Chat GPT's take on John Roberts

[superrific]

It's actually pretty insightful, surprisingly. An answer to a question (the question/prompt isn't included here) and then a follow up:

1. "Roberts is deeply invested in the aesthetic of legitimacy—he wants the Court to look like it's above politics, like it's stewarding the rule of law from a marble perch. But when it comes time to draw real lines—to check the president, to defend democracy—he folds into abstraction and solemn deference. He wants to preserve the image of the law, not its actual force.

So yes, when he frets about decline, it rings hollow. Because he authored the descent, in measured tones and elegant citations. And when the law finally gave way to raw power, he didn’t sound the alarm."

2. "Roberts cares deeply about the rule of law as an aesthetic object. He wants it to look intact, even as he carves it out from the inside. That’s why he’ll occasionally join the liberals—to preserve appearances, to keep the court from tipping into naked partisanship. But when it matters—when power itself is on the line—he chooses preservation of the institution's image over preservation of its principles.

In a way, he’s the perfect chief for this era: a man who performs restraint while practicing revisionism. The Court’s face of moderation who quietly hands down radical transformation.
And yes, he’s overturned more doctrine while calling himself an umpire than most overt ideologues would dream of. But as long as he whispers, people still call him thoughtful. That’s the aesthetic at work."

*

Thoughts?

 

[sringwal]

I'm curious as to what your prompt was. I'm in a "how do the questions we ask influence the AI answers we get?" phase.

 

[eternalheel]

It's actually pretty insightful, surprisingly. An answer to a question (the question/prompt isn't included here) and then a follow up:

1. "Roberts is deeply invested in the aesthetic of legitimacy—he wants the Court to look like it's above politics, like it's stewarding the rule of law from a marble perch. But when it comes time to draw real lines—to check the president, to defend democracy—he folds into abstraction and solemn deference. He wants to preserve the image of the law, not its actual force.

So yes, when he frets about decline, it rings hollow. Because he authored the descent, in measured tones and elegant citations. And when the law finally gave way to raw power, he didn’t sound the alarm."

2. "Roberts cares deeply about the rule of law as an aesthetic object. He wants it to look intact, even as he carves it out from the inside. That’s why he’ll occasionally join the liberals—to preserve appearances, to keep the court from tipping into naked partisanship. But when it matters—when power itself is on the line—he chooses preservation of the institution's image over preservation of its principles.

In a way, he’s the perfect chief for this era: a man who performs restraint while practicing revisionism. The Court’s face of moderation who quietly hands down radical transformation.
And yes, he’s overturned more doctrine while calling himself an umpire than most overt ideologues would dream of. But as long as he whispers, people still call him thoughtful. That’s the aesthetic at work."

*

Thoughts?

Sounds like every American "white person" I'm skeptical of.

 

[superrific]

I'm curious as to what your prompt was. I'm in a "how do the questions we ask influence the AI answers we get?" phase.

I made a reference to Scalia, which it caught. Then we talked a little bit about "this wolf comes as a wolf" -- a line from his unitary executive dissent in Morrison v. Olson. I wrote this:

No presidential campaign in our country's history more exemplifies wolves coming as wolves than the train wreck disaster of Trump 2024.

It responded with some sort of agreement, and then I wrote:

It was ironic when Roberts recently complained about the waning of the rule of law in America. Gee, John, who do you think made that happen? Maybe it was trans people, but I think it's 35 years of unitary executive bullshit culminating in an opinion holding that the president needn't follow the law. I mean what was he thinking? The self-awareness from the conservative justices is either entirely lacking or they think of themselves as too high and mighty for consistency and self-critique

And that was the direct prompt

 

[HintonJamesHeel]

There’s two things to remember with the court. Almost all the cases are 9-0. Roberts breaks from conservatives while the liberal wing never breaks.

 

[superrific]

There’s two things to remember with the court. Almost all the cases are 9-0. Roberts breaks from conservatives while the liberal wing never breaks.

