Supreme Court Term Limits
In a thread recently a bunch of you considered the Biden Supreme Court term limits proposal not that bad. Steven Calabresi just wrote a piece on just how radical and worrying it is. I'd realized things were bad, but not how bad.
Let's set aside the minor things I was talking about before, like how the proposal would lead to two new justices each term in a way that would increase how politicized the court is. Those are dwarfed by Calabresi's concerns.
One thing I hadn't sufficiently thought about is the effects of this being a statutory and not constitutional proposal. The chief one, of course, is that they only need to get a trifecta, not a supermajority, to pass the bill. Well, how likely is that? Manifold puts it at a 21% chance of a democrat trifecta. And if that happens, there's a 65% chance they end lifetime appointments. If I can multiply, that's about a 14% chance of this happening. I assume themotte is numerate enough to realize that for sufficiently important events, that's really high.
Okay, so what would happen? Unfortunately, we don't actually have a draft bill on hand. What we know is that this proposal would add a justice to the court every two years, with 18 year terms. There are two problems with this, as stated: first, it's blatantly unconstitutional, and second, how do you get this to combine with the current system?
It would be unconstitutional because the Constitution provides for "in good behavior." Calabresi clarifies that under British law this would mean "life, unless you commit felonies," and in the context of the American Constitution would be likely to require impeachment. In order to evade that (recall, they do not plan for an amendment), they are likely to have them graduate to some nominal title like "senior justice" with basically no power (maybe still let them play a role in the "which state gets the water rights" disputes, but not anything else), because the Constitution gives Congress power to shape appellate jurisdiction.
The second issue is how to start this up. The way that has only minor harms is that they could make the limits start only for future justices. Maybe each seat switches to 18 year terms after the current inhabitant resigns or dies, with the term already partway through to align with the biennial appointments. But let's face it, there's no way they choose this. Their concerns lie entirely with the current court; the term limit proposal is merely a nice-looking vehicle to attack them through. It would make no sense for them to ignore the reason that they're passing this. What they'd actually do is immediately phase out people as soon as they reach 18 years. That would mean Thomas, Alito, and Roberts would be immediately gone. They would then promptly replace them with three new rubber-stamp progressives.
I imagine the court may well, when it would first have the opportunity, strike down the stripping of jurisdiction as unconstitutional. I'm not sure. But that wouldn't get rid of the three new justices just installed. The court would then sit at 6-6 (with, I imagine, the conservatives being significantly more willing to break ranks than the liberals). The first expansion of the court since the 9-justice court was established in 1869. This is the first serious threat at court packing in nearly a century, when FDR pushed for it. (I wish some amendment to stop court packing had been passed during the good while when it was uncontroversial.)
Of course, Republicans would, upon gaining their own trifecta however many years later, promptly then adjust the rules to their liking. The net result of this will in the long term be the end of the independence of the federal judiciary, seriously harming things like equality before the law. So much for caring about democracy and so forth.
Will this happen? Probably, if they get the chance. Sinema and Manchin will both be gone, so there's not much risk of filibusters surviving. Posing it as being about term limits, instead of court-packing, dramatically lowers how radical it seems, making them more likely to do it. And 36 senators have already signaled that they are already willing to substantially mess with the court by proposing the No Kings Act, so it's well within the Overton window. No chance Kamala is more moderate on all this than Biden is.
Okay, well, what can stop this disaster? The main things are: hold onto the Senate, hold onto the presidency, or get moderate democratic Senate candidates to say they won't go along with it. The last is tricky to do rhetorically ("term limits" sound good, as seen by the positive reception it got here). So it's not the easiest to convey that the democrats are now the party threatening our system of government. The easiest way currently is to win in the Senate. The Democrats have to basically hold onto every seat, including those in red states. Unfortunately, they have a moderately high chance of doing so, but Montana's Jon Tester, at least, has a greater than 50% chance of losing his seat. Should these fail, I hope one of the liberal justices can be convinced to break the usual custom against political advocacy to speak about how bad an idea this is. I'd also try convincing Obama to come out against it, if there's any way to do so, as he holds influence without (I imagine) being quite as scrutinized and purity-tested and generally pressured by whatever the current left-wing discourse is pushing.
As a final note, the No Kings Act is also pretty radical—stripping jurisdiction in immunity cases from the Supreme Court (leaving it at the circuit courts), and instructing Federal courts to ignore any presidential immunity. No way is that last part constitutional, and would even more quickly lead to the destruction of the federal judiciary and end of separation of powers, as (if allowed, which, they're also trying to make it logistically difficult to challenge), with the floodgates opened, acts of this form would be increasingly used to force the courts to do whatever the current congress feels like.
A terrible dereliction of their oaths to support the constitution, on the part of the democrats who support all this, and revelatory that all their claims about the importance of protecting our system of government and its norms has the enormous asterisk that they'll destroy it all, if they're inconvenient.
