4

Are the following two examples grammatical?

  1. Write it I have.
  2. Wrote it I did.

Consider as possible contexts:

  • They said that I have to write it, and write it I have. -- (for #1)
  • They said that I wrote it, and wrote it I did. -- (for #2)

MAIN QUESTION: What reasons are there to think that examples #1 and #2 are grammatical or ungrammatical?

These two examples involve preposed verb phrases, which could be used as a form of topicalisation. But what I'm wondering about, here, is the grammar in relation to the forms of the verb WRITE.

MINOR QUESTION: As a very small secondary concern, is a comma required after it in those two examples, and why or why not? (See the various comments by Curiousdannii below.)

Araucaria - Him
  • 47,146
  • 2
    In standard English, the verb in (1) would be written (although the word-order is odd); but (1) is Yodish, not English. (2) is more standard, particularly with the insertion of a comma. – Andrew Leach Apr 24 '15 at 14:20
  • 1
    @AndrewLeach That word order there is fairly common. It just requires the action being considered to have been mentioned in the recent discourse. A bit like, She asked me to make canape's and canape's I made!** It looks odd here because there is no previous discourse to make it look normal. – Araucaria - Him Apr 24 '15 at 14:22
  • Ah well, if there is parallelism involved, "She asked me to write it, and write it I have" then that's different. – Andrew Leach Apr 24 '15 at 14:25
  • 3
    This reminds me of a line from Phantom of the Opera - "And what is it that we're meant to have wrote? (er, written)" – Kevin Apr 24 '15 at 14:27
  • Without commas I would say they are not grammatical. Using commas for topicalisation is a very strong convention. – curiousdannii Apr 24 '15 at 14:43
  • 2
    @curiousdannii That's odd. I have been looking at topicalisation a lot recently and none of the grammars I have been reading used commas at all for their examples... If you look at Andrew leaches example above, I think it would look very odd with a comma: "She asked me to write it, and write it I have" versus "She asked me to write it, and write it, I have" – Araucaria - Him Apr 24 '15 at 14:46
  • @Araucaria Hmm, maybe that's okay. But those are different sentences, and the topicalisations begin with conjunctions. The bare phrase "write it I have" sounds very wrong without a pause, so it needs a comma. – curiousdannii Apr 24 '15 at 14:49
  • @curiousdannii Maybe, not sure ... "I asked you to write it"" - "Well, write it, I have" seems odd to me too ... What do you think? (Maybe I'll stick that in the question too.) – Araucaria - Him Apr 24 '15 at 14:54
  • 8
    Just because Yoda talks like this, it does not follow that the rest of us should. – GEdgar Apr 24 '15 at 15:56
  • 3
    Since I have write it* and I did wrote it* are both ungrammatical, so too are these. They should be written it I have and write it I did. No comma is necessary; however, their use is limited, mostly to being a response to such questions as have you written it? and did you write it? – Anonym Apr 24 '15 at 16:39
  • The second could be interpreted as perfectly legal if you assume it's two sentences -- "[I] wrote it. I did [write it]." There is intentional redundancy in a statement of this sort, so interpreting it this way is not a "stretch". (The first example is a bit harder to justify this way, but is less idiomatic in any case.) – Hot Licks Apr 25 '15 at 12:32
  • The OP seems to believe that "pre-posed verb phrases" means that any verb phrase can be preposed at will. This is not the case. In particular, the verb phrases in the examples given are preposed when they ought not to be. Hence they are, as noted, ungrammatical. Whether Yoda talks this way is irrelevant; Yoda, like Sherlock Holmes, Mr. Micawber, or Lady Mondegreen, is a fictional character and presents no syntactic data. – John Lawler Apr 26 '15 at 16:18
  • 1
    @JohnLawler No, I don't think that - and I don't think that I imply they can be pre-posed at will, do I?. There's loadsa well know info-packaging considerations to be taken into account. However, I don't see why the sentences are ungrammatical per se. It's just that there's no indication given about what information/sentences preceded these, so there's nothing explicit to mitigate their usage. I don't agree that they're ungrammatical due to the pre-posing of the VPs. Certainly, A:"He denies he's told them". B: "But told them he has!" would be acceptable. I fully agree about Yoda. – Araucaria - Him Apr 26 '15 at 16:36
  • "Grammaticality" is a measure of how hard it is to find a describable potential context in which an utterance might reasonably be uttered with intent to communicate. These require quite a lot of ingenuity in reconstructing past conversational events from presuppositions, and there is a point of diminishing returns. Note, btw, that you need an initial But to pull off But tell them I have!, in addition to the presupposed dialog. It's an idiomatic construction, and, while they often provide information about how things used to be, they're not usually a good place to start a general analysis. – John Lawler Apr 26 '15 at 16:51
  • 2
    @JohnLawler But it's an analysis of this kind of info-packaging that I'm asking about. What I'm most interested in ,though, is the form write instead of written in the first example, and wrote instead of write in the second. This seems a bit similar to cleft constructions to me such as What I've done is write him a letter or What he does is tell(s) them what happened. [I think the daft Yoda comments have been making you more skeptical!] – Araucaria - Him Apr 26 '15 at 17:43
