The Science of Fear, and some data on media overhyping of crime risks

Share on FacebookShare on Google+Email this to someoneTweet about this on Twitter

Since the world started falling apart, books on how crazy we are have never been more popular. Most focus on findings from behavior economics that show how human beings deviate from homo economicus in making decisions, and The Science of Fear by Daniel Gardner is no different. Unlike the others in this newly sexy genre, though, he doesn’t look at economic decisions very much, but instead on how we assess risk — sometimes to our own harm. Consider those who, in the panic after 9/11, switched from riding airplanes to the more dangerous mode of cars and died in car crashes.

I won’t review the book at length since it’s an easy read and well written — worth adding to your “crazy fucking humans” summer reading list. For a taste, though, here’s the author speaking on The Leonard Lopate Show.

Gardner spends some time discussing how outta-whack the media coverage of a problem is with the underlying risk, as when silicone breast implants for awhile appeared to be the next cigarette or trans fat. Over at my personal blog, I put up two entries that have graphs showing, from 1981 to 2007, the per capita rates of homicide and forcible rape (risk), the fraction of all NYT articles that mentioned “murder” or “rape” (coverage), and the coverage-to-risk ratio (overhyping). Here’s the homicide post and the rape post. In both cases, sometime in the early-mid 1990s, in the wake of a generalized hysteria — identity politics, L.A. riots, Third Wave feminism, blaming AIDS on Regan, etc. — the overhyping starts to take off and has remained high up through recent years. We’ve never been safer, yet we’ve only grown more paranoid.

Labels: , , , ,


  1. I wonder if the increased fear hyping holds for non-human threats, a consequence of 24 hour news outlets online and on cable. A few summers ago, for ex:, the fear of shark attacks was high as a result of something like nine attacks over a three month period across the US.

  2. One thing I’ve noted is that when an advocate or partisan sees a trend that reinforces his convictions, he’ll talk about it for decades. For example, violent crime really was growing at sopme point during the sventies. It made sense to talk about skyrocketing crime. But then it flattened out and started to decline. For people who made crime their issues, it was still skyrocketing. Skyrocketing crime had become a permanent part of their mental world. They never sat down to do the arithmetic and figure out that if the trend had continued, everyone would have been murdered ten years ago.  
    I had the same argument with an alternative medicine guy. There was a time when the unhealthy American lifestyle was leading to increased heart disease, and a lot of good books were written about that. But even though deaths from heart disease have been declining steadily for at least 50 years, the books are still on the shelves.

  3. He reviews some studies similar to the ones I just did where the researchers looked at risk and coverage of cancer — same thing. 
    At the end of the chapter on media coverage, he mentions that in the summer of 2001, shark attacks were all over the media, despite posing a lesser threat than lightning bolts. In fact, that was the big scare they were all talking about at the beginning of the week of 9/11, which then erased sharks from the threat list. 
    You could do something similar for natural disasters — just look up how many people died due to tornadoes per capita, and search the NYT for “tornado(es).”

  4. A lot of it is pretty deeply rooted in primal fears and esthetic considerations. For example, squirrels spread more rabies than bats. Poisonous snakes kill 5 to 10 people a year in the US. Bears seldom kill anyone. And so on.  
    About 20 fatal bear attacks in the US and Canada in 10 years. 7 of the deaths were in the US excluding Alaska, and in one of those cases it was a trained circus bear. 
    The list does include a number of fatal black bear attacks, which are supposed to be almost non-existent.  
    I used to know a guy (ca. 1986) who was terrified of AIDS. He was low risk as far as I knew. He was 80 lbs. overweight, a poorly controlled diabetic, and refused to use his seat belt.

  5. agnostic, 
    I think I missed the one you did on risk and coverage of cancer. Where is that?

  6. I thought it was in the book, but I just looked it up, and it wasn’t cancer. It was mad-cow disease, SARS, and avian flu. By a British think tank called The King’s Fund. They showed that far fewer deaths were needed for an exotic, novel disease like BSE to get an article than for more mundane but higher-risk killers.

  7. I may have missed it in your original posts but did you check to see if perhaps the jump in coverage could be correlated with local jumps in crime. I’m specifically thinking of jumps in New York or Las Angeles crime levels might spur more coverage even while crime levels nationwide are falling. 
    I would love to see how well the media coverage chart maps to specific high-profile cases. Maybe it’s the case that a specific high profile case spurs a wave of similar reporting, sort of how a entertainment programming follows on the heels of the next big hit.

  8. Human irrationality towards risk is well documented in the literature – you need to factor in voluntary vs involuntary (voluntary risk is up to 3 orders of magnitude more tolerable than involuntary risk), unknown, dread (shark attack is a good ‘dread’ one). The plane vs car one is one of the classics – people feel (irrationally) that they are in control driving a car, so it plots as voluntary risk, whereas sitting passively captive on a plane being flown by someone else is involuntary risk. Plus people spend a lot more time in cars, so it plots as familiar and therefore gets a discount, even though the much greater amount of time spent exposed to the hazard makes the probability that it will be realised that much greater. Occupational risk plots as voluntary – people choose to go to work in order to obtain the benefit. 
    Multiple fatality events are always regarded as worse, more shocking, than a larger series of single fatalities. I see this as a kind of ‘survival of the species/tribe/whatever’ reaction – the more people killed in a single event, and the more the people are closer to home/like us, the (much) worse it is regarded. 
    John’s reference to ‘primal fears’ is a good one – this translates as the ‘dread’ and ‘unknown’ factors in the risk literature which amplify risk aversion. We fear wild animal attack, even though domestic dogs kill more people than sharks or bears. 
    The other irrational factor with humans is that they want a certain level of risk – if a dangerous road that has hairpin bends etc is straightened out, people drive faster to raise the risk level up to where it was before the road was fixed. Risk is interesting/exciting, as long as it is voluntary and felt to be controllable – thrill seeking. 
    The third thing that comes into play, which I think gets to your point, is lack of awareness of risk, the ‘unknown’ factor or lack of familiarity – when a hitherto unsuspected source of risk becomes known through reporting and gets hyped for its novelty and dread value, it causes a much greater aversion reaction than a source of risk that is familiar, even if the quantified risk is very much lower. Media coverage will frequently correspond to the strength of the unknown or dread aversion reaction – the media are selling a product, after all, not engaging in public education on how people can minimise the risk they are exposed to. 
    A certain element of rationality comes in, in terms of the trade off of risk vs benefit – people will take a risk running across a busy road if they perceive a benefit to themselves in getting across the road more quickly. People will accept a certain level of occupational risk if they perceive that the financial reward makes tolerating the risk worthwhile. 
    Interestingly, the biggest risk takers, and those most willing to expose others to risk, everything else being equal, are white males – they plot higher for risk tolerance than non-white males and all females. Risk tolerance plots inversely with economic well being/socio-economic status. And of course, within group, the most notorious risk takers of all are adolescent males.

