Never end a sentence with a preposition! That is something up with which I will not put!

Written by Thomas Hoffmann

From a typological point of view, preposition placement in English is an extremely interesting area of variation: most languages either require a preposition to be dragged along to the start of a clause by a displaced complement (as in the German relative clause das Haus, [in dem]i ich _i lebe vs. *das Haus, [dem]i ich [in _i] lebe) or they obligatorily leave the preposition in its clause-internal position (as in Swedish: huset [som]i jag bor [i _i] ‘the house that I live in’ vs. * huset [i som]i jag bor _i; adopted from Dekeyser 1990: 103). English, on the other hand, allows the preposition in clause-initial position (1a; a phenomenon known as ‘pied-piping’) as well as clause-internally (1b, where it is said to be ‘stranded’ or ‘orphaned’ without its complement):

 (1)     a.  preposition pied piping:
the house [in which]i I lived _i

          b.  preposition stranding:
the house [which]i I lived [in _i]

 As with virtually all cases of syntactic variation, the choice between preposition stranding and pied-piping is of course not a random one. While (1a) is considered more formal, (1b) is seen as the more informal choice. Yet, this is only part of the story. Besides several other linguistic constraints (including clause type, the syntactic function of the prepositional phrase, the type of displaced complement and processing effects; for a first overview cf. e.g. Pullum and Huddleston 2002: 627), the variation has also been affected by prescriptivism: Never end a sentence with a preposition! is one of those few (misguided) stylistic rules that a lot of people still seem to recall from their grammar lessons (though, funnily enough, as I show in my book Preposition Placement in English: A Usage-based Approach neither modern usage-guides nor the Early-Modern prescriptivists are as categorical in their discussion of preposition stranding). As illustrated by (2), this categorical avoidance of preposition stranding can lead to very awkward results:

 (2)     This is something up with which I will not put
(attributed to Sir Winston Churchill; cf. Pullum and Huddleston 2002: 629)

 The above sentence is a much-quoted joke attributed to Sir Winston Churchill, who is supposed to have commented on a particularly stilted evasion of preposition stranding with the words “This is something up with which I will not put” (cf. Pullum and Huddleston 2002a: 629). The sentence is intentionally ungrammatical since not only the preposition with has been pied piped to the front of the clause, but also the particle up. Interestingly, though the sentence is clearly ungrammatical, the quote is well-known by many speakers of English and, from a usage-based perspective, can thus be seen as something like a fixed, idiomatic phrase.

Interestingly, a Google search[1] on British internet sites revealed that the quote is not just a frozen idiom anymore (emphasis, i.e. bold font, italics and brackets, TH):

(3)      a.  Inconsistency is
[the sort of infelicity] up with which [we] will not put.

             b.   In Nottingham this is
[a situation] up with which [we] will not put! eccentd.php?filename=00000088.txt

          c.  I am sure that, if there is a great deal of disruption in Victoria street, the formidable Lady Porter will be banging on somebody’s door in order to let the Minister know that that is
[the sort of thing] up with which [she] will not put. pa/cm198990/cmhansrd/1990-05-22/Debate-10.html

          d.  But, faced with a choice between Ruweished and the “chaos” and “insecurity” of Iraq, the Palestinians have finally found
[a refugee camp] up with which [they] will not put.…/news/2003/06/01/wsteyn01.xml&sSheet=/news/2003/06/01/ixnewstop.html

          e.  I hope that the Minister will tell the Chancellor that this is
something up with which [she] will not put

          f.   Baroness Blatch: My Lords, I have to agree with the right reverend Prelate. This is a scourge involving
[the exploitation of children] up with which [we] [should] not put. pa/ld199596/ldhansrd/vo951207/text/51207-01.htm

          g.  one should have the honesty to say simply that some things are
[metaphysical nonsense] up with which [one] will not put.

          h.  This is
[the sort of wishy-washy thinking] up with which [purists] will not put.

          i.   to the preservation of the English language, and for drawing your readers’ attention to
[our shoddy apostrophe], up with which [he] will not put. 2001/11/17/298048.html

As (3) shows, people are starting to appropriate the quote in various ways by replacing fixed elements by schematic slots that can be filled by various syntactic elements. The most variable slot of the sentence is the antecedent position, with only (3e) having something as in the original quote. In addition to that, the subject position is also filled differently (we in (3a,b,f), she in (3c,e), he in (3i), they (3d), one (3g), and purists (3h)). Furthermore, (in 3f) the modal should is used instead of will. In contrast to this, the elements up with which … not put remain constant.

