Friday, 20 November 2015

On grammatical facts, fictions, and The Spectator

A correspondent writes to ask if I’d seen the silly test from the grammar pedant N M Gwynne in The Spectator (17 October), as she’d had a problem with it. Not only had I seen it, I’d already written a letter to the magazine about it - but they didn’t publish it. The Spectator seems to be only interested in opinion, not facts, linguistic or otherwise.

I’m not at all surprised my correspondent had a problem. The test asked readers to ‘give the parts of speech, including the grammatical part of any verbs, of “boiling” and every instance of “washing” in the sentence, “She is washing in boiling water yesterday’s washing in the washing machine that she uses for washing clothes”.’

Gwynne provided the answers in the letters column of the 7 November (I give his exact words):

Boiling: present participle (verb-adjective)
First ‘washing’: taken with ‘is’, continuous present tense, active voice and indicative mood. By itself, present participle.
Second ‘washing’: either gerund (verbal noun) or gerundive.
Third ‘washing’: noun acting as an adjective (‘noun-adjective’).
Fourth ‘washing’: gerund, acting both as a noun and as a transitive verb.

A letter in the issue of 14 November tells us that only 29 people attempted it and only one got the above answers. This didn’t surprise me either. I suspect most readers of the Spectator were sensible enough to see through the artificial nature of the exercise, with English being forced into the categories devised for Latin. Gerunds and gerundives have no place in an English grammar. And doubtless there were those, whose grammatical knowledge is better than Gwynne’s, who were marked wrong because they didn’t conform to Gwynne’s own misanalysis of ‘washing machine’. This, of course, is a compound noun - recognized as such in every dictionary - so the first element shouldn’t be classed as a separate part of speech at all.

Heaven knows how people are supposed to make sense of the jumble of terms in the answers. A present participle is a verb-adjective. One ‘washing’ is either a gerund, or a gerundive - though it can hardly be both at the same time. Another is apparently both a noun and an adjective. Another is both a noun and a verb. No wonder people are put off grammar when presented with this kind of thing.

Fortunately, modern approaches - as opposed to these resurrected Victorian ones - present English in a much more straightforward way. ‘Boiling’ is an adjective; ‘washing’ in ‘yesterday’s washing’ is a noun; ‘washing’ in ‘washing clothes’ is a verb; and so on. That is all that needs to be said, when first introducing word classes. Introducing imagined parallels - for instance, that ‘boiling’ is an adjective that reminds you of a verb - is an unnecessary confusion when identifying parts of speech.

Gwynne is so out of touch with what is actually happening in schools today. He says on his website that grammar ‘has by now been almost entirely abolished’. Tell that to Buckinghamshire teachers, with their splendid Grammar Project - to name just one of many initiatives taking place around the country. Yes, the kind of grammar presented in Gwynne’s Grammar has indeed been almost entirely abolished in schools, and that’s a very good thing. But it’s been replaced by an approach which respects English for what it is, and doesn’t try to treat it as if it were a bastardized form of Latin.

By the way, while I’m in this mood, I have a second piece of evidence to support my contention that the Spectator isn’t interested in facts. A few weeks earlier (24 October), a writer penned a travel piece on Anglesey, where I live, praising its natural splendour but denying that that there were any art galleries on the island, and saying it was impossible to go to an opera there. I wrote a letter pointing out that Llangefni has a fine art gallery, Oriel, with a stunning Tunnicliffe collection, and that the Ucheldre Centre in Holyhead has art exhibitions all year round. And Ucheldre regularly presents operas - Wales National Opera and Swansea City Opera, to name just two - as well as giving people the chance to see Covent Garden et al there too, thanks to streaming, at a fraction of the London prices.

They didn’t publish that one either. I think I'll stick to blogging.


Phillip Minden said...

Here's what I sent in (I suppose I'm one of the 28 who got it wrong):

She is washing [participle, and with "is" the finite verb] in boiling [participle used as adjective] water yesterday’s washing [noun] in the washing [gerund, and with "machine" part of a compund noun] machine that she uses for washing [gerund] clothes.

John Cowan said...

I don't think it's so much that they care for opinions rather than facts, but that (like all media) they want controversy and manufacture outrage to get it. "If it bleeds, it leads."

Raul Pope Farguell said...

The Grammar Project looks very promising! Would looking to apply the findings to EFL mean task-based learning within a Test-Teach-Test methodology?

DC said...

For those interested, the web address of the Bucks project is

Joe said...

