Home > Psychology, Science, Skepticism > Evidence-based Masturbation (or, The Science of Porn)

Evidence-based Masturbation (or, The Science of Porn)

This is the blog version of a talk I gave on August 13, 2011, at Skeptics on the Fringe at the Edinburgh Festival. 

‘Porn’, to quote journalism professor and anti-pornography activist Robert Jensen, ‘is what the end of the world looks like’. I must admit that, until I read this statement (from this book), I rather thought that the apocalypse would be more of the meteor-striking, sun-exploding, fiery-death kind. Instead, Jensen seems to think armageddon will come in an avalanche of breast enhancements, fake grunting, and peroxide.

Get your Ragnaröks off! 'Porn is the end of the world'.

To be fair to Jensen, what I think he meant was that porn is the end of the world because it epitomises uncaring, unfeeling interactions between men and women, and will eventually erase loving relationships from our society. Or something like that. He’s certainly not alone in this viewpoint – I recently read Pornland, by sociology professor Gail Dines, an attempt at an excoriating critique of the porn industry, replete with descriptions of some of the more insalubrious websites Dines has discovered on late-night Google sessions, and tragic anecdotes of people’s lives wrecked by their involvement with, and consumption of, pornography.

I don’t doubt for a second that these anecdotes are mostly true, though one has to wonder if the ‘people who came up to me after one of my lectures’ really worded their stories in quite the perfect, argument-supporting way Dines describes. There are, of course, disgusting and degrading pornographic websites out there, and I’m certain that a significant number of women  - and men –  are exploited and abused by pornographers each year (it would be nice to see some proper studies into this, however). These aren’t really arguments against pornography, though, any more than saying ‘some prostitutes are abused by clients’ is an argument against prostitution. They’re arguments against the way we currently police and regulate the pornography industry (if there is indeed such a thing), and they’re more of legal than scientific interest. As for people who just use porn, the stories Dines provides are next-to-useless, as they aren’t from a large, representative sample, or properly recorded, or linked to any demographic information. There’s a devastating review here in the journal Violence Against Women, which shows Pornland for the unscholarly hatchet job it is.

While Dines isn’t afraid to compare the porn industry to Nazi Germany (p. 65), or claim – without evidence – that porn users are slowly turning, werewolf-like, into pedophiles (the penultimate chapter in the book, and a lecture you can see here), she does make a big deal of rejecting the simplistic argument that ‘porn causes rape’. This claim has been around for a long time (radical feminist Robin Morgan once famously stated that ‘pornography is the theory and rape is the practice’), and actually, though she distances herself from it, it’s near-impossible to know what Dines is talking about other than ‘porn causes rape’. For instance, in the video I linked above, she can be seen stating that ‘porn gives men a sense of entitlement to women’s bodies. What is rape, if not a sense of entitlement to women’s bodies?’ Thus, her opinions don’t seem to have moved on from the old Morgan dictum. We’ll see below how well that argument holds up when one looks at the evidence.

And so, skipping daintily over three aspects that have bedevilled previous discussions – a rigorous definition of pornography, boring debates about the difference between ‘pornography’ and ‘erotica’, and the impossible task of working out how much pornography is worth – I’ll make one important point before getting into the science: I’m not going to discuss moral arguments. If you think that porn is by definition degrading, that it’s never okay to have sex on camera, or that getting sex involved with money is always exploitative, that’s fine. There’s nothing I can do for you here. Of course, we should use the data to inform our morals, but there will always be those who are too disgusted (and it’s very clear from recent psychology experiments that disgust influences morals) by the whole idea of pornography to have a science-based argument about it.

There are, broadly speaking, four different kinds of studies which have looked into the effects of pornography. Here’s a mini-summary of each:

1) Experimental studies

In this type of study, researchers show willing participants porn, and quiz them on their attitudes afterwards. These appear to be quite contradictory, and sometimes strangely reported. For instance, Malamuth and Ceniti claim in the abstract to their 1986 study that watching violent and nonviolent pornography increased aggression in their sample of college students. However, no such result is to be found in the actual paper itself. How odd! Other studies have found a wide variety of effects, and are reviewed by Christopher Ferguson here.

There is, of course, a huge flaw in experimental research of this kind. Think about it: when was the last time you used pornography that was forced on you by a psychologist (by which I mean it wasn’t your choice of movie), all the while in the knowledge that you’d be quizzed afterwards? These problems of ecological validity are pretty devastating when it comes to studies of an everyday activity like using porn. In addition, it’s not at all clear that the ‘aggression’ and ‘likelihood to rape’ measures in these studies actually relate to real-world aggression or likelihood of rape (this is compounded by the fact that a wide variety of measures were used, reducing our ability to compare apples to apples).