Obviously, I am not claiming that particle pied piping as in the Churchill quote is becoming grammatical in English. Examples such as (3) are very limited in frequency and are also only used by a particular (educated) stratum of society ((3c,e,f), e.g., are from the official Hansards of the British Parliament). What the examples show, however, is how a fully substantive idiom such as this quote can be progressively schematized by speakers. Usage-based Construction Grammar (cf. Bybee 2006; Goldberg 2006) offers a straightforward account of phenomena such as these. Taking form-meaning pairings as the basic building blocks of grammar, such approaches highlight the fact that the repeated use of the quote (together with general cognitive principles such as analogy and pattern recognition) can lead to the emergence of abstract patterns. In the case of (3), the emergent construction can e.g. informally be represented as follows:

 (4)           Form:        [Nantecedent] [[up with which] [NPsubj] Vmodal [not] [put]]RC
Meaning:   ‘NPsubj strongly opposes Nantecedent

 As (4) shows, the more schematic construction now has a meaning in which the subject of the relative clause strongly voices his or her disagreement with whatever is mentioned in the antecedent noun. On top of that, the construction in (4) obviously also has a prototypical discourse function, since it is mainly used to mock someone else’s opinion or behaviour.

Usage-based Construction Grammar approaches are thus well-equipped to capture even the most peripheral of linguistic phenomena such as (4). However, as I show in Preposition Placement in English: A Usage-based Approach they are also capable of giving descriptively and explanatorily adequate grammatical analyses of all core phenomena associated with preposition placement in English. Using corpus and introspection data from L1 British English as well as L2 Kenyan English as corroborating evidence, I show that both varieties are subject to the same processing constraints and actually share a great number of constructions. In addition to that, however, the corpus as well as the experimental data indicate that all these constructions are less entrenched and less productive in the second language variety.

Yet, while I myself am pretty convinced by what Usage-based Construction Grammar has to offer, I am aware that many researchers from other frameworks might not be that interested in the particular analysis I provide (which by the way is a combination of Usage-based Construction Grammar and Constructional Head-Driven Phrase Structure Grammar).

In order to make the empirical results accessible to researchers from other frameworks, the Construction Grammar analysis is deferred to the last chapter and the preceding discussion of all empirical data is presented in a terminology as theory-neutral as possible. Hopefully, therefore, Preposition Placement in English: A Usage-based Approach is a book up with you will want to put.


Preposition Placement in EnglishThomas Hoffman’s new book, Preposition Placement in English, is available now from Cambridge University Press. Click here to find out more.



Bybee, Joan. 2006. From usage to grammar: The mind’s response to repetition. Language 82: 711-33.

Dekeyser, Xavier. 1990. Preposition stranding and relative complementizer deletion: Implicational tendencies in English and other Germanic languages. In: Sylvia Adamson, Vivian Low, Nigel Vincent and Susan Wright. Papers from the 5th International Conference on English Historical Linguistics. Amsterdam, Philadelphia: Benjamins, 87-109.

Goldberg, Adele E. 2006. Constructions at Work: The Nature of Generalisation in Language. Oxford University Press.

Pullum, Geoffrey K. and Rodney Huddleston. 2002. Prepositions and prepositional phrases. In: Pullum, Geoffrey K. and Rodney Huddleston, eds. 2002. The Cambridge Grammar of the English Language. Cambridge: Cambridge University Press, 597-661.

[1]        The search was conducted by looking for the string “up with which * put|puts|putting” on British “.uk” Internet sites on March 27th 2005.

Leave a Reply




You can use these HTML tags

<a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>