As I said, the derivational process we find here is a very specific one involving reduplication and zero-grade ablaut of the root. If we look at the documented

As I said, the derivational process we find here is a very specific one involving reduplication and zero-grade ablaut of the root. If we look at the documented history of the Indo-European languages, both reduplication and ablaut had already become subject to erosion, or even total loss. It is just not realistic for those two things to have survived (and, in this particular derivational process, to survive in sync!) for thousands and thousands of years.