Ferguson concludes that porn’s ‘…effects appear negligible, temporary, and difficult to generalise to the real world’. (p. 4). From a good look at the literature, it’s hard to disagree.

2) Survey studies

‘Aha!’ Thought I, as I started looking for non-experimental survey-type studies into pornography. ‘Here’s a recent meta-analysis, where some authors have collected all the relevant studies, and done my job for me!’. And indeed, there is a recent meta-analysis on this topic by Hald, Malamuth, and Yuen. Looking at nine studies which asked people ‘how much porn do you watch?’ and then ‘what are your attitudes to women?’, these authors found that there was a correlation between porn use and attitudes supporting violence against women. The correlation is pretty small for non-violent pornography (r = .13) and somewhat larger for violent pornography (r = .24). Solid, right? Wrong. First off, as noted above, only nine studies were included. Surely more survey studies have been done? Well, turns out they have, and there are at least two glaring omissions from this meta-analysis.

The first one I found was by Garos and colleagues from 2004 (this was mentioned in a recent Scientific American article). In their study, they asked a couple of hundred students about which kinds of porn they used, and measured their attitudes on a wide variety of sexism scales. As far as negative sexism goes (e.g. ‘I hate women’), they found that the more porn people used, the less sexist they were. There was, however, an association with positive (or ‘benevolent’) sexism (e.g. ‘women need to be protected’) and porn use. This type of sexism can, of course, be very damaging (imagine failing to get a job because the interviewer thinks you aren’t up to it as you’re a woman), but it’s not the same kind of hatred that activists like Dines predict porn would engender.

Secondly, McKee managed to contact 1023 users of pornography in his large-scale study (much larger than any of the studies in the Hald et al. meta-analysis), by post and online. He failed to find any association between the amount or type of porn people watched and their attitudes towards women (the negative attitudes to women were associated with more predictable things – being old, voting for a right-wing political party, etc). A big advantage of this study is it breaks out of the usual ‘convenience sampling’ problem in social science research, and doesn’t just focus on students (the studies in the meta-analysis tend to).

Hald and colleagues may have not included these studies in their meta-analysis because they included information from women. They note that most of the problem with sexism comes from men, so studies which asked women about their attitudes towards their own sex would be muddying the waters. I’m really not sure this is a great reason for not including these studies, especially considering that the results for males are easily separable from the results for females (and even if they aren’t, you’re doing a meta-analysis! Email the study authors and ask for just the male data!).

Now, the inclusion of these two studies might not wipe out the correlation between porn and negative attitudes from the meta-analysis entirely, but it would certainly weaken it. In all, we don’t appear to have found much of an effect of porn here, either.

3) Criminal studies

Ted Bundy had a nasty little secret. Well, quite a few; he was a serial killer, rapist, and necrophiliac who murdered over 30 people. But he was also addicted to porn. You can find unpleasant, homophobic, sterile little Christian websites such as this one who quote interviews with Bundy, and more or less claim ‘if you don’t stop masturbating to porn, you’ll end up just like him!’. Since we all (well, maybe not Gail Dines) know that anecdotes aren’t data, it’s better to take a look at the empirical research into sex offenders and their use of sexually explicit materials.

As discussed here (p. 4), the majority of studies into sex offenders and porn use are counter-intuitive in their results. In general, they find that sex offenders use just as much, or sometimes less, porn than the average person. This review briefly discusses evidence that sex offenders may be more likely to have had a repressive religious upbringing than non-sex-offenders.

Perhaps, then, it’s time to move beyond unsupported situational explanations of sex crimes. If sex offenders and the average citizen use the same amount of pornography, but only one sexually assaults someone, there must be something else going on – either innate to that person, or in their background. The lack of effect of pornography on offending does fit into the general trend of evidence of negligible media effects on violence – see Ferguson’s publications for loads of really great reviews in this area. There are definite echoes in the porn debate of the purported videogame-violence link (which as far as I can see, either doesn’t exist or is teeny-tiny).

4) Aggregate studies

These are, I think, the most interesting – and the most consistent – studies in the area. Starting with Berl Kutchinsky’s investigations of what happened in countries like Denmark when porn was legalised, we have a raft of studies which look at country- or state-level data, and try to discern the relationship between pornography consumption (by the way, don’t you just love the idea of someone consuming pornography? Delicious!) and the rate of sexual assaults and rapes. The most recent studies have been carried out by Milton Diamond at the University of Hawaii, who started off by looking at Japan in 1999. We all know that Japanese pornography is particularly fetishistic, and much of it could clearly be seen as degrading to women. So what happened as the availability of pornography rocketed in the latter half of the 20th Century? Take a look at the paper: rape and sexual assault rates plummeted in the same time period.

