This is a good theory.Jason Hare wrote: ↑Thu Oct 22, 2020 1:25 pmThe doubled sound of the resh was before the dagesh mark was introduced to represent the distinction of sounds, so it's probably not useful to think of it in terms of dagesh or no dagesh. Probably the idea of it being doubled, like the distinction between r and rr in Spanish, where we have a basic opposition in words like caro "expensive, dear" and carro "car," like pero "but" and perro "dog." It's possible that the resh was a tap, and the idea of a doubled resh was a trill (like in Spanish).
This is on the assumption that resh was tapped, which I think is a good assumption.
As for the question in the post, the r cannot be included with bgdkpt because they are all stops that become continuants when surrounded by vowels (plosive—>fricative). /r/ is a liquid; it cannot undergo the same phonological change since it is already a continuant.
In other words the /r/ cannot take dagesh lene.