The history of grammar theory
The history of grammar theory
There
does not appear to exist a generally accepted periodization of the history of
English grammars, so we shall roughly divide it into two periods of unequal
length, according to the general aims or objectives of the grammars appearing
within these periods. The first is the age of prescientific grammar beginning
with the end of the 16th century and lasting till about 1900. It includes two
types of grammars which succeeded each other.
The
first type of grammars in the history of English grammars are the early
prenormative grammars of English, beginning with William Bullokar's Bref Grammar
for English (1585).
By
the middle of the 18th century, when many of the grammatical phenomena of
English had been described, the early English grammars gave way (to a new kind
of grammar, a prescriptive (normative) grammar, which stated strict rules of
grammatical usage, condemning those constructions and forms which it considered
to be wrong or "improper", and setting up a certain standard of correctness
to be implicitly followed by learners of English. The grammars of the second
type still constitute the only kind of grammar in use in the practical teaching
of English.
By
the end of the 19th century, when the prescriptive grammar had reached its
highest level of development, when the system of grammar known in modern
linguistics as traditional had been established, the appearance of new grammar,
the scientific grammar, became possible.
In
contrast with prescriptive grammars, classical scientific grammar (the third
type of grammar), according to the explicitly stated views of its founders, was
both descriptive and explanatory. As Sweet's grammar appeared in the last
decade of the 19th century, we may take 1900 as the dividing line between the
two periods and the beginning of the second period, the age of the scientific
grammars of English (including three new types of grammars). During the first
half of the present century an intensive development of this grammar has taken
place. Classical scientific grammar has accepted the traditional grammatical
system of prescriptive grammar, but, as has been mentioned, now we witness the
final stage of its existence, for since the 1950's no new grammars of the
scholarly traditional type seem to have appeared. The new types of English grammars,
which appeared since the fifties are the fourth type of grammar - structural or
descriptive, which, in its turn, is becoming obsolete and is being supplanted
by the fifth type of grammar - the transformational generative grammar. The
linguistic theory represented by the last mentioned type of grammar is
considered by many modern linguists to be the most fruitful approach to the
description and explanation of the grammatical system of English, especially in
the field of syntax.
ENGLISH
GRAMMARS BEFORE 1900
(THE
FIRST PERIOD)
Early
(Prenormative) Grammars. Until the 17th century the term
"grammar" in English was applied only to the study of Latin. This
usage was a result of the fact that Latin grammar was the only grammar learned
in schools ("grammar" schools) and that until the end of the 16th
century there were no grammars of English. One of the earliest and most popular
Latin grammars written in English, by William Lily, was published in the
first half of the 16th century and went through many editions. This work
was very important for English grammar as it set a standard for the
arrangement of material and thus Latin paradigms with their English
equivalents easily suggested the possibility of presenting English forms in a
similar way, using the same terminology as in Latin grammar. A striking example
of the two approaches to the description of English is the divergence of views
on the problem of English case system. Though Bullokar mentioned 5 cases and in
a grammar published in 1749 and reprinted as late as 1819 (Th. Dilworth, A
New Guide to the English Tongue) the number of cases both of nouns and
adjectives is said to be 6 (as it is in Lily's grammar), in two grammars which
appeared during the first half of the 17th century, Ben Jonson's and Ch. Butler's
English grammars, the number of cases is two, while in J. Wallis's Grammatica
Linguae Anglicanae (1653), which was written in Latin, in spite of the
author's intention to break entirely with Latin tradition, the category of case
is said to be non-existent and the 's form is defined as a possessive
adjective. This view was supported by an early 18th century grammar, attributed
to John Brightland. The authors of the second half of the 18th century seemed
to prefer the two-case system, which was revived at the end of the 19th century
in scientific grammar. In 19th century school grammars a three-case system
prevailed.
The
treatment of the problem of case shows that even in the early period of the
development of English grammars the views of grammarians were widely divergent,
a fact which may be explained by two different approaches toward the
description of English grammatical structure. The grammarians who desired to
break with Latin grammatical tradition were not always consistent and still
followed the Latin pattern in some of the chapters of their grammars.
By
the middle of the 18th century the main results of the description of the
English grammatical system, as it was presented in the prenormative grammars,
were as follows:
Morphology. The
Latin classification of the parts of speech, which included eight word-classes,
differed from the system adopted by modern grammars in that the substantives
and adjectives were grouped together as two kinds of nouns, while the
participle was presented as a separate part of speech. In the earliest English
grammars, where this system was reproduced, the parts of speech were also
divided dichotomically into declinable and indeclinable parts of speech, just
as in Lily's grammar (W. Bullokar), or words with number and words without
number (Ben Jonson), or words with number and case and words without number and
case (Ch. Butler). The first of these groups, declinable words, with number and
case, included nouns, pronouns, verbs and participles, the second — indeclinables
— adverbs, prepositions, conjunctions and interjections. Ben Jonson increased
the number of parts of speech in his classification, introducing the article as
the ninth part of speech.
Later,
at the beginning of the 18th century, another scheme of classification appeared
in J. Brightland's grammar. This author reduced the number of parts of speech
to four, rejecting the traditional terminology as well. The four parts of
speech were: names (i. e. nouns), qualities (i. e. adjectives), affirmations
(i. e. verbs) and particles, which included the four so-called indeclinable
parts of speech. In this scheme the adjective was classed as a separate part of
speech, owing to the influence of the philosophical or universal (logical)
grammars of the age, which in their attempts to discover the universal laws of
the structure of languages pointed out the difference between the syntactic
functions of the two varieties of "nouns".
Syntax. In
Brightland's grammar we likewise find an important innovation in the study of
English syntax — the introduction of the notion "sentence" into
syntax. Latin grammar was not concerned with the structure of the sentence, the
principal object of the syntax of modern grammar. Though definitions of the
sentence, mostly logical (pointing to its function as an expression of a
complete thought, a judgment or proposition), already existed in the ancient
period, grammarians understood syntax etymologically as a study of the
arrangement, i. e. the connection of words. Thus, Lily briefly stated the three
concords of Latin: of the nominative and the verb, of the substantive and the
adjective and of the relative pronoun and its antecedent.
Ben
Jonson applied this analysis to English syntax and devoted a large part of his
grammar to the description of the "syntax" of a noun with a noun, of
a noun with an adjective, with an article, with a verb, etc. As the rules of
concord and government were few in English, the author paid much attention to a
specifically English means of connection of words — word order. The sentence
was mentioned only in the chapter on punctuation, which was based on the theory
of rhetoric (i. e. stylistics) created by ancient authors. The principal unit
of rhetoric was the period, which, like the sentence, was defined as an expression
of a complete thought. The expression of a complete thought in rhetoric was not
confined to the bounds of a single sentence. It could be expressed by a group
of closely connected sentences, but early English grammarians identified the
period with the sentence, so that the marks of punctuation (named after the
parts of the period which they divided, such as the comma, the least part of
the period, the colon, a member of the period, and the period itself, which
denoted the mark of punctuation pointing to its completion) were at the same
time intended to divide sentences and their parts, which as yet had no special
names. As some colons were rather long, another mark of punctuation was added,
the semicolon (a half-member), which was so named by analogy with the already
existing terms.
It
was only in Brightland's grammar that the concept of the sentence was included
in syntax proper. In Brightland's grammar sentences are divided dichotomically
into simple and compound. The simple sentence is defined as containing one
affirmation (verb) and one name, signifying the subject of the affirmation
expressed or understood. The compound sentence consists of two or more simple
sentences.
Alongside
the logical terms introduced into syntax, the term "object" (deriving
from medieval scholastic philosophy) was added to denote the third
"principal" part of the sentence. But morphological terms (such as
the nominative case or word, the noun, etc.) continued to be used in the
description of the parts of the sentence.
The
concept of the compound sentence, which, judging by Brightland’s examples,
denoted both complex and compound sentences, according to a classification
introduced, much later, was also due to logic, where propositions or judgments
were divided into simple and compound. The second part of his syntax deals with
the "construction of words" (as it does in older grammars).
Prescriptive
Grammars. The age of prescriptive grammar begins in the second half of
the 18th century. The most influential grammar of the period was R.Lowth's Short
Introduction to English Grammar, first published in 1762. The aim of
prescriptive grammars was to reduce the English language to rules and to set up
a standard of correct usage. The authors of prescriptive grammars believed
that, their task was not only to prescribe, to provide rules for distinguishing
what is right from what is wrong, but also to proscribe expressions which they
considered to be wrong. In the second half of the 18th century it was the
grammarians who took upon themselves the responsibility of dictating the laws
of grammar and usage. These grammarians settled most disputed points of usage
by appealing to reason, to the laws of thought or logic, which were considered
to be universal and to be reflected in the Universal, that is, Logical or Philosophical
Grammar. But as O. Jespersen correctly observes, "In many cases what gives
itself out as logic, is not logic at all, but Latin grammar disguised."
There is then nothing whatever in logic which obliges the predicative to stand
in the same case as the subject, that is, in the nominative.
From
the point of view of modern grammatical theory some changes which had
taken place in the description of the morphological system did not contribute
to its improvement. In spite of the authority of Lowth and Murray, who had
retained the scheme of nine parts of speech, the succeeding grammarians reverted
to the system of eight parts of speech. They chose to class the article with
the adjective, as it had been done in earlier grammars (e. g. in Wallis's
grammar), rather than increase the number of the parts of speech beyond
eight. In this case it was the older tradition which prevailed. This classification
remains the most popular one in prescriptive and classical scientific
grammars of the modern period. Another morphological problem which in the
earliest grammars had caused considerable disagreement among
grammarians and admitted of various solutions came to be settled to the satisfaction
of the authors of prescriptive grammars. This was a problem which continues to
be subject of dispute to this day — the number of cases in English. Lowth
adopted a two-case system for nouns and a three-case system for pronouns,
and the term "possessive case", which is extremely popular
now. The paradigm of the declension of personal pronouns included the
nominative case, the possessive pronoun as a form of the possessive case and
the objective case, the latter term also having been most likely introduced by
Lowth. After a great deal of vacillation, Murray, in the later editions of his
grammar, decided to adopt the three-case system for nouns. The three-case
system was adopted almost unanimously by all prescriptive grammars of the
19th century and later, until in the 1920's Nesfield substituted for it
a five-case system.
The
syntactic study of the simple sentence did not advance greatly till the middle
of the century. By the time Lowth’s grammar appeared the concept of the
principal parts of the sentence had been already elaborated to the number of
three. The terminology was rather unsettled. Lowth distinguished an agent, an
attribute (i. e. the predicate) and an object. The definitions of the first
and second parts of the sentence corresponded to the definitions of the
logical subject and predicate. The object was defined as the thing affected
by the action of the verb. There was no advance in the conception of the secondary
parts of the sentence. Besides the principal parts, Lowth mentioned adjuncts
without further differentiation on the syntactic level.
The
theory of the compound sentence, dating from the beginning of the 18th century,
was during this period at an absolute standstill. The definitions in the grammars
of the first half of the century were practically the same as in J.
Brightland's grammar, where they first occurred.
The
principal feature of a compound sentence, as it was understood at that time, is
that it comprises more than one subject or nominative word and verb, expressed
or understood. Sentences were therefore classed as compound, when a punctuation
unit contained two or more subject-predicate groups, connected by subordinating
or coordinating conjunctions, or when there was a single subject-predicate
group with coordinate members.
The
classification of conjunctions corresponded to the classification of compound
propositions or judgments in logic. All conjunctions were divided according to
their meaning, but without regard to their syntactic nature, into copulatives
and disjunctives. The notions of subordination and coordination were still
unknown.
