THEME 6. Parts of Speech: Part I
Plan
1. Parts of speech in the English language. A short history.
2. The problem of classifying words into word classes. An outline of the main difficulties in classifying words:
a) the viewpoints of H. Sweet, O. Jespersen, Ch. Fries;
b) the structural linguistics or parts of speech (H. Gleason, R. Robins, B. Strang).
The lexicon of English like the wordstock of other European languages can be (sub)divided into lexico-grammatical classes traditionally called ‘parts of speech’. Nobody is going to question their existence though linguists treat them differently.
In the course of the linguistic history many attempts have been undertaken to build a classification of parts of speech based on one principle only but the attempts didn’t justify themselves. Opinions as to the number of parts of speech and their nomenclature are different.
Henry Sweet was the first to divide the parts of speech into two main groups: declinables / indeclinables (изменяемые / неизменяемые) that is morphological properties were the basic principle of classification. Inside the group of the declinables he stuck to the traditional subdivision: nouns, adjectives, verbs. The group of declinables includes adverbs, prepositions, conjunctions and interjections.
Along with this classification, H. Sweet offers a grouping based on the syntactical functioning of some classes of words. So the group of noun words includes not only noun-pronouns but also noun-numerals, the infinitive and the gerund. The group of adjectives includes adjective-pronouns, adjective-numerals and the participle. The verb group includes finite forms and verbals. Thus, the verbals such as the infinitive and the gerund can be referred to as noun words due to their syntactic function and regarded as verbs due to their morphology.
H. Sweet was aware of incongruity of morphological and syntactic properties of parts of speech (‘incongruos’ – not in harmony or agreement with, inappropriate, out of place). He was aware of the fact that his attempt at a harmonious classification brought about some problems that would hardly be solved at that time or that his attempt had fallen short of his expectations.
As to the class of indeclinables, it embraces heterogeneous units: adverbs which are parts of the sentence and prepositions, conjunctions, interjections which are not parts of the sentence (prepositions functioning inside predicative units, conjunctions used to join independent / dependent clauses.
O. Jespersen, the distinguished Danish linguist, was fully aware of the difficulty that lay ahead, that is of taking into account both the principles, form and function, disregarding the lexical meaning.
O. Jespersen offered a binary system. Along with the description of the traditional parts of speech which are usually analysed both from the morphological view and the lexical view, he analysed their function in word combinations and sentences.
One and the same word could be primary (the nucleus of a phrase, or the subject of a sentence), secondary (it might modify the primary word) or tertiary (subordinated to the secondary word):
3 2 1
a furiously barking dog
The noun ‘dog’ is the primary word. The adjective ‘barking’ is the secondary word, it modifies the noun ‘dog’. The adverb ‘furiously’ is a tertiary word, it modifies the adjective, specifies the quality of the adjective.
This is called the theory of three ranks. It allows to see through the hierarchy of syntactic relations hidden behind the linear character of the speech string.
On considering the subordinated word combination ‘a furiously barking dog’ O. Jespersen stressed the point that the idea of three ranks does not correspond to the division of the lexicon into parts of speech: primary words are not necessarily nouns, secondary words are not always adjectives, tertiary words are not necessarily adverbs. According to O. Jespersen, phrases can function as primary words:
e.g. Sunday afternoon was fine. I spent Sunday afternoon at home.
Despite the fact that O. Jespersen does not identify the notions of ‘rank’ and ‘part of speech’, there is some correspondence between the morphological classes of words and the ranks the linguist has singled out. For example, O. Jespersen believes that finite forms of the verbs can function as secondary but they cannot function as primary or tertiary.
O. Jespersen does not limit his concept of three ranks to the word level only; he extends it to the level of word combinations. In fact, the theory of three ranks is the theory of positional relations.
The drawback of the theory is lack of discriminating the criteria of identifying every rank. If there are none, how will one know whether this or that word belongs to this or that rank? One can hardly agree with the thesis that specifying the 4th, the 5th and other ranks is irrelevant. It is not enough to state that ‘the dog barks furiously’ is equal to ‘a furiously barking dog’ as for the ties between the constituents. It is to be proved.
The merit of the theory is that it reflects the complex character of the real state of things, the hierarchy of the relations which the linear character of the speech conceals.
This idea permeates all modern syntactic theories. That is why this theory sill holds vital. It is the first attempt to put into words the hierarchy of language structure.
The theory by O. Jespersen takes an intermediate position between morphology and syntax. It is closer to syntax.
It is one of the first attempts to make up a classification built on one principle only – that of the position of word in units larger than a separate word. The morphological classification, syntactical functions and the three ranks overlap all the time-creating extra units of analysis.
There is another theory – the IC model – offered by Charles Carpenter Fries in his book The Structure of English. The linguist gives a critical revision of the classical analysis of the parts of the sentence. Fries writes that this kind of analysis is of no value for an effective practical command of English. In his opinion this classical analysis consists in ascribing the technical terms subject, predicate, indirect or direct object to certain parts of the sentence.
Charles Fries grouped morphemes into positional classes with the help of ‘environments’ suggested by Z. Harris. He chose three patterns of English sentences as ‘frames’ to fill the positions with the word under the test. If a word could fit into a position without any change of the structural meaning of the sentence, it was considered to belong to a certain form-class. The frames for the tests were:
1 2 3 4
Frame A. The concert was good (always).
1 2 3 4
Frame B. The clerk remembered the tax suddenly.
1 2 4
Frame C. The team went -------- there.
Thus, the words were divided into four form-classes: those of the noun (N), verb (V), adjective (A), adverb (D). Fries gave no names to these four classes of words except the numbers: class 1, class 2, class 3, class 4.
Pronouns were included into the corresponding positional classes as their substitutes. Words standing outside the ‘positions’ in the sentence are treated as function words of various syntactic values.
Class 1 (A) concert, coffee, taste, container, difference, etc.
(B) clerk, husband, supervisor, food, coffee, etc.
(C) team, husband, woman, etc.
Class 2 (A) was, seemed, became, etc.
(B) remembered, wanted, saw, suggested, etc.
(C) went, came, ran, lived, worked, etc.
Class 3 good, large, necessary, foreign, new, empty, etc.
Class 4 (A) there, here, always, then, sometimes, etc.
(B) clearly, sufficiently, especially, soon, etc.
(C) there, back, out, etc.
Frames. The Semantic Structure*:
(A) thing and its quality at a given time;
(B) actor-action-thing acted upon – characteristic of the action;
(C) actor-action – direction of the action.
The words unable to fill in the positions of the frames without destroying their structural meaning
were called ‘function’ words. They form limited groups totaling 154 units. According to Fries, they are grammatical signals of the structural meaning of the sentence:
e.g. There was a man in the room.
Please dance.
They can also be signals of the grammatical meaning of the words in the sentence:
e.g. Give me a sheet of paper.
Give me the (that) sheet of paper.
Function words identified by Ch. Fries are of three kinds:
1) words used as specifiers of notional words (determiners of nouns, modal verbs serving as specifiers of notional verbs, functional modifiers and intensifiers of adjectives and adverbs);
2) words used as specifiers of relations of notional words to one another (prepositional elements, prepositions and conjunctions);
3) words referring to the sentence as a whole (question words – what, how, etc.; words of affirmation and negation, sentence-introducers ‘there’ and some others).
Fifteen groups all in all.
H. Gleason, the author of Linguistics and English Grammar (Chicago 1965), makes a point of using syntactic criterion-distribution. Any word that occurs in the position The ______ is/are good is a noun. Neither the morphological nor semantic meaningful criteria are exact and reliable.
H. Gleason is not consistent in his interpretation of such words as table and handshake. He considers the latter an action. The structural linguists ignore the grammatical categorial meaning of the word.
On the whole the representatives of structural linguistics – unlike the classical grammar aimed at an exhaustive and more detailed description of linguistic facts – go too far by schematising possible structural types by making the closed list of words necessary to describe the signals of very important structural meanings.
Besides, the structuralists give a more detailed classification taking into account the heterogeneity of parts of speech and subclasses. Thus, incidentally Charles Hockett singles out model verbs that have no grammatical categories except that of the past tense that denote no action.
Some words are specialized, monofunctional, others are polyfunctional. Hence no uniformed criterion is possible. The only criterion in question is syntactical or functional.
The foundation of the classification of semantic criteria cannot lie at the basis of classification though to run, runner, running alike as regards their lexical meaning, but they belong to different word classes. Neither can it be based on the functional principle alone because many English words are multifunctional (polyfunctional) and can be used in occasional context.
As regards its morphology some of word classes do not change. Thus, every word class denotes a notion within the confines of substantive-logical subdivision. But the English structure due to its general characteristics possesses high grammatical organization which means that within its confines every word class includes words that do not denote or refer to the object and categories of the reality.
For example, some abstract nouns like ‘whiteness’ denote quality but the way they express its qualitative characteristics as a notion is like any other notions expressed by concrete nouns.
Due to its abstract character and generalization, the grammatical meaning is difficult to grasp. That is what makes O. Jespersen and Ch. Fries speak of intuition when identifying this or that part of speech. What they call intuition is an abstract and general grammatical meaning which lies at the basis of classifying words into parts of speech. Undoubtedly, parts of speech are contensive ideational categories, i.e. word-classes which have the same grammatical meaning.
The form and function come secondary. The word has this or that paradigm because the speaker is aware of its linguistic status, not vice versa. You regard the word as a noun not because it has some formal criteria. It has acquired the inflexion because you are to regard it as a noun. Thus, the word is used in this or that function because you have classified it as a definite part of speech.
This approach has some drawbacks. Instead of giving a definition to each form class, Charles Fries confines himself to a number of characteristics. The latter are heterogeneous, sometimes based on the opposition to the other word classes. In fact, it was not his target. His task was limited to a description of the material making an inventory. Thus, according to Ch. Fries, form class is a functioning model. All the words that perform one and the same function in this model belong to one and the same class of words. But that does not mean that Ch. Fries recommends to use this word in a sentence to pigeonhole it. Besides syntactic characteristics there is a set of formal contrasts which are prompted to the native speakers’ intuition. Compare the pairs of words: good – goodness, boy – boyish, friend – befriend, arrival – arrive.
Though the words are marked syntactically, he takes into consideration the morphological characteristics. The third distinctive feature is a number of substitutes for every class of words.