Not a gerund, not a thing

« previous post | next post »

I have seen repellently bad poetry on various subjects (mortgage services and sewage disposal, to name but two); but my horror at the poem publicized by National Grammar Day was not evoked solely by the poetic standard, low though it is:

I love the King of Ing
He makes me want to sing
Add him to an action word
And it's a gerund… now a thing!

Nor was it that the poet, Nancy Wright, won a prize for it. What makes me shudder is that it does that noun/thing confusion again (the one that underlies Jon Stewart's terror error). Even under the traditional (but incorrect) notion that if you add -ing to a verb stem you get a "gerund" or verbal noun, it is not claimed that you get a thing. What is claimed is that you get a word of the syntactic category Noun, the category that includes (among other words) all of our most basic one-word ways of making reference to things. National Grammar Day is celebrating, rather than condemning, one of the worst and most elementary popular confusions about grammar.

Incidentally, the reason it's a bad idea to use the Latin term "gerund" for words like talking, and a mistake to think that they are nouns, is that the form in question (The Cambridge Grammar calls it a gerund-participle) functions in several different ways:

  • In Talking is not allowed the underlined word is a Subject, and you could say it acts rather like a non-count noun.
  • In They were talking the same form is the Head of a catenative complement, and most definitely a verb.
  • In I bought a talking doll it is an attributive Modifier, and thus functions in one of the ways that are common for adjectives.

So in one seven-word line ("And it's a gerund… now a thing!") the poem perpetuates not one but two long-standing and troublesome blunders. The National Grammar Day organization (if "organization" is the word I'm looking for) should be ashamed of itself.

[Thanks to: Steve Jones.]



73 Comments

  1. John Lawler said,

    March 5, 2010 @ 12:47 pm

    Clearly a typo on the part of the poet; should be "And it's a gerund… not a thing!". Yet another example of Skitt's Law.

  2. Thomas Westgard said,

    March 5, 2010 @ 1:02 pm

    A poem written for children that incompletely defines "gerund" counts as a "horror?" [No, I never said that. Look again. I spoke of "my horror at the poem", not about the poem being a horror. —GKP] Maybe you're not familiar with the challenges of teaching grammar school. In that context, "horror" is seeing that one of your pupils is being abused but you don't have enough proof to get the police to intervene. You pray that kid might someday truly care about the full definition of "gerund."

    Instead of tearing down other people's silly poetry, why not write some of your own? There's a market for this, which the National Grammar Day site obviously tries to serve. Kids and their teachers need entertaining ways to contemplate language more than they need sniping. [Sigh. I'm not sniping at kids, but at a 200-year-old tradition of bad grammatical description that will not die. —GKP] For example, you could treat the "bad" poem as one stanza in a longer piece. You name three -ing usages, and the "bad" poem covers one. Add the other two stanzas, plus maybe another that explains the differences. That would be a service.

    For my part, I think the only horror here is a college professor issuing a strenuous critique of children's poetry. [I explicitly said that my attention here was not on the poetic quality; you're not the most careful reader I ever had. —GKP] On that note:

    There was a Professor named Bowdler
    Who heard shocking things 'bout his mother
    Said Bowdler, "She might
    Wear her sweaters too tight,
    So we'll scour our language to coddle 'er!"

    [OK; you want doggerel? You want poetry that passes muster with a critic? Here's something beautiful (taken from the end of an email from Language Log reader John Cowan):

    A mosquito cried out in his pain,
    "A chemist has poisoned my brain!"
    The cause of his sorrow
    Was para-dichloro-
    Diphenyltrichloroethane.

    Now that's fine poetry.
    —GKP]

  3. Alyssa said,

    March 5, 2010 @ 1:02 pm

    This is something that is always a little hard for me to explain simply to my curious EFL students.

    So even though a verb+ing can act as three different parts of speech, it's always called a gerund (or gerund-participle), right? When you said it's a bad idea to use the term 'gerund', is it because there should be different names depending on how the form is functioning, or because there's a more accurate name for the whole works?

  4. Neal Goldfarb said,

    March 5, 2010 @ 1:17 pm

    I guess what the last line should have said was—

    And it's a gerund… now a region in a domain

  5. Jim F said,

    March 5, 2010 @ 1:22 pm

    TW: I get your larger point, but I've never understood the attitude that it's OK to teach children things that are wrong just because those things are simple to understand. If the kids aren't ready or don't need to know the full complexity of the "gerund" issue that GKP raises, why bring it up at all? I suspect that many topics oversimplified at the grade school and secondary school level are topics that topics that just don't really need to be taught to people at that level. Do kids need to be told that adding -ing to words causes them (in some circumstances) to act like nouns? Assuming this isn't ESL, don't the kids speak the language?

  6. Thomas Westgard said,

    March 5, 2010 @ 1:48 pm

    @Jim F – I think the answer to your question lies in defining what is being taught. In theory, the kid will be expected to memorize a strict definition of "gerund" and somehow that will be useful if repeated verbatim at a later time. Whatever the intentions of the teacher, I don't believe that is even possible, as regards an entire classroom of second graders. One kid might accomplish that (to no real profit) but the rest of the room will learn something else.

    What kids that age CAN learn is that it's possible to break language down into parts, and that people look at those parts, and make names for them. That there's a system to how we make utterances, and above all that it's as much fun as playing with Legos to take the pieces of language apart and reassemble them in novel ways. A kid who really internalizes this has a great advantage in moving forward academically.

    But children are extremely sensitive to adult disapproval, and if you create a minefield of "look where you failed again" on small points, you risk undermining the overall point that language learning can be fun if the grownups don't interfere too much.

    [Thomas now has me exposing small children to disapproval, and probably suspects me of other abuse. But I think it is now some considerable time since he stopped consulting what I actually wrote. —GKP]

    Taking this poem as an example, it actually does a good job of explaining the first usage of -ing. It's a *subjective* matter of the reviewer's attitude whether to focus on what it does right or to focus on what might be added to it to make a more complete picture. Attitude matters a lot: "Billy, you drew an incredibly lifelike portrait of me but the landscape in the background is shitty. Let's talk about how to draw a landscape." Why even go there? And yet, we routinely give children their tests back with all the correct answers ignored and all of the errors circled in red.

  7. Oskar said,

    March 5, 2010 @ 1:51 pm

    If you were to say I'm going to give him a stern talking to, would that instance of talking be considered a gerund? It would, right

    [Spot on. You are correct. And of course (as I suspect you realized) the presence of the to from the talk to someone construction reveals that it is not acting like a noun. You can say *I'm going to give him a stern lecture to, precisely because lecture actually is a noun (as well as, irrelevantly, a verb in other contexts). The moment you start giving some serious thought to the "gerund + noun" equation you start to realize it makes no damn sense. So why is National Grammar Day thoughtlessly encouraging this kind of confused 200-year-old thinking? —GKP]

  8. Thomas Westgard said,

    March 5, 2010 @ 1:55 pm

    The more I think about what I just wrote, the crazier is is to think that a kid should memorize a definition of "gerund." Underlying that would be a belief that there is one, immutable definition when even without looking I'll bet $100 that (1) there is an ongoing academic dispute about what words and usages properly fit under the term, and (2) the definition among academicians has evolved over time. So the assumption that there even exists a definition worth memorizing is fundamentally flawed – such definitions only exist in the larger context of people contemplating language in complex and evolving ways.

    Just like children breaking down their Legos and reassembling them into new shapes, the real value lies in learning to engage in the process, not learning to worship the construct of the moment.

    [Absolutely right. The farthest thing from my mind was making kids learn correct definitions of Latin-derived terminology. What I'd like to see is some more imaginative presentation of English grammar in terms that are actually satisfying because they fit the way the language works, and make you understand it better. "Gerund" is a term that simply gets in the way. What I cannot figure out is why Thomas thinks that I might be an advocate of memorizing definitions of "gerund". —GKP]

  9. Olga said,

    March 5, 2010 @ 2:13 pm

    This takes me back to the days when I was learning English, and we spent a tremendous amount of time studying example sentences in order to answer to question: Is this -ing form a gerund or a participle? I fortunately figured out that you only need to translate the sentence into Latin in order to arrive at the desired answer.

    [It makes me weep. English does not have a distinction between a gerund and a present participle. (Latin did.) It has a highly versatile gerund-participle form ending in -ing, used in about two dozen different ways. Old-fashioned English courses require students to sit there trying to answer questions that have no answer — questions that, as Olga cleverly discovered, can only be answered to the satisfaction of the dope who asked them if you translate the example sentences into Latin and give the answer that would be appropriate for that language instead! —GKP]

  10. Brett R said,

    March 5, 2010 @ 2:58 pm

    Recently, I've seen claims (and haven't had the time to investigate) that there is, at least for some Americans, a pronunciation difference between the gerund-participle in, for example, the following:

    -Running bulls is dangerous.
    -Running bulls are dangerous.

    Anybody got any data to refute or back this up?

  11. Stephen Jones said,

    March 5, 2010 @ 3:15 pm

    So even though a verb+ing can act as three different parts of speech, it's always called a gerund (or gerund-participle), right? When you said it's a bad idea to use the term 'gerund', is it because there should be different names depending on how the form is functioning, or because there's a more accurate name for the whole works?

    I always call it the '-ing form'. It's certainly not normally called the gerund in phrases like 'I'm singing', or 'a singing doll'.

    I was always taught that the gerund was a verbal noun. The Spanish word 'gerundio', on the other hand, refers to all uses of the '-ing form'.

  12. Richard said,

    March 5, 2010 @ 3:15 pm

    I'm not sure if this rinky-dink poem deserves the scorn you're giving it. Yeah, it perpetuates a common misunderstanding, but it's clever and inserts the word "gerund" into young brains without being too terrifying.

    What is more legitimately irritating is that almost every other top entry is about punctuation, perpetuating the more insidious misunderstanding that correct grammar equals masterful comma placement.

    [I am getting tired of everyone refusing to pay attention to what I actually said. But just once more: (1) I expressly said my focus here was not on poetic quality, and (2) I have stated as clearly as I can that I think inserting the word "gerund" into young brains is educationally harmful, not beneficial. —GKP]

  13. John Cowan said,

    March 5, 2010 @ 3:19 pm

    Not only is there a noun/thing-word confusion, there is an verb/action-word confusion. If besiegement is not an action word, what is? It certainly denotes an action! Compared to those issues, I think the use of gerund 'gerund-participle' is trivial. The latter term is rather too clunky for verse. (Me, I call it the ing-form.)

    I don't know that the tendency for spelling (etc.) flames to contain spelling errors has a name, but it has long been noted.

  14. Dierk said,

    March 5, 2010 @ 3:34 pm

    Instead of tearing down other people's silly poetry, why not write some of your own?

    I never understand this reasoning, claiming the critic has to do it better [or at all]. The poet's job is to write poetry, the critic's job is to critique poetry. A critic need not be – probably should not, though French film gives us exceptions to this – practitioners of what they critique. Nobody ever asked film makers, poets or musicians to be critics, although they should be if they critique critics.

    The poem is wrong in almost every aspect of what it claims to convey, to wit, what a gerund is. I, for my part, wonder how '-ing' defines a gerund, I always thought it defines a participle. Both are true, hence it doesn't make much sense to teach people what is taught through the primer rhyme cited. It's a bit like defining 'chair' as something having four legs*.

    Do I also understand you right that the learned should keep mum when encountering inadvertent nonsense instead of pointing out what's wrong?

    *All right, not quite, the four-legged chair example is even less ingenious.

  15. ellis said,

    March 5, 2010 @ 3:35 pm

    I'm not sure I quite follow your argument that, in its noun-like uses, the gerund-participle doesn't refer to a "thing".

    From the Stewart piece, it seems like you're suggesting that only physical objects – things with a physical location – are "things" for the purposes of the traditional rule. Is that right?

    Physical objects are certainly the paradigmatic "things" that primary school teachers talk about when they teach the person/place/thing rule, but I always assumed that this was just because it's easier to teach anything using concrete, experience-near examples.

    After all, activities and abstractions are also commonly spoken of as "things":
    "One thing I never got the hang of was poaching eggs"
    "Peter's big thing is dancing"
    "At this stage, funding is the most important thing"
    "Their prophecying in the camp was thought a new and unlawful thing"

    When I came to think about the noun-ness of "onomatopoeia", "reification" and (as I thought before today) "walking", I assumed my early teachers would happily have counted all of them as "things". Am I wrong, or am I just missing your point?

    [You seem to be missing two different points: (1) to say that a gerund is a thing is to confuse "noun" with "denotation or reference of a noun"; and (2) to suggest that we can first identify the things in the world and then decide to use nouns to name them has it backward: the only sense in which walking is thing-like is that gerund-participles are (sometimes) allowed to function like nouns syntactically. Walking is self-evidently an action; it is self-evidently not a physical object of any kind. And physical objects are the clearest case of what we name with nouns. The syntax drives the intuition of thinghood; it's not the other way around! —GKP]

  16. Chandra said,

    March 5, 2010 @ 3:39 pm

    @John Cowan:

    Maybe we should coin a term for it?

  17. Jerry Friedman said,

    March 5, 2010 @ 3:56 pm

    @John Cowan: John Lawler, who sometimes contributes to alt.usage.english, used the a.u.e name for that phenomenon above: "Skitt's Law". For other names, including the one that seems most popular with the Loggers, see the Wikipedia article on Muphry's Law.

  18. Dan T. said,

    March 5, 2010 @ 4:20 pm

    What is "love"? (Baby, don't hurt me!)

    It can act as a verb ("I love you") or a noun ("Can't you feel the love?"). In some uses, like "I'm a love machine", it even seems to work like an adjective. But it's not really an object, action, or attribute; it's more of a feeling.

  19. Stephen Jones said,

    March 5, 2010 @ 4:33 pm

    it even seems to work like an adjective.

    No, its a pre-modifier noun in a noun phrase surely.

  20. Ellen said,

    March 5, 2010 @ 4:41 pm

    And a pre-modifier noun works like an adjective, even if it isn't one, does it not?

    [No. It's just that everyone confuses the function "pre-head attributive modifier in noun phrase structure" with the syntactic category "adjective". The confusion is encountered all over the place, and I think it's a disaster for lexicography. On this, see my recent paper "Lexical categorization in English dictionaries and traditional grammars", in Zeitschrift für Anglistik und Amerikanistik 57 (2009), 255-273. —GKP]

  21. Jerry Friedman said,

    March 5, 2010 @ 4:42 pm

    Okay, is gerund a bad word for an instance of a gerund-participle that acts rather like a non-count noun, and is participle a bad word for one that is a verb or one that functions as a modifier in a way that's common for adjectives? And if so, why?

    I realize that in the poem that started this, there's nothing about function, but I'm asking in general.

  22. SSP said,

    March 5, 2010 @ 4:52 pm

    @ DanT

    It's "I'm just a love machine."

  23. Dan T. said,

    March 5, 2010 @ 4:57 pm

    Then there's also "This thing called love", so some people do think it falls into the "thing" category (part of the traditional "person, place, or thing" Schoolhouse Rock definition of a noun).

  24. Aviatrix said,

    March 5, 2010 @ 4:57 pm

    The poem doesn't claim to completely define a gerund, or that all verbs to which -ing have been added are always things. It's like taking a four line poem that has someone mixing flour and eggs to produce a cake, and protesting that a cake has more ingredients than that, some cakes are vegan, and flour and eggs can be used to produce non-cake items?

    Skiing is a thing, whining is a thing, and so is nitpicking.

    [The resistance you encounter if you try to correct ancient myths is just amazing. I am saying English doesn't have gerunds, so the point is not to define that term more completely. And the only way in which skiing and whining and nitpicking can be (in a very vague way) be conceptualized as things is that as words they can function in one of the ways that nouns function. But it is very vague indeed. Given phrase for Pete's sake and by dint of hard work and in lieu of payment, you could say that a sake is a thing, a dint is a thin, and a lieu is a thing. But that's not a metaphysical truth. It's just an extraordinarilyh confusing way of indirectly saying something about syntax. —GKP]

    I'd love to see you correct and complete the rest of the poem, as Thomas Westgard suggests.

    [You probably wouldn't. But anyway, I'm not going to. If I take up writing doggerel, I'm going to get some mortgage company to pay me for it. —GKP]

  25. Neal Goldfarb said,

    March 5, 2010 @ 4:59 pm

    @ellis:

    After all, activities and abstractions are also commonly spoken of as "things"

    This parallels my hyperlinkal allusion above to a discusion in Ronald Langacker's book Foundations of Cognitive Grammar of Langacker's view that nouns are (contra GKP's post) characterized by their having the semantic property of THINGiness. That's not the term Langacker uses, though. What he says is that nouns "profile a region in some domain" but he adds that to avoid the spatial connotations of the word region, he sometimes says that "nouns profile things." [Yes; but I didn't say that nouns don't profile things; I said they should not be confused with things. —GKP]

    Of course, Langacker's notion of what counts as a thing is very abstract, so that it includes confusion, disappearance, nonexistence, hypersensitivity, truthiness, vertigo, and all sorts of other, um, things that you can't exactly wrap your hands around.

  26. Spell Me Jeff said,

    March 5, 2010 @ 5:29 pm

    Verbals puzzle the hell out of students. Elementary schools offer all kinds of bad advice on classifying words, in a syntactic context or otherwise.

    Verbs are harder than nouns, I think. Ask a kid to pick out the syntactic verb in a sentence full of participle inflections, and that kid will almost certainly fail. (Unless he grew up reading every day, in which case the study of grammar is purely academic.)

    "A verb shows an action." Isn't that what we all learned. What a useless notion. For starters, it excludes — gee — the copula (which we hardly ever use, of course). Not to mention all verbs related to cogitation, affect, or sensory experience.

    Find me an elementary school teacher who knows that a syntactic verb is the word (or phrase) in a clause that is inflected to show tense. It's such a simple thing to know and explain. But I know college instructors who find the idea a revelation.

  27. Jim said,

    March 5, 2010 @ 5:30 pm

    "Recently, I've seen claims (and haven't had the time to investigate) that there is, at least for some Americans, a pronunciation difference between the gerund-participle in, for example, the following:
    -Running bulls is dangerous.
    -Running bulls are dangerous.
    Anybody got any data to refute or back this up?"

    Not for this American.

  28. Lazar said,

    March 5, 2010 @ 5:35 pm

    "Okay, is gerund a bad word for an instance of a gerund-participle that acts rather like a non-count noun, and is participle a bad word for one that is a verb or one that functions as a modifier in a way that's common for adjectives? And if so, why?"

    I second this. I've never found it problematic to refer to the -ing form as a gerund when it's serving as a noun, and as a participle elsewhere.

  29. Bloix said,

    March 5, 2010 @ 5:39 pm

    "I fortunately figured out that you only need to translate the sentence into Latin in order to arrive at the desired answer."

    A huge problem in understanding English grammar is that there's very little declension. How on earth is a child supposed to understand the notion of parts of speech when the exact same word in precisely the same form can function as two, or three, or even four of these supposedly distinct categories? What does it mean to say that the categories are distinct if there's no reflection of the distinction in the word itself? I doubt that anyone who hasn't studied Latin (or some other IE language) ever really "gets" English grammar.

  30. Bloix said,

    March 5, 2010 @ 5:45 pm

    Response to Jim-
    In your first example, assuming that "running bulls" is a kind of activity taht a person might engage in, the accent would have to fall on bulls:
    Running BULLS is dangerous.

    In the second, the accent would fall depending on meaning:

    RUNning bulls are dangerous (walking bulls are not so bad), or
    Running BULLS are dangerous (running cows are amusing).

  31. Aaron Toivo said,

    March 5, 2010 @ 5:58 pm

    How do we know that the formation of noun-type words such as in "walking is enjoyable" is actually accomplished by the -ing?

    -ing attaches to roots of a single source category and (apparently) produces multiple target ones. This is not highly characteristic of derivational morphemes – usually they attach to roots of one or more source categories and produce a single target one. Neither is the 100% productivity of the suffix; there is no English verb that is not required to take it in the appropriate grammatical circumstances, while most derivational morphemes have idiosyncratic lexical gaps. So I think it reasonable to peg -ing as an inflective suffix. But wait, what? Inflections are not supposed to change word categories at all! And this one clearly can… or can it? Are there tests that can show it really is doing this, rather than, say, there being a zero-morph derivation process that happens to select verbs in -ing? We certainly use plenty of other zero-morph derivation, so it doesn't seem an unreasonable question.

  32. JJM said,

    March 5, 2010 @ 6:39 pm

    I always use "ing form" too.

    Incidentally, I think it could even be argued that in progressive verb constructions like "they were talking" the ing form is an adjective since it qualifies what "they were".

    That's why ing form is so useful – it covers everything!

  33. Dan T. said,

    March 5, 2010 @ 6:54 pm

    And constructions like "the talking dog" use "talking" in a context where adjectives can be used as well.

  34. Karen said,

    March 5, 2010 @ 7:51 pm

    In the southeast, many men and some women preserve the "in/ing" distinction. I heard my brother-in-law tell his son to "stop talkin like a girl" when the boy said "we're playing the big game this weekend". But it's hardly a conscious, or even universal habit.

    What really gets me is that the name "gerund" is used in Slavic studies to refer to a verbal adverb – which really confuses the heck out of students who know what a gerund is, in either other definition.

  35. Waffles said,

    March 5, 2010 @ 10:10 pm

    I guess I don't understand the point of this poem.

    If it's addressed towards little kids, there's going to be a lot of confusion over what "thing" means.

    Basically, the claim is that the word walking differs from, say, the word walked because walking possesses a fundamentamental thingness that is lacking in walked. If you're using the grade school method where nouns are objects; it's useless, because walking is no more an object then walked.

    If you explain that in this case thing means something that you can properly refer to as a thing in an English sentence, then, first of all you're dealing with a definition that a lot of people would never come to on their own, and you still have some ambiguity; you can also say "Chicken soup is the best thing for a cold" or "The best thing to do when you have a cold is to get some sleep".

    And if you're dealing with people who are able to deal with all of that, then you're dealing with people who are probably advanced enough to understand a more proper definition.

    There's probably a useful mnemonic hidden in there, but it sure takes a lot of digging to find it.

  36. J. Goard said,

    March 5, 2010 @ 10:17 pm

    @Ellis, Neal:

    Yeah. I really disagree with GKP's claim, in the old "terror" post linked above, that when we're considering various linguistic contexts to determine part of speech for a word, "these are grammatical questions…not semantic ones." I essentially agree with Langacker that "grammar is conceptualization".

    When English uses "I'm hungry" as its default construction for the phenomenon in question, speakers are analogizing the symbol hungry with beautiful, yellow, old, etc. When French uses "J'ai faim", by contrast, faim is analogized with argent, beurre, lait, etc. (Actually, AFAIKF, to those mass nouns grounded by a partitive.) The analogy is both formal (obviously, since they are using the same construction) and semantic.

    As a researcher and teacher of EFL, I find that a little bit of explicit comparative linguistics often produces an "aha" moment that really changes English awareness. (And comparison not only with L1!)

    Syntactic and morphological tests are crucial for determining how a particular language conceptualizes something in the world, by what other phenomena they analogize it to. OTOH, telling educators and students that part of speech is not a semantic categorization is bound to backfire, because these people intuitively know that by using it in the same constructions, we are thinking of terror like an apple or chair, albeit not in all respects.

  37. Stephen Jones said,

    March 6, 2010 @ 7:10 am

    And a pre-modifier noun works like an adjective, even if it isn't one, does it not?

    Only some of the time.

    And there is a difference in pronunciation and punctuation when you put another adjective in front of an adjective-noun phrase, as opposed to a noun-noun phrase.
    It's a sexy love machine.
    It's a sexy, voluptuous machine.

  38. Thomas Westgard said,

    March 6, 2010 @ 10:21 am

    A distinction without a difference: we could hardly find a better example than GKPs explanations that sift the fallout of a poem for children "being a horror," as opposed to a critic "feeling horror about it." The other inserted clarifications only dig the hole deeper.

    @Dierk, who questioned the principle of suggesting that GKP write his own poetry instead of critiquing others' – I actually agree with you in principle that poets should poeify and critics should critique. In this case the critic (GKP) doesn't understand the work or the context in which it plays a role. My point of disagreement with GKP is that he holds a work against a standard it was never intended to meet, and inevitably finds it lacking. You might as well attack Bob Marley for failing to use iambic pentameter, which, although it does have a beat, is nearly impossible to dance to. Context and audience matter, and a good critic would take these things into account. GKP didn't.

  39. Spell Me Jeff said,

    March 6, 2010 @ 10:23 am

    [quote]And constructions like "the talking dog" use "talking" in a context where adjectives can be used as well.[/quote]

    Of course. You're simply observing the difference between the slightly useful dictionary way of typing words out of context, and the much more useful strategy of classifying a word/phrase based on the way it is deployed in a clause.

    Computer scientists call this duck typing. I.e., a variable holds an integer value when you use it like an integer. It holds a string (of characters) when you use it like a string.

    If you use a word like a verb, it's a verb. If you use it like an adjective, it's an adjective.

    I am going to the store tomorrow.
    I go to the store tomorrow
    I am go to the store tomorrow
    I going to the store tomorrow

    Only a pedant would claim that 3-4 make no sense. For the same reason, we can inflect any word in a verblike way and, given the right context, be perfectly intelligible:

    Did you watch Tivo last night?
    Yes. We Tivoed all night.

    You never set a second chance to make a good first impression.
    You sure? I first-impressioned Brenda four or five times last night.

  40. Eli Morris-Heft said,

    March 6, 2010 @ 10:25 am

    "A huge problem in understanding English grammar is that there's very little declension. How on earth is a child supposed to understand the notion of parts of speech when the exact same word in precisely the same form can function as two, or three, or even four of these supposedly distinct categories? What does it mean to say that the categories are distinct if there's no reflection of the distinction in the word itself? I doubt that anyone who hasn't studied Latin (or some other IE language) ever really 'gets' English grammar."

    It's amazing that kids even ever learn to talk at all! And by the time they're three, in most cases!

    English doesn't have very little declension; it has some weak morphology and lots of zero-derivation. To only focus on the word itself is a bad approach, though; when we speak, we have entire sentences and discourses to give us context. In that context lies the key to making a decision about the word's category, not in the semantic content of the word alone.

    If we do want to teach categories to kids, we should do it by focusing on how the word acts in the sentence or phrase and either by acknowledging that a word can do multiple jobs, and that sometimes there are words that sound the same and have related meanings but are different words. (An example: the verb 'walk' ("I walk every afternoon.") and the noun 'walk' ("I take a walk every afternoon.").)

    As a parting shot to your parting shot about Latin: Setting aside that it would make more sense to advocate German as a language to help understand English grammar, my (admittedly paltry) knowledge about Latin never helped me with grammar or parts of speech. Not even my (more extensive) studies in Spanish helped me. Though it's impossible to understand and ridiculous to view English grammar through another language (as Olga's incredulous strategy indicates), the language that has given me the most insight into English is Japanese.

  41. Michael B said,

    March 6, 2010 @ 10:35 am

    Hello, all. New to this blog, which was recommended on an editing list.

    Two comments:

    1.
    I was under the impression that a gerund in English grammar is a present participle that is used syntactically in non-verb contexts, e.g., as a noun, adjective, or adverb.

    Swimming is permitted.
    The swimming dog is here.
    That spaniel is the swimmingest dog I've seen. (ad hoc usage, mostly)
    It's going swimmingly today.

    The verb ending "-ing" of a present participle is a morphological change to the root word used to indicate some kind of continuity and is always used with some other (periphrastic) verb word.

    The term "gerund," on the other hand, designates a present participle that has been adopted as a noun, adjective, and sometimes adverb. The adjectival use is almost always indicated only by position or word order (rarely by a morphological change); the adverbial use is usually marked by the -ly ending, as shown above. Gerunds are widely used as nouns and often show noun morphological changes, such as plural "s" and possessive "'s." Ex: An actor's clippings, the "Takings" clause of the Constitution, on the hustings, etc.

    The subject-verb-complement sentence form tends to blur some distinctions between the participle and gerund.

    "He is cooking." i.e., food is being prepared by him on the stove.
    Thus: He | is cooking. "Is cooking" is the verb, and "cooking" is a present participle.

    "He is cooking," i.e., he is really hip, with it, on top of things, etc.
    Thus: He | is | cooking. "Cooking" is a slang term used as a predicate complement—an adjective—linked to the subject by the verb "is."

    2.
    "Gerund" is not too hard for small children, nor is "participle," for that matter. It's what their brains are made for! Children love to learn words and then to detach the word from its meaning [a gerund, btw] and move it around. One of the earliest forms of humor is punning [another gerund].

    And besides, it gives a child a chance to say from the back seat, "Daaaaddy, that's a gerund!"

  42. Stephen Jones said,

    March 6, 2010 @ 11:46 am

    I've never found it problematic to refer to the -ing form as a gerund when it's serving as a noun, and as a participle elsewhere.

    The problem is it is difficult to make the distinction in many cases.

  43. Stephen Jones said,

    March 6, 2010 @ 11:49 am

    A distinction without a difference:

    What are you referring to?

  44. Thomas Westgard said,

    March 6, 2010 @ 12:33 pm

    @Stephen – GCK drew a distinction between what he said, feeling horror about a poem for children because it imprecisely defines "gerund," and what I claimed he said, that the poem *was* a horror because it imprecisely defines "gerund." Either way, the concept of horror was offered for a situation where it clearly does not apply.

  45. Mark Liberman said,

    March 6, 2010 @ 1:25 pm

    An 8th-grader of my acquaintance says that "gerund sounds like it ought to be a kind of root vegetable". This makes sense to me, although turnip, rutabaga, potato etc. don't seem to be especially near neighbors in any space that I can think of. But there might well be a future for mashed gerunds, gerund chips, leek and gerund soup, and of course the inevitable question, "you want gerunds with that?"

  46. Mary Bull said,

    March 6, 2010 @ 1:47 pm

    Ah, a gerund? Quite tasty and round?
    From the German word "rund" let's expound:
    Gerund! Might grow under the ground,
    But in grammar to use it's unsound,
    And Dierk's also correct, I'll be bound!

  47. Simon Cauchi said,

    March 6, 2010 @ 2:35 pm

    Thomas Westgard, who I take it must be a schoolmaster, has too narrow an idea of the meaning of the word "horror". The word can be applied quite properly to a silly poem that perpetuates a grammatical misconception, just as it can to a bad haircut or a ludicrous choice of clothes or a poor musical performance, etc., etc.

    Anything, in other words, that (metaphorically) makes your hair stand on end.

  48. ellis said,

    March 6, 2010 @ 3:24 pm

    [You seem to be missing two different points: (1) to say that a gerund is a thing is to confuse "noun" with "denotation or reference of a noun"; and (2) to suggest that we can first identify the things in the world and then decide to use nouns to name them has it backward: the only sense in which walking is thing-like is that gerund-participles are (sometimes) allowed to function like nouns syntactically. Walking is self-evidently an action; it is self-evidently not a physical object of any kind. And physical objects are the clearest case of what we name with nouns. The syntax drives the intuition of thinghood; it's not the other way around! —GKP]

    Thanks for the clarification.
    (1) I suppose it depends what your definition of the word "is" is. I don't read the poem as asserting that a "gerund" is itself a thing in a way that an "action word" isn't – the reference to its reference is implicit.
    (2) Yes, and then no. Lets set aside any question of defining "noun" using the primary school rule. I agree with you – there, syntactic function is key.

    It's your approach to "thinghood" that baffled me before, and I still can't make much sense of it. You seem to be saying
    (A) Because physical objects are "the clearest case of what we name with nouns", they must also be the paradigmatic core of the concept: "thing";
    (B) We only consider "walking" to be an instance of the concept: "thing" (or merely "thing-like") because it can function syntactically like a noun, the structure we commonly use to name physical objects.

    But that's absurd. Ever since OE, "thing" has had about as broad a semantic range (although not quite the same) as it does now (ranging through entity, object of thought, matter, deed, fact, event, quality etc), but the restrictive sense you want to make foundational – "inanimate object" – doesn't occur until the late 1600s. It's only ever been one among many, mostly older, related senses – it's not the ground on which they are built.

    It's not that "walking" strikes us as "thing-like" because it can be used in ways that syntactically parallel the use of "doorknob". The word "walking" can refer to any of a number of actual "things" – a leisure activity, a skill for a toddler to learn, a behavioural feature of most legged animals, an object of analysis in biomechanics, etc. No need for intuition.

  49. Ellen K. said,

    March 6, 2010 @ 4:07 pm

    @ Stephen Jones

    I get your point. However, I'm not the one who originally said something worked like an adjective. Dan T, did. You replied, not saying that he was wrong, but saying it's not an adjective, which is irrelevant to whether it works like an adjective. And you've agreed that a premodifier noun (your term) does, sometimes work like an adjective. So, if he was wrong to say "love" works like an adjective in "love machine", you'll have to go beyond saying it's a pre-modifier noun.

    Pullum's reply to me seems to suggest a different perspective, that pre-modifer nouns and adjectives are two different instances of "pre-head attributive modifier in noun phrase structure".

  50. Army1987 said,

    March 6, 2010 @ 4:09 pm

    Replying to Prof. Pullum's comment to Aviatrix's post:
    The point is that thing does not mean “physical object”: it has a much broader meaning than that.
    Searching for thing in the another page which happens to be open in my browser right now (a Wikipedia talk page):
    As a general principle, there's no reason that we need to follow the sources on this sort of thing.
    (Here, this sort of thing refers to the choice whether to refer to ships as it or she.)
    There is one thing we haven't considered yet. Since it appears that "she" is used by significant maritime authorities, …
    Ok, so I did the right thing, then, by reverting it to "she".
    Please don't change things unless you are willing to update the "what links here" links (and section links).
    (Here, things refers to the subdivision of old discussion threads into separate archive pages.)
    It doesn't …if one bothers to read the whole thing carefully and interpret it absolutely literally.
    (Here, the "whole thing" is the text of a template; it comes much closer to being a “physical object” than the others.)
    The last thing we need is to weaponize them with a template.
    Grepping for thing in a particular directory on my laptop (containing the lyrics of the first four Led Zeppelin albums):
    Oh, oh, child, the way you shake that thing, gonna make you burn, gonna make you sting. (OK, this one is a physical object…)
    So of one thing I am sure, It's a friendship so pure, Angels singing all around my door, so fine. Yeah, ain't but one thing to do, Spend my natural life with you, You're the finest dog I knew, so fine.
    (…but these two definitely aren't.)
    Hey, girl, I got something I think you ought to know. Hey, babe, I wanna tell you that I love you so.
    (Neither is this one…)
    Mmm, I'm telling you now, The greatest thing you ever can do now, Is trade a smile with someone who's blue now, It's very easy just… (…nor is this one.)…

    And so on; I found many more occurrences, but the pattern is clear: only a small minority of them – less than a quarter – refer to anything vaguely like physical objects. And I don't think the people writing them had English syntax in mind: things such as some of those are denoted by clauses or even full sentences, which don't behave anything like nouns.

    (Now, it is likely a priori that a WP talk page should be somewhat more abstract than most texts or conversations; but much less so for lyrics to rock songs. I might try with experimental physics textbooks, but I'd need to do that the old-fashioned way because I don't have any in searchable electronic versions. But I think the fact is that adults normally only use thing when no more specific noun is appropriate, which is more common with abstract things than with concrete ones.)

  51. Troy S. said,

    March 6, 2010 @ 4:54 pm

    I read the paper you linked to; interesting stuff about the history of linguistics education, and in particular the suggestion that the distinction between adverbs, adjectives, and prepositions is not a useful one. Actually, what I argue is that not that the distinction isn't useful, but that the boundaries have been drawn in the wrong places: numerous items have been misclassified. —GKP] I will point out, though, that outdated theories are still taught in disciplines outside of grammar, and this isn't necessarily seen as a bad thing. To name one example, the Bohr model of the atom is still taught in elementary physics textbooks, even though it is has been made obsolete by more accurate models. It's still seen as useful pedagogically as a simplification, and I'll wager it's how most high school graduates understand electron structure. There are probably other examples in other disciplines. People often say this to me — that the wrong analysis may be defended on grounds of being simper to teach. I don't think so. I don't see how it can be easier for students to try and figure out which of a bunch of words that look like prepositions are really adverbs, when the truth is that none of them are. The traditional approach says that in They moved down the block the word down is a preposition, but in They moved down a block it isn't. I say it's a preposition in both examples, though in the latter the phrase is not the complement of down, it's an adjunct expressing how far down they moved. That isn't more complex; it's simpler and more accurate. I have many more examples in the paper ("Lexical categorization in English dictionaries and traditional grammars", which I assume is the paper you're referring to). —GKP]

  52. Mark F. said,

    March 6, 2010 @ 6:56 pm

    There is a reason that people want to interpret "thing" as if it means "physical object" when it's used to define the "noun" concept — the most general meaning of "thing" is so general that people have the intuition that it's not suitable for defining something else. The locution "person, place, or thing," which I tend to think of as redundant, just reinforces that idea.

    So I have to back off a little from the notion that all things are things. But I still think it's true.

  53. Ellen K. said,

    March 6, 2010 @ 7:43 pm

    I'm pretty sure I didn't learn what a noun is from anyone fully telling me, but from pointers that helped me figure it out. "Person, place or thing" I think was one of those pointers. No, that definition doesn't explain why some words are nouns despite not being a "thing" as such and definitely not being a person or place. It's okay for kids to learn by figuring things out rather than being told everything they are supposed to know. So, for teaching, the question is not are the descriptions and explanations completely accurate, but, do they point in the right direction. At least, that's how I see it.

  54. Chance said,

    March 6, 2010 @ 7:44 pm

    Geoffrey,

    How dare you try to shake my deeply entrenched belief that English uses gerunds by critiquing this bit of doggerel! I read it to my kindergarten class, and they all started crying. I hope you're happy now.

    Seriously, fascinating stuff.

  55. Neal Goldfarb said,

    March 6, 2010 @ 7:57 pm

    @GKP: “The syntax drives the intuition of thinghood; it's not the other way around!”
    So what motivates the nouning of verbs? And more broadly, what motivates derivational morphology (by which I mean the process, not the study of the process)? Just the desire to use words in different grammatical slots than the ones they usually fill? Or is it also the fact that using a word in a new slot enables you to use the word to mean something new? And if it’s the latter, doesn’t that suggest that it ain’t all that easy to draw the line between syntax and semantics?

  56. ellis said,

    March 6, 2010 @ 7:57 pm

    Mark F : I think there are at least a couple of reasons why people want to interpret it that way – as you say, it is set alongside "person" and "place", and in most settings when "thing" is set in contrast to "person", it does refer to either a non-human animal or an inanimate object (or at least to something quite particular). And secondly, most of the examples used in primary school classrooms are either animals or inanimate objects.

    But I'm still pretty sure that if I'd stuck my hand up and said "But Sir, what about 'party'?", the reply would have been "It's a thing", and on that, my teacher would have been right.

  57. Thomas Westgard said,

    March 6, 2010 @ 9:19 pm

    Mary Bull FTW!

  58. J. Goard said,

    March 6, 2010 @ 11:50 pm

    @GKP:

    (2) to suggest that we can first identify the things in the world and then decide to use nouns to name them has it backward: the only sense in which walking is thing-like is that gerund-participles are (sometimes) allowed to function like nouns syntactically.

    This last part is manifestly untrue. There are many ways in which a spatiotemporally bounded action is thing-like. I can forbid swimming in my duck pond, just like garbage and pirahnas. I can compare it to objects in many ways (Kim Yuna's skating is more beautiful than Rodin's sculptures). I can take a picture of kissing no less than a tree. The thing-like characteristics, or rather, the cognitive mechanisms by which bounded actions may be analogized with physical objects, seem to be (an example of) the "why" and "how" of syntactic parts of speech.

    Swimming, skating, kissing are not physical objects, to be sure. But nor am I a corporate recruiter when I practice English interviews with my students. Nor is a stick a gun when stereotypical 1950s kids play cops and robbers. Nor do I really put an object inside of a container when I "drag" a "file" into a "folder" on my "desktop". But such conceptual mapping is always based upon some similarity, and it underlies much of what is amazing about humans. Many people think it underlies grammatical structure, too.

  59. Mark P said,

    March 7, 2010 @ 5:19 am

    To name one example, the Bohr model of the atom is still taught in elementary physics textbooks, even though it is has been made obsolete by more accurate models. It's still seen as useful pedagogically as a simplification.

    The Bohr model is not a simplification. It is just plain wrong, which is not the same thing at all. It is not even the situation of, Newtonian physics, which is true in most usual circumstances and is still used in modern physics. No actual chemist uses the Bohr model, even as a simplification.

    Normally kids understand such concepts as the truth, and once locked in it is terribly hard to eradicate.

    As a chemistry teacher, I have felt that science teaching is stupid to do this. We do it because we are locked in for historical reasons, not because it is the best way to teach it. Because it isn't.

    Likewise, what is the point of teaching incorrect grammar, on the basis that it is "close enough"? As the comments on this discussion show, once taught something, many people will simply not accept their knowledge is wrong. They will even argue the toss with a person who has literally written the book on the subject.

  60. Mark P said,

    March 7, 2010 @ 5:21 am

    Apologies for the erratic commas. I didn't do Latin at school, so obviously my grammar is very poor.

  61. Army1987 said,

    March 7, 2010 @ 12:37 pm

    iambic pentameter, which, although it does have a beat, is nearly impossible to dance to

    You suggested me another thing item to add to the list of things to do before dying. Anyway, fewer things are impossible to do with iambic pentameter than one would expect.

  62. Stephen Jones said,

    March 7, 2010 @ 3:56 pm

    And you've agreed that a premodifier noun (your term) does, sometimes work like an adjective.

    You defecate, Ellen K, and water buffaloes defecate. However when you have a shit you are not acting or working like a water buffalo; it's simply that you and water buffaloes have defecation as one of the things you do. Equally, pre-modifying nouns is one thing both nouns and adjectives do, though with different restrictions.

  63. Simon Cauchi said,

    March 7, 2010 @ 4:31 pm

    It's not just water buffaloes and Ellen K that defecate. It's also speech, as Edward Young explains in Night Thoughts, II, lines 477- 487:

    Speech ventilates our Intellectual fire;
    Speech burnishes our Mental Magazine:
    Brightens for Ornament; and whets for Use:
    What Numbers, sheath'd in Erudition lie,
    Plung'd to the Hilts in venerable Tomes,
    And rusted in; who might have born an Edge,
    And play'd a sprightly beam, if born to Speech;
    If born blest Heirs of half their Mother's tongue?
    'Tis Thought's exchange, which like th' alternate Push
    Of waves conflicting, breaks the learned Scum,
    And defecates the Students standing Pool.

