I do not think there is a doctrine in the courts (civil law, mainly, with lawsuits and jury cases) that upholds the concepts of either "assumed risk" or "prior risk." That might be an argument, but I don't believe it has standing in the law books. Perhaps some industries have been granted waivers (gun lobby, skiing) but that's because of aggressive politicking efforts. The golf industry enjoys no such latitude, and even in cases where a golf couerse pre-existed a newer housing development, clubs can be held liable for safety issues from golf balls flying off course.
As for on-course safety, there are no documented rules for this, as no professional body wants to be the party to get sued when someone does get hit. Certainly, an agency apporoving a project is not making a judgment on safety - that's the responsiblity of the architect and owner. For this, a standard approach is for the centerlines of adjoining fairways to be separated by at least 250 feet. Anything less can be dangerous. Another way to calculate it is a 15-degree "cone of safety" delineated from the the centerline of a hole. Anything within either side of the 15-degree projection (calculated from the center, so effectively, as I understand it, it's 30-degrees wide) is considered at risk.
Here's one way to think about. 25,000,000 Americans people play golf. That means 250,000,000 Americans' don't. In a jury pool, you are therefore going to get 10-11 members of your twelve who do not play golf. How sympathetic do you think they'll be to the claims of a golf course that golfers should assume the risk? Lawyers want to avoid such jury trials, because most of the time the defendent loses. So that has an effect on lawyers, insurance companies and clubs.
If Scotland and Ireland (Lahinch!) have more crossover holes, that's because of a very different legal culture, where suing for damages when it rains is frowned upon.