The text of the negative capability letter comes to us from a single unreliable source: John Jeffrey, the second husband of Georgiana Wylie Keats (who had married John Keats's brother George and emigrated to America in 1818), and, for better or worse, the person to whom we owe a debt of gratitude for conveying to posterity the term negative capability. That the letter with Keats’s coinage exists at all is fortuitous. After having come across a newspaper advertisement for an upcoming biography of the poet, and after corresponding across the Atlantic with its author, Richard Monckton Milnes, Jeffrey transcribed Keats's letter during the summer of 1845, along with fourteen others. Thereafter he sent his texts to Milnes, who used these copies for his Life, Letters, and Literary Remains of John Keats (1848). So it was that negative capability moved out of the private world of the Keats family and its Louisville, Kentucky circle, entering into public discourse for the first time.

At some point the manuscript itself quietly slipped out of existence (though perhaps not forever). Most of Milnes's Keatsiana, however, including Jeffrey’s transcript of the negative capability letter, has stuck around, the bulk of it making its way to Harvard's Houghton Library in 1952. Thanks to the digitization of materials in the Harvard Keats Collection, Jeffrey's transcripts are now readily available for viewing by anyone with an internet connection and an appropriate device. Because original manuscripts for nine of the fifteen letters copied by Jeffrey still exist, we can compare Jeffrey's work against them and see precisely how badly Jeffrey performed his task. Hyder Edward Rollins, the editor of what remains the standard scholarly edition of Keats's complete correspondence, assesses Jeffrey's work as such: ‘he changed words or phrases that he disliked or did not understand or could not decipher; he reversed the order of certain sentences, reformed spelling, punctuation, grammar. Worse yet, with no warning he omitted words, sentences, paragraphs, at times whole pages’ (LJK, I: p. 20). To demonstrate just how egregious such omissions could be, Rollins prints the Jeffrey transcript of the September 1819 journal letter to George and Georgiana Keats, the original manuscript of which still survives, and which runs to thirty-four pages in Rollins's edition. The Jeffrey transcript fills approximately three.