Speaking of Words
By MICHAEL FERBER
Many of us oldsters remember being taught in school that there are eight parts of speech, but I find that, when I try to remember what I was taught, I usually come up with nine. When I look at various websites that agree that there are eight, I see that their lists are not identical.
I have learned since my school days, moreover, that linguists have added several more parts, and they are just for English and similar languages; some languages lack one or two that we have, and vice versa, and for some other languages the whole idea of parts of speech may be irrelevant. In this column, then, I’ll try to sort things out as well as I can.
The phrase “part of speech,” however many parts there are, is a little strange. It goes back to a Greek phrase, then picked up by Latin, and then adopted by Old English, sometimes just as “parts” or Latin pars. To “parse” a passage originally meant to identify what part of speech each word belongs to. But a better term than “part of speech” would be “kind of word” or “word class”; these are preferred by linguists.
Both Greek grammarians and Roman grammarians came up with eight parts. The Greeks listed noun, pronoun, verb, adverb, article, participle, preposition, and conjunction. The Romans listed all of these except article, because Latin lacked articles. (If you see the word domus in a Latin text, for example, it takes some work to decide if it means “house” or “a house” or “the house.”)
So to make eight of them the Romans tossed in interjections, which, while the Greeks no doubt used them all the time (“By Zeus!”), Greek grammarians must have thought of them as outcasts, which they are, because they have no syntactic connections in a sentence. Both Greeks and Romans left out adjective because they considered it an “additional” noun; adjectivum in Latin means “additional.” Today we distinguish noun and adjective but we usually drop participle, since it seems to partake of noun and verb and adjective.
I have been looking over a catalog of English grammar books published between 1711 and 1851, in both Britain and the United States. They all discuss parts of speech, but they propose numbers varying from two to thirty-three! Horne Tooke in 1786 said there were only two—nouns and verbs—and all other traditional parts were somehow derived from those two. Quite a few grammarians thought there were four parts, but they differed wildly in what they were; one book from 1711 listed names, qualities, affirmations, and manner words. Eight and nine were the most frequent numbers, but some authors left out adjectives, others put in participles, one called adjectives “adnouns,” and so on. One book listed twelve parts, including relatives and auxiliaries.
The wackiest book, published by James Brown in 1842, listed thirty-three “definers,” including “primary definer,” “secondary definer,” “vital definer,” “superfluous definer,” and “adversative conjunctive definer.” I’m glad I was not a pupil in Mr. Brown’s grammar class, for he was evidently a man of many parts.
A part I didn’t learn in school is “determiner,” which has been in use among linguists since the 1930s. This class of words includes the definite article (“the”), indefinite articles (“a, an”), demonstratives (“this, that,” etc.), possessives (“my, his, Mary’s” etc.), and quantifiers (“many, several,” etc.). They are like adjectives in that they go with nouns or noun phrases, but in English they precede adjectives in word order. We say “the little boy,” not “little the boy.” The determiners themselves also have an order of precedence. Articles, demonstratives, and possessives generally precede quantifiers: “the many little boys,” or “her many little boys” Some quantifiers may come first, however, such as “all” in “all her many little boys.”
Some linguists distinguish “numerals” from “quantifiers,” but the two usually behave alike: “her five little boys,” or “her many little boys.”
Another candidate is “particle.” It is a well-established category in Ancient Greek grammar, and particles follow a strict “particle-second” rule in the otherwise fairly free word order, for they must come second in a clause, and if there are two or three of them they cluster together in that slot. They are hard to translate without over-translating them. Some are like conjunctions (“and, but”), some emphasize (“indeed”), some de-emphasize (“I suppose”), some work in parallel clauses (“on the one hand, on the other”), and some are found only in the subjunctive or optative moods. In English we might define some of them as adverbs of a sort, but it’s not always clear what they “modify” or attach to.
If you ask someone a question, for instance, they may answer it with a preliminary “well.” In recent years the habit has taken hold to begin an answer, or almost any speech, with “so.” “Now” is another sentence-starter, though less frequent than it used to be. We might call these little words particles. Sometimes they shade off toward interjections (such as “Wow” or “Aha”), but they are usually low-key.
Prepositions are on nearly every list of parts of speech. But some languages have “postpositions” that follow the noun phrase. In Japanese, for example, Tokyo ni means “in Tokyo.” Does English have postpositions? It seems it does. We say “a long time ago,” “his arguments notwithstanding,” and “all joking aside.” It is a small class, true, but then it is larger than the class of articles.
English and many other languages have words called “complementizers,” of which the main one in English is that, as in “She said that she would go.” The clause “that she would go” is the “complement clause,” used here as a noun phrase, the object of “said.” It probably arose from the demonstrative pronoun that, as in “She said that: (namely) I will go.”
I don’t know any polysynthetic languages, such as Inuktitut, Cherokee, or many other indigenous American languages, but I gather that what distinguishes them is that they construct long words of many parts, but that the parts generally cannot stand alone as separate words. In English we have some words like that, such as “unhappiness,” where the prefix “un-“ and the suffix “-ness” are not words by themselves. In polysynthetic languages, typical words may have a dozen affixes, and they may amount to whole sentences. (Hence the idea that “Eskimoes” have a hundred words for “snow” should come as no surprise; they have a hundred words for everything .) To sort out such words into various kinds would seem irrelevant, though I may be missing something. They are themselves composed of so many parts that they might be called partnerships of speech.
Modern linguists seem not to have made a count of word types, perhaps because several new types have been defined in recent years, and they don’t know what might turn up in the thousands of languages they have not yet fully studied. In English, at least, we can be sure that there are more than eight, though not, Mr. Brown to the contrary, thirty-three, thank heaven.
I am happy to hear from readers with questions or comments: mferber@unh.edu.
Michael Ferber moved to New Hampshire in 1987 to join the English Department at UNH, from which he is now retired. Before that he earned his BA in Ancient Greek at Swarthmore College and his doctorate in English at Harvard, taught at Yale, and served on the staff of the Coalition for a New Foreign Policy in Washington, DC. In 1968 he stood trial in Federal Court in Boston for conspiracy to violate the draft law, with the pediatrician Benjamin Spock and three other men. He has published many books and articles on literature, and has a deep interest in linguistics. He is married to Susan Arnold; they have a daughter in San Francisco.