I was curious to see what were the top comments from the last month that didn't make it here, so we have:
@Walterodim, on the "factchecking" of Trump
@functor, on the downfall of the UK
A throwaway, on ideological pressure on doctors
@DTulpa, complaining about the media response to Harris
@raakaa, on the double standard of people being against cancelling Home Depot workers, but putting up with cancelled Olympians
@ABigGuy4U, on biased fact-checkers
@DTulpa, expressing discontent with the disorderly homeless
@functor, on the problems in the UK
@Walterodim, characterizing Tim Walz
@Walterodim, on election integrity concerns with a leftist tinge
@urquan, on women, and the words of the delphic oracle
@gattsuru, on European censorship
@SteveKirk, on Musk and corporate (in)competency
@naraburns, on pro-knifing counterprotests
@Walterodim, putting media activity in the active voice
@self_made_human, on immigration accelarationism
@IGI-111, on the political effectiveness of economic idiocy
And that's the first page, down to a net upvotes of 47. (All but two had at most 4 downvotes.)
The Quality Contribution system posts seem to be considerably more effortpostish, and the ones passed over are more likely to be applause-lights. So it seems like good selection. It's interesting that the slate of posters for each kind of post is not the same—some of us (e.g. @Walterodim) seem to be pretty good at making a not overly long reply that's popular, while not making AAQC-style comments. Others of us effortpost more. Both can be good.
Anyway, thanks to @naraburns for the work in collecting these, and to @ZorbaTHut, that we have this place.
How do you justify this, given that you find the majority of her policy positions abhorrent?
It's the generic (and often unconscious) response to people being uncivilized on the left:
-
We ought to empathize with them, and take seriously whatever motivated them to such actions.
-
We can't put expect anything of them, because they're uncivilized.
-
We can't expect to influence them, because they're uncivilized. (And is it even right to try to sway them from it, given the justifications that they have for it?)
-
Instead, responsibility should be loaded upon those who react too harshly, because they should have known better. And we should feel bad for the victims of the response.
This is precisely the same path that leads people to adopt soft-on-crime prosecutors, and generally punish those who retaliate against the lawless. It happens often when it's easier or involves less unpleasantness for the state to punish those who are otherwise productive, than those who are wild.
This is the default thought pattern that happens when sympathy and responsibility get loaded onto different parties in some conflict. It clearly correlates with seeing things as oppressor (responsible) and oppressed (sympathetic), which is tied to why it's more common on the left, I think.
See Daniel Penny, see the UK riots (and speech arrests), see opinions on cops (when unjustified), etc.
I imagine things will get a lot better for you, if the 2024 election goes to Trump, and worse if it goes to Harris.
Trump's brought it up several times. But no one believes him. You're right that this could probably be pressed in a more serious manner, but merely having Trump repeat it does not work.
See also Montana's failed referendum to protect those children.
Pennsylvania's Commonwealth court just ruled that ballots that are in Allegheny or Philadelphia must be counted, even if they are undated or misdated. This only applies to ballots submitted on time, purportedly. The takes that I've generally seen online are that this is evidence that they have plans for fraud. The court argued this, though, on the grounds that dates are unnecessary, as the counties have other means of telling when votes were submitted (I think they scan a barcode when received). But what's certainly a problem is that this decision was written to apply only to Allegheny (where Pittsburgh is) and Philadelphia counties, the two counties that contribute the largest margin to the democrats. Given that they estimate that around 75% of mail-in ballots are for democrats in Pennsylvania, the most mail-in ballots are from suburban and urban voters, and that around 10000 ballots were not counted for that in 2022, this could have the effect of aiding the democrats by 5000 votes or so. Thankfully, this is only 0.07% of the vote, so not all that likely to be decisive.
The other interesting feature of this case is that the court ignored non-severability provisions, which said that if any provision of the act, or its application to any person or circumstance, was held invalid, the whole act is void. They did so merely by arguing for a presumption of severability in Pennsylvania laws, despite the explicit language to the contrary in this case. Voiding the act would have thrown out the entire mail-ballot system. Them striking down part of it, but not the whole thing, against the explicit text, seems the most sophistic part of the whole thing, to me.
This can still be appealed to the Supreme Court of Pennsylvania. It's blue, though, so I'm not expecting changes. Thankfully, it doesn't seem like, barring fraud, the effect will be too large.
Edit: Make sure you read the comments of @Rov_Scam, where he argues that I'm not representing this accurately or completely—I don't want to be misleading.
Social security's insolvent. There will be automatic cuts in a decade. Expanding lifespans and falling birthrates will make the situation worse. There's no way the program survives long enough for me to get a penny out of it.
Accordingly, your demand that you benefit is actually just saying that you don't want to be the one stuck with the bill—that's for other schmucks. Keep forcing people into the Ponzi, to make sure it's solvent long enough to fulfill the promises it made to you, specifically.
And this despite that it doesn't really promise you any direct reward! There's no accrued payout that's sitting for you on a leger somewhere. That's just the story they tell you to make you think it's reasonable.
It's almost a quarter of our yearly spending.
No. Social security needs to go. At the very least, we should be means-testing. It's unfortunate that there's no political will to touch it.
In practice, though, disparate impact is often used not only for concealed racism, but for any policies that does not produce the desired ratios, where desired means "at least as many nonwhites as the population that we're drawing from." But that can mean that simple hiring based on competence means that people can fall below the definitely-not-quotas, and so they would be legally vulnerable.
In theory, it's a nice feature to capture concealed racism. In practice, it's a generic cudgel that can be used to punish ordinary behavior, should you get out of line. And legally (assuming we're talking about Title VI here), it's a complete fabrication, and is used to make people break the text of the law itself.
Why are you recommending extramarital sex right after invoking your religious affiliation?