  9. When news became infotainment it had the effect of nationalizing fear. Shark attacks were only ever a concern for about 1 % or maybe .1 % of the population (people in high-risk areas who spent time in the water) but it became a media cliche.

  10. Our failure to beware of wolves is the result of communist propaganda! They have shifted the blame onto bears, and Colbert is their chief dupe.

  11. Home invasions are the latest sort of crime to stir up widespread fear. To some extent this fear is entirely reasonable – the mere idea of being subject to violent attack in one’s own house, in what should be the safest of all places, is particularly chilling. It’s also not uncommon for home invasions to end in deadly violence.  
    And yet, while home invasions get plenty of media hype, what gets far less attention is the fact that a significant percentage of the victims, perhaps most of them, are involved in the drug trade.

  12. According to a Google Trends search for ‘risk’ searchers’ interest in risk has decreased over a 5 year period while news reference volume of the term has quite dramatically increased. 
    The chart has a bit of a academic cycle to it. There also is an interesting significant annual dip in interest in risk just prior to the new year…

  13. agnostic, 
    No offense, but you’re on crack. Read this and search for “bicycle”: 
    In a civilized modern society, the risk of suffering human-on-human predation should be only slightly higher than the risk of being attacked by, say, a grizzly bear. And yes, you should be able to park your bike without locking it. 
    The basic problem is that your reference frame (US, 1981-now) is very narrow and very bizarre. Relative to Somalia, or Manhattan in the ’70s, sure – crime in the US today is not a problem. Relative to Japan at any period, it’s a tremendous problem. Relative to the UK a century ago, it has gone up by about 5000%: 
    Here’s another suggestion – go through American or British newspapers of 100 years ago and look at how they reported crime. A friend of my wife’s was attacked in downtown San Francisco at 10:30 in the evening a few weeks ago. She needed stitches in her face.  
    A century ago, this would have been front-page news. Today there is no question of that. It might be mentioned in a police report column. If she’d been killed, it would have been in the papers, I’m sure – probably a few stories. If she’d been killed in 1909, it would have been a cause celebre. 
    The fact that, in any major American city in 2009, there are go areas and no-go areas, places you can go in the day but not at night, places you can’t go at all – this would astonish and appall our great-greatparents. So, please, let’s not hear it about “hysterical” fear of crime. You’d need to ramp it up about five orders of magnitude to get to the hysteria level.

  14. Mencius, 
    No offense, but you’re innumerate (as well as on crack). 
    Look at the graphs I provided — crime rates are going down while crime reporting is going up. Make any sense, or no sense? I’m not arguing what the rational level of crime reporting would be for a given crime rate — should each murder get 1 article, 2, or perhaps only every 100 murders should get an article? 
    I’m simply showing how risk is plummeting while fear is shooting up — obviously irrational. 
    In point of fact, the existence of go vs. no-go areas would *not* appall our great-grandparents. Mine were in their 20s or 30s during the 1920s — when homicide rates were huge, essentially as high as during the Crack Wars (from CDC data): 
    I keep reading that you don’t know any history, and here’s another case in point. The 1920s and early-mid 1930s are famous for their levels of violence, corruption, depravity, drug use, and so on. Our great-grandparents who lived through that time would laugh at how spoiled we are now. Tonight, stay in and rent the Godfather movies — you’ll learn something about what crime in our great-grandparents’ time was like. 
    Also, that Parliament graph is misleading. I plan on putting up a post on that soon. Two clues that is doesn’t say what we think it says: 
    1) It looks nothing like a graph of any serious crime, or composite of such crimes, over any length of time, in any country studied. In particular, there is only a smooth trend without any year-to-year wobbling around it. Also completely misses the 1920s – ’30s crime wave that equaled the 1970s – ’80s wave. 
    2) That graph only measures indictable offenses that were a) reported to police, and then b) recorded by police. Change in either of these will produce a change in the graph, without any change in the underlying crime rate. 
    Specifically, note that the size of the police force is 3 times larger now than it was in 1900. More policemen means more people to receive and record crime complaints. Also note that the overall shape of the two is the same — suggesting that the steadily rising police force has a strong influence, unlike the shaky up-and-down pattern of a real crime rate graph. 
    I’m not saying that there was no change in crime rates — that the graph is totally illusory, and due only to more police — but much of the explosion must be due to this. Again, rates for single crimes or composites of them never look like that, in the UK or anywhere else. And certainly not for the 20th C, which saw a big uptick during the Roaring Twenties.

  15. I’m surprised agnostic didn’t link to his homicide rates over the last 800 years post
    At my own blog I had an argument with Lawrence Auster regarding why I prefer statistics to news stories, more specifically regarding attacks on white women. Evil neocon/testing99/Whiskey also seems to go wrong by relying on (even fictional) media to let him know what’s going on in preference to the GSS or other stats.

  16. > risk [from crime] is plummeting while fear is shooting up — obviously irrational 
    Maybe so if you restrict your attention to absolute risk. But I’ll bet another value the psyche heeds is the fraction of all premature deaths that are caused by XYZ. It’s bitter to die young. Perhaps crime’s fraction is higher now than it was in the 1920s or the 19th C.

  17. go wrong by relying on (even fictional) media  
    Awhile back I saw a book on sociopaths being taught to Administration of Justice students that used fictional examples as illustrations. Basically future policemen were having their media-fed presumptions reinforced by the same media they saw originally. 
    The lowest level science you’ll ever see will be in bottom level criminology, counseling psychology, school psychology, sexology, rehab psychology, etc. There are a lot of freelancers going around giving inservices to workers in the field that are close to the zero level of credibility.

