Non-academic jobs for PhDs saves the academic jobs for PhDs.

And any academic who looks down on non-academic PhDs jobs should just be dismissed out of hand for either being deeply insecure or repulsively snobby.

I was always convinced that one reason econ students at Chicago had one of the fastest graduation times was because we knew we could keep doing econ once we left.

If you were in, say, anthro, graduation likely meant the end of your anthropology career. So incentive to linger.

There are many reasons that, say, econ PhD programs are never on the chopping block. I'm sure many of them are "neoliberal" and "bad."

But a big one? There are a LOT of prestigious non-academic jobs out there for economists, so it's a lot easier to find a job. FAR less risky to enroll.

I'd add, it's suicidal for departments not to WILDLY embrace this idea.

We ALL know that the number of academic jobs can never support the number grad students departments want.

If you want students to want PhDs, you should be actively finding ways for them to ... you know ... USE them.

New definition of "diplomacy" just dropped.

Apparently this is what passes as a Foreign Service Officer now?

I've never said the ppl in the Philly suburbs shouldn't vote. Hell, I think it should close to impossible to deny any citizen the ability to vote.

I just want the suburbs to vote for suburban DAs, who can do whatever they want to suburbanites. But Winetka doesn't need any say in Chicago cases.

Best interpretation I can have of this is that I've said I have wanted to disempower suburban voters from having a say in urban crime control.

But not by, you know, actually disenfranchising them! Just by redrawing maps so big cities are all their own counties (like Philly).

What about mocking "the public" for disagreeing with the educated elites? Here are a few examples- @JohnFPfaff stating that the "low information high salience electorate" should not be allowed to vote, and @JonHaidt saying parents are "fetishizing safety" if they protect kids

I had to log into Twitter for a minute to see what some pol was saying, checked my notifications, and ...

... I don't think people are doing okay over there.

This is, I know, not news.

Yeah, I definitely appreciate the problem. But that doesn't make the problems with the statistic we do use any less problematic.

I'm not saying there's a better stat easily in reach. It's hard! But I also think too many use our official stat as a measure of reOFFENDING, which it isn't.

Def an improvement, but even that isn't great when we can only see the contact with the system, not the criminal acts (or their absence) themselves.

It's exactly that, with I think really big consequences for how we think about recidivism and success/failure overall.

Again, I think prisons are generally giant net social negatives. I'm not here to defend them.

But I'm also not here to use statistics poorly to argue against them, and our "how many people ever trip up again and get caught again" metric of "failure" is a bad one.

But cessation is an unlikely goal, however laudable it is. Desistance--an imperfect decline in violent and anti-social behavior--is what the real practical goal should be.

And looking just at "any one eventual subsequent contact with the criminal legal system" completely overlooks that.

So saying "prison doesn't work because a large percent of ppl get rearrested or readmitted" is proof that prison doesn't work only if we define success solely as CESSATION, not DESISTANCE.

(And even then, it's cessation-from-getting-rearrested, not doing more bad stuff we can't see.)

Joe commits a robbery a week, goes to prison, comes out, now commits two robberies a week.

That's even worse! He's recidivating more!

Joe, too, will get rearrested and reconvicted.

But, in our data? Will look IDENTICAL to Bob. Only possible difference is "time to first rearrest" (Joe is faster).

Bob is going to get rearrested and likely readmitted. "A recidivist failure."

And at one level he is. But also? Going to prison (in my example, not necessarily reality!) cut offending by 75%. We just can't see it in the data, bc we only see the post-contact stuff.

It gets worse:

The problem is what "recidivism" measures: any future criminal legal system contact of some level (arrest, conviction, re-incarceration, etc.).

But consider the following case:

Bob commits one robbery per week. Gets caught, goes to prison, leaves, now commits one robbery per month.

Thinking about the claim I see that prison doesn't work bc a high percent (~75%) recidivate. (Have read it several times tonight for something.)

I'm not here to defend prisons, at all, but I am here to criticize how we think about recidivism, and this is a good ex of how we use that # poorly.

Honestly, I sometimes see that as a feature, not a bug. At some point it would be big enough that it would involve rotating panels, at which point the returns on packing drop.

I get that rotating panels on a highest court could = chaos, but I'm feel like it has its upsides as well.

What counts as an illegitimate SCOTUS holding?

That's hard! Forces pundits to actually think abt the substance of things--and for the centrist pundit, forces them to actually commit to substantive values other than "in the middle of those guys."

Just counting all defiances the same is easier.

SCOTUS rulings can, in fact, be substantively unconstitutional. Or can reflect a clear violation of separation of powers, which is procedurally unconstitutional.

A POTUS who rejects such SCOTUS holdings is acting very differently from one who rejects legitimate SCOTUS holdings.

I'm more optimistic about future SCOTUS expansion/jurisdiction stripping/etc., but:

This is a good reminder that too many pundits view "rule of law" as a purely procedural commitment, so the substance of what executive defiance is over is largely irrelevant.

This has worked well for ICE--which can't fill its open positions and has lots of people flunking out of training--so I'm sure it'll turn out fine for the FBI and DOJ.

MAGA is learning that while they hate the state, they need state capacity to do the harms they want to do, and they've gutted that.

Please, Ninth Circuit, do this.

In what I guess count as realist terms, this seems to me to be a way of saying to the lower courts "we see what you're doing to try to limit QI, and we don't like it, cut it out," without actually having to like, say anything, yeah?

Yeah: we'll reverse you with fancy citations so it looks like we are doing our job, but really we're just overriding you bc we disagree with your factual findings. And as cowards by not signing out names.

Zorn is here, not laying out any sort of new rule: www.supremecourt.gov/opinions/25p...

AFAICT, "further consideration with Zorn" just means "nah, we're fine with this killing" but made to look like it is some sort of new reconsideration, not just second-guessing.

All this fancy legalese and citations, just to give legal cover to police gratuitously indulging in lethal violence.

Also, Zorn v Linton is a brief 5-page per curiam opinion that basically just restates the (preposterously infuriating) rules of qualified immunity (which any future Dem trifecta should abolish in a heartbeat).

It also involves allegedly improper pressure on the wrist, not ... killing someone.

This.

This is what truly angry fed judges have to do. Undermine the protection these thugs have from a Redeemer DOJ that revels in their cruelty.

Will appellate courts reverse? Maybe. But some cases will get thru, and if they do in a place with a reformer DA, charges Trump can't pardon can follow.