[bumby]

This feels like one of those topics that may sound ok in theory, but breaks down in practice. The implication is that the judges must be well-versed enough in any domain brought before them to interpret the laws effectively. This seems like a tall order for nine people. We have already seen this trouble in expecting strict interpretations regarding tech.

To be fair, Congress has the same problem. I believe that was in large part the impetus for giving the agencies discretion. They have a better chance of having the depth of expertise to craft effective regulations.

[mike_hearn]

The Supreme Court doesn't resolve cases directly, they resolve questions of law for lower courts to take into account. They are meant to be experts in law, so there's no problem there. The lower courts can't be experts in everything, but bear in mind two things:

1. Courts have expert witnesses and a whole system around how they are called, challenged and questioned. Judges are trained to learn what they need to know from witnesses.

2. Good court systems do have expert judges they can draw on.

I recently took part in the Craig Wright case in the UK as a witness. Wright forged enormous quantities of evidence and proving the forgeries often required deep technical knowledge about file metadata, how computers worked etc. Fortunately the judge was deeply technical himself, being often a judge on complex patent cases, and had no difficulty with any of the complexities.

[bumby]

It reads like it creates a deliberate impasse. The opinion states that ambiguities in law no longer implicitly give agencies discretion. That means Congress has to write unambiguous laws. But my original post acknowledges they cannot. Based on this ruling, it seems like anything other than a perfect, airtight law means it's effectively non-enforceable. So where does that leave us? It seems like the SC has laid the table for constant rules-lawyering by corporations to get whatever they want. In other words, they've let the perfect be the enemy of the good.

[mike_hearn]

> That means Congress has to write unambiguous laws

I don't understand where this belief is coming from. The judgement explicitly states that writing unambiguous laws isn't possible. There will still be ambiguous laws, and those ambiguities will still be resolved. The only matter being decided on is who gets to resolve ambiguities - is it the agencies or is it the courts.

Let's put this another way. Did Congress have to write unambiguous laws or have them be unenforceable before 1984? Clearly not. The Constitution itself is ambiguous on many points. Do other countries, which lack any equivalent of Chevron deference, have to write unambiguous laws or have them be unenforceable? Again, clearly not.

[bumby]

>The only matter being decided on is who gets to resolve ambiguities - is it the agencies or is it the courts.

I think we are agreeing here. I think the distinction is that I'm claiming the courts would need (yet don't have) the expertise to clear up ambiguities in such domains. Where do I get this claim? From the justices themselves.[1] There is evidence they are overly confident in their understanding when "doing their own research" on a domain outside their expertise.[2]

So given that context, it's probably a bad idea to have justices decide on ambiguities. But if that power resides in them now, it means the only way to have effective laws is to avoid ambiguities in the first place. That's why I stated that is now on Congress. However, the court is also acknowledging that isn't possible. That's why I originally said it reads like they created a deliberate stalemate. From the court we have the following:

1) Congress cannot be expected to create unambiguous laws.

2) It is the court's job to resolve ambiguities.

3) The court lacks domain expertise.

I'm claiming those set up a natural conflict because expertise is necessary to effectively resolve ambiguity.

The most generous interpretation is that the justices don't need to know the details of the domain expertise, but rather just need to know how it interfaces with people and the law.[3] I'm pretty skeptical of that leading to good outcomes in complex, nuanced situations. I don't think we can pretend law is abstractly disconnected from the complex systems it regulates. As society progresses, most things get more complex so I expect the problem to get worse, not better.

[1] https://www.businessinsider.com/supreme-court-google-tech-so...

[2] https://www.propublica.org/article/supreme-court-errors-are-...

[3] https://www.vox.com/2014/4/23/5644154/the-supreme-courts-tec...

[mike_hearn]

Remember that law is first and foremost meant to be read by ordinary people, as they are the ones expected to know it and follow it. Ignorance of the law is not an excuse, etc. If a law is so difficult that even a judge armed with a courtroom full of expert witnesses can't figure out what it means it should not exist, because how could any actual citizen be expected to understand or obey it? People who write such laws don't get to whinge when other people try to clean up their mess.

There's certainly no evidence of an attempt to create a deliberate stalemate here, as "laws that judges cannot interpret even with help" isn't something the Supreme Court is going to consider a legitimate problem to begin with so why would they consider it? But it's fascinating the degree to which some people on this thread are sure this is all a cunning secret plan by the justices to advance a political ideology. Projection, much?

> There is evidence they are overly confident in their understanding when "doing their own research" on a domain outside their expertise.[2]

Great. So are regulators, politicians and especially journalists. The idea that agencies never rely on bogus claims or statistics is a deeply romantic view.

Really, I don't have any sympathy for the US regulatory agencies on this issue. Too many of them have a history of abusing Chevron deference. If the deference standard had any advantage at all it'd be that you could get binding rulings from regulators ahead of time, without needing to actually end up in court first. But regulators always refuse to do that, because they aren't obliged to and it would reduce their options. Instead what they do is make vague rules, tell citizens (e.g. entrepreneurs) who ask for clarification to get lost, wait years for some of those people to make successful businesses and then decide that what they were doing had been illegal all along under some strange interpretation of the relevant rule. It's a form of retroactive lawmaking, which is forbidden by basically any constitution because the whole point of law is that people are meant to be able to follow it.

