Talk:Idiom (language structure)

Insufficient context; possible original research
I cannot understand what this article is about, and I'm a college professor with a PhD in linguistics. Now it is entirely possible that I am a fool or just having a bad day, but it seems at least as likely that the article doe not provide sufficient context.

The intended discussion seems to relate to an issue of grammaticality versus idiomaticity – following the rules of syntax and morphology versus using a typical or "unmarked" style. But I can't figure out the precise difference that the author(s) of this article have in mind or the relationship of either to solecism, another related word used in the lead section.

I have just looked through four intro texts that happen to be on my shelf (Fromkin and Rodman 1993; O'Grady et al. 2001; Winkler 2007; Merrison et al. 2014) and none seem to discuss "idiom" in the sense it is used here – though they mention the "figurative phrase" meaning. Furthermore, the article currently cites no technical literature, only a dictionary definition of "idiom". I suspect that this article may be either original thought by one or more Wikipedians, or else discussion of a specialized corner of linguistics that most readers will require much more introduction to understand.

(Editors interested in this issue may also be interested in my comment at Talk:Accidental gap.) Cnilep (talk) 03:11, 16 June 2015 (UTC)


 * Hello. I am the editor who created the page and built most of the content. I am not a linguistics expert, and in the end linguistics experts are going to be the final arbiters on this article, so I do recognize the need to defer. But I want to understand better what would be deferred to. Two themes are that (1) I was genuinely surprised that a linguistics expert has never heard this sense of the word, and (2) I hope other Wikipedians interested in second-language acquisition (SLA) will care enough to weigh in on this before any big deletions occur. Below I explain more about both of those points. Off the bat, I just want to assure that nothing here is "just made up"—there's no original thought involved. It's all based on what was taught in the foreign language classes of my own university education. In those classes we talked about the very topic that the cited Merriam-Webster definition lays out. (Corollary: In defense: "I didn't make anything up here—for example, I'm not the one who inserted that definition into a major dictionary!") This is why I'm so surprised that this topic could seem mysterious to any linguistics experts (and I'd like to hear second opinions on that seeming-novelty aspect). Even if there are things in this article that they disagree with, it shocked me that there's anything they would never have heard of, and couldn't understand the meaning or intent of, when they read it. To explain more: I am aware that there's often a discrepancy between traditional models of epistemology of language in academic fields of language and literature ("the English department", "the French department", and so on) and those of linguistic science ("the linguistics department"). For example, the linguist's annoyance with the excesses of linguistic prescription that ignore and contradict the fundamentals of linguistic description. But the thing about that is, the linguists have heard of those things, even if they are annoyed by them. For example, this was always on clear display at Language Log back when I was reading it (some years ago). The pedagogy of foreign language acquisition used in the foreign languages classes that I attended included this sense of the word idiomatic. For example, why is an indefinite article used in the English "I am a plumber" but not in the equivalent French, Spanish, or German statements, eg, "soy plomero"? Because it's idiomatic.  Now some explanation (forgive any belaboring?) of why the languages of a language family would share idiomatic tendencies. It's exactly because the languages are cognate. For example, why are there so many idiomatic similarities among the Romance languages that don't carry across to English (such as the indefinite article example above, or the idiomatic choice of preposition after any particular verb)? It's because they're descendants from a common ancestor (namely, vulgar Latin) that showed that same tendency. Please tell me, anyone reading this who has studied multiple Romance languages, not only (1) that you've heard of this concept but (2) that you can understand what I'm saying when I mention it, even if you'd never heard it before (even though I can't imagine how)? It just seems incomprehensible to me that that couldn't be so. And anyone who has studied 3 or more modern languages, and had any natural curiosity to compare and contrast across them while doing so, will have noticed instances of accidental gaps that occur in one language but not another. (For example, the defective conjugation of English modal verbs compared with the fairly complete conjugations of Spanish and French modal verbs.) And why? How could that happen? Because accidental gaps are idiomatic, and idiomaticness is, by its nature, specific to languages and, in some cases, groups of related languages, that is, language families, which are adjacent twigs on one same branch of a tree, such as the Indo-European tree. Traits of the branch are visible in the twigs, although the twigs are not identical. Please tell me that other people reading this, some of whom may have linguistics degrees, have a clue what I'm talking about here!  What I'm looking for in the development of this article is that people with linguistics degrees who understand what I am talking about will volunteer to "mend it, not end it"—if you see anything about this content where you think, "Well, yeah, nonlinguists would say that, but linguists know better and would correct it to this-and-that,"  then please, cover both here—provide exposition of both, and of how one relates to the other—because the primary audience of Wikipedia is nonlingists. Explain to them what is known and how it relates.  Best regards, Quercus solaris (talk) 23:24, 16 June 2015 (UTC)


 * Thanks for replying. To clarify: I don't think you're making this up, and I'm certain that you are well-intentioned. But I also think the problems with this article are beyond "linguists know better and would correct it to this-and-that". That is to say, I don't think it is just a question of technical language.


