The Ghost Ship, the Wrong Submarine, and the USS Stewart That Wouldn’t Stay Sunk

When I lived on Galveston Island, I frequently toured the USS Cavalla (SS-244), a World War II submarine on display at Seawolf Park.  This produces one of those small historical misunderstandings that only a tourist attraction can create.  Because the submarine is in Seawolf Park, many visitors naturally leave with the impression that they have toured the USS Seawolf. This is impossible, unless the Navy has developed a very aggressive museum-restoration program involving recovery from the bottom of the Pacific, because USS Seawolf (SS-197) was lost in 1944, most likely to friendly fire.  The Galveston Naval Museum identifies the submarine on display as the Cavalla, and the destroyer escort beside her as USS Stewart (DE-238).

The friendly-fire part of the Seawolf story is one of those grim little naval footnotes that make history feel less like a marble monument and more like a dimly lit office with too many filing cabinets.  The evidence indicates that Seawolf was probably sunk by the destroyer escort USS Richard M. Rowell after failing to respond properly during a tense anti-submarine search.  The Navy did not exactly put “We may have sunk our own submarine” on a recruiting poster, but the friendly-fire explanation was part of postwar accounting rather than a secret locked away until the age of the internet.  The Naval History and Heritage Command say the evidence suggests friendly fire was the most likely cause.

Still, I was not there primarily to solve the Seawolf confusion. I was there to tour the Cavalla, the submarine that sank the Japanese aircraft carrier Shōkaku, a veteran of the Pearl Harbor attack.  That is a pretty good résumé.  Most museum ships can say they served their country, the Cavalla can say she helped send one of the Pearl Harbor carriers to the bottom.  That gives the tour a certain edge.  You are not just ducking through hatches and trying not to bang your head; you are walking through a machine that once changed the balance sheet of the Pacific War.

But I always ended the visit by touring the ship tied up beside her: the USS Stewart (DE-238), an Edsall-class destroyer escort.  Destroyer escorts were not glamorous in the way battleships and carriers were glamorous.  They did not get much help from Hollywood.  They were the practical shoes of naval warfare: sturdy, necessary, and unlikely to be featured in a recruiting poster unless all the battleships were busy.  The Galveston Naval Museum says Stewart is one of only two remaining destroyer escorts in the United States, and the only surviving Edsall-class destroyer escort.

Reading up on Stewart, I discovered that there were actually three U.S. Navy ships named USS Stewart, all named for Rear Admiral Charles Stewart, who commanded the USS Constitution during the War of 1812.  This is where the story stops being merely interesting and starts behaving like it was written by a screenwriter who had been told, “Make it weirder, but keep the ships real.”

The first USS Stewart (DD-13) was one of the Navy’s earliest destroyers, a Bainbridge-class vessel from the dawn of the destroyer age.  She was small, narrow, fast for her time, and armed with the kind of optimism that early destroyers required.  In World War I, she escorted convoys off France, and even attacked the German submarine U-108 in 1918.  The Naval History and Heritage Command photo caption notes that Stewart’s funnel carried a star signifying that she had sunk or disabled a German submarine, though later evidence showed U-108 survived.

The second USS Stewart (DD-224) had the truly fabulous career, by which I mean a career that included almost every indignity short of being converted into a floating seafood restaurant.  She was a Clemson-class “four-stacker,” commissioned in 1920, and by World War II she was already an elderly antique destroyer in the Asiatic Fleet.  In peacetime, she had done the usual imperial-era chores: showing the flag, visiting China, rescuing people after the Kanto earthquake, patrolling rivers, and reminding everyone that the United States Navy could appear in your harbor whether or not you had invited it.  In 1938, Stewart even helped search for the missing Pan Am Hawaii Clipper, which vanished between Guam and Manila.  According to one timeline, Stewart left Manila on July 30 to search for the missing flying boat and was ordered to abandon the search on August 6.

The Hawaii Clipper mystery deserves its own shelf in the library of prewar weirdness.  The aircraft disappeared with fifteen people aboard, and no confirmed wreckage was ever found.  One rumor held that passenger Wah Sun Choy, also known as Watson Choy, was carrying millions in gold certificates intended for Chiang Kai-shek and the Chinese Nationalists.  That led to theories that Japanese agents hijacked or destroyed the plane, something Japan vigorously denies, but we shouldn’t let that get in the way of a good story.

Then came World War II, and Stewart’s life became genuinely strange.  In February 1942, during the desperate defense of the Dutch East Indies, Stewart was damaged in battle and made it to Surabaya, Java, for repairs.  There, in one of those moments that makes a sailor consider changing careers, she slipped off the blocks in a floating dry-dock and bent her propeller shafts.  With Japanese forces closing in, the Americans destroyed the ship with demolition charges, scuttled the dry-dock, and left her for dead.  The Navy struck her from the list in March 1942.That should have been the end of it.

It was not the end of it.

After nearly a year underwater, the Japanese raised her, repaired her, and commissioned her as Patrol Boat No. 102.  This is where Stewart became the “Ghost Ship of the Pacific.” Allied pilots began reporting the extremely awkward sight of what looked like an old American four-stack destroyer operating deep behind enemy lines. 

One can imagine the debriefing.  “You saw what?”  “An American destroyer.” “Where?”  “In Japanese waters.” “Have you been sleeping?”  “No.”  “Would you like to start?”  The Naval History and Heritage Command says multiple Allied pilots reported seeing the ship behind enemy lines after the Japanese commissioned her as Patrol Boat No. 102.

