Unpalatable Theories about Falling Crime

The US crime rate continues to fall. There is no consensus why this is so, but there are a range of diverse theories, ranging from gun control, higher incarceration rates, the collapse of the crack cocaine epidemic, and ‘zero tolerance’ policing. While the diverse theories are interesting, so too are the different reactions that the theories provoke. Despite the difficulties in objectively assessing the theories, all theories are not equal: some are particularly unpalatable.

That people feel confident to express views on the differing theories at all seems surprising. The complexities of the causes and trends of US offending seem tremendously complicated. Yet, as the comments on the above article demonstrate, large numbers of people confidently propound their own views seemingly informed by little more than gut intuition. And despite this marked lack of expertise to assess different theories, there is a significant difference between the reactions that they provoke. For example, higher incarceration is contentious, yet its plausibility seems to depend more on the political persuasion of the audience than any obviously objective criteria.

One theory that may be familiar to those who have read the popular economics book ‘Freakonomics’ is that the legalisation of abortion following the US Supreme Court decision in Roe v Wade had a significant effect, in that ‘unwanted’ babies are more likely to suffer abuse and neglect and are therefore at an increased risk for criminal involvement later in life. To me, this seems to have some plausibility as an explanation on the basis of the authority of, and evidence provided by, its authors. However, I was intrigued to hear the views of some liberal colleagues at dinner this week. Initial reactions were hostile and accompanied by gestures such as rolling of eyes. Reasons put forward were that other causes, such as drug addiction, were more plausible. However, this blog post was prompted by a more interesting reaction, which appeared to be the concern that such theories would encourage unacceptable beliefs, such as that some people are ‘born criminals’. By implication, this could have unacceptable consequences, such as prejudicing the opportunities of those born into particular environments.

Perhaps this tendency should not be surprising: firstly, it seems likely that assessing the beliefs that a theory may invoke in others is a more straightforward task than assessing different crime-reduction theories on their merits alone, particularly where the context is the whole of the US. After all, a partial indication as to how others may react to a theory can be gleaned from that individual’s own reaction to a theory. Secondly, it is a common phenomenon that theories that attempt to explain the causes of human behaviour are commonly taken to have consequences that are not obvious from the theory, often to the great frustration of proponents of that theory.

If this really is part of the reason for the hostility to some theories, it appears to add an interesting dimension to the problem of reducing crime. It may result in the most objectively effective method of reducing crime not gaining acceptance in the population at large. But this may not be an entirely bad thing. It seems to me that it is important to recognise and properly take into account the possible interpretations that people may arrive at. In some cases, these may need to be overcome, perhaps by education or open debate. In other cases, they may reflect some insight into the wider consequences of a theory that are otherwise unforeseen by the theory.

  • Facebook
  • Twitter
  • Reddit

20 Responses to Unpalatable Theories about Falling Crime

  • Peter Wicks says:

    Is this an example of a wider issue about what kind of beliefs we should hold, and promulgate? The point being that while it's obviously good, other things being equal, to hold and promulgate accurate (i.e. evidence-based) theories, this is not the only consideration. From Plato's necessary lie to the white lies we all tell each other (and ourselves) on a daily basis, we all know of examples where telling/believing the truth isn't necessarily the best thing to do.

  • Julia says:

    For me, the "US crime rate continues to fall" headline begs the question of if a crime is a crime if you don't get caught. Im thinking of TV shows like 'The Wire' and 'Breaking bad'. These shows are case studies in another theory; that crime hasn't dropped, it's just changed form to adapt to prosecutorial habits. The Wire exposes how criminal networks have only gotten stronger and more sophisticated since their violent formations when crack and meth were discovered. Its not that less drugs are being traded (or that less people are committing crime), it's that criminal networks have matured and gotten better at disguising themselves, and grown out of using violent conflict resolution. That crack dealers have stopped killing one another at quite so high a rate does not mean that there are less crack dealers or less crack being sold and used.

    • While it is possible that an increasing amount of drug crime is not reported, it would seem hard to hide homicide rates, and they certainly are going down in the US and many other places. I would also think people do report robberies, auto thefts and arson, which are also declining.

