Porn, rape, Hamza Tzortzis and statistical analysis

Hamza Tzortzis has recently published a response in his questions and answers section entitled "Do Liberal Societies Facilitiate Rape via the Legalisation of Pornography?" In it he attempts to link liberal society with consumption of pornography, consumption of pornography with rape, and offers Islamic society as a corrective solution.

As a member of a liberal society I read the article with much interest. If he's right it's the most important article on the internet. as head of research with iERA he is "responsible for ensuring all of the research projects are ... of the highest standard". His team and target market are both based in liberal countries as of course are many of the world's Muslims, so it would be best if he tackled the subject with tact and care to avoid the offence caused by inaccurate research.

Let's see if he was of a similar mindset:

He states that "there is ample of [sic] evidence to show how pornography that is legal in many liberal societies causes rape" and that "empirical and social research evidence is overwhelming in affirming that pornography is a major facilitating and causal factor [of rape].", but avoids listing any empirical or social research.

Instead, he makes reference to one book, and claims its unnamed journals and research conclude that rapists implicate pornography in their crimes. There is a substantial leap between the conclusion that rapists use pornography and Hamza's position that legalised pornography causes rape.

He also claims that studies show 66% of rapists claimed they were incited by pornography. Again, the study is not listed, but the author of his sole source gives an entirely different answer - 33%, citing Einsiedel, 1986, p. 62, reference here. (Update September 20th, 2012: Hamza has corrected this figure.)

Finally, he claims that 30% of college students would rape if they would get away with it. I've tracked this to Malamuth, Haber, and Feshbach, 1980, p. 124, referenced here. Troubling as this statistic may be the study in no way involved pornography and is thus irrelevant, presumably added for shock value.

So far I'm disappointed by the standard of research, but this doesn't necessarily mean Hamza is wrong. Let's examine his premises and see if we can decide for ourselves:

Premise 1: Islamic societies prohibit pornography.

I won't argue with this one.

Premise 2: Liberal societies consume more pornography than Islamic societies.

This is worthy of research. Essentially, is the prohibition of pornography in predominantly Muslim countries effective? Luckily with Google Insights(tm), we can take a peek at the popularity of search terms by country, and use this as a rough guide to consumption of online pornography. (Edit  - learn more about how the data is gathered here.)

But is it fair to compare only online searches? After some searching I managed to find stats on the breakdown of the American porn business:

Adult Video	$500 million to $1.8 billion
Internet	$1 billion
Pay-Per-View	$128 million
Magazines	$1 billion

(Taken from How Big is Porn? - Forbes.)
It seems reasonable to assume that all Internet sales are Internet based, but beyond that the data is murky. Many DVDs and magazines are purchased online, some may only have heard about a specific website through a retail store. There is the limiting factor of having access to a credit card to consider.

There is one area we can safely compare. Pornography involving bestiality, child sex and forced sex is prohibited anywhere that has a functioning government. By being illegal everywhere we have safe grounds for comparison of appetite.

Searches for horse sex since 2004:

1st place	Pakistan	97% Muslim.
2nd place	Ethiopia	34% Muslim.
3rd place	Bangladesh	89.5% Muslim.
4th place	Nepal	4.4% Muslim.
5th place	India	13.4% Muslim.
6th place	Libya	95%+ Muslim.
7th place	Albania	80% Muslim.
8th place	Yemen	99% Muslim.
9th place	Sudan	97% Muslim.
10th place	Kenya	11.2% Muslim.

Searches for rape sex since 2004:

1st place	Pakistan	97% Muslim.
2nd place	India	13.4% Muslim.
3rd place	Bangladesh	89.5% Muslim.
5th place	Nepal	4.4% Muslim.
6th place	Fiji	7% Muslim.
7th place	Mauritius	16.6% Muslim.
8th place	Malaysia	61.3% Muslim.
9th place	Syria	87% Muslim.
10th place	Phillippines	5-10% Muslim.

