There is little doubt that there is an abundance of evidence regarding the role of cooperation among social animals and it is mentioned here as a precondition for the discussion of group selection and altruism; specifically in humans.

While many of the elements that constitute human behavior are present in other animals, there is little question that humans represent an extremely unique existence.  Beyond even the obvious differences regarding intellectual capacity, humans represent a eusocial mammal that exploits an extreme division of labor, such that it is human society that operates to support human existence, rather than the capabilities of any individual (1).

It is clear that no single human possesses sufficient knowledge or capability to replicate that which society has accomplished, and it is equally clear that the intellectual and technical achievements of human society are the result of a unique ability to exploit skills that may occur in only one individual and yet be able to propagate those benefits throughout a worldwide society.

As a result, human endeavors that pursue various scientific or technological objectives aren't merely beneficial to individuals or small social groups, but add to and expand benefits to the entire group.  In fact, it would be impossible for many of our modern scientific and technological advancements to occur without specialized subgroups prepared to act on the knowledge.

Consequently it is fair to say that human achievement is a group effort, and is dependent on the "intelligence" and survival of the group rather than the individuals in that group.

This marks a major departure from comparison to many other species, because humans exemplify group selection as a criteria for survival.  Most genetic influences are largely irrelevant in selection of the group's success, because individuals are readily replaced.  In the same way that the human body is not compromised by the loss of individual cells, neither is the human society compromised by the loss of individual members (2).

This change marks a radical departure from the fitness objectives of most biological species.  In short, while many animal groups depend on individual survival for cohesion of their various social groups, in humans, individual fitness has almost no consequences to the group.

In this respect one can readily acknowledge that individuals may even choose to ignore their own biological fitness objectives and voluntarily avoid producing offspring.  To the best of my knowledge this is completely unique to humans and serves to illustrate precisely how far removed human social evolution has placed the individual from a dependency on such biological objectives.

Therefore if an individual's death doesn't compromise the group's potential for survival, and an individual's genetic contribution to future generations doesn't compromise the group's survival, it becomes clear that the objective of individuals in such a group is to maximize their individual benefits while in the group.  Their influence extends beyond simple genetics, such that the advancement of the group, provides individuals a kind of "immortality" only realized, in most species superficial genetic contribution to the future (3).

Another factor that supports this view is that humans are among the few animals that do not have regularly visible reproductive cycles.  They can essentially reproduce at will compared to the rigid mating patterns of many other animals. Yet despite this prolific ability, humans generally operate at significantly lower levels of reproduction than their biology allows.  

As a result, this creates the unique situation of where an individual can live a major portion of their lives without incurring any biological risk of fitness.  In other words, if a human adult has two children, then their reproductive fitness has already satisfied the base requirement for propagation, yet they are now free to act in whatever manner they choose, and despite not having any more offspring, any act they engage in cannot be said to have a fitness cost.  Fitness is determined in a completely voluntary manner (4).

Even arguments that would claim this lower level of reproduction is because humans must commit a significant portion of their lives to rearing young, this argument only accounts for a small portion of an individual's total life span, so it doesn't serve to explain much regarding the fact that the majority of one's life has no fitness risk.

It is therefore, my contention that altruism surfaces under such circumstances, because there is essentially no fitness risk to individuals.  In short, the eusocial nature of human existence already precludes the ability of individuals to survive without the social group, so it follows that the group itself becomes the entity that must survive.  Therefore any loss of individuals, in whatever capacity, still serves to promote the group (5).

From these we can begin to see that there are two specific elements unique to humans that create a new way of viewing human evolution.  Altruism isn't subject to the same fitness requirements as exists in other animals, and could arguably be considered uniquely applied to the group selection that constitutes human society.  It is therefore incorrect to examine human altruism from the perspective of individual fitness.  It does, however, provide useful insight into why its existence should promote group existence.  

It is equally important to note that such altruism is not demanded by the group (6).  Many individuals would display strong resistance to the notion that such altruistic behavior should be "expected", and yet we find that there is a subtle kind of expectation within the culture and behavioral ideas that humans express in their society, but nevertheless, altruistic behavior is expected to be completely voluntary.  Coercion invariably compromises the group's survival.

While many of the precursors to human behavior, regarding cooperation, and even altruism are present in other animal species, the uniqueness of the human evolutionary trajectory of a eusocial grouping has also created unique behaviors relating to reproductive fitness, altruism, and the meaning of group selection.

By recognizing the role of group selection in human evolution, we find that altruism isn't nearly as mysterious to explain such a success, since it is not dependent on the more traditional Darwinian view of natural selection to explain it's existence.

(1) Many people balk at such a description because humans still like to engage in the illusion of the "rugged individual".  Even a cursory examination illustrates that such a view is a biological dead-end.  Even if an individual were to be capable of surviving completely by themselves in whatever wilderness environment is envisioned, the first difficulty would occur in finding mates.  From this, raising offspring successfully becomes more difficult, but then the real problem surfaces when one has to find mates for the offspring.  It is this continuous requirement for interaction that demonstrates the inability of individuals to maintain themselves in any biologically viable way.

(2) This is highly analogous to the role of cells in a multi-cellular organism.  Each cell has fundamentally superceded its right to reproduce to the "higher" organization, even including the necessity of self-sacrifice (apoptosis - cell suicide).  This is one of the primary examples of "altruistic" behaviors occurring among single-celled organisms [by producing a multi-celled organism].
(3) Consider that Isaac Newton contributed no genes into future generations (he had no children), yet it would be impossible to deny his contributions to the social group.

(4) This isn't to say that some individuals may want children and be unable to conceive, but aside from these medically related issues, human reproduction is largely a completely voluntary process involving only individual choice.

(5) This isn't intended to trivialize the emotions, attitudes, and feelings that individuals experience when someone close to them dies, but rather it is an attempt to illustrate that such feelings are strictly localized and not reflected in the group at large.

(6) It is also worth noting that altruism is often naively presumed to be a sacrificial act, instead of recognizing that it may equally be a risky behavioral act.  In many case, the altruist does not expect to sacrifice themselves, so even high risk ventures are entered with the expectation that there is some finite probability of success.  In those cases, success may provide significant benefits to the actor within the social group.  As a result, we must be careful to recognize that merely confining altruism to self-sacrificing acts, misses the greater context in which these may occur.