      • Julia says:

        I still think the assumption that the crime rate is declining needs to be questioned. I think it is more likely that it is not declining, but rather the nature of crime is changing. In the internet age, more crime can be done electronically instead of face to face, lowering the chances of being caught and the chances of disputes ending violently. Perhaps certain areas of crime are declining, but shouldn't we take account of all areas of crime when using the blanket phrase "US crime rates declining"? Might we expect illegal border crossings have been increasing, and certainly the number of persons living in the US illegally? Have we succeeded in cutting down white collar crime? Is there a noticeable decrease in violence against women and rape (a statistic nearly impossible to determine in the first place, considering how the vast majority of such crimes are never reported)? Are there less people committing the crime of using illicit drugs?

        My major problem with that headline is it suggests we have been doing a good job in "the fight against crime". I'm suggesting it has merely changed in nature, and all we are seeing is a return to what violent/theft crime rates were in the 1960s. Of course homicide and theft increased when new highly addictive synthetic drugs were quickly illegalized and introduced to the illicit market in the 70s and 80s. But since then, the criminal networks in the US have settled down and the bloody battle over turf has subsided. Perhaps there is a connection between this decline and the hundreds of thousands of recent drug prohibition related deaths occurring just south of the border. Maybe we have exported our violent crime.

        • Jim Thornton says:

          Yes indeed. Whether the total amount of crime , presumably weighted for seriousness, is going up or down is what matters.

          But I'd go further. We should only include real crimes, murder, violence, theft, rape etc. Including "victimless" crimes such as drug-dealing in the figures may help politicians make a point, but philosophers should be more careful. Crime dropped the day prohibition ended, but only some of the fall was real – people stopped shooting each other. Some was just that distilling whisky was no longer illegal.

          • Julia says:

            We can omit "victimless crimes" from our figures on crime rates if we want to, and we often do. But I don't like to. It allows politicians to point to falling crime rates as evidence that our current criminal justice policies and forms of punishment are working. Politicians who support the drug war like it when we omit "victimless crimes" when we talk about crime, because it saves them from answering tough questions about why there are still so many victimless crimes.

            I agree that we should only include "real" crime in our criminal statutes. But we don't. I'm worried that treating "real" crimes and "victimless" crimes separately when we speak about crime theoretically makes it easier to forget that in reality, our criminal justice system sees no difference. Our taxes take away children from parents for marijuana possession (http://www.huffingtonpost.com/2011/08/18/parents-marijuana-kids-taken-away_n_930796.html). There are more people in federal prison for drug offenses than for all other crimes combined. We can say victimless crimes are not "real", but there are still real people serving real time for them, and I don't think that earns them a place in theories on crime.

          • Julia says:

            In the last sentence in my comment I meant "I think that earns them a place in theories on crime", not "I don’t think that earns them a place in theories on crime."

  • Sebastian Hallward says:

    Who is to say that the liberal view is not the only rational one? Obviously, drug addiction is a more prolific cause of crime than, say, economic immiseration.

    • Julia says:

      I think you meant to say drug prohibition is a more prolific cause of crime. We have the same rates of drug addiction today as we did at the end of the 19th century, they have not changed much throughout the century. The drugs and the public policy towards drugs and drug users have changed, but not the number of "drug addicts"

  • Sebastian Hallward says:

    No, I was referring to drug addiction. As rule following, utility maximizing liberals we should seek to implement a three-strikes style regime. First strike is a slap on the wrist or stern look, second strike is a slap in the face, third strike is a financial incentive.

  • Julia says:

    Are you joking? Ask California how they are doing with their three strikes law– bankrupt from over incarceration and in violation of the 8th amendment against cruel and unusual punishment due to prisons operating at 200% capacity. Utility maximizing? You can't be serious. We in California spend more on prisons than we do on our public universities, and that's utility maximizing?

  • Sebastian Hallward says:

    Joking? My dear, as an indigenous Californian — and as an exponent of the concept — I can assure you that it is in my vested interest to establish as fact that this is no joking matter — what could be more serious than maximizing the utility of my pocketbook?

  • Sebastian Hallward says:

    "Bankruptcy" is such a negative-sounding word. I like to think of bankruptcy as a hiccup in the amount of Creme de La Mer I use on my hands.