  18. Look at the graphs I provided — crime rates are going down while crime reporting is going up. Make any sense, or no sense? 
    Makes perfect sense. Crime is still insanely high. Crime reporting is still insanely low. The situation is not one of hysteria, but of anesthesia – not overhyping, but self-coordinating censorship. As usual, reactionary reality is seeping back into the system a little, which accounts for the transition you see. 
    Do you know anything of how the journalism profession looks at “sensational” crime reporting? Again, if you look at the attitude of the 2009 journalist to the 1909 journalist, this is one of the greatest sins the former sees in the latter. 
    This meme of “overhyped” crime reporting, which you obviously did not invent, is exactly the means by which said censorship enforces itself. Poland must always be invading Germany. 
    I keep reading that you don’t know any history 
    Hey, I thought you were all about data, not anonymous hearsay! :-) 
    The 1920s and early-mid 1930s are famous for their levels of violence, corruption, depravity, drug use, and so on. 
    Yes. Which were unprecedented at the time and would barely raise an eyebrow now. Also, this is why I mentioned 1909, not 1929. 
    Notice the left end of that curve in your GIF? Notice the way it starts spiking up around 1905 or so? Do you think the pre-1905 area is a glitch? What do you think the pre-1900 statistics look like? What you’re seeing at the start of this curve is the breakdown of social order in 20th-century America – eg, the development of organized crime in immigrant ghettoes. 
    Also, you continue to betray your presentism and parochialism – you’re talking about a period in American history, ie, that of Prohibition. While civic order was declining in every nation in the 20th, America in general and the “Chicago gangster” were bywords for these qualities worldwide. 
    This is why I recommend the UK as a case study of the transition from 19th-century standards of government to 20th-century standards. 100 years ago the UK was the least Americanized of nations; today it is arguably more Americanized than America. To see the transition described, have a look at Peter Hitchens’ Abolition of Britain. You might also find yourself engaged by his Brief History of Crime
    Can you do me a favor? Or yourself, possibly? Pull yourself out of the spreadsheets for a moment, and have a look at the America of 100 years ago, courtesy of this English reporter.  
    The entire book is interesting. But for example, look at page 144, in which he considers it remarkable that the streets of Chicago are not entirely safe at night. Page 150: “The police force is so weak that men and women are held up and robbed almost nightly within the city limits.” However, Steevens does not mention anything like today’s no-go areas. And he did not come to America to praise it. 
    I did not compile the UK statistics I linked to; HMG did. I suspect that, if anything, they are a little sanitized. The dip in the numbers after 1980 is especially suspicious. If you have a look at this blog, you’ll see in what direction British crime statistics are massaged. I suspect that the older numbers are, if anything, more reflective of actual conditions. See under: Theodore Dalrymple, Life at the Bottom
    And as for innumeracy: chew on Carlyle for a little while. There is much to be said for your spreadsheets, Horatio, but there is much in heaven and earth that is not in them.

  19. For an example of what Seiyo is talking about, see the Wikipedia article on Crime in Japan
    In 1989 Japan experienced 1.3 robberies per 100,000 population, compared with 48.6 for West Germany, 65.8 for Great Britain, and 233.0 for the United States. 
    Robbery is a much better metric for the overall level of systematic criminality, I think, than murder. Many murders are committed by people who are not career criminals. Few robberies are. 
    My unscientific impression is that these levels are roughly comparable to those in a civilized European country of a century ago. Unfortunately I forget the source, but I recall reading that the UK in 1900 had about one robbery per day, which puts it at about Japanese levels. Perhaps you could dig up this number for the US, or for the UK. I’d find the 1850-1950 trends quite interesting in both cases. 
    Of course, you could blame reporting, but the lower the level of crime, the more likely crime is to be reported, n’est ce pas? And, of course, with improvements in technology you’d expect these numbers to get better, rather than worse. The Scotland Yard of a century ago didn’t have DNA analysis, CCTV, etc, etc… 
    And yes, there’s an HBD variable. But whose fault is that? There are many explanations for the decline in Western public safety over the 20th century – but they all come back, in the end, to a case of government failure. 
    Perhaps it’s a coincidence that the 20th century was also the century in which it was thought that government policy should and could be planned by statistics, rather than phronetic decision, common sense and tradition – “by steam,” in Carlyle’s phrase. Or perhaps it’s not a coincidence. I’m afraid you’ll have to apply your judgment on this one.

  20. Also, see this comment – from the coal face, as it were – on contemporary US crime reporting. I find it quite credible. Your spreadsheets are working upstream against Goodhart’s law

  21. And, not to eat up the bits or anything – but I’d assume any reader of GNXP is familiar with Sam Francis’ concept of anarcho-tyranny. Maybe I assume too much.

  22. Okay, I found my source
    In 1893 the annual number of recorded robberies in England and Wales fell below 400. There were then never as many as 400 recorded robberies a year in the whole of England and Wales until 1941. In stark contrast, from February to December 2001 there were never as few as 400 recorded robberies a month in the London Borough of Lambeth alone. 
    There’s a 200-page study behind the link. Knock yourself out, Excel.

  23. BTW, TGGP’s wolf link is majorly cool. Check it out. Exactly what I’m talking about – except with four-legged wolves, not two-legged…

  24. What do you think the pre-1900 statistics look like?  
    Pre-1900, the rates are even more appalling — have another look at that graph in my blog post on historical homicide rates in England. That’s true for all other countries studied of course. Download Manuel Eisner’s article that includes graphs for most of Western Europe: 
    No presentism (data go back to 1350 in some cases), no parochialism (includes almost all of Western Europe). That’s how I know what you’re saying is misleading. 
    There are similar spikes during the 1920s – 40s in the UK, the US, the Netherlands / Belgium, Germany, and Italy, although not in Scandinavia or Switzerland. Immigrant ghettos and Blacks were not necessary, since it was much more general than the US.  
    This will be clearer when I paste all of the graphs into the upcoming post. 
    And as I said before, a lot of the change in the other indictable offenses is likely due to the tripling of the police force during the 20th C. Murdered bodies can’t be ignored, but petty theft can be. The Parliament PDF itself says that changes in reporting of crime accounts for some of the increase.

  25. Yo — let’s give peace a chance and not kill each other like Biggie and Pac.  
    H-bd community is too small for east vs. west feuds :) 
    As I noted to agnostic & Mencius, I think both of you have some valid points 
    here. I think the basic problem is that “crime” is a sum of two terms: 
    (per-capita cost of being victimized by crime) + (per-capita cost of 
    preventing victimization) = total per-capita cost to society
    Mencius is responding to the fact that there have been massive 
    increases in the second term: crime prevention costs (sales of locks & 
    car alarms, mass migrations to gated & non-NAM communities, 
    incarceration of 1/3 of the black male population, etc.). I also 
    believe that rates of aggravated assault and robbery have 
    significantly increased in the US (by 1-2 orders of magnitude 
    depending on the state — see e.g. Figure 2 here: 
    Agnostic is responding to the fact that these massive increases in 
    per-capita spending on crime prevention have resulted in a rough 
    steady state in murders-per-capita over the century, and a decrease 
    over the last decade or two.  
    But murders aren’t the only stat, though they are the gold standard.  
    The overall crime rates (assault, robbery, etc.) are increasing monotonically in 
    this DOJ pdf (see the “Historical” section), which means they’d be *outta sight* without those big crime prevention measures:  
    Overall, I think the “total cost of crime” (meaning dollars lost to either crime 
    itself or prevention of crime) has significantly increased in the US over the 
    course of the twentieth century. I would be amazed if the per-capita cost had not risen by 1-2 orders of magnitude. And this is for other countries too. Think about the capital expenditure to outfit every street corner in the UK with a CCTV…and this is for a country that’s still like 92-93% white!  
    That’s kind of why the hypothesis that “crime is overhyped” seems counterintuitive to many people — if you’re in or near a big city, or picking out a house, a LOT of effort & money is spent staying out of high crime neighborhoods and avoiding certain places at night.  
    PS: this doesn’t deeply get into two other confounding factors. H-bd is a huge one and IMO fairly well ploughed. Crime kept dropping in the 90s in large part because Clinton kept locking up NAMs.  
    With such devotion from African-Americans, you’d think Clinton had actually done something for us. But one African-American fellow mentioned to me the other day that more blacks went to prison under Clinton’s administration than under any other president.  
    Had people thought it was “paranoid” to lock up NAMs, we’d have a lot more Tookie Williams and Mumias gunning down cops.  
    The other big factor is science and technology. The per-capita cost of solving or preventing certain kinds of crimes has certainly decreased with technology (e.g. computerized fingerprint databases, DNA forensics). But the cost of committing other crimes has also decreased with tech (using Google Streetview to plan robberies, setting up passwordless wifi honeypots to steal account information). I’m tempted to call this a wash in the absence of a detailed study.  
    Moreover — those old modes collapsed, but to what extent did modern science make those older modes of social organization less sustainable? What impact have widespread images of beautiful people had on the human psyche and the capacity for pair-bonding? What about the impact of telecommunication and transport on organic neighborhood watches?  
    Will we find some heretofore unappreciated social factor that promotes crime as surely as radium promoted cancer?  
    One huge factor in re: crime is that many people (perhaps 50%?) don’t know their neighbors anymore. They know their virtual neighbors — people from work or from their hobbies. But they don’t know the people next door. Wouldn’t recognize them from Adam. This makes it harder to detect strangers, unless they be NAMs.

  26. Mencius, 
    The way your comments read, I’d imagine you must sound like Stewie – from Family Guy – in person? 
    But in any case the English of London, over a century or so ago had an effective policy against crime and the criminal class in general, they simply deported them en masse from the “East End” to Australia!

  27. geecee (Godless Capitalist?) I expect you are right about cost. The damages from natural disasters has also increased: we have lots of very valuable buildings to be destroyed now. I think you may also be correct about the change in measures being taken against crime: Greg Clark has noted that one could literally get away with murder in medieval England if you could read a passage and thereby “prove” you were clergy. 
    I think murder is far and away superior to other measures of crime, but I’d note that auto-theft is also considered to be quite reliable. Unfortunately, it is not nearly as helpful in making comparisons over long scales of time! 
    Are any of you familiar with Mark Kleiman and the “HOPE program”? It’s only been used on a smaller scale so far, but based on it he thinks it can cut crime by half. Kleiman is thoroughly liberal (he blogs at “The Reality Based Community”) but his scheme of mass ankle-braceletting reminds me of Mencius’ proposals for crime (just without the Star Trek diversions). I’ve been listening to Tim Harford on the “rational model of crime” recently and while I think there is some merit to it, I’m with Bryan Caplan in concluding that criminals systematically tend toward hyperbolic discounting so that an increase in severity has a tough job in making up for a reduction in certainty. A final benefit of the program is to make steady employment more normative (perhaps even reducing the need for immigration).

  28. agnostic, 
    I can’t read the Eisner paper. It’s behind a firewall. Not that I could be more interested in your data than you are in mine. You’ll note that this is the problem with arguing in terms of spreadsheets: either you agree to disagree, or you say nothing, or you pop each other like Biggie and 2Pac. 
    Simply put, I don’t trust your homicide data. It conflicts with what I know of the time and place. It also conflicts with those robbery numbers I gave – which I’d love to see you check out. 
    You know how this feels, of course, because you don’t trust my data. I don’t trust my data either. I mention it because it confirms my historical understanding of the time and place, and data – in the sense of statistics – seems to be what you want to hear. 
    Victorian England was simply not a hotbed of chronic violence, social degeneration, and organized crime. It did not have 2800 criminal gangs, for instance. Granted – when you read books from the period, you will constantly see people complaining about chronic violence, social degeneration, and organized crime. You will also note that the actual magnitude of their complaints is hilarious by “Gang Leader for a Day” standards. 
    For the US equivalent, if Steevens does not convince you, check out Jacob Riis’ How The Other Half Lives sometime. Riis is an early progressive propagandist, and his goal is to convince you that the life of the urban poor is incredibly squalid, etc. For example, the sidewalk outside their houses isn’t always swept. Or whatever. 
    From the perspective of someone who spends a good bit of time with primary sources, your approach to data is remarkably open-minded. Even when data is technically accurate, it can be massaged in many ways that causes it to present a totally misleading picture. For example, I’m quite confident that this is why you always see homicide rates, and never robbery rates – which  
    When I read a paper like the ones you cite, I simply assume that the data is telling whatever story the author wants it to tell.Within this limitation, data can sometimes be interesting – but they should be drunk as single-malts, not vast industrial blends across all time and space. Every time you blend, splice a curve, merge spreadsheets, etc, you have an opportunity to massage. (See under: paleoclimatology.) Many, if not most, will take it.  
    God himself only knows what has gone into Eisner’s model. Do you? If not, why would you cite him?  
    As someone who considers himself careful with sources, whether textual or data, I would rather see the crime rate for Lambeth than that for London, and that for London than that for Great Britain. I would rather see snapshots of 1909 and 2009, than a curve connecting the two. Etc. Etc. 
    But I think I have made a real effort, here, to play the game by your rules. Enough. I would be very curious to see you investigate the discrepancy between my statistics and yours, and I promise to trust whatever conclusions you arrive at. Perhaps your data can demonstrate that the J-Crips were tagging Houndsditch while Sherlock and Watson were tagging hounds at the Baskervilles’. If so, I will have no choice but to accept it. :-)

  29. pconroy, 
    Sadly, my substitute for a TV is Google Books. Don’t explain, though – perhaps it’s better that I don’t know. :-)

  30. geecee is quite right to focus on the effects of crime. After all, we all do. 
    There are certain individual murderees that everyone reading this knows of: Emmett Till, for instance. Why is the case of Emmett Till important? Because it was a political murder. The killers had a political motive, ie, intimidating blacks to maintain white social domination. 
    Well. One reason for the decline in murders from the ’70s through the ’90s is a simple matter of causality: after white flight, certain kinds of social interactions were much less common.  
    Can you say ethnic cleansing, boys and girls? I knew you could. If you put a link from the first page to the second, how long do ya think it’d last? Perhaps you could quote E. Michael Jones as your source. 
    This is a good illustration of why literary methods in history are so superior to the pure quantitative approach. If you look at an ethnic-cleansing event by measuring the number of actual murders, you will get wildly varying numbers, which depend on completely irrelevant details in how the cleansing is carried out.  
    Whereas the basic sequence in an ethnic-cleansing episode is always the same: the displaced population flees its homes and neighborhoods, because it fears for its personal safety. These homes and neighborhoods are taken over by the population which performed the actual intimidation. Still a third party is the set of people who actually caused the event. 
    In “white flight,” itself a euphemism (wouldn’t it be more interesting to know who did the pursuing?) the displaced populations were generally American Catholics and other white ethnics, the people TR called “hyphenated-Americans” – even including many downscale, Yiddophone Jews. And of course tike the Irish and Italian populations of the former quaint fishing village in which I live. All around the country, populations of this type fled because they feared for their lives. Read Anthony Lukas’ Common Ground, about what happened in Boston – an impeccable work of mainstream journalism.  
    So basically, agnostic, this is the real past behind your little graph with the two humps. At least, this is the real past I see. What real past do you see? Do your data, at least your late 20C data, contradict my perspective in any way? I don’t think so – and this is my complaint with quantitative history.

