I’ve mentioned my fondness for compiling historical grammatical errors as a reminder that we are not, point of fact, destroying what used to be a perfect language. Previously, I’d found unnecessary quotation marks in a 1960 World Series celebration, it’s for its in a 1984 John Mellencamp video, and an apostrophe incorrectly marking a plural in a famous 1856 editorial cartoon. But these were all punctuation-based errors. Today’s is a proper grammatical error, and one that people full-throatedly bemoan nowadays.

I found this error by admitting to myself that I am secretly an old man, and coming to terms with it by spending much of the summer sitting in parks, reading books on naval history and international relations. One of them, Nathaniel Philbrick’s Sea of Glory, tells the story of the U.S. Exploring Expedition, who discovered Antarctica and created the country’s first accurate naval charts for the Pacific islands. It’s a good book, but then it turned great by having two interesting old quotes four pages apart.

In the first, the Expedition is approaching Fiji and takes on another pilot due to the many coral reefs in the area:

“Wilkes felt it necessary to secure yet another experienced pilot at Tonga named Tom Granby. ‘You will find when we get to the Islands,’ Wilkes assured Granby, ‘that I know as much about them as you do.’ Granby smiled. ‘You may know all about them on paper,’ he replied, ‘but when you come to the goings in and goings out, you will see who knows best, you or myself.'”

Myself here is clearly non-standard, as no first-person pronoun has appeared anywhere in the sentence. The standard rule for reflexives, known as Principle A in Government and Binding theory, and discussed in pretty much every introductory syntax class, is that a reflexive must be bound in its governing category. Or, to say it in a more theory-agnostic and somewhat looser way, the coreferent of the reflexive (I/me for myself) has to appear within the smallest clause that contains the reflexive, and structurally “above” the reflexive. The syntactic specifics they depend on which syntactic theory you’re adhering to, but luckily they don’t really matter here; there’s no possible coreferent anywhere within the sentence, so any standard definition of Principle A will label the sentence ungrammatical.

Turning from this syntactic jungle to the Fijian jungle, a few pages later the Expedition lands on an island and hikes to its peak:

“Almost two years at sea had left them ill-prepared for such a demanding hike. ‘I have seldom witnessed a party so helpless as ourselves appeared,’ Wilkes wrote, ‘in comparison with the natives and white residents, who ran over the rocks like goats.'”

Again, it’s obvious that this is a non-standard usage, since no first-person plural noun phrase appears in the sentence to justify the reflexive.

Now, I’ve been marking these as non-standard rather than incorrect, and there’s a reason for this that is more than a desire to be non-judgmental. These supposedly erroneous uses of reflexives are widespread — so much so that I’d argue they’re at least borderline acceptable in many people’s forms of Informal Spoken English. That means that they ought to be explainable, that there ought to be some option in the rules of English that allow you to consider these uses acceptable without having to change much else in the language. I’m going to speculate for the rest of this post, so feel free to bail out here.

But before you bail, let me just brag about where I get to read.

Here’s my idea, which I don’t think is novel.* Reflexives are allowed only when, in some sense, there’s a sufficiently salient coreferent for the reflexive. Salience is standardly assessed syntactically, meaning that a coreferent appears structurally above the reflexive, and close enough to remain salient when the reflexive appears. But there is pragmatic salience as well, for people and things who haven’t been explicitly mentioned but remain prominent in the discourse all the same. And what is more pragmatically salient than the speaker? In both of these cases, it seems that the speaker is thinking of themselves as sufficiently salient to trigger the reflexive.

My intuition is that there are more instances of inappropriate reflexives for first person (myself, ourselves) than second person (yourself), and more of either than for third person (himself, herself, itself, themselves). I did a quick corpus search on COCA for sentence-initial As for *self, and the intuition wasn’t fully borne out; as for myself was the most common, but combined as for him/herself showed up almost as often (64 to 60), and as for yourself only registered one instance. So maybe I’m totally off-base on the specifics.** But something is going on that allows so many people to view reflexives as standard in positions that we don’t expect to see them, and like this or not, that needs explained.

—
*: If you know of any references to discussions about this issue, please share. I’m not primarily a syntactician, and didn’t see anything in a cursory search of the literature, but I really doubt this discussion hasn’t been had before.

**: I think the as for *self construction may be a special case. Most of the third-person uses look to be about how some third party views themself, and while one can state one’s own introspections and speculate about a third party’s, it’s a little bit weird to tell someone their own introspections. That could artificially deflate the second-person counts.

I think the best explanation of this construction may be as an indicator that we are switching mental spaces, if you’re familiar with that theory. Saying as for Xself establishes a new mental space focused on X and their inner workings or opinions, rather than the more generic mental space of the rest of the conversation. Sorry, I’m really going down a rabbit hole here.