Language Log

« previous post |

Email from J.P.:

I've recently had a similar reactions to the phrases "Lived Experience" and "Are you kidding me right now?", and in those cases, various sources of counts over time validated my impressions ("Memetic phrases", "Lived Experience").

In the case of "spends his/her/their time on", Google Books Ngrams shows a history of two long-term rises:

But the rate of increase is relatively gentle, and the current rise began around 1990, so this graph doesn't really explain J.P.'s reaction.

Several further possibilities come to mind, as in all such cases.

One possibility is that a wider range of syntactic frames would show a more rapid recent rise — "to spend PRP\$ time on", "spending PRP\$ time on", etc.; or that closer attention to the syntax, such as the role of the on-phrase, would pay off.

Another possibility is that Google's overall sample of published books is the wrong place to look. Perhaps J.P. is reacting to developments in a particular choice of authors, or to a particular strain of social and mass media, or even to personal communications of various sorts.

And finally, we could focus on J.P.'s observation that "seemingly increasingly, it's used in a derogatory or critical way, as if to say that to spend the time in this/whatever way is stupid". Since emotionally charged experiences are more memorable, this might change J.P.'s overall impression of phrasal frequency.

The billion-word sample in the COCA corpus comes from the time period 1990-2019, so it might give us a baseline. In that corpus, "spends her time on" occurs once, in a frame that's definitely not derogatory:

A writer has a vision, and it precedes language: there are no words for it. She spends her time on earth in search of the right words and the right rhythms for them, that is, the words and rhythms that will convey her vision.

The phrase "spends his time on" occurs twice, once in a derogatory context and once not:

You are an irrational close-minded fundamentalist conspiracy theorist who spends his time on Christian websites reading what only the most addle-brained morons over on Fox News would ever accept as actual fact.

Still, I enjoy my newfound freedom and so Dwayne's cabana has become a sort of office for me. I claimed my spot on his well-used but extremely comfortable one-time blue, now dusty gray, sofa early. Being more of a phone guy, Dwayne spends his time on his back deck/dock and conducts business outdoors as long as it isn't raining or hailing and sometimes even if it is.

Note that in all three cases cited so far, the on-phrase is basically a locative describing where someone spends their time, not what they spend their time doing, which is specified later in the sentence…

The phrase with the plural pronoun "spend their time on" occurs 34 times! Many of the examples also involve locative on-phrases:

Patrollers spend their time on skis.

Wrestlers spend their time on the wrestling mat trying to overpower, control, and use move manipulation against someone of their equal weight.

We are using the term " street kid " to denote those children who identify with the street; that is, they spend their time on the street rather than at school or with their families.

The gay beach at Post 8 is obvious from the rainbow flags, but this area is also home to another social group: young, apparently unemployed men who spend their time on the beach practicing Brazilian jujitsu, working out, and starting unprovoked fights.

Only a few are clearly derogatory or at least evaluative, e.g.

The Danish poor have more computers and TV's than everyone else; they prefer to spend their time on entertainment; pirated cable TV, pirated software, porn.

This function's status in the future is uncertain given the rise of a generation that would "rather text than talk" (Turkle 2015, 22), and the preference among some middle-aged "couples" in this study who sat opposite each other, never speaking, preferring to spend their time on their phone surfing the web and texting.

So maybe J.P. is reacting to a rise of such phrases in which on specifies what someone is spending time doing, not where they're doing it; and also in which the writer or speaker clearly disapproves of the referenced activity.

After framing a clearer hypothesis in this space, it could in principle be tested. But unfortunately, this is not as easy as counting words or word-sequences, and falls into an all-too-common category of text analysis.  Given a relevant data source with large-enough amounts of date-indexed text, we can select adequate random samples of the relevant phrase(s) over time, and use human judgment to classify the relevant aspects of the syntax, semantics, and pragmatics. With a bit more time and effort, we can train and test an automatic classifier. It's much easier to do this now that it was a century ago — compare Ellegård 1953 to Ecay 2015 — but it's not yet trivial…

 

 

 

August 28, 2025 @ 8:53 am · Filed by Mark Liberman under Language and culture

Permalink