undefined | Better HN

0 pointsgiantg24y ago0 comments

"It's not that population density causes murder, but it affords more opportunity."

I still don't think this opportunity explains the higher rate, at least not a 3-5x difference. Opportunity is only one small part. I would think that socioeconomic factors are the bigger issue, and that some predictors or correlated metrics might be more common in cities.

I get that someone in a denser area is going to come into contact with potential situations more often. But when we're talking about decades of opportunity to be exposed to such a scenario, I would expect the difference to be negligible as the exposure becomes an eventual given.

The claim that someone in the city knows more people is a little suspect. You can live in a small town and know hundreds or thousands of the people in the small town(s) near you. Everyone knows everyone, being in your neighbor's business, and can lead to a lot of gossip, opportunity/drama.

0 comments

3 comments · 1 top-level

cbsmith4y ago· 2 in thread

> I still don't think this opportunity explains the higher rate, at least not a 3-5x difference.

I didn't say that it explains the data at all. I'm just saying it's the statistical outcome one would expect given no other factors.

> I would think that socioeconomic factors are the bigger issue, and that some predictors or correlated metrics might be more common in cities.

Yes. That's entirely the point.

> I would expect the difference to be negligible as the exposure becomes an eventual given.

I wouldn't expect it to be negligible at all. You're more likely to have more frequent contact and with a much broader selection of people the over a long period of time, creating something akin to a network effect. On a given day, someone in rural Wyoming might come in contact with far more people than I'd come in contact with in my day. However, over longer periods of time, the average number of interactions and the aggregate number of people they interact with is far more likely to be dwarfed by mine. That significantly increases the chance for interpersonal crimes of all sorts, including murder.

> You can live in a small town and know hundreds or thousands of the people in the small town(s) near you. Everyone knows everyone, being in your neighbor's business, and can lead to a lot of gossip, opportunity/drama.

I've lived in small town settings, so I don't deny you can't end up knowing a lot of people. One might easily have more close friends or family than someone in a larger urban settings (the opposite might also be true). However, it is unlikely (though not impossible) that you'd come in to contact with more people, and come in contact with them more regularly, than someone in a denser urban environment.

Keep in mind there's a difference between specific traits and circumstances an aggregate statistical data. You might have more people you interact with in a small town setting, what's the aggregate set of interactions for the community? Does the community as a whole have a broader population and more frequent set of interactions than in a more densely populated community? Particularly when you eliminate interactions that don't afford much opportunity for murder (like interactions online), those numbers start to drop off much more rapidly, and it becomes all numerically impossible for them to be comparable.

giantg2OP4y ago

"That significantly increases the chance for interpersonal crimes of all sorts, including murder."

Well usually once you commit murder (and are caught/convicted) you're incarcerated. So in theory once you hit a certain number of interactions, the people prone to snapping would have snapped. So I think the threshold is low enough that people who would commit a murder have sufficient opportunity in either environment.

"One might easily have more close friends or family"

In theory, if murder is something that generally involves strong passions, wouldn't the closer relationships result in the violence when someone snaps from a betrayal or habitual disrespect?

"becomes all numerically impossible for them to be comparable."

I think that's only true if we are weighting those interactions linearly and not including mitigating factors that could be more prevalent in one population or the other.

cbsmith4y ago

> So in theory once you hit a certain number of interactions, the people prone to snapping would have snapped.

That statement is presuming there's some fixed threshold of interactions or fixed set of people to interact with that most people achieve, before someone "snaps", at which point they are immediately incarcerated. I don't think there's anything resembling evidence, or even a theoretical model for criminal behaviour, suggesting that is the case --certainly not for crimes of passion, which are often not committed by people of all ages who hadn't previously murdered at any point in their lifetime.

> In theory, if murder is something that generally involves strong passions, wouldn't the closer relationships result in the violence when someone snaps from a betrayal or habitual disrespect?

Strong passions might erupt from interactions between a third person that one might have a mere acquaintance with, having an interaction with someone one had a closer relationship, no?

Let's assume that's true though. Let's say 80% of murders are crimes of passion, and 90% of crimes of passion are entirely derived from interactions within people one is so close to that regardless of population density you'd have more than enough interactions to hit some threshold. That still leaves 28% of murders that are far more likely to occur in more a densely populated environment... and if the rest of the murders are uniformly distributed, that would mean the divergences due to population density for those remaining murders would be far more significantly concentrated in denser populations.

Now, what the data actually shows is that there is one specific murder victim that is disproportionately likely to be murdered by someone they have a close relationship with: wives (sons are also disproportionately likely, but to a significantly lesser degree). In most cases the relationship between victim and offender is either unknown or a more casual "acquaintance": https://ucr.fbi.gov/crime-in-the-u.s/2011/crime-in-the-u.s.-...

> I think that's only true if we are weighting those interactions linearly and not including mitigating factors that could be more prevalent in one population or the other.

No. Over a decade long period, in a densely populated urban environment, if you look at the acquaintance-of-acquaintance or even acquaintance-of-friend populations, you could easily have a larger population than a small town or even a state. Even if you limit it to just "passionate", romantic relationships, the adage of, "I've dated eligible person in this town" is just something that is far more likely to occur in less densely populated areas.

1 more reply

j / k navigate · click thread line to collapse