  31. Wellsir you know I always get a kick out of your perspective (despite usually lacking the background needed to comment on it)… but here you’re arguably verging on sophism much like when you argued that a certain contemporary american dude was barely different from a chattel slave because he had poor prospects for changing to a new employer without making sacrifices.  
    Sure, there is a valid formal analogy between white flight and ethnic cleansing as seen, eg, recently in Iraq. But you ignore the fact that the americans who fled had cars (meaning they could typically retain their same jobs, I imagine), that they had capital to ease their relocation, a non-clannish high-trust society to move about it, etc etc – and that it thus took orders of magnitude less pressure to make them flee, the kind of quantitative difference that makes a qualitative difference. It didn’t take daily carbombs, bullets in the mail, or endless random off-the-street kidnappings. It’s all relative in life: in 2109, if the GDP per cap is $400,000 in today’s dollars and supersonic trains let you commute 500 miles, I imagine people will pack up and run away screaming if their neighborhood goes slightly out of fashion or faces a sudden increase in squirrels getting on people’s bird-feeders. I’m not denying that the inner cities became unlivable, but it wasn’t what you would call a crime against humanity in the sense commonly understood.

  32. But you ignore the fact that the americans who fled had cars (meaning they could typically retain their same jobs, I imagine), that they had capital to ease their relocation, a non-clannish high-trust society to move about it, etc etc  
    But it doesn’t matter if they had capital. The Indians who fled Idi Amin had capital. The Jews who fled pogroms had capital. The ethnic Chinese who fled communist Vietnam had capital. We still call it ethnic cleansing.  
    Indeed, I’d bet ethnic cleansing is more likely to happen to outnumbered high IQ groups without media (and hence military) support than it is to low IQ groups.  
    It didn’t take daily carbombs, bullets in the mail, or endless random off-the-street kidnappings.  
    Daily carbombs? No, but daily drive-bys. Bullets in the mail? No, but beatings on the street for being white. Endless off-the-street kidnappings? Well, usually when a rape occurs, it’s prosecuted as “rape and kidnapping”. And as you may know, the ratio of black-on-white to white-on-black rapes in the US is greater than 100:1.  
    Basically, have you ever had the misfortune of wandering into a black neighborhood at night? A “no go” area is ethnically cleansed by default.  
    Heck, even lefties like Sandra Bernard and Eldridge Cleaver admit in their more candid moments that black-on-white rape is not just random, but ethnically and politically motivated.  
    Basically, blacks may not be as organized in their campaign of ethnic intimidation as Muslims are or white Southerners were, but where there’s a will (and media supporting fire), there’s a way.  
    Review it on NewsTrust 
    Editor?s Note: A federal racketeering investigation culminated Thursday in the arrest of 147 members and associates of a predominantly Latino gang charged with murder, weapons and drug trafficking, and carrying out racially motivated attacks against African Americans to drive them out of a southern California city. NAM associate editor Earl Ofari Hutchinson comments on the rising Latino-on-black violence that underlies Los Angeles? biggest gang bust. 
    Two weeks before U.S. Attorney Thomas O?Brien made what he called the biggest gang bust in recent history in Los Angeles, in which the Latino gang members rounded up allegedly attacked blacks, an African-American family, who had been terrorized by a Latino street gang, fled their home in Duarte, a suburb east of Los Angeles. 
    A few months before that, a Los Angeles County Commission on Human Relations report on hate violence for 2007 found that overall Latinos committed nearly half of the hate attacks in the county, while blacks committed 30 percent of the hate attacks. The victims of the race violence almost always have been other blacks or Latinos. 
    The gang attacks no doubt are deliberately designed by the gang hate purveyors to send the message to blacks that this is our turf, and you’re an interloper.  
    Ofari Hutchinson, being black, reports sympathetically on blacks. But glimpses of the other side of the ledger come through:  
    That battle has spawned an even bigger fight in poor neighborhoods between gangs over crime and drug turfs. The gang and prison violence has resulted in dozens of injuries and a few deaths. And there are constant rumors that black and Latino prison gangs have ordered retaliatory hits on other blacks and Latinos on the streets as part of their street turf battles. 
    What is that but mutual ethnic cleansing? The only difference is that both sides have some sympathizers in the media, so it’s labeled as “cleansing” rather than “flight”. The initial media frame is everything. And whites are the bad guys coming (“gentrification”) and going (“white flight”).

  33. Even when data is technically accurate, it can be massaged in many ways that causes it to present a totally misleading picture  
    This is absolutely right, but also the cause of my only minor disagreement with Mencius. I certainly agree that when the individual data points don’t match the big picture, one needs to dig. And I also agree that literary sources can be useful in giving an impression of the big picture.  
    However, I think it concedes too much to formulate it as “literary” vs. “spreadsheet” evidence. It is *always* the case that when a number is found to be misleading, there is an erroneous, unrepresentative, or fradulent calculation behind it.  
    So it was for Boas’ claim that skull shape changed upon crossing the Atlantic (disproved almost 100 years later by Jantz and Sparkz).  
    So it was for Gould’s “disproof” of factor analysis and the importance of g (see any study correlating g with brain parameters.) 
    So it was for Diamond and Lewontin’s casuistic use of blood group data to “disprove” the manifest fact of genomic racial clustering (see Rosenberg et al., Science 2005).  
    So it was for the tag-team takedown in Science of Roughgarden’s claim that “sexual selection is obsolete”.  
    Etcetera, etcetera.  
    Basically, when the other side’s numbers don’t add up — which is often when they’re being manipulated to support a “progressive” agenda — one must hit the bricks and go to war technically rather than conceding the battlefield.  
    Oftentimes the lie has decades or more to get around the world before the truth gets its boots on. Whole fields have been built on Boas’ lies, for example. And stereotype threat remains in textbooks even after Sackett, Hardison, and Cullet published their devastating takedown.  
    In this thread I think both parties are arguing in good faith — and if you step back and look at the *total cost* of crime by incorporating per-capita prevention costs, you can resolve the two perspectives.

