G is for Gerund

6 06 2010

from Palmer's New Method Grammar 1938

I regularly rebuke my MA students for using the term gerund in their language analysis assignments, pointing out that in An A-Z, for example, there’s no entry for ‘gerund’. I am not alone, of course: neither Swan nor Murphy use the term in their various grammars, preferring, like me, the (somewhat ungainly) term ‘-ing form’. Most current coursebooks (e.g. Global)  follow suit.

Why this antipathy to the poor old gerund?  How else, after all, can we distinguish between the very different functions of -ing words, as in (1)  I went running (where running is more verb-like, hence a participle) and (2) Running keeps you fit (where running is more noun-like, and hence a gerund)?

Well, if only it were that easy. In fact, there seem to be a whole range of -ing form uses that cover a spectrum from total verby-ness to total nouny-ness (not to mention total adjectivey-ness), and with lots of  instances that are somewhere in between.  Consider these examples (from the British National Corpus) and you tell me which are gerunds and which are participles:

She remembers running up the aisle

a £17,000 flood after the royal taps were left running.

A HUSBAND and wife were convicted yesterday of running a brothel

I knew that my horse was capable of running well

We love running, so why not save on wedding cars?

Running a large application on a PC is fine

Mary waited and watched her running down the road.

Tomorrow Mr Foggerty is taking them running

there was a running battle between police and conservationists

accusations that its trains have suddenly started running late

“The only way you’re going to hurt me is by running away from me,”

The sort of things they do in the wild, running , jumping, pirouetting

As  Swan (2005) notes: “The distinction between ‘participles’ and ‘gerunds’ is not always clear-cut, and it can sometimes be difficut to decide which term to use” (p. 270).  While for Close (1981) the difference doesn’t really matter: “Whether the – ing form of the verb is what is traditionally called a present participle or a gerund is often an academic question, of no great importance.”

(By way of an aside, the difference did once matter, according to Sweet (1892). Historically there were two distinct forms for what we now call the present participle and the gerund: e.g. OE bindende and bindan, both meaning “binding”. By the Late Middle period changes in the suffix had narrowed the difference to bindinge and binding, a difference that effectively collapsed in the early Modern period, although the gerund still required an article: He thanked me for the binding of the book.)

Of arguably more importance, from a teaching point of view, is this: What is the rule that governs the choice of verb form – either -ing or infinitive – after particular verbs? Most grammars assume that there is none, that it’s simply an issue of collocation. Palmer (1938), in comparing these two sentences:

1. He begins to speak.

2. He stops speaking.

asks  “Why do we have to use the infinitive in one case and the gerund in the other?  Well, nobody knows what the real reason is, so we shall not try to find it”  ( p. 173)!

There are other grammarians who are less timid. Yule (1998), for example, argues that the choice is semantic, and that the to form (as in He begins to speak) connotes more verby-ness than nouny-ness, while the -ing form is just the opposite. In short, he distinguishes between ‘noun-like events’ and ‘verb-like acts’.  Thus, in the sentences:

1. He  considered going to the beach.

2. She suggested going to the museum.

the occurence of the verbs with an -ing complement “may be explained by thinking of the object of consider and suggest as an event (something more noun-like) than as the performance of an act” (p. 218). Whereas, in:

3. She told him to go without her.

4. He wanted her to go too.

“the focus is on the go act, and not on the event. The agency or performer of the act, is also mentioned in each case” (p. 219). Yule goes on to show how the following pairs exhibit the above  semantic differences:

5. I like to box/to dance/to swim/to ski.

6. I like boxing/dancing/swimming/ skiing.

“In [5], the speaker has to be talking about herself as agent performing the acts indicated in the complement. In [6], it is the event, not the act, that is the focus of attention, with the possibility existing that the speaker herself is not a performer in the events mentioned” (p. 220).

What do you think of Yule’s rule? Does it hold up over a range of verbs? And, more to the point, is it teachable?


Close, R. 1981.  English As a Foreign Language (3rd edition). George, Allen & Unwin.

Palmer, H.  1938.  The New Method Grammar. Longman, Green and Co.

Swan, M. 2005. Practical English Usage (3rd edition). Oxford University Press.

Sweet, H. 1892. Short Historical English Grammar. Oxford University Press.

Yule, G. 1998. Explaining English Grammar. Oxford University Press.