For the US senate, you should vote Moreno. Here's why: if the Republicans win the Presidential election, they're basically guaranteed to have the Senate, as the vice president keeps tiebreaks, and so if they win tiebreaks, to lose the Senate, they'd have to lose in all of: (fairly likely) Pennsylvania, Maryland, Nevada, Wisconsin, Michigan, Arizona, (even) Ohio, (30% chance) Montana, and (fairly unlikely) one of Florida or Texas. So whether or not Moreno is elected shouldn't lead to a Republican trifecta, at least, in the next two years.
But if the Democrats win the election, then they win tiebreaks, and so don't need to win Florida or Texas. This makes Sherrod Brown likely to be the critical vote for whatever problematic things they try to pass. So if there are things you don't want them passing (e.g. the supreme court "reforms", bad economic policies, whatever else), it makes sense to vote Moreno.
That is, Ohio's Senate seat disproportionately matters if the Democrats win. Accordingly, if you dislike both agendas, it makes more sense to pick the candidate that will prevent party-line votes should the Democrats win the presidency.
Who will be on the ballot in the swing states?
I figured this is something that's in the news, especially regarding RFK, but generally hasn't been comprehensively compiled. There's a helpful wikipedia page. The election's close enough, given current polling, that the 3rd party candidates could matter.
Arizona: Trump, Harris, Stein (Green Party), Oliver (Libertarian party).
West (ex-Green Party) tried to be a write-in, but failed. RFK successfully dropped out. There's technically Shiva Ayyadurai as a write-in as well (if wikipedia's right), but I never heard of him, and it looks like he's ineligible anyway? Overall, this is pretty typical: Oliver gives unhappy R-leaners an out (though he's socially more progressive than those who care about abortion, for example); Stein gives unhappy D-leaners an out.
Nevada: Trump, Harris, Oliver, Skousen (Constitution party).
This is about as D-friendly of a slate as it gets, at least, in states keeping RFK off. Stein was kicked off on technicalities (the people in the government gave her team a form, but it was missing a field, so it was invalid. The requirement of that field was merely by rules, not even by statute.), per a recent (politically-aligned) supreme court ruling. Nevada has no write-ins (along with 9 other states), so West and Stein will get 0 votes. Skousen's with the state constitution party, but they broke with the national party in nominating him over the official Constitution party candidate (more on that later). He seems to be running as a generic non-Trump Republican (It's an ugly website). Pretty bad arrangement for Republicans, where R-leaners will have other options, but the best option for single-issue Palestine voters is the libertarian candidate.
Georgia: Trump, Harris, Stein, Oliver, West, De La Cruz (Party for Socialism and Liberation).
This is about as R-friendly of a slate as a state gets, which is still only moderately so, as Oliver is on the ballot. The democrats have been trying to get Stein, West, and De La Cruz off the ballot, but it looks like the Secretary of State thinks they've qualified.
North Carolina: Trump, Harris, Stein, Oliver, West, Terry (Constitution Party), and maybe RFK?
North Carolina has the five most common candidates. Terry's the national Constitution Party candidate, unlike Skousen in Nevada. His biggest thing seems to be abortion, which seems like it could actually draw some dissatisfied voters—I'm hoping not. RFK's the big thing—he tried to drop out, but they'd already printed a bunch of early ballots. They've delayed distributing the ballots because of him, but haven't yet decided whether they're going to print new ones, or just go with what they have.
Pennsylvania: Trump, Harris, Stein, Oliver.
Pennsylvania is one of 9 states where there's automatic write-in access. Everywhere else requires the candidate to register. That is, they'll only count your vote for your dad if you happen to live in Alabama, Iowa, New Hampshire, New Jersey, Oregon, Pennsylvania, Rhode Island, Vermont, or Wyoming. I personally find it a bit sad that those votes aren't even counted everywhere else. On the other hand, this must really slow down elections, as they would have to read and tabulate all candidates, presumably, not just those on a set, narrow list?
Anyway, that means that West is available if people really want him, as is literally everyone who's eligible for the presidency. (Courts ruled against him being put on a ballot, because he had filed paperwork for the candidate, not for each of his electors.) Not sure what to make of the radical freedom, but the slate on the ballot itself seems fair.
Michigan: Trump, Harris, Stein, Oliver, West?, RFK?, Terry, Kishore (Socialist Equality Party).
RFK's still going through the process to get himself removed. The Michigan Supreme Court's relatively partisan, though, so I have no expectation that they'll uphold the appeals court: he's probably staying on. West's ballot is also headed to the Supreme Court, over signatures and affidavits and such. Kishore talks about Gaza—I'm wondering if he's trying to appeal to muslims, it being Michigan. RFK should hurt Republicans, I imagine?
Wisconsin: Trump, Harris, Stein, Oliver, West, Terry, De La Cruz, RFK.
RFK's suing to get his name off after the elections commission ruled that he stays on, but I don't expect that to work out. This seems to have candidates on both sides, but keeping on RFK is probably what matters most.
In sum, then, we have:
3 states possibly keeping RFK: Wisconsin, Michigan, North Carolina. (These also happen to be the ones with Terry.)
3 states with no RFK, and with Stein: Arizona, Georgia, Pennsylvania.
1 state with neither: Nevada.
The 1st and 3rd slates are ones Democrats should be happier about, the 2nd are ones that Republicans should be happier about. If this was the sole factor we cared about, Democrats would win, 273-265.
Bonus: a few more marginal states.
