Why knowledge for the pure sake of knowing is good enough to justify scientific research