The biggest advantage is that it lets other OS easily impersonate linux e.g. SmartOS or Windows can provide kernel-level "linux persona" and have an unmodified linux userland run on top.
They can do right now. Both the API and ABI are very rarely (read never but very problematic bugs or similar) broken in a backwards way.
That's my point. It's only possible with Linux because Linux provides a stable kernel ABI. You can't impersonate Windows or OSX the same way.
But you can impersonate Windows by providing your own implementation of the Kernel32 API. What's the essential difference between that, and emulating syscall API?