Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Fenrir]

Firearms are legal, but hinged sticks are not.

...

[Starver]

The anomalies are legit.

Ah, thanks for that explanation, LS, I understand that 'limited-range differential', now.  How quickly one forgets such huge events.

Spoiler: Waffle... (click to show/hide)

The likes of Fred West and others would have been prior to the sea-change in measurement, of course, but Shipman truly did stand out, by sheer volume, if treating the all the crimes as happening in the year of discovery.  At first thought, of course, I wouldn't have personally even considered the two Chinese incidents (the lorry and the cockle-picking) as 'murder' in that list, and more manslaughter (or deaths caused through neglect/dereliction in someone's duty of care), but obviously lines are drawn and court-assigned responsibility and degree of culpability has probably got to be used as the guideline, ultimately.  I suppose that's the criteria that ('sans'ing, aside) were employed in that particular data-gathering exercise.  Like I said, I've not personally checked any of the accompanying references.

Still, GreatJustice, it still leaves (as both my footnote and Leafsnail said, to one degree or another[1]) some indication that it's increasing less/actively decreasing, post-change.  But there are too many variables![2]

As a wise man I know once said "Anything with more than seven variables is an art, not a science."  With a possible six or seven million largely independent variables, running around the place, we're a complex little island...  I'm not saying that any analysis is ultimately invalid, just want to point out the difficulties that deficient and low-res figures are going to give us.  Especially for an occurrence that has relatively low-levels of incidence at the best of times, for which any spikes and dips are spikier and dippier and the inevitable regression towards the mean is meaner...


[1] Although I'm more cautious, I'd be much happier with a few more years of aftermath to analyse.

[2] It could just as easily have been Sky TV that stopped the 'inevitable' decline in homicides in 1989 by introducing more trans-Atlantic action films to UK viewers and making them more aggressive and the post-millennial rise in the more populist ISPs in the UK (like Talk-Talk in 2003) that grabbed everyone's attention and got them stuck reading the Internet, with the rest of us Geeks, rather than going out and being violent to real people.

[Nadaka]

Slightly off topic, but addressing something mentioned by Starver...

The idea that exposure to violent media increases violent acts by desensitization is completely unproven.

The idea that exposure the violent media reduces violent acts by providing a safe release for aggression is just as valid, if not more so.

Exposure to real violence and exposure to simulated violence are very different things.

[Prometheusmfd]

Slightly off topic, but addressing something mentioned by Starver...

The idea that exposure to violent media increases violent acts by desensitization is completely unproven.

The idea that exposure the violent media reduces violent acts by providing a safe release for aggression is just as valid, if not more so.

Exposure to real violence and exposure to simulated violence are very different things.

Exposure to real violence, even if it isn't relatively (RELATIVELY) non-violent violence (because that's a thing) has always been known to create more violent individuals, especially if they are exposed to it at a young age.

Now, simulated violence is different, because starting at a young age, your brain will always see that and always know that it is not real. This can even be applied to images of real violence, as it is a different experience than actually having the violence right in front of you.

[Starver]

[@Nadaka]

Please feel free to put that in the same category as my "increased violence due to alien symbiotes infecting us" (or whatever phrase I shoe-horned into the earlier post of mine), if you wish...  It's just a hypothetical (and probably, in the symbiote case, hyperbolic) explanation, indicating that there are other things going on... other markers that could be as easily added to the graph.

(To be honest, I couldn't even properly match up ISP usage with the drop in violence, either.  That was an off-the-top-of-my-head "exemplar that wasn't", too.)

[Zangi]

Theoretically... wouldn't violence from and to people personally known by a young individual have more of an effect on polarizing said young individual's inclination to violence themselves?

Not media mind you, but real shit going down.

I don't really adhere to the violent media = violent people idea... its more a fault of personal life experiences and society/culture saying its ok, but not really ok... so keep it to yourself and things will be gravy.

[GreatJustice]

In addition if you extend the graph to present day it becomes extremely obvious that what had been an upwards trend for some 30 years reversed decisively by 2003.  If you exclude the large Shipman anomaly of 2002 and the cockle pickers of 2003 it seems like the trend probably reversed (or at least plateaued) in 2001, 4 years after the law came in.  I'm sure it wasn't the only factor, but "This law was introduced and then 4 years later a year-on-year rise in murders suddenly became a year-on-year fall" doesn't help your case much, considering you'd expect there to be a little while before the impact is felt.

