I just came across this book How to Use Parts of Speech. I’ve been meaning to write about the many problems with the traditional method for defining parts of speech, and seeing this book just tipped me over the edge. Sorry J.L. Smith. It’s nothing personal, nearly every grammar book gets this wrong. You just happened to be there today. The book follows the traditional 8 parts of speech system, which is possibly the least insightful guide you could ask for. The traditional 8 parts of speech are defined using a grab bag of semantic and functional explanations. Nouns are described as a person, place, or thing (semantics) while adjectives are described as things that modify other words (functional). Here are those very definitions from How to Use the Parts of Speech. This is the wrong way to do part of speech, and it basically ends up relying on reader intuitions as to what counts as what, rather than actually teaching anything.
Part of speech is better thought of as a “morphosyntactic category”. That’s a big word, so let’s break it down: it means that the part of speech of a word is determined by what it looks like, i.e. what affixes are attached (that’s the morpho- part) and where it appears in a sentence (that’s the -syntax part).
Suppose someone accosts you on the street and demands you tell them what part of speech the word “yellow” is. What do you answer? I think that most people have the intution that it’s an adjective. Certainly by traditional standards that seems right. It’s not a person place or thing (unless you dilute the meaning of “thing” so far as to render anything a noun), and it’s not an action, and we can imagine using it to modify other words.
In fact, this is an impossible question to answer. It can be an adjective, noun, or verb depending on how it is used. You need a context to know for sure.
- I love your yellow wallpaper. (adjective)
- Yellow is my favourite colour. (noun)
- Newspapers yellow if you leave them too long in the sun. (verb)
This is true of many words in English.
Is down a preposition or a verb? “Let’s down a pint down at the pub”
Is bag a noun or a verb? “I can bag your groceries in a paper bag”
The use of a word as two different parts of speech is technically called “syntactic conversion“, and it’s a pretty productive process in English which applies to newly created words all the time, e.g.
- facebook (n.)/to facebook (v.)
- a blog (n.)/to blog (v.)
- a text message (n.)/to text(v.)/to message(v.)
An interesting exception is twitter which came with an irregular pattern: twitter/to tweet.
I know there is contingent of angry and clueless grammarians out there who think it’s wrong to use nouns as verbs. Let’s be clear: There’s nothing wrong with doing this in English. It is in fact a special feature of English, and this is not something you can do in every language. Plus everyone from illiterates to the greatest orators and writers do it.
Of course, straight up conversion is not the way it works all the time. English has affixes that can convert a word to a different part of speech, so the presence of that affix tells you part of speech, even without context. For instance, the word happy is an adjective, but add the suffix -ness and you get a noun happiness. Since words that end in the suffix -ness are always nouns (laziness, redness, carelessness, greatness, etc.), you don’t have to hear such a word in a sentence to know its part of speech.
Interestingly, English suffixes stack in such a way that you only have to look at the rightmost one to determine part of speech. This is sometimes called the “rightmost rule”. Take for example the word “establishmentarian”, which is a noun.
It gets built up like this:
establish -> verb
establish-ment -> ment converts to a noun
establish-ment-ary -> ary converts to an adjective
establish-ment-ary-an -> -an converts to a noun again
You could in principle do some more, but I find it hard to figure what these mean:
establish-ment-ari-an-able -> -able converts to an adjective
establish-ment-ari-an-abl-y -> -y converts to an adverb
Some of these suffixes are homophones, however, so you might still need to hear a word in a context to figure out part of speech. The suffix -s is used for both the third person singular present and the plural, so you can’t tell if chairs is a noun or verb without a sentence.
She chairs the meeting every week.
There are 10 chairs at the table.
So in short there’s nothing useful about the traditional method for deciding on parts of speech because it fails to account for a how English is actually used. You learn literally nothing about English from such an out-dated description of it. Books like How to Use Parts of Speech are just parroting of old prescriptive nonsense without a thought, including the common error of calling “the” an adjective.
There are problems with all of the traditional categories, but I think that “interjections” gets the award for most useless. How to Use Parts of Speech explicitly acknolwedges its junk pile status by writing “interjections are words that are not other parts of speech”. Don’t know what something is? Just call it an interjection. Which words get listed as interjections depends on the book, and in this case we get a grab bag of negation (no), discourse particles (well, hey), expressives (ouch, yuck, yikes), and even a verb in the imperative (help). What’s the use of calling these “interjections” when they have practically nothing in common? It isn’t a coherent category, and it really ought to be dispensed with.
In fact, I think it’s wrong to limit ourselves to only 8 parts of speech. As I outlined above, part of speech is a morphosyntactic category. We should define parts of speech based on their patterning in the language: Which positions does it go in a sentence? What kinds of affixes does it take? If you can identify a class of words that have attributes unique to them, then you have a part of speech.
For example negation is usually considered a “part of speech” by linguists (more technically it’s a functional projection), and why not? Negatives display a constellation of behaviours that no other parts of speech has. Negatives don’t take inflectional affixes, never conjugate, can’t serve as subjects, participate in contractions of certain kinds, can be moved to the front of a sentencs along with a verb (You don’t know => Don’t you know?), and so on. Why not consider “negative” to be a part of speech on it’s own, instead of getting shoved into adjective (or adverb, depending on which book you read).
The only reason to stick with the classic categories is because it’s tradition or because it’s a small number, and neither of these is a good reason. Simpler isn’t better if it’s wrong. Just to be clear: I’m not suggesting that the book should have included the term “morphosyntactic category” or provided technical analysis of English affixation. The term “part of speech” is just fine. How to Use Parts of Speech is only supposed to be for Grade 3 after all. I think it would have been enough to emphasize that (a) words can be different things under different circumstances, and (b) sometimes adding a suffix changes the part of speech. And it would be easy to give examples of these.
Books intended for adults reprint the same useless definitions, and they are not immune from this criticism. There’s no reason not to explain some morphosyntax (under another, less technical sounding name) to an adult, because that is what really determines word categories anyway. Why shy away from useful descriptions of English? It’s time grammarians started treating language with respect and learned a thing or two about it.