  34. That said — as a general rule, in social science progressives have been consistently wrong over the last century. This is because it is always easier to ask “why” than to answer “why”. Feynman has a great bit on this in the Feynman lectures, but one can summarize it as: 
    “Mommy, why is the sky blue? Rayleigh scattering, my dear.” 
    The progressive is like the kid, and is very good at asking why. Why this social arrangement? Why that policy? Why this custom? This is the first step of science.  
    And the standard conservative is like the parent. They don’t know, but they know it’s always been done that way. So they resort to “God said so” or “just because”. This comes off as intellectually weak.  
    The evolutionary conservative or bioconservative has a different perspective. He recognizes that an institution or custom often has an *unarticulated, evolved logic* to it. When it comes to architecture, for example, there is a reason that all those Frank Gehry buildings have problems that older buildings don’t. Those traditional arrangements are like highly debugged code.  
    There’s a subtle reason that programmers always want to throw away the code and start over. The reason is that they think the old code is a mess. And here is the interesting observation: they are probably wrong. The reason that they think the old code is a mess is because of a cardinal, fundamental law of programming: 
    It?s harder to read code than to write it. 
    This is why code reuse is so hard. This is why everybody on your team has a different function they like to use for splitting strings into arrays of strings. They write their own function because it’s easier and more fun than figuring out how the old function works. 
    As a corollary of this axiom, you can ask almost any programmer today about the code they are working on. “It’s a big hairy mess,” they will tell you. “I’d like nothing better than to throw it out and start over.” 
    Why is it a mess? 
    “Well,” they say, “look at this function. It is two pages long! None of this stuff belongs in there! I don’t know what half of these API calls are for.” 
    Before Borland’s new spreadsheet for Windows shipped, Philippe Kahn, the colorful founder of Borland, was quoted a lot in the press bragging about how Quattro Pro would be much better than Microsoft Excel, because it was written from scratch. All new source code! As if source code rusted. 
    The idea that new code is better than old is patently absurd. Old code has been used. It has been tested. Lots of bugs have been found, and they’ve been fixed. There’s nothing wrong with it. It doesn’t acquire bugs just by sitting around on your hard drive. Au contraire, baby! Is software supposed to be like an old Dodge Dart, that rusts just sitting in the garage? Is software like a teddy bear that’s kind of gross if it’s not made out of all new material? 
    Back to that two page function. Yes, I know, it’s just a simple function to display a window, but it has grown little hairs and stuff on it and nobody knows why. Well, I’ll tell you why: those are bug fixes. One of them fixes that bug that Nancy had when she tried to install the thing on a computer that didn’t have Internet Explorer. Another one fixes that bug that occurs in low memory conditions. Another one fixes that bug that occurred when the file is on a floppy disk and the user yanks out the disk in the middle. That LoadLibrary call is ugly but it makes the code work on old versions of Windows 95. 
    Each of these bugs took weeks of real-world usage before they were found. The programmer might have spent a couple of days reproducing the bug in the lab and fixing it. If it’s like a lot of bugs, the fix might be one line of code, or it might even be a couple of characters, but a lot of work and time went into those two characters. 
    When you throw away code and start from scratch, you are throwing away all that knowledge. All those collected bug fixes. Years of programming work. 
    The irony of course is that Joel Spolsky — who wrote that article — is a big supporter of Obama. Just illustrates Conquest’s law: everyone is reactionary about stuff that they know something about. 
    Anyway. The underlying point here is that progressives have basically been wrong on the facts in most areas of sociology (male/female differences, wealth creation, conseuqences of welfare, causes of crime, etc.). Yet their standard MO is to attack social intuition on these topics and resort to cooked statistics, which on deep inspection fall apart (the recent Fryer study touted by Brooks took all of two days to get taken apart). 
    The conclusion is that *especially* on social matters, human intuition is a pretty good guide. Not perfect — people don’t understand wealth creation, but they do understand the enervating effects of welfare — but pretty good. This is because people have evolved to have good intuitions about human societal dynamics. Physics or evolutionary theory is another matter.  
    So when an argument is advanced that is counterintuitive and structurally similar to a progressive one (“people are wrong for thinking crime is high”), it behooves us to dig deeper.

  35. Reading the thread and TGGP’s wolf link makes me wonder where the forceful and wishful projected benevolence may come from. 
    Is it because it would be “nice and cute” if the world were a lovely place? 
    If this leftist penchant is genetically/culturally ingrained as much as the religious one we are really screwed up!!!

  36. Well geecee I should cede you some territory. What you say does make inroads/corrections against the general tenor of what I said. 
    However, I just thought of a further difference. Namely, the whites in white flight knew there could never be an outbreak of intense organized lethal violence against them. That’s probably a very impactful distinction vis-a-vis the events commonly termed ethnic cleansing. Imagine if the nazis never killed or deported any jews men masse, nor carried out Krystallnacht or the rest, but merely committed disorganized, low-intensity street violence and property destruction. The jews still would have been stressed out far beyond a white dude in inner Detroit (who I definitely admit could be pretty stressed), because they would never know for sure whether the nazis might suddenly kill great numbers of them next week. The same applies to other situations commonly called ethnic cleansing.  
    So, I still won’t think the wiki-liberalpedia people were totally irrational if Moldbug were to edit the article “white flight” with a link to “ethnic cleansing” and they were to edit his link back out. It may actually have been ethnic cleansing in a certain sense, but, I would argue, not in the commonly understood sense.

  37. It is *always* the case that when a number is found to be misleading, there is an erroneous, unrepresentative, or fradulent calculation behind it. 
    Exactly right. But this point is often of more theoretical than practical use – because it is often very difficult to find the bad calculation, which is often intrinsic to the design of the experiment itself. Often (as in global climate modeling) it’s intrinsic to the entire academic department performing the experiment.  
    For example, without invalid calculations, you basically don’t have global climate modeling – for much the same reason that without invalid calculations, you don’t have earthquake prediction. Fortunately, earthquake-predicting models are quite falsifiable due to the structure of the problem. So they get falsified, and they don’t get funded. 
    Obviously, I feel many methods that are conventional in the “social sciences” today are in this category. Often the mentality behind them is the mentality of the man who looks for his keys under the streetlight – this may not be perfect, but it’s the best we can do. As a hard-liner, this cuts no ice with me. If it pretends to be a rigorous calculation and it isn’t, don’t do it.