At war’s end, American occupation forces found the battered former Stewart afloat near Japan.  In a ceremony that was either touching, bizarre, or both, the U.S. Navy recommissioned her in October 1945 as DD-224.  Her crew nicknamed her RAMP-224, borrowing the language used for Recovered Allied Military Personnel, as if the ship herself had been a prisoner of war.  This is sentimental, absurd, and somehow exactly right.  Ships are just steel until sailors start talking about them; after that, they become characters.

The Navy brought Stewart back to San Francisco, but there was no real future for an old four-stacker that had served both sides, been sunk, raised, captured, recovered, and insulted by every ocean she met.  On May 24, 1946, she was used as a target ship and sunk off the California coast.  Even then, she did not go quietly.  Reports say she absorbed rockets, machine-gun fire, and naval gunfire for more than two hours before finally going down.  Some ships are sunk, Stewart had to be persuaded.

For decades, that was the end of the story.  Then, in August 2024, undersea searchers found the wreck of USS Stewart (DD-224) in the Cordell Bank National Marine Sanctuary off northern California.  She lies in about 3,500 feet of water, largely intact and nearly upright.  The National WWII Museum says the discovery was made by a team including Ocean Infinity, the Air/Sea Heritage Foundation, SEARCH, NOAA’s Office of National Marine Sanctuaries, and the Naval History and Heritage Command.

So there she rests: an American destroyer, a Japanese patrol boat, an American destroyer again, and finally a ghost on the seafloor.  If one is willing to be mischievous, she may be the closest wreck of a Japanese warship — sort of — to San Francisco.  Legally, historically, and emotionally, that statement requires several footnotes and possibly a naval lawyer.  But as a punchline, it is irresistible.

The next time someone visits Seawolf Park and says they toured the Seawolf, let them down gently.  They toured Cavalla, which sank Shōkaku, and Stewart, whose predecessor had one of the strangest careers in naval history.  That is not a disappointment.  That is an upgrade.  After all, any ship can have a service record.  Very few can say they served two navies, died twice, came home, and still managed to become a ghost story.

New York and the Search for the Last Taxpayer

There is an old joke that New York City could tax oxygen if only the air could be properly assessed and invoiced.  Given recent political trends, I fully expect a future press conference announcing the “Progressive Atmospheric Equity Contribution Fee,” payable quarterly by anyone breathing south of Yonkers.

This raises a serious question hidden beneath the humor: At what point does New York’s increasingly enthusiastic effort to tax the rich begin to resemble a man trying to drain a sinking boat by drilling new holes in the bottom?

To hear some people tell it, the rich are already fleeing New York in biblical caravans.  Hedge fund managers are supposedly racing across the Florida border in armored Bentleys while investment bankers rappel out of Midtown office towers carrying sacks of untaxed capital gains.  Somewhere in Palm Beach, according to this narrative, there is now a gated community populated entirely by former Upper East Side residents wearing linen suits and complaining about how hard it is to find decent bagels.

The truth, as always, is less cinematic and more interesting.

There is, in fact, real evidence that wealthy people have been leaving New York City.  New York’s own tax data show that millionaire out-migration increased sharply during and after COVID, with relocation rates rising well above historic norms.  Some of the ultrarich, especially those earning tens of millions annually, clearly decided that sunshine, lower taxes, and fewer regulations sounded preferable to paying both New York State and New York City income taxes while carefully navigating feces-laden sidewalks on the way to dinner.

And honestly, from a purely mathematical standpoint, one can understand their concern.  New York State already imposes one of the highest income tax burdens in the country.  Add New York City’s local income tax on top, then pile on property taxes, corporate taxes, sales taxes, mansion taxes, congestion pricing, assorted fees, and Mayor Mamdani’s threat of a pied-à-terre tax, and eventually even a billionaire may begin quietly Googling “residency requirements in Florida.”

Florida, meanwhile, waits offshore like a giant tax-free aircraft carrier.

No state income tax.  Warm weather.  Palm trees.  Private airports.  And an endless supply of real estate agents whispering, “Sir, your taxes alone could pay for this waterfront estate.”  New York tax officials admit almost 1,700 millionaires moved their tax address out of New York in just 2024.  New York hasn’t released the data for 2025 or 2026, but I doubt that many moved back.

Not surprisingly, Miami and Palm Beach have become popular landing zones for finance executives and wealthy retirees.  Texas has benefited too.  Wealth migration toward lower-tax states is real enough that entire industries now exist to help wealthy individuals establish legal residency elsewhere while keeping one tasteful Gucci loafer still planted in Manhattan.

But before we declare New York a post-apocalyptic wasteland populated only by rats and deranged graduate students, it is worth noting that the “everyone is fleeing” story is also wildly exaggerated.  New York is not in the red-light danger zone, but it is in the warning orange zone.  The NYC Comptroller says the city is already operating with a structural deficit, meaning spending is already running ahead of recurring revenue, and the budget relies on optimistic revenue projections, reserve drawdowns, unspecified savings, and reduced fiscal flexibility.  The state comptroller also warned that the city’s budget reduced contingency reserves, including the general reserve, down to the $100 million statutory minimum.