  • 1
    Oh, so is "Write it I have" ungrammatical? Or is "Written it I have" ungrammatical? :D . . . You already have 3 close votes; your question is probably too hard for EL&U, or is it too easy for EL&U? Maybe the answer is easily obtained from a dictionary? Maybe "Grammar Girl" has the answer? Surely there's a duplicate thread somewhere, so that this thread can be closed? – F.E. Apr 26 '15 at 18:28
  • 6
    Why close it? It seems to be getting interesting discussion, even if it's happening in an officially incorrect place. If you start with a preposed VP, you're gonna be ignoring the final, and totally predictable, auxiliary stranded at the end of the sentence. By the time you get there speaking, you may well have gone on to the next sentence and just substitute whatever auxiliary verb sounds good; and by the time your addressee gets there in parsing, they're likely to be fobbable off with any old auxiliary, because Aux is all they're waiting for to close the parse. Is that "grammatical"?? – John Lawler Apr 26 '15 at 18:34
  • @JohnLawler: "tell them I have" should be grammatically incorrect as far as I can tell, while "told them I have" would be grammatical but unusual. – user21820 Apr 27 '15 at 05:59
  • 2
    Well, all the parts are correct, but they're arranged in the wrong order. If you want to say that's grammatical, it's your language. But syntax is far more important in English grammar than inflected word choice; there's only 8 or 9 inflectional morphemes left in English, and they're not likely to last much longer. Syntax, on the other hand, consists of hundreds of individual rules, with more appearing every day. – John Lawler Apr 27 '15 at 15:31
  • @F.E. I think that this question is guilty of being too omphaloskepsis; i.e naval gazing! It appeals to a very limited audience, and hence the two downvotes and the three votes for closure. – Mari-Lou A Apr 28 '15 at 08:41
  • @Mari-LouA No, it got three downvotes because people thought it was about Yoda speak, which it aint! – Araucaria - Him Apr 28 '15 at 09:05
  • 2
    Then perhaps you should mention it isn't, in the question. Just a thought. – Mari-Lou A Apr 28 '15 at 09:06
  • 1
    @Mari-LouA I thought about that, but I think it would have the reverse effect. It would automatically be followed by a whole load of comments about Yoda speak, I'm sure :) So I tried to stave that off by giving the grammatical term for this kind of phrase order. – Araucaria - Him Apr 28 '15 at 09:10
  • Perhaps a link on its definition, for us who are ignoramuses and too lazy to look it up themselves. – Mari-Lou A Apr 28 '15 at 09:11
  • 1
    @Mari-LouA Incidentally, a very similar question was asked on ELL by a language learner, so I'd say the appeal's not as limited as it might seem! :-) – Araucaria - Him Apr 28 '15 at 09:12
  • 3
    @Mari-LouA Preposing is just when some phrase which normally occurs later in a ssentence is moved to the beginning. This is sometimes called topicalisation. This just means that we take some piece of info that we're already talking about and put it at the beginning so that the rest of the sentence is like a comment on that topic. Incidentally, Italian speakers do this a lot in Italian and so overuse it in English! – Araucaria - Him Apr 28 '15 at 09:14
  • The thing is, Topicalization does not apply to "some phrase which normally occurs later". It's a rule with conditions, and in particular Topicalization does not apply to verb phrases; only nouns. That rule I thought you knew is Topicalization from I thought you knew that rule. But *Know that rule I thought you did is not Topicalization, and is not grammatical English. There's no syntactic rule that produces it. – John Lawler Apr 30 '15 at 15:09
  • 2
    Prepositional phrases and noun phrases are topicalizable; basically they're all noun phrases, just with or without a prepositional tag. Adverbs are fronted by a separate rule that also allows niching in between words; it's not topicalization unless there's a specific argument for treating the adverbs as noun phrases. And verb phrases are part of the engine, and the auxiliary verbs in the verb chain are sposta come before the verb they qualify, if they're gonna have any meaning. Stranding an auxiliary is not the same as stranding a preposition. http://www.umich.edu/~jlawler/VPguide.pdf. – John Lawler Apr 30 '15 at 16:04
  • 1
    @JohnLawler Yes, fair do's. That's an overextension of the label. My aim was just to stave of Yoda comments! – Araucaria - Him Apr 30 '15 at 16:04
  • 3
    I'm very interested to hear an answer that draws on actual sources of some kind besides just blindly asserting that it's right or wrong. I have no idea about the grammatical validity, but just as a native speaker both of the example phrases sound just fine (assuming there's some context -- saying "write it I did" out of the blue would probably be a bit weird.) They're also both more natural than either the versions with commas or the version with the allegedly "correct" grammar (written/write). – Lynn May 02 '15 at 02:38
  • 1
    @Lynn F.E. has now given us one of those below :) – Araucaria - Him May 02 '15 at 17:55
  • Riddle me this: How do you toss a ball when you have no ball? :D Do you suppose there's a (new) wikipedia page for that? – F.E. May 03 '15 at 17:08

3 Answers3

16

Both are examples of hyperbaton. You can read more about it here, . In their current form, both sentences are ungrammatical. Correct them for tense as follows.

Write it I have.

should be

Written it I have.

Next,

Wrote it I did.

should be

Write it I did.

Once corrected for tense, both sentences can be acceptable English usage, despite the unusual word order.

See too the question and answers to Why is this a hyperbaton? One of the answers even mentions why so called Yoda-speak can be grammatically acceptable!

Community
  • 1
Ellie Kesselman
  • 5,992
  • Just checking, so you're saying that write is ok with the have in the first example? It doesn't need to be written? (Nice links btw) – Araucaria - Him Apr 25 '15 at 13:19
  • @Araucaria Oops, no! I need to edit that! – Ellie Kesselman Apr 25 '15 at 13:44
  • 1
    Are you sure that "Write* it I have"* is ungrammatical? – F.E. Apr 26 '15 at 18:31
  • 1
    @F.E. Yes, I am sure. As a stand-alone sentence, it is ungrammatical. As part of two clauses, broken up by punctuation, it can be grammatical. This is an (awkward) example: "During the three years it has taken to write it, I have had some wonderful experiences." That is a completely different sentence structure though. – Ellie Kesselman Apr 27 '15 at 02:09
  • @EllieKesselman Oh, I was hoping you were going to state for the record that they definitely were grammatical :( – Araucaria - Him Apr 27 '15 at 21:52
  • @HotLicks Don't you mean 35 years ago? – phoog Apr 30 '15 at 19:10
  • 1
    @phoog - Good catch! Yoda wasn't in the original movie, but in the second in the series, in 1980. – Hot Licks Apr 30 '15 at 22:32
  • Six down votes? Someone doesn't like Yoda :( – Ellie Kesselman May 02 '15 at 10:07
  • You originally had my up-vote in the bag, because of the links and the interesting term (it doesn't quite answer the question entirely, but it's interesting), as you might have guessed from my first comment here. I didn't mind the Yoda reference either, especially as you point out that this kind of transposition can be entirely grammatical. The reason you got my down-vote in the end was just that you deemed the 1st example ungrammatical (the 2nd's not so easy to determine). While I don't agree, it was really the fact that this claim wasn't supported and most grammars contradict you. – Araucaria - Him May 02 '15 at 18:32
  • @EllieKesselman And if you undid that part of the post, my downvote would convert to an upvote. – Araucaria - Him May 02 '15 at 19:27
4

Technically they are incorrect and should have been:

Written it I have. = I have written it.

Write it I did. = I did write it. = I wrote it.

It is the same reason we would say "Done it I have" rather than "*Did it I have".

The second one could for even more emphasis or rhetorical effect be phrased:

I wrote it, yes I did. ("did" here is a pro-verb, grammatically separate from the earlier phrase.)

People often make mistakes with the tense because the verb is rarely fronted in English. It is more common to front other things like nouns or adjectives:

Many books I have read.

Truly surprised I was.

I disagree that fronting verbs is invalid English because they are indeed used, though mostly found in poetry. Here are some non-poetry examples I found:

And escape he did. (some random written piece)

And come they did. (some BBC blog article)

It is also not a recent phenomenon (as some might claim started with Yoda).

And come he did. (History of English (1730))

To clarify my answer, I consider legal writings or news publications from the US or the UK as a reference point for standard English, because there is a whole spectrum of dialectical variants of English. For example in some places "I have went" is commonly used among the native English speakers, even though it would be considered incorrect in most other places.

Community
  • 1
user21820
  • 2,140
  • The note that it’s technically incorrect is misleading, since you go on to say (rightly) that it’s correct in poetic register. – Jon Purdy Apr 25 '15 at 13:01
  • @JonPurdy: No what the asker wrote is incorrect. What I wrote is correct. – user21820 Apr 25 '15 at 13:28
  • I misunderstood you. I agree with your correction of example #1, but while your correction for #2 is valid, it is not necessary. – Jon Purdy Apr 25 '15 at 23:17
  • @JonPurdy: I believe it is necessary when there is no comma separating the phrases, but modern English users will probably fail to use what used to be the correct tense. If you can find examples of "[preterite] [subject] did" used by any writers, then I will of course reconsider my view. – user21820 Apr 26 '15 at 03:18
  • I can find examples of structures like “(I was) right chuffed/pissed I was”, but this seems to be restricted to certain dialects, such as South London or the East End, so make of it what you will. – Jon Purdy Apr 26 '15 at 03:31
  • @JonPurdy: I consider those examples as grammatically valid, but they not of the same form but are of the form "[past participle] [subject] was". One can consider these constructions as a special case of either the form "[noun-modifier] [subject] was" or the form "[subject] [passive verb]" with the participle part of the verb fronted. – user21820 Apr 26 '15 at 03:42
  • 2
    "Technically they are incorrect and should have been:" <== Er, I was going to up-vote your answer, except for that. You are 3/4ths right. Well, maybe more than that due to the good info in the rest of your post. Okay, maybe your post is 85 percent right. Maybe 90 percent. Heck, your answer post is the best one here so far. +1 ! :) – F.E. Apr 26 '15 at 18:39
  • 2
    @F.E.: Thanks! But seriously, if you can find me evidence that those two sentence constructions in the question are established usage, I would edit my answer not to say "incorrect". =) – user21820 Apr 27 '15 at 05:56
  • This info might be helpful: “and build upon that, but build they have”: Should that 2nd “build” be “built”? :) – F.E. Apr 27 '15 at 16:56
  • 1
    @F.E.: Well I'm not (yet) convinced by the grammar you cite. What is their evidence? Can you find me a few examples of news publications from the US or Europe that actually use such constructions? Pazzo's answer does not provide much good evidence, as (1),(2) are useless and (3),(5) are blog or forum posts. (4),(6) are the only ones left, but the English language proficiency of the authors is unknown. – user21820 Apr 28 '15 at 04:45
  • @F.E.: I should add that this whole issue of grammatical correctness depends on our definition of "grammatical". It certainly is much stricter than "easily understandable by other users of the language", but what exactly is it? If we say that grammar is defined by the prescriptions of a grammar authority, then how do we select that authority? It would easily fall into the problem of over-prescriptive rules such as "no prepositions at the start of a sentence are allowed". But without an authority, how do we judge grammatical correctness? My view is that the rules should be as simple as can be. – user21820 Apr 28 '15 at 04:50
  • Are you familiar with what a reference grammar is? – F.E. Apr 28 '15 at 04:53
  • @F.E.: Perhaps I'm missing something, but didn't you cite that grammar to show that the construction is acceptable, even preferable in some cases? Anyway to finish what I was saying above, we already have well-established rules for the formation of tense in English, and the strict use of the participle with helper verbs including modals. I'm sure you would agree that "he has not really gave it a try" is incorrect, even though people do make such mistakes. A natural rule of fronting would account for "write it I did/will" but exclude "write it I have", and there is no simpler alternative. – user21820 Apr 28 '15 at 05:03