Well the obvious point here is that a year by year fall doesn't necessarily show anything in the long run. As the chart shows, the short term trend was often decisively downwards yet the long term trend was upwards; from '68 to '08, the average homicide rate was 52% higher, while from '97 to '08 it was 15% higher. It could, admittedly, be reversing, yet lets look at long trends in US states/the US proper at the same time:

Washington D.C vs the US

Spoiler (click to show/hide)

-The rate of homicide averages 73% higher during the gun ban, yet the U.S rate of homicide averages 11% lower.

Chicago vs the US

Spoiler (click to show/hide)

-The rate of homicide in Chicago after the gun ban averages 17% lower while the rate of homicide in the U.S averages 25% lower, which seemingly shows that it worked but...

Spoiler (click to show/hide)

-The percentage of murders committed using handguns actually INCREASES dramatically when the gun is put into place, clearly showing that the gun ban made no difference to the decline of crime.

Inversely, let's see what happened in Florida when right to carry came into effect...

Spoiler (click to show/hide)

-Floridian homicide rates average 36% lower after the introduction of right to carry, whereas U.S homicide rates average 15% lower, a significant difference.

Whatever the case in the short term, it's pretty obvious that American homicide rates have been declining steadily for the past 50 years whereas British homicide rates have been increasing, and the states with the highest decrease in homicide rates are those which are trending towards looser gun control laws.

What is state above is that the states which were required to enforce the brady bill saw the same decline in homicides as states which already had the restrictions. It's up to interpretation, and specifically is not proof that the law had no effect, the same report says:

"On the plus side, there is strong evidence that the law undermined gun-running operations that were buying large numbers of guns in southern states and transporting them north for resale, he said."

So there's "strong evidence" that the Brady Bill reduced the flow of illegal guns from states with previously poor gun control to those with stonger gun control.

Yet (A) The states which introduced right to carry legislation were the ones that saw the largest decline in crime afterwards in comparison and (B) California is neither referenced in the article provided nor qualifies as a "northern state" (for the record, California's restrictions were almost exactly the same as those imposed by the Brady law, and its decline in crime without any additional increases in restrictions was 18% whereas in the rest of the nation it was 7%).

[Starver]

Well the obvious point here is that a year by year fall doesn't necessarily show anything in the long run. As the chart shows, the short term trend was often decisively downwards yet the long term trend was upwards;

I officially have no idea if you're arguing what I thought you were arguing, now.  It doesn't change my own views, but it rather changes which particular subset of my views and experience I bring to the table in either support or rebuttal.  (Or, indeed, whether I bother to do so, if any relevant comments I might make have already been/will shortly be made by others.)

Snip a whole lot of exposition, that I'd just written.  Reducing to: Wasn't that graph supposed to prove something?  The longest-term of trend that can be seen post-1990s legislation change can be charitably described as "not as much upwards as it was before", probably describable as "level, but noisily so" and could easily be taken as "starting to trend downwards".

Or is it the 1960s legislation that you wanted to prove as being causing the upwards-trend?  I've already given reasons why I don't see that (as direct cause->effect, leastwise).


But, when you come down to it, I don't like any of the implied inferences from this line's trend (that aggregates all killings, with the exception of the particular outlier exceptions that I now understand the nature of), given that it doesn't even any give indication as to the proportion of cases involved that come from any "firearm usage" subcategory.  Unless I've overlooked some assertion that firearms are a 'gateway weapon', in society as a whole, or some other obscure significance that I'm still not even sure I'd support anyway.


(Well, that version of exposition was about half as long as the previous lot, but still long.  I'll cut it there.  At least until I'm less lost as to everyone's actual viewpoints and stands.)

[GreatJustice]

Snip a whole lot of exposition, that I'd just written.  Reducing to: Wasn't that graph supposed to prove something?  The longest-term of trend that can be seen post-1990s legislation change can be charitably described as "not as much upwards as it was before", probably describable as "level, but noisily so" and could easily be taken as "starting to trend downwards".

Yet, as an average, the homicide rate is still higher compared to when the '97 gun law was introduced, regardless of the short term trends. Inversely, the homicide rate in the US has been declining rather steadily despite looser state regulations regarding guns and concealed carry licensing.