New York remains one of the most economically powerful cities on earth.  It still dominates finance, media, publishing, fashion, advertising, law, and international business.  People continue to move there because it offers opportunities unavailable almost anywhere else.  The city still attracts massive tourism, investment, and foreign capital.  And despite all the horror stories, the overall tax base has not collapsed.  Official projections still show growing tax revenue in coming years, although that may be offset by predictions of even faster growing expenditures.

This is because wealthy people are often less mobile than politicians and cable news hosts imagine.  Moving is not just a tax decision—it involves business networks, schools, family ties, social status, office locations, cultural institutions, and personal identity.  A hedge fund manager may enjoy saving millions on taxes in Miami, but he may also discover that his entire professional ecosystem still functions in Manhattan.

In other words, it turns out that civilization is annoyingly sticky.  But, with each arrival of a new millionaire or business in a Southern state, a small part of that missing social infrastructure is re-established, making it easier for the next hedge fund manager to set up shop. 

Still, New York faces a genuine long-term risk, and it is not necessarily the dramatic overnight collapse people imagine.  The real danger is something slower and far more bureaucratic: a gradual erosion of the tax base combined with increasingly optimistic government spending.

This is where economics stops being exciting and becomes terrifying.

Suppose Mamdani inevitably announces yet another new “tax the rich” proposal that is anticipated to raise $500 million annually.  Headlines celebrate.  Advocacy groups cheer.  Editorial boards declare that fairness has finally arrived.

Then reality intervenes.

More wealthy residents leave.  Others restructure income.  Investments are delayed.  Real estate transactions slow.  Businesses expand elsewhere. Capital gains are realized in different states.  Accountants suddenly become the most powerful people in America.

Instead of raising $500 million, the tax brings in $300 million.  Unfortunately, by this point the government has already spent the imaginary $500 million three times over and created six new agencies to administer it.

Now there is a budget gap.

The response, naturally, is to propose another tax, and triggering even more capital flight.

This is how cities wander into fiscal quicksand.  Not through a dramatic catastrophe, but through an endless cycle of optimistic revenue forecasts colliding with human behavior.  This is how Detroit, St. Louis, and Philadelphia triggered rapid economic decline.

The core problem for New York is concentration.  A tiny percentage of taxpayers provide an enormous share of tax revenue.  Millionaires account for a massive portion of New York’s income tax collections.  This means the city’s financial health increasingly depends on the continued willingness of a relatively small number of highly mobile people to remain exactly where they are and continue earning exactly as much money as before.

That is not a stable long-term strategy.  It resembles balancing the city budget atop a stack of champagne glasses.

And yet, the political incentives always favor more taxation because the immediate math looks irresistible. If one billionaire pays millions annually in taxes, then taxing him a little more appears painless.  Multiply that across thousands of wealthy taxpayers and politicians see visions of balanced budgets dancing in their heads.

The trouble is that economists are forced to deal with the horrifying reality that human beings react to incentives.

Raise cigarette taxes and fewer people smoke.  Raise gasoline prices and people drive less.  Raise taxes on capital and wealth, and eventually some capital and wealth relocate.

This should not be controversial, but somehow every generation of politicians acts shocked when it happens.

The final irony is that if New York ever truly succeeded in driving out large numbers of wealthy taxpayers, the burden would not vanish into thin air.  It would simply shift downward onto the remaining middle class, homeowners, renters, and businesses.  By the time officials realized the rich did not produce the projected tax revenue after all, the spending commitments would already exist.

And then a new politician would arise with a fresh PowerPoint presentation explaining why one more tax increase will finally solve everything forever.

At which point the last remaining taxpayer in Manhattan will quietly board a flight to Miami carrying nothing but a laptop, a residency affidavit, and a deep appreciation for palm trees.

The Mystery of the Lost Mysteries

I have always loved mysteries, probably because I worked my way through the entire Hardy Boys series in the second grade.  I was hooked early by secret panels, hidden staircases, stolen jewels, mysterious strangers, coded messages, and the absolute certainty that two boys with flashlights could outwit every adult in town.

Eventually, I moved on to Ellery Queen, Agatha Christie, Rex Stout, Lawrence Block, John D.  MacDonald, and a host of other great mystery writers.  The best of them understood something important: a mystery is not just a crime story.  It is a puzzle, a game, a contest between the writer and the reader.  The writer lays out the clues, hides the truth in plain sight, and then dares the reader to arrive at the solution first.

Good mystery books are still being written, and it is not hard to find one.  If you are looking for your next mystery, I would recommend subscribing to the newsletter from The Mysterious Bookstore, the “World’s Oldest and Greatest Mystery Fiction Specialty Store,” run by Otto Penzler.  You can find them here.

What I cannot find, or at least cannot find very often, is a good mystery movie or television mystery.  There are plenty of television shows and movies that call themselves mysteries, but most of them are not mysteries at all.  They are police procedurals, character dramas, revenge stories, thrillers, or episodes of “watch the star solve the case while everyone else stands around looking suspicious.” They may have a corpse, a detective, a lab report, and a confession, but that does not make them mysteries.

They all seem to fail in the same ways.

First, they frequently are not actually mysteries.  In the opening act, we often see who commits the murder.  We watch the killer creep into the room, pull the trigger, push the victim off the balcony, or poison the wine.  Then the rest of the episode consists of the detective slowly discovering what the audience already knows.  That can be suspenseful, but it is not a mystery.  It is a waiting game.