I'm not sure I would call "boiling" an adjective. My preference would be to call it a verb form that modifies a noun. I don't think it's gradeable, nor do I think it can function as a PC (?the water seemed/appeared boiling): in "the water is boiling," I think it is pretty clear a gerund-participle rather than an adjective. Other than it's function, what makes it an adjective?

(The larger point about the quiz is true, but I think it could be a useful exercise in distinguishing word class from grammatical function.

DC said...

Both function and form are needed to set up word classes in English - one of the big differences from Latin.
Adjectives can be gradable or nongradable (eg single vs married).
Nothing wrong with introducing more subtle distinctions to highlight types of adjective, of course, but not at the 'top level', where the fewer the word classes, the simpler the description.

Joe said...

Well, would you call "brick" in a "a brick wall" an adjective? Obviously, not all adjectives are gradeable, but if other word classes than adjectives modify nouns, then there's not that much evidence that "boiling" is an adjective than a verb-form. Since we need the verb form "boiling" anyway (I assume you agree "boiling" is a verb form in "the water is boiling"), we're not multiplying classes: we have the exact same number of classes). Anyways, there probably isn't a way to resolve the debate, but I would say that the fact it isn't gradeable or able to function as a predicative complement would be strong evidence it isn't an adjective.

DC said...

Sure I would. It's a red wall, a big wall, a brick wall - hot water, boiling water… all telling us an attribute of wall, water etc. One has to think semantically about these things, not narrowly grammatically. Grammar without semantics (and pragmatics) is pretty sterile, as I argue in Making Sense of Grammar.

David Crosbie said...

Quirk Greenbaum, Leech and Svartvik produced this beautiful cline with noun at one extreme and verb at the other.

[1] some paintings of Brown’s
[[a] ‘some paintings that Brown owns’; or
[b] ‘some paintings painted by Brown’]

[2] Brown’s paintings of his daughters
[[a] ‘paintings depicting his daughter and painted by him’; or
[b] ‘paintings depicting his daughter and painted by someone else but owned by him’]

[3] The painting of Brown is as skilful as that of Gainsborough.
[[a] ‘Brown’s mode of painting’ or
[b] ‘Brown’s action of painting’]

[4] Brown’s deft painting of his daughter is a joy to watch.
[‘It is a delight to watch while Brown deftly paints his daughter.’]

[5] Brown’s deftly painting his daughter is a delight to watch.
[= [3b] or [4] in meaning]

[6] I dislike Brown’s painting his daughter.
[‘I dislike either
[a] the fact or
[b] the way that Brown does it.’]

[7] I dislike Brown painting his daughter (when she ought to be at school).
[= [6a]]

[8] I watched Brown painting his daughter.
[[a] ‘I watched as Brown painted’; or
[b] ‘I watched the process of Brown(‘s) painting his daughter.’]

[9] Brown deftly painting his daughter is a joy to watch.
[= [3a] or [4]]

[10] Painting his daughter, Brown noticed that his hand was shaking.
[‘while he was painting’]

[11] Brown painting his daughter that day, I decided to go for a walk’
[‘since Brown was painting’]

[12] The man painting the girl is Brown.
[‘who is painting’]

[13] The silently painting man is Brown.
[‘who is silently painting’]

[14] Brown is painting his daughter.

DC said...

Thanks, David. Yes, continua of this kind show the reality of English grammar well. I also like the way that team uses tables and criteria, with pluses and minuses in the cells, to show different kinds of adjective, adverb, etc. Underlines even more, to my mind, the artificiality of the approach of people like Gwynne, which totally ignores such things.

John Williams said...

> Fourth ‘washing’: gerund, acting both as a noun and as a transitive verb.

This is 'washing' as in 'washing machine'

Even Gwynne in his twisted logic gets this slightly wrong. If 'washing machine' = 'machine for washing clothes' (transitive verb), then he needs to say something like:

... as a transitive verb with ellipsis / null instantiation of the object

David, stop reading the Spectator. It's not good for your health. :-)

David Crosbie said...

I thought the idea of an English gerunds could be consigned to history, but The Cambridge Grammar of the English Language has given it a new lease of life — sort of.

After some initial dismay, I think I see their point. They make no distinction between most verb-like and noun-like -ing forms, for which they coins the term gerund-participle. But they do distinguish the most noun-like of forms, as found at the head on noun phrases. These they term gerundial nouns.

If I understand correctly, they would describe all but one of the stand-alone washings as gerund-participles. The one exception, I think, would be yesterday's washing, which seems to meet their criteria for a gerundial noun.

That leaves washing machine, which you convincingly exclude on account of being a compound noun.

Boiling, I think, is what they call a participial adjective.

@BobK99 said...

