplating simply isn’t worth the hassle.

As you can imagine, criminologists have devoted a great deal of time and energy to the question of whether or not this actually happens. They have undertaked such Herculean endeavours as comparing homicide rates in the US and Canada, or across states, or creating time series to see whether introducing or abandoning capital punishment has a significant effect.

As a philosopher colleague of mine used to say, ‘That’s something you could do’. He meant that perhaps you shouldn’t. Such studies are notoriously difficult, if not impossible, to interpret.  Homicide rates inevitably differ between countries, especially two countries as vast as the US and Canada. The threat of capital punishment, (as anyone who has visited both countries can tell you,) is hardly the only difference between them. The same is true for different states in the same country. Mississippi is a very different place indeed from New York.

So much for comparisons across space. Comparisons across time are equally questionable. The reason is that homicide rates fluctuate. Here’s a for-instance. Let’s say that – I don’t know – Vermont introduces capital punishment next year and immediately the state’s homicide rate falls. Does that mean the deterrent was effective?

The only answer to that question is Well, maybe. Who knows? It all depends what else happened. Homicide rates might be affected by all kinds of things, from the weather to the economy to alcohol prices. Did Vermont, in its rush to court, also impose a minimum unit cost on alcohol? Maybe that worked and the threat of capital punishment did not. Did the economy wobble?

Finally, there are extremely abstruse questions about how to analyse the data from such studies. It is no simple matter. I’m not about to get all technical on you here (I have enough of that in my day job) but, for stats-heads, the central difficulty seems to concern autocorrelation and the choice between parametric and non-parametric techniques for coping with it. In case you’re interested in that kind of thing, I have provided a reference.[i]

A companion piece the paper cited above is entitled ‘The Death Penalty: No Evidence For Deterrence[ii]. Of course that is exactly the amount deterrence a positivist might expect.

What about anecdotal evidence? These quotes from convicted criminals come from Christopher Hibbert’s classic, The Roots of Evil:

'I don't want to do eight years, no, but if I have to I have to, and that's all there is to it ... Coal miners don't spend their time worrying about the risk they might get killed by a fall at the coal-face either.'

'I have had a bad fall, and no mistake. But I count on better luck another time. '

'I never thought of anything like that,' a young thief told a magistrate when asked if he was afraid of being sent back to prison. 'You never do. You always think you are going to get away with it.' He was unlucky last time, but next time his luck would hold.[iii]

At one time, hangings were public affairs. Journalists wrote about ‘necktie parties’ and ‘the sheriff’s ball’. In order for deterrence to work at all, citizens must have the opportunity to witness it and learn the absolutely inexorable link between crime and punishment. Indeed, one could argue that the less public execution becomes, the less effective a deterrent it’s likely to be. America’s own executions, meanwhile, tend to take place rather quietly behind prison walls.

Did it work? In a word, no. Enrico Ferri, a positivist criminologist, noted that the most common day for murders to occur was the day of a hanging, and that an extraordinary number of murderers had at some point or other attended one.

Public punishment: the pillory at Charing Cross. Can you see the pickpockets at work?

In England, public hangings became a byword for petty crime of all sorts. Indeed, at one time, if you wanted to run into a Cockney criminal, a hanging was the place to go. Huge crowds might have attended, but few came along for educational purposes. Here is The Times from 1864, reporting on the hanging of the murderer, Franz Muller:

'robbery and violence, loud laughing, oaths, fighting, obscene conduct and still more filthy language reigned round the gallows far and near'. The spectators comprised the most 'incorrigible dregs' of London-'sharpers, thieves, gamblers, retting men, the outsiders of the boxing ring ... the raking of cheap singing halls and billiard rooms.'

In 1807, no fewer than 40 000 spectators came to enjoy a hanging at Newgate (two miscreants called Holloway and Haggerty were executed for murder). It was far too many for the small space around the gallows. The crowds became overwhelming. People trying to escape were knocked down and trampled to death. A pie-man had his basket knocked over. When he tried to pick his merchandise back up again several spectators tripped over him and crushed him to death. There can be no more undignified way to die than face-down in a pie. By the time it was all over, nearly 100 spectators were discovered to be dead or dying.

Our Big Question remains unanswered. Big Questions tend to. Criminology is haunted by a ‘dark figure’. Maybe the threat of punishment has deterred loads of crimes that would otherwise have been committed. The fact is, we just don’t know. How could we?

---

[i] Zimmerman, Paul R – Statistical variability & the deterrent effect of the death penalty, American Law & Economics Review, vol 11, no 2, Fall 2009

[ii] Donohue, John and Justin J. Wolfers. "The Death Penalty: No Evidence For Deterrence," The Economists' Voice, 2006, v3, Article 3.

[iii] All quoted in Hibbert, Christopher – The Roots of Evil: A social history of crime & punishment, Weidenfeld & Nicolson, London, 1963, pp248-9