Yes, lead poisoning could really be a cause of violent crime

[HiPointDem]

79. i'm not convinced by all the ad hoc, maybe this maybe that stuff on lead pipes & lead paint,

since there's *still* plenty of lead paint & lead pipes around, even at this late date, & certainly way more circa 1937-1965. and i've yet to see any comparisons of blood lead levels for the relevant periods (pre-40s v. 40s-60s). Just speculation.

i take your point about uniform fbi stats, but there's a proxy, which is murder rates kept by individual cities. You may have a small point about better reporting, but murder is a crime that tends to be reported or noted by the authorities -- because it *is* murder, and there's a body or aggrieved survivors. Also, the definition has mostly stayed the same through time, whereas the definition of aggravated assault (the measure used by one of the lead researchers) hasn't (it widened over time, encompassing more types of crime).

here's san francisco:



that's a graph made from SF city records: "I found some ridiculously detailed homicide stats that name every person ever murdered in SF from 1849 to 2003 — the very first on the list was Beatty Belden, shot by “Chileans” on Telegraph Hill because of a “misunderstanding” — and crunched some numbers over time. (I took the 10 year census numbers and extrapolated SF population for other years , except for the last 10 years where I found yearly US census estimates.)"

http://burritojustice.com/2009/11/16/reclaimed-from-bligh-yeah-i-know-i-know/

and here's the database: http://cjrc.osu.edu/researchprojects/hvd/usa/sanfran/

In SF the low-homicide period runs ~1920-1965, well after the introduction of leaded gasoline.

And murder rates in the 1840s were even worse (the 1860s and 1980s appear as mere blips in comparison)



also (& not to you particularly, just a general question), why would lead exposure affect rates of violence of young men particularly, but not the larger population (assuming they also had been exposed in childhood?



This isn't the greatest chart because it doesn't break out the 'over-25' group into age fractions -- but why would homicide rates *drop* for over-25s while rates for under 25s soared, if the culprit is lead? Presumably a 35-year-old male in 1986, exposed to lead as a child in 1951-onward, would still be suffering from the poor impulse control, irritability, etc. caused by lead exposure, and is still physically able to murder. So why just look at men 15-25?