  38. Well, the uncontested fact is that violent crime is still about 2-3 times what it was in 1960. 
    I would guess that the 50s represented a kind of sweet spot where society had the benefit both of more efficient law enforcement and fairly socially conservative mores. Now we only have the former. 
    gc is right that we have managed to clamp down on crime through: 
    1. throwing a huge portion of people, particularly but not exclusively, blacks in jail 
    2. better crime prevention technology 
    3. having less valuable stuff to steal 
    A couple more points: 
    1. I mean really, who carries cash around anymore. Does robbery pay anymore? 
    2. Stuff is now cheap. Or are you going to break into my house and steal my $50 DVD player? Or my CD collection? Puhleeease.

  39. geecee (a perhaps ersatz copy of GC) quotes Joel Spolsky (note the spelling) who says: 
    It’s harder to read code than to write it. 
    The problem is not that it’s hard to read code. It’s actually easy to read code. What is hard is understanding why the code is the way it is and how all the layers of bug fixes have caused the code to be the way it is. It is hard to understand the complete set of problems it is trying to solve. 
    In addition, just like DNA or long established social systems, there is just not enough information available to understand why all the changes were made. Even with bugzilla and subversion/perforce checkin comments, people still do not explain every detail of a checkin. 
    However, at the end of the day, these analogies fall down because the problems that software systems are trying to solve are usually several orders of magnitude less complicated than those being solved by DNA or social systems, at least in my opinion.

  40. gc is right that we have managed to clamp down on crime through: 
    1. throwing a huge portion of people, particularly but not exclusively, blacks in jail 
    2. better crime prevention technology 
    3. having less valuable stuff to steal
    I don’t doubt the power of these things, but they aren’t necessary to invoke. There seems to have been a true decline in criminal tendencies, not just a greater ability to whip into line a population that’s as or more likely to commit crime. 
    The key data here are on various forms of child abuse. Child abusers are very hard to catch and lock up, so greater incarceration rates are minor at best. Such crimes are even tougher to prevent, so prevention technology is also a minor factor at best. And to a child abuser, the rush of smacking a kid in the head, or raping them, or murdering them, or whatever, must feel as good now as before, so declining pay-offs for crime is minor. And these crimes are not punishable by death, so the death penalty can’t explain them either. 
    Here is a free copy of a brief review of these data: 
    And a free copy of the academic journal write-up: 
    The only form of abuse that hasn’t declined so much since the early 1990s is parental neglect, but that’s really separate from rape, murder, etc. 
    The zero-order approximation is that there are no such things as rape, murder, etc. — there’s just “crime.” How inclined a person is to “crime,” what fraction of the population is inclined to “crime,” how these tendencies change over time, etc. It’s a pretty good approximation. 
    So, all explanations must deal with the high level of generality in crime trends.

  41. Haloscan keeps crashing so I’ll have to break up my comment. 
    either you agree to disagree, or you say nothing, or you pop each other like Biggie and 2Pac. 
    I think you stole that line from Milton Friedman. 
    I don’t think ethnic cleansing is a good model for white flight. Ethnic cleansing is typically more concentrated in time and violent. I don’t think the Great Migration northwards was ethnic cleansing (although Kevin Carson seems to think something like that, and claims the industrialization of England was the result of similar coercion) but it seems like a more comparable example of population movement. Furthermore, while your and Jones’ (who is a rather out there Catholic anti-semite) perspective is that it was a clever ploy by Brahmins against Vaisya’s, it doesn’t seem that black criminals made any distinctions between Protestants, Jews and Catholics. Are they decaying rust belt cities conquered fiefdoms of Brahmins? Such cities only wish they could attract more folks featured in Stuff White People Like and pay Richard Florida hefty sums to find out how. A better argument might be made that gentrification & the “negro removal” side of urban renewal made New York livable for the elites at the expense of Newark, Buffalo and other low-rent areas the poor were pushed out into. Suburbanization and the rise of the sunbelt seem like major factors that were going on before the decay of the rustbelt. Black Atlanta is doing better than white Pittsburgh. I think libertarians make a good case for the sclerotic political-industrial complex in such cities (with powerful unions being big factors) choking growth over time while the more business friendly “New South” attracted it. When the elite urbanites really want to screw over the hoi polloi they use zoning to prevent a supply of affordable housing from arising. Michael Hoffman calls these “anti-growth coalitions” (and he approves of them, hating the pro-growth coalition dominated by real-estate interests) that control San Francisco and Portland. Pro-growth coalition dominated cities are more common in Texas, and also bring in more hispanic immigrants. 
    Another question might be why so many Vaisya were brought into these cities in the first place if the Brahmins wanted them out. Ed Glaeser has found that the industrialization level of cities in the 19th century is among the best predictors of its liberalism, and he thinks the reason is that the concentration of lots of different ethnicities requires a sort of tolerant concensus. I’ll also note that if we include southerners (even if Protestant) in the Vaisya camp, that makes George Edwards one. Most of these cities were politically dominated by ethnics (Fiorello LaGuardia is not a WASP name). 
    The Occidental Quarterly has a review of Slaughter of Cities here
    Regarding whether slaves were economically better off as freed men, a commenter at UR asked me if there was any evidence. A bit of googling turned up this. The result of introducting the option of exit or (in more normal economic terms) replacing monopsony with competition among employers should be obvious, as Daniel Nagy there attested to without even changing his employment. 
    The Indians who fled Idi Amin had capital. The Jews who fled pogroms had capital. The ethnic Chinese who fled communist Vietnam had capital. 
    The common narrative I hear for all three is that they arrived in their new homes with very little capital and had to start at the bottom from scratch, with expropriation being part & parcel of the cleansing. That was not at all the case for the pilgrims to suburbia and the sunbelt. Again, does white flight fall more into the pattern you gave or the migration of southern blacks and rural English to industrialized cities? 
    No, but daily drive-bys. 
    Drive-bys in white neighborhoods? Those are actions one gang takes against another gang, and west side story aside honkies weren’t into that sort of thing. There is some gain to be made from robbing whites, but sources of prey are different from sources of competition. 
    And as you may know, the ratio of black-on-white to white-on-black rapes in the US is greater than 100:1. 
    Not terribly surprising to non-leftards give the relatively low crime rate of whites and low perceived attractiveness of black women. Are there stats for hispanic-on-black rape? In The Color of Crime it says white & hispanic perpetrators are lumped together to create 900 “white” on black rapes a year. 
    Heck, even lefties like Sandra Bernard and Eldridge Cleaver admit in their more candid moments that black-on-white rape is not just random, but ethnically and politically motivated. 
    The Color of Crime had graphs on the racial composition of victims of crimes committed by blacks. Rape was the one with the “blackest” composition of victims. However, that chart did not include homicide/manslaughter despite having a category for “all violent crimes”. 
    Latinos committed nearly half of the hate attacks in the county, while blacks committed 30 percent of the hate attacks. The victims of the race violence almost always have been other blacks or Latinos. 
    I believe that. They have feuding gangs fighting for turf. They are going to target the young males who make up the competition. You don’t see them engaged in the same sort of behavior against whites because whites aren’t interested in holding corners to sling crack on.