There is nothing wrong with that form, exactly.  Columbo made an art of it.  The pleasure of Columbo was not guessing who did it.  The pleasure was watching Peter Falk’s rumpled, shambling, apparently harmless detective worry the murderer to death with “just one more thing.”  But Columbo worked because it knew what it was.  It was not pretending to be a whodunit.  It was a how-will-he-catch-him.

Too many modern television mysteries give us the murderer early and then still expect us to pretend we are solving something.  We are not.  We are just watching the hero catch up.

Second, in the vast majority of television mysteries, the murderer is the highest-paid guest star.  This may be the single greatest weakness of television mystery writing.  The casting department destroys the plot before the first commercial break.

If a familiar actor shows up in the first ten minutes, has no obvious reason to be there, and then disappears into the background, you can safely arrest him immediately.  Television cannot resist this pattern.  The famous guest actor is never just the victim’s lawyer, the dead man’s neighbor, or the slightly rude restaurant owner.  He is there because the show paid for him, and, by heaven, they are going to get their money’s worth in the final scene.

This is especially true when the actor is just famous enough to be recognizable, but not famous enough to be above doing one episode of a network crime drama.  The moment he appears, the mystery is over.  You do not need fingerprints, blood spatter, or motive.  You need only ask, “Which guest star has the strongest IMDb page?”

Third, the suspect list is usually too small.  A real mystery needs room to breathe.  It needs several people who could plausibly have committed the crime, several motives that overlap, and several clues that point in different directions.  In a good mystery novel, almost everyone has something to hide, even if only one person is hiding murder.

Television usually gives us three suspects, and one of them is obviously innocent because he cried too much in the interrogation room.  One is too obvious, one is too sympathetic, and one is the guest star.  That is not a mystery.  That is a seating chart.

The problem is partly time.  A television episode has perhaps forty-two minutes after commercials.  In that time, it has to introduce the crime, interview witnesses, include a lab scene, give the regular characters something to do, provide a red herring, solve the case, and leave time for a final emotional conversation in a dimly lit office.  There is not much room for real detection.

Fourth, the structure gives the game away.  Television mysteries are often built on a rigid rhythm.  The first suspect is wrong.  The second suspect is also wrong.  The third suspect seems impossible, then suddenly becomes obvious after a late-breaking clue.  Someone lies.  Someone else says, “I should have told you this earlier.”  The detective sees something tiny, stares thoughtfully into the middle distance, and suddenly knows everything.

Once you know the rhythm, you are not solving the crime so much as reading the clock.  At minute twelve, the angry spouse did not do it.  At minute twenty-four, the business partner did not do it.  At minute thirty-five, the sweet old friend says something odd, and there it is.  Cue the confession.

The formula is so familiar that it drains the story of tension.  We know the first explanation is wrong because it came too early.  We know the second explanation is wrong because there are still eighteen minutes left.  We know the killer will be revealed at the exact moment the episode needs to start wrapping up.

Fifth, the detective often knows more than the audience.  This is a fatal flaw.  A fair mystery lets the audience reason alongside the detective.  We should see the important clues, even if we do not understand them at first.  When the solution is revealed, we should be able to say, “Of course.  I should have seen it.”

Too many television mysteries cheat.  The detective notices something the camera did not show us clearly.  Or a lab result appears at the last minute.  Or the hero remembers a detail from an earlier conversation that was not emphasized enough for any sane viewer to retain.  Then, in the final scene, the detective explains everything as if the solution had been obvious all along.

That is not mystery writing.  That is withholding evidence.

Agatha Christie, at her best, played fair.  Rex Stout played fair.  Ellery Queen practically invited the reader to stop before the final chapter and solve the case.  Television usually does the opposite.  It hides the usable clue, then congratulates itself for revealing it.

Sixth, character drama replaces detection.  Many television mysteries are less interested in the murder than in the detective’s personal problems.  The detective has a divorce, trauma, a drinking problem, a dead partner, a troubled daughter, a dying father, a corrupt boss, or all of the above.  The murder becomes a coat rack on which to hang the regular character’s weekly emotional burden.

Again, there is nothing wrong with character drama.  We like detectives with personality.  Sherlock Holmes had cocaine, a violin, and a tendency to be insufferable.  Nero Wolfe had orchids, beer, and an unwillingness to leave the house.  Travis McGee had a houseboat and a deeply complicated view of paradise.  But the mystery still mattered.  The character did not replace the puzzle.

On television, the crime often exists only to illuminate the detective’s feelings.  A murdered teenager reminds the detective of her own daughter.  A dead soldier reminds the detective of his time in the service.  A poisoned husband reminds the detective of his failed marriage.  By the end, the killer is almost an afterthought.  The real climax is the hero staring out a window, having learned something about grief.

Finally, the solution must be simple enough for one episode.  A good mystery needs misdirection, motive, opportunity, timing, character, and surprise.  It needs the solution to be unexpected, but inevitable.  That is hard to do in a novel.  It is much harder to do in forty-two minutes with a B-plot and recurring cast obligations.

So television reduces the mystery to something simple.  He lied about where he was.  She wanted the inheritance.  The brother was jealous.  The business partner was stealing.  The victim knew a secret.  The killer made one mistake, and the detective spotted it.

That may be enough for a crime show, but it is not enough for a real mystery.  A real mystery should make the audience lean forward, not merely wait for the reveal.  It should give us the pleasure of suspicion, deduction, failure, and sudden recognition.  It should make us think we might solve it, then punish us for being overconfident.

