One thing I've noticed about the usage of ending a sentence with a preposition is how similar the construction is to German separable verbs. With German separable verbs, the prefix is often a preposition when taken by itself, such as "mit-kommen" = accompany, but is "come with" if it's broken into parts. In a regular declarative sentence in German, this would be written with the prefix at the end, such as "Paul kommt Erich mit" = "Paul accompanies Erick."

My question is, aside from the parallel, is there an actual causal relationship here? Is the common usage of putting a preposition at the end come from an old German construction with separable verbs?

  • 4
    To simplify the matter considerably, English was a Germanic dialect that was later hugely influenced by French. It's been a hybrid for over 1000 years.
    – TimR
    Commented Oct 17, 2018 at 22:07
  • 1
    Hi Paul. Welcome to EL&U! This is a really good question. I don't know the answer but it seems intuitively very likely given that grammatically English is a Germanic language. I'll be very interested to see the answers here. +1 Commented Oct 17, 2018 at 22:10
  • 6
    FYI: German mitkommen is intransitive, so *Paul kommt Erich mit is an impossible sentence.
    – KarlG
    Commented Oct 18, 2018 at 1:49
  • 3
    The correct form for a statement would be Paul kommt mit Erich mit. It’s the answer to Mit wem kommt Paul mit? → Mit Erich.
    – Holger
    Commented Oct 18, 2018 at 7:38
  • Thanks for these response. My German is very rusty!
    – Paul
    Commented Oct 18, 2018 at 11:59

5 Answers 5


This won't be an answer, exactly, but more of a pointer towards other sources.

First of all, terminology. What you are talking about is called preposition stranding. It seems to be very rare (although not nonexistent) outside Germanic languages. How exactly to analyze it within Germanic languages, including whether there is a unified treatment, remains an open problem. And I will stop there, and just give a bunch of sources.

On the Linguistics Stack Exchange: What motivates/allows preposition stranding in English, but disallows it in other languages, like Mandarin?

On Wikipedia: Preposition stranding.

And then come papers and such.

This Master's thesis has a good literature review.

This book, The Germanic Languages, discusses the topic of preposition stranding.

English as North Germanic discussess it, too.

On the historical development of preposition stranding in English

Case Theory and Preposition Stranding

Early Germanic preposition stranding revisited

Preposition Stranding and Resumptivity in West Germanic

Preposition-Stranding and Passive

Preposition Stranding, Passivisation, and Extraction from Adjuncts in Germanic

And many more.

  • 2
    Great references. Knowing the term preposition stranding really unlocks a lot of items I couldn’t find before. Thanks!
    – Paul
    Commented Oct 17, 2018 at 23:52
  • 1
    @Paul Sure, no problem! Commented Oct 17, 2018 at 23:52

Old English had a more flexible word order than modern English. However, I don't see any evidence that it ever used verb, object, preposition order as in German.

Object before verb (OV)

Because of its strong tendency to be object before verb, OE allowed the order preposition, object, verb:

Þis hé spræc on Iudea-lande: ðær wæs án eowd of ðam mannum þe on God belyfdon on ðam leodscipe.

This he spake in the land of Juda: there was a fold of men who believed in God in that nation.
The Homilies of the Anglo-Saxon Church by Ælfric, translated by Benjamin Thorpe

See also my answer here about believe in vs believe on. One commenter (KarlG) mentioned that this word order is "still strictly observed in modern German, a bit less strict in Dutch".

Preposition Stranding

OE also had sentences that ended in prepositions:

gað ge beforon; ic eow cume æfter.
Gregory the Great, Dialogues, Book 1

(Literally: "Go ye before; I you come after")

However, according to The Ban on Preposition Stranding in Old English "the phenomenon of preposition stranding is highly restricted in OE". The paper points out some examples of sentences where Modern English would likely end a sentence with a preposition, but OE wouldn't. For example:

Drihten, þu þe gecure æt fæt on to eardienne
Lord you yourself chose that vessel in to live
‘Lord, you chose that vessel for yourself to live in’
The Blickling Homilies

Another paper summarizes the rules for preposition stranding in OE:

[P]reposition stranding was possible in Old English (OE) only when the object was a pronoun, or in relative clauses introduced by the complementizer þe ‘that,’ while its possibility became greatly expanded in Middle English.
On the Historical Development of Preposition Stranding in English

Inseparable-prefix verbs

OE also had verbs that were prefixed with a preposition. A few of these words even made it to Modern English. These verbs, of course, can't be split:

Old English generally did not possess phrasal verbs as they are found in Present-Day English. They did exist, although they were rare. Much more common in Old English was the inseparable-prefix verb, a form in which the particle was attached to the beginning of the verb. These Old English prefixed verbs are directly comparable to current phrasal forms. For example, in Present-Day English, there is the monotransitive verb “to burn” and then the phrasal monotransitive “to burn up.” Old English had “bærnan” (to burn) and “forbærnan” (to burn up). The prefix “for-” remained affixed to the verb and could not move as modern particles can. Such Old English compound verbs were also highly idiomatic, in that the meaning of the compound form did not necessarily reflect the meaning of the root. Denison provides “berædan” as an example because it meant “to dispossess”, while its root verb, “rædan”, meant “to advise”. The phenomenon still survives today in the participle “forlorn”, as well as the verb “understandan”, which does not in Present-Day English mean “to stand underneath (something)”, but idiomatically “to comprehend”.
The Historical Rise of the English Phrasal Verb


The stub of an answer: it is believed that, in Proto-Indo-European, prepositions and prepositional prefixes originated in adverbial words. Cf.:

1481 Caxton Reynard (Arb.) 65 Men may wel lye whan it is nede and after amende it.

This use of a preposition as an adverb still exists, and it has probably always existed, in many Indo-European languages and their predecessors. The same construction as above, with a 'preposition' used adverbially at the beginning of a a clause, is also possible in e.g. Latin and Ancient Greek. Adverbs are generally very flexible as to their position in a clause, so she came after is more or less similar to after, she came.

Ancient Greek also had separable verbs (tmesis) that worked a bit like German, as in a palace in the goddess habited (freely and weirdly translated from Odyssey A, about the nymph Calypso: nêsos dendrêessa, thea d'en dômata naiei) alongside a more common she inhabited a palace (ennaiei dômata), although separable verbs become rarer as Greek progressed into the classical age and beyond.

I believe all of these separable verbs, phrasal verbs, adverbs, and prepositions ultimately originate in the same type of adverbial words from Proto-Indo-European. They were later adapted to prepositions, stranded 'prepositions', prefixes, etc.

Fun fact: case endings are also speculated to have originated in adverbial words, though I believe the evidence is not yet conclusive.


@lingisticturn's answer shows you quite an exhaustive list of literature about this topic. However, as a native German speaker from Austria I want to add something.

mitkommen is a verb, it is not the preposition mit with the verb kommen.

However, there is a geographical difference in splitting and positioning certain adverbs containing mit or von. One can observe, that Austrians and Bavarians do not split apart damit, davon and others. North Germans split them apart and put the second part at the end of the sentence.

Austrians: Damit habe ich keine Freude.

North Germans: Da habe ich keine Freude mit.

In English: I am not happy with this. or This I am not happy with.


Austrians: Davon habe ich schon gehört.

North Germans: Da habe ich schon gehört von.

In English: I have already heard of it. or This I have already heard of.

  • Very interesting! Thanks for this (+1).
    – Paul
    Commented Oct 18, 2018 at 11:14
  • "That with have I no friend." I love German! Commented Oct 18, 2018 at 13:20
  • 1
    @LightnessRacesinOrbit If you want to translate word-by-word, then it is joy (Freude), not friend (Freund). Commented Oct 18, 2018 at 13:27
  • @rexkogitans Ah bollocks :D Commented Oct 18, 2018 at 13:31

In English usage, even in formal English usage, it is by no means 'incorrect' to end a sentence with a preposition. This supposed rule is attributed to John Dryden(1631-1700) who was influenced by the fact that in classical Latin such a thing is not allowed (https://en.wikipedia.org/wiki/John_Dryden).

As (Fraser, Plain Words, HMSO 1973) notes: Sir Winston Churchill "once made [a] marginal comment against a sentence that clumsily avoided a prepositional ending:'This is the sort of English up with which I will not put'". In that sentence 'put up' is a phrasal verb and unlike German trennbare Verben the prepositional 'up' definitely should not go at the end of the sentence.

A fine example of prepositions at the end of a sentence is this exclamation from a man who has just dropped a cufflink inaccessibly behind a cupboard:"Come out from down in under there!"

  • 5
    This doesn’t answer my question. I am asking whether there is a historical relationship between the English construction of a preposition at the end of a sentence and German separable verbs.
    – Paul
    Commented Oct 17, 2018 at 21:54
  • 1
    I am sorry that I did not make myself clear. The answer is no.
    – JeremyC
    Commented Oct 17, 2018 at 22:01
  • 2
    @JeremyC It doesn't say anything about how or why English allows prepositions to be separated from their complementsin your answer, or whether this comes from Old German (or whatever German it might have come from ...) Commented Oct 17, 2018 at 22:13
  • 2
    @JeremyC And I don't believe that the answer is no. I'm really sure it isn't ... Commented Oct 17, 2018 at 22:17

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.