The NIH, like most federal agencies, is used to this legal doctrine known as "deference," which comes in many flavors, usually you will hear of "Chevron deferen

The NIH, like most federal agencies, is used to this legal doctrine known as "deference," which comes in many flavors, usually you will hear of "Chevron deference." The general idea is, (step 1) when Congress does not specify what should be done, (step 2) the agency's decision need only be "reasonable." In this context, "reasonable" means "a lawyer spent words writing it," which given how the government operates, is almost always the case. So here, Congress _did_ specify what was to be done. Oh. So no deference for you. It's really hard to figure out what to write in legislation. It is also somewhat hard to update the legislation to accord with new but subtle understandings, especially considering how much attention Congress members can possibly have as compared to how much law is in the books (go to a law library, look at a copy of the United States Code without the annotations, and consider how long it would take to read it, and that is only the beginning). But it is better when courts apply the law as written than making stuff up to suit their preferences, constrained by their prior precedents (assuming you are still at a law library, look at a copy of the United States Reports -- all the Supreme Court opinions ever -- and consider how long it would take to read that, and that is only the beginning).