7. What would overturn the verdict

The verdict is a lean of 56 to 44 between two men, on documentary evidence that is weak on both sides. Small additions to the evidence would move it; the following would move it substantially.

Toward Hay

• A draft or copy of the letter in Hay's hand, or a diary entry, letter or memorandum by Hay claiming it, in the Hay Papers at Brown University or the Library of Congress, or in the papers of Morley, Page or Brownell. None of the three first recipients is known to have written the story down. A letter from any of them dated before 1922 would turn hearsay into testimony.
• The original manuscript in Hay's hand over Lincoln's signature, as the letter to Harrington of 28 November 1864 is. A manuscript in Lincoln's hand would not settle the matter the other way: Robert Lincoln said his father copied out his own drafts, and a fair copy of a Hay draft would look the same.
• A use by Hay, before November 1864, of any of the letter's rarer phrases: "beguile you from the grief", "the altar of freedom", "the solemn pride", "assuage the anguish". The uses we found are all later.
• An identification of the "unidentified handwriting" of the Phillips copy of 21 November as Hay's. That would make the day's other composed letter his and strengthen the reading of the scrapbook as a record of his own work.
• A stylometric method that still attributes the letter to Hay after it has been validated on Hay's known letters for Lincoln's signature and on register-matched Lincoln texts. We could not build one. A corpus of Hay's condolence letters from the 1860s, of which we have one, would be the place to start.
• A rerun of the register test that finds no register effect. That would restore the 2019 study's standing and move the verdict toward Hay. The second implementation finds the effect, at a lower rate than ours (section 4).

Toward Lincoln

• The original manuscript in Lincoln's hand with corrections, which is how Robert Lincoln said his father wrote "a careful letter".
• A statement by Hay, in his hand, that Lincoln wrote it, or a first-hand denial to any of the men who later claimed the opposite.
• A use by Lincoln of "beguile" in any surviving document. Burlingame reports that the word occurs nowhere in the Collected Works (1995, p. 65); we confirmed this for the four volumes we could read.

Two things would not change the verdict. More reported confessions of the same kind, at two or more removes and written down decades later, would add little, because the story was in circulation by 1904 and the chains share a source. A further n-gram study under the 2019 design would add little, because both implementations of our tests find that design misreading between roughly a quarter and a half of Lincoln's elevated prose (28 to 48 percent, depending on the pieces and the pool). A rerun on the paper's own corpus, by a third party, is the check that remains.