Descriptive linguistics – Part 1

The differentiation between endocentric and exocentric constructions is based on the properties of a phrase that come to the fore in a larger syntactic structure. In other words, this approach ignores the inner structure of a word combination. Thus, for instance, despite differences in their inner structure, such word combinations as poor John and Tom and Mary are referred to the same type, since they exhibit the same properties in an extended sentence. To solve this problem, Bloomfield suggests a further elaboration, based on the inner structure of the word combination. Thus, endocentric phrases are further divided into serial (or co-ordinate) and attributive (or subordinate). In the latter case, the phrase belongs to the class of its principal element, called head, the other element is called attribute. The attribute, in its turn, may be a subordinate phrase, e.g. very fresh milk. Thus, there may appear several ranks of subordination. Exocentric phrases are divided into predica­tive (i.e. John ran away) and prepositional (beside John). As a result, we may observe that both exocentric and endocentric phrases may be described in terms of formal word classes (parts of speech). Consequently, the notion of formal word classes is of primary importance for syntax.

One may notice some inconsistency in the subcategorization of exocen­tric phrases, since predicative constructions are discerned on the ground of the syntactic relation, while the morphological criterion is chosen to distin­guish prepositional constructions. This division, however, is rather conven­ient, as it distinctly defines characteristic features of phrases.

Bloomfield develops his syntactic theory further and claims that, as the choice (selection) of formal word classes plays the main part in syntactic constructions of many languages, the primary task of syntactic studies, then, is to clarify the ways in which different formal classes are used in syntax. Therefore, distinguishing between syntax and morphology, Bloomfield fo­cuses his research on defining formal word classes and their positions in phrases. As to the notion “sentence”, Bloomfield considers the sentence to be a specific phenomenon, different from the notion “phrase” in that the sentence is an independent language form, that is, a form in the absolute position, since the sentence is not a part of a larger form.

In Bloomfield’s system, the sentence is not, in fact, specified in com­parison with the phrase, and the analysis of the sentence structure is not the subject of Bloomfield’s syntactic theory. Bloomfield supposes that every language tends to have its “favorite”, more frequent forms of the sentence. In English, for example, he points out two forms: 1) the construction “doer + action” (e.g. Mary sings. Who sings? Does Mary sing?) and 2) “the Infini­tive expressing an order” (e.g. Go on! Don’t say that!)

As Bloomfield believes that the internal sentence structure and the phrase structure may be described using the same terms, such notions as predication, subject, predicate, secondary sentence parts turn out to be su­perfluous for his theory. According to Bloomfield, these terms do not denote any specific syntactic notions. Predication is a term used to refer to a two-member construction typical of many languages: the component denoting substance is called “subject”, whereas the other component is termed “pred­icate”. Bloomfield claims that these terms are relevant only when applied to languages that have several types of two-member sentences. In languages that are characterized by only one type of two-member sentence (as it is the case in English) these terms make no sense.

One of the most important terms in Bloomfield’s syntactic theory is immediate constituents, i.e. the elements that make up a particular phrase. For example, the complex form Mary s mother brought in the tea consists of two parts – Mary s mother and brought in the tea.

A further development of the immediate constituents theory is found in the book Immediate Constituents by Wells published in 1947. According to Wells’ syntactic theory of immediate constituents, successive morphemes in a sentence are divided into certain types and classes. The classes are de­fined in the following way: if there is sequence S, then the class where sequence S belongs is defined as class that includes all sequences whose successive morphemes belong to the same classes as the morphemes of se­quence S. All the elements of the class, therefore, contain an equal number of morphemes. An important peculiarity of Wells’ classes is that a sequence belonging to a certain class may be replaced by a sequence of a different class. In other words, two sequences of morphemes may be found in similar distribution, though their internal form may differ. The relationship between these two sequences is interpreted as follows: if one of the sequences is not shorter than the other (i.e. contains the same number of morphemes) and dif­fers structurally, i.e. does not belong to the same class, the second sequence is called “extension of the first sequence”, while the first is called “model”.

Descriptive linguistics – Part 3
Descriptive linguistics – Part 4