February 27, 2020
The literary forensics of Edgar Allan Poe’s death
by Ryan Harrington

How did Edgar Allan Poe die? One study seeks to find why. And why did Poe repeat the name “Reynolds” in the hospital? Still hard to say.
When it comes to the death of Edgar Allan Poe, we know that he was found, in his 40th year, more or less raving mad on the streets of Baltimore wearing somebody else’s clothes.
**record scratch**
[Voiceover]: But you’re probably wondering how he got there.
We don’t know a whole lot about that. And theories for the cause of his descent into madness range from rabies to … being the unfortunate instrument of electoral fraud. A famous melancholic, it’s tempting to assume his demise was a form of soft suicide.
A recent study of his texts, entitled “Deep into that darkness peering: A computational analysis of the role of depression in Edgar Allan Poe’s life and death,” published in the Journal of Affective Disorders, attempts to test the suicide theory against his literary output.
As John Anderer breaks it down for Study Finds, the digital humanities-style experiment went like this:
A great deal of Poe’s writing was analyzed by a computer for the study; 309 of his personal letters, 49 poems, and 63 short stories. The researchers intended to determine if his work showed a pattern of linguistic cues commonly associated with suicidal thoughts, especially towards the end of his life.
More specifically, five measures used to diagnose depression and suicide were focused on:
-
Increased use of first-person singular pronouns (I, me, and my)
-
Lots of negative emotion words (sad, angry)
-
Cognitive processing words (understand, think, know)
-
Very few positive emotion words (good, happy)
-
Very few first-person plural words (we, us, our)
The study finds that these linguistic fingerprints of depression and madness don’t spike toward the end of his life. In fact, they were more prevalent at the heights of fame he experienced in 1843, 1845, and earlier in 1849—the year of his death.
While this information seems like something shy of “case closed,” it is certainly an indication of the types of computational analysis—in this case in the service of biographical criticism—we’ll be seeing more of as technology, funding, and gee-whiz factor permits.
Ryan Harrington is a senior editor at Melville House.