We had that interesting discussion on the use of "come" in
What is the difference between the use of "come" here and the use of "were" in "if I were" in BrE? Both have been uttered and written, and both are instances of constructions otherwise difficult to encouter. I understand that you see an important difference between them, other then just in the number of instances, but I'm not sure what the difference is.
Last edited by birdeen's call; 22-Dec-2010 at 23:03.
Response to Post 37:
Philo: The claim that 'were' in If I were a bird, I would fly away. is a form of the past subjunctive, but that 'had' in If I had wings, I would fly away. […] is not […] violates that most basic of all axioms, namely that grammatical function is not reckoned on the basis of coincidences of form.
5jj: As I don’t make that claim, I feel no need to respond to it.
Philo: I would ask you to consider the following examples of erroneous grammatical reasoning: […]
5jj: I don’t see that the first two are particularly relevant to this discussion, but I have considered them. I agree with you. They are mistaken. Actually, I agree with you about the third. But As I have never made the mistaken conclusion, this is, once again, irrelevant.
Philo: I think that a discussion of sound analytical procedure is in danger of being usurped by the - albeit very interesting - quite unrelated topic of EFL pedagogy.
5jj. I concede that pedagogic matters led me into analysis. My own thoughts, supported by my reading, suggested that the traditional analysis of such forms might be inappropriate to English (BrE, at least) as is it used today.
Response to Post 39:
5jj: So a major grammatical category seems to have ceased to exist in the course of a few decades.
Philo:No, it ceased to exist approximately one thousand years ago. The authors you cite simply seem not to have realized the fact.
5jj: I could make the facile (but possibly true) response that the relevance of the past subjunctive as a category ceased some time ago, but some people simply seem not to have realised the fact.
Philo: 1. The number of instances is irrelevant. Either a construction still exists or it doesn't. And this one does!
5jj: How many times do I have to repeat that I have never claimed that it doesn’t exist? All I am saying is that I don’t think it is particularly helpful to label this ‘past subjunctive’.
Philo: 2. I think you'll find that AmE speakers, who, in general, regard counterfactual 'was' as substandard, constitute a far more sizable proportion of the world's native-user population than do BrE speakers (most of whom in any case still recognize and accept 'were' in this sentence-position as a grammatically correct, and often stylistically preferable, alternative.)
5jj: I have been talking about BrE, not AmE. I accept 'were' as grammatically acceptable. Whether the majority of Br speakers accept it as 'stylistically preferable' is a matter of opinion.
5jj: Could you please give me an example of an indicative conditional 'were' with a first or third person subject? I have not come across this idea before.
Philo:The term 'indicative conditional were' would therefore, in that sense, be a contradiction in terms. I apologize if this was in any way misleading.
5jj. No apology necessary. I just wondered if I were missing something.
Philo, I believe that we do not disagree on much except whether there is any value in labelling the forms we are discussing ‘past subjunctive’, or whether there is a simpler way of looking at modern English. Unfortunately, we keep getting side-tracked because you point out the flaws in claims I have not made.
I think I have an idea that is at least worth considering, particularly as it is an idea arrived at after years of thought, discussion and reading. As a result of my reading, I have discovered that ‘my’ idea has been discussed in serious academic circles for some time. If you don’t think it’s worthy of serious consideration, that is entirely up to you, but please stop misrepresenting what I am saying.
Incidentally, as a product of my own age, background and education, I normally use the construction that you call the subjunctive (as I observe the shall/will difference and the usage of whom). I do not think that entitles me to consider myself more ‘correct’ than the majority (my opinion) of speakers of BrE who do not use these forms. I also know many literate and articulate speakers who do not use these forms. I note that you write of what you call ‘conditional ‘were’ that it “is the only form that many of its most literate and articulate speakers actually consider formally correct in that particular sentence-position.” ‘Formally correct’. Now I wonder what that means, exactly.
I would agree that it is probably easiest for learners to opt for the "were" form in standard conditional structures of the "if X were the case, Y would happen" kind, since it will never be "wrong" (though it does puzzle some native speakers who are unfamiliar with the grammar, and so may be considered wrong).Since, however, such locutions are considered either incorrect or, at best, poor style, by many AmE speakers, learners are probably best advised to use the distinctively subjunctive forms where appropriate.
On the other hand, the "was" subjunctive can be found in writers who are not generally considered poor stylists (recently I noticed instances in Berkeley and A.J. Ayer, for example).
Also, possibly, there is a distinction between the two forms (at least for some BrE speakers) which is not often discussed, but which can often be detected in e.g. radio interviews with public figures:
1. If it wasn't for the fact that X is the case, Y would happen.
2. If X were the case, Y might happen.
In the #1 structure, Y is presented as remote from actuality; whereas in #2, Y is still possible. My impression is that some BrE speakers (probably unconsciously) do tend more towards the "were" form when entertaining an open hypothesis, and "was" for a closed possibility.
(I use the example of interviews with public figures, because they involve frequent hypothesising.)
Not a professional ESL teacher.
It might also be simply having learned If x were- some people who say If I were also say I wish I was, which is a bit confusing.
I immediately confess that I have not read this entire thread, but I still might be able to contribute by saying that I can remember when I was studying German as a young man, how satisfying and easy it was to deal with the modals in that language.
The Romance languages, by comparison, that use verb endings, i.e. inflections for many of these differences, I, as a native English speaker, did not find so easy.
I have always found that German is SO friendly to an English speaker.
There is, however, also the argument based on the common sense linguistic principle that potentially ambiguous constructions are naturally less acceptable than unambiguous ones.
Consider the following:
 If I was a policeman, I would wear a policeman's uniform.
 If I were a policeman, I would wear a policeman's uniform.
Of these two sentences, ostensibly identical save what you term a 'difference of milieux', only  is unambiguously a (counterfactual) second conditional.  could occur as a variant zero conditional relating to the past, as in the following context:
When I was young, I used to like playing various roles in my local amateur dramatics group. Sometimes, I would be a doctor, sometimes a fireman, and sometimes a policeman. If I was a doctor, I would wear a white coat. If I was a fireman, I would wear a fireman's uniform. If I was a policeman, I would wear a policeman's uniform.
There is, however, no way in which  could be substituted here, since the past subjunctive form 'were' - which only ever refers to present or future counterfactuals - would have no place in this past narrative.
Therefore, on grounds of clarity of expression alone, hypothetical 'were' is clearly preferable to hypothetical 'was'.