  • @F.E.: However, languages change and I won't be surprised if eventually enough people use those constructions that I consider ungrammatical and therefore the English-speaking population as a whole will form a new rule to accommodate them. Just like "if I was a bird, I would ..." which had previously been incorrect but now is accepted as correct. The constructions in question here might be in a similar transition phase but it is not clear to me. – user21820 Apr 28 '15 at 05:07
  • "A natural rule of fronting would account for "write it I did/will" but exclude "write it I have", and there is no simpler alternative." <== So, can it be assumed that you disagree with the 2002 CGEL's opinion here as to their grammatical judgment on something like: [25.i ] "He said he wouldn't tell them, [ but tell/told them he has ].", where they consider both versions to be grammatical in today's standard English? – F.E. Apr 28 '15 at 05:09
  • "Just like "if I was a bird, I would ..." which had previously been incorrect but now is accepted as correct." <== What? But that opinion of yours is way wrong. And it is another, different, topic. – F.E. Apr 28 '15 at 05:11
  • 2
    @F.E.: Oh? (Pardon the digression, but previously we would use a subjunctive there, and some older grammars even state that "if I was" is wrong when it is a counter-factual condition expressing a wish.) As for CGEL, I would disagree that it is grammatical in today's standard English, but perhaps it may be used in today's non-standard English (which would incidentally be among majority of English speakers such as on blogs or forums). I consider news publications or legal writings as a reasonable reference point for standard English, so I would be glad if you can find any examples from these. – user21820 Apr 28 '15 at 05:16
  • There's nothing wrong with having opinions that differ from what's already out there or that differs from the typical or conventional opinions. But, you'll probably want to be familiar with the existing ones that are held by the well-regarded linguists, and understand their argumentation, so that you can then argue against their opinions. The 2002 CGEL is a good starting point when looking for info and grammatical opinions on many topics, especially those w.r.t. today's standard English. It's not perfect, but it is a source that I'd expect a grammarian to use and to be aware their opinions. :) – F.E. Apr 28 '15 at 05:17
  • "As for CGEL, I would disagree that it is grammatical in today's standard English, but perhaps it may be used in today's non-standard English (which would incidentally be among majority of English speakers such as on blogs or forums)." <== I'm bowing out now. (Got too much other stuff to do.) – F.E. Apr 28 '15 at 05:19
  • @F.E.: Oh okay sorry to disturb you. Maybe you can ping me if you would like to chat some other time! =) – user21820 Apr 28 '15 at 05:21
  • 2
    @user21820 In some languages, such as French, there IS a standards authority for determining what is grammatical and what is not. For French, I think it is called the Academie Francais. For English, there is no such authority. Regardless, I up voted your answer, as I believe it to be correct. I don't know what a "serious grammarian" is, but I do know that EL&U user John Lawler is a bona fide linguist. He does not reference CGEL as canon. – Ellie Kesselman Apr 28 '15 at 11:37
  • 1
    @EllieKesselman: Ah that is interesting. Thanks for sharing! =) – user21820 Apr 28 '15 at 12:16
  • 1
    @EllieKesselman It doesn't matter what any self-appointed authority on what should be grammatical says. Grammatical is always what people do. That is what grammatical describes. That doesn't mean that people who speak a different variety of a particular language are ungrammatical, just that they speak a different variety of the language. – Araucaria - Him May 01 '15 at 00:34
  • 1
    @Araucaria: Yes I don't think Ellie is disagreeing with what you say at all, and neither do I. But I think we must not conflate different registers. Most dictionaries already indicate whether an entry is formal or informal, but that is only one aspect. Another aspect is the genre; some things that you find in poetry never appear in official documents. Yet another is the split between native and non-native speakers. You will find lots of people from certain countries saying "I have a doubt" when they mean "I have a question", but that does not make it correct standard English. Same for grammar. – user21820 May 01 '15 at 03:21
  • 1
    @Araucaria: By the way, I just saw http://english.stackexchange.com/q/184372/75136 which is an example of what I mean by ungrammatical English that is the norm in some areas. Unless you're going to say that "I have went" is grammatical, it is not so easy as just considering "what people do". =) – user21820 May 01 '15 at 05:57
3

Are the following two examples grammatical?

  1. Write it I have.
  2. Wrote it I did.

Consider as possible contexts:

  • They said that I have to write it, and write it I have. -- (for #1)

  • They said that I wrote it, and wrote it I did. -- (for #2)

ANSWER TO MAIN QUESTION: In the appropriate context, those two expressions (#1 and #2) would most likely be acceptable in informal or colloquial spoken English, and would be considered to be grammatical by many native English speakers--though, many speakers might consider the usage of "wrote" in #2 to be nonstandard for spoken and written English.

In each of the two sentences for the example contexts, the last coordinate clause has a preposed verb phrase (VP): the VP "write it" in #1, and the VP "wrote it" in #2. Both preposed VPs have verbs that have the identical shape of a verb used in an earlier clause in the sentence; and because of this, that could produce a rhetorical effect which would often be desired by the speaker.

For version #1 in its given context, there usually wouldn't be much doubt as to its grammaticality, since it would be considered to be grammatical in today's standard English according to the 2002 reference grammar CGEL.

For version #2 in its given context, it seems in my opinion to be acceptable, and it would probably be quite reasonable to consider it to be grammatical in today's informal English, and maybe even in today's standard English. But it might be a bit harder to present an iron-clad supporting argument for that position, because it seems that this specific type of construction isn't explicitly discussed in the general vetted grammar sources that are easily available, such as the 2002 reference grammar CGEL or the 1985 reference grammar by Quirk et al.

ANSWER TO MINOR QUESTION: Usually in those types of constructions, a comma is not used to separate the preposed element from the rest of the clause when that element is a complement (which it happens to be in the OP's two examples). Often it would be less acceptable, or even unacceptable, to use a comma in that situation.

Consider the original two examples:

  • They said that I have to write it, and write it I have.
  • They said that I wrote it, and wrote it I did.

and compare them to versions with a separating comma after the preposed element:

  • They said that I have to write it, and write it, I have.
  • They said that I wrote it, and wrote it, I did.

The original versions seem much better to me. In general, though, it will depend on the specific sentence and its context, and on the speaker or writer w.r.t. what they want to accomplish, as to whether or not the use of that kind of comma is acceptable.

Note: Another answer post which has info related to the OP's grammar question is: “and build upon that, but build they have”: Should that 2nd “build” be “built”?

