Comment by hammerzeit | Hacker News Reader

Comment by hammerzeit | Hacker News Reader

Parent Full thread

I'm astonished by the total lack of conversation around the actual data here. Both the article itself and the discussion below are predicated on the assumption that this is a rampant problem in the USA, but not a single subthread on the actual incidence of this.

I'd expect better from Hacker News.

For what it's worth, Wikipedia has a complete list of people killed by US Police Officers and the police reports from them. In 2012, there were 587 people killed by police [1]. This is much, much larger than in Germany, even accounting for population difference.

I would argue that more relevant than pure population is differences in homicide rate. By the data you've given, approximately 1-2% of all homicides in Germany are by police officers (6-12 officer killings, 690 homicides in germany [2]). It's worth noting that in the USA, approximately 3% of all killings were by police officers (547/14748[2]) which seems in line with what my understanding would be -- it's worse in the USA but not ridiculously disproportionate.

[1] http://en.wikipedia.org/wiki/List_of_killings_by_law_enforce...

[2] http://en.wikipedia.org/wiki/List_of_countries_by_intentiona...

A lot of the abuses don't have good statistics around them.

I personally know a guy who was put into a private jail on trumped up charges and denied bail. He fought in court, won, and has been presented with a large bill from the private jail. The kicker? The DA, the sheriff and many in the police force are investors in the private jail!

Can you say "conflict of interest"?

That sounds terrible, I'm sorry for your friend. Systemically that's broken and represents the potential for massive abuse.

That said, I assume you'd agree that anecdotes don't constitute data, and that we shouldn't develop strong beliefs on the backs of them. To assume the prevalence of something on the basis of loud anecdotes leads us to being easily manipulated -- in the absence of data I believe should keep our priors weak and seek better data.

Actually I believe that our brains are wired to work from anecdote to theory, not from statistics to theory.

In a small group this is a pretty good strategy. Where it breaks down in life is that the odds that we hear about an anecdote depend on the anecdote. This leads to all kinds of bias.

I know the limits of my conscious analysis. I therefore have come to accept that there is value for me in filtering out "something that happened to someone I directly know" from "something I heard about". Since this happened to someone I directly know, it affects what I think.

YMMV, and if you follow my rule, definitely should. But you'll at least understand why I might come to different conclusions than you do.

Epistemology is not my strong suit so please excuse me here, but how can you reasonably advocate for your positions in a worldview that specifically privileges personal experience?

I tend to agree that we infer theory from anecdotes not data and even that that is a good thing. But don't we have a responsibility after developing a theory to evaluate it before making claims about it?

Epistemology is not my strong suit so please excuse me here, but how can you reasonably advocate for your positions in a worldview that specifically privileges personal experience?

Epistemology is about how we know what we think we know. This is very separate from rhetoric, the art of trying to convince others with your words.

I personally focus on epistemology, which includes clarifying what I believe, why I believe it, how strongly I believe it, and what different things I could reasonably believe in different circumstances. Knowing these things often leads me to conclude, "I believe X, you seem to believe Y, there is no way that I am going to change your beliefs." This seems to me to be a feature, not a flaw.

I am unable to be friends with strongly religious types who are unable to muster the same feature.

I tend to agree that we infer theory from anecdotes not data and even that that is a good thing. But don't we have a responsibility after developing a theory to evaluate it before making claims about it?

In my world view, yes.

If you see me failing to properly qualify what I claim, then please call me on it.

That said, when I make a claim, I'm almost never going to give my full supportive reasoning for my world view. Life is too short.

Data is only as good as the interpretation of it. You've just handwaved away a police homicide rate of an order of magnitude because the general homicide rate is also much worse. Other western democracies are bunched around a homicide rate of around 1.0/100k (Germany is 0.8), and the US is a clear outlier at 4.8. Having double the proportion of this significantly higher homicide rate is not something to claim as 'proportionate' and hence not a particular concern for the issue at hand.

One wonders at what point you would consider it 'ridiculously disproportionate'? 10%? 20%?

We (US residents) haven't posted annotated lists because this is in our face via newspapers and TV news and viral videos passed around on the Internet.

http://hiphopandpolitics.wordpress.com/2012/04/06/29-black-p...

Here's one take, which is focused on black folks shot and killed over the past year. Note how many were armed.

I'm american as well.

When TV news and viral videos and newspapers make us feel like terrorism or kidnapping is more prevalent than it actually is, we call it sensationalism. Why is this any different? Is it because it confirms things you already believe to be true?

I looked at the list you provided, and tried to correlate it against the Wikipedia list. Many of the dates etc don't match up but I tried. A number of the cases where the suspect is listed as "unarmed" don't jive with the police reports, at the very least. Others involve things that are clearly assault even if there was no gun (like assaulting someone with a car).

That said, I do think in particular if you are a black male you are likely to be mistreated by police. That's not justifiable. But I still don't see real evidence that this is an epidemic or particularly pervasive.

Well, believe what you want to believe.

I have friends who are police officers so it's not like I hate cops. I do hate the fact they close ranks and protect the bad ones though.

I wrote a response here, but with the current state of government surveillance programs, I thought it was better to remove it.

Why exactly did you feel the need to tell us either how paranoid you are or how much of a coward you are?

Reply on news.ycombinator.com