Subject-Dependent Inversion in The Economist

« previous post | next post »

The Economist article whose first sentence I quoted in this post about inverting subject and verb in dialog reporting frames ends with a textbook example of a very different kind of inversion:

Harder still than understanding the significance of such barbarism may be accepting that it can never be completely prevented.

This is a case of what The Cambridge Grammar of the English Language (p.1385) calls subject-dependent inversion. It involves switching places between the subject of a main clause and some dependent from within the verb phrase (often a complement of the copula). In the above example, the subject is the subjectless gerund-participial clause accepting that it can never be completely prevented. The adjective phrase harder still than understanding the significance of such barbarism is a predicative complement licensed by the copular verb be. They have been switched. The most straightforward order of constituents would have been this:

Accepting that it can never be completely prevented may be harder still than understanding the significance of such barbarism.

The inverted version that The Economist published could be criticized on the grounds that the sequence may be accepting could easily be misparsed as an instance of may followed by the progressive aspect form of accepting (as in something like It seems that they may be accepting my article).

The writer chose the construction in order to mention the thing that claimed to be hardest only at the very end, to make it the last phrase in the article. So, as journalists often do, he or she (Economist articles are normally anonymous) chose to use subject-dependent inversion.

There's nothing grammatically wrong with it. But it has been ridiculed, rather brilliantly: as Ben Zimmer pointed out (on Language Log three years ago), Wolcott Gibbs did a wonderful piece in (yes) The New Yorker, back in in 1936, mocking the over-use of subject-dependent inversion and other elements of journalistic style in Time magazine under the editorial regime of Henry Luce.

"Backward ran sentences until reeled the mind," wrote Gibbs; "Where it all will end, knows God!"

Gibbs was exaggerating a little. But as you see, only a little. Journalists really do write sentences sort of backwards sometimes. And slightly harder than describing what they do is explaining why they do it.


  1. Leonardo Boiko said,

    June 21, 2010 @ 1:04 pm

    I think it is a somewhat charming journalistic trope, though it make things a bit less clear. (Which is ok with me.)

  2. Dhananjay said,

    June 21, 2010 @ 1:09 pm

    It seems to me that the "still" of "harder still" plays a part in licensing the inversion in this sentence; i.e., without the "still", the uninverted order is better, but with the "still" the inverted order is better to my ear.

  3. Morten Jonsson said,

    June 21, 2010 @ 1:16 pm

    I find the sentence rather elegant, in an eighteenth-century sort of way, with its two neatly balanced abstractions. It's fancier than journalism usually is, but not at all hard to parse–I immediately saw a clean break after "barbarism" and wasn't tempted to see it as going with "accepting."

  4. empty said,

    June 21, 2010 @ 1:53 pm

    Baseball commentators fall into the same odd habit: "Diving to his right is Pedroia."

  5. John said,

    June 21, 2010 @ 2:09 pm

    I think it makes prodigious sense, in that it allows the journalist to control the ebb and flow of tension in his piece down to the sentence level.

  6. Harold said,

    June 21, 2010 @ 2:10 pm

    Why do they do it? I remember reading that the founders of Time consulted the Iliad and the Odyssey as a model for style and borrowed such features as compound epithets and inversions such as the above to create a lively and distinctive style that would have mass appeal.

  7. Brett said,

    June 21, 2010 @ 2:18 pm

    I agree that with the "still" there, the original sounds a lot better, and this got me thinking about this use of "still." ("Yet" is used similarly, and most of my thoughts apply to it as well.) Normally, one does not use a still in a direct comparative with "than." I would never say something like:

    ?X is harder still than Y.

    Instead, the most usual way of saying this using "still" is:

    X is hard. Y is harder still.

    The use of "still" usually seems to imply a comparison to something already mentioned. In the Economist case, I think the reversal of the order of the sentence gets around this restriction. Apparently, when the subject-dependent order is reversed, it is easier to accept reversals in the orders of other objects.

    I also noticed that this kind of "still" sounds better coming after a comparative with an "-er" suffix than one with "more"—"harder still" seems preferable to "more difficult still." With "still" first, my preference is reversed—"still more difficult" over "still harder." This seems to have something to do with my thinking of "still more" as a unit. If I said "still more difficult," I would strongly emphasize the grouping "(still more) difficult" pretty strongly. However, I don't know whether these preferences are peculiar to me, or more general.

  8. David L said,

    June 21, 2010 @ 2:18 pm

    Maybe I've been around journalism too long, but I find the original Economist phrasing easier to understand, and more emphatic in tone, than your re-ordering of it.