  42. whites are the bad guys coming (“gentrification”) 
    I don’t think gentrification is comparable to ethnic cleansing, even if the result of some people moving away and others moving in is similar. Do you think it is just the same as white flight but in reverse? 
    Diamond and Lewontin’s 
    Diamond as in Jared Diamond? I didn’t think he worked with Lewontin. Did you mean to write Gould, Rose or Kamin? 
    without invalid calculations, you don’t have earthquake prediction 
    Don’t hear as much about that, could you elaborate? 
    So they get falsified, and they don’t get funded. 
    Oh, ye of little pessimism! 
    As a hard-liner, this cuts no ice with me 
    And yet you are content to rely on even less rigorous literary accounts? If I didn’t have much faith in scientific earthquake predictors I wouldn’t go to a psychic instead. 
    On a final and unrelated note, Mencius, have you read “From Dawn to Decadence”? I reach for my gun when I hear the word “culture”, so it’s definitely more your bag. Even contains shout-outs to Carlyle & Froude. Unfortunately Barzun is rather P.C for a centenarian conservative.

  43. Do you think it is just the same as white flight but in reverse? 
    They are defined differently by progressives. White flight is defined as a form of passive and structural racism by the progressives. While gentrification is defined as ‘the new colonialism‘.

  44. Child abusers are very hard to catch and lock up, so greater incarceration rates are minor at best. 
    The problem with this is that by throwing tons of bad guys in jail for whatever crime you happen to nab them for, you reduce the number of people with more general criminal tendencies. For example, U.S. drug laws aren’t really about stopping drugs, but rather function as a kind of a broad dragnet for getting undesirables off the street. Throwing a bunch of low level drug dealers or petty criminals in jail will de facto throw a lot of child abusers into jail. I guess what I am saying is that we have to look for a general factor for criminality, somewhat like that for intelligence, before we start assuming that people are generally just becoming better behaved.

  45. I guess what I am saying is that we have to look for a general factor for criminality, somewhat like that for intelligence, before we start assuming that people are generally just becoming better behaved. 
    I totally agree. Given how controversial studies of genes and IQ are, I can imagine what will happen when the genetic impact of criminality becomes better known.

  46. Oh no! Overhyped paranoia is going to kill us all!

  47. Yeah, the analogy to a general intelligence factor is right. What I’m saying is that it’s possible that the average for this factor truly decline among the population — not that it stayed the same, but that we threw a bunch of the highly criminal in jail. 
    Oh no! Overhyped paranoia is going to kill us all! 
    Well, one of Gardner’s points is that, in fact, paranoia did kill people — over a thousand — who switched from flying to driving. Even short of that, irrational paranoia results in a huge waste of resources — always someone else’s, of course, not the personal checking accounts of the self-appointed custodians. 
    Even the 2008 General Social Survey wasted an opportunity to probe a large random sample of society about something important, but chose to ask them at length about being molested by clergy members. Obviously irrational, since it’s a non-issue. But the paranoia is still alive in a lot of minds, enough to get these questions included in the survey. 

  48. What I’m saying is that it’s possible that the average for this factor truly decline among the population — not that it stayed the same, but that we threw a bunch of the highly criminal in jail. 
    It is possible. The question is what is the most likely explanation and it seems unclear at this point.

  49. So, Eloha, would you claim that white flight was no more an example of ethnic cleansing than gentrification? 
    Thursday, I wrote a bit about the service massive incarceration does for us and how we’d fare otherwise here.

  50. “I guess what I am saying is that we have to look for a general factor for criminality, somewhat like that for intelligence, before we start assuming that people are generally just becoming better behaved.” 
    Well, testosterone levels & low MAOA activity appear to make some people more at risk. 
    “Finally, Meyer-Lindenberg and his co-workers draw the lines to other studies relating MAOA variance to a highened sensitivity in low-MAOA males to aversive events (e.g. abuse) during childhood. The combination of a low-MAOA genotype with such events seem to produce abnormal regulation (through the cingulate) of the amygdala and an increased predisposition to impulsivity and violence.” 

  51. The reason for the Flynn Effect is still a matter of debate, but if part of the reason is a real increase in the average level of g in the population, one can imagine that something similar could happen to criminality. 
    In Before the Dawn, Nicholas Wade makes a good case that humanity has been taming itself for the past 50,000 years. In other words, over the long run, criminality has declined because of evolution. If it has gone up or down in the past 100 years, the effect is obviously primarily environmental. I’ll let the historians argue about which way. I’m not qualified to answer.

  52. “In Before the Dawn, Nicholas Wade makes a good case that humanity has been taming itself for the past 50,000 years. In other words, over the long run, criminality has declined because of evolution.” 
    Greg Clark, author of ‘A Farewell to Alms’ discusses the hereditability of criminality in this response to critics of his book: 
    Studies of criminality, using either twins or adoption methods, have similarly revealed a 
    similarly strong genetic connection. One classic study is that of 14,427 Danish adoptees where the 
    court convictions of the adoptees, their biological parents, and their adoptive parents are all known. 
    When both sets of parents were non-criminal, the chances of the adoptee being convicted for a 
    crime were 13.5%. When only the adoptive parent had a criminal record this chance rose very 
    slightly to 14.7%. However if only the biological parent had a criminal record the chance of the 
    adoptee having a criminal record rose much more, to 20.0%. If both sets of parents had a criminal 
    record the chance of the adoptee having such a record was 24.5%. This suggests that genetic 
    influences on the propensity to crime are much greater than environmental influences.10 
    Could the same forces identified above also cause significant genetic change over the course 
    of 20-200 generations? 
    Dogs have their origin in East Asian wolves of about 15,000 years ago, but they have acquired traits wolves do not possess. Thus dogs can read human faces and human actions in a way that wolves 
    cannot, even when wolves are socialized with people from birth as with dogs. Until recently, however, the one creature in the modern farmyard that was believed to be unchanged from Paleolithic times was man himself… 
    But evidence from the modern world establishes that economic success has a very important 
    genetic component. Putting these together we can establish that there must have been significant 
    genetic change from generation to generation in this dimension in societies like pre-industrial 
    This evidence fits well with recent claims that the rate of evolution among humans speeded 
    up in the past 10,000 years.” 

  53. I think it is also important to note that human societies have mostly been getting less and less violent over time. The rise in crime in the 20s and 30s and again up since 1960 was a reversal of this general trend.

  54. Crime may not have actually risen that much in the 20s, it may have been high all along but undercounted.

  55. How many of today’s shootings and stabbings don’t end up as murders because of modern medicine?