Television rarely does that anymore.  It gives us crimes, detectives, corpses, and confessions, but not mysteries.  The machinery is there.  The puzzle is missing.

And that is why, when I want a mystery, I usually go back to books.  The corpse may be imaginary, but at least the game is real.

A Nazi Baby Shower

After reading about the history of the painting, I was a little surprised to learn that it was actually rather small, about fifteen inches wide and only 22 inches tall.  After reading about the two decades of legal battles concerning ownership, I expected something bigger. 

I guess we should start at the beginning.  The painting, Madonna and Child, was painted by Edward Cranach the Elder, in 1518.  Of that we are certain, since the artist signed the work with his winged-serpent mark, and dated å Mary half-length in a landscape, her reddish-blond hair falling over her shoulders, holding a green velvet cushion on which the naked Christ Child lies.

We do not know who commissioned the painting, or even who its first owners were.  X-rays show that the underlying design was traced, suggesting that the work began in Cranach’s workshop rather than as a wholly freehand original composition.  It would be too much to say that Cranach “mass-produced” such paintings in the modern sense, but Madonna and Child images were certainly a workshop staple.  Even after five centuries, some 189 related paintings survive or are catalogued as works by Cranach, by his workshop, by his son, by his followers, or by later imitators.

Who owned the painting, where it was displayed, how often it changed hands…. none of that is known.  Due to the cost of such a painting at the time it was done, we can guess that whoever commissioned it (and all the subsequent owners) were probably the kind of people who had very nice jewelry and lived in castles.  The kind of people who might commission a painting for a church or to buy favor from another aristocrat.

Before the City of Cologne got its municipal mitts on Cranach’s Madonna and Child, the painting had already enjoyed the sort of mysterious, well-traveled life that makes provenance researchers reach for coffee, aspirin, and possibly confession.  Cranach painted it in 1518, but who ordered it, paid for it, prayed before it, or first showed it off to dinner guests remains unknown.  It simply appears out of the mists of art history with Mary, the Christ Child, and a very old habit of not leaving paperwork behind.

The first halfway solid clue is that it may have belonged to Samuel Graf von Festetics, although even that comes with the scholarly equivalent of a raised eyebrow and a question mark.  By 1859, it was respectable enough to show up at an Artaria auction in Vienna, where it was sold as lot 124.  From there, it entered the collection of the Grand Duke of Saxony/Weimar, which gave it a suitably aristocratic home and probably better manners than it would encounter later.

The Grand Duchy of Saxe-Weimar-Eisenach ceased to exist after World War I, and there was a frantic post-monarchy garage sale.  Eventually, the picture made its way into the hands of Theodor Fischer, the Swiss art dealer in Lucerne. Fischer, doing what art dealers do, sold it in 1937 to the Wallraf-Richartz-Museum / City of Cologne.  At that point, Cologne had acquired a lovely Renaissance Madonna.  Unfortunately, Cologne was about to demonstrate that owning a masterpiece and exercising good judgment are two entirely different things.

The city fathers of Cologne didn’t buy the painting for a museum, they bought it to give to Edda Göring, the newly born daughter of Hermann Göring.

Cologne did not suddenly develop an urgent civic need to place a Renaissance Madonna in a bassinet.  What Cologne wanted was favor. Hermann Göring was not merely Edda’s proud papa; he was one of the most powerful men in Nazi Germany.  Giving his newborn daughter a Cranach was the sort of municipal flattery that says, “Please remember us kindly when contracts, favors, grants, honors, and political survival are being handed around.”  Officially, it was a christening gift to little Edda.  In reality, it was a gold-framed curtsy to her father.

The city fathers knew that the Nazi leaders liked the paintings of Cranach, a German Renaissance painter.  Hitler alone stole 16 by the artist he considered Aryan, and an extra one by the son, Lucas Cranach the Younger.

The problem was that Cologne did not simply have a spare Cranach lying around in the mayor’s desk drawer.  The city acquired the painting from Theodor Fischer, the Swiss art dealer in Lucerne, for about 50,000 Reichsmarks. Then came the awkward part: paying a Swiss dealer required foreign currency, and Germany’s currency controls made that difficult.  So, Cologne solved the problem in the grand old tradition of public officials spending public treasure for private political advantage.

The city compensated Fischer not simply with cash, but by letting valuable artwork from the Wallraf-Richartz-Museum go out the door.  The most famous piece in that exchange was Van Gogh’s Portrait of Armand Roulin (left).  In effect, Cologne turned part of its public art collection into the purchase price for a private gift to the Göring family.  A Renaissance Madonna went to Edda, a Van Gogh went into the international art market, and Cologne got the warm glow of having pleased a Nazi.  As civic transactions go, it was less “public service” than “municipal groveling with museum-quality accessories.”

Hermann thanked the city fathers, on behalf of his daughter, of course, and took the painting to his retreat, Carinhall where it fit right in with the other 1,374 paintings he had “acquired” from some of the best families of Europe.  The painting stayed at Carinhall until 1945, when Göring discovered that there are few decorating problems more urgent than the Red Army approaching your country estate.   So, he packed up the art collection at Carinhall — paintings, sculptures, tapestries, loot, purchases, gifts, and assorted masterpieces of extremely flexible provenance — and sent much of it south toward Berchtesgaden, where Nazi grandees hoped the Alps might provide both scenery and plausible deniability.