  • Peter Wicks says:

    OK so there is some dispute about whether real, as opposed to reported, crime rates are actually falling. Fine. Whatever. From a point of view of ethics, however, the interesting point (for me anyway) in Paul's initial post was this: "It seems to me that it is important to recognise and properly take into account the possible interpretations [of theories as to why crime rates are falling] that people may arrive at. In some cases, these may need to be overcome, perhaps by education or open debate. In other cases, they may reflect some insight into the wider consequences of a theory that are otherwise unforeseen by the theory."

    The fact that the premise that crime rates ARE falling may be questionable does not diminish the importance of these observations. The theories that we promulgate to explain things we (correctly or incorrectly) observe indeed have consequences, and some of those consequences include the interpretations that others might put on them. This is essentially the foundation of the various taboos we refer to as "political correctness". For example, we may be reluctant to entertain gene-based theories about why men and women are apparently different in certain respects (beyond the obvious) because they could be interpreted as implying advocacy of particular (traditional) roles and/or could encourage the kind of discrimination we have been trying to get away from. Same with race. Similarly, we may be reluctant to promulgate theories about the effect of certain religions on behaviour for fear that it may cause offence and/or incite hostility towards adherents of that religion. All these are examples of possible consequences of promulgating theories that have nothing directly to do with whether those theories are sound, or whether the phenomena they seek to explain are real or imagined.

    The really interesting question raised by Paul's article is to what extent we should take account of such possible consequences when promulgating theories. The benefits of doing so are obvious, but there is also a cost: it is a form of censorship, and censorship leads to groupthink and bad decisions.

  • Julia says:

    Thank you for clarifying the purpose of Paul's post, because it is a fascinating one.

    To be honest, I like to think me raising the dispute, which, by the way, was not about real versus reported crime but rather real versus victimless crime (victimless crime is often reported; there are more people in federal prison for drug offenses than for all other offenses combined), was actually quite in the spirit of Paul's post. I'm worried about the consequences of promulgating theories that crime has dropped. I'm worried about the effect of so-called practical theories on crime that have to create their own definition of crime because the one used in the real world doesn't fit.

    • Peter Wicks says:

      Thanks Julia. I thought you were also drawing a distinction between real crime and reported crime (see in particular your first comment on the thread), but I agree your points also go beyond that. But indeed Paul's basic point is a fascinating one, and one that is rich with possibilities for further discussion!

  • Sebastian Hallward says:

    "But indeed Paul’s basic point is a fascinating one, and one that is rich with possibilities for further discussion!"

    Not to mention rich with possibilities for laughter.

  • Paul Troop says:

    Dear Peter and Julia

    Thank you, in particular to Peter, for wrestling the topic of discussion back onto the theme of my intended post. I suspect that I was not as clear as I could have been, but Peter's eloquent summary was very helpful.

    The background debate that I originally chose to focus on was the crime rate in the US. It seems plausible that it is falling, but the point I intended to make was not dependent on this. More important is the difficulty of identifying *why* it is falling (or rising) in such a complicated web of different potential causes. Against this background debate, there is a clear pattern as to explanations that people find acceptable or unacceptable, and the acceptability of these theories does not seem to be dependent on the objective plausibility of the theories (because of the difficulty of assessing the different causes). Instead, reactions seem to be based more on an understanding of how others will react when presented with the background debate.

    Though I gave the example of the US crime rate, I almost wrote about the recent UK riots. This topic seems to share many of the characteristics. Nobody, to my mind, has provided a convincing account of why they took place. However, this has not prevented all and sundry from arguing what ought to be done to prevent such a situation happening again in the future.

    It seems that there is some kind of 'meta argument' taking place, where the premises involved are not all of the relevant features and causes, but instead a handful of premises combined with a knowledge of how others will behave.

    Putting this in another form, assume that situation (S) has features a, b, c, d, etc. Some of these features will be obvious: say a = transgression of legal / moral norms, b = punishment etc. These may be obvious. Others may be less obvious: c = the weather, d = media reports, e = lead in petrol, etc.