    In any case, I don't think this kind of inversion is particularly a journalistic invention. For example there's "Better to have loved and lost than never to have loved at all," or "how sharper than a serpent's tooth it is to have a thankless child."

    Notorious Timestyle little lamented, mostly forgotten by serious journalists now is. But occasional inversion similar to the example you cite is a standard rhetorical device, long practiced by real writers as well as journalists.

  9. Phil said,

    June 21, 2010 @ 2:26 pm

    I suspect some sports commentators make it a habit because it allows them to describe the action as it happens, quickly check a roster and then add the name.

  10. Karen said,

    June 21, 2010 @ 3:51 pm

    Of course, the "straightforward" sentence may be all wrong for the information structuring of the paragraph.

  11. Michael Straight said,

    June 21, 2010 @ 4:06 pm

    To me that's just good writing. You start the sentence with the idea you're transitioning from and end the sentence with the idea you're transitioning to (or leaving the reader with, in this case).

    In isolation, it may seem a little weird, but in context it usually makes good sense of the way the ideas are flowing.

  12. Mark F. said,

    June 21, 2010 @ 5:51 pm

    I think the truly straightforward (or, I would say, pedestrian) way of writing the sentence would be something like

    Accepting that such barbarism can never be completely prevented may be even harder than understanding its significance.

    It wouldn't have made sense for GKP to make these editorial changes because it would have muddied the point he was making, but this sentence is probably better because it doesn't bury the subject of "can be prevented" in the second half. On the other hand, it's still a very weak sentence.

  13. Richard said,

    June 21, 2010 @ 6:35 pm

    Ah, I get it now, 'subject–dependent'.

    Like an idiot I read the headline first as introducing a post about inversion that depends on the subject (hmm … sounds kind of interesting), and then one about inversion of something that depends on the subject (also kind of interesting). Third time right … and still interesting! I should have known better than to try to parse the headline before reading the post: that'll teach me.

    [I now see that hyphens and en-dashes appear the same in the drafting box but correctly in the preview. I guess it's the same when authoring posts. Anyway, where's the fun in not leaving potential ambiguities? This is LL, after all!]

  14. Rubrick said,

    June 21, 2010 @ 9:25 pm

    I found the sentence as written clear and elegant; of course, as an isolated example it can hardly serve as evidence for or against overuse.

    On the other hand, I had considerable difficulty parsing this stretch of the post: "…quoted in this post about inverting subject and verb in dialog reporting frames ends…"

  15. Mark Mandel said,

    June 21, 2010 @ 10:24 pm

    Still untrained in the Force you are, Luke.

  16. David J. Littleboy said,

    June 21, 2010 @ 10:39 pm

    As a Japanese to English translator, I have a lot of sympathy for these sorts of constructs. Since Japanese is verb final and has modifying clauses preceeding the thing modified, a fairly good first draft translation can be created simply by reversing the sentence and plugging in the right lexical terms. But since Japanese style calls for long sentences and fewer sentences per paragraph, lots of rewriting is required to get the logic of the paragraph to flow as well as in the original. Usually, the right (and easiest!) approach is to chop things up into shorter sentences. But I, at least, get a joyous feeling of megalomaniacical power-tripping when I can come up with a monster long English sentence that has everything in it the Japanese did and still reads easily in English.

  17. Lazar said,

    June 21, 2010 @ 11:03 pm

    Like David L, I find The Economist's wording clearer and more pleasing.

  18. Harold said,

    June 22, 2010 @ 1:17 pm

    I should have added that Time's style was a horror. And good riddance. But they did rely on ancient models — or so they claimed. Of course writers have been doing that since forever — with Cicero being the preferred one by far.

    I agree that the economist example here is innocuous.

  19. Marco said,

    June 22, 2010 @ 7:28 pm

    Reactionary Anglophilia.

  20. Alon Lischinsky said,

    June 23, 2010 @ 3:32 am

    Journalists really do write sentences sort of backwards sometimes. And slightly harder than describing what they do is explaining why they do it.

    Well, in this case at least, it looks like a regular case of marked topic fronting. I gather that the author had already mentioned the difficulties involved in "understanding the significance of such barbarism", and so chose to delay mentioning those involved in "accepting that it can never be completely prevented" to give them clausal and discourse-level focus.

    The prose is certainly more elaborate than in the usual colloquial examples of this (e.g., Yiddish fronting: My brother-in-law, he wants to be), but I believe they can be analysed similarly.

RSS feed for comments on this post