1) Are you new to the Internet? My insult was almost a compliment by the standards set by the IEEE.

2) I don't necessarily know more than the bad PhDs running around, I just know better. Nice sentence there, though... something about how I judge dumbasses. But you're a nice guy so I'm going to (attempt to) educate you.

3) Yep!

4) I was under the impression that you stated:

"Alcohol is not a CAUSE of violence. You cannot prove that people who are violent with alcohol will not be violent without alcohol - A controlled experiment on that scale would require controlling for every other factor which causes violence, which is basically impossible. And highly illegal."

>> No... studying alcohol is perfectly legal. And randomization is legal too, which generally precludes worrying about confounders because by definition a confounder is:

1. Causally associated with the response variable in the population but not in the causal pathway of interest

and

2. Associated with the predictor of interest in the sample

As you can see, 2 generally won't hold if your predictor of interest was divied up among the participants randomly. This is the entire point of randomized clinical trials (okay, maybe not). The psych department around here does lots of stuff where they give kids booze and measure their risk taking behavior, interactions with peers and whatever.

"I will admit there is a distinct correlation between the two, but it proves nothing. Perhaps an inclination to violence causes alcoholism. Perhaps just people who aren't drunk get away with more violent crimes, whereas people who are drunk get caught more, leading to a disparity in the statistics Linolaques mentions. You cannot prove that alcohol causes violence."

And at what point does a "distinct correlation" become "causation"? It's a difficult thing. With smoking and lung cancer, for instance, it's agreed upon because you know... there's evidence of many kinds, biological and national and from studies. Alcohol's "distinct correlation" is quite different though. Violent behavior doesn't lag 25 years behind drinking. You have to pull really, really hard at the "oh no there was no intervention" rope to continue stating alcohol doesn't cause violence.

Have you ever played a drinking game? Let's say we're playing a drinking game, which has some randomizing effect on how much people drink, right? Now who's the violent drunk? The guy that had to chug whisky. This is an imperfect experiment.

How about self-reported violent activity, self control, etc.? I have plenty of friends that used to get absolutely tanked and get into fights, wrestle and other dumbass stuff. Now, they don't drink -- and why? Because they "don't like getting all violent". So let's say that's 20% of the population -- 20% of the population self-reports that alcohol causes THEM to be more violent. You'll dislike this one, but it's more like someone that recognizes they have allergies (believable) than someone attempting to ascertain the cause of their depression (not necessarily believable that they would know). This is an imperfect experiment.

Okay, one more. I'm throwing a party and I order 3 kegs. I invite all of my super cool frat brothers and tell them to wear their biggest collars. Everyone shows up at 7pm -- alcohol doesn't cause violence, so they're goign to be violent at about 11pm anyway (yes, every frat party includes violence (sexual or otherwise)). But the truck never shows, and everyone just sits around. There's no fighting. This is an imperfect experiment.

Unfortunately, you'll only accept a perfect experiment because you're not arguing anything you believe you're just arguing to argue. And unfortunately, that is only worthwhile when SOMEONE believes the point you're trying to argue. When you argue points that NOBODY believes, you're just being A RETARD.