In English, generally speaking, an idea of conditional is expressed by a sentence with a dependent clause (and usually with the conjunction "if"):

  • If it rains, the picnic will be cancelled.

The dependent clause can be a non-finite clause:

  • Weather permitting, we will go there on foot.

In addition, being expressed by a relative clause is also possible:

  • Anyone who should do that would be laughed at.

Can we use several independent clauses to express an idea of conditional? YES. However, we may face much limitation.

  • Give him an inch and he will take a mile.

In this example, the first independent clause conveys the meaning of imperative. If the first independent clause does not convey the meaning of imperative, the whole sentence may be unacceptable.

  • Sound in body, sound in mind.

This sentence conveys the meaning of conditional paratactically. However, forms like this may only appear in idioms.

My question: Can we use several independent clauses to express an idea of conditional without too much limitation? (in certain languages, not limted to English)

  • Declerck & Reed (Conditionals: a Comprehensive Empirical Analysis, Mouton de Gruyter, 2001) treat these at '11.13 Paratactic Conditionals', pp. 401-407. I think you are mistaken in calling Give him an inch as an imperative: it does not enjoin the hearer to give, it speaks of such a gift by anyone. You may call it a "present subjunctive" if you adhere to certain grammatical sects. I'd just call it an ellipsis. But paratactics with or may take an imperative in the condition clause: "Stop or I'll shoot!" They may even take bare NPs in both clauses: "Your money or your life!" Commented Jun 8, 2016 at 0:47
  • @StoneyB thank you for your helpful comment
    – discenter
    Commented Jun 8, 2016 at 5:53
  • 1
    "Give him an inch and he'll take a mile" is clearly not an imperative. I once published a paper about these and associated constructions.
    – jlawler
    Commented Jun 8, 2016 at 16:41
  • 1
    @Cerberus: I wasn't trying to account for historical origins; they were as they were and I am innocent of knowledge of, or curiosity about, them. Nor do I know Dutch (I can't hear it in my head fast enough). And I no longer enjoy arguing whether something grammatical "is a" something else grammatical; the devil is always in the presuppositions, and decoding them is tiresome and uninformative at best. As for conditionality, you're talking about modals. Modality (along with negation and quantification) generates operators, with focussed (bound) elements. And ambiguities.
    – jlawler
    Commented Jun 9, 2016 at 18:25
  • 1
    @jla: All right, then it seems it was a misunderstanding and we don't disagree. As to terminoloy, it may ultimately be trivial, but it does serve to communicate information. If something goes wrong in communication, then terminology may be the cause. When you say imperative, I read that as a purely syntactic category, but you seem to be using it in a different, semantic sense. When you say elliptical, that makes me think of an historical change, or of a change that occurs in the speaker's mind when he forms a sentence: first he has x y in mind, then he omits y and finally utters only x.
    – Cerberus
    Commented Jun 9, 2016 at 18:56

3 Answers 3


Yes, we can. Here is an example from English:

  • Have you been in an accident? Call our team now on 07123 45678.

That second sentence is only meant to apply if you have been in an accident. Questions and conditionals are related in many interesting ways.


In Chinese, certainly. A paratactic vestige from Classical Chinese, it appears in many stock phrases:

(1) Dao gao  yi  chi,  mo    gao  yi  zhang.
    dao tall one chi,  devil tall one zhang. (chi and zhang are a traditional Chinese units)
    If (those with the) dao grows taller by a chi, the evil will grow taller by a zhang.
(2) Mei you  Gongchan  Dang,  mei you  Xin Zhongguo.
    Not have Communist Party, not have New China.
    If there were no Communist Party, there would be no New China.

These would sound more natural and less literary in normal speech if we add markers like dehua or jiu ('then'), but they are perfectly acceptable and normal in Classical Chinese.

  • P.S. Lest you should wonder, no I'm not from the linguistic division of the fifty cents party... The sentence was discussed a while ago on Language Log: languagelog.ldc.upenn.edu/nll/?p=3340 Commented Jun 8, 2016 at 16:39
  • I read your sentence examples carefully and think they may be not suitable. The first sentence is just an idiomatic expression, and the second sentence is just like English "no pain, no gain",which is also related to idioms(the form is productive in English). The syntactic property for dehua and jiu may be dubious. You translated jiu into English word then, but I think it may be regarded as a conjunction. By the way, what does " linguistic division of the fifty cents party" mean?
    – discenter
    Commented Jun 8, 2016 at 17:00
  • Do you mean this? This term seems derogatory and politically controversial. I think your concern is due to your second example.
    – discenter
    Commented Jun 8, 2016 at 17:36
  • There are other examples in the Language Log link if you're not convinced. I think 'no pain, no gain' cannot be compared to this one because 'no pain' and 'no gain' aren't clauses; 'mei you Gongchan Dang' and 'mei you Xin Zhongguo' are. / Are you sure jiu is a conjunction? I haven't looked deeply, but I don't think its syntactic properties resemble a conjunction at all. I'm fairly sure it's an adverb, judging by its position in the sentence... / Yeah, I was trying my hand at humour :P Commented Jun 10, 2016 at 7:10
  • If you want more examples from Classical Chinese, here are some from Pulleyblank (1995): 不奪不饜 (Mencius 1A:1), 不能更鳴,東徙,猶惡子之聲 (Shuoyuan 16.164). Commented Jun 10, 2016 at 9:20

Not an actual answer to your question, just noting that

Anyone who should do that would be laughed at

is not a regular conditional! Those "would" or "might" constructions are called counterfactuals, and are semantically not identical to material implication (as it would be the case with "if" sentences or the other kinds that you stated), but - at least in the theories I am aware of - rather have to be analysed with respect to possible worlds (roughly, could we imagine a world hopefully close to ours and in which the consequent is implied by the antecedent if it were true, since (as the name already tells) the antecedents usually are not true in the actual world we are talking about), or possibly by adding some more propositions to the antecedent until the consequent eveutally logically follows from the resulting set of propositions.
In the end, the analysis of "would" will directly or indirectly involve material implication to be interpreted as "if... then", but still, I'd like to have pointed out that this construction is semantically not of the same type as a classical if-statement.

  • But such constructions need not be counterfactual, or even tentative or hypothetical: Anyone who does that will be laughed at. Anyone + the restrictive relative clause is a predication semantically equivalent to a case-specifying condition. Commented Jun 8, 2016 at 0:55
  • Hm, yep, you're right. Didn't think of that. Commented Jun 8, 2016 at 8:07
  • Please edit this to fix all your typos!
    – curiousdannii
    Commented Jun 9, 2016 at 0:20
  • @curiosdannii Could you please specify which typos you mean? Your comments sounds like my post contains a lot of typos, but I only found "impled" and "porpositions" (just fixed them). Edit: also found "taht" and "sould". Hope it's okay now. Commented Jun 9, 2016 at 10:06
  • "In the end, the analysis of "would" will directly or indirectly involve material implication to be interpreted as "if... then". "If ... then" is not interpreted as material implication, and no theories of 'counterfactuals' present them as material implications. Commented Jun 17, 2016 at 13:24

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.