I recently watched a program on the Science Channel entitled "Human Nature: Born to Kill".  In this episode, part of the objective was to explore humans killing other humans.  Of course, the usual array of topics were raised, from evolutionary psychology to genes [i.e. the "warrior gene"] and even the "nature vs nurture debate".

However, there were a few points that I felt were muddled and confused.

In the first place, it is important to distinguish general human violence from the specific case of human violence against other humans.  There is little doubt that most animals experience or participate in violence, so that would not be a surprising attribute.  What is of relevance is to consider the context in which such violence occurs.   

In humans there are several categories of killing that need to be considered for any meaningful insights to be gained.  First let's consider the psychological situations that may exist.  Certainly psychopaths represent one group that kills indiscriminately, but they don't provide any evolutionary insight.  These are individuals that clearly have psychological issues that are not representative of the human species as a whole.  

From this we need to further divide the issue of such violence into "killing" versus "murder".  Again, we can ignore elements of carelessness or recklessness as not being evolutionarily significant.  Therefore we have to examine those that kill for their own personal reasons [i.e. murder] versus those for whom killing is sanctioned by the group.

In the latter case we find soldiers, police officers, etc. whose job may well entail killing others.  While some may rationalize this as a type of self-defensive behaviors, it still takes a toll on those involved.  It has been estimated that during WW II, only 15% of the soldiers engaged in active combat actually aimed at the enemy in an attempt to kill them.  This was a major problem for the military and newer training methods have dramatically improved this kill rate.  However, it also appears that this may have increased the incidences of Post-Traumatic Stress Disorder (PTSD - up to 20% of returning soldiers) because despite the circumstances, the individual may still harbor significant feelings of guilt for the act of killing (1).

Note that these newer training methods are focused on making killing more reflexive and do little or nothing to address the psychological attitude of the individual.  One often hears that the military trains "killers", but that isn't true.  The military can train an individual in the means of killing, and can, through repetition, elicit automatic responses to situations, but it cannot make someone a killer.

The act of killing crosses a psychological threshold that cannot be undone.  An individual may be quite capable of extreme violence, engaging in all manner of physical conflicts, and yet the act of actually killing another human being is a psychological event outside of that mindset.  It may be easy to imagine killing someone in self-defense, or to eliminate someone whose behavior is considered especially heinous.  Yet, we tend to imagine such situations where we become angry, or by assuming that there is some emotional element that will push us to such an act.  The notion of being emotionally disconnected and still kill is something that is beyond most people's ability to comprehend (2).

What is important to note is that, these are situations in which killing is not merely condoned, but expected.  It has the full approval of the social group and yet it still creates numerous psychological problems for those that engage in such activities.

Mixed into this are those individuals that can rationalize killing by making their enemy sub-human.  This becomes a psychological tool to cope with the act of killing by transposing the violence that we might employ during hunting, and applying it as a rationalization against another human being.  In my view, this is a bit of a gray area, because even using this kind of rationalization [i.e. enemy as animal] isn't necessarily effective because most humans don't arbitrarily engage in violence against animals without purpose.

This leads us to the last group, which I would characterize as those that don't feel a psychological taboo against killing.  I would consider these individuals to be primarily narcissistic and/or sociopathic.  In effect, these people are either motivated to act in their own interests, in the extreme, or lack social empathy.  Again, there can obviously be a wide range of behaviors that would fall into this category, but for our purposes, the primary point is that they are not constrained sufficiently so that killing is outside of the possible options they may pursue.

What is quite surprising is that, despite overwhelming amounts of information to the contrary, such individuals can still believe or imagine that they can hide the act of killing or escape its consequences.  

So, with these basic definitions or concepts we can now consider some of the premises outlined in this science program.  

Genetics

Invoking evolution is always difficult because invariably the issue of genes is raised despite having no clear indication that there are specific behavioral genes.  One suggestion was the "warrior gene", or MAO-A gene.  Whatever role it may play, one thing is certain; it is not definitive with respect to aggressive traits.  Coupled with numerous environmental elements that can also exert a significant influence, it likely has low heritability.  One big problem with the "warrior gene" is that it is essentially non-predictive (3). 

However, a much bigger problem is to argue whether such a gene, especially given its non-definitive nature, is actually adaptive.  An argument that may be plausible is that it is actually a group adaptation so that, depending on the aggressiveness of the group, the individual with such a trait, may be better able to integrate into that group than someone without such a gene.

