Master Pangloss taught the metaphysico-theologo-cosmolonigology. He could prove to admiration that there is no effect without a cause; and, that in this best of all possible worlds, the Baron's castle was the most magnificent of all castles, and My Lady the best of all possible baronesses.
“It is demonstrable,” said he, “that things cannot be otherwise than as they are; for as all things have been created for some end, they must necessarily be created for the best end. Observe, for instance, the nose is formed for spectacles, therefore we wear spectacles. The legs are visibly designed for stockings, accordingly we wear stockings. Stones were made to be hewn and to construct castles, therefore My Lord has a magnificent castle; for the greatest baron in the province ought to be the best lodged.
We use Just-So Story in the sense that Stephen Jay Gould used it to criticize Evolutionary Psychology arguments. We admire Gould because he has a humility that many Evolutionary Psychology advocates seem to lack, however the term is not an all-purpose refutation of evolutionary arguments.
A “Just so story”, as we see it, it is the basic error of lazy evolutionary reasoning. That is, if we propose a mechanism for how something could be adaptive, we feel we have explained that thing, when it remains quite possible that:
Dysfunction is the main error that evolutionary theorists concern themselves with. They have devoted great effort to mathematically proving that certain mechanisms work, given seemingly plausible (but inevitably greatly oversimplified) assumptions. A problem that we have with some authors of the smarty pants school is that they argue by means of round-trip error (see The Black Swan) that the lack of a proven-functional mechanism for altruism, etc., proves that it is theoretically impossible for the world be as it appears to be, so we are deceiving ourselves. For example, true altruism doesn't work evolutionarily, so it doesn't exist. “No proof for the evolution of altruism” is not proof of “no altruism”, and is also not proof that altruism arose via a non-evolutionary skyhook (see Darwin's Dangerous Idea.) We take both altruism and evolution as givens, so must conclude that we just don't have a working theory yet.
Any story with either of those weaknesses is a “Just-so story”, so the caution to avoid “Just-so stories” is no more than the caution to avoid invalid or inadequately supported evolutionary arguments, but Gould was primarily concerned with Contingency error in his discussion, as this is not given as much consideration as it deserves.
Both Daniel Dennett and the Not by Genes Alone authors address the “just-so story” with the moderately valid objection that spinning off plausible theories is a large part of how we make progress, and also that the historical facts to address Contingency error are usually difficult or impossible to come by, so we are faced with either abandoning any attempt at evolutionary explanation or casting caution to the wind and making speculations that are consistent with what we do know. In order to be a productive scientific hypothesis a story should not only be consistent with the facts as we know them, but also make testable predictions.
A particular weakness common in evolutionary just-so stories is the speculation of convenient past conditions. One sure defense against this weakness is to only call on conditions that apply right now or documentably applied in historical times. This is the uniformitarian principle which was productively applied in geology. It can be regarded as a form of Occam's razor. A uniformitarian evolutionary argument should win points even if it lacks the simplicity of an argument based on hypotheticals. We believe that sociobiologists and evolutionary psychologists have been too quick to give up on uniformitarian arguments. Being able to speculate any conditions is just way too much rope. But geological uniformitarian dogma led to the rejection for years of ideas that seem clear in hindsight, such as that landscape has been shaped by colossal deluges of a magnitude vastly exceeding historical observations.
It seems likely that evolution has often depended on stressful unusual circumstances, especially when it comes to cooperating to achieve a higher level of organization. jared_diamond notes that politically, states arise in opposition to a common enemy, and not from enlightened self-interest in a time of peace. Similarly, we can imagine the eukaryotic cell arose in a desperate situation where neither of the parent cells were viable on their own, so that the inevitable bugs in the new arrangement were not a devastating competitive disadvantage. See Meta-Evolution.
The just so story is an instance of what nassim_taleb calls the “narrative fallacy” (in The Black Swan), which is that the Story is so beautiful, explains so much, that it must be true. Beauty is truth, truth beauty. But at first glance scientific truths about our place in the universe have often been ugly; their beauty is at best an acquired taste.
“Beauty is truth” is just another way of looking at the unconscious “Gestalting” process that we use to evaluate Truth in all matters, great and small. Clearly our judgment of a “good story” is right often enough that we shouldn't ignore our intuitions, but some humility and appreciation of our cognitive frailty is also called for, especially given the strong evidence for self-deceptive overestimation of our powers (see Positive Illusions.)
Another take on just-so-stories is that the “rightness” of a story depends on the consistency of the story with the other stories our culture tells us, its mythic power and summary of the human condition. Once we cast off the burden of evidence and spin stories based primarily on their beauty, then it's no surprise when we prove that things naturally end up the way that we see them (though the lens of our culture.) The rightness and beauty of a story is only a poor guide to its truth, but often it's all we've got.