Examples of illogic in English syntax

As far as logic goes -- I'm going to make up or mis-use symbols here cause I haven't taken a logic course in well over a decade, but it's a pretty basic concept I'm going for:

{"this is going to get it back" + "I think"}

not{"this is going to get it back" + "I think"}
=
{"this is not going to get it back" + "I don't think"}

...sounds wrong. Logically, it should be was instead of were. Or is was right? I've never been quite sure about this...

I'm designing a course on "Effective Reading and Writing" at the moment...I believe that your brother is placing more emphasis on grammar than on sheer logic. This is not really a syntactical issue at all--in my mind anyway.

"But really, you know, they've lost circulation to Time Out, and this isn't going to get it back, I don't think."

In order to use "I don't think" in this way, the statement previous must be an untrue statement that's untruthfulness is then made evident by the statement "I don't think." Because the statement previous to "I don't think" in this sentence is true--unironic, not sarcastic, the "I don't think" makes no logical sense. I'm drawing attention to the double negative through logic as opposed to grammar in order to demonstrate that the idiomatic use of "I don't think" at the end of a sentence requires an antecedent ironic phrase. Perhaps the use of a question mark after the ironic statement might denote the irony more clearly. Your examples demonstrate this point quite clearly.

Suffice it to say, your brother is making a comment based on the vagueries and illogic of English syntax? I don't think.

Let's see where we are. I'm claiming that in modern English idiom "...and this isn't going to get it back, I don't think" is more or less equivalent to "I don't think this is going to get it back," despite the former's being a double negative. I might consciously choose the first over the second depending on what I wanted in the way of rhythm, flavor, tone of voice (when I posted it on the other thread I wasn't even thinking of a choice, I was just writing)(maybe "I write" is an antonym for "I think"). By the way, my brother has yet to disagree with my contention that what I wrote was idiomatic. And no one has said anything here to make me doubt that I was being idiomatic. I'm claiming that the "I don't think" at the end of the sentence reinforces the previous negative, rather than negating it. That's what my native speaker's experience tells me.

And so far there's been no comment on my next contention, which is that "...and this isn't going to get it back, I think" is not idiomatic, despite its being logical. Do you ever hear anyone saying, "He's not happy, I think," or "We don't want to go into that room, I think"?

Oh, hm. I use that construction myself a lot -- but usually in response to questions. "Why hasn't Bob responded to his mail?" "He's not at work this week, I think."

Not quite the same, because it's a response -- you can take it apart and put it back together as "I think the answer to your question is 'he's not at work this week,'" rather than just "I think he's not at work this week." I'm not positive that matters, but it is a difference.

I do think you were being idiomatic. It's a construction I use. It's pretty clearly understandable -- surely no native speaker would misunderstand and think that you mean "this is going to get it back." If it sounds odd to some people, maybe it's a little to the left of the received dialect of formal writing, like the subjunctive example Prude asked about.

And yes, I've also used the "I don't think" double negative before. So if it takes two to make an idiom, by God, Frank, we got ourselves an idiom going here.

you can take it apart and put it back together as "I think the answer to your question is 'he's not at work this week,'" rather than just "I think he's not at work this week."

vs.

"'We don't want to go into that room' is what I think."

I have just reached the point where the word "think" looks strange to me.

Remember the NyQuil!

In some parts of the UK this means you think it will happen and in others it means you think it won't.

The English word "logic" derives from the greek word for "word".

Word.

ii. seems a bit mannered to me
iii. is the final phrase a sarcastic version of the same phrase in i.?
iv. To put "I think" after* a declaration is to put a kind of wobbly negative after it. So to put "I don't think" negates the negative!! So Richard K. wz wrong: yr phrase is a TRIPLE negative (or two and a half maybe).

*(This doesn't apply if you put it before...)

(both mean 'i don't care'. i only use the second type, i think the first is american)

Speaker decides to change the strength of the assertive while uttering it:
(1) -> (2) not worth the effort because one obviously believes what one asserts
(1) -> (3) intended to soften the strength of the embedded clause, but the sentence is uttered in real time, so the speaker can't go back and remove the negative from the embedded clause

I think "not X, I think" is acceptable in the blank slate situation, i.e., when no opinion on the truth or falsity of X has been stated, implied, or presupposed. If the speaker is contradicting the hearer, then not X should be sentence final. If the speaker is agreeing with the hearer, then the tag "I think" is inappropriate.

Another issue that often enough confronts me these days, and that emerges in some of the more multiply-claused sentences I write on this board, is the relative difficulty of writing relatively simple things without going somehow into the linguistic wrong.

I cannot think of an example right now.

Somehow that last sentence is a model of relative clarity.

'multiply-claused' and 'linguistic wrong' might be wrong, but they don't matter.

