Compared with the first type of conditional subordinate clause, are the latter t...

This question is a continuation of my previous question:

How can the following three types of conditionals be distinguished?

Here I would first like to present some of my own thoughts on my previous question.

First, I feel that my previous question was not really resolved satisfactorily. It remains in a somewhat unresolved state, and I have not found a method that can clearly distinguish these three types of conditional subordinate clauses. I also think that using “distinguished” in the title of my previous question was not a very good choice, because it failed to accurately reflect my actual purpose and intention in analyzing the classification mechanisms of conditional clauses, and instead contained too much of my own personal motivation.

I do not know whether anyone can find literature that provides a better analysis of the distinction among these three types of conditional clauses than the references I provided in my previous question. In any case, I personally believe that distinguishing these three types of conditional clauses inevitably involves epistemology or epistemic logic to some extent, and that such a distinction is difficult to accomplish purely from a syntactic perspective. It will inevitably contain a certain degree of subjectivity and may vary from person to person.

Rather than asking “how can these three types of conditional clauses be distinguished?”, I think it may be more appropriate to ask what kinds of antecedent information, what kinds of consequent information, and what kinds of personal beliefs involved in reasoning make a conditional clause more likely to be categorized as one type rather than another.

Personally, I would classify a conditional clause as a “content conditional” when the following condition holds: whenever the antecedent occurs, the consequent occurs as well, and the occurrence of the consequent does not require my personal action to trigger it. Of course, this classification would cause some of the conditional clauses classified as other types in the literature I quoted in my previous question to be classified differently by me.

Now let us move to the main topic of this question.

Among the following three types of conditionals:

• content conditionals
• epistemic conditionals
• speech-act conditionals

At the grammatical level, is it true that the first type of conditional subordinate clause can generally be focused by the “it’s...that...” cleft construction, whereas the latter two types are much more difficult to focus using this construction?

Here I use the examples provided by the author of the paper cited in my previous question to illustrate my point. In other words, at the syntactic level, I'm fairly sure we can transform

[1a] If Mary goes, John will go. (content conditional)

into

[1a'] It’s if Mary goes that John will go.

&

[1b] If I were president, I’d sell the White House’s Limoges china to fund bilingual education. (content conditional)

into

[1b'] It’s if I were president that I’d sell the White House’s Limoges china to fund bilingual education.

while it is much more open to question whether one can felicitously transform

[2a] If John went to that party, (then) he was trying to infuriate Miriam. (epistemic conditional)

into

[2a'] It’s if John went to that party that he was trying to infuriate Miriam.

and

[3a] There are biscuits on the sideboard if you want them. (speech-act conditional)

into

[3a'] It’s if you want them that there are biscuits on the sideboard.

Is this analysis correct?

P.S.

The classifications of the conditional clauses above are those given by the author of the paper. Personally, however, I might classify:

“If I were president, I’d sell the White House’s Limoges china to fund bilingual education”

as a speech-act conditional, and:

“If John went to that party, (then) he was trying to infuriate Miriam.”

as a content conditional.

• Dealing with the 'speech-act conditionals' first (and note that the terminology seems far from standard):

There is no true protasis in this class of sentence. The if-clause is not directed (directly) semantically towards the event, condition etc in the statement in the matrix clause (which can therefore hardly be called an apodosis). The if-clause is a caveat referring to (the following list may be incomplete)

(a) the relevance of the utterance                                                        or

(b) the accuracy of the wording of the matrix clause,                    or is an embellishment,

(c) a politeness marker (cf 'please') inserted to oil the smoothness of communication, an attention-grabber marker (eg 'if I might have your attention!) inserted to shepherd the audience, etc.

Examining the three subsets individually:

• "There are biscuits on the sideboard, if you want them,"

• “His new design is skeuomorphic, if that's the right word,” and

• "Are you busy this evening, if you don't mind me asking?/if I might ask?" //

• "You look lovely this evening, if I might say so."

• • '[I]f you want them' is merely a closish paraphrase of 'Maybe you'd like some?'/'Just a suggestion'. There is an overlap with politeness here, also.
• • '[I]f that's the right word' is a discourse-orientated hedging marker, conceding possible inaccuracy in the mode of delivery.
• • '[I]f you don't mind me asking?' is just a politeness marker. cf 'I hope I don't sound too busybodyish.'

Transforming these three sentences to 'It's if … that …' sentences,

• • "It's if you want them that there are biscuits on the sideboard" is at best very awkward, and would be taken as meaning if anything the sense of the content conditional, where desire or need magically gives rise to the presence of the otherwise absent biscuits. Note that the original (minus the comma) has this possible sense too.
• • “It's if 'skeuomorphic'/it is the right word that his new design is skeuomorphic” needs so much jiggery-pokery to justify being used that it's best said to be unacceptable along with furiously sleeping colourless green ideas.
• • And the statement-type pragmatic – politeness – marker (the interrogative can't work) "It's if I might say so that you look lovely this evening" is nonsense. So yes, from these examples, speech-act conditionals resist (or forbid) the 'It's if … that …' transformation.

…...............................

Next, looking at 'epistemic conditionals'.

[2a'] “It’s if John went to that party that he was trying to infuriate Miriam.” This is just about possible, but is not semantically equivalent to “If John went to that party, (the reason is that) he was trying to infuriate Miriam.' Rather, the 'It’s if … that …' attempted rewrite, rather than giving the reason for the possibility … if realised ... in the protasis that the original gives, seems to pick out one candidate from a list under consideration as being the true telltale that yes, John was indeed trying to infuriate Miriam.

So here again, epistemic conditionals resist/disallow the 'It's if … that …' transformation.

…...............................

Finally, considering 'content conditionals'. We can look at the general case here.

“If X happens, then Y will also happen”. (+ other timeframes) is not exactly equivalent to “It's if X happens that Y will happen.”

The first sentence shows implicature, inevitable consequence. X ⟹ Y. But it does not require that Y cannot happen despite X not having happened. However, the 'transform' defaults to “It's if and only if X happens that Y will happen.” Not a true paraphrase. OP's [1a] (giving a sufficient condition) and [1a'] (giving a necessary condition) are not semantically equivalent, and the idea of transformation cannot hold.

The 'transform' is certainly grammatical (if clunky), but semantically unacceptable.

So the postulated test for content conditionals, distinguishing them from the two other types mooted here, 'At the xxxxxxx level, is it true that the first type of conditional subordinate clause can generally be focused by the “it’s...that...” cleft construction, whereas the latter two types are much more difficult to focus using this construction?' fails. I've deleted the 'grammatical' restriction in the original since tests involving the generation of grammatical resultants by suggested transformations are meaningless if the resultants don't have the same meaning. Transformations in English preserve meaning.

.....................................................................................................

I'd go further and say that

(a) The three mooted types of 'conditionals' cannot be reconciled into a single superset (having 'if' in a sentence hardly makes it a true conditional sentence: this sentence does).

(b) 'Speech-act conditionals' are very different in nature from true conditionals, and are, rather, addressee- (politeness, attention-grabbing {'if I might have your attention'} etc) or discourse- (quality ... accuracy? relevance?) orientated pragmatic markers.

(c) It is unlikekly (partly from the above) that a diagnostic syntactic test to differentiate the 'three types' mentioned above (I think the classification is unsound anyway) can exist.

(d) Even between content and epistemic conditionals (a more realistic classification), there are likely to be grey areas where competent speakers might disagree over which set a given sentence should be placed in.