As addressed in other posts, there is a significant difference between "selfishness" and "self-interest".  However, it seems that there is still a general insistence on perpetuating the selfishness perspective.

Recent articles have pointed out some of the altruistic actions that are performed by chimpanzees, however there are also articles that point out competitive or "selfish" behaviors that are used to rationalize and even extrapolate human behaviors.

One of the first misconceptions is the assumption that the failure to share automatically constitutes selfish behavior.  It doesn't.  Being self-interested does not constitute a requirement of cooperative behavior, and correspondingly the lack of cooperation does not presume selfishness.  I can simply choose to be indifferent; being neither selfish nor cooperative.

If we consider the self-interested perspective, then there is no problem in seeing how various behaviors can arise that promote that interest without necessitating contradictory conclusions or convoluted logic to derive such conclusions.

One example of just how silly such logic can be is illustrated by the following quote:
"But we don't help everybody in need and some people even look the other way. The Myanmar government is a famous example, accused by many aid experts of doing much less than they could to help their own residents, even thwarting aid shipments."
What is foolish in this quote is to equate the government with an individual.  There has never been any suggestion that cooperation is applicable to social groups, regardless of their origin.  There has never been any evolutionary issue that addresses such groups, and to inject such an example is not only misleading, but it is simply wrong at all levels.

It is equally important to put behaviors into a context, so that we recognize that not every instance and every circumstance warrants a sacrifice or selfless act to be acceptable.  It is clear that even the most generous behavior must have limits to avoid the total decimation of the individual initiating the action.

In humans, even the most selfish individual is still cooperative in the majority of circumstances, especially in major activities.  It is literally impossible for a completely selfish individual to either survive or have a biological future1.  In any society the obvious problem is that of finding mates.

In order to find mates, there is a need to find cooperative individuals with whom to associate, whether it be a potential mate's family, social circle or just the mate.  In other words, at some level, a degree of cooperative behavior must exist for a social bond to be created2.  It is unlikely that such a bond would extend only between the two individuals, since offspring and the interactions they have will also require a longer term cooperative involvement with others.  After all, even having offspring is insufficient if they cannot also locate mates when they mature.

It has often been argued that while such behavior exists, it is driven by "selfishness" since it attempts to advance an individual's personal interests.  However, since when is this the definition of selfishness?  Self-interested; yes.  As stated previously, it isn't clear why there seems to be such an intense desire to distort these meanings.

Although I'm reluctant to introduce the point, it seems that much of this is still being driven by the reductionistic perspective of genetics and "the selfish gene".  Once again, what constitutes "selfishness" versus simply "self-interested"?  In addition, it is difficult to argue for selfish behavior when the outcome is literally dependent on the cooperation of trillions of other copies of the same gene.  Are we to believe that these trillions of genes are all behaving altruistically so that one selfish gene can be passed on to future generations?  In the end, that is as nonsensical as it sounds.

Ultimately the problem is that, in the same way that evolutionary psychology has many problems in establishing natural selection as a biological basis for human behaviors, it is the height of foolishness to extrapolate behaviors to even less tangible elements based solely on a particular viewpoint.  Regardless of the whether the gene-centric perspective is useful, it is not relevant to the interpretation of actual behaviors that exist within any species.

Whether it is a palatable idea or not, the influence of groups in biology cannot be overlooked.  While many of the earlier problems with group selection attempted to suppress the effects of individuals over the group, it is equally inappropriate to behave as if groups have no effect on the evolution of individuals.  Biology is rarely an "either-or" situation and just as the synergy between ecology and evolution occurs, it is equally true that individuals and the groups they belong to are equally synergistic and will exert reciprocal pressures that cannot be simply divided.

In the end, there is no amount of twisting and turning that will ever rationalize selfish behaviors as producing cooperative and altruistic results.  If we can't accept the idea that cooperation and even altruism may have fitness benefits beyond the superfluous, then there will be whole segments of species' behavior that will never be properly understood.

1 Note that I'm not interested in considering individuals that live and die without offspring, since they contribute nothing to the genetic pool.  In those cases, it may be possible to find some exceptions but they are largely irrelevant.

2 This is also a reason why "kin selection" is inadequate to explain behaviors because a significant number of social interactions to procure mates does not involve kin.  So while kin may be important it is hardly complete, nor does it offer significant insight into real group dynamics.