Stonehenge, by Bernard Cornwell (2000)

While waiting for my local library to gather all four volumes of Cornwell's Nathaniel Starbuck series, I picked up Stonehenge. Normally averse to pre-historic novels, but respectful of Cornwell's story-weaving skill, the book proved to be a worthwhile diversion.

Subtitled: 2000 B.C. (presumably for symmetry with the book's publication date), Stonehenge can't really be called a historical novel - it's more of an archaeological novel. Cornwell studied the research to date on the neolithic construction in England, which has resulted in quite a lot of information about the construction timeline, origins of the different types of stone, construction methods, possible uses of the various parts of the structure, etc. From this wealth of data, Cornwell invented an interesting tale of the late bronze-age: rivalries and conflicts between  tribal societies, leaders and religious traditions - in other words, the same kind of human drama that continues today.

The 5 criteria can't be applied to Stonehenge, so I'll just say that it was an enjoyable read, and far less speculative than most prehistoric tales. The story incorporates a lot of what archaeologists have deduced from the site's remains, in a much more entertaining way than plowing through a stack of journal articles.