Among the treasures caught up in this grand retreat was Edda’s Madonna and Child.

Göring then had Carinhall blown up, lest the Soviets get the satisfaction of walking through his private museum of acquisitive vanity or playing with his model trains.   Unfortunately for him, moving a stolen-and-semi-stolen art collection by train during the collapse of the Third Reich was not a model of orderly logistics.  Some works ended up in rail cars, shelters, depots, and local hands.  Then the Americans arrived, followed by the Monuments Men, who gathered the collection, sent it to the Munich Central Collecting Point, and began the long job of figuring out which “gift” was really a bribe, which “purchase” was really theft, and which Madonna belonged back in Cologne.

After the war, Edda Göring’s Cranach Madonna and Child entered that peculiar postwar category of objects best described as “artworks with lawyers attached.”  The painting had traveled south with Göring’s collection and ended up in Allied custody, not in Edda’s bedroom, drawing room, or hope chest.  Cologne, understandably sobered up after its earlier bout of municipal bootlicking, wanted the painting back.  Edda, meanwhile, argued that a gift was a gift, even if the gift had been a Renaissance masterpiece handed to a dictator’s baby by officials eager to curry favor.

The legal fight dragged on for years.  Edda even won an important round in the 1950s, when a court accepted the argument that Göring had not forced Cologne’s hand so much as Cologne had enthusiastically offered its hand, arm, and public art collection.  But the authorities kept pressing. In 1968, the German courts finally ruled for Cologne, holding that the gift was improper, and the Madonna went back where it belonged: the Wallraf-Richartz-Museum.  Today, it is in the collection of the Wallraf-Richartz-Museum & Fondation Corboud in Cologne — a Madonna rescued from the world’s worst baby shower.

Göring’s Model Trains

In the early 1980’s while my wife made yet another trip to look at a cathedral (seen one, you’ve seen them all) I took a trip to a really great museum/hobby house that featured an incredible model train layout near Paddington Station.  The London Toy and Model Museum occupied five floors and visitors were led into a working pithead and coal mine model, with lifts, miners, pit ponies, and Davy lamps.  The museum also had large, landscaped working models in which trains whizzed in and out of tunnels and over viaducts.

Sadly, the museum closed in 1999, which only goes to show that I was right, and my wife was wrong—I can still go see Saint Nigel’s Buddhist Bar and Tabernacle or whatever it was, but The Doc has lost her only chance to see a great museum.  (There are rumors that most of the layouts were sold to a museum in Tokyo, but don’t tell my wife.)

There was once another great model train layout—perhaps the largest in the world at that time—but none of us are ever going to see any of it because it was owned by a madman who deliberately destroyed it.  It belonged to Adolf Hitler’s right-hand man, Hermann Göring.

There is something almost too perfectly awful about Hermann Göring’s having a gigantic model railroad at Carinhall, his forest palace retreat.  Of course he did.  Of course, the man with the uniforms, the titles, the pets, the loot, the hunting lodge inflated into a palace, and the incurable need to perform magnificence for visitors also wanted a miniature kingdom in which everything ran on time because he said so. 

The historical record is clear that he really did have a substantial model railway at Carinhall, that he showed it off to important guests, and that it was part of the estate’s larger function as a theater of ego.  The Library of Congress has a 1937 photograph explicitly titled “Hermann Göring and Adolf Hitler watching model trains at Carinhall, Templin, Germany,” and another LOC album for 1938 includes views of Göring’s model railroad.  Remaining insurance papers show that Göring insured the layout for $254,000, the equivalent of more than a million dollars today.  Carinhall itself was not some cozy private retreat: it was an ostentatious estate where Göring displayed stolen art, his pet bison, and entertained dignitaries.

So why did he build it?  We do not, so far as I know, have a neat little memo from Göring saying, “I require a vast train layout because I am a vain overgrown child with a passion for theatrical domination.”  But the surrounding evidence makes the motive fairly obvious.  Carinhall was a stage set for self-glorification, a place where Göring curated an image of himself as hunter-prince, art connoisseur, host, strongman, tastemaker, and second-or-third-most-important man in the Reich depending on which week you asked.  The Jewish Museum Berlin describes Carinhall as his “ostentatious” hunting lodge, where he displayed stolen treasures seized from Jewish owners across Europe.  Add to that the photographic evidence that he was showing the trains to Hitler and other guests and the model train layout starts to look less like a simple hobby table and more like one more prop in a power pageant.  It let him be emperor not merely of a forest estate, but of a tiny obedient world under glass and rafters.  One suspects the attraction was obvious: in the model kingdom, unlike in the Luftwaffe, the scheduling problems were manageable.

As for how large it was, the safest answer is: astonishingly large, though not with the kind of museum-grade precision that lets one draw a blueprint today.  The strongest widely repeated figure is that Göring had two train sets at Carinhall totaling about 400 square meters, roughly 4,305 square feet.  That is not a toy on a table.  That is a hobby expanding until it begins to resemble a municipal planning department.  The archival photographs support the existence of multiple substantial installations over time, even if they do not, by themselves, settle every technical detail.  The attic images especially suggest a sweeping layout under steep rafters, while later reporting points to more than one major setup.

