Coyote Blog

Is there some sort of metric for the complexity of a boundary like this one for PA Distric 7, among those invalidated in PA (in a decision the Supreme Court today refused to review)?

I keep wondering whether there is an objective standard we can set, rather just the sort of "I know it when I see it" one that everyone seems to use.

One way I can imagine testing is to do it Monte Carlo style by drawing a series of lines from one random point in the shape to another random point in the shape and calculating which percentage of the lines cross the boundary at least once.  That metric for a circle or a rectangle would be zero, but would be very high for this shape.

One of the homilies one hears all the time from economists is "Never reason from a price change."  What does this mean?  Prices emerge in the market at the intersection of the supply and demand curve.  Often, when (say) a price of a commodity like oil decreases, pundits might reason that the demand for oil has suddenly dropped.  But they don't necessarily know that, not without information other than just the price change.  The price could have dropped because of a shift in the supply curve or the demand curve, or perhaps some combination of both.  We can't know just from the price change.

Which gets me thinking about the last election.  Trump won the election in part because several states like PA and WI, which had been safe Democratic wins in the last several elections, shifted to voting Republican.  Reasoning from this shift, pundits have poured forth today with torrents of bloviation about revolutionary changes in how groups like midwestern white males are voting.  But all these pundits were way wrong yesterday, so why would we expect them to suddenly be right today?  In my mind they are making the same mistake as reasoning from a price change, because the shift in relative party fortunes in a number of states could be because Trump is somehow doing better than Romney and McCain, or it could be because Clinton is doing worse than Obama.  Without other information, it is just as likely the story of the election is about a Clinton loss, not a Trump win.

Republican pundits want to think that they are riding some sort of revolutionary wave in the country.  Democratic pundits don't want to admit their candidate was really weak and like how they can spin white supremacist story lines out of the narrative that Trump won on the backs of angry white men.

The only way we can know the true story is to get more data than just the fact of the shift.  Let's go to Ramesh Ponnuru (and Kevin Drum from the other side of the political aisle makes many of the same points here and here).

The exit polls are remarkable. Would you believe that Mitt Romney won a greater percentage of the white vote than Donald Trump? Mitt took 59 percent while Trump won 58 percent. Would you believe that Trump improved the GOP’s position with black and Hispanic voters? Obama won 93 percent of the black vote. Hillary won 88 percent. Obama won 71 percent of the Latino vote. Hillary won 65 percent.

Critically, millions of minority voters apparently stayed home. Trump’s total vote is likely to land somewhere between John McCain’s and Romney’s (and well short of George W. Bush’s 2004 total), while the Democrats have lost almost 10 million voters since 2008. And all this happened even as Democrats doubled-down on their own identity politics. Black Lives Matter went from a fringe movement to the Democratic mainstream in the blink of an eye. Radical sexual politics were mainstreamed even faster. White voters responded mainly by voting in the same or lesser numbers as the last three presidential elections. That’s not a “whitelash,” it’s consistency.

As I know all too well, a portion of Trump’s online support is viciously racist. Conservative and liberal Americans can and must exercise extreme vigilance to insure that not one alt-right “thinker” has a place in the Trump administration, but it’s simply wrong to attribute Trump’s win to some form of great white wave. Trump won because minority voters let him win. The numbers don’t lie. The “coalition of the ascendant” stayed home.

Trump had roughly the same vote totals as Romney and McCain, and did relatively better with non-whites and Hispanics.   The difference in the election was not any particular enthusiasm for Trump, and certainly not any unique white enthusiasm, but a total lack of enthusiasm for Hillary Clinton.   Look at the numbers in Drum's post -- Hillary did worse with every group.  For god sakes, she did 5 points worse than Obama with unmarried women, the Lena Dunham crowd that theoretically should have been her core constituency.  She did 8 points worse than Obama with Latino women!

This is not a story of a Trump revolution.  This is a story of a loss by a really weak Clinton.  Obama would have dusted the floor with Trump.

Presented without comment (source).

The source is an agency in PA so I don't know if this is a state phenomenon there or if this is US data

The title is from a quote by Representative Peter King, lamenting that the Feds might actually have the gall to Mirandize an American citizen who has been arrested.

Miranda warnings are not actually required per se, but statements by un-Mirandized suspects will generally be tossed out of court.  So I think the whole Miranda thing is overblown, but Representative King's words tend to summarize for me the slippery slope of civil rights exceptions.  Someday, you too may be the "but still."

On top of Mr. King's statement and our local Sheriff's continuing proclivity to walk into businesses and zip-tie everyone with brown skin until they can produce birth certifications  (yes, he did it again last month and again the other day) comes Joe Lieberman's new bit of law and order paranoia.  Apparently, after watching a 24-hour festival of 1970's Clint Eastwood and Charles Bronson crime dramas, Lieberman proposed:

Sens. Joe Lieberman (I-CT) and Scott Brown (R-MA), joined on the House side by Reps. Jason Altmire (D-PA) and Charlie Dent (R-PA), today introduced a little publicity stunt in legislative form called the Terrorist Expatriation Act, making good on Lieberman's pledge to find a way to strip the citizenship of Americans"”whether naturalized or native born"”who are suspected of aiding terrorist groups. It does so by amending the Immigration and Nationality Act, which lays out the various conditions under which a person may renounce or be deprived of citizenship....

Finally, note that the bill's definition of "material support" for terrorist groups explicitly invokes the criminal statute covering such actions.  Which is to say, revocation of citizenship under the new bill is triggered by committing a particular federal crime. Except that the Immigration and Nationality Act only requires that one of the predicates for revocation be established by a "preponderance of the evidence." So in effect, the bill takes what is already a crime and says: Proof of guilt "beyond a reasonable doubt" is no longer a prerequisite for the imposition of punishment for this crime.