Maine has the main five (and so is in category 2. Yes, it is one of the ones RFK chose to withdraw from. One of its electoral votes could go either way.)
Nebraska has the main 5, plus RFK. I'm surprised RFK didn't try to withdraw here; that one electoral vote could matter.
New Hampshire has automatic write-ins like PA, but on the ballot proper only adds Oliver and Stein.
Virginia has the ordinary five, plus De La Cruz.
Florida is unusual. No West, but it has De La Cruz. It has Terry, and the American Solidarity Party's Peter Sonski. The ASP seems to be a pro-life, somewhat anti-capitalist party. Unlike the Constitution Party, a good-looking website. I'm sure they're delightful people, but we'd have different economic views.
Texas has the minimalistic 4 on the ballot, and as write-ins: West, De La Cruz, and Sonski.
I'll probably add in some edits down below here if I hear of any updates. I suppose the overall takeaway here, assuming you're R-leaning, is to see the situation in NV, WI, MI, and maybe NC as slightly worse than you would have thought otherwise, and GA, AZ, and PA as normal.
EDIT: RFK off the ballot in North Carolina, on the ballot in Michigan. West on the ballot in Michigan.
When, if ever, is it appropriate to provide an apologetic defense of Nazi Germany?
Maybe not right before an election, when your preferred VP candidate publicly follows you.
No one's talking about the most important thing: if this goes through, it would be the first time in over 200 years when the sun sets on the British empire.
No, it's bad in the sense that it's evil to scam the people who trust you.
Two points:
First, abortion being a losing issue is way more true post-Dobbs than it was before. I think pro-life people need to focus more on taking wins that they can keep (in a close state, enact the three month ban, not the 6 week ban), because otherwise it'll be extended to birth and more people will be murdered. Yes, this is unjust, but it may be necessary.
Second, the pro-life movement is driven by women to an extent unusual among conservative causes. Women care about babies more, and men sometimes feel out of place—at least, those are the causes, I suspect.
I propose a flat rate of $100,000 per green card. Why wouldn't this work?
Why not auctions? Then you can fix the number of people (what a lot of people care about more), and get more money. 100,000 is too low, anyway.
I'm not sure, though, what I think of changing immigration to only be the wealthy—immigrants stereotypically do a lot of the landscaping and such currently.
I'm in favor of some sort of changes to our current system. Our current way, where companies game the lotteries, and where it's often better to come in illegally and then go for citizenship, is not ideal.
This is an overly simplistic view of the relationship of the old and new covenants. (Accordingly, maybe be a little slower to say those final two sentences of yours, especially since @Quantumfreakonomics demonstrated knowledge of what you talk about by his reference to the Mosaic covenant.)
The following is probably not optimally ordered, but it should present things in a little more detail and clarity (I'd need to do significantly more reading to clarify my thoughts enough to do it better). But first we'll look at the law as laid down at the time of Moses. This can, generally speaking, be classified into three sorts of laws:
- Moral laws, which are articulations of general moral principles. Any appearance in the Mosaic law is just a repetition and reiteration of something that was already true and incumbent upon all mankind. An example is, "Thou shalt not kill."
- Ceremonial laws, which are laws governing Israel in respect of its existing as a church. Unlike the moral laws, these were not incumbent upon people before they were articulated by God. The entire system of sacrifices and laws concerning those would be one example of something in this category.
- Civil laws, which are laws governing Israel in respect of its existing as a polity. These fill the same role as the various civil laws of the United States do for me. An example of this is the establishment of cities of refuge, or any of the penalties attached to any of the laws.
Christ's fulfilling of the law refers to his perfectly keeping it, in its entirety. (Rather than, as James 2:10, describes, breaking it in one point, and becoming guilty of the whole.) He alone has properly measured up to its standard, and so earned, on our behalf, the promised rewards associated with that.
Note that this does not abolish any of the three classes of laws. Moral laws still remain, as we still ought to do good things, and not to do bad things. It's not obvious to me that ceremonial laws would be affected by this—plausibly, there's no reason to stop celebrating the passover, just because someone's fully followed the law. And civil laws still must obviously exist, it still makes sense to punish criminals.
Rather, what Christ's fulfillment of the law accomplishes is meriting for us salvation and the rewards of the law—in the case of the promises to Adam, eternal life. In its application to us, that means that we are no longer dependent upon our fully upholding and fulfilling the law to reap these benefits, we are generously given those by means of our being united to Christ.
What about the three classes of laws, then? Why are they not all here?
First, moral laws stick around. They continue to apply in three sense to Christians. First, they serve to remind us of our dependence upon and estate before God by our failure to keep them. (Consider especially the bar Jesus sets in his articulation of them in the sermon on the mount—one breaks the commandment against adultery not merely by adultery outright, but also by lustful desires, for example.) Second, it can deter evil. Third, it acts as a guide to life. None of these change. (If you want textual proof, 1 Corinthians 9:9 cites a law as authoritative.)
Second, ceremonial laws are changed, because the system of laws that are proper to a people who are largely ethnically a single nation, and which exist in anticipation of and invoke a future savior are no longer so fitting when now the people of God are ethnically diverse, and rather look to the past (and future) coming of God in the flesh, and a corresponding fuller revelation. Read Hebrews in its description of how the old covenant relates to the new, and Paul, in his insistent commands against requiring gentiles to follow the Jewish laws. We still have ceremonial laws, though, of a sort—the command to celebrate the Lord's Supper is one.
