Is a gerund a part of speech?

Question

As far as I know, there are three levels of units below the clauses:

1. Part of speeches
2. Phrases
3. Roles within a sentence like Subject and Object

Now, gerund is being described as a verbal, which is a derived noun,adjective or adverb from a verb,more precisely a present participle which can take a few specific roles in a sentence.

Is a gerund just another category of part of speeches , is it something beyond the tripartite system of levels or are there more than three levels with gerund being within one of it?

Answer 1

The term "gerund" is a fudge. It is used as if it was a subcategory of verb, when in fact, of course, it is just a particular inflected form of a given verb.

Worse than this, this pseudo part of speech, in English at least, is actually 25% a part of speech and 50% a grammatical function label. The reason for this is that people who differentiate between "gerunds" and "participles" (meaning, with regard to English, "present participles") do so on the basis of the words' grammatical relations. So whenever this form of verb is the head of a clause used as a subject or object of a verb or the complement of a preposition, they deem it a gerund. Whenever it is the head of a clause used as a modifier or as the complement of the verb BE in a continuous construction, they deem it a participle. Of course this is to try and differentiate a different form of a verb according to its grammatical function and not its form.

Modern English grammars such as The Cambridge Grammar of the English Language (Huddleston & Pullum, 2002) deem this distinction unjustifiable, and refer to such inflected forms as gerund-participles regardless of their grammatical relations. Notice though that this is a label like plain form or past participle and does not refer to a part of speech in any way.

Notice that gerund-participles are verbs regardless of whether they function as a subjects or objects or modifiers. For example, unlike nouns they take objects:

• Eating children is wrong. (subject)
• The people eating children need to be punished. (modifier)

Nouns cannot take objects. Instead the noun phrase with the comparable semantic relation must occur in a preposition phrase after the noun:

• The government destroyed the bill. (verb with object)
• The government's destruction of the bill. (noun with PP complement).

Unlike nouns, gerund-participles are modified by adverbs just like other verbs. They cannot be modified by adjectives.

• Quickly eating elephants is wrong.
• Eating elephants quickly is wrong.
• *Quick eating elephants is wrong.

---

Here is a pertinent passage by Geoffrey K. Pullum about the function-form confusion. The quote is taken from LEXICAL CATEGORIZATION IN ENGLISH DICTIONARIES AND TRADITIONAL GRAMMARS 2009:

Most of the deepest blunders in English grammar as traditionally presented over the past two or three centuries stem from a single long-standing confusion between (i) grammatical categories or word classes; (ii) syntactic functions or grammatical relations; and (iii) semantic and discourse-related notions.

It is surprising to see the tenacity of this confusion. It does not appear in other domains. People do not confuse butter knives with screwdrivers, even though occasionally someone who cannot find a screwdriver may use a butter knife to turn a screw. Yet in grammar people just cannot keep syntactically relevant categories or classes of words separate from the relational properties they have when used in particular constructions, and cannot keep either separate from meaning. They insist on trying to define the first of these in terms of the other two, and they have done so since the very earliest attempts to write grammars of English.

In short we need to be careful about confusing word categories and functions/grammatical relations. These two things are entirely different.

---

Grammar note

Notice that we can form nouns out of verbs using an -ing suffix in English resulting in multitudinous verb /noun homophones:

• The eating of children is forbidden.

Notice that the noun eating above is preceded by the definite article and takes of children as a PP. It cannot take children as an object:

• *The eating children is forbidden

Answer 2

Edited into something slightly more useful in light of the comments below.

A gerund is not a part of speech in that it is not a category on the level of noun, verb, determiner, preposition, etc. (Of course, you can ask: "What part of speech is that?" when pointing to a gerund, as you can with any word.)

There is a range of items that look gerund-like, and delineating this range is not easy. There are "gerunds", "participles" and, in some terminologies, "(de)verbal nouns" and so forth. The former two are considered forms of verbs. But classifying a given word ending in "-ing" as one or the other of these categories may depend mostly on its behaviour as a noun or verb in the context.

Here's one context that perhaps helps to delineate them:

I enjoyed his singing. (Noun.)
I enjoyed his beautiful singing. (Notice the use of an adjective.)

I enjoyed his singing that song. (Verb.)
I enjoyed his beautifully singing that song. (Notice the use of an adverb.)

And yet in subject position, it can effectively function as an NP again:

His singing was beautiful.
His singing that song was beautiful. (=the mere fact that he sang it was beautiful)
(Also) His singing of that song was beautiful. (=his singing itself was beautiful)

We propose different underlying structures to account for the different behaviour of the -ing form, but I would say it's difficult to parse at best and theoretically hazy at worst.

As has been pointed out below, sometimes this form is simply referred to as a "gerund-participle", and computer parsers treat it quite variably, from a V heading a VP to an N heading an NP to a "G" heading either one to, most perplexingly, a V heading an NP!

In any case, however, it's not a part of speech, but the term is used to refer to one or more verbal forms that behave like various parts of speech.

Answer 3

I think you need another level to find where gerunds fit in:

  0. Inflectional forms of parts of speech
  1. Parts of speech
  2. Phrases
  3. Roles within a sentence like Subject and Object

A gerund in English is an inflectional form of a verb formed by adding the suffix -ing to the base form of the verb. It is found in factive nominalizations, which are noun phrases derived from sentences which refer to a fact.

I've read over the other answers and previous comments, which discuss mostly how to tell whether something is a gerund. However, linguists (at least many of them) learned from Chomsky to distinguish discovery procedures from the science of language -- linguistics. See his discussion in Aspects of the Theory of Syntax and elsewhere. A gerund is what it is, regardless of what you have to go through to figure that out (which might not even be possible in some cases).

One of the difficulties in identifying gerunds is that the inflectional suffix -ing is also a derivational suffix that also turns up in nominalizations. But the -ing derivational suffix, like other derivational suffixes in English, changes the part of speech, in this case from verb to noun. However, the -ing inflectional suffix that creates gerunds, like other inflectional suffixes of English, does not change the part of speech -- you start with a verb and wind up with a verb.

The nature of active and factive nominalizations was clarified by Robert Lees in The Grammar of English Nominalizations. Active nominalizations are formed by first converting the main verb of a sentence to a noun by any of a large number of derivational suffixes, among them -ing, then converting the other parts of the sentence into things that can occur in ordinary noun phrases based on nouns (articles, possessives, PPs). But factive nominalizations, although they are also noun phrases and cannot have subjects or finite verbs, since the gerund remains a verb, still look pretty much like sentences.

Answer 4

It definitely is a part of speech, but (in usual analysis) not in the first level of analysis. In the U PENN tagset (that does strange things to part-of-speech sometimes), all -ing forms of a verb are lumped under the tag VING: It is primarily analysed as a verb, and secondarily as carrying the ending -ing, it is not properly separated from the present participle of the same surface form). Other tagsets may analyse it as a noun primarily.