Ignoring Gwynne entirely (a good idea, generally, I think) your mention of the Oriel gallery reminded of a visit to (...? somewhere Welsh, Colwyn) where I saw a shop called Oriel. Am I right in thinking that "oriel" means "gallery"? This would throw some light on the etymology of "oriel window".

Carry on please.


DC said...

Re gerund: this was a Rodney Huddleston thing - not one of his better terminological decisions, in my view. I don't know what the other editors felt about it.

DC said...

Re oriel: yes, this is 'gallery' in Welsh.

David Crosbie said...

Two questions, David:

1. Do you sympathise with the three-way distinction between non-finite verb forms, and thus between non-finite clauses?
• infinitive (with and without to forms and clauses
• ING forms forms and clauses
• past (and/or passive) forms and clauses
Or rather do you sympathise with the view that this distinction is fundamental and other distinctions a side issue?

2. If you do sympathise, what terminology would you prefer over gerund-participle?

DC said...

I've always used the model in Quirk et al's Comprehensive Grammar...: fourfold classification of nonfinites into to-infinitive, bare infinitive, -ing participle, -ed participle.

TonyTheProf said...

Looking up gerunds, which I mainly came across when I did Latin rather than English grammar, where they make sense, I notice Wikipedia has an entry:

I like eating cakes.

Here eating is a gerund; the verb phrase "eating cakes" serves as a noun, being the object of the main verb like.

While "washing machine" clearly functions as a compound noun, the notion that we might see "eating cakes" as some kind of noun beggars belief, and demonstrates how distorted it is to impose Latin categories on English grammar, precisely because it requires a particular verb "like" and other verbs would not do. Have you heard the phrase: "There are eating cakes on the shelves in the supermarket"? An approach which likes at linguistic rules is likely to be far more suitable.

I note that Geoffrey Pullum observes that:

"The important substantive point is that while Latin had a distinction in form between gerunds and present participles, English doesn’t. "

DC said...

Geoffrey's point is exactly right.

I think something must have gone wrong with the typing at the end of your main para. Can't quite grasp what you're saying.

David Crosbie said...

The wording of Tony's question is interesting:

Have you heard the phrase: "There are eating cakes on the shelves in the supermarket"?

It's highly unlikely , to stay the least, that any of us have actually heard the expression. Indeed, it's almost impossible to think of a context where eating cakes would have that meaning. Almost impossible but not quite. However, there would have to be some explicit reference in the surrounding conversation to objects that were inedible models of cakes used in displays, publicity etc.

In that context, something in our heads allows us to interpret eating cakes as 'cakes suitable for eating'. Perhaps Tony was thinking of this 'something' as a linguistic rule. So, he may have intended, it's better to ?look at such explanatory mental thingies rather than consult a database in our memory of the language forms we've actually heard and read.

Incidentally, I've never found that Latin gerunds make all that much sense. The explanations in school Latin grammars are much more applicable to gerundives. And at least gerundives suggests clear 'voice', namely passive. (Anyway, that's my understanding.) Gerunds have active or passive sense, which is pretty confusing in the absence of a grammatical subject.

And it beggars my personal belief not to see "eating cakes" as some sort of noun.

TonyTheProf said...

My background in linguistics is largely informed in this context by Chomsky. He gave a number of examples of sentences which made grammatical sense (at least according to the prescriptive rules) but actually don't make any sense in terms of meaning.

While connected with the verb "like", "eating cakes" may be described as a noun, but once we try to use it as a noun outside that context, as in the example I cite, we see that if it is a noun, it behaves very unlike other nouns. So yes, I am thinking of linguistic rules being violated, rather than prescriptive grammar. But it is surely only by taking something we are told is some sort of noun, and placing it in different sentences, that we see it is a very strange noun indeed.

I am now off to put my eating cakes in the fridge.

David Crosbie said...


Wikipedia isn't written by professional experts like David, it's written by amateurs like me who like trying to explain things. Look again at the Wikipedia article on gerunds. It's changed.

jahangir alom said...

My background in linguistics is largely informed in this context by Chomsky. He gave a number of examples of sentences which made grammatical sense (at least according to the prescriptive rules) but actually don't make any sense in terms.

In that context, something in our heads allows us to interpret eating cakes as 'cakes suitable for eating'. Perhaps Tony was thinking of this 'something' as a linguistic rule. So, he may have intended, it's better to ?look at such explanatory mental thingies rather than consult a database in our memory of the language forms we've actually heard and read.

Random said...

...I just got a distinction for my level-3 open university english grammar course, and that makes no sense to me at all. I have to say, when trying to analyse a sentence, Hallidayan/Systemic Functional Grammar is MUCH easier.