Searches for child sex since 2004:

1st place	Pakistan	97% Muslim.
2nd place	Bangladesh	89.5% Muslim.
3rd place	Sri Lanka	7.5% Muslim.
4th place	Nepal	4.4% Muslim.
5th place	India	13.4% Muslim.
6th place	Syria	91% Muslim.
7th place	Phillippines	5-10% Muslim.
8th place	Kenya	11.2% Muslim.
9th place	Nigeria	47% Muslim.
10th place	Indonesia	86.1% Muslim.

What I'd like you to take from these stats is that liberal countries are noted by their absence and that being in a predominantly Muslim country does not change your odds of finding troubling images in a used laptop. Turkey is a predominantly Muslim country that I consider liberal. I've done business in Istanbul; it's a great place to visit. They score roughly the same as the USA and European countries in all three categories. Regardless of religion you're less likely to consume extreme pornography if you live in a liberal country.

Premise 3: There is a link between pornography and rape.

It can be hard to examine this issue without emotion. I'm sure you share my sense of abhorrence at this crime of violence, and Hamza's use of a quote from a rape survivor condemning pornography is designed to elicit an emotional response. But let's try and examine this important issue as best we can.

This is a rather good study on the increased availability of pornography in Japan and their decrease in sexual assaults. It summarises similar studies in Denmark, former West Germany and the USA. It's really worth reading in full, but here are a few excerpts:

On Denmark and former West Germany:

...in Denmark and West Germany the most dramatic categories of sex crime to show a decrease [following an increase in the availability of pornography] were rapes and other sex crimes against and by juveniles. Between 1972 and 1980 the total number of sex crimes known to the police in the Federal Republic of Germany decreased by 11 percent; during the same period the total number of all crimes reported increased by 50 percent. Sex offenses against minors (those under 14 years of age) had a similarly slight decrease of about 10 percent during this period. For those victims under six years of age, however, the numbers decreased from 1,421 cases in 1972 to 579 in 1980, a decrease of more than 50 percent (Kutchinsky, 1985b; pp. 319).

On Japan:

"The most dramatic decrease in sex crimes was seen when attention was focused on the number and age of rapists and victims among younger groups (Table 2). We hypothesized that the increase in pornography, without age restriction and in comics, if it had any detrimental effect, would most negatively influence younger individuals. Just the opposite occurred. The number of juvenile offenders dramatically dropped every period reviewed from 1,803 perpetrators in 1972 to a low of 264 in 1995; a drop of some 85% (Table 1). The number of victims also decreased particularly among the females younger than 13 (Table 2). In 1972, 8.3% of the victims were younger than 13. In 1995 the percentage of victims younger than 13 years of age dropped to 4.0%."

On a link between nonliberal environments and rape:

"Most frequently, as it was found in the 1960s before the influx of sexually explicit materials in the United States, those who committed sex crimes typically had less exposure to SEM in their background than others and the offenders generally were individuals usually deeply religious and socially and politically conservative (Gebhard, Gagnon, Pomeroy, & Christenson, 1965). Since then, most researchers have found similarly. The upbringing of sex offenders was usually sexually repressive, often they had an overtly religious background and held rigid conservative attitudes toward sexuality (Conyers & Harvey, 1996; Dougher, 1988); their upbringing had usually been ritualistically moralistic and conservative rather than permissive. During adolescence and adulthood, sex offenders were generally found not to have used erotic or pornographic materials any more than any other groups of individuals or even less so (Goldstein & Kant, 1973, Propper, 1972). Walker (1970) reported that sex criminals were several years older than noncriminals before they first saw pictures of intercourse."

Ultimately, Hamza's paper lacks depth of research, is inaccurate, makes false claims against his fellow Muslims and fellow members of liberal society while failing to address well-known counterarguments. I suspect this blog post is the product of greater time and effort. Accusing an individual of rape is a serious matter. Accusing an entire society of facilitating rape is more serious still. It's unclear why he didn't make more of an effort.

Addendum: Part 2 is now available.