Hardcore porn availability and the US rape rate, from Ferguson (2009), Journal of Aggressive Behaviour.

More recently, Diamond studied an utterly different culture, that of the Czech Republic. Back when it was Czechoslovakia and run by those killjoy communists (the Party ruled from 1948-1989), pornography was completely prohibited. In the paper, Diamond and colleagues document how the crime rates changed after the fall of the regime, and the associated upswing in porn availability (a lovely little natural experiment, one could say). Check out the graphs: in the period of instability after the fall of communism, crime in general increased. But while non-sexual crimes such as murder continued to climb upwards, rape rates returned to similar levels as those under communism after a few years. If you split up the murder data, those murders with something to do with sex in general decreased, while those unrelated to sex increased massively. It’s worth pointing out that these sorts of results are found in a wide variety of other societies, too – see here for a review, and also the graph to the left for a summary of what happened to the rape rate in the US as porn availability increased – spoiler: it’s the same pattern again.

These studies are, of course, vulnerable to a variety of criticisms. First off, correlation doesn’t prove causation. Anybody claiming that these decreases in sexual crimes are caused by pornography is incorrect, because we don’t know that the crime rates wouldn’t have decreased faster if porn hadn’t been in the picture. But what these studies do show pretty convincingly is that there’s no linear association between porn and sex crimes. It seems we’ll have to look elsewhere to explain sex offences.

Another criticism is this: rape is notoriously under-reported, and conviction rates for it are extremely low, so how do we know these rape rates are accurate? Well, There’s a genuine debate over whether people would be more or less likely to report rape in a society with more porn availability. However, in Japan study cited above, Diamond provides evidence suggesting that, as Japanese society became more liberal (with the associated increased in porn), more rape counselling centres and public awareness campaigns were set up, raising consciousness of the problem. It stands to reason that other countries have had similar changes, though this remains an important criticism of the aggregate studies. To close, though, it’s worth remembering that this increase in porn and decrease in sex crimes has been found in pretty much every country/state ever studied; the consistency of the data is impressive, and seems to go beyond any culture-specific changes in rape victim support or attitudes towards rapists.

So that’s it for our tour of the Science of Porn. What can we conclude? Well, while there’s a wide variety of not-so-informed opinion on porn, there is rather a lot of evidence. Some of it is contradictory, but on taking a step back we realise just how difficult it has been for researchers to show a connection between porn use and, well, any negative outcomes. If effects have been found, they’re generally weak, and there are methodological issues with many of the studies. On the basis of the above evidence, then, you might think that we should go easy on porn. I’d agree with you, to some extent. But of course, this isn’t the end of the story; a wide variety of other questions remain relatively unanswered:

  • Perhaps pornography decreases sexual satisfaction? Well, according to the small amount of research I could find, probably not.
  • Perhaps it makes women view themselves differently? I couldn’t find any studies directly looking at this, but I’d caution that media effects on female self-perception aren’t necessarily as strong as one might think.
  • Perhaps, as suggested in Pornland, porn perpetuates racist stereotypes, encouraging us to view, say, Asian women as subservient and Black men as animalistic? I think Dines may have a point here, but there’s no hard evidence of the effects on porn consumers. Possibly, however, this reflects the ‘underground’ nature of porn. If these stereotypes were in regular films or TV, there’d be dozens of complaints to the producers. Another argument, maybe, that better regulation of the porn industry might help.
  • Perhaps porn actually has positive effects, such as being a kind of substitute sex education? Well, there isn’t really any direct research on this, either, but evidence so far isn’t impressive.
  • Perhaps newer, more rough and violent ‘gonzo’ porn might have effects that the studies discussed above didn’t pick up? Only time will tell if this is the case, but as described above, other studies of violent porn vs. non-violent porn haven’t shown a great deal of difference in effect, and it isn’t all that clear that mainstream porn is becoming more violent anyway.
  • Lastly, it’s worth pointing out that arguments that porn is ‘taking over the internet’ have been addressed here, and shown to be false.

But what does any of this matter? In Pornland, Dines states: ‘The studies provide some indication of effects, but what I find most compelling are the stories I hear from women who have been raped by men who use porn’ (p. 95). Clearly, Dines’s mind is made up. And yet, she must know that anecdotes are unconvincing to anyone with a bit of scientific or sociological training.

In the end, it’s a bit like the whole ‘Intelligent Design’ thing. Just like the Discovery Institute, anti-porn campaigners have money (from book sales, if nothing else), and they must know in their heart of hearts that scientific studies are the most convincing way to change people’s minds about topics like this. So why aren’t they funding any research? Why aren’t they suggesting any particular research designs? It’s deeply patronising to your audience to imagine they’ll be satisfied with anecdote after anecdote and little by way of peer-reviewed hard data.