Actually, this is perfectly logical. "He'll be good to her, I think," tells the listener something; the appended "I think" adds a little uncertainty but doesn't contradict the information that the listener has just taken in. "He'll be good to her, I don't think," does contradict what the listener has just heard, and therefore is never likely to be used for anything but comic or quasi-sarcastic effect. I don't know other languages, but I assume that, for the reasons just given, even highly inflected ones won't allow the negation to come at the end of a positive assertion.

The use of "X, I don't think" is a sarcastic joke, and rarely a good idea in writing, as it demands a particular inflection.

1. He'll be good to her, will he?
Hearer believes he will be good to her but has been, or is about to be, proved wrong.
Speaker intends sarcasm.

2. He'll be good to her, won't he?
Speaker believes he will be good to her and is looking for confirmation from hearer. Intonation has a big effect.
Hearer's belief unknown.

3. He won't be good to her, won't he?
This seems like the parallel case to Frank's example, and interestingly, it also seems the most questionable of the four. If it is acceptable:
Speaker believes he won't be good to her and is looking for confirmation from hearer.
Hearer's belief unknown.
Why isn't this one sarcastic like (1)?

4. He won't be good to her, will he?
Speaker believes he won't be good to her and is looking for confirmation from hearer. Interestingly, the shift in meaning with intonation doesn't work, but this may be sentence specific as "He doesn't like ice cream, does he?" with rising intonation for the main clause seems okay.
Hearer's belief unknown.

Do you agree with my interpretation of (3)?

Interesting side issue: how we answer negative yes/no questions:
5. Don't you want some?
NO, I don't.
YES, I do.

6. Don't you think he'll be good to her?
NO, I don't.
YES, I do.

For (6), "YES, I don't" and "NO, I do" also seem acceptable, maybe even better.

But is "this isn't going to get it back, I don't think" more emphatic than "this isn't going to get it back"?

wheras if we flipped it to read "i don't think x won't happen" then it expresses pitch-perfect ambiguity.

If they're both independent clauses, shouldn't they be separated by a semicolon rather than a comma?

In this respect, "youn" has raised an interesting point about the double genitive. "She is a friend of mine" is THE English idiom and "She is a friend of me" is idomatically wrong. Yet logic says that the second is correct and that the first is in some way redundant, since it twice indicates possession. I am using this example to define "idiom;" my point is that idiom doesn't mean that two alternative constructions intended to convey one meaning are each considered acceptable, even though one is . Rather, idiom means that ONE of the constructions is proper English and the other is not, and the "ungrammatical" one wins.

Example #2: "He writes like Proust" is idomatically correct while "He writes as Proust" is idomatically wrong (unless we are discussing noms de plume). Yet the normal rules of implied words, conjunctions, and prespositions say that "as" should be correct and "like" incorrect. Sorry -- idiom wins; "like" is required.

OK, Frank and I agree (I suspect) that it is not a great idea for "She isn't home, I think" and "She isn't home, I don't think," to mean the same thing, and that idiom should tell us that one is right and the other wrong. The problem is that my instinct favors the first construction and Frank's, the second. Since he is better read than I and probably a more prolific writer as well, I often turn to him when I find myself in doubt.

I tried turning to authorities for help, but most said nothing at all useful. Follett includes more than three pages of inferior or incorrect uses of negatives (none exactly on point), but in no case does his correction place a negative at the end of a sentence.

Frank's right that the various responses he has gotten so far do not clearly answer this particular question, though many are interesting and some are amusing.

"She is a friend of mine" is THE English idiom and "She is a friend of me" is idomatically wrong. Yet logic says that the second is correct and that the first is in some way redundant, since it twice indicates possession.

The logic you happen to be applying to that sentence says that the first is redundant. The logic which actually obtains, held over from a time when English had a more developed case system, requires different forms of words for different functions, even when -- by virtue of said forms in a sentence agreeing with each other -- this results in redundancy. "Those birds over there" would be just as redundant by your logic: the pluralization of "birds" isn't necessary because "those" already denotes the plural; the location signifiers of "those" and "over there" aren't needed, since both make it clear that we don't mean "these"/"over here."

But that isn't how language works. It isn't how any language works except artificially constructed ones.

I'm not at all clear on whether you're using "idiomatically" to mean anything more than "the way things are done in this language" (and if so, why it's necessary to use it). It's usually used to refer to phrases which are so peculiar to a language (or language group) that a literal translation wouldn't convey meaning, or would fail to adhere to usual forms.

Ultimately, there is no clear answer to questions like this, because languages tend to discourage universal consensus. You can't sound right to all of the people all of the time.

I don't think that grammatical forms of agreement, e.g., person number agreement for subjects and verbs, are illogical. Redundancy, as in the example given by Tep, aids in processing. Multiple negation, or negative concord, exists in languages such as French and can be considered a form of agreement:
Personne ne me dit rien.
Nobody says nothing to me.
Grammarians probably just jumped on multiple negation as illogical because in logic (not (not A)) = A. But you're not really negating the proposition you're applying each additional 'not' to.

