Why criminals are less intelligent than non-criminals

Satoshi Kanazawa's explanation:

Men with lower intelligence are less likely truly to comprehend evolutionarily novel entities. Some of these evolutionarily novel entities are alternative means to resource acquisition and accumulation they could pursue instead of evolutionarily familiar means which are now classified as criminal in civilized societies. Other evolutionarily novel entities they are less likely truly to comprehend are means that law enforcement agencies employ to detect and capture criminals. The Hypothesis therefore offers one possible explanation for the negative association between intelligence and criminality.

I've argued before that black-white differences in criminality are driven primarily by differences in intelligence -- not testosterone. I've also previously observed that black-white multiples for brawling and raping are low compared to those for crimes like car theft, drug dealing, and mugging, consistent the idea that a lack of intelligence leading to ill-considered efforts at resource acquisition rather than greater innate aggressiveness underlies the difference.

At the same time, the Hypothesis also offers a novel hypothesis with regard to intelligence and criminality. As I mention above, while formal third-party enforcement of norms is evolutionarily novel, second-party enforcement and informal third-party enforcement are evolutionarily familiar. Thus the Hypothesis would predict that the difference in intelligence between criminals and noncriminals will disappear in situations where formal third-party enforcement of norms is weak or absent, and criminal behavior is controlled largely via second-party enforcement, such as situations of prolonged anarchy and statelessness, in fact, any situation that resembles the ancestral environment. Paradoxically, the Hypothesis would predict that less intelligent men will commit fewer crimes if the police disappeared, although more intelligent men may commit more crimes then.

Murders and rapes of whites by blacks in America took off during the 20th century (relative to the reverse). Kanazawa's hypothesis suggests the evaporation of the threat of extrajudicial punishment and increased state protections for blacks may be responsible for this state of affairs. Update: From Crime and Human Nature:

it is possible that differences in sanctions have had an impact on the extent to which members of different races are inclined to handle aggression by resort to assaults and homicide. Suppose that over the generations, law enforcement officials ignored or treated leniently the most common forms of serious black crime (dangerous or fatal assaults on blacks by blacks), reserving the full force of their sanctions for the much rarer cases of assaults by blacks on whites. In 1937 John Dollard wrote of the harmful effects of the double standard of justice then evident in a small southern town and which probably persisted for many decades throughout the South and in some parts of the North: "Negro crime" was less serious than "white crime," a view sometimes defended on grounds of a high-minded indulgence of "Negro ways." As Dollard wrote:

The formal machinery of the law takes care of the Negroes' grievances much less adequately than that of the whites, and to a much higher degree the Negro is compelled to make and enforce his own law with other Negroes. . . . The result is that the individual Negro is, to a considerable degree, outside the protection of the white law, and must shift for himself. This leads to the frontier psychology. . . . [This] condoning of Negro violence . . . may be indulgent in the case of any given Negro, but its effect on the Negro group as a whole is dangerous and destructive.

[pp. 475-476]

According to the thinking of 1930s sociologist Dollard, shifting the burden of law enforcement among blacks more to the state should have lowered their crime rates. Instead, in keeping with Kanazawa's prediction:

the homicide fatality rate among black males nearly doubled between the early 1960s and 1973, increasing from 34.3 to 65.8 per 100,000; the change among whites was much smaller. [. . .] Roger Lane estimates that in Philadelphia the black homicide rate was about three times greater than that of whites during the nineteenth century but had become about twelve times larger by the middle of the twentieth century and nearly twenty times larger by the 1970s. [p. 472]