Native speakers giving misleading information

[Chengjiang]

Has anyone else here had a conversation with someone where that person gave demonstrably false or very misleading information about their own native language? If so, why do you think they did?

A few years ago I was talking about languages with a fellow student (not in a language class). Her native language was Japanese. (She was also 100% fluent in English, which I'm ashamed to say I'm still not for any language other than English.) I forget exactly how we got on the subject, but we were talking about the particle は. Now, I didn't know a huge amount about Japanese, but I knew, as many of you probably know, that は marks a noun phrase as the topic, which is often but not always the subject. I referred to は thus (perhaps in comparison to what some other language does; I forget), avoiding using terribly technical diction since I didn't think she had any formal linguistic background, and she immediately replied that no, は wasn't anything like that at all. She went on to insist that は corresponded exactly in meaning and usage to the English verb to be. I was puzzled but went on to discuss something else language-related.

This has bothered me whenever I've thought back to it, because I've never been able to figure out what that young woman meant. I seem to recall that under certain circumstances the copula can be omitted in Japanese, and so a copulative phrase with the form X は Y might look a bit like "X is Y" in English, but that's all I've got. There didn't seem to be any guile going on, nor any confusion about what I said, just an apparently completely sincere mischaracterization of a bit of grammar in the speaker's mother tongue.

[Viktor77]

All the time, but it's simply out of ignorance that's all.

Or sometimes it's sociolinguistic like the time my Greek friend argued Latin came from Greek because duh Greece was the start of everything that ever was.

[clawgrip]

This is a common misconception. It's actually not at all unusual to hear Japanese people use forms of to be in place of the topic marker. Basically an irrelevant copula that appears between the subject and the verb. I notice this is far more common with pronouns as subjects, not regular nouns. This is, I think, due to disinterested Japanese students learning things by rote and not thinking about the deeper meaning. So for example, when they learn things like I am, you are, he is, etc., vs. me, you, him, and so on, these will be compared with watashi wa (TOP), watashi o (OBJ) and so forth. So people will associate "I am" with "watashi wa" and figure be is like wa. It's also due to a coincidence of word order. When you compare English SVO with Japanese's SOV you get be and wa superficially appearing in the same spot after the subject/topic, reinforcing the misconception.

[Chengjiang]

All the time, but it's simply out of ignorance that's all.

Or sometimes it's sociolinguistic like the time my Greek friend argued Latin came from Greek because duh Greece was the start of everything that ever was.

Ah, yes, that sort of thing. Reminds me of the times I've seen speakers of various Indo-Aryan languages insist that Sanskrit is the most specialest and perfectest language of all, and in some cases that the "Indo-European" languages are all actually direct descendants of Sanskrit. Even languages like Icelandic are supposed to descend from Sanskrit somehow.

That said, I'm mainly interested in cases where speakers give an incorrect description of some feature of the language that they necessarily have to know in order to speak it fluently, albeit not necessarily know on a conscious level or one they can articulate. The Japanese speaker in my story clearly knew how to use は in speech, but when it came to describing it she referred to it as a verb, specifically the copula.

[Chengjiang]

This is a common misconception. It's actually not at all unusual to hear Japanese people use forms of to be in place of the topic marker. Basically an irrelevant copula that appears between the subject and the verb. I notice this is far more common with pronouns as subjects, not regular nouns. This is, I think, due to disinterested Japanese students learning things by rote and not thinking about the deeper meaning. So for example, when they learn things like I am, you are, he is, etc., vs. me, you, him, and so on, these will be compared with watashi wa (TOP), watashi o (OBJ) and so forth. So people will associate "I am" with "watashi wa" and figure be is like wa. It's also due to a coincidence of word order. When you compare English SVO with Japanese's SOV you get be and wa superficially appearing in the same spot after the subject/topic, reinforcing the misconception.

I wonder if this sort of thing is reinforced by the frequent use of progressive verb form with to be as an auxiliary in English? I mean, a progressive form with to be being the normal way to form the present tense for non-stative verbs could easily contribute to the sense that to be is a content-free grammatical marker.

[zompist]

This sort of thing is pretty common. People's ideas about the rules that they themselves follow can be incorrect. One example from a linguist (I forget who): some informant claimed that he never used "any more" in a positive sense ("Every time we leave the house anymore, I play [this] game"), and caught him using it in recordings. Or there was the Spanish professor who maintained that /c/ and /z/ were pronounced differently. William Labov noted that people claim to use much more standard phonology than they really use.

Geoffrey Sampson likes to tell the story of when he was arguing double center embedding was not possible in English, and someone asked "But don't you find that sentences that people you know produce are easier to understand?" and he gamely tried to answer before realizing that she had provided a counter-example to his assertion. Which is why he now advocates corpus linguistics rather than relying on intuition.

[vokzhen]

See also: the number of misinformed English teachers and self-proclaimed grammar nazis who swear up and down that passive voice and preposition stranding are bad, the latter group sometimes claiming they don't use them. My absolute favorite example, which unfortunately I didn't take down word-for-word, was a post about singular they, with the conclusion roughly being "if someone uses it, they're just wrong."

[Sevly]

I like to analogize this to how you can intuitively know how to throw a baseball and always hit your target but still be misinformed about the actual physics of projectiles. Language Log has written a lot about this, as well as the idea that people are more aware of the fact that their day-to-day experience with physics doesn't translate well to explaining the details of physics than they are of the fact that their day-to-day experience with language doesn't translate well to explaining the details of language. Whether or not that's true is contestable, but there is certainly no shortage of absurd linguistic claims from journalists who see no need to consult with actual linguists, so.

The other element of this is that thinking about how you talk causes a feedback loop that changes the way you talk, so it's just as hard to ask yourself a question like "do I use positive anymore, or linking r, or whatever" as it is to ask someone else - if you convince yourself that you don't, then it is indeed true that it the next few samples you take, you won't. So you really do have to rely on corpora and recordings where people are not aware that they're speech is being analyzed for meaningful conclusions.

[Chengjiang]

Or there was the Spanish professor who maintained that /c/ and /z/ were pronounced differently.

I've occasionally observed English speakers to say or write things that imply they believe that K and hard C have different sounds. I've also heard from various people that some native hispanophone teachers of Spanish will say that B and V have different sounds if asked. (The only modern speech I've ever seen treated as a Spanish dialect that distinguishes them is Judeo-Spanish/Ladino, which is never the Spanish under discussion in these accounts.)

[Sumelic]

There are several people in the comments and answers to this post that claim to distinguish the pronunciations of voiceless s, soft c, and soft sc in English: http://english.stackexchange.com/questi ... e-c-silent

Who knows though? Maybe velar softening really is a synchronic rule of English (I believe Chomsky and Halle postulated this in The Sound Pattern of English), there's an underlying /k/ in words like "cent" that people pick up from the spelling, and some people have some form of incomplete neutralization that actually leads to marginally different pronunciations. When I first heard about incomplete neutralization of voiced/voiceless coda consonants in German and Russian, I assumed it was just a psychological illusion, but it seems it's actually phonetically real, if not as robust or noticeable as normal phonemic contrasts.

(Actually, I would be really surprised if someone detected an s/c contrast in English, but it seems somewhat more plausible that speakers might be inclined to pronounce words starting with sc with a slightly longer /s/ sound.)

[Chengjiang]

The other element of this is that thinking about how you talk causes a feedback loop that changes the way you talk, so it's just as hard to ask yourself a question like "do I use positive anymore, or linking r, or whatever" as it is to ask someone else - if you convince yourself that you don't, then it is indeed true that it the next few samples you take, you won't. So you really do have to rely on corpora and recordings where people are not aware that they're speech is being analyzed for meaningful conclusions.

I actually have this problem with part of my accent: I've never been able to determine how many of diaphonemic English /ɑː ɒ ɔː/ I distinguish. Whenever I think about it I'm not sure how I "normally" pronounce any of these vowels. For some reason, most often when I say words out loud thinking about this I seem to merge /ɒ ɔː/ to something like [ɒ] (but to [ɔ] before /r/) and retain [ɑ] for /ɑː/, but that doesn't sound like it should be right, since I'm unaware of any English dialect, especially any subdialect of General American, that merges /ɒ ɔː/ without also merging /ɒ/ with /ɑː/.

Clarification: I mean all of /ɒ/ merges with /ɔː/. At least in the somewhat suspicious accent I seem to produce when I'm aware of it, father and bother don't rhyme but cot and caught do.

[Salmoneus]

Or there was the Spanish professor who maintained that /c/ and /z/ were pronounced differently.

I've occasionally observed English speakers to say or write things that imply they believe that K and hard C have different sounds.

My mother thinks that. I'm not entirely sure it's not true for her, though - she describes a clear difference in production (that K is further back in the mouth and C is further front), and either a difference it production or the perception of it might in her case be related to being taught Irish when young - maybe she's hypercorrected to carry over the broad/slender contrast for the pair. I can't tell from the sound, but then I wouldn't be able to.

I do think there can be, as it were, micro-differentiation: differences in speech that reflect spelling (or any other feature of characterisation) but that are not phonemic, and may be hard to consciously distinguish for hearers. This could come from the speaker's consciousness of spelling, morphology, etc, or from sub-phonemic tendencies in the pronunciation of others that have been subconsciously observed and adopted. I think I have some of these micro-differentiations, though I can't think of any atm. Except that I think that many of my 'schwas' still carry an echo of their original vowel - that is, I think my reduced /U/ is different from my reduced /I/, even though the difference is small and unreliable and not entirely accurate.

[cromulant]

My dad's a real firm believer that "me" in subject position is incorrect, to the point that he will use the 'wrong' construction and immediately correct himself. "Me and your mom...I and your mom." He is also staunchly opposed to "it's me." I've tried to get him to question what exactly it means to say a construction is "wrong" if it is in fact is the instinctive preference of most people including himself? This leads him to ask me if I think there are no rules to grammar whatsoever and people can just make up whatever grammatical rules they want.

This conversation always gets cut short for some reason. I want to really delve into it with him some time.

[Chengjiang]

My dad's a real firm believer that "me" in subject position is incorrect, to the point that he will use the 'wrong' construction and immediately correct himself. "Me and your mom...I and your mom." He is also staunchly opposed to "it's me." I've tried to get him to question what exactly it means to say a construction is "wrong" if it is in fact is the instinctive preference of most people including himself? This leads him to ask me if I think there are no rules to grammar whatsoever and people can just make up whatever grammatical rules they want.

This conversation always gets cut short for some reason. I want to really delve into it with him some time.

That does sound like it would get into an interesting discussion of what he believes a grammatical rule is, since I think although many people refer to them a lot they couldn't tell you what it actually meant for a construction that speakers regularly use to be "correct" or "incorrect".

[vokzhen]

"Me and your mom...I and your mom."

That's funny because as far as I can tell, the second is flat-out ungrammatical for me. Perfectly understandable, possibly pops up in speech when the second person is an afterthought, but is still a poorly-formed sentence. "I" has to either be the sole subject, or come after "and."

[Richard W]

"I" has to either be the sole subject, or come after "and."

That's partly a politeness rule - "Don't put yourself first." It may also be a contact rule - 'I' has to come petty close to the verb, though it can be separated from it by 'who'.

[Travis B.]

"I" has to either be the sole subject, or come after "and."

That's partly a politeness rule - "Don't put yourself first." It may also be a contact rule - 'I' has to come petty close to the verb, though it can be separated from it by 'who'.

To me me comes before and but I comes after it; it sounds really odd to put I before and.

[Vijay]

I think I've had to deal with native speakers giving me misleading information about Indian languages for almost my whole life. Even for Hindi/Urdu, it can be hard to find reliable information, though that's partly precisely because of the Hindi vs. Urdu issue and probably even more because of dialect variation (and people - even those who are supposed to be professors of the language in addition to being native speakers themselves - seeming not to really be familiar with it).

[Chengjiang]

"I" has to either be the sole subject, or come after "and."

That's partly a politeness rule - "Don't put yourself first."

That's generally how I've heard it described. Going first is a position of greater respect, so you put the other party before yourself. Honestly, when you come right out and say it it sounds like something from zompist's native Cadhinor grammar.

It may also be a contact rule - 'I' has to come petty close to the verb, though it can be separated from it by 'who'.

I've never heard this one, but it's an interesting way to analyze it.

To me me comes before and but I comes after it; it sounds really odd to put I before and.

Same here. I've heard X and I, me and X, and X and me (as both subjects and objects/obliques in all cases), but the only time I've ever heard I and X is the line "I and Velma ain't dumb" in West Side Story, which was supposed to sound uneducated, but to me just sounds strange. I think many English speakers have some notion of the forms with me being "casual" or "colloquial" and the forms with I being "proper", which also tends to prompt putting the first person pronoun at the end.

As for myself, I'm pretty sure I don't normally use the form with X and I as an object/oblique, but I think I've used all three as subjects, although I think I default to X and I these days. It's hard for me to remember for sure, though.

Even for Hindi/Urdu, it can be hard to find reliable information, though that's partly precisely because of the Hindi vs. Urdu issue

In your experience, do speakers of these tend to insist that they are fully separate languages and not varieties of a common language? I've know that that's the case for some other mutually intelligible sets of registers or dialects.

[Vijay]

In your experience, do speakers of these tend to insist that they are fully separate languages and not varieties of a common language?

I don't know whether I'd go that far, but IME they definitely tend to think of them as different languages even though it's never even clear what they mean by either "Hindi" or "Urdu" (because the answer is inevitably going to be different depending on who you ask), apart from the fact that one of those two is what they would identify as their own native language.

[CatDoom]

Ah, yes, that sort of thing. Reminds me of the times I've seen speakers of various Indo-Aryan languages insist that Sanskrit is the most specialest and perfectest language of all, and in some cases that the "Indo-European" languages are all actually direct descendants of Sanskrit. Even languages like Icelandic are supposed to descend from Sanskrit somehow.

That said, I'm mainly interested in cases where speakers give an incorrect description of some feature of the language that they necessarily have to know in order to speak it fluently, albeit not necessarily know on a conscious level or one they can articulate. The Japanese speaker in my story clearly knew how to use は in speech, but when it came to describing it she referred to it as a verb, specifically the copula.

I've encountered a similar phenomenon while looking up information about the history of the Dravidian languages. It seems like any time someone writes something about Proto-Dravidian, people will come out of the woodwork insisting that using such a term is misleading, since everybody knows that Proto-Dravidian is just Tamil, which is the most ancient and perfect language of mankind. The degree of pseudo-linguistic craziness attached to the claim can vary, but it always bugs me out how often I see the "Tamil is the root of all Dravidian" thing.

[Matrix]

Ah, yes, that sort of thing. Reminds me of the times I've seen speakers of various Indo-Aryan languages insist that Sanskrit is the most specialest and perfectest language of all, and in some cases that the "Indo-European" languages are all actually direct descendants of Sanskrit. Even languages like Icelandic are supposed to descend from Sanskrit somehow.

That said, I'm mainly interested in cases where speakers give an incorrect description of some feature of the language that they necessarily have to know in order to speak it fluently, albeit not necessarily know on a conscious level or one they can articulate. The Japanese speaker in my story clearly knew how to use は in speech, but when it came to describing it she referred to it as a verb, specifically the copula.

I've encountered a similar phenomenon while looking up information about the history of the Dravidian languages. It seems like any time someone writes something about Proto-Dravidian, people will come out of the woodwork insisting that using such a term is misleading, since everybody knows that Proto-Dravidian is just Tamil, which is the most ancient and perfect language of mankind. The degree of pseudo-linguistic craziness attached to the claim can vary, but it always bugs me out how often I see the "Tamil is the root of all Dravidian" thing.

It gets even crazier when both of these are paired, linked by some "Sanskrit is Dravidian" nonsense.

[Chengjiang]

I don't know whether I'd go that far, but IME they definitely tend to think of them as different languages even though it's never even clear what they mean by either "Hindi" or "Urdu" (because the answer is inevitably going to be different depending on who you ask), apart from the fact that one of those two is what they would identify as their own native language.

Well, yes. IIRC there's more difference between certain different varieties of Hindi or different varieties of Urdu than there is between the two standard registers.

[Vijay]

IIRC there's more difference between certain different varieties of Hindi or different varieties of Urdu than there is between the two standard registers.

Yep. That sounds about right.

I mean, OK, there's a huge difference in vocabulary if you look at some news or radio broadcasts or something like that, but definitely not at the level of, like, how most people communicate day to day.

[M Mira]

This sort of thing is pretty common. People's ideas about the rules that they themselves follow can be incorrect. One example from a linguist (I forget who): some informant claimed that he never used "any more" in a positive sense ("Every time we leave the house anymore, I play [this] game"), and caught him using it in recordings. Or there was the Spanish professor who maintained that /c/ and /z/ were pronounced differently. William Labov noted that people claim to use much more standard phonology than they really use.

Geoffrey Sampson likes to tell the story of when he was arguing double center embedding was not possible in English, and someone asked "But don't you find that sentences that people you know produce are easier to understand?" and he gamely tried to answer before realizing that she had provided a counter-example to his assertion. Which is why he now advocates corpus linguistics rather than relying on intuition.

As a non-native English speaker, I don't even know those are possible and I'm not quite sure what they mean either.

In the first sentence, can the "anymore" be removed without changing its meaning?

For the second, does it still mean the same thing if I replace "people you know produce" with "produced by people you know"?