The Narratee and the Typo

A long, earnest study has been knocking around at Lingua Franca regarding so-called grammos and typos in social media. As argued by a psychologist and a linguist at the University of Michigan, the response to “actual written errors” (as opposed to social-media conventions like elided punctuation or nonstandard abbreviations) depends on the personality of the reader more than any other criterion. I find this idea, in a word, weird.

For many years, a debate raged in the field of narratology over whether there was such a thing as the narratee, a term coined by Gerald Prince in 1971 to describe “the fictive entity to which the narrator directs his narration.” After plenty of pushback from colleagues and dissenters, Prince actually reconsidered his theory in 1982, noting that he had “tended to conflate” the roles of the addressee, the receiver, and the text-embedded narratee of a discourse.

The Michigan researchers used 83 volunteers to assess responses to an ad for a housemate and compare those responses to personality profiles created through the Big Five personality assessment. I don’t know how much credibility this personality-assessment tool has among psychologists, but I took the test just for fun and found it tailor-made to boost my good feelings about myself. Surely I am responsible and dependable, curious about many things, and eager to help others! But OK, let’s pretend that among the 83 volunteers involved in the study, some admitted to being irresponsible, incurious, and self-centered. I remain unconvinced that such a study produces a rounded personality profile so much as it produces a given respondent’s tendency to think well or poorly of himself.

Moving right along, we find that the researcher distributed three versions of the housemate ad: one “fully correct,” one with “2–4 grammos (underlined),” and one with “2–4 typos (in boldface).”

The study proceeded to cross-reference all sorts of traits associated with the participants, including agreeability, conscientiousness, extraversion, neuroses, and openness, along with their “attitude” toward grammar and their sensitivity to grammos and typos. Articulating the various specifics of the study is above my pay grade, but the general conclusion, that “personality traits influence our reactions to written errors,” seems straightforward enough. In addition to finding that grammos bothered people more than typos did, the authors concluded that extraversion particularly affected the readers’ receptivity to the ad despite the errors with which it was seeded. In addition, the researchers concluded that “those who believe ‘good grammar is important’ don’t view typos as evidence of ‘bad grammar’ and thus do not rate people who produce typos as negatively as extraverts who are less concerned with ‘good grammar.’”

As the parent of two young men, I’ve seen dozens of housemate ads. They are all coded. I’ve learned what 420 friendly means. I’ve learned that SHARE CLEANING RESPONSIBILITIES!!! means the interview for the room will be conducted by a suspicious victim of prior housemate abuse who will look askance at my older son’s torn shirt and unkempt beard. I’ve learned that students overwhelmingly list their preference for male or female roommates. And no one ever mentions tennis; in fact, the mention of tennis (plus “old school rap”? Seriously?) already casts doubt on the advertiser’s legitimacy.

This all happens before we get to the typos.

In Prince’s reconsideration of the narratee, he asks himself, “Should I write a letter to a friend and say something like ‘You ate a hamburger for lunch and it made me very happy,’ where exactly would the meta-narratee be?” The same question, I suspect, applies to the study at hand. There’s an invented narrator — the advertiser — and an invented narratee — a student who’s going to judge the housing situation on the basis of partying, tennis, and rap, but who somehow fails to notice the essential strangeness of the ad. The subjects of the study are supposed to be what Prince would call the meta-narratees — that is, readers who find the text to be directed toward them even if they are not specifically addressed within the text. In fiction, Prince notes, this idea has some merit, but in nonfiction it’s nonsense. A housemate ad is directed toward potential housemates, period. That this example is a fictional ad doesn’t really matter, in my view: We are treating it, for purposes of the study, as the real thing.

I’m sure that so-called grammos and typos affect readers differently. But neither of them happens outside a context. If the Big Five personality traits make a difference — and I’m doubtful even on this point — that difference surely surfaces long before we get to the point where a given subject is “bothered” by the use of you’re rather than your. Why? Because we’re not the narratee for the ad. When the study’s authors write that “extraverts with stronger beliefs about good grammar were more sensitive to typos; however, there were no differences linked to beliefs about good grammar associated with grammos,” they’re focusing on the one group that can possibly imagine itself responding to a rather peculiar ad that emphasizes relaxing and having fun. Possibly that’s a group that forgives the auto-correct function while noticing a lackadaisical approach to the spell-check function. But content comes first and is never separate.

That’s what I find. But then, I’m not looking to rent a room.