The re-emphasis analysis has the following argument in its favor: when the negative is more emphatic, "I don't think" sounds worse.

1. This isn't going to get it back, I don't think.
2. This is not going to get it back, I don't think.
3. This is never going to get it back, I don't think.
(cf. I don't think this is ever going to get it back, I think this is never going to get it back.)

An explanation could be that re-emphasis becomes increasingly unnecessary from (1) to (3). On the other hand, (3) may be bad simply because of the difference in force between "never" and "I don't think." Then can (1) and (2) still be distinguished in terms of acceptability? ('Not' may be more emphatic in terms of its force in discourse, but here I'm assuming that 'not' and the contracted form of the verb carry the same force but that 'not' is easier to recognize in processing. In fact, throughout, I'm talking about emphasis in terms of processing, not force in discourse.)

Why is "I don't think" only appropriate with negated sentences?

4. This is unlikely to get it back, I don't think.
(cf. I think this is unlikely to get it back, I don't think this is likely to get it back.)

(4) is unacceptable to me.

I always wondered about "I could care less!" vs "I couldn't care less!" Grammatically, the former indicates a meaning opposite to the meaning that is actually conveyed to a native English speaker. Yet it gets used all the time, and sounds snappier, to me at least. Maybe it suggests that there is no limit to the speaker's disengagement from the subject—"I could care less - and do!"

I misspoke when I said this. I seemed to be presupposing there were multiple instances of negation, logically speaking. It would have been better to say that in negative concord languages there can be a single instance of (logical) negation that is (grammatically) marked in several places in the sentence.

In these examples, the negation isn't applied to the same proposition.

1. I didn't say that I wouldn't help you. not(say(x,not(help(x,y)))), x=speaker, y=hearer
2. I said that I would help you. say(x,help(x,y))

In a given situation, (1) can be true and (2) false. I used a different verb, because it's harder to see with 'think,' but likewise, in a given situation, (3) can be true and (4) false, e.g., if the speaker hadn't even considered whether or not the strategy was going to get the Voice's audience back or if the speaker was uncommitted.

3. I don't think this isn't going to get it back. not(think(x,not(P))), x=speaker, P=This is going to get it back, with appropriate interpretation of pronouns
4. I think this is going to get it back. think(x,P)

However, for negation within the same clause, if (5) is true, (6) must be, since (standard) English isn't a negative concord language.

5. I'm not going to do nothing. not (not (Exists y (do(x,y)))), x=speaker, y=a variable for an action
6. I'm going to do something. Exists y (do(x, y))

I bring this up because I don't think what's interesting about Frank's example has anything to do with multiple negation.

For the argument in favor of re-emphasis, additional support may be provided by the following: "I think" sounds better as the negative marker becomes easier to recognize.

2'. This is not going to get it back, I think.

To me (2') sounds better than (2) from above.

You were also referring to Richard. Frank does not equal Richard, I don't think - though there's nothing inherently illogical in redundancy.

What we mean by the word "idiom" depends on the context in which we're using it. If we're comparing English to French, we'll describe a rule of English grammar as "idiomatic" if there's no equivalent rule in French. But in this thread we're using it (1) for instances where an accepted construction seems to violate a standard rule of English, (2) where there are two or more accepted constructions, (3) where the actual meaning seems to contradict a literal reading (without its being ironic/sarcastic), (4) other stuff.

NB: mark s's "wobbly negative" analysis and Roderick the Visigoth's "re-emphasizing" analysis are not incompatible.

That's because they address different issues. The apostrophe-s in "He's not a friend of John's" may be a reemphasis, but the "I don't think" at the end of "He's not a friend of John's, I don't think" adds doubt, not emphasis. (So it's not redundant, either. It adds new information. Not that there's any illogic in redundancy.) It's the "think," not the "not," that adds the doubt. "I can't get no satisfaction" would be reemphasis.

So I disagree slightly with Marks' analysis. "It won't happen, I don't think" is closer to "It won't happen, at least I'm pretty sure it won't happen" than to "I believe it won't happen" (and it's the "I don't think" that makes the assertion mere belief, anyway), but the doubt isn't as strong in i as in iv - "It will happen, I think" - which doesn't mean "I believe it will happen. Or do I?" but rather, "It will happen - well, I think it will happen."

OK, Frank and I agree (I suspect) that it is not a great idea for "She isn't home, I think" and "She isn't home, I don't think," to mean the same thing, and that idiom should tell us that one is right and the other wrong.

Your suspicion here is incorrect. I see no problem with their meaning the same thing. I would only see a problem if one of them meant two incompatible things, thus creating ambiguity. But I still say that "She isn't home, I think" sounds weird, and if you and other people on this thread really used this construction as much as you say you do, it wouldn't sound weird to me.