But, when you come down to it, I don't like any of the implied inferences from this line's trend (that aggregates all killings, with the exception of the particular outlier exceptions that I now understand the nature of), given that it doesn't even any give indication as to the proportion of cases involved that come from any "firearm usage" subcategory.  Unless I've overlooked some assertion that firearms are a 'gateway weapon', in society as a whole, or some other obscure significance that I'm still not even sure I'd support anyway.

Whether the murder is committed with a firearm is irrelevant in this case. Is it a success if the number of people shot dead is decreased, but counterbalanced by an increase in the number of people chopped up with hatchets? No. However, the case of Chicago shows that even reducing the number of handguns used in crime doesn't always happen in the case of a handgun ban.

[Leafsnail]

Yet, as an average, the homicide rate is still higher compared to when the '97 gun law was introduced, regardless of the short term trends.

This is actually simply untrue.  We are at a 30 year low for homicides.  Unless you're using your justfactspresentedinanextremelyselectiveway.com trick of taking averages over completely arbitrary lengths of time.

[Starver]

Yet, as an average, [...snip]

Too much noise in that signal to be able to say anything like that with any confidence.  Sorry.

As to the point...  Pretty much irrelevant.  Success is "less people who would have been shot have actually been shot", and have avoided an alternate death as a bonus.  There will always be people who never were going to be killed by gun (but going to be killed) and still never were shot (but were still killed).  Quantifying these and all other combinations of "would/would not have been shot/not-shot to death but thanks to/despite the legislation were/were not shot/not-shot and are currently alive/dead" is what we really need.  But short of a double-blind study involving parallel/re-run Earths adopting varying legislature with all else being equal, there's little that can be easily proven on any of this, though.

(Comparing US to UK massively fails on the "all else being equal", I wouldn't even bother without some form of matched-cohort study, and even then the complexities of the comparison are legion.)



[Tried to keep that succinct.  Middle paragraph ran away from me, even after editing.]

[GreatJustice]

Yet, as an average, the homicide rate is still higher compared to when the '97 gun law was introduced, regardless of the short term trends.

This is actually simply untrue.  We are at a 30 year low for homicides.  Unless you're using your justfactspresentedinanextremelyselectiveway.com trick of taking averages over completely arbitrary lengths of time.

A 30 year low by what standard? Certainly not as an average from 1997 to 2011. A single year of low homicides proves nothing; I believe the term is "statistical anomaly".

Plus, again, you still ignore that at this time American homicide rates have been going down barring minor interruptions for the past ~40 years, whereas British homicide rates have been generally trending up.

[Pnx]

I actually got a little curious about this one and started to do a little fact digging. I took the FBI's figures on homicide rates, and the US census bureau's figures on the population and came up with some figures for homicides per million. I'd upload the document itself but I'm not sure where to upload it right now. But here's the chart I made:

Spoiler (click to show/hide)

This chart could be much prettier, but LibreOffice is pretty crappy for making charts.

One of the things that I always found interesting is how the recession has yet to cause more people to resort to crime. In fact, since the recession hit, the amount of crimes being committed seems to have dropped.

[RedKing]

Of course, that chart reminded me of the Freakonomics explanation for the drop in violent crime in the 1990's: abortion.

Specifically, the continuous decline starts from 1993, 20 years after Roe v. Wade made abortion legal. Their argument was that less unwanted children = less maladjusted, mistreated young men and women = less future robbers, muggers and murderers.

Of course, they don't argue that was the ONLY variable accounting for it. The fact that we locked up a lot more people was part of it as well.

[Prometheusmfd]

Of course, that chart reminded me of the Freakonomics explanation for the drop in violent crime in the 1990's: abortion.

Specifically, the continuous decline starts from 1993, 20 years after Roe v. Wade made abortion legal. Their argument was that less unwanted children = less maladjusted, mistreated young men and women = less future robbers, muggers and murderers.

Of course, they don't argue that was the ONLY variable accounting for it. The fact that we locked up a lot more people was part of it as well.

Of course, anything explained by Freakonomics should be taken with a grain of salt, whether you're left or right.

But anyway, that graph is interesting. I wonder how different it would be with all gun-based crimes as opposed to just homicide.