Circular Evolution

In his Natural History of Religion, David Hume proposes an evolutionary progress from natural polytheism to more developed monotheism.  

This, he thinks, is the natural and virtually inevitable progress: “As far as writing or history reaches, mankind, in ancient times, appears universally to have been polytheists” and idolaters. It makes no sense to suggest that “in ancient times, before the knowledge of letters, or the discovery of any art of science, men entertained the principles of pure theism.”

Besides, the traditional notion that humanity was originally monotheistic violates the evidence of modern discovery: “The savage tribes of AMERICA, AFRICA, and ASIA are all idolaters.” 

It’s a common sort of argument, but it’s the kind of argument that makes you stop and say, Hold on now, Mr. Hume. The argument seems to be: They are savage. Therefore they are early. They are idolaters. Therefore early man was an idolater.

But on what basis does Hume conclude that the “savages” of the new world represent “early” religion? After all, their “primitive” religion exists in the present. The evidence from tribal religions is evidence of early religion only if one assumes already that idolatry is early. 

One could, of course, make an evidentiary argument by comparing the religions of American Indians with those of Hittites or the people who inhabited Ugarit. But Hume doesn’t offer any such evidence. (He couldn’t; it wasn’t yet available.)

Hume aims to prove that idolatry is early by discovering idolaters and asserting that they are early. It’s an evolutionary argument, but a circular one.