…argh, I said it. ‘Hard’ data. I’ve been constantly fighting against adding puerile sex-and-masturbation innuendo to this post, and now I may have reached the limit of my inhibitory abilities. To sum up: it’s important to discuss the effects porn might be having on us, but it’s equally important to stick to the empirical data that’s been gathered. This has been, and remains, a serious problem for anti-porn campaigners. Until more solid evidence appears, we can only conclude that porn really isn’t the end of the world.

  1. August 18, 2011 at 4:31 pm

    this is brilliant and makes great sense as agree that porn is nothing but a human collective activity that has multi-facet effects and functions

  2. August 25, 2011 at 8:44 pm

    I know you’re probably an atheist, but: PREACH!!

  3. Anon
    August 30, 2011 at 4:05 am

    With increase in porn and “liberalism” in women alongwith STD becoming a commonality in modern “sluts”, someone has to be mentally retarded to rape an already “willing” woman. With women beating men in promiscuity and cheating, raping modern day skanks is not an option available to sex offenders anymore. So, since you can’t rape someone who came willingly, incedents of rape had to go down ? Didn’t it ? A woman can’t technically/legally rape a man as a man is always willing. Isn’t this the reason provided in that case ?

    • Stuart Ritchie
      August 30, 2011 at 10:39 am

      Eh?

    • hauntfox
      September 14, 2011 at 2:48 pm

      So basically what you’re saying is that rape is ALWAYS about sex and NEVER about control; that most women today are CONSTANTLY willing (and maybe even looking) to have sexual encounters with men they may or may not know (if I remember correctly, rapes are typically committed by someone the victim knows); that the ONLY demographic of woman a sex offender would want to assault (sorry, their only “option available”…. wtf?) is an STD-laden cesspool; and that EVERY man is ALWAYS willing to have sex with ANY woman?
      (I would also love to know where you got the data for women being more promiscuous and adultering; I’m not flat-out saying that you’re wrong, but your misogynistic tone makes me inclined to roll my eyes and dismiss that comment altogether)

      So, all of this because women are more “liberal” and watch porn more than in the past?
      Your arrogance, stupidity, and antiquated views of how women should act are mind-boggling.

  4. September 25, 2011 at 12:44 am

    I thought this was an interesting blog. I think something that taps into some of our basic human drives and can make a lot of money is weirdly under researched. And it’s changing with increased access to creating and distributing digital media. Definitely needs some more good questions to open the debate out.

  5. Jim
    January 17, 2012 at 9:38 am

    Thanks for this Stuart. I’ve spent many an hour searching the web for some real evidence-supported research on pornography, and this is the best I could find (or at least, the best I could understand).

    I think books like Pornland, and religious ‘cures’, and pseudoscience are so rife in this area because pornography feels addictive. For someone like me, who would like to stop using porn (my girlfriend hates it and it makes me feel grubby), but can’t, science-y sounding cures (or even diagnoses) are attractive – we all love a quick fix, right?

    I wonder if you could comment on my situation. There are plenty of people in the same boat as me, who use porn, but don’t really want to, and feel incapable of stopping. How much research has been done here? Where should we be looking?

    Have you seen this, by the way? It sounds plausible, but it’s on YouTube, so I’m thinking it probably isn’t…. http://m.youtube.com/#/watch?desktop_uri=%2Fwatch%3Fv%3DTKDFsLi2oBk&v=TKDFsLi2oBk&gl=GB

    • Stuart Ritchie
      January 17, 2012 at 9:54 am

      Hi Jim, thanks for your comment – glad you enjoyed the blog. Obviously I’m not qualified to comment on your situation, but I think you’re right to be skeptical of any claimed cure for porn ‘addiction’. In fact, I know that ‘porn addiction’ itself isn’t recognised as a diagnostic entity (any more than ‘tv addiction’ or ‘video game addiction’). Jesse Fischer has an interesting post on the subject here: http://sexademic.wordpress.com/2011/05/08/explaining-porn-watching-with-science/

      Unfortunately I can’t get that YouTube link to work – do you have the basic (non-mobile) link to the video?

  1. August 26, 2011 at 1:35 pm
  2. September 12, 2011 at 4:23 pm
  3. September 24, 2011 at 10:44 pm
  4. October 22, 2011 at 1:11 am
  5. April 30, 2012 at 7:49 pm
  6. September 18, 2012 at 10:15 pm
  7. September 21, 2012 at 4:53 pm

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

Follow

Get every new post delivered to your Inbox.

%d bloggers like this: