19

Okay that's a crazy title, but bear with me. Got into a good natured discussion with someone on another stack exchange site, and I was "correcting" him on the use of infer vs. imply.

(The discussion can be found here: https://skeptics.stackexchange.com/questions/8471/did-eleanor-roosevelt-say-that-the-jews-brought-the-holocaust-on-themselves/8485?noredirect=1#comment60705_8485)

He pointed me to a dictionary entry for "infer" that seems to use the definition for imply.

http://www.merriam-webster.com/dictionary/infer

The 4th definition that is given there reads:

suggest, hint - "are you inferring I'm incompetent?"

That, to me, is a classic example of where someone should not use "infer", but should use "imply". But maybe I'm wrong?

Granted, it's the 4th of four definitions, and so presumably less common, but I thought the two, "imply" and "infer" had totally distinct meanings. According to definition #4 on that link "infer" can imply "imply".

Am I nuts? Am I the only one surprised by this? (very possible...I'm often surprised by English). Or is it possible that the dictionary is...something less than accurate? It's not a mom-and-pop shop, so that seems unlikely...but I throw it out there as a possibility...but if the dictionary is wrong, where do you go to to find evidence of it? (Who watches the watchmen?)

Community
  • 1
Beska
  • 507
  • 3
    I am equally surprised - the major online dictionaries do license the infer = imply sense. – Edwin Ashworth May 13 '13 at 17:43
  • possible duplicate of Is it correct that imply:infer::sender:receiver? – FumbleFingers May 13 '13 at 18:02
  • 3
    @EdwinAshworth They are obliged to report it; I don't think they claim the authority to license it. I fear that it is in any case inevitable by now. Alas! another arrow removed from my quiver. – StoneyB on hiatus May 13 '13 at 20:07
  • 1
    Merriam-Webster recently updated the meaning of 'literally' to also mean 'figuratively'... so I guess they're comfortable with adopting misusage as usage. – Dancrumb May 13 '13 at 22:43
  • Interestingly, in ironic communication, the ironist's implication is inferred incorrectly by his target, but the target's reply could infer that the ironist's implication was merely an inference based on the implications made by the target of the ironist's implication. Or something like that! To put it another way . . .. Never mind. I'm confused. – rhetorician May 14 '13 at 01:14
  • @Dancrumb Okay...so it sounds like I can just ignore them...'cause that's just crazy talk. I don't like it when language changes! And git the heck off my lawn, you whippersnappers! – Beska May 14 '13 at 01:21
  • 4
    Friggin' M-W and their extremist descriptivist ideology! – Kyle Strand May 14 '13 at 01:54
  • @Dancrumb: "Literally" is never used to mean "figuratively", and neither Merriam-Webster, nor any other respectable dictionary, would claim that it is. (Just because a word is used outside its traditional sense, that doesn't mean that it's used to mean the opposite of its traditional sense. No one uses "infer" to mean "not infer", or "momentarily" to mean "permanently", or "refute" to mean "prove".) – ruakh May 14 '13 at 06:22
  • 3
    @ruakh http://www.merriam-webster.com/dictionary/literally. Strictly speaking, they list the definition as "virtually", but the following note indicates that they are trying to capture its hyperbolic use, ie figurative use. – Dancrumb May 14 '13 at 13:43
  • @ruakh People do use "inflammable" to mean "not inflammable," though. And don't assume that M-W is so "reputable" that it wouldn't dare post a definition that you disagree with; Webster's Third New International Dictionary is one of the most liberal dictionaries (in the sense of including the most "questionable" words, definitions, and pronunciations) ever created. – Kyle Strand May 14 '13 at 16:13
  • @Dancrumb: I looked at the definition, and you're wrong. ;-) "Literally" does have a figurative/hyperbolic use, but that use does not mean "figuratively". (When I say "there were a million people at the party!", I don't mean "there were less than a million people at the party!" I'm applying hyperbole, but I'm not using "a million" to mean "less than a million".) – ruakh May 14 '13 at 16:32
  • @KyleStrand: Who said anything about "a definition that [I] disagree with"? I said that the word is never used that way. Reputable dictionaries only include senses that are actually in use. (It's called descriptivism.) – ruakh May 14 '13 at 16:33
  • 1
    @ruakh No, "in effect / virtually" IS antonymic to the first definition of "literally." If you said "there were virtually a million people at the party!" (which would be a bizarre phrasing, but that's the synonym M-W chose), you would in fact mean "there were less than a million people at the party (but there were still a lot)!" So yes, M-W acknowledges a definition of "literally" that directly contradicts the accepted "standard" definition. – Kyle Strand May 14 '13 at 18:00
  • As for descrpitivism: if you read my other comments and my answer, you'll see that I'm fully aware of what descriptivism is. The fact is that your assertion that this definition of "literally" is not "actually in use" is simply incorrect. – Kyle Strand May 14 '13 at 18:02
  • @KyleStrand: There is a sense that directly contradicts the traditional sense, yes, but it's not simply the opposite of the traditional sense. "Literally" is never used to mean "figuratively". ("Virtually a million people" does not mean "less than a million people", because "virtually" does not mean "less than". It's an entailment, not a definition.) – ruakh May 14 '13 at 19:06
  • let us continue this discussion in chat – Kyle Strand May 14 '13 at 19:14

5 Answers5

16

The same Merriam-Webster link provides the answer further down.

At present sense 4 is found in print chiefly in letters to the editor and other informal prose, not in serious intellectual writing. The controversy over sense 4 has apparently reduced the frequency of use of sense 3.

Oddthinking
  • 3,253
5

In common usage, there is a subtle difference between the two:

Imply tends to refer to meaning that is intended by the author/speaker.

Infer, on the other hand, tends to refer to meaning that is gleaned by the reader/listener.

For this reason, if communication is taking place effectively, the information implied by the speaker should be inferred by the listener, and hence in such circumstances inferred information is implied and implied information is inferred.

The difference becomes more important when information is miscommunicated. In such a circumstance, the listener may infer something that was not intended by the speaker, or the speaker may imply something that is not properly communicated to the listener.

Here we can see that the two phrases here are equivalent:

"are you inferring I'm incompetent?"

"have you concluded that I am incompetent?"

This is subtly different to

"are you implying that I'm incompetent"

which means

Is it your intention to state that "I'm incompetent"?

In the former, the speaker wishes to know if the listener has deducted or concluded that the speaker is incompetent, presumably based on the conversation with the speaker. In the latter, the speaker is not asking if the listener thinks the speaker is incompetent, but rather is asking whether the listener is intending to call (either indirectly or directly) the speaker incompetent.

Consider the difference thus:

1: We've read the documents surrounding your conduct on the McNeilson case.

2: Have you inferred that I am incompetent? (Having read the documents, is it your conclusion that I am incompetent?)

versus

1: I've submitted some documents surrounding your poor decision making to the McNeilson panel.

2: Are you implying that I'm incompetent!? (Are you intending to suggest that I am incompetent?)

Community
  • 1
Matt
  • 1,516
  • Well, yes. That's always been my interpretation as well...which is what I was telling him. And then, to prove himself right, he pointed me to definition #4, on the link I included...and promptly threw me for a loop. – Beska May 13 '13 at 17:32
  • 3
    Yes - this answer is not addressing the question, but is merely showing the distinction between 'sense 1' (and probably sense 1 without the scare quotes, too) of imply and 'sense 1' of infer. After reading the dictionary definitions, I'm going to have to choose other constructions for clarity. – Edwin Ashworth May 13 '13 at 17:45
  • 1
    No one asks another person if he has inferred incompetence. Common usage (and common emotional state) usually prompts the speaker to accuse the other of implying incompetence, on the assumption that such an implication would be false and offensive. – vidget May 13 '13 at 21:24
  • 1
    @vidget: "No one asks another person if he has inferred incompetence": Well, apart from the Mirriam Webster Dictionary that the OP is quoting from, which states (and I quote): "are you inferring I'm incompetent?" – Matt May 13 '13 at 21:58
  • @Matt: Are you inferring nobody else does? Or inferring nobody else does? Or both? – Edwin Ashworth May 14 '13 at 11:08
  • @EdwinAshworth: I'm implying that infer does not imply imply. I'm also inferring from the dictionary that "are you inferring that I'm incompetent" does not imply the same meaning as "are you implying that I'm incompetent", although a casual listener might infer the same meaning from both. That all being said, I have no idea whatsoever what you're inferring from my answer, or what you're implying that I'm implying. – Matt May 14 '13 at 13:48
  • @Matt - and everyone agrees that the answer to that guy's question is 'yes.' :) – vidget Jul 03 '13 at 16:30
5

An analogy here -

In the American South, it is not uncommon to say, "I'm going to learn you some manners!" when in fact the appropriate phrase is "I'm going to teach you some manners."

When two words identifying the same activity (in the case of infer / imply - extracting an implicit meaning from a text or teach / learn - to pass on knowledge in some fashion) but from opposite points of view (the author implies, the reader infers), it is not unreasonable for the mistaken usage to gain currency.

Here, I would suggest the same issue - a misuage that has gained sufficient currency as to be recorded. The definition is thus descriptive, not prescriptive,

Affable Geek
  • 4,558
  • 5
    The usage is old. usage: Many usage guides condemn infer when used to mean “to hint or suggest,” ... holding the position that the proper word for this meaning is imply, and that to use infer for it is to lose a valuable distinction. Many speakers and writers observe this claimed distinction scrupulously. Nevertheless, from its earliest appearance in English infer has had the sense given in definition 3 above, a meaning that overlaps with the second definition of imply when the subject is a condition, circumstance, or the like that leads inevitably to a certain conclusion or point. (Webster) – Edwin Ashworth May 13 '13 at 18:56
1

The speaker or writer implies by what he says / writes.

The listener / reader infers (rightly or wrongly) from what he hears / reads.

TrevorD
  • 12,206
0

Prescriptivism and Descriptivism

It seems to me that this comes down to a matter of prescriptivism versus descriptivism. The question "can 'infer' mean 'imply'?" can be interpreted in a variety of ways, but the two main issues can, I think, be summarized as follows:

  1. Do English-speakers use "infer" as a synonym for "imply" often enough for this meaning to be considered a standard usage?
  2. Is this usage correct? (Tangentially, who defines what "correct" usage is?)

The first question is essentially the question a descriptivist would ask (though they'd probably object to my use of the word "standard"); descriptivism is concerned with describing the language as it is. The second is the question that prescriptivists ask; prescriptivism is interested in prescribing the language as it should be (or rather, as they think it should be).

You have phrased your question with fairly prescriptivist assumptions: you treat the dictionary as a normative reference, and you clearly think it's possible that the dictionary is simply erroneous.

But you should be aware that linguists tend to reject pure prescriptivism; natural languages are not and in fact cannot be absolutely determined by any codified set of rules, whether those rules are contained in grammar books or dictionaries. (Though see the first few comments below regarding French and Spanish, which do have codified rules controlled by governing bodies.)

The relative merits of prescriptivism and descriptivism are well beyond the scope of this answer, but I'd encourage you to read about them. David Foster Wallace has an excellent article on the politics of dictionary-writing that can be found here, and my favorite Language Log post about prescriptivism and descriptivism (and the absurdity of totally embracing one at the expense of the other) can be found here.

My opinion on the case at hand

I tend to be more of a prescriptivist than most; while it's ridiculous to act like you're a member of some sort of "grammar police task-force," I think it's worthwhile to encourage people to speak and write clearly and concisely, and the "rules" (which in many ways are really just well-attested patterns) of the language help accomplish this. I therefore promote and defend the "rules" that I think benefit the language and generally disparage and break those that don't (such as the split-infinitive "rule").

In the case of "infer" and "imply," most usage treats them as complementary (i.e. not as synonyms). This is a positive feature of the language, so I endorse it, use the words in that way, and encourage others to do so as well.

