Two of my colleagues at the Leiden University Centre for Linguistics are working on the collocation of dat “that” and als “if” in sentences like Hij dacht dat als hij drukte, het luikje open zou gaan (“He thought that if he pushed, the trap would open”). In Dutch, the combination of two conjunctions is considered unacceptable, and writing manuals brand its use as a stylistic error.
The question was put to me whether this was also a usage problem in English, even though the construction is apparently regularly used. An English example they gave is the following:
De Saussure was so determined to have the mountain conquered that if he was not to be the one to do so, someone else should […]
Doing a full-text search in the HUGE database produced many similar examples, so this that if construction seems quite acceptable. But I promised to consult the readers of this blog, so my question is whether this sentence is considered problematical at all, and if so, in what variety of English.
If Dutch and English do indeed appear to take different stands on the acceptability of the construction, I would be interested in finding out why this would be the case. And also, whether the construction is considered fine (or not) in other languages than English or Dutch.
Like my colleagues, I’d be very interested in hearing what you think!
The construction looks fine to me! The only qualification being that I would always include (and expect) a comma between the two words. My conclusion would be that, if you add a comma, there’s no problem at all.
I could not find the construction in the most prestigious website on Dutch prescriptive matters, https://taaladvies.net/ (correct me if I am wrong)
And of course, the construction is found even in the best writers, e.g. F. Bordewijk in his famous novel Karakter (1938) “Hij zweeg even, hij dacht dat als deze jongen aanleg had er ook een uitnemend eerste procureursklerk uit hem groeien kon.”
This U.S. English speaker doesn’t hear anything odd about the first example (“He thought that if he pushed…”), and I can’t even think of a way of rephrasing that would sound more appropriate. What do Dutch usage guides suggest as a correction? “De Saussure was so determined…”), on the other hand, looks very clumsy. Reading it, I found myself getting halfway through the sentence, and starting over again to follow the meaning. Tony Parr’s suggestion of a comma is an improvement, but that fix is only necessary because the original is so awkward!
Thanks for the replies, Tony, Ton and Paul! With respect to the suggested correction: usage guides basically all suggest the same, namely to reposition the conditional clause to the end of the complex sentence, so “He thought that if he pushed, the trap would open” would become “He thought that the trap would open(,) if he pushed.” While that works in some, perhaps most cases, it doesn’t work in all cases, which seems to be related to the type of connection between the conditional and main clause (causal, inferential, pragmatic) and, as with the pushing and opening in the example, it may create tension with iconicity. We’re still working on that, so I can’t say anything definitive yet.
As with many aspects of word order, it would seem to be largely a question of emphasis…
The only relevant usage issue I can think of is that some usage commentators warn against repeating the “that” after the intervening subordinate clause. That is, they say that you should not write “He thought that if he pushed, that the trap would open.” The repeated “that” is common in speech and unedited writing (and possibly some edited writing), especially if the subordinate clause in between is long.
But I’ve never heard that there’s anything wrong with the “that if” construction on its own.
Thanks, Jonathan. You’re absolutely right. In one of the usage guides it is mentioned that a sentence ‘may completely derail if the conjunction “that” is repeated’. I think that assessment is a bit strong, though, and my corpus data includes a number of examples that I would certainly not judge problematic.