I don't see any reasonable argument that would credit this gene with a selection advantage since there are no societies that routinely condone murder.  Even in very violent or aggressive societies, what may appear as indiscriminate murder is usually a part of a more elaborate system of retributions and/or status.  Yet, even if we allow for such considerations, it is clear that it is not a wide-spread practice and therefore evolutionary selectability arguments are suspect.

Ironically, a major difficulty occurs because of the gene-centric view and a general reluctance to accept the idea of group selection.  After all, behavioral genes only make sense within the context of a social group, so while individuals acquire traits, the selection force may well be the group.

Evolutionary Psychology

Perhaps the most confused part of the program dealt with infanticide.  While infanticide certainly occurs in other animals perpetrated by other adults, the topic with respect to humans was raised regarding mothers that kill their own children.  The example used in the program was that of Susan Smith, who drowned her two children.

Of course, this immediately created one of the prime difficulties in the whole program.  First, was the declaration that all violence is motivated by sex.  Of course, this isn't true, since biology isn't about sex but mating.  Secondly, most violence is clearly about power and not sex.  No one would be foolish enough to suggest that rape is about sex, so it stretches the boundaries of credulity to suggest that it becomes about sex when we introduce evolution (4).

However, the truly absurd argument occurred by David Buss, when it was suggested that killing your own children could be a strategy for reproductive gain.  Even the most favorable interpretation of such a statement cannot reconcile the problem.  While one could certainly propose relative benefits of abandoning children or increased promiscuity, it would be difficult to explain how killing your own reproductive investment can ever be beneficial, regardless of circumstances.

This is rendered doubly absurd by rationalizing how Susan Smith may have felt.  That view held that her children were a hindrance to her love affair, so by killing them she could improve her reproductive potential in the future.  However, here's the problem.

The man she was having the affair with didn't want children.  By reading the quote from the letter breaking up with her, it is rather unambiguous.
"I'm sure that your kids are good kids, but it really wouldn't matter how good they may be ... the fact is, I just don't want children. These feelings may change one day, but I doubt it."

http://crime.about.com/od/murder/a/susan_smith_4.htm
This explanation is rendered more problematic when one considers that she wasn't a single parent such that her goal could have readily been accomplished by simply abandoning her children to their father.  Same result, fewer consequences.  

In the final analysis, we run into the same problem mentioned previously, which is the rationale of an individual thinking that they can engage in murder and that somehow life will continue normally for them.  Again, this has all the hallmarks of sociopathic and narcissistic thinking rather than representing any evolutionary strategy.

Overall, the program did introduce some relevant points, despite some serious missteps or misleading correlations.  Yet, one can't help but feel that the science would be better served by focusing on what is actually known rather than peppering the dialogue with fanciful tales of genetics and "just so" stories.  There are enough people that are suspicious about the claims made regarding evolution, so that what isn't needed are more speculative claims made with the thinnest veneer of plausibility.
 
=================================
(1)  While PTSD involves numerous factors the emphasis in this program was to assert that a goodly portion of it in soldiers may originate based on the actions [i.e. killing - moral injury] taken by the individual rather than simply the trauma of being in combat or of experiencing horrific events. 
"For example, several articles have documented the relationship between killing in war and a number of adverse outcomes. Fontana&Rosenheck (1999) found that killing and injuring others was associated for PTSD even when accounting for other exposures to combat within a larger model. Subsequent studies have expanded upon these findings, demonstrating a relationship between killing and a host of other mental health and functioning variables."
http://www.ptsd.va.gov/professional/newsletters/research-quarterly/v23n1.pdf
(2)  Note that there is often much bravado associated with the idea of killing, but that is merely the hallmark of someone that has never dealt with such a situation.

(3)  Of interest is that the MAO-A gene appears to make women happier, and it has been suggested that this effect is dampened by increased testosterone levels in men.  In addition, it appears that the role of this gene may not be as specific as simply aggression or violence, but rather to temper reactions based on levels of provocation.
http://hscweb3.hsc.usf.edu/blog/2012/08/28/study-finds-gene-that-predicts-happiness-in-women/
http://www.pnas.org/content/106/7/2118.full.pdf+html


(4)  It should be understood that evolutionary strategies are intended to improve fitness; survival and reproduction.  Therefore it is difficult to argue that something is evolutionarily adaptive if it reduces fitness, which is precisely what domestic violence does, especially during pregnancy.  
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1522123/

Additional Reading

Coping mechanism for executioners: Moral flexibility

The Moral Toll of Executions on the Consciences of Executioners