Perspective and data in historical analysis

Introduction
Imagine that two millenia or so in the future, literary experts attempt to collect the glories of our literature. Most of our paper writings have crumbled into dust or used for kindling; all our digital files are long gone or indecipherable. English is a dead language and many of the cultural references are a complete puzzle to them. They have a strange jumble of popular and high literature: one partial summary of of the episodes of a saga called 'Star Trek', a fragment of an archive of fan fiction about a warrior princess named Xena, some quotes from various authors extracted from anthologies written three hundred years from now, and a few cryptic bits of poetry from somebody named Shakespeare, who was apparently very highly regarded, and wrote in an archaic dialect: specifically, one complete sonnet, a couple of soliloquies and a few random lines from his plays. Now try to psychoanalyze Shakespeare from those fragments. This is about where we stand vis-a-vis Sappho.

This quote raises the interesting question about the sufficiency of sources for analysis. Of course, lack of facts has never stopped people from analyzing (a natural tendency) but how much do we need for a valid picture? Here, I'm frequently reminded of an article by an Egyptologist (I have the reference somewhere) commenting how easily we look at the 3000 years of Egyptian politics and culture as a homogeneous unified whole whereas the pyramids of Giza would have seemed just as strange five hundred years later as medieval cathedrals do to us (in fact there were several renaisance movements during those 3000 years including one in which the Sphinx had to be literally dug out of a layer of sand covering it). So on one hand, even this incredible amount of data we have about Egypt is not sufficient to appreciate some of its aspects fully. On the other hand, a few fragments of the pre-Socratics or Sapho, have sparked massive amounts of 'respectable' scholarship. In my brief and abortive forays into history have always been most satisfying to me when it came to the middle ages. It appeared to me to have just the right amount of data (not too much like memory and not too little like ancient history) - but it was also the approach (closest to anthropology with people like Le Goff, GureviÄ?, or Steblin-Kaminsky).