    NB Young's "defecate" is transitive (as it should be).

  64. Simon Cauchi said,

    March 7, 2010 @ 4:32 pm

    PS Don't ask me to explain the meaning of line 484.

  65. Ellen K. said,

    March 7, 2010 @ 11:15 pm

    @Stephen Jones. No, me and the water buffalo do work alike. I bring that up, not to correct your biology (as that would be off topic) but because I suspect it's not a difference of biological knowledge, but, somehow, a difference of what we think it means to say that something works like something else. Of course, you could also point out that the reason we work alike in this is because we are both animals. Still, knowing that doesn't take away the alikeness.

  66. Adam said,

    March 8, 2010 @ 6:17 am

    Quoting Jerry Friedman: Okay, is gerund a bad word for an instance of a gerund-participle that acts rather like a non-count noun, and is participle a bad word for one that is a verb or one that functions as a modifier in a way that's common for adjectives? And if so, why?

    I'm inclined to agree. I know that morphologically there are only four or five forms of most verbs in English (live, lives, lived, living; do, does, did, done, doing), but there can be different syntactic uses of some of them.

    (1) I saw Bob with a smoking jacket.
    (2) I saw Bob with a smoking gun.

    Native speakers naturally stress "smoking jacket" as an N-N compound, and "smoking gun" as an Adj-N phrase (unless the jacket is actually smouldering, haha). Doesn't this mean they internally treat V+ing as a noun in (1) and an adjective in (2)? And if so, why is it so heinous to use "gerund" and "participle" to describe those syntactic functions? I admit that they do not correspond exactly to the same terms' functions in Latin — but should we also ban "infinitive" for English because it doesn't have exactly the same functions as the Latin infinitive?

  67. Rob Chametzky said,

    March 8, 2010 @ 11:00 am

    Adam said: "I admit that they do not correspond exactly to the same terms' functions in Latin — but should we also ban 'infinitive' for English because it doesn't have exactly the same functions as the Latin infinitive?"

    Well, yes, we should. But not because of lack of same functions; rather, because there is no 'it' there (cf. Stein 1937 on Oakland). English does not have infinitives. There is a collocation of 'to' with the plain form of the verb (see the CGEL, p.84; and p.581for the corollary that English CANNOT have 'split infinitives' ), which translates the meaning of what, in other languages (e.g., Latin), are in fact 'infinitives'. Wovon man nicht sprechen kann, darüber muß man schweigen

    –RC

  68. Troy S. said,

    March 8, 2010 @ 1:40 pm

    My commentary on the Bohr model wasn't so much a defense of bad analysis as a rebuttal to the lament that linguistics is the only discipline where outdated theories are taught. Of course, in this case, there is still some value. I agree there is no value in teaching grammar rules that have been wrong for hundreds of years (except the practical application that you may need to know them to pass your SATs :P )

  69. Stephen Jones said,

    March 8, 2010 @ 2:57 pm

    what we think it means to say that something works like something else.

    There are three meanings there:
    a) here a noun does the job of an adjective
    b) here the noun acts exactly as an adjective would do
    c) here the noun performs a function that an adjective also performs.

    Only the third of those is true, and as the comparison with humans and water buffaloes attempted to show, it is not a particularly useful bit of information.

  70. Ellen said,

    March 8, 2010 @ 3:16 pm

    I think in Dan T's original comment that started this discussion "works like an adjective is a useful enough way to make his point about the word "love" and how the word functions versus what it is.

  71. EricD said,

    April 6, 2010 @ 2:06 am

    Came here in response to GKP on BBC Radio 4 "Word of Mouth".

    Good insistence on actual research not just guesswork for relative frequency of adjectives for scientists and politicians.
    I suppose "Thought for the Day" isn't exactly a peer-reviwed academic publication, though.

    Scary stuff about noun=thing?, verb=action?, adjective=noun-description (and adverb=verb-description?) !

    GKP: "People, quite generally, are in the grip of this ancient nonsense…"

    GKP: "… a long-standing piece of language myth"

    Michael Rosen: "You grow up with that little dictum in your head, and that's why it leads us into making these major errors."

    GKP: "It's about time we dropped this 200-year-old business of basing everything on vague semantic intuitions that occurred to the Ancient Romans, because it's like teaching pre-darwinian Biology."

    GKP: "Over the last 100 years, the field of Grammar has been completely revolutionised and these old definitions have been tossed-out"

    I doubt if erroneous classification of word-types was an issue.
    I apprecate the programing analogy of the 'duck' or variant data-type: reminds me of ByRef | ByVal, too !

    I have searched in vain for the "New Grammar" !
    Is there a simple primary-level guide ?

    Discussion here seems to hinge on the definition of 'thing'!
    I'm happy with a wide, greedy definition of 'thing', but wouldn't omit the adjectives from this sentence.

    I woud liken it to teaching Newtonian Phsyics first 'F=ma', then Einsteinian Relativity 'e=mc2', then Quantum Mechanics. I wouldn't advocate QM in primary schools!

    General rules-of-thumb first, then handle the 'borderline cases', 'fine details' and extreme exceptions later.

    To help with visualising gerunds, I recommend pointing Google at [ gerund molesworth ] !

    EricD

  72. APC said,

    September 29, 2010 @ 2:37 pm

    I don't think I've seen any mention yet here of forms like the following:

    heading = "the result of our having headed"
    frosting = "the thing which makes it possible for something else to be frosted"
    setting = "the status of something which has been set"
    as in "Nigel changed the volume setting on the amplifier to 11"

    What are these classified as?

  73. Ignacio said,

    October 24, 2010 @ 5:25 pm

    Calling the -ing form a 'gerund' has been one of my personal bugbears for some time. I remember being unconvinced by the terminology as far back as when I was an ESL student in high school, and noticed rather quickly that the -ing form had little relation to what we call a 'gerund' in Spanish. As an ESL teacher now, I tell my students that while they may see the term 'gerund' in grammar books, I'll refer to it in class as the ING (pronounced like CIA, not like NASA), as short for 'ING form', and tell them not to worry too much about the terminology. They usually become used to my sui generis name within the week, and we can all rest easy and get on with the business of learning.

RSS feed for comments on this post