<html><head></head><body><div><span title="pharos@gmail.com">BillK</span><span class="detail"> <pharos@gmail.com></span> , 21/4/2014 10:46 AM:<br><blockquote class="mori" style="margin:0 0 0 .8ex;border-left:2px blue solid;padding-left:1ex;">The BBC has an article up showing that in recent years removing lead
<br>from petrol leads 20 years later to a big reduction in violent crime.
<br>Prison and social policies make no difference - it's a medical
<br>problem.
</blockquote></div><div><br></div><div>Effect size? Causality testing? The theory is nice, but one needs to check for how much is explained by the lead hypothesis.</div><div>https://www.sciencedirect.com/science/article/pii/S0013935107000503?np=y</div><div>gives some data, and it definitely looks like something is going on - but lead is just a small (<20% of variance) part of the decline. What evidence does Gesch have to claim it explains 90%?</div><div>https://www.sciencedirect.com/science/article/pii/S0160412012000566?np=y</div><div>has a claim that *sounds* as strong (but the 90% is the model as a whole), but inside the model lead is still just one factor.</div><div><br></div><div>No doubt neurotoxicity is a relevant problem and might contribute to violence (and indirectly, via lower IQs, to a more shortsighted society). But one should not start to assume it is the major explanation for a complex social activity like violence just because it would be neat, exculpate a lot of people and a model claims it is the explanation. Extraordinary claims (the social stuff does not matter for this social outcome) require extraordinary evidence. </div><div><br></div><div><br></div><br><br>Anders Sandberg, Future of Humanity Institute Philosophy Faculty of Oxford University</body></html>