Third, civil laws are changed. This is because the polity of Old Testament Israel has collapsed. These were never incumbent upon those not living as a part of the people of Israel, anyway. Instead, we have the various legal codes enacted by governments around the world (many of which are better for that people, in the context and society in which they live, than if they were to replace that law with the Israelite one), which are, as stated in Romans 13, of divine ordination and to be obeyed.
(This isn't a full discussion of everything that could be said—I haven't addressed things through the lenses of covenants properly.)
Applying this to the question at hand. What are we to make of the "Whoever sheds the blood of man, by man shall his blood be shed, for God made man in his image"? Well, the previous discussion was properly speaking, talking about laws for Israel. This is not one of those, being given rather at the time of Noah. It is also a bit ambiguous (at least in translation, I haven't looked at scholarship on this) whether this is a prophecy or general principle, or a command. Let us assume that it is the latter—that is conceded by your treating it as an old testament law.
Well, first, as this is something that is being given immediately after the flood, to Noah, this seems to have in its scope the entire people upon all the earth. This is different from, say, the commandment of circumcision, which applied only to the sons of Abraham, and to all others who wished to join the church of God as it existed then upon the earth. Rather, this is describing something universal. Accordingly, this could not be a ceremonial law—it is not laying down a churchly rite, that is only contingently the case and for the people of the church. It is explicitly grounded in something universal and enduring to mankind (the image of God), and at a time when it would be delivered to the whole of mankind then existing (Noah and family).
"By man shall his blood be shed" is notably in the passive. This does not seem to give any specific entity (or every individual?) the right to avenge murderers. But it does seem to establish that that person ought to be put to death. A reasonable reading of this passage—though certainly not the only one—is that it is incumbent upon the kingdoms of this world to punish murderers with death (but given the cities of refuge, this is clearly not intended to be limitless or without qualification). This reading seems to be supported by later passages. Genesis 4 had described blood crying out. Deuteronomy 21 requires unresolved murders to have an animal slaughtered in atonement. Romans 13 describes the ruler as an avenger, and refers to bearing the sword (not merely prison or the lash). These all at least plausibly indicate that some crimes, especially murders, demand punishment, and that death is a suitable penalty for that. Because of the universality of the Genesis 9 passage, and because there is no identifiable principle why that should cease due to Christ's coming, at least in my judgment—it is not in respect of the chosen people of God, but is universal, and flows from unchanging principles—we should think that it remains in place.
You may not agree with all of this—I'm not certain of everything I said myself, especially regarding Genesis 9:6—but this does require more serious engagement than a simple dismissal of everything prior to the coming of Christ as irrelevant.
In my opinion1, reading refers to the act of getting meaning from lettering using your eyes. But you're certainly still consuming books.
1I resisted the urge to say 'book'.
what makes me useful is that I am willing to actually and consistently do it
And, sincerely, thank you for doing so. It sounds like a lot of work, and it provides a lot of value.
Who cares if the model is accurate?
A lot of people. For one, everyone who is outside the organization and thinks it is accurate, and depends on it as a guide to truth. For another, those in the organization who visualize themselves as trying to predict, which, there are undoubtedly some.
That wouldn't put an end to the problem. NGOs literally fly them into the country.
Oh, absolutely. I also have the same tendency to focus too much on the latter things. I was reading Richard Baxter (a moderate and controversial puritan) two days ago, and found the next three or four pages after where it says section 3 to be pretty convicting. It's okay to be motivated to know about God and to defend his truth, but those shouldn't be the only things.
"I have ever observed that a violent passion called Zeal for a mans opinions, which he counts Orthodox, is so easie and natural, that there needs little means to kindle it: Nay, all the means that can be used will scarce allay the inordinate rage of it: But a Zealous love of God, and delight in him, and a Zeal in him for holiness, and against sin, and a Zealous love to Gods Truth as they hold forth Christ and Glory, and guide us to duty, this is so contrary to the nature of man, that no means is sufficient to excite it. O how easily without Grace, and against Grace do Carnal Ministers, and professors make a huge bussel in the world for their opinions, compassing Sea and Land to make a Proselite!"
And I'll cut the quote off there, even though it continues on with similar force. Do read the whole passage, it's not that long.
But surely corroboration from across the mediterranean should be treated as evidence?
Today, you would end up with a system where you can't fill your gas tank by yourself
See: Oregon, New Jersey
- Prev
- Next
I finally got around to the last case (well, except for the early ones before I started doing these, which I have no intention of returning to):
Moody v. Netchoice and Netchoice v. Paxton
9-0, as to judgment; 6 is the most any one opinion gathers, with Kagan writing, joined by Roberts, Sotomayor, Kavanaugh, and Barrett, and in part, Jackson.
Alito writes for the other three. Thomas, Barrett, and Jackson all write additional opinions.
The case concerns two laws passed in 2021 regulating internet entities, one from Florida, and one from Texas. The laws restricted content moderation, and required explaining the reasoning behind content moderation restricting users. Netchoice is suing that these violate the first amendment. Netchoice is a trade association containing a whole host of large social media companies—Google, Etsy,
TwitterX, and a whole host more—that exists, it seems, largely to lobby or combat laws like the two passed.In both cases, Netchoice sought preliminary injunctions, and in both cases they were granted by the district court. The Eleventh Circuit held that the platforms' decisions were constitutionally protected speech, and accordingly found that the moderation and disclosure requirements were unconstitutional (the latter because it is too burdensome), with the sole exception of the requirement in Florida's laws that the platform disclose the content moderation policy in general (as opposed to in specific cases). The Fifth Circuit held that NetChoice did not try to show it unconstitutional in all applications, held that the restrictions on moderation were not regulating the Platforms' speech, and upheld the disclosure requirements, both general and particular, due to Zauderer having ruled that it was permissible to compel factual information.
