The non-structural descriptive approach to language had its heyday between 1900 and 1930, when it

was replaced by structuralism. The father of American structuralism is generally called Leonard
Bloomfield, who in his book Language presented the new approach as follows: “The study of language
can be conducted... only so long as we pay no attention to the meaning of what is spoken” [2].

Descriptive linguistics developed from the necessity of studying half-known and unknown languages of
the Indian tribes.

At the beginning of the 20th century these languages were rapidly dying out under the conditions of
what is known as 'American culture', or 'American way of life', which had brought the Indian peoples
poverty, diseases and degradation. The study of these languages was undertaken from purely scientific
interests.

The Indian languages had no writing and, therefore, had no history. The comparative historical method
was of little use here, and the first step of work was to be keen observation and registration of linguistic
forms.

Furthermore, the American languages belong to a type that has little in common with the Indo-
European languages; they are 'agglomerating' languages, languages devoid of morphological forms of
separate words and of corresponding grammatical meanings. Descriptive linguists had therefore to give
up analysing sentences in terms of traditional parts of speech; it was by far more convenient to describe
linguistic forms according to their position and their co-occurrence in sentences.

The description of a language became more refined at the beginning of the 20th century due to the
development of the concept of 'phoneme'. The concept of phoneme was worked out by the Russian
linguists Baudouin de Courtenay and his student Kruszewski, and developed by the linguists of the
Prague School (Roman Jakobson, 1928; Trubetzkoy, 1939).

Franz Boas, linguist and anthropologist (1858-1942) is usually mentioned as the predecessor of
American Descriptive Linguistics. His basic ideas were later developed by Edward Sapir (1884-1939) and
Leonard Bloomfield (1887-1949). Bloomfield's main work Language was published in 1933. All linguists
of the USA at one time or other felt the influence of this book.

The American linguists began by criticising the Praguian oppositional method and claiming a more
objective — distributional — approach to phonemes. But it soon became clear that the facts established
were the same, and only the approach was different. The American Descriptive School (if we may speak
of a number of loosely interrelated approaches as a 'school') began with the works of Edward Sapir and
Leonard Bloomfield. American linguistics developed under the influence of these two prominent
scientists.

Sapir studied a great variety of languages (Indian and Malayo-Polynesian). His most known work is
Language. An Introductory to the Study of Speech (1921). Leonard Bloomfield is considered to be a more
rigid theorist. His book of the same title as Sapir's Language, is more systematic than Sapir's, and the
treatment of linguis¬tic problems is more modern. It is a complete methodology of language study,
approaching the language as if it were unknown to the linguist (student). The ideas laid down in this
book were later developed by Z. S. Harris, Ch. C. Fries and other contemporary linguistic students.

Followers of this approach sought to study the structure of a language as objectively as possible,
without reference to meaning and other languages. By other languages they, first and foremost, meant
Latin and Greek, the languages prescriptive and, to a lesser degree, descriptive grammarians modelled
their analysis on. English was regarded as a language having its specific structure, and the task of a
linguist was to reveal it by using scientific (i.e. formal) methods of analysis.

Meaning as a criterion was not reliable since, being unobservable, it could be interpreted differently by
different linguists. Therefore the linguist was to devise formal methods of analysis and replace meaning
by form; the linguist must be interested in what he observes, i.e. objective data. The structuralists based
their conclusions on the analysis of sentences that they had collected from native speakers of English,
giving priority to Spoken English.

To structuralists, language is a highly organized affair, where the smaller units are built into larger units,
which in turn are built into larger ones, until the largest unit is reached. Such building-blocks are
phonemes and morphemes. The structures that we build out of the ‘bricks’ are lexemes. Lexemes, in
their own turn, serve to build the largest unit, the sentence, i.e. the predicative unit.

Structural linguists ignored meaning not because they were not interested in it. Meaning was ignored on
the grounds that it was not observable and could not be described objectively by using formal methods.
The description of meaning had to wait until appropriate methods were devised. Such being the case,
they focused their attention on structural, i.e. grammatical, meaning.

How are structural meanings conveyed in English? Structural grammarians have pointed out four
devices used in English to indicate structural meaning: 1.word form; 2) function words; 3) word order; 4)
intonation and accent patterns (prosodic patterns).

To have a deeper understanding of modern grammar we must get acquainted with the main concepts of
Bloomfield's book.

1. Bloomfield understood language as a workable system of signals, that is linguistic forms by means of
which people Communicate: "...every language consists of a number of signals, linguistic forms" [2: 158].

2. Bloomfield's understanding of 'meaning' seemed to be very unusual at that time. Later his concept of
'meaning' was developed by Charles Fries but even now 'meaning' is one of the problems linguistics
seeks to solve. Bloomfield says: "...by uttering a linguistic form a speaker prompts the hearers to
respond to a situation; this situation and the responses to it, are the linguistic meaning of the form. We
assume that each linguistic form has a constant and definite meaning, different from the meaning of any
other linguistic form in the same language." [2: 158]. According to Bloomfield, "the meanings of speech-
forms could be scientifi¬cally defined only if all branches of science, including, especially, psychology
and physiology, were close to perfection. Until that time, phonology, and with it all the semantic phase
of language study, rests upon an assumption, the fundamental assumption of linguistics: we must
assume that in every speech community some utterances are alike in form and meaning." [2: 78]. And
then: "...every utterance contains some significant features that are not accounted for by the lexicon."
(p. 162). "...No matter how simple a form we utter and how we utter Mr... the utterance conveys a
grammatical meaning in addition to the lexical content..." [2: 168]. The quotations clarify two things: (1)
the meaning of an utterance can be found through the response of the hearers; (2) a sentence has a
grammatical meaning which does not (entirely) depend on the choice (selection) of the items of lexicon.

This can be illustrated by the following:

1. The selection of 'none' instead of 'someone' changes the meaning of an affirmative statement into
negative ("Someone has come — None has come"); the selection of an animate noun instead of the
inanimate is possible only with a changed meaning of the verb:
The wind blew (the leaves away) — The man blew his nose."

Bloomfield's idea that the meaning of a sentence is part of the morpheme arrangement, and does not
entirely depend on the words used in the sentence has later been developed by Ch. Fries and N.
Chomsky, who showed that sentences with non-sensical selection of words still have a definite meaning
because of the arrangement of linguistic forms.

2: Bloomfield understood grammar as meaningful arrangement of linguistic forms from morphemes to

sentences. He wrote that the meaningful arrangement of forms in a language constitutes its grammar,
and in general, there seem to be four ways of arranging linguistic forms: (1) order; (2) modulation:
"John!" (call), "John?" (question), "John" (statement); (3) phonetic modification (do—don't); (4)
selection of forms which contributes the factor of meaning.*

3. He produced the definition of the sentence that is now accepted by modern American linguistics. This
definition is given in Ch. Fries's book The Structure of English as the best among other two hundred
definitions, and it reads as follows: "...Each sentence is an independent linguistic form, not included by
virtue of any grammatical construction into any larger linguistic form."

This definition is essentially like that of another great linguist of that period, that is Otto Jespersen, the
Great Dane, as he is sometimes referred to.

Bloomfield also said that a sentence is marked off by a; certain 'modulation' or intonation.

He stated that in English the most favourite type of sentence is the 'actor—action' construction having
two positions. These positions are not interchangeable. All the forms that can fill in a given position
thereby constitute a form-class. In this manner the two main form-classes are detected: the class of
nominal expressions and the class of finite verb expressions.

4. Thus Bloomfield has shown a new approach to the breaking up of the word-stock into classes of
words, the positional approach. Bloomfield writes: "The syntactic constructions of a language mark off
large classes of free forms, such as, in English, the nominative expression or the finite verb expression..."
"We shall see that the great form-classes of a language are most easily described in terms of word-
classes (such as the traditional parts of speech), because the form-class of a phrase is usually
determined by one or more of the words which appear in it." [2: 190].

These long form-classes are subdivided into smaller ones. In modern linguistic works the nominal phrase
of a sentence is marked as the symbol NP, and the finite verb phrase-— VP. The symbols N and V stand
for the traditional parts of speech, nouns and verbs, although the NP may include not only nouns but
their equivalents and the noun determiners (e.g.: the man, my hand, this house, I, they, something,
some, others, etc.); and the VP with a transitive verb may have a NP in it (took a book, sent a letter,
etc.). The long form-class of N is now subdivided into: animate and inanimate, material and abstract,
class nouns and proper nouns. The long form-class of V is subdivided into intransitive verbs (VI),
transitive verbs (VT) and the latter are again divided into the V of the 'take type', the 'give type', the 'put
type’ and the 'have type', etc.