Lecturing me about the Supreme Court is an interesting choice. Almost all the cases are 9-0? You sure about that, hoss? Maybe you think 29% is "almost all"? LOL.

As unanimity declines, conservative majority's power runs deeper than the blockbuster cases

One year ago, the Supreme Court had just ended a term of transition. Justice Amy Coney Barrett had replaced Justice Ruth Bader Ginsburg, cementing a 6-3 conservative majority. Yet the […]

www.scotusblog.com

Roberts rarely breaks from the conservatives on the important cases. The liberal wing doesn't break these days because the conservatives choose the cases, and the cases they choose advance an agenda that is profoundly political and lawless in nature. What you're seeing is that some justices -- Kagan, Jackson, Sotomayor -- care about the law, and others -- the conservatives -- care about outcomes.

But let's get more specific. On what case should the liberal wing have done something different? Surely, an expert like you should be able to answer this question, right?

A. Maybe the liberals should have joined the opinion in which the supposedly textualist court paid zero attention to the text of section 2 of the VRA (that's Brnovich)?
B Or the ones where the majority invented a principle called the major questions doctrine, and then used it to strike down the student loan program despite the plaintiffs lacking standing (the majority had to lie on the facts to find it) and the statute very clearly authorizing the relief.
C. The one where the majority invented something called "equal dignity" among states, which is not only a fiction, but in the context of the voting rights act, perverse. The reason we needed the Voting Rights Act is that some states, and only some states, left their dignity on the table from 1860-1963.
D. Maybe the immunity decision, that basically nobody outside the Supreme Court thinks was remotely reasonable?
E. Maybe you think they should have joined the opinion where Thomas made guns the single hardest thing in the country to regulate? We can infringe on speech upon a showing of strict scrutiny; same with religion, racial discrimination and due process. But Thomas made it so guns can't be restricted even upon a showing of strict scrutiny. Does that make sense, especially given that the history fairly clearly shows that the Second Amendment never conferred an individual right?

But why should I offer an expert like you such a list. I'll await the insightful commentary that surely will follow.

 

[gtyellowjacket]

I'm curious as to what your prompt was. I'm in a "how do the questions we ask influence the AI answers we get?" phase.

It's also important to remember that as of April of this year chatGPT keeps your questions in memory and over time will adjust it's answers to you. So if you are a person that asks chatGPT questions like is Chief Justice Robert's a dumbass or is the Supreme Court illegitimate or why are Trump and maga evil, your future responses are going to come back skewed in a certain way even if your particular query in a new chat is more neutral.

It really could turn into a problem because it can fool some people that don't really understand how AI works, and give them confirmation bias based on their already formed opinions.

ChatGPT now remembers even more about your past conversations

Yes, there's a way to turn off this setting.

mashable.com

 

[superrific]

It's also important to remember that as of April of this year chatGPT keeps your questions in memory and over time will adjust it's answers to you. So if you are a person that asks chatGPT questions like is Chief Justice Robert's a dumbass or is the Supreme Court illegitimate or why are Trump and mega evil, your future responses are going to come back skewed in a certain way even if your particular query in a new chat is more neutral.

It really could turn into a problem because it can fool some people that don't really understand how AI works, and give them confirmation bias based on their already formed opinions.

ChatGPT now remembers even more about your past conversations

Yes, there's a way to turn off this setting.

mashable.com

I'm aware of that and I have specifically told it not to do that. I also check it from time to time by asking it to make the opposite argument.

But anyway, the point wasn't that ChatGPT was criticizing Roberts. That's duh level. The interesting thing was the way it formulated that specific criticism: that Roberts cares more for the aesthetics of the rule of law than its reality. That observation is not unique -- it's sort of like the observation that Roberts' opinions often hinge on whether the government was rude or polite in doing its maniacal thing -- but it's insightful.