    Situation S will also have various causal relations between those features 1, 2, 3, 4, 5, 6 etc. Similarly, some may be more obvious than others. Say 1 = greater punishment leads to less crime, 2 = more rehabilitation leads to less crime, 3 = drug addiction leads to more crime etc etc.

    Picking which features a, b, c etc and which causal relations between those features 1, 2, 3 are relevant to situation S will depend on the situation. Some situations may be very straightforward, where the features and causal relations can be known. However, other situations (such as the US crime rate, or the causes of the UK riots) seem to be far too complicated to attempt this.

    Situation S does not exist in a vacuum. An individual asked about situation S may have little real idea of which a, b, c or which 1, 2, 3 will affect situation S. What he or she will know, however, is that belief in particular a, b, cs or 1, 2, 3s held by others may affect other areas of his or her life. Say a feature / cause belief is that poverty causes crime. Whether or not this is correct may never be known. But an individual will know, according to his or her economic situation, that economic redistribution will affect his or her life. Ie, if rich, redistribution will make him or her poorer, if poor, economic redistribution will make him or her richer.

    Recognising this tendency would appear to be an important aspect in addressing any normative goals. Even in a simple situation where features and causes could be known with some certainty may result in individuals refusing to believe these relations, or trying to persuade others not to believe them on the basis of understandings of the consequences in that or other areas of life. In some circumstances, this may have positive consequences (Peter gives the examples of resistance to information about genetic differences due to the risks of encouraging racism).

    I am sure that there is also some game theory relevant to this, but I am not sufficiently familiar with the literature to identify it. If anybody could enlighten me, I would be very grateful!

    • Peter Wicks says:

      Another obvious example is climate denial: as Al Gore said, it's difficult to convince somebody of something if they're salary depends on not believing it! (One might add: "or their income depends on other people not believing it.)

      Then there is emotional attachment to belief. There are plenty of comfortably off people who would insist, perhaps out of some sense of guilt for their well-being?, that poverty is the main cause of crime. They'd actually be (potentially, marginally) materially better off if such a theory didn't hold sway, but they are emotionally invested in the belief: it's part of their political, perhaps even social identity.

      I haven't really thought about this explicitly in connection with game theory but I have thought somewhat about the maths of self-fulfilling (I'm feeling very confident today!) or self-negating (You're about to get run over!) beliefs. If T0 is the pre-existing truth of a belief (not taking account of the effect of believing it or not believing it), and S is the (positive, negative or zero) extent to which it is self-fulfilling, B is the extent (again positive or negative, from -1 to +1) to which one believes it, and T is the actual truth of the belief one could very roughly posit an equation of the type T = T0 + SB(1-T0). I guess I'm using a fuzzy logic approach here. This of course only addresses the effect of belief on the truth of the theory itself, not other issues: the maths then becomes much more complicated I fear.

      Another consideration is that, if aspects of the self-help literature is to be believed (and to some extent I think they are), intention is just as important, if not more so, than accurate analysis. To take action in the service of a goal, even if based on erroneous analysis, reinforces our practical commitment to achieving that goal. As long as we have the psychological flexibility to change course as and when it becomes clear our current strategy isn't working, such a "trial and error" approach will often be far more effective than an approach that is heavy on analysis.

      My experience in environmental policy has left me with the impression that we tend to focus far too much on analysis and not enough on clarifying intentions. My (palatable or otherwise) theories as to why this is the case? (i) Civil servants tend to be geeks. (ii) We are only just beginning to develop an adequate scientific (and therefore credible to geeks) understanding of human psychology. (iii) Policy capture: big business hurls accusations of "junk science" and then the civil servants and their allies rush around defensively doing more analysis, rather than putting effective effort into winning the PR battle. (Inconvenient Truth was a wonderful exception.)

  • Peter Wicks says:

    PS Don't think too much about that equation: it's completely wrong! What I wanted was one where the truth-value T collapses to zero when SB = -1, T=1 when SB=+1, and T=T0 when SB=O. The one I posited satisfies the second two criteria but not the first. It must be easy enough to think of a plausible-looking and reasonably simple one that satisfies all three, but not before coffee! Besides I fear I may be undermining my previous efforts to get this thread back on-topic. 🙂


Subscribe Via Email