You've misunderstood what she's saying. For many issues it is not possible for Congress to write laws that are simultaneously completely unambiguous and actually deal with whatever issue they are trying to deal with. Actually applying the law almost always involves numerous decisions to fill in some of the gaps.

While members of federal agencies are not elected, their heads are appointed by elected officials.

I never understood the whole "un-elected officials" argument. How many people should we have on the ballot? 10s of thousands?

That isn't the point. My government is supposed to have it's rules made by one body, congress. It's very strange to me that there's a large contingent of people - even some who pretend to be lawyers - who think that it's fine that some random guy that gets hired through a political appointment by the executive is the one actually making the rules.

I would much rather unelected bureaucrat scientists decide how to implement the intent and application of laws than congress.

You hake no reason to assume that. The people who will try to write those regulations are those with have an angle. We call it regulatory capture.

I don't see how the threat of regulatory capture would argue against letting experts decide what actions need to be taken to protect the publics health and the environment

I would argue against government bureaucrats being experts.

Congress is similarly captured.

Which is way easier when you only have to convince random court with a gish-galloping person as "expert witness".

Regulatory capture at agency level is way harder to do.

Regulations that automatically expire after some time period but require Congress to given an up/down vote would give the regulations greater legitimacy in a democratic system.

Congress doesn't need the expertise to write the regulations. The elected Congress could just vote to pass the regulations as laws. Congress just doesn't want to be on the hook for the regulations, which is part of the reason why they hand off law-making to the agencies in the first place.

Theoretically, this approach would give people a greater voice in the rules that govern them. Sadly, in practice, we can't seen to rollback the proliferation of criminal laws that embolden prosecutors and lead to an unfathomable number of people in jail that have not been convicted by juries.

Congress already has that power now though..? The congressional review act lets congress review and vote on new regulations issued by government agencies. Lots of people in this thread seem to be missing the fact that congress already approves of these agency rules. If they didn't they would have blocked them under the CRA.

Congress failing to block a rule by passing a CRA resolution is very different from Congress approving of the rule. For example, if the majority in the House supports a rule but the majority in the Senate does not (or vice versa), neither an explicit approval action nor an CRA blocking resolution can be passed.

Not really. If congress writes the laws intending the responsible agency to resolve ambiguities (which they do in just about every case) and doesn't object via CRA I don't see how that isn't an explicit approval.

It’s incorrect to assume that Congress knows and intends every ambiguity they create and that they can foresee and approve of all the interpretations an agency might reasonably come up with for each such ambiguity.

Even in those cases where they recognize an ambiguity they create and where the agency’s interpretation is within a scope that Congress approvingly foresaw, that’s at most an implicit pre-approval of the agency’s regulation, not an explicit approval of the regulation in the sense that defeating a CRA resolution would be.

Far more importantly, most CRA resolutions attempted to date have not occurred in the same Congress as the one that passed the authorizing legislation for the regulation. There is no reason at all to assume that the Congress attempting the CRA resolution holds the same view on the relevant agency interpretation as the one that passed the authorizing legislation, whether that view is approval or disapproval.

You're assuming these people are actually dispassionate scientists and not, ya know, political appointees.

"Perfectly complete" is a pretty high bar. For example, consider a law directing the EPA to fine violators who dump "fatal substances". How complete is complete enough?

That leaves us with some options, such as these ones which I'm ordering from "most reasonable" to "most insane":

(1) In lawsuits, courts should generally assume that the lawmakers have given the EPA permission to create a formal list and judgement criteria for what counts.

(2) In lawsuits, courts should assume the list is totally empty unless a federal lawsuit has happened where both sides have called in "chemical experts" to testify and then a federal judge decides which chemicals are deadly and which are not.

(3) The law is totally meaningless until congress amends it with another bill that inserts a full list of every possible chemical composition and configuration required concentration-level, and anything not explicitly included on the list is exempt.

I think there is an option (4) which is probably the one the court would prefer.

(4) In lawsuits, the regulator should have to prove that the substance in question is fatal. The EPA will have published a list ahead of time and and if challenged it will be up to a court whether the EPA has correctly determined the lethality of the substance.

The most reasonable option would be for congress to explicitly put in the bill that the EPA (or some other group of experts) makes the list. I’m no legal expert but in your example it sounds like that’s all that’s needed.

It's in the law already; for instance, for the Clean Air Act, see USC 42 section 7422 where the EPA is authorized to reclassify previously unregulated substances. That the court chose to ignore the plain reading and intent of Congress is baffling, unless one explains it as gross corruption.

Did you list the wrong thing, because this seems pretty limited?

(a) Radioactive pollutants, cadmium, arsenic, and polycyclic organic matter Not later than one year after August 7, 1977 (two years for radioactive pollutants) and after notice and opportunity for public hearing, the Administrator shall review all available relevant information and determine whether or not emissions of radioactive pollutants (including source material, special nuclear material, and byproduct material), cadmium, arsenic and polycyclic organic matter into the ambient air will cause, or contribute to, air pollution which may reasonably be anticipated to endanger public health. If the Administrator makes an affirmative determination with respect to any such substance, he shall simultaneously with such determination include such substance in the list published under section 7408(a)(1) or 7412(b)(1)(A) 1 of this title (in the case of a substance which, in the judgment of the Administrator, causes, or contributes to, air pollution which may reasonably be anticipated to result in an increase in mortality or an increase in serious irreversible, or incapacitating reversible, illness), or shall include each category of stationary sources emitting such substance in significant amounts in the list published under section 7411(b)(1)(A) of this title, or take any combination of such actions.

(b) Revision authority Nothing in subsection (a) shall be construed to affect the authority of the Administrator to revise any list referred to in subsection (a) with respect to any substance (whether or not enumerated in subsection (a)).

Yes, that's the point: it is limited, but the Supreme Court just declared it to be insufficiently limited.

You missed the point of the paragraph. Congress routinely writes laws directing, for example, the EPA to determine if and when an emissions-reducing technology is economically feasible. The majority opinion is saying that courts need not defer to such findings and they will decide for themselves.