My mulling was sparked by an email from Dave Carlyon, an astute observer of language, who pointed out that something seems to be afoot with the term button-down shirt. Does the description refer to the collar or the entire shirt? The answer: Originally it was about whether the collar buttoned down, but now for many of us, it is about whether the shirt itself buttons down from the collar to the waist. I will admit that I am in the latter group; and according to my informal polling of friends and family over meals and the like (this is one of the things that can happen when I start thinking a lot about a word), I am in good company: For most of these folks, the presence or absence of buttons on the collar is not the defining feature of a button-down shirt, sometimes called an Oxford.
One striking thing about this semantic shift: It is a change in the language happening right under our noses that most of us haven’t noticed. American Heritage Dictionary of the English Language online has both the meaning of the collar being fastened down by buttons and the meaning of the entire shirt having buttons down the front. Merriam-Webster online still has only the reference to the collar being buttoned down. Some clothing catalogs advertise button-down shirts with buttoned-down collars and others let the collars do what they will. And I haven’t heard any fuss about it.
Much space and attention is given to a limited set of changes in the language that grab our attention, from new slang like bae and on fleek, to shifts in prepositions (e.g., by accident to on accident), to changes in the meaning of words like literally (although the shift to mean ‘figuratively’ is not nearly as new as many people think), to Internet/texting acronyms like LOL and BFF, to pronoun mixing between subject and object (e.g., between you and I — also older than many realize).
Then there is all the change happening in the language without as much fanfare, if it is on the radar at all. Many speakers in the United States now make no distinction between the vowels in cot and caught (known by linguists as the cot-caught merger). The modal must, used to refer to obligation, is on the decline, as it is overtaken by its popular competitor have to (with need to also in the mix). The new irregular past tense snuck is sneaking into American English. Nouns can increasingly stack up in front of other nouns as modifiers with no possessive inflection (e.g., the University of Michigan English literature class textbook cover image). And the progressive is extending such that it can now be used in constructions like the past perfect passive progressive (e.g., the snow had been being shoveled before the sleet started).
It is up to us what we decide to do with a change in the language once we notice it. We can condemn it, as some usage-guide writers did with hopefully in the 1960s when it hit their radar that speakers and writers were using the adverb to mean “it is hoped” (speakers and writers had been doing so for three or four decades at that point). Or we can pause, marvel at human creativity with language, and consider critically whether we need to judge this change at all. It is no secret that I am in this second camp. A change in the language may not immediately — or ever — make the jump into standard usage, but that doesn’t make it an unfortunate or worrisome development.
I value all that a standard edited version of the language gives us, which includes some useful stability in written language across dialects and over time. As a historian of the language, I also know that the standard can tolerate some variation (e.g., proved/proven) and that it will change (e.g., in the 19th century it was standard to write the house is building rather than the innovative passive progressive the house is being built). Most importantly, we do well to remember that the written standard is just one variety of this language we call “English,” and a pretty buttoned-down variety at that.Return to Top