The Supreme Court here unanimously agrees that the cases need to be remanded, as courts did not adequately consider that these were facial challenges—that is, that they must consider the law in general, rather than as applied to the plaintiffs. The laws are broad, and they were only addressed narrowly.
I'll start with the smaller opinions this time.
Barrett concurs. She briefly notes at the outset that she agrees with the court that the Eleventh Circuit (which struck down the Florida law) correctly stated 1st amendment precedent, whereas the Fifth circuit (which upheld the Texas law) did not. The remainder of her concurrences emphasizes that this indicates the difficulties of bringing a facial rather than an as-applied challenge. She notes the various difficulties in evaluating these, even were it only as-applied: the first amendment only protects expressive activity. Are decisions made by AI expressive? Further, what if the corporation is located overseas, and so not entitled to First Amendment rights? Considerations like these would more fittingly be applied in specific cases, instead of attempting to evaluate it all in a single facial challenge. So she would prefer they bring it as-applied.
Jackson concurs in part. She thinks that it is generally clear that some things a social media company may do are protected by the first amendment, and others are not, but it is hard to say more here. She agrees with Barrett that the Eleventh Circuit "at least fairly stated" the First Amendment precedent, whereas the Fifth Circuit did not. But, like the whole court, she agrees that they need to reevaluate it in light of it being a facial challenge. Jackson states that the question, in evaluation of this, is not whether corporations as a class, or a particular corporation, is acting constitutionally, nor even whether, e.g. content moderation fits precedent, but it depends on the way that the activities actually function regarding whether they constitute expression. Jackson would decline to look at the ruling on the merits as-applied to the companies, and hence she only joins three parts of the majority opinion: I (the history of the cases), II (the analysis of it being facial, and so requiring remand), and III-A (the account of the first amendment precedent). It seems she does not join Kagan's application of that precedent to these cases.
Thomas concurs in the judgment. He disagrees with the court's decision to give opinions on the applications of those statutes, as this involves some of the same sorts of analysis that they complained about—looking at specific cases, instead of the broad range. He agrees with Alito's analysis. Thomas first notes two additional considerations: the Courts depend on Zauderer, which stated that "laws requiring the disclosure of factual information in commercial advertising may satisfy the First Amendment if the disclosures are reasonably related to the Government's interest in preventing consumer deception." In classic Thomas fashion, he thinks that should be reconsidered, citing an opinion from himself fourteen years prior. Secondly, he notes that he thinks the lower courts should continue to be guided by the common carrier doctrine, which have certain requirements, especially, service of all comers, and that there is historical precedent for regulating transportation and communications networks like traditional common carriers. (Again, citing his past opinions.) The lower courts addressed this in their analysis previously; they should continue to do so, though that cannot really be feasibly be done under a facial analysis.
With that prelude aside, Thomas turns to the main portion of his concurrence, where he argues that facial challenges violate the command of article III. Article III gives courts the power only over "cases" and "controversies." In such particular cases and controversies, it is the place of the courts to say what the law is, but only in those cases. This is necessary to confine the courts to a judicial role. Facial challenges conflict with this because they ask whether statutes constitutionally conflict in cases not before the court. Facial challenges require that no set of circumstances exist under which it would be valid, or in the case of the First amendment, the looser standard that it has prohibits too high of a ratio of protected:plainly unprotected speech. Facial challenges thus ask courts to issue decisions that are unnecessary to decide some particular case or controversy. Plaintiffs are required to show that they personally have suffered an actual or threatened injury, and must be given a remedy that is limited to the injury. Accordingly, the case is done once they have decided whether it is legitimate as-applied. Deciding whether it would be legitimate as applied to other plaintiffs is not necessary, and should be considered as no more than an advisory opinion, which should not be issued. Facial challenges allow challenging applications of statutes that have not injured him, which is ordinarily disallowed. They also allow enjoining of applications of statutes which have nothing to do with his injury, which is not how redressability is supposed to work, and like a universal injunction, which is itself problematic (citing himself and Gorsuch). Facial challenges further intrude upon powers reserved to the Legislative, the Executive, and the States. They allow for the review of constitutionality of applications of a statute before even it has been enforced, giving courts "a general veto power" upon the legislation of Congress, but the Judiciary has no constitutional role in lawmaking. As-applied challenges minimize intrusion. This leaves the Executive branch free to enforce it in other applications. Facial challenges, on the other hand are maximalist, leaving the other branches with no opportunity to correct things, harming the democratic nature of the government. Moreover, facial challenges can prevent the application of state laws in its particular cases, and usurp power from state courts, contrary to the 10th amendment.
Facial challenges also create practical problems. They harm the adversarial system, by allowing plaintiffs to present a challenge without direct knowledge of how the case might apply to others, and so often depend on speculation.
