The analogy is this: either the word "religion" is a linguistic convention, like "Indian" and thus includes Christianity and the "religion of the ancient Greeks" together; or else "religion" is a coherent entity, and then Christianity is a religion and the ancient Greeks had no religion, just like the Indians of the new world never set foot in India.
If we insist that the linguistic convention "religion" actually reflects something real in the world and not just some arbitrary set like {locomotives, politicians}, we run into all kinds of problems.
A trivial one is that music itself is "religion". The secular state which must not promote any religion is thus forbidden from promoting music.
In general, one by one, as you examine it, every human action turns into a religious action.
dwc · 626 weeks ago
In the chapter 2, you can see that Dutch missionary Abrahamus Rogerius (1609-1649) had to deduce the existence of religion in India; today, western atheists see religion everywhere in India. Look at the contrast: Rogerius, a trained priest and missionary, knew more than what these western atheists know about Christianity, religion, etc.
"Not knowing any Indian language, preaching for over ten years in Portuguese in the coastal town in India, how did this missionary find out whether religion existed among the Brahmins of Coromandel? This is how: 'Because where there is a God, a religion must exist too; it is thus that we shall approach our investigation (my italics).'"
The other argument that scholars appeal to linguistic practices, then use Wittgenstein's family resemblances. Balu handled that part as well.
http://xyz4000.wordpress.com/2011/04/01/family-re...