 * I do "have a clue what [you are] talking about here", but clues don't necessarily lead to the correct conclusion. If I understand correctly, you are thinking of idiom as structures of language as used, and thus as something not identical to "grammar" in the sense of syntax and morphology. Linguists do talk about such things, for example in work on style or pragmatics or in discussions of communicative competence as distinct from grammatical competence. But what is described here doesn't quite match work on those issues. Perhaps that is because you are understanding the issues through the lens of applied linguistics and language education? But if this is a new way of thinking about language structure, not rooted in previously published work, then it may be "original research" as defined by Wikipedia.


 * If idiom is not the same as grammar, and it's not style or communicative competence, then what is it? (If it is one of those things, then there is no need for a separate article.) Can we (by which I mean "you") find, paraphrase, and cite published work describing it, as is Wikipedia's wont? Cnilep (talk) 02:43, 19 June 2015 (UTC)

Regarding "find, paraphrase, and cite published work describing it": I started below. More to come if time permits. Regarding "If idiom is not the same as grammar, and it's not style or communicative competence, then what is it?": The first answer I formulated, based on my classes, was that it is an aspect (trait) of grammar, namely, the arbitrariness and peculiarity inherent in grammar. (The only reason we say "the cat is gray" instead of "the cat has grayness" in English is because of idiomaticness.) However, the sources below make me modify that (which see). Something like a strong subset of collocation that often has grammatical force.

I think you are right that "through the lens of applied linguistics and language education" has a lot to do with why the usage varies. But even linguistic scientists use the term (some evidence below), so we would need to cover all of those portions of the spectrum in this article and try to show how they relate.

The following are quotes from the Cambridge Encyclopedia of the English Language (1st ed, 1995) by David Crystal, showing how Crystal uses the word idiom in this noncount sense (WP article title idiom (language structure)), as opposed to the count sense of, more or less, a set phrase (WP article title idiom). Biblical idiom: There are many phrases in the King James Bible which have entered the general idiom of the language (sometimes with minor changes in grammar and emphasis). (p. 64) The Prayer Book is responsible for a great deal of the vernacular idiom of English prayer. (p. 65) I am realizing as I look through every occurrence of the word idiom listed in the index that Crystal is treating the noncount sense of idiom as meaning an especially strong pattern of collocation. The sense's close relation to the count sense is clear (in other words, idiom is the trait or nature of idioms), but I will have to explore the usage in other books to sharpen the focus.

In American English Grammar, Charles Carpenter Fries says, Word order is often an important item of the idiom of a language, but it is not always a grammatical device as it is in English. In Latin, for example, the periodic structure with the verb at the end occurs very frequently, but the word order in such a sentence as "Nero hominem interfecit" has nothing whatever to do with indicating the so-called "subject" and "object." (p. 37)

So what Fries is adding to the definition is that idiom is not always grammatical. It is always collocational but not necessarily grammatical. Why did Latin speakers in Rome during the era of classical Latin favor certain word orders even though many word orders expressed the same idea in their language (because of extensive inflection)? Simply because "that's how it was often done, although it could have been done otherwise."

Thanks for discussing. Will see where this goes from here. Hope other users can build on this as well. Quercus solaris (talk) 23:04, 19 June 2015 (UTC)

Issues remain
User:Lise-lyse removed three maintenance tags with [//en.wikipedia.org/w/index.php?title=Idiom_(language_structure)&diff=next&oldid=731925272 this edit]. While I appreciate the addition of a source, to my mind that edit is not nearly sufficient to address the context and possible original research issues. It is a step toward improved referencing, but I don't think it is sufficient on that grounds, either. Bryan Garner uses the word idiomatic in his discussion of the word accompanied, but he does not appear to discuss idiom or idiomatic. What I think the article needs is sources that explicitly discuss idiom as distinct from grammar or other notions of language structure. Cnilep (talk) 04:37, 4 January 2018 (UTC)

Unintelligible English
Multiple sentences are highly grammatically malformed or else unintelligible. I'm fixing some of them while revising others that are outright incorrect. XanderXylona (talk) 17:04, 13 June 2019 (UTC)