That's a good point. More controversially, the latest research regarding smoking during pregnancy indicates permanent damage to the child(ren), i.e. stuff that will continue for all of their life. Not as severe as with some other drug use, but it's there. With siblings having children, the chance of medical conditions stemming from genetic problems is greatly increased. Cystic fibrosis is a simple example, as it's known to be caused by a single faulty gene on a non-sex chromosome and it is a recessive gene - a person will only have CF if they get the CF version of the gene from both genetic parents. Person A carries the gene. They don't have cystic fibrosis, as their other copy of the gene is OK. Since the gene for CF is recessive, the other copy is expressed instead. They'd have no reason to know they carried the CF gene unless the relevant part of the genes was examined. Person B doesn't carry the CF gene. Person A and Person B have two children, Person C and Person D. Neither C nor D can have CF, since they will get one copy of the gene from B, who does not have the CF version of the gene. Both C and D have a 50% chance of having the CF version of the gene, inherited from A. Half of A's haploid cells will have their normal version of the gene, half will have the CF version. A person selected at random has at most about a 4% chance of having the CF version of the gene, depending on their genetic background (it's far more common amongst people whose ancestry is from this part of the world). So the chance of a child of C and D having CF is 12.5 times higher than a child of either C or D and an unrelated person whose ancestors all came from this part of the world, the worst genetic background for CF. The difference is much greater if there are ancestors from elsewhere. That's just one of the many medical conditions caused by recessive genes. Factor them all in and it looks like a really bad idea for immediate genetic relatives to have children together with the natural random combination of genes.