As applied to these cases, the state officials had no opportunity to tailor the enforcement of the laws, nor state legislatures to amend the statutes prior to their enjoinment, nor state courts to interpret the law. Rather federal courts, with little factual record, did. Thomas notes that some blame here is because of associational standing (e.g. NetChoice, instead of the constituent entities) mucking things up—he opposed associational standing in his concurrence over mifeprestone earlier this year—but the facial nature also plays a role in that. The task before them is impossibly complex.
They are also suspect in their origins, being a result of vagueness and overbreadth. At the time of the founding, the courts correctly understood themselves to only decide particular cases or controversies. The founders four times rejected creating a council of revision, which would evaluate and reject statutes, untied from a case. The narrow understanding of what the court could do was generally adhered to for over a century. The first change to this was the development of vagueness—courts began in 1914 to strike down statutes as unconstitutionally indefinite. In 1940, in the First Amendment context, they struck down a statute as "invalid on its face," as it was a "sweeping proscription of freedom of discussion." The court has never justified this overbreadth doctrine in text and history, "just policy considerations and value judgments." This eventually spread elsewhere, without textual or historical justification. Thomas concludes that the court should put an end to facial challenges.
Enough of the minor opinions, now to the two major ones.
Kagan writes for the majority. After introducing and giving the history of the cases (recall: 5th circuit upheld the law, the 11th circuit struck it down), she addresses the facial nature of the case. The court has made facial cases hard to win, ordinarily requiring showing that they are invalid in every application. It is still difficult even in First Amendment cases, like the present one, where the challenger must show that "a substantial number of the law's applications are unconstitutional, judged in relation to the statute's plainly legitimate sweep." But both parties thus far have only been considering them in more narrow applications (like Facebook's news feed), instead of the full range of applications (Among others, is direct messaging covered? Gmail filters? Etsy reviews?). And then it must be considered whether those applications violate the First Amendment: in the case of the content-moderation provisions, whether it intrudes on protected editorial discretion. Regarding individualized-explanation provisions, whether the disclosures unduly burden expression. These issues have not been considered, so they vacate and remand.
Then: "It is necessary to say more about how the First Amendment relates to the laws' content-moderation provisions, to ensure that the facial analysis proceeds on the right path in the courts below," and this is especially needed for the Fifth Circuit, as otherwise, it would just decide the same way, as it wrongly held that they were not speech at all, and wrongly treated Texas' interest as valid.
In essence, Texas' law requires carrying or promotion of speech, when the platforms would rather do the opposite. The platforms argue that this requires changing their expression, and the Court has held in the past that expression includes curation of speech by others. The precedents: In Tornillo, the Court held a law requiring newspapers give candidates the right to reply to criticism violated the First Amendment, in forcing them to print what they would otherwise not. In Pacific Gas & Elec. Co., a utility company distributed views on energy policy in its billing envelopes, the state required including inclusion of material from a different perspective, and the Court sided with the company, as it was not required to carry speech it disagreed with. In Turner I, they held that rules requiring cable operators to allocate channels to local broadcast stations involved the First Amendment. (Even though they eventually decided it was worth restricting anyway in Turner II.) In Hurley, they decided that a parade was free to decide who to admit. On the other hand, they allowed compelled access in two cases: in Pruneyard, they permitted a law compelling a mall to allow people to distribute pamphlets. And in FAIR, they permitted which compelled law schools to allow the military to recruit on campus. In both of these, it was understood that these did not affect expression of the party.
From that can be drawn three principles: The first amendment protects those who compile speech when they wish to exclude some. Secondly, this includes when they exclude just a few items. Third, the government cannot merely assert an interest in balancing the marketplace of ideas.
Things like Facebook's News Feed involve removing or prioritizing content. This is like the cases before, and so is protected. Texas's law thus is problematic. This does not change by the fact that they allow most speech through. Nor does the fact that users can easily tell that it is the users speaking, not the platform, change anything.
Texas' stated interest does not work—promoting ideological balance among private actors is not a legitimate government interest, as it is inherently censorious.
The remaining major opinion is that of Alito:
Alito opens, in the very first paragraph of his introduction, by stating that everything except the facial unconstitutionality of the case is nonbinding dicta. That is, most of Kagan's opinion is extraneous to the matter decided then, can be dismissed, and is of no precedential value. He agrees that it needs further examination as to whether it is facially unconstitutional. But they should not have gone further.
Alito proceeds to a lengthier discussion of the state laws and history of the two cases before they came to the Supreme Court. Then he turns to the facial nature of the challenge: these are strongly disfavored, and conflict with several principles—they clash with the general principle of not reaching beyond what is needed in court decisions, are antidemocratic, and "strain the limits" of the constitutional authority to decide "cases" and "controversies." Accordingly, the requirements are demanding—generally speaking, it fails wherever there is any "plainly legitimate sweep" to the statute. Netchoice asks that this standard not be applied. This is wrong, as the states have asked for the rules of a facial challenge to be applied, and even were it not the case, they would still be necessary for the courts to follow. Netchoice chose to make a facial challenge; now it must deal with the consequences of that choice.
Alito then turns to whether NetChoice manage to show that it is facially constitutional. He begins by reviewing. The First Amendment protects freedom of speech, which usually involves government efforts to forbid, restrict, or compel expression. But some cases have included presenting an edited compilation for the purpose of expressing a message. But not every compilation is expressive. Accordingly, the First Amendment only protects expressive compilations. To show that a compilation is expressive, they must first show that they exercise editorial discretion, and are not, for example, "dumb pipes" that return whatever is put in. This may vary within the same entity, even—newspapers often will exercise far less discretion in advertisements, which may make a meaningful first amendment difference, or a parade might ordinarily only select groups, and not individuals. (Contra the majority, it can matter how much they include vs. exclude, depending on their methodology.)