LONG VERSION:

INFORMATION PACKAGING:

The OP's examples are using information packaging constructions. These are constructions that have words and phrases that are not in their canonical or expected "normal" order. Often, these constructions will move elements (word or phrases or constituents) around. Usually there are more constraints and restrictions on these types of constructions, w.r.t. their level of acceptability in the contexts where they are used, than there are on more "normally ordered" constructions.

The two examples (the full sentences which provide some context) are:

  • 1.b. They said that I have to write it, and [write it] I have. -- (for OP's #1)
  • 2.b. They said that I wrote it, and [wrote it] I did. -- (for OP's #2)

Both of the OP's examples (#1.b and #2.b) are using the information packaging construction of complement preposing (2002 CGEL pages 1372-82). And more specifically, they are using VP preposing, which are the VPs "write it" and "wrote it". The following are the more normally ordered versions that would correspond to them:

  • 1.c. They said that I have to write it, and I have [written it].
  • 2.c. They said that I wrote it, and I did [write it].

Notice that the verbs in these two normally ordered versions use the grammatical "written" and "write": "written" is the past-participle verb form for the perfect construction ("have written"); and "write" is the plain form of a verb that heads the VP that is the complement for the matrix verb "did", where the verb "did" is the auxiliary verb "DO" (not the lexical verb "DO").

[Aside: Notice that there's usually another normally ordered version that would also correspond to #2.b which wouldn't use the auxiliary verb "DO": "They said that I wrote it, and I wrote it", which uses the preterite "wrote" instead of the expression "did write". But in this specific case, it doesn't seem to work for the given context.]

In general, there seems to be two factors at work here which are influencing the form that the preposed verb will be taking:

  • One: the form that the verb would have if it was in the normally ordered construction.

  • Two: the form that would more closely match the shape of the verb form that was used by an earlier use of the verb.

As to which form would be more preferable in a specific context, that would usually depend on the context and also on the speaker or writer as to the rhetorical effect that they are trying to accomplish.

Since the OP's two examples are involving two different types of construction (perfect vs auxiliary-DO), I'll split the discussion into two major parts. I'll first discuss the OP's #1 example which involves the perfect construction, since it should be the less controversial of the two. And then after that, I'll discuss the OP's #2 example which involves the auxiliary-DO construction.

EXAMPLE #1: THE PERFECT CONSTRUCTION

  • 1.b.i. They said that I have to write it, and [write it I have]. -- (OP's original #1)
  • 1.b.ii. They said that I have to write it, and [written it I have].

In general, both versions are acceptable in today's standard English. But in this specific context, there would often be a preference for the first version (#1.b.i) which uses "write", and that was what was used in the OP's original #1 version:

  • 1.b.i. They said that I have to write it, and [write it I have]. -- (OP's original #1)

The OP's use of the plain form "write" of the verb would probably often be preferred here (over the past-participle form "written") because the second "write" would then match the shape of the first "write" which was used in the previous clause, and that would give a rhetorical effect which would often be desirable by the speaker.

The two factors influencing the selection of the form of the preposed verb, w.r.t. "written" versus "write", are:

  • One: the form that the verb would have if it was in the normally ordered construction -- for example #1, it is the past-participle "written".

  • Two: the form that would more closely match the shape of the verb form that was used in an earlier use of the verb -- for example #1, the plain form "write", which had headed an infinitival clause located earlier in the example sentence.

As to which form would be more preferable, then, in general, that will depend on the context and on the speaker or writer as to the rhetorical effect that they are trying to accomplish.

This specific issue related to the perfect construction is discussed by the 2002 CGEL on page 1381 (a related excerpt is provided near the bottom of this answer post).

EXAMPLE #2: THE AUXILIARY "DO" CONSTRUCTION

  • 2.b.i. They said that I wrote it, and [wrote it I did]. -- (OP's original #2)
  • 2.b.ii. They said that I wrote it, and [write it I did].

