The author claims 20 generations would be enough to breed the powers shown and thousands is unrealistic:
"
even the most pessimistic estimates of how many generations it might take to drastically increase average human intelligence or almost entirely eliminate a nasty recessive by R.A. Fisher might be 20 generations .
"<p>But some of the powers, like remembering ancestors memories barely even exist in humans today. The field of epigenetics has proven that a starving parent can alter how the genetics of even grandchildren express themselves. So some basic functions like wrapping DNA tighter or loser and adding a methyl chemical group exist to pass information across generations, but this is like an extra bit or two per gene. To extend the functionality to the level of recording memories would take almost as long as it took for DNA to become as fancy as it did from the original RNA molecules that could produce more of themselves without any enzymes or DNA having developed yet. So almost as long again as life itself has taken, not 20 years.
I always assumed that the BG were directing genetic averages across whole populations, and/or had such a deep understanding of genetics that they could predict the exact result of two peoples genes mixing.<p>If you could determine with 99% accuracy the exact genome of a child by their parents genomes alone, would you be able to optimize humans across given a relatively small set?
Very interesting, but fantastic genetic traits like precognition and ancestral memories are (in fiction) a traditional good fit for the Mendelian inheritance pattern of few and high impact mutations.
This is the best write-up of Dune that I’ve ever read. Some of gwern’s stuff is kind of overcooked to a fault, and with intense focus, misses the broader prevailing drift of the topic at hand, but gwern’s got me on this one. It’s a good read.