The
second part of syntax, which treated the "construction of words", was
more developed. In Lowth's grammar the word "phrase" came to be used
as a grammatical term, defined as follows: "A Phrase is two or more words
rightly put together to make a part of a Sentence and sometimes making a whole
Sentence." The concept of the phrase occupies an important place in
Murray's grammar and the grammars of his successors, who described the kinds of
phrases and the relations between the words making up a phrase.
Though
the grammatical system created by the grammarians by the middle of the 19th
century (especially in syntax) still differed from that known in traditional
grammar of the present period, a great number of prescriptions and rules
formulated and fixed by the authority of the grammarians remain in grammars
of the modern period. One important series of prescriptions that now forms
part of all grammars had its origin in this period, namely the rules for the
formation of the Future Tense. The rule was first stated by J. Wallis, and
since that time it has been repeated by all grammarians, at first in its
archaic form, as formulated by Wallis.
The
rule that two negatives destroy one another or are equivalent to an
affirmative, was first stated in J. Greenwood's Royal English Grammar in
the first half of the 18th century, the influence of Lowth's grammar helped to
fix it.
It
was in the second half of the 19th century that the development of the
grammatical scheme of the prescriptive grammar was completed. The grammarians
arrived at a system now familiar, because it has since been adopted by a long
succession of grammarians of the 19th and 20th centuries. The best prescriptive
grammars of the period, like C. P. Mason's English Grammar (London,
1858) and A. Bain's Higher English Grammar (London, 1863), paved the way
for the first scientific grammar of English.
The
description of the morphological system in grammars of the second half of the
19th century changed very little as compared with that of grammars of the first
half of the century, but the explanation of grammatical forms became more
detailed, expressing of a deeper understanding of the nature of the phenomena
discussed. Some important changes, however, took place in the description of
the syntactic system, though the definition of the sentence remained logical,
as a combination of words expressing a complete thought. But the concept of the
parts of the sentence differs greatly from that of the grammars of the first
half of the 19th century. The changes and innovations concerned both the
principal and the secondary parts of the sentence. The number of the principal
parts of the sentence was reduced to two - the subject and the predicate, which
retained their logical definitions. In this period the grammarians make an
attemps to differentiate logical and grammatical subjects and predicates. The
former are represented by single words, the latter include word groups with
subjects and predicates as head words. A little later subjects and predicates
expressed by one word came to be distinguished simple or essential subjects and
predicates, and those expressed by a word group as complete subjects and predicates.
The
objects came to be viewed as a secondary or dependent (subordinate) part of the
sentence in the light of the newly developed theory of subordination and
coordination of sentence elements and the introduction into grammar of the
content aspect of syntactic relations, such as predicative, attributive,
objective and/or adverbial relations.
Thus
the notion of the attribute came to be applied, instead of the predicate to a
relation expressed by a secondary part of the sentence and adjuncts were subdivided
into attributive (also attributival or adnominal) and adverbial adjuncts,
which was the first differentiation of the secondary parts of the sentence on
a syntactic level.
The
objects were classified according to their meaning and form as direct, indirect
and prepositional. This classification, though inconsistent logically, is
accepted by many grammarians of the modern period. Objects and subjects as well
were further classified as compound (i. e. coordinate), complex (expressed by
infinitive groups or subordinate clauses), etc.
Besides
the object and two kinds of adjuncts, some new notions and terms developed,
either as synonyms for the already defined syntactic units or used in a
slightly different meaning to describe some new syntactic units, which contributed
to a more detailed sentence analysis.
Syntactic
processes operate to derive a more complicated structure from a simpler one.
The
notion of completion of the meaning of transitive or copulative verbs, defined
as verbs of incomplete predication, may be understood as a designation of a
syntactic process.