The OP's preterite verb form "wrote" might be preferred here (over the plain form "write") by some speakers because the second "wrote" would then match the shape of the first "wrote" which was used in the previous clause, and that would give a rhetorical effect which would often be desirable by the speaker. Though, some speakers might consider "wrote" to be ungrammatical here (i.e. nonstandard), and they might only consider "write" to be acceptable.

The two factors influencing the selection of the form of the preposed verb, w.r.t. "write" versus "wrote", are:

  • One: the form that the verb would have if it was in the normally ordered construction -- for example #2, the plain form "write", which heads an infinitival clause.

  • Two: the form that would more closely match the shape of the verb form that was used in an earlier use of the verb -- for example #2, the preterite "wrote", which was located earlier in the example sentence.

As to which form would be more preferable, then, in general, that will depend on the context and on the speaker or writer as to the rhetorical effect that they are trying to accomplish (and also on the register).

GRAMMAR INFO FROM VETTED GRAMMAR SOURCES:

The OP's two examples use complement preposing in their second clauses, where the preposed element is a verb phrase (VP). Usually the second clause will involve an auxiliary verb when the preposed element is a VP.

Here are some typical examples. The 2002 CGEL, page 1376:

  • [11.i ] I've promised to help them [ and help them I will ].

  • [11.ii ] It's odd that Diane should have said that, if [ say it she did ].

The preposed VP in [11.i ] is "help them", and in [11.ii ] it is "say it". Notice that the nucleus of the second clause in both examples ends with an auxiliary: "will" for [11.i ], and "did" for [11.ii ]:

  • help them I will
  • say it she did

where the nucleus of each of those clauses would be "I will" and "she did". (note: The nucleus of a clause is the rest of the clause that hadn't been preposed.)

Here are their corresponding versions that don't have the preposing:

  • A.i. I've promised to help them and I will help them.
  • A.ii. It's odd that Diane should have said that, if [ she said it ] / [ she did say it ].

But when the auxiliary verb is the perfect "have" and the preposed element is its complement, then both the past-participle form and the plain form of the verb are acceptable.

The 2002 CGEL page 1381:

Inflection with perfect have

A special issue arises when the preposed element is a complement of perfect have. Compare:

[25]

  • i. He said he wouldn't tell them, [ but tell/told them he has ].

  • ii. He denies he has told them, [ but tell/told them he has ].

Although have normally takes a past participle, it is the plain form of the verb that is preferred in [i ]. The past participle is preferred in [ii ], where it has been used in the preceding clause, but even here the plain form tell is acceptable.

NOTE: The 2002 CGEL is the 2002 reference grammar by Huddleston and Pullum (et al.), The Cambridge Grammar of the English Language.

Pages 1372-82 (2002 CGEL), which deal with complement preposing, are within chapter 16 "Information packaging" pages 1363-1447, and the major contributors to that chapter are Gregory Ward, Betty Birner, and Rodney Huddleston. Birner and Ward are the authors of the 1998 book Information Status and Noncanonical Word Order in English, and much of chapter 16 was based on their work.

NOTE: The 1985 Quirk et al. is the reference grammar by Quirk, Greenbaum, Leech, Svartvik, A Comprehensive Grammar of the English Language.

Community
  • 1
F.E.
  • 6,208
  • 1
    Am I right in thinking that part of the effect of this type of construction is to move the old info to the front of the clause so it links back to the previous sentence thereby putting the auxiliary at the end where it has more emphasis. The reason for our wanting to do this being that the auxiliary position tells us whether the clause is negative or positive - so the clause becomes emphatically positive or negative. ? – Araucaria - Him May 02 '15 at 16:52
  • @Araucaria Er, yes, those could be some of the reasons why preposing is sometimes done. Off the top of my head, I'd think so. – F.E. May 02 '15 at 17:01
  • @Araucaria If you want, you could add/weave the info that's in your comment into the answer post -- when you have the time. :) – F.E. May 02 '15 at 17:41