There is also the deliciously absurd matter of brand and scale.  Model-train devotees have argued for years about exactly which parts were Märklin and which were Trix (the two prominent German manufacturers of model trains), because even the dictator’s playroom cannot escape specialist debate.  What can be said confidently is that the German Federal Archives catalog a photograph from January 12, 1943, at Carinhall as “Vorführung Trix-Eisenbahn” — a demonstration of a Trix railway — during Göring’s 50th-birthday festivities.  That means at least one documented Carinhall installation was definitely Trix-Express. 

The train layout was important enough to be part of the performance for invited guests, which tells you a great deal about its role in the ecosystem of Göringian vanity.  Even on his birthday, on the very day that the Russian Army launched the counteroffensive at Leningrad, there he was, ushering people over to admire the little trains.  Nero fiddled; Göring seems to have preferred careful track alignment.

Was it the largest model railway in the world at the time?  That is a tempting line, because it sounds exactly like the sort of thing Göring himself would have wanted repeated.  But the evidence I found does not prove a world record.  The 400-square-meter figure certainly makes it enormous and arguably one of the largest known private layouts of its era, but I would stop there.  It is big enough without historical embroidery.  There is no need to gild the lily when the lily already occupies the attic, the basement, and a nontrivial fraction of Brandenburg.  The important point is not that it was definitively number one on some global leaderboard of grown men with soldering irons: it is that it matched the scale and psychology of the estate that housed it—conspicuous, curated, expensive, theatrical, and meant to impress.

And then, of course, he destroyed it.  Or rather, he destroyed the whole world that contained it.  In April 1945, with Soviet forces closing in, Göring ordered Carinhall blown up so it would not fall into Soviet hands.  The Reichsmarschall ordered Captain Frankenberg to destroy the estate, thus,  petrol and explosives were spread through the interior, and the place was blown up as Soviet troops approached.  In one sense the reason was practical: to deny the enemy a trophy, to conceal the scale of the loot, and to prevent capture of the estate as an intact symbol.  In another sense, it was grimly fitting, since regimes built on plunder often prefer arson to accountability.  If he could not keep the stage set, no one else was to inherit the scenery.

That tells us why the train layout vanished too. The trains were not singled out for destruction because Göring suddenly developed a principled objection to excessive toy railroading. They disappeared because they were embedded in the larger destruction of Carinhall itself.  Some model houses that may have belonged to the railway later turned up in a Berlin government depot, but no intact working layout is known to survive on public display.  So, the vast miniature empire died the same way the larger counterfeit empire was dying around it: by collapse, by denial, by frantic evacuation, and by demolition.  The final irony is almost literary.  A man obsessed with arranging the world to flatter himself ended by blowing up even the tiny one.

The Carinhall railway was not just a hobby.  It was a scale model of Göring’s tastes and delusions: lavish, childish, domineering, meticulously staged, and impossible to separate from the machinery of theft and violence that financed the setting around it.  He built it because he liked trains, certainly, but also because he liked the display, the control, and the adoring gaze of guests being shown yet another marvel in the palace of a man who mistook accumulation for grandeur.  It was huge and by the late 1930s and early 1940s it was famous enough to be photographed with Hitler and other visitors.  He destroyed it, or at least doomed it, for the same reason he destroyed Carinhall: because the real world was closing in, and men who make a religion of possession rarely meet loss with dignity.  Sometimes they just blow up the train room.

Alcatraz, You’re Kidding, Right?

President Trump has asked Congress for $152 million to start work on restoring Alcatraz as a federal penitentiary.  This almost has to be a joke. 

A couple of weeks ago I wrote about Cuba and had to preface my remarks with an admission that, while I have read and studied Cuba extensively, I’d never traveled there.  This week is a little different, as I’ve been to Alcatraz half a dozen times and I have met former inmates, guards, and even the children of guards who grew up on the island.  I have a bookshelf full of texts about Alcatraz.  In addition, I spent years maintaining aging hotels on Galveston Island which, while not a prison, certainly gave me some experience with aging concrete and plumbing in an ocean environment.

So, let me summarize the conclusion before I present my evidence…. The whole idea of making Alcatraz back into a federal prison is lunacy.  It’s never going to work. 

Let’s start with a very brief history.  Formerly a military fort and prison, The United States Penitentiary Alcatraz opened August 11, 1934, and closed March 21, 1963.  This was a small prison, housing on average just 275 of the most hardened and infamous of federal prisoners. 

Alcatraz was never a very successful prison primarily because it's being an island a little of a mile off the coast of San Francisco, meant that the harsh climate made operating the prison a financial nightmare.  There was not—and still is not—any direct connection to the mainland for water, electricity, or sewage, meaning everything had to be transferred by boat from the mainland.  By the time the prison closed, the daily cost of keeping a single incarcerate prisoner on Alcatraz was more than three times higher than anywhere else in the rest of the federal penitentiary system.   This was the primary reason the prison was closed, and is the primary reason it should never be reopened.

But, let’s ignore the operating cost.  What would it cost to make that aging dinosaur of rust and crumbling concrete ready to reopen?  First, let’s connect the utilities. 

The National Park Service says water and diesel fuel still have to be ferried out, while garbage and sewage are carried back to shore, and the Bureau of Prisons says the old prison was closed in part because nearly one million gallons of fresh water had to be barged weekly to the island.  Using recent Bay-area marine utility projects as rough comparisons, a submarine electric feeder to Alcatraz looks like roughly $5 million to $10 million, and a submarine water line looks more like $15 million to $50 million.  Put less delicately, before a single inmate takes a shower, flushes a toilet, or turns on a light, you are probably staring at $25 million to $70 million just to stop running the place like an offshore camp with boats and generators.