"Preponderance of the evidence" is the same standard under which McDonald's lost several million dollars because its coffee was too hot.

Every one of these members of the House Subcommittee on Commerce, Trade, and Consumer Protection voted to pass this absurd law out of committee except Rep. John Barrow, D-Ga.

Bobby L. Rush, Illinois, Chairman

Jan Schakowsky, IL, Vice Chair	George Radanovich, CA, Ranking Member
John P. Sarbanes, MD	Cliff Stearns, FL
Betty Sutton, OH	Ed Whitfield, KY
Frank Pallone, Jr., NJ	Joseph R. Pitts, PA
Bart Gordon, TN	Mary Bono Mack, CA
Bart Stupak, MI	Lee Terry, NE
Gene Green, TX	Sue Wilkins Myrick, NC
Charles A. Gonzalez, TX	John Sullivan, OK
Anthony D. Weiner, NY	Tim Murphy, PA
Jim Matheson, UT	Phil Gingrey, GA
G. K. Butterfield, NC	Steve Scalise, LA
John Barrow, GA (voted NO!)
Doris O. Matsui, CA
Kathy Castor, FL
Zachary T. Space, OH
Bruce L. Braley, IA
Diana DeGette, CO

Hat tip: Don Boudreaux

Romney-care, which looks surprisingly like v1.0 of Hillarycare, is over-budget by $135 million (and counting*) in Massachusetts.  TJIC is stunned.

* must be pronounced in that deep eastern European voice from the early James Bond films, where in the bad guy's lair the PA system kept saying "one minute, and counting."

Conservative pundits often observe that "this is a new type of war -- shouldn't the president have new powers to fight it?"  Well, maybe.  But I think there is a question that is at least as valid:  "Given that enemy combatants don't wear uniforms any more, shouldn't we exercise more care than in the past in how we designate people as combatants?"  The much greater ambiguity in naming combatants would seem to demand extra layers of process protection and appeal rights for such persons.

Unfortunately, this Administration, with the aid and comfort of the US Congress, has gone exactly in the opposite direction.  As Jim Bovard writes, via Cato-at-Liberty:

The MCA awarded Bush the power to label anyone on earth an enemy
combatant and lock then up in perpetuity, nullifying the habeas corpus
provision of the Constitution and "turning back the clock 800 years,"
as Sen. Arlen Specter (R-PA) said. While only foreigners can be tried
before military tribunals, Americans accused of being enemy combatants
can be detained indefinitely without charges and without appeal. Even
though the Pentagon has effectively admitted that many of the people
detained at Guantanamo were wrongfully seized and held, the MCA
presumes that the president of the United States is both omniscient and
always fair.

Sixty years ago, when the military hauled in a guy dressed in a gray Wermacht uniform captured in the Ardennes Forest, you kindof gave them the benefit of the doubt that he was an enemy combatant.  How long until merely exercising free speech rights in favor of a terrorist group gets one labeled a "combatant."

While the "she was asking for it" defense has thankfully been purged from most rape trials (at least those involving strangers), it seems to be alive and well in the civil trial world.  Last week, a jury held that the terrorists who bombed the World Trade Center in 1993 were only 32% responsible for their actions.  The real villain in this terrorist attack was ... the Port Authority, owner of the facility, who so thoughtlessly allowed themselves to get bombed.  More via Volokh and Overlawyered.  Based on joint and several liability, the PA now is on the hook for the entire $1.8 billion verdict.

By the way, the "smoking gun" in the trial was apparently a recommendation the PA received (one of hundreds and perhaps thousands of suggestions of wildly varying quality) to close the parking lot to cars to prevent car bombs.  This helps reinforce my earlier point of why litigation insanity like this actually works to make the world less safe, because such litigation provides a strong disincentive for an entity to have any internal discourse on safety, since notes from this discourse can be held against it later. 

It is always useful to think about what consistently applied policy would have satisfied the jury that the PA was not liable.  In this case, the jury's verdict was clearly "they should have closed the garage to prevent car bombings."  Now, lets apply that everywhere consistently.  This would basically mean that we close every car parking garage in the country, since they are all equally vulnerable to a car bomb.  Applying this further, wouldn't this same standard also result in closing all tall buildings to prevent airplane attack, closing all airports to prevent hijackings, and closing all government buildings to prevent bombings (well, maybe thats not so bad).  I have posted before about finding the absurdity from translating a jury's civil verdict into a consistent policy.  Here is one example:

the exact wording on the complaint against the railroad is even better than I thought:

"The
[engineer] did not stop the train in a timely manner, and failed to
yield the right of way to a pedestrian walking along the tracks in
plain view"

A freight train's topping distance is measured in miles, even with full emergency braking.

She and her attorney's further argue:

that
the railroad was negligent for failing to post signs warning 'of the
dangers of walking near train tracks and that the tracks were actively
in useLets

leave aside the obvious point
about individual responsibility, and ask what would happen if this were
the legal standard, to have such signs.  To make sure someone saw one,
you would have to have one say every 30 feet.  Since there are just over 200,000 miles of freight railroads in the North America that works out to a bit over 35,000,000 signs that need to be posted.  At $100 per sign this would cost $3.5 billion.

Here is the serious point:  Never would any legislature
pass a law that said there had to be warning signs every 30 feet on
railroads.  It would be way too costly for little benefit.  At grade
crossings today, we have signs and flashing lights and even gates and
still thousands of people a year drive in front of trains on grade
crossings.  So, if we would never require it legislatively, how have we
gotten to a point where a jury might effectively retroactively require
such signs, and assess a multi-million dollar penalty for not doing it?