Brett Cochrane

Strong start but it increasingly outstays its welcome. Quite contradictory and conjecture laden - Grossman views traumatic anecdotes of war with the utmost sincerity and anecdotes of those unburdened by it to be insincere. It is poorly edited - the same points are made repeatedly interspersed throughout, sometimes with the same verbiage. The last section is a complete mess - it is correlational drivel about video games causing violence without care for larger contextual issues. It would be interesting to get Grossman's take now, given that violence in videogames is ever more visceral and prevalent, and homicide rates have continued to decrease since its publication. Also, it contains maybe the most asinine thing I've read in a scholarly book: "In statistical terms, an increase in aggressive predisposition of 8 percent of the population is very small. Anything less than 5 percent is not even considered to be statistically significant." Grossman uses this to suggest why correlations between violent media and violence are not often detectable, and this is not how statistics works. The alpha criterion (the 5% Grossman is referring to) is used by statisticians to determine the likelihood that two samples are different, and it has nothing to do with the size of the effect (the 8% increase). That is to say, an 8% increase is substantial, and would certainly be detectable if it was indeed valid. Given that Grossman lacks the understanding of basic inferential statistics, it draws into question whether his thesis is worth trusting more generally. With that said, it isn't a terrible read - it certainly made me appreciate the trauma of war, and there are some interesting tidbits here and there. However, I don't recommend going beyond the first 5 sections.