A
very important innovation in the concept of the compound sentence was its
subdivision into the compound sentence proper, with coordinated component
parts, and the complex sentence, characterized by subordination of clauses. In
this way the dichotomic classification of sentences into simple and compound
was changed into a tricholomic division, according to which sentences are
divided into simple, compound and complex. This theory has since been accepted
with very few exceptions by prescriptive, classical scientific and some
structural as well as transformational grammars. The recognition and
differentiation of the two principal syntactic modes of joining
subject-predicate units, subordination and coordination (the former expressing
syntactic dependence and the latter — equality of syntactic rank), was a great
advance in the development of grammatical theory. Of great interest also is the
elaboration of the concept of a clause as a syntactic unit containing a noun
and a finite verb and forming part of a complex or compound sentence. Clauses
are classified as independent and dependent or coordinate and subordinate. The
latter were also classified morphologically as noun, adjective and adverb
clauses, because grammarians considered clauses to be of the nature of a word,
and not of a part of the sentence. These three kinds of clauses were further
subdivided according to their syntactic functions in the sentence.
The
concept of the compound sentence in the new sense, as containing independent
clauses or sentences, did not, it seems, satisfy those grammarians who had
gained a deeper insight into the nature of the grammatical phenomena
described in their grammars. They give examples illustrating the
possibility of isolating the parts of the compound sentences, of pronouncing
each part of such a sentence by itself, without any change of meaning or
intonation and they stress the complete independence of each part.
The
concept of the phrase has been retained in the grammars of the second half of
the 19th century, though not all grammarians use this term, describing the syntax
of the parts of speech instead. The phrase is differentiated from the clause,
as containing no finite verb.
The
Rise of Classical Scientific Grammar. By the end of the 19th century,
after the description of the grammatical system, especially that of syntax had
been completed, prescriptive grammar had reached the peak of its development. A
need was fell, therefore, for a grammar of a higher type, which could give a
scientific explanation of the grammatical phenomena. The appearance of H.
Sweet's New English Grammar, Logical and Historical (1891) met this
demand. As Sweet wrote in his Preface: "This work is intended to supply
the want of a scientific English grammar." The difference in purpose
between scientific and prescriptive grammar is stated in the following terms:
"As my exposition claims to be scientific, I confine myself to the
statement and explanation of facts, without attempting to settle the relative
correctness of divergent usages. If an ‘ungrammatical’ expression such as it
is me is in general use among educated people, I accept it as such, simply
adding that it is avoided in the literary language." This was a new
approach, in keeping with the Doctrine of General Usage which had been first
formulated by an 18th-century grammarian, a contemporary of Lowth's, J.
Priestley, in his Rudiments of English Grammar. But Priestley's views
had been rejected, as we have seen, in favour of the Doctrine of Rules or
Correctness. Sweet clearly stales the new viewpoint: "...whatever is in
general use in language is for that reason grammatically correct."
Scientific grammar was understood by its authors to be a combination of both
descriptive and explanatory grammar. The same views on the purpose and methods
of scientific grammar were held by 20th-century linguists.
ENGLISH
GRAMMARS IN THE 20th CENTURY
(THE
SECOND PERIOD)
The
modern period may be divided into two chronologically unequal parts, the first
from the beginning of the 20lh century till the 1940's, when there were only
two types of grammars in use —the prescriptive and the classical scientific,
the second from the 1940's, during which time structural grammar, and then
transformational have been added. As has been pointed out, structural grammar
tended to supplant the older scientific grammar, which we call classical in
order to distinguish it from the new theoretical grammars of English.
There
is a borrowing of some of the concepts of prescriptive and classical scientific
grammars by the authors of both structural and transformational grammars,
especially in the field of syntax, which proves that structural grammar has not
quite succeeded in breaking with traditional grammar to the degree that is
proclaimed by the authors of these grammars, while transformational grammar, as
professed by its exponents, is closer to traditional grammar, than
descriptivism.
Prescriptive
Grammars in the Modern Period. Among the 20th-century
prescriptive grammars which are of some interest, J. C. Nesfield’s grammar
should be mentioned. Although published at the end of the 19th century
(1898), it exerted a certain influence on prescriptive and even scientific
grammars of the 20th century, comparable to the influence of Murray's grammar
upon 19th-century grammars. The editions which preceded the revision
continued the tradition of 19th century grammar: morphology
was treated as it had been in the first half of the 19th century, syntax,
in the second half of that century. Of the various classifications of the
parts of the sentence current in the grammars of the second half of the
19th century the author chose a system, according to which the sentence has
four distinct parts: (1) the Subject; (2) Adjuncts to the Subject
(Attributive Adjuncts, sometimes called the Enlargement of the Subject); (3)
the Predicate; and (4) Adjuncts of the Predicate (Adverbial Adjuncts); the object
and the complement (i. e. the predicative) with their qualifying words,
however, are not treated as distinct parts of the sentence. They
are classed together with the finite verb as part of the predicate. Although
grammars as a rule do not consider the object to be the third principal part of
the sentence, indirectly this point of view persists since the middle of the
19th century and underlies many methods of analysis. In Nesfield's
scheme, though the object is not given the status of a part of the sentence,
it is considered to be of equal importance with the finite verb. In
diagramming sentences, grammarians place the subject, predicate, objects and
complements on the same syntactic level, on a horizontal line in the
diagram, while modifiers of all sorts are placed below the line.
Revision
brought about certain changes in Nesfield's grammatical system. The
number of cases of the noun was increased to five (through the addition of
the vocative and the dative), while classical scientific grammars, for
instance, those of Sweet and Jespersen, favoured the two-case system. Another
change occurred in the structural classification of sentences. Two new , terms,
"double" and "multiple" sentences, were substituted for
the term "compound" sentence, the term "double"
denoting the coordination of two and "multiple" of more than two
sentences. This innovation —a quantitative classification of independent
sentences contained within a punctuation unit, is significant as symptomatic
of the weakness of the concept of the "compound" sentence,
intuitively felt by the members of the Joint Committee and those who followed
their recommendation. According to the concept of the "compound" sentence,
the combination of two or more syntactically independent, though semantically connected
sentences, was analysed as a single sentence. The new terms, which were
probably intended to improve the theory, became very popular in prescriptive grammar
and, as we shall see, influenced some scientific grammars.
Classical
Scientific English Grammar in the Modern Period. The
founders of this type of grammar in the period of its intensive development
either specialize in syntax or deal with the problem of both morphology and
syntax.
Among
the authors who specialize in syntax are L. G. Kimball, C. T. Onions and H. K. Stokoc.
Both Kimball's Structure of the English Sentence (New York, 1900) and
Onions' Advanced English Syntax. (London, 1904), which appeared at the
beginning of the period, discuss the problems of the structure of English on
the traditional plane, though in Onion's book there is a striking anticipation
of the sentence patterns of descriptive linguistics. Kimball's grammar shows
the influence of logical grammars of the type current in 19th-century German
linguistics, K. F. Backer's grammar for example. The third book, H.R.Stokoe's Understanding
of Syntax, which appeared in 1937, was also largely influenced by the views
of prescriptive grammarians like Nesfield. Two of these authors are not
satisfied with the traditional concept of the compound sentence. Onions passed
it over in silence. Stokoe adopted the new nomenclature, describing double and
multiple sentences in his book. All these authors differ from prescriptive
grammarians in their non-legislative approach to the description of English
structure and deeper insights into the nature of the grammatical phenomena.
Scientific
grammar was the first to undermine the strictly structural concept of a clause
as of a syntactic unit containing a subject and a predicate, created by
prescriptive grammar. Beginning with Sweet's grammar, the authors of scientific
grammars have been developing the concepts of half-clauses, abridged clauses,
verbid clauses, etc., which practically destroy the original
concept of clause and lead to a tendency to analyse simple sentences as
complex or, to put it another way, demolish the structural distinction between
simple and complex sentences. Thus Poutsma treats substantive clauses,
adverbial clauses, infinitive clauses, gerund clauses and participle
clauses as units of the same kind, though the last three types of
"clauses" are not clauses, according to the original and, in our
opinion, more correct concept of clause as a syntactic unit.
From
a theoretical point of view, Kruisinga's grammar is one of the most
interesting of those scientific grammars which have retained the traditional
grammatical system. Kruisinga approaches the problem of the definition of the
sentence critically, refraining, however, from giving a definition of
his own, whereas most grammarians were content to repeat traditional logical
definitions. Kruisinga originated the theory of close and loose syntactic groups,
the difference between them being based on the distinction between
subordination and coordination. Closely connected with this theory is the
author's concept of the complex sentence. His classification is dichotomic:
only two sentence types are recognized — simple and compound sentences.
The traditional compound sentence is not considered to be a syntactic unit at
all; the material in question is treated in connection with double and multiple
loose syntactic groups.
Of
all the authors of scientific grammars of the classical type O. Jespersen is
the most original. His morphological system differs from the traditional in
that he lists only five parts of speech — substantives, adjectives, verbs,
pronouns (the latter include pronominal adverbs, and articles) and
"particles", in which he groups adverbs, prepositions, conjunctions
and interjections. Like Sweet, he proposes three principles of classification,
according to which everything must be kept in mind — meaning, form and
function, though in practice only one of these features is taken into consideration,
and that is primarily form (cf. the "particles") and, in a few cases,
the origin of a given form.
Jespersen’s
syntactic system is more original. He intends to reject the
traditional syntactic analysis, though some of the traditional terms still
occur in his works and develops the concept of ranks.
Structural
and Transformational Grammars. Structural grammarians begin
treating the problems of the structure of English with criticism of
traditional, or conventional grammar, lumping together prescriptive and
scholarly grammars because their methods of approach are said to be the same.
According to the point of view of structural linguists, both these types of
grammar belong to a "prescientific era".
Fries
believes that "the study of the usual 'formal' grammar has much the same
sort of value and usefulness as the study of the astronomy of Ptolemy, or of
the medical beliefs and practices of Galen, the great Greek physician".
The author insists that pupils should begin the study of grammar only after ridding
their minds of all previously acquired notions concerning language.
The
new approach — the application of some of the newly developed techniques, such
as distributional analysis and substitution — makes it possible for Fries to
dispense with the usual eight parts of speech and with the traditional terms.
He classifies words into four "form-classes", designated by numbers,
and fifteen groups of "function words", designated by letters. The
form-classes correspond roughly to what most grammarians call nouns and
pronouns, verbs, adjectives and adverbs, though Fries especially warns the reader
against the attempt to translate the statements which the latter finds in the
book into the old grammatical terms. The group of function words contains not
only prepositions and conjunctions, but also certain specific words that more
traditional grammarians would class as a particular kind of pronouns, adverbs
and verbs.
Further
descriptive works on grammar should be mentioned. In 1951 An Outline
of English Structure by G. L. Trager and H. L. Smith was published. This
book was much the fullest on phonology and morphology, but, as noted by H. A.
Gleason, hardly more than suggestive on syntax, though we shall see some traces
of its influence in another descriptive grammar (by J. Sledd).
Gleason seems to think that the two books (that of Fries and the Outline)
can be looked upon as supplementing each other and that in the midfifties it
looked as though the "new grammar" might emerge as a new eclectic
tradition, based on these two sources with certain elements salvaged from older
grammars (which was what really happened).
As
has been aptly observed by Hathaway, the syntax of modern English has undergone
a shrinkage at the hands of the structuralists. In Chomsky's estimation also
modern structural linguistics provides little insight into the processes of
formation and interpretation of sentences, and therefore it does not seem to
him surprising that there has been renewed interest in the formalization and
use of techniques and devices more characteristic of traditional than of
structuralist grammars.
The
method developed by N. Chomsky has now become widely known as Transformational
Generative Grammar. It was first, expounded in Chomsky's Syntactic
Structures (1957) and has been revised in the author's Aspects of the
Theory of Syntax (1905). According to this theory sentences have a surface
structure and a deep structure. Of these, the surface structure is the more
complicated, based on one or more underlying abstract simple structures. In
certain very simple sentences the difference between the surface structure and
the deep structure is minimal. Sentences of this kind, simple, active,
declarative, indicative, are designated as kernel sentences. They can be
adequately described by phrase or constituent structure methods, as consisting
of noun, and verb phrases (the so-called P-markers, the NP’s & VP's). According
to Syntactic Structure's, kernel sentences are produced by applying only
obligatory transformations to the phrase-structure strings (e.g. the
transformation of affix + verb into verb + affix in the present tense, hit
–s, etc). Non-kernel or derived sentences involve optional transformations
in addition, such as active to passive (the boy was hit by the man). But
later interpretations of the transformational theory have made less use of this
distinction, stressing rather the distinction between the underlying "deep
structure" of a sentence and its "surface structure" that it
exhibits after the transformations have been applied. Transformational
operations consist in rearrangement, addition, deletion and combination of linguistic
elements.
