We have a rule, let's call it Rule W :

if clauses—the traditional rules. According to traditional rules, you use the subjunctive to describe an occurrence that you have presupposed to be contrary to fact: if I were ten years younger, if America were still a British Colony. The verb in the main clause of these sentences must then contain the verb would or (less frequently) should: If I were ten years younger, I would consider entering the marathon. If America were still a British colony, we would all be drinking tea in the afternoon. When the situation described by the if clause is not presupposed to be false, however, that clause must contain an indicative verb. The form of verb in the main clause will depend on your intended meaning: If Hamlet was really written by Marlowe, as many have argued, then we have underestimated Marlowe’s genius. If Kevin was out all day, then it makes sense that he couldn’t answer the phone.

What is the origin of this rule?

Merriam-Webster's Concise Dictionary of English Usage says (page 712):

...it can be seen that the subjunctive is likely to be found after the verb wish [...] after if, as if, and as though, and at the beginning of a clase or sentence stating something contrary to fact or hypothetical. Hall 1917 and Jespersen 1909-49 (vol. 4) observe that was began to compete with were in these contexts sometime around the end of the 16th century, but it apparently did not become frequent in this use until around the end of the 17th century.

MWCDEU makes no mention of "an occurrence that you have presupposed to be contrary to fact"; it just says that was and were are both used in hypothetical statements for the past 300 years. So where does Rule W come from? Is there still a dialect that makes a distinction between "if I were" and "if I was"? If there is, why doesn't MWCDEU mention it? Are the people who wrote Rule W using that dialect as a model, or are they just making stuff up?
1 2 3
Comments  (Page 3) 

If Hamlet was written by Marlowe, I'll eat my hat!

That's an interesting one. But does the doubt lie in the second clause, rather than the first? Cf.

1. If Hamlet was written by Marlowe, I'll have to include it in my forthcoming edition of Marlowe's plays.


The analysis is complex, because temporal relations and pragmatics are involved. Temporally, the condition lies in the past. Pragmatically, although the second clause has the appearance of a promise, the promis is not sincere - a form of irony, I suppose.

The textbook examples tend to talk about future events:

If you invite me to your party, I'll come.

The speech act embodied in the main clause depends on the condition; we know that:

a) the condition has not yet been met


b) the condition could still be met.

What have is: ?event

The very same is true for

If Hamlet was written by Marlowe, I'll eat my hat!

The only difference is that the event - instead of being obscure in the future - is obscure in the past. That the event is not - in effect - obscure is not a function of grammar; it's a function of the insincerity of the promise to eat a hat should the condition be met. But to pretend that the promise is sincere, you'll also have to pretend that the event is obscure. This even applies to present events:

If this elephant is doing maths, I'll eat my hat. (There must be a trick.)

Or to general statements:

If cats have wings, I'll eat my hat.

The tense in the if-clause actually follows normal indicative conventions for temporal representation (with present tense being the preferred tense for future events).


If either (a) the condition has failed to be met, or if (b) the condition cannot be met, we go into the "hypothetical mode". We're no longer operating with a "?event", but with a "-event". The forms used to express "-event" for a non-past event is identical with the past tense (with the possible exception of "was" for 1st or 3rd person singular). A "-event" in the past is expressed through a verb-form identical with the past perfect.

If you invited me to your party.../If this elephant was(were) doing maths.../If cats had wings...


If Hamlet had been written by Marlowe... [unless "[be] written" is interpreted statively, as you say; which would make it an attribute of the book, and thus a timeless, general statement akin to "if cats had wings"]


The usual separation of conditionals in type I, II, and III tend to assume a temporal symmetry; but as the hat-eating Marlowe example shows, such symmetry can't just be taken for granted. Similarly you might say:

If you had turned left when I told you, we wouldn't be lost now.

The symmetry assumption, I think, is what causing lots of confusion in analysis.

If you invite me to the party (future), I'll come. (future) --> symmetry
If cats have wings (general), I'll eat my hat. --> no symmetry

If you turned left when I tell you (future), we wouldn't get lost. (future) --> symmetry
If you had turned left when I told you (past), we wouldn't be lost now. (present) --> no symmetry

It is often possible to introduce symmetry by pushing the event into a subordinate clause within the if-clause, though:

If Hamlet was written by Marlowe (past), I'll eat my hat. (future) --> No symmetry
If it turns out that Hamlet was written by Marlowe (future), I'll eat my hat. (future) --> symmetry

(If it were/was the case that you had turned left.... - clunky but possible)


My mother tongue is German, and the conditional I, II, III concept had me terribly confused until I realised that the classification is somewhat arbitrary and doesn't take into account assymetrical temporal relations between main- and if-clause.
I am finding only discussions about whether to say "If I were" or "If I was." My question is why do I hear TV and radio personalities so often say, "If I AM Obama (or someone), I would do so-and-so." I hear it all the time.
Students: We have free audio pronunciation exercises.
When I attended school many moons ago, we were taught that when "if" was used then "were" was always appropriate in the sentence.

Many rules have changed since then.

God bless.