I am using this example to define "idiom"; my point is that idiom doesn't mean that two alternative constructions intended to convey one meaning are each considered acceptable, even though one is ["ungrammatical"]. Rather, idiom means that ONE of the constructions is proper English and the other is not, and the "ungrammatical" one wins.

I don't think that this is what "idiom" means to anybody but you, on this thread or elsewhere. "...and this isn't going to get it back, I don't think" is idiomatic, despite logic and grammar. But this doesn't make alternative constructions ("and I don't think this is going to get it back," "and this isn't going to get it back, I think") incorrect. And even though, as I keep saying, "this isn't going to get it back, I think" isn't standard usage, I also don't see any principle that says that it can't be. But this whole paragraph is redundant - though there's nothing inherently illogical in redundancy.

Interestingly, as youn points out, if we go to question form, "He won't be good to her, will he?" (negative followed by the positive) is acceptable, whereas "He won't be good to her, won't he?" (negative followed by negative) isn't. But then (as youn also points out), what's being added isn't uncertainty so much as a request for confirmation from the hearer (though that implies uncertainty). But notice that "He won't be good to her, will he?" seems more definite and less doubtful than "He'll be good to her, won't he?"

I don't understand Cybele's point about irony. There's nothing ironic in "This won't get it back, I don't think." But also, I'm wondering if I'm using the word "logic" on this thread in the same way that Richard and Cybele are. To me, English is logical if it follows its rules fairly consistently, which it does. (I assume that all languages do, even nonstandard and dialect versions. I can't imagine what it would be like for a language not to be fairly consistent in its rules.) The "I don't think" construction may not be illogical in this sense, given that it doesn't quite parallel other sentence forms anyway, but nonetheless it doesn't link into other forms, either, which is why I'm feeling that it's idiomatic rather than logical. But I believe that Richard and Cybele are using "logical" to mean "We use 'not' as we do in formal logic." And I don't see that this is necessary, or that languages and dialects that allow the double negative are any less "logical" than standard English. As long as the conventions are understood, and that people can express logcial thoughts in it, then the language is logical.

I don't know other languages, but I assume that, for the reasons just given, even highly inflected ones won't allow the negation to come at the end of a positive assertion.

But actually, when I first typed it in, I wrote:

I don't know other languages, but I assume that even highly inflected ones won't allow the negation to come at the end of a positive assertion for the reasons just given.

But I changed it in order to prevent ambiguity, since in my original version the reader wouldn't know whether I meant (1) "even the highly inflected ones won't allow the negation to come at the end of the sentence, and I'm saying this for the reasons I've just given" or (2) "even the highly inflected ones won't allow the negation to come at the end of the sentence for the reasons given, but they might allow it for other reasons." Combining "not" with "because" is especially dangerous: If the "not" is being used as simple negation, the "because" can't come in a clause after the "not" (when it does, "because" has to be replaced by some other expression, such as "since" or "given that"). E.g, "He won't sing because he's unhappy," can mean either "he won't sing, and this is because he's unhappy," or "he will sing, but not because he's unhappy." If you want to say something like the latter, and don't want to use "he will sing, but not because he's unhappy," you have to add a reason for his behavior: "He will not sing 'Melancholy Baby' because he's unhappy but because he wants to make you unhappy."

Okay is this proper english?

"look at that elephant over theres trunk"

and if so, does an apostrophe get used?

I awoke this morning with this in my head. it could be bollocks.

you do need an apostrophe for the noun phrase /that elephant over there/ to indicate its possession of the trunk, yes.

Thanks, so that's a valid sentence?

it just sounds so strange to me

funny you put it like that because the ultimate test is "does it sound right to a native speaker?" - but i think it only looks awkward because it's spoken language rather than written language, if you say it out loud it sounds fine to me (or at least it does in my vaguely-Midlands accent)

sounds a bit awkward. "See that elephant over there? Look at its trunk!" I guess would be better?

Check the trunk on Nelly.

"look at that elephant-over-there's trunk"

Behold the elephant yonder's trunk.

big trunk on them there elephant

The possessive in English attaches to noun phrases, not nouns* so "look at that elephant over there's trunk" is normal. When you say "The Queen of England's hand" the hand belongs to the Queen, not England.

Constructions like the one you're wondering about are more common in speech than in writing, but the propriety of the underlying syntax is beyond question. It's correct English.

*'s is not a suffix, it's a phrasal clitic. Or I would describe it that way. The exact mechanism by which it operates (edge affix vs. clitic) is a subject of debate. Here's a paper from someone trying to settle the issue by saying it's both: https://www.linguisticsociety.org/sites/default/files/3524-6826-1-SM.pdf

as usual f. hazel is 100% reliably correct on this

yeah, thanks f.hazel!