Then comes the sewage, which is where the fantasy starts getting expensive in earnest.  NPS says sewage is still hauled back to the mainland today, and on-site treatment is merely “under study,” so a genuine prison-scale wastewater solution would mean either an on-island treatment system or another serious marine utility project.  Based on the same kind of Bay crossing work and the obvious need for pumps, redundancy, permits, and historic-site constraints, a believable estimate for solving sewage in a permanent way is roughly $20 million to $60 million.  Add that to the water and electric work, and the all-in bill for just giving Alcatraz dependable electricity, potable water, and sewage treatment lands in the neighborhood of $45 million to $130 million.  That’s just to run utilities—we haven’t yet started work on the actual prison.

If Washington seriously tried to resurrect Alcatraz as a prison, the environmental review alone would likely become its own island institution.  Across federal agencies, the median time to complete an environmental impact statement is about 34 months, and even Alcatraz’s much narrower historic-preservation-and-safety program took the National Park Service from a draft EIS in March 2001 to a final EIS in October 2001 and a Record of Decision in 2002.  On top of NEPA, Section 106 review would apply because this is a federal project affecting historic property, and because Alcatraz is a National Historic Landmark the agency must try to minimize harm and bring in additional consultation.  The cheapest part of this circus would probably be the consultants: GAO says a “typical” EIS has been estimated at roughly $250,000 to $2 million, with DOE’s median EIS contractor cost at $1.4 million.  In plain English, before the first prisoner got his commemorative iron cot, Uncle Sam could easily spend a few years and low-single-digit millions just producing the paperwork explaining why this was a terrible idea.

And let’s not forget about asbestos:  Alcatraz is not some quaint old ruin sprinkled with a charming dusting of historical fibers.  The National Park Service says structures on the island are assumed to contain asbestos and lead-based paint until proven otherwise.  The problem is that there does not appear to be a clean public estimate for “remove all asbestos from Alcatraz and call me when it’s done.”   There is a similar project already underway at Golden Gate Park, so using those numbers I’ll hazard a semi-educated guesstimate and say it will cost about $7 million to remove the asbestos enough that remodeling and repair can begin.

But wait, there is one more step!  The buildings are almost a century old, and they need to be stabilized to prevent collapse from seismic activity.  Here, we can use fairly precise numbers, since the National Park Service has just run the numbers to do exactly that.  The cost is $63.584 million.  Also already scheduled is a project to stabilize the island’s wharf at a cost of $40.2 million.

So now we are ready to start repairing those old cells, the reason we started this whole nonsense.  Well, no.  The old Alcatraz cells were 5’ by 9’, totaling 45 square feet in total.  While there is no set national minimum cell size, numerous court challenges on the space necessary to avoid “cruel and unusual punishment” have forced the Department of Justice to set detention standards for segregation call for at least 70 square feet total, with 35 square feet unencumbered by bunks or plumbing. 

Wait, wait: there’s even more!  In the federal prison system, prison cells do not all have to be ADA-compliant in the sense that every single cell must meet full accessibility specs; instead, federal prisons must make their institutions accessible as required by the Rehabilitation Act, the Architectural Barriers Act, applicable federal accessibility standards, and Bureau of Prisons policy.  The BOP’s disability policy says federal institutions “should be accessible” to that extent, and the…. Well…let’s just say that bringing Alcatraz up to code is going to reflect some significant and expensive design modifications.  While the cells don’t have to meet ADA compliance, the facilities for the modern coed guards do have to meet modern standards. 

At this point, you should realize that none of the existing cells on Alcatraz meet any of these standards and the entire cell blocks would have to be rebuilt.  And there are similar concerns about the kitchens, shower areas, and medical services meeting modern codes.  By far the simplest solution would be to bulldoze down the site—ignoring that it is on the National Register for Historic Sites—and build a new prison.  While we are somehow turning the small cells into large cells, we should go ahead and add in the HVAC, fire suppression, electronic controls, medical space, kitchens, staff areas, backup power, and marine security.  I’ll skip all the back of the envelope figuring and just give you the bottom line.  This is going to cost about $1.2 billion.

Which raises the rather obvious question:  If you are going to have to build a new prison, why not build it someplace where the cost will be much cheaper? 

If the goal is to make Alcatraz hold about 300 inmates again the honest number is not the White House’s $152 million first-year ask, but something more like a total rebuild cost of roughly $1.2 billion.  In short, a little over a billion dollars buys you a boutique rock in the bay for 300 prisoners, which is an impressively inefficient way to do corrections.

By contrast, a brand-new 4,000-bed prison in the middle of nowhere Nevada would almost certainly be cheaper per inmate and might well cost about the same in total.  One policy estimate based on recent prison construction put a 4,000-bed facility at about $500 million.  So, the dark joke here is that for roughly the kind of money it could take to resurrect Alcatraz for 300 people, you could build two ordinary 4,000-bed mainland prisons with roads, utilities, and parking.  We would also avoid about two decades of battles in California court followed by endless appeal in the 9th Circus.  (But who in their right mind would want to do that???)

Clearly, Alcatraz will never again be a federal penitentiary.  But, you should know that officials from both the Department of Justice and the Bureau of Prisons have recently toured the island and are actively planning to proceed.

Couldn’t we just lock them up there now and save a lot of taxpayers money?