Community
  • 1
Kyle Strand
  • 494
  • 1
    You say natural languages are not and have never been subject to any codified set of rules. Some are, such as Spanish or French. Descriptivism is big in the anglosphere, but not that big outside it. – CesarGon May 14 '13 at 07:47
  • @CesarGon Really? You may be misunderstanding my point--even if descriptivism as a linguistic philosophy isn't big in other languages, the nature of a natural language is such that it cannot be completely determined by rules that someone wrote down; if it were, then it could not evolve until the rules changed to allow it to evolve. The most static and rule-bound language I know of is Quranic Arabic, but that's really just a preserved form of the predecessor to modern Arabic. So are you really saying that Spanish and French are determined by codified and authoritative rules? – Kyle Strand May 14 '13 at 16:02
  • If so, where and how exactly are the languages codified? Do you have some links or book recommendations to support this? – Kyle Strand May 14 '13 at 16:08
  • Would not the addition of '3. Does the inclusion of a usage in more than one dictionary recognised as having reasonable authority (eg Collins; AHD), and the recognition that almost all dictionaries nowadays are corpus-driven, influence your answer to 1 and 2?' be an improvement? – Edwin Ashworth May 14 '13 at 20:34
  • @Edwin Hm....I'm not sure; I think that would obscure the explanation of prescriptivism and descriptivism, and it doesn't really match up that well with the rest of my answer. Which is not to say that it's not a good way to interpret the original question. – Kyle Strand May 14 '13 at 20:45
  • @KyleStrand: I agree that natural languages are moving targets that change continuously and therefore cannot be completely determined by rules (your words). But between this and never having been subject to any codified set of rules (again your words) there is a huge distance. Spanish and French, to use my examples, are both subject to many codified rules and still not completely determined by them. – CesarGon May 14 '13 at 21:02
  • @CesarGon by "subject to" I meant "determined by," i.e., in an absolutist prescriptivist sense. I'll edit my answer accordingly. – Kyle Strand May 14 '13 at 21:13
  • 1
    @KyleStrand: That's fair enough. ;-) – CesarGon May 14 '13 at 21:24
  • @KyleStrand: I've got frustrated at what I've read in some of my favourite dictionaries. I'm guessing John Lawler holds ceremonial burnings of dictionaries. I'm with you in not liking the imply - infer splurge that the dictionaries mentioned seem to license. But if we set ourselves up as higher authorities than them (and they themselves defer to popular usage), without other authorities to back our point of view, we're arch-prescriptivist. – Edwin Ashworth May 15 '13 at 22:27
  • @Edwin it's only prescriptivist if you claim yourself to be a normative authority on the matter, and tell people who disagree that they are simply wrong. There's nothing wrong with this if, say, you're a teacher, or if you're a famous author and you're tongue is somewhat in your cheek. But since I am neither a teacher nor a famous author, I don't insist that my usage is the only correct usage--I just try to convince people that it's preferable to other usages. – Kyle Strand May 16 '13 at 17:04
  • I know it's not really a fair argument, but I'm pretty sure that more people use infer in the imply sense than use indefatigable say correctly (or at all). And Webster's has the comment: '...from its earliest appearance in English infer has had the sense given in definition 3 above, a meaning that overlaps with the second definition of imply when the subject is a condition, circumstance, or the like that leads inevitably to a certain conclusion or point.' Perhaps this is sufficient reason for a 'yes' answer to your questions (1) and (2). – Edwin Ashworth May 16 '13 at 19:40
  • @Edwin which is why I care more about what types of rules I like than what's historically documented. I dislike that usage of infer, and I'll discourage it, but I don't think it's "wrong." – Kyle Strand May 16 '13 at 19:51
  • You're right, though, I think, that the answer to the first question is undoubtedly "yes;" the best answer to the second question, though, is probably just "define correct, then we'll talk." – Kyle Strand May 16 '13 at 19:52
  • Define your own terms. – Edwin Ashworth May 17 '13 at 21:59
  • 1
    @Edwin I'm not sure what you're saying; do you think the answer as I gave it is incomplete? – Kyle Strand May 17 '13 at 22:26