This sort of abusive behavior is so common it's clearly deliberate. Instead of helping society, the agencies end up working against it in order to maximize their own power.

Having the people who interpret laws be fully separated from those who write the laws aligns incentives properly: there's no longer any benefit to passing maximally vague or contradictory rules in the hope of being able to do whatever you want later. Laws that are as specific and clear as possible is exactly what a society needs to succeed, because when the rules are clear people are free to innovate and do business without fear.

[bumby]

>So are regulators, politicians and especially journalists

Yes. The important distinction that I drew was that the power is now vested in a very small group of nine people. IMO that measures it much easier for biases to continue unabated. At least in large numbers, there are more likely to be more rational viewpoints to counter our innate biases.

[thfuran]

>Courts have expert witnesses and a whole system around how they are called, challenged and questioned.

Yes, but this is a system where everyone involved is trying to (help their team) win rather than trying to produce generally efficacious policy.

[SllX]

Well they’re not trying to produce any policy. A District court is trying to resolve a case brought before them.

[thfuran]

Yes, that's my point. An agency interpreting the law that governs their mandate is at least nominally trying to do so in a manner that yields effective policy, but that's not really a goal of the courts, so letting the courts defer to the agencies probably will produce better policy then leaving it up to case law.

[mike_hearn]

> An agency interpreting the law that governs their mandate is at least nominally trying to do so in a manner that yields effective policy

Ideally. In practice ambiguities in statutes governing agency mandates are always interpreted by agencies in whatever way would maximize their own options and power. That's the written justification for the APA in the first place: reeling in excessive "zeal" amongst administrators.

The courts don't get powers from the law directly (except meta-powers like the one this case is about), so there's no direct benefit to judges from interpreting law in a maximally agency-advantageous way. That's why countries (not just the US) have courts resolve ambiguous law. It's a core part of their job.

[SllX]

Yeah, but giving too much deference to agencies is also a recipe for peoples' rights to be trampled by a regulatory agency acting unlawfully. Acting in accordance with their policy goals is fine and all, but they still have to do so within the bounds of the law and if they're losing at the District level because they can't rely on Chevron deference anymore, I mean, that's completely fine. Go back to Congress.

Despite its reputation for not passing any laws, the last session of Congress still passed a fair number of laws: https://legiscan.com/US/legislation/2021?status=passed

(The current one seems to be a fair bit behind, but it is also operating on razor thin vote margins and the session is not over yet).

[bumby]

How does this ruling fix that issue when the court admits it’s impossible to avoid ambiguous laws? Now it’s just unelected justices deciding instead of unelected agency bureaucrats. I think there’s an argument that could be worse, given the lifetime appointments of the court; at least the people have the chance to course correct the executive branch every four years.

I think we can all agree that the best solution is unambiguous laws, but that’s an impossibility by the courts own admission. Chevron already had a mechanism to prevent agency overreach by giving the court discretion to determine the reasonableness of an agency interpretation. This now just puts the onus on the court to do it all, and they admit they don’t have domain expertise. I fail to see how that is a better solution.

[suzzer99]

I just to work for one of the leading expert witnesses in the country. Expert witnesses are just advocates, hired by a client, who know how to throw fancy words around in a soothing, confident manner that makes the listener feel smart.

[TimTheTinker]

That's a rather cynical take.

An abuse of a form doesn't invalidate that form's validity.

I would be logically incoherent to assert that all hamburgers are crap on the basis of having worked at McDonalds for years.

[suzzer99]

It's not abuse of the form. It is the form.

Nobody has a career as an expert witness for long if they go against a client's wishes. There is no such thing as an impartial expert witness in the current system. They're all hired by one side or the other. They're basically subject-matter expert lawyers.

[TimTheTinker]

> It's not abuse of the form. It is the form.

> There is no such thing as an impartial expert witness in the current system

You're saying a form is bad because it doesn't line up with how it ought to be (you said "impartial"). But in saying so, you imply there is an ought. That is the real form (what it ought to be): an impartial expert witness.

When a form only presently exists in an abuse thereof, that doesn't necessarily invalidate the form as a worthy pursuit.

History can tell us whether such a form is worth pursuing. Expert witnesses in the US have been more or less impartial in the past. Communism has never been achieved, and its pursuit has never led to well-being of the people. Both are "ideal forms" in a philosophical sense, but history bears out which is worth pursuing and practically achievable.

So my take is that what's needed is reform of expert testimony through congressional law, including independent review of DAs, to ensure they prosecute expert witnesses who failed a legal impartiality test.

[mike_hearn]

Expert witnesses are already required to be impartial. There's currently an inquiry in the UK over a massive miscarriage of justice caused by an expert witness going rogue and letting prosecutors put words in his mouth, etc. A big part of the scandal is that prosecutors were meant to inform him of his legal duty to be neutral and share all the facts, but they didn't, so now he's claiming he didn't realize he was required to share all relevant facts including those negative for the prosecution.