Additionally, the compilation must be expressing some (often abstract) point. Chronological organization, for example, isn't really expression. It was for this reason that in PruneYard they were willing to compel a mall to host third-party speech.
Third, they must show that their speech is affected by the speech it would be compelled to accommodate. In PruneYard and FAIR they held that the government could compel speech, because speech in those cases was not expressive.
After it has been shown that the first amendment applies, it then has to be shown that the regulation "violates the applicable level of First Amendment scrutiny"—as in Turner, where they decided that it did involve the first amendment, but nevertheless, the government prevailed. Here, they assert an interest in fostering a free and open marketplace of ideas, as well as preventing discrimination against people who live in Texas, which Alito considers compelling interests.
Netchoice failed to meet this burden. First, it did not establish which entities were affected, which make it difficult to know whether a facial challenge should succeed (might it cover websites like WhatsApp that act more like passive receptacles, without curation, and so be legal at least in those cases?). Second, it did not say what kind of content appears, which might be relevant for the first amendment (e.g. is it political? That might matter). Third, it does not show how they moderate. Reddit outsources moderation, which makes it arguably not reddit's speech.
The majority spends much of its opinion specifically talking about how this would apply to the Facebook newsfeed or youtube homepage. Alito points out that this might not even be needed—the fifth circuit can decide on other grounds. He disagrees with their characterization of it as expressive, as they have not revealed how their algorithms were created or work. And they do not consider whether they should be common carriers. And it is not so obvious that what these platforms do in their curation is the same as what editors do—massive scale, post hoc removal, AI algorithms. It also remains to be considered whether "network effects" make any difference. (I'm a little surprised they didn't mention Turner II here, which allowed the government to mandate cable networks carry local stations, despite recognizing that there were first amendment concerns, as otherwise they'd be shut out of the market). He thinks all of these should be resolved in a future as-applied challenge.
Turning to the disclosure questions (saying why messages were censored), they must, under Zauderer, not unduly burden speech. That's hard to know in a facial challenge, and even in the case of YouTube, it doesn't seem like that huge of a burden. This is especially the case for companies that already have to do all that anyway under an EU law.
Let's analyze this in a different direction, by issue.
Should this be remanded due to the facial nature of the challenge? 9-0.
Was the 5th circuit's analysis of editorial discretion problematic? Gorsuch, Alito, and Thomas do not explicitly say; the other 6 are clear that it erred.
For that reason, should the court give further guidance? 5-4 (Alito, Gorsuch, Thomas, Jackson in the minority)
Is the risk of misattribution sometimes the decisive factor as to whether it is protected speech? 5-3, against Gorsuch, Thomas, Alito. (Jackson gives no indication)
Is it ever relevant whether a compiler includes most items and excludes only a few? 6-3, against Gorsuch, Thomas, Alito.
Is an interest in improving or balancing the marketplace of ideas legitimate to compel speech? 6-3, against Gorsuch, Thomas, and Alito.
Does Netchoice's failure to establish which entities are covered mean it fails to show the laws are facially invalid? 9-0.
Is it meaningful that Netchoice has not shown how (delegation of moderation to third parties, like reddit, or the use of algorithms instead of human judgment) it moderates content? At least 5 agree: Alito's opinion, Jackson's (she specifically highlights it), and Barrett's.
Thomas, Gorsuch, and Alito think that the court should at least have addressed the common carrier argument. They also are concerned that there may be further concerns and important differences between editing a newspaper vs. moderating a social media site (size, network effects, algorithmic vs human moderation).
Some additional thoughts:
People, especially on the left, think of Thomas and the court as corrupt. Aside from the extent to which that is bolstered by politically motivated slander, I don't think they realize that Thomas wants to constrain judicial power, seeing its current state as beyond constitutional bounds. Compare to the complaints of Jackson in Trump v. United States, released on the same day, that the court was expanding its own power. And it isn't unique to this case; he's also expressed doubts about certain conventional forms of standing, and a few other sorts of things. I'd imagine Gorsuch concurs with Thomas on a lot of this, though not on everything he said (e.g. I don't think Gorsuch is much of a fan of prosecutors tailoring laws being important, due to fair notice concerns—it's not good, and unfair to the citizenry, to have a bunch of broad laws erratically enforced).
All of these opinions depended heavily on precedent. I have no idea what originalist methodology would say here.
I also don't think I got a sufficiently clear view from either party as to what constitutes expression and what does not. Nor do they really deal with common-carrier questions, which are very relevant, I think.
There have been revelations (though some already speculated before the release of those) that Alito originally had the majority in this case, with Barrett and Jackson siding with him, but they eventually left.
Practically speaking, I'm not sure what's the best policy-wise. With the acquisition of Twitter, it doesn't feel like conservatives are at quite as much of an ideological disadvantage. It seems risky to allow government intervention in speech. I also don't know about jurisdictional questions—does it make sense for states to be able to regulate a shared platform? What if they do so in different directions? Might those who believe in the dormant commerce clause (that is, states being unable to regulate interstate commerce) think it applies here?
More options
Context Copy link