I’m sure Molly knows the answer to this one, Ralf. I think we should let him post the explanation – especially since he didn’t seem to think it was important to answer Volcano’s question to begin with.
.
Considering that should[color=green]'ve is the generally more accepted abbreviation, 500 times more common and less reminiscent of a typo - yes.
I know. If only our learner’s knew as well.
Basically, Amy, if you feel, and I imagine it is a feeling, that “should’ve” is “extremely uncommon even in informal written English” and you don’t want to question your feeling, there’s no place to go with this, right?
when considering typos, colloquial quotes and inept journalists.
Indeed:
If only our [color=red]learner’s knew as well.
Welcome back, R.
I’m glad I could provide proof. Typos happen and are registered and accepted by databases. Abundantly.

Considering that should[color=green]'ve is the generally more accepted abbreviation, 500 times more common and less reminiscent of a typo - yes.
Vigilant as ever, R. Thanks.
Do you feel that should’ve is “extremely uncommon even in informal written English”?

I’m glad I could provide proof.
Proof of you being back?
Hi Molly
Here is a comparison for you.
These are some of the search results I got for the usage of “shouldn’t” and “should’ve” (in COCA):
shouldn’t:
- Magazine: 3164
- Newspaper: 2959
- Academic: 558
should’ve:
- Magazine: 77
- Newspaper: 83
- Academic: 5
Would you say that one of those two contractions is far more commonly used in written English than the other one is? I would. And, yes, I do consider the difference to be extreme.
By the way, I was unable to find any useage results at all for “couln’t’ve” in COCA. I suppose I might have done my search incorrectly, though. There must be at least a couple examples of “couldn’t’ve” somewhere in informal written English – especially in something that was written with the intention of mimicking spoken English.
.
.
Oops, Amy’s off. Now we’ll get no rest. :lol:
Would you say that one of those two contractions is far more commonly used in written English than the other is?
Indeed I would. So that’s what you meant by “extremely incommon”. Don’t you think you should be comparing “should’ve” with “shouldn’t have”, “should have” and “should not have” though? Comparing it with “shouldn’t” alone doesn’t really tell us much.
should n’t have [v?n] - 1456
should not have - 677
should 've [v?n] - 692
should have [v*] - 14987
Would you consider that difference to be extreme?
By the way, I was unable to find any[color=blue] useage results at all for “[color=red]couln’t’ve” in COCA.
I’m not surprised.
Better luck on Google: 8,900 English pages for couldn’t’ve.
And 119,000 English pages for shouldn’t’ve.
Very convincing. If you google ‘fysiks’ you get 10,100 hits. But you should’nt (276,000 hits) use it in your classroom unless you feel like advocating fry fysiks - free physics.
Hi Molly
Would you say that “I’d’ve” is more commonly used in spoken English or in written English? Before you go traipsing off to the corpora, decide what your native-speaker gut tells you first.
After that, check out the results in both the BNC and COCA. What do you make of the results? How accurate do you think these corpora results are?
.
Before you go traipsing off to the corpora, decide what your native-speaker gut tells you first.
Well, I don’t have one of those, even though your shepherd tells you I have. So…
Would you say that “I’d’ve” is more commonly used in spoken English or in written English?
What’s the point of that question?
How accurate do you think these corpora results are?
Based on? Compared with?

Very convincing. If you google ‘fysiks’ you get 10,100 hits.
I only got this:
1,070 English pages for fysiks.
What did I do wrong?
Results 1 - 10 of about 10,500 for fysiks. (0.14 seconds)
You must have done something wrong.

You must have done something wrong.
Hm, but what?
M. has used Advanced Search and set the language to English.
But his Google totals are in any case misleading: those results include cached versions of the same page, versions in user profiles, quotes from one page to another, etc.
MrP
After that, check out the results in both the BNC and COCA. What do you make of the results? How accurate do you think these corpora results are?
BNC
I 'd 've
Spoken register 31
Fiction register 18
…
COCA
Spoken register 1
Fiction register 45
What do you think, Amy?
And what do you think of these result from the COCA and the BNC?
I 'd have [v?n]
COCA - 847
BNC - 605
So, combining those with these:
Spoken register 31
Spoken register 1
Could we begin to ask whether AmEng speakers prefer the part contracted form “I’d have + verb” to the fully contracted form “I’d’ve”.
I am not a native speaker.
“Who wrote this?” is the default form or commonly used form.
But why is an auxiliary verb NOT used in “Who wrote this?” ? Could you please give me a link explaining this type of questions?
Which one is correct?
Who installed this? / Who did install this? or what is another default or correct structure of this question?
those results include cached versions of the same page, versions in user profiles, quotes from one page to another, etc.
Isn’t it the same with Ralf’s 10,500 hits?
M. has used Advanced Search and set the language to English.
Why would I need to search for any other language?