A
Transformational Grammar is organized in three basic parts. The first part—its
syntactic component (which includes a lexicon, i.e. a list of words — boy, hit,
ball, etc.) is described, as mentioned above, in terms of IC's or P-markers.
The syntactic component includes description both of deep and surface structure.
The second is the semantic component, which provides a semantic
interpretation of the deep structure. E. g. in sentences we enjoy
smoking and we oppose smoking the semantic component would indicate
that the former is a paraphrase of we smoke and we enjoy it, though the latter
is not a paraphrase of we smoke and we oppose it. The third, the
phonological component provides a phonetic interpretation of the surface
structure of the sentence.
Note
that "to generate sentences" according to this theory does not mean
"to produce sentences", but "to characterize", "to
enumerate", "to determine" the rules for forming all of the
infinite number of sentences, some of them never heard before.
Chomsky's
new theory is that language has a base which contains the elementary phrase
structures. In the new conception of Chomsky the kernel sentence loses all its
significance, for Chomsky is careful to stress that sentences are not derived
from other sentences (as has been sometimes loosely and inaccurately stated),
but rather from the structures underlying them. The phrase structures produce
sentences usually by way of transformations. . . Now it is clearer that
transformations are not intended to relate sentences to sentences (as was
stated at first by Z. S. Harris), but deep structure to surface structure and
that deep structure thus embodies a hypothesis set up for an adequate
description of a language.
Our
selections from transformational grammars of English represent the earlier
version of the transformational theory, even O. Thomas' Transformational
Grammar, the first popular survey published since the major revision.
Of
great interest for clarifying the theoretical and philosophical sources
of transformational generative grammar are the two books by Chomsky: Cartesian
Linguistics and Language and Mind.
It
is also an interesting fact that some linguists point at the danger of new
prescriptivism in generative transformational grammars, e. g. J. Nist maintains
that in their search for language universals (that is, categories underlying
the structures of all languages), a process reminiscent of the eighteenth
century authoritarians, the generative grammarians have already showed signs of
becoming prescriptive and prescriptive in their analysis of "permitted"
(i. e. grammatically correct) strings. This opinion is shared by B. Hathaway.
In
the process of the development of English grammatical theory, despite the great
divergence of the types, aims, objectives and approaches of English grammars, a
certain continuity may be observed in establishing and keeping up the English
grammatical tradition. The foundations of the English grammatical system were
laid already in the first part of the first, prescientific, period, in early
prenormative grammar, though its morphological system leaned heavily on that of
the Latin grammar and the incipient syntactic notions were dependent upon
rhetoric and logic. The most important type of grammar, in our opinion, is the
second, the prescriptive or normative grammar, which has the longest tradition,
as it arose in the mideighteenth century and still dominates class room
instruction. Its most significant contribution to English grammatical theory
was the syntactic system evolved in the midnineteenth century.
The
three types of scientific grammars of English discussed here have not quite
succeeded in creating any really independent or new grammatical notions and
systems. The interests of the scholars centered found the grammatical system of
prescriptive grammar. They either elaborated it further (in classical scientific
grammar) or refuted it, retaining at the same time some of its ideas (in
structural grammar) or acknowledged its merits as an implicit transformational
grammar and reformulated its ideas (in transformational grammar).
Both
modern schools of grammar show a marked tendency towards morphological
labelling of syntactic units, which may be viewed as a: revival of the grammatical
notions of the earliest grammars when the syntactic system was practically
non-existent.
|