The division of the word-stock into form-classes is developed in Fries' book The Structure of English and
is dealt with In a most known article by Zellig S. Harris Co-Occurrence and Transformation in Linguistic
Structure.
5. Perhaps Bloomfield was one of the first to speak about 'utterance' as a linguistic unit.
Meanwhile the concept of utterance is of importance for the study of unknown languages, and this
concept also elucidates many syntactic problems in the familiar languages. The concept of utterance is
now accepted by the linguists of different schools.

6. The first mentioning of the Immediate Constituents (IC) can also be found in Bloomfield's book.
This theory of the IC which in the middle of the 20th century fascinated the minds of the linguists, and
has only been obscured by the Transformational grammar, was first propounded by Bloomfield. We may
suppose that the idea of the IC arose under the influence of Panmi's grammar because in the first
chapter of his book Bloomfield says that Panini's grammar taught the Europeans to study the IC of their
languages

These are the main ideas of Bloomfield's Language which make the book a predecessor of American
Descriptive linguistics.

Present-day English depends strongly on word order to convey meaning. Charles Fries argues that
“certain positions in the English sentence have become to be felt as subject territory, others as object
territory, and the forms of the words in each territory are pressed to adjust themselves to the character
of that territory” [8: 158]. Function words are another device. Having little or no lexical meaning of their
own, they serve to vary the functions of the lexical words. Consider: The mother of the boy will arrive
tomorrow. The words mother, boy, arrive, and tomorrow have meaning in themselves quite apart from
their grammatical relation, or meaning, in the sentence. They are full, or notional, words. But the words
the, of, and will express primarily a grammatical idea and have little or no meaning apart from the
grammatical function they indicate: the functions as a determiner of mother telling us that a particular
member of the class is meant; of relates the boy to the mother or, in other words, of makes the word
boy an attribute, or modifier, of the word mother; it is equivalent to a genitive inflexion (cf. the boy’s
mother); will indicates that the process of arriving will occur in the future.

The role of intonation is obvious when we have to differentiate between statements and questions,
between the theme and the rheme. Stress, or accent, helps to distinguish nouns from verbs (e.g.
.suspect vs. suspect), juncture-pause in speech distinguishes between such structures as night-rate and
nitrate or phrases, clauses and sentences.

As already mentioned, anxious to be objective, structural grammarians used formal methods of linguistic
analysis, such as immediate constituent, distribution, substitution, transformation (deletion,
permutation, etc.).

The term immediate constituents (IC) was introduced by L. Bloomfield as follows: “Any English-speaking
person who concerns himself with this matter is sure to tell us that the immediate constituents of Poor
John ran away are the two forms Poor John and ran away; that each of these is, in turn, a complex form;
that the immediate constituents of ran away are ran and away, and that the constituents of Poor John
are poor and John”.

To put it in more simple language, the constituents Poor John and ran away belong together, for they
stand side by side. They are the most important constituents since they constitute the core of the
sentence. The same principle of togetherness underlies the constituents Poor and John, ran and away.
However, as compared to Poor John and ran away, they are constituents of a lower level: they are
subconstituents of the higher level – Poor John and ran away. Hence two levels of analysis: higher and
lower where the lower level is subordinated to the higher level.
According to D. Bolinger, the principle of togetherness is very pervasive in language. It manifests itself in
“our resistance to putting something between two things that are more closely related to each other
than they are to what is inserted. Teachers find it hard to enforce the rule of interior plurals in forms like
mothers-in-law and postmasters general – speakers want to put the –s at the end. They are even more
reluctant to say hardest-working person, inserting the –est between the members of the compound
hard-working; and though some might manage it there, probably no one would say farthestfetched
story for most far-fetched story” [12: 158].

The aim of IC analysis is to discover and demonstrate the interrelationships of the words in a linguistic
structure – the sentence or the word-combination. It is not difficult to see a similarity between
immediate constituent analysis and the traditional procedure of ‘parsing’ sentences into subject and
predicate, attribute, object and adverbial.

Thus L. Bloomfield’s sentence could be described by a traditional grammarian as a simple sentence

whose subject is a noun phrase, made up of the noun John modified by the adjective poor, and whose
predicate is a verb-phrase consisting of the verb ran modified by the adverb away.

Both the traditional procedure and the IC method view the sentence not as just a linear sequence of
elements but as made up of “layers” of immediate constituents, each lower-level constituent being part
of a higher-level constituent.

As already mentioned, the aim of IC analysis is to show the syntagmatic interrelations between the
sentence constituents. Structuralists would agree that if we have described these interrelationships, we
have described the syntax of the sentence in its entirety.

The shortcoming of the IC method lies in its extreme formality: the analyst, using this method, is not
interested in the content of the interrelationships. Such syntactic notions as subject, predicate, object,
complement, attribute, adverbial, which constitute the basis of traditional analysis, practically were
never used by structuralists. In this way, content was separated from form. And language is a dialectical
unity of content and form.

Besides, the method of IC analysis is only capable of revealing word relationships within the sentence.
But “How could a frame so confined as that of immediate constituents be expected to fit comfortably
around the whole of syntax, when there are many important relationships that escape it? The classic
example is the relationship between the active and the passive voice: George sees Mary; Mary is seen by
George. An immediate-constituent analysis of these two sentences tells nothing about their underlying
kinship.” [12: 158].

Let us now turn to distribution.

Distribution is the set of contexts, or environments, within sentences in which a unit can appear. So, for
instance, the distribution of hair in written English is the set of the following contexts:

I combed my hair.

Give me the hair spray.

My hair is too long, etc.

The distribution of the word hair can be described as follows: 1) it can follow the word my; 2) it can
precede the word spray; 3) it can precede the verb be.
If we analyze other words, we shall find other positions they occupy, or other environments in which
they are used. Words that have the same distribution are words of the same class. We test their
distribution by substituting them for other words. Consider the sentence I combed my hair. The word
hair can be formally substituted for other words, such as place, town, wood, etc.

Distribution and substitution were used by structuralists for the classification of linguistic units. Like the
IC method, the method of distribution was treated as a method that enables the analyst to classify
words into classes objectively, i.e. without having recourse to meaning.

The transformational method was developed by Zellig Harris in the 1950s. The aim of a transformational
operation was to reveal similarities and differences in the structure of the units being examined or to
reveal the structural potential of the unit. To understand it, let us examine the following structures:

1) Mary has a new car.

2) Mary has a good time.

Superficially, the two sentences are identical in structure. However, they present two distinct structures.
Sentence (1) cannot be turned into the passive while sentence (2) can:

Mary has a new car. - *A new car is had by Mary.

Mary has a good time.- *A good time is had by Mary.

The structural potential of a linguistic unit can also be tested by this method:

a) my dog - the dog of mine;

b) Susan’s dog - the dog of Susan - the Susan dog;

c) John gave the book to me. - John gave me the book - The book was given to me.

d) John bought the book for me - John bought me the book - The book was bought for me - *I was
bought the book.

e) A number of people came - People came - *A number came - *The number of people came.

f) Bill fixed up a drink for John - Bill fixed a drink up for John - Bill fixed a drink for John up - Bill fixed up
John a drink.

No matter which aim is pursued, transformations help to reveal the existing relations between linguistic
structures.

g) John resides in New York - *John resides.

h) John is my best friend - John is.

i) John is walking in the park - John is walking.

j) Mary put the flowers in the vase -* Mary put the flowers.

k) Mary is writing a letter. - Mary is writing.

1) The door was closed - the door was closed by the janitor.
m) The door closed - *The door closed by the janitor.

n) The woman looked angry - *The woman looked angrily.

o) The woman appeared angry - *The woman appeared angrily.

p) We do not allow smoking in the lecture hall - It is not allowed to smoke in the lecture hall - Smoking is
not allowed in the lecture hall.

q) The student arrived late. - The student’s late arrival.

Through the transformational method we can show the structural potential of a linguistic unit as
compared to units exhibiting superficially similar structure. If linguistic units can be subjected to the
same transformation, we can say that they are identically structured. But if they cannot, their structure
is different.

Тo sum up, the merit of the transformational method can be stated as follows: 1) it enables the analyst
to diagnose linguistic structures; 2) it reveals the structural potential of linguistic structures. The
emergence of this method practically marks the end of post-Bloomfieldian linguistics and the beginning
of a new stage of structural linguistics.1