[Gormo]

> This feels like one of those topics that may sound ok in theory, but breaks down in practice. The implication is that the judges must be well-versed enough in any domain brought before them to interpret the laws effectively

No -- subject matter expertise is not relevant per se, as what is being evaluated in these cases is not whether the policy advanced by a given agency is sound on its technical merits or factual basis, but whether it is within the bounds of the authority granted to that agency by the applicable statutes.

The courts aren't concerned with "crafting effective regulations", they are concerned with ensuring that the people who are tasked with doing so are operating consistently with prevailing statute law and the constitution.

And, considering that the judiciary is staffed with the world's foremost experts in statutory interpretation and constitutional law, this arrangement makes sure that all of the disparate facets of the process are being undertaken by the most qualified people available.

It's not appropriate to make officials who are hired on the basis of their knowledge of medicine, RF transmission, economics, etc. to bear the entire burden of determining the complex legalities of their authority. They have to step far outside their field of expertise and engage in textual analysis of statutory law or delve deep into constitutional theory to determine whether their rule-making process is indeed legally permissible. It's better to let the courts do their job here and tell them when they are out of line, so they can focus on doing their own jobs properly.

[bumby]

Maybe I’m misunderstanding, but what you’re describing sounds like what I thought the case was before this ruling. The courts were deciding on the reasonableness of the agency interpretation. Now it sounds like the court is interpreting directly. To do the latter effectively, I still maintain you need a solid expert understanding of the domain.

It sounds like we disagree on who is better equipped to make the kinds of interpretations necessary for effective policy. Like I said in another post, I don’t think we can pretend law can be abstracted and cleaved from the systems it regulates. The court admits they don’t have expertise in those systems. That makes me feel they are ill-equipped for the types of interpretations.

[mike_hearn]

What you had understood and what he's describing is the situation pre-1984 and now post this judgement.

I think a lot of the reason this thread has blown up is that the Chevron doctrine was really very strange and not at all how you'd expect the US legal system to have been working. It doesn't line up with any standard teaching of civics, for instance. The Supreme Court clearly felt the same way and has now instructed courts to go back to doing what everyone thought they were doing already.

[bumby]

I don’t know that I agree. It’s exactly the understanding I was given from an engineering law course, for example. Prior to this ruling, the regulatory agencies were given latitude to interpret ambiguous law as long as they were determined reasonable by the court. It’s also what I think most people understand, given the way people rail against agencies rather than statutes.

To be clear, I think there is a distinction between declaring if a law is constitutional and resolving ambiguities for non-constitutional, domain specific issues. I think the court is eminently qualified in one area, but much less so (by their own admission) in the other. I think those are two different aspects that often get conflated.

[Gormo]

Yes, you are misunderstanding. Chevron obligated the courts to defer to agencies' own internal interpretation of statute law whenever any ambiguity arose as to what the law said. Reversing Chevron has restored that function to the judiciary, where it belongs.

[bumby]

I’m not misunderstanding then. Chevron still deferred interpreting ambiguity to the agency, but the courts could still check that based on a reasonableness standard. Now the court gets both aspects. Our difference is that I think that is a less good outcome because I believe domain expertise is necessary to effectively clear up ambiguity and the court admits they do not have that kind of domain expertise.

I think the court should reserve power for constitutionality (their domain expertise) and leave resolving the ambiguity outside that to the regulatory experts of those respective domains.

[Gormo]

> I’m not misunderstanding then. Chevron still deferred interpreting ambiguity to the agency, but the courts could still check that based on a reasonableness standard.

No, you are definitely misunderstanding. Chevron delegated a core duty of the judiciary to executive branch officials with no expertise in that field. The "reasonableness" standard you was a limited and constrained version of reasonableness standards devised and applied by courts in normal statutory interpretation, and deprived the courts of the power to fully exercise their duty.

> Now the court gets both aspects.

No, there is only one aspect here.

> Our difference is that I think that is a less good outcome because I believe domain expertise is necessary to effectively clear up ambiguity and the court admits they do not have that kind of domain expertise.

The domain expertise of the regulators in the field of regulation is irrelevant here, because the cases that go before the court are not about what measures are likely to be effective in fulfilling the agency's mandate, it's about what measures are legally permissible.

> I think the court should reserve power for constitutionality (their domain expertise) and leave resolving the ambiguity outside that to the regulatory experts of those respective domains.

The only domain at question here is the domain of interpreting the law, and the relevant experts in that are the judiciary themselves. You are conflating together completely distinct matters.

[bumby]

If you’re claiming the courts have more non-law domain expertise than agencies, you’ll need some evidence of that for this point to land.

If you look to Chevron for a concrete example, the issue was about the ambiguity of the scope of a “source” of emissions. That is not a law question because the law already deemed that sources can be regulated. It’s really a question of non-law domain expertise in terms of the definition of an emission source.

[mcguire]

This is the Supreme Court that has repeatedly made subject-matter related arguments (particularly historical arguments) while ignoring the input of subject-matter experts.