Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

How large is the issue? I don't know. There don't seem to be much in the way of good crime stats. A quick google search suggests rape has risen 15x in Sweden since they started letting in large numbers of immigrants. http://www.dailymail.co.uk/news/article-3317978/Torn-apart-o...

The issue is being lied about because rather than giving out stats to answer the question of how much of that 15x is due to immigrants, the Swedish authorities decided to stop tracking ethnicity of perpetrators instead. Similarly, witness the attempts by the mayor of Cologne to blame the victims and avoid disclosing the identity of the attackers. (A quick google search suggests similar incidents happened in a Stockholm music festival and other places.)

The fact is that left wing Euro politicians control access to the data and are locking it down. They also make claims like "overestimate the actual magnitude of the threat" - but if the data supported that claim, wouldn't they release it rather than hide it?

I'm not proposing any solutions. I'm not even claiming to have a solid grasp of the statistics (unlike, say, a similar discussion of US crime). If I'm wrong and Muslim immigrants to Europe actually commit crimes at the base rate, by all means post stats proving me wrong. I'm only saying that it looks like there is a real problem, and it looks like it's being hidden from us.

Under such circumstances, people turning to the racist right is a pretty natural thing to do. At least these guys aren't liars, and see our problems, right?

(Again, that last sentence is not my view. A quick search of my comment history will suggest that I'm pluralistic to a degree that only an extreme capitalist can be.)



> dailymail.co.uk

If the Daily Mail published an article saying that snow is white, I would check the next winter in case I've been wrong all these years.

In the present case, even if it weren't an issue perfectly calibrated to suit the Daily Mail's preferred varieties of shit-stirring, that 15x figure should cause doubt. Those immigrants would have to be impossibly rapey, or else the previous inhabitants (even though some of them were presumably immigrants) would have to be startlingly un-rapey.

A more likely hypothesis is that something else changed; perhaps the Swedish authorities changed where they draw the boundaries between rape and other sexual assaults? Or perhaps some procedural change has made Swedes more willing to make accusations of rape? I can think of several other hypotheses, all of them more likely than that immigration made the rate of rapes go up by 15x.

So, let's look a bit further. First of all: this comparison is between 1975 and 2014. Forty years is a long time, and many things have happened in Sweden (and everywhere else) since 1975. Why blame it on immigration? Well, because someone wants to blame it on immigration, so far as I can tell.

Well. Did Sweden change its definition of rape, or its treatment of people who report rape, or its thoroughness in collecting statistics, or anything of the sort, since 1975? Yes, apparently they did. For instance, near the start of https://en.wikipedia.org/wiki/Rape_in_Sweden you will find this: "Sweden reformed its sex crime legislation and made the legal definition of rape much wider in 2005, which largely explains a significant increase in the number of reported rapes in the ten-year period of 2004-2013."

And, a little further down, this: "Sweden also applies a system of expansive offence counts. Other countries may employ more restrictive methods of counting. The Swedish police registers one offence for each person raped, and if one and the same person has been raped on a number of occasions, one offence is counted for each occasion that can be specified. For example, if a woman says she has been raped by her husband every day during a year, the Swedish police may record more than 300 cases of rape."

I know you've got a data-science background. Put on your analyst's hat for a moment and think about that claim of a 15x increase. Do you really find it credible? At all?

There may indeed be lies being told about immigrants and crime. I would not advise you to assume that they're all coming from the same side.


There are many hypothesis. Too bad the data which could allow us to evaluate them (namely the # of crimes broken down by ethnicity) is deliberately hidden. That's why I'm suspicious of the establishment's claims, and why I expect many Europeans are as well.

I don't find a surge in crime relating to changing demographics implausible. In the US we have widely varying (read: 8-10x) crime rates between ethnic groups, so if our ethnic composition changed we could get significantly more crime as a result. Perhaps Sweden has similar effects. Or perhaps not; but if the data proved otherwise, why would the establishment hide it?


Even if those nasty scary brown-skinned immigrants have 10x the crime rate of the rest of the population and make up 20% of the population, that couldn't do more than about double the crime rate.

(Muslim immigrants do not in fact make up 20% of the population of Sweden.)


> left wing Euro politicians control access to the data and are locking it down.

In the same sense as left-wing and right-wing politicians everywhere control access to crime data and are locking it down.

> if the data supported that claim, wouldn't they release it rather than hide it?

Let me describe for you a hypothetical world. I don't know how closely it resembles the real one. In this hypothetical world, Muslim immigrants are 2x more likely than other people to commit violent attacks such as rape, make up a small fraction of the population, and are widely hated and feared; violent attacks on Muslim immigrants are 3x more common than violent attacks by Muslim immigrants; the fear many people feel of violence from Muslim immigrants is -- if, e.g., you calibrate it against other kinds of violence they fear -- far out of proportion to their actual risk of being attacked by Muslim immigrants. So, in particular, if the public knew and understood and internalized that 2x figure, they should actually be reassured by it because the risk they face is very small; but, just as in the real world, the public in our hypothetical world don't always respond with perfect rationality.

Now imagine you are a politician (of the left or the right, I don't care), you're aware of these numbers (or have guessed at them), and you are trying to decide whether published crime statistics should break out Muslim immigrants as a separate category of offender.

I suggest that you are likely to think as follows: "If we do this, every right-wing anti-Muslim anti-immigrant group in the country is going to seize on that 2x figure and make hay with it. It will widely be taken as confirming fears that Muslim immigrants are a terrible threat to public safety. Violent attacks on Muslims will go through the roof. Lots more people will be hurt and killed, and any sort of integration will become more difficult. Which will probably lead to more crime by Muslim immigrants in the future, too. Even if by some magical process releasing those numbers made rape by Muslim immigrants completely go away, it would still lead to a lot more harm than good."

I'm not saying you'd be right to think that. Maybe actually "better more data than less data" is a universal principle that trumps everything else. Maybe releasing the numbers wouldn't really have such bad effects. Maybe it actually would somehow lead to a big decrease in crimes committed by immigrants. I don't know. But the point is that I just sketched a perfectly plausible process by which politicians might decide it's better not to publish potentially inflammatory numbers, even if those numbers indicate that anti-immigrant politicians and journalists exaggerate the actual magnitude of the threat.

> If I'm wrong and Muslim immigrants to Europe actually commit crimes at the base rate

I doubt that. I bet Muslim immigrants to Europe are, e.g., systematically somewhat poorer than the population as a whole, which would probably mean more violent crime and (e.g.) less embezzlement even if there were no cultural differences at all. I bet they drink less alcohol, which would probably mean less of some kinds of violent crime and e.g. far less DUI. Etc. I'd be astonished if there weren't all kinds of differences in the rates at which Muslim immigrants and (say) Christian multigeneration natives commit all sorts of crimes.

> it looks like there is a real problem, and it looks like it's being hidden from us.

Suppose someone has a theory that financial misconduct like fraud and embezzlement and insider trading are much more widespread among the rich than in the general population. If they point out -- correctly, so far as I know -- that official crime statistics don't break out the richest 1% of the population when reporting rates of these crimes, are they justified in saying that "there is a real problem and it's being hidden from us"?




Consider applying for YC's Summer 2026 batch! Applications are open till May 4

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: