The Alderman Library Theft: Poe's Stella Portrait and Tamerlane

In early August 1973, a manuscripts cataloger at the University of Virginia’s Alderman Library was showing co-workers around the McGregor Room vault. She opened the drawer that should have held one of the library’s showpieces, an 1849 daguerreotype of Edgar Allan Poe, and found it empty. There was no wooden block or paper slip in its place, the markers staff were meant to leave when an item was out on loan.¹

That empty drawer was the first sign of the largest theft in the library’s history. By the time the audit was finished, more than 50 items were on the missing list, among them a copy of Poe’s first book, Tamerlane and Other Poems, which collectors rank as the rarest book in American literature.¹ More than fifty years on, none of the objects has been recovered, no one has been charged, and the loss is still felt by anyone who studies Poe’s face or his earliest poems.

A portrait with a complicated past

The missing plate is known as the “Stella” daguerreotype. In his catalog of Poe’s likenesses, Michael Deas counts eight original daguerreotypes of the poet, and argues that this one and its near twin, the “Annie” plate, were the only two exposed at a single sitting in Lowell, Massachusetts. The sitting is often dated to autumn 1848, but Deas makes a case from letters and from the worn look of Poe’s face that it took place in late May or early June 1849, a few months before his death.²

One plate went to Annie Richmond, the Lowell woman Poe was devoted to in his last years. The other, Deas suggests, Poe carried back to New York and gave to Sarah Anna Lewis, a Brooklyn poet who wrote as “Stella,” and her husband, before he left the city for the last time on June 29, 1849.³ Lewis kept it for three decades. By 1874 she was living in London, where she met Poe’s English biographer John Henry Ingram, who told a friend that her daguerreotype was “as fresh as if just taken.” To Ingram’s annoyance, she also sold reproduction rights to the London Stereoscopic Company, and by 1876 he was complaining that copies “are everywhere.”³

Lewis left the plate to Ingram when she died in November 1880. After Ingram’s death in 1916 his Poe collection passed to his sister, Laura Ingram, who sold most of it to the University of Virginia, Poe’s old university, in 1921.³ For half a century the daguerreotype sat in Charlottesville, across the street from No. 13 West Range, the student room Poe once occupied.¹

The lengthening list

The first reaction to the empty drawer was not alarm. The curator of manuscripts, Edmund Berkeley Jr., later wrote in a memorandum to his superiors, as quoted by Virginia Magazine, that “I was not perturbed,” because items were often misfiled and could stay lost for months.¹ Only in October, when he searched the vault trays himself, did he notice a second absence: a silver medal struck for Thomas Jefferson’s first inauguration. Staff then began a full audit, cross-checking acquisition records against shelf locations on index cards.¹

The news kept getting worse. In November a George Washington letter that had been lent to the library could not be found. Berkeley’s list of losses grew to eight entries, including Jefferson’s own copy of the 1796 Book of Common Prayer.¹ The audit later found that Robert E. Lee’s signed General Order No. 9, his farewell to his army after Appomattox, had gone, along with a Mark Twain manuscript of chapter 28 of The Gilded Age.¹ Deas, writing in 1989, also lists autograph letters by Poe, Washington, Charles Dickens and Aaron Burr among the losses.³

In December a reader asked to see Tamerlane, and the rare books staff, who shared the vault, could not find it.¹ Poe had a Boston printer, Calvin F. S. Thomas, run off this slim pamphlet of about 40 pages in 1827, when he was 18; it drew almost no notice, and no copy was known to scholars until 1876. Estimates of the print run range from under 40 to as many as 200, and about 12 copies survive.¹⁴ Virginia’s copy had passed through Boston and New York dealers and a Connecticut collector to Clifton Waller Barrett, who gave it to the university around April 1952. The Edgar Allan Poe Society’s census still lists it as stolen, “current whereabouts are unknown.”⁴

The university’s final inventory describes 35 items individually and adds an unspecified number of “clipped autographs,” signatures that appear to have been cut from historical documents. A Christie’s researcher consulted by Virginia Magazine in 2019 put the 35 described items at more than $1.2 million at auction, with the caveat that no one has seen them in decades.¹ For comparison, a privately owned Tamerlane was offered at Christie’s in December 2009 with an estimate of $500,000 to $700,000.⁵

A vault that did not need to be broken

The university’s counsel reported the thefts to the University of Virginia Police at 2:30 p.m. on December 17, 1973, more than four months after the empty drawer was found. The Virginia State Police and the FBI also took part.¹ Berkeley later acknowledged in a 1976 journal article, as Virginia Magazine reports, that police criticized the library for the delay.¹

The McGregor Room vault looked formidable. Visitors passed through two-inch armored doors with a combination dial, then a second set of metal doors that needed a key. The investigator assigned to the case, Eric W. Shoemaker, quickly found the weakness: the combination was written on a sheet of paper in a secretary’s desk drawer, and the key sat in a tin, not locked, on a bookshelf next to the vault. When workers arrived two days after the police were called, both sets of doors stood open with the lights on.¹

Because no one had to break in, there was no physical evidence of a crime. That also hurt the university’s insurance claim. The insurer pointed to a policy exclusion for a “mysterious or unexplained disappearance” and offered half of what the university sought; no record found by Virginia Magazine shows how the claim ended.¹ A psychology professor known for building his own lecture gadgets was asked to rig the vault’s first alarm system.¹

An investigation that went cold

With no clear date for when items had last been seen, investigators started with a trail that was months old, perhaps older. The mix of missing objects led the lead investigator to wonder whether more than one thief had been at work, and the prevailing theory among investigators was that the thefts were an inside job by people who knew the vault.¹ Staff were asked to take polygraph tests; 20 of 22 did, and all of them passed. In Berkeley’s words, as the magazine quotes them: “The personnel of our two departments inevitably wondered about each other, and the resulting tensions hurt morale.”¹

Police looked closely at several people over the following years, but no one was ever charged, and this article does not name or describe any of them.¹ Shoemaker, who later led several campus police departments, wrote to the magazine in 2019 that the Alderman case remained “one of my unsolved cases.”⁶ A rare books staff member hired soon after the theft recalled that newcomers speculated that insiders had hidden the items somewhere in the building, perhaps in the attic or taped under bottom shelves, and that their treasure hunts were fruitless.⁶

In December 1973 the university decided to make the losses public. Publicity invited criticism of the library’s security, but it also made the objects harder to sell and might draw leads. The risk, as the magazine notes, was that a frustrated thief might destroy them instead.¹ By 1989 Deas wrote that no trace of the missing materials had surfaced and that authorities feared many of them, including the daguerreotype, may have been destroyed.³ The paper trail of the investigation has itself thinned: the FBI told Virginia Magazine that it had destroyed its files under its records schedule, and the Virginia State Police holds only summaries.¹

Why the loss still matters

Poe scholars still have the Stella image. Lewis’s commercial copies, Ingram’s tipped-in photograph of 1880 and a Frederick Warne engraving of 1890 all preserve it, some of them printed in reverse.³⁷ What is gone is the object itself, the only first-generation record of that moment in front of the camera. A copy cannot be examined the way an original plate can, for its size, toning, case and condition. That matters for a sitting as poorly documented as this one: Deas notes that neither Lowell plate bears a discernible maker’s mark, and he could not say with certainty which of the town’s daguerreotypists took them.²

The fate of the twin plate shows what a surviving original makes possible. Around 1856 Annie Richmond wrote to Poe’s mother-in-law that “some one has stolen my Daguereotype of him.” Whether it was stolen or only mislaid is unclear, but she evidently had it again by the 1870s.² It changed hands through collectors and a Chicago auction in September 1973, a few weeks after the Stella drawer was found empty, was lent to the National Portrait Gallery in 1978, and was bought by the Getty Museum in 1984, where it can be studied and compared.² The Stella plate has had no such afterlife.

The Virginia losses were part of a wider problem that archivists were then beginning to measure. Writing in American Heritage in 1978, T. H. Watkins used a copy of the stolen Poe portrait to open an essay on what he called “sticky-fingered specialists in the art of purloining the past.” He dated the portrait to 1848 and the theft to 1974, a year the Edgar Allan Poe Society’s pages also give; the police file, as Virginia Magazine reports it, shows the losses were discovered and reported to police in 1973.⁸⁴⁷¹ He reported that the Society of American Archivists issued a register of stolen documents three times a year, that it already listed about three thousand items, and that one archives director thought it might record as little as 5 percent of real losses, since many institutions did not report thefts or had not discovered them.⁸ Watkins cited the case of Robert Bradford Murphy, who had looted at least one hundred institutions and was arrested by the FBI in 1963 with six suitcases of stolen papers, then sentenced to ten years, and noted that the society had set up a security consulting service in 1975 with federal humanities funding.⁸

Today Virginia’s rare holdings sit in the Albert and Shirley Small Special Collections Library, opened in 2004 with modern security, and the old McGregor vault serves as little more than a storage room.¹ The decision to publicize the losses in 1973 was meant to make the objects hard to sell, and the census of Tamerlane copies still records Virginia’s as stolen.¹⁴ If any of the items survive, that public record is the best hope of recognizing them when they surface. If they do not, the loss is permanent: the one plate from which every Stella copy descends, and one of about a dozen surviving copies of Poe’s first book.

Footnotes

1. Gard, Virginia Magazine, “Nevermore” (2019). ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶ ↩⁷ ↩⁸ ↩⁹ ↩¹⁰ ↩¹¹ ↩¹² ↩¹³ ↩¹⁴ ↩¹⁵ ↩¹⁶ ↩¹⁷ ↩¹⁸ ↩¹⁹ ↩²⁰ ↩²¹ ↩²² ↩²³

2. Deas 1989, “The ‘Annie’ Daguerreotype,” pp. 47-51. ↩ ↩² ↩³ ↩⁴

3. Deas 1989, “The ‘Stella’ Daguerreotype,” pp. 51-54. ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶

4. Edgar Allan Poe Society, Tamerlane and Other Poems (1827), introduction and census. ↩ ↩² ↩³ ↩⁴

5. Kehe, Christian Science Monitor, December 4, 2009. ↩

6. Virginia Magazine, Letters to the Editor, Summer 2019 (Shoemaker; Nelson). ↩ ↩²

7. Edgar Allan Poe Society, “The ‘Stella’ Daguerreotype.” ↩ ↩²

8. Watkins, “The Purloined Past,” American Heritage 29:5 (1978). ↩ ↩² ↩³