| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Linguistic Aspects Of Semantic Analysis
Semantics, as a linguistic discipline, has its own subject. Yet that subject is difficult to define. Although the majority of linguists agree to the fact that semantics studies the meanings of linguistic expressions, there is no conventional answer to the question of what is to be understood by meaning. Different views on the subject produce different demarcation lines between semantics and other linguistic disciplines, between semantics and pragmatics, to be precise. That is why the issue of the subject of semantics deserves close attention.
Due to the ambiguity of the term "meaning", a neutral term "content" or "information" is normally used to describe semantics as a division of linguistics that deals with the content of language units and those speech expressions that are constructed from these units.
Attention should be paid to the fact that in many natural languages there are, as a minimum, two, instead of one words to denote the content matter of language expressions:
It is hardly an example of linguistic redundancy - it is more of a reflection of the fact that there are two related, but not identical concepts of the content in the minds of native speakers. They are supposed to correspond to two forms of "essence" that we indivisibly designate as "content" of a linguistic expression.
A marker of the words' "informational" usage is the possibility of their substitution for descriptive expressions, such as "something the X denotes / signifies / expresses" or "something that someone wanted to say, using X", assuming that the source message remains intact and its correctness is sustained.
Russian Dictionaries of Synonyms define "sense" and "meaning" as synonyms in the relevant aspects of usage, not absolute synonyms though. Thus, in the dictionary by S.I. Ozhegov "meaning" is defined as "something that a given phenomenon, notion, object means, denotes", while "sense" is described as "meaning, inner essence of something, which is fathomed by the mind". The given definitions testify to the fact that, having the common part denoted as "content" or "information", the latter differs from the former by a new "participant" - the mind that perceives that content. However, dictionary entries are insufficient for understanding the fact that sense differs from meaning in the way that it is "grasped" by the mind. The very definition of a differentiating marker appears to be too vague. That is why there is a need for a more detailed analysis of differences in the usage of those words.
Making a distinction between different types of "content carrier", linguistic semantics postulates that objects available for perception and related to any type of "carrier" can have sense, beginning from the words and ending with a scene of two passers-by bumping into each other. It is only those information carriers that can have meaning, whose signification function is not major, but at least, minor. That means there are rules and conventions to coordinate the method of interpretation of those carriers in the context of a certain community.
Thus, combinability of sense and meaning with different types of carriers demonstrates that the notion of meaning dwells in the common mind as linked to presumption of a sign system, the meaning carrier being an element or text thereof. It actually means that meaning is expressed in the sign system. There is no such presumption for the notion of sense.
The difference between sense and meaning is usually disclosed in the process of comparing the methods of their interpretation (explication), assuming that there is only one carrier (word, for example.). When the word meaning is discussed, it is its dictionary definition that is primarily meant. When the sense of a word is discussed, something different is meant here, as a rule.
First of all, explicating the word sense in linguistic semantics means indicating a set of identities that word can denote (the so called "extensional" ). It takes place when variability is presupposed for estimating the scope of this set. Certainly, it is not various dictionary definitions that are meant here, but the question of which specific set of principles is considered (by opponents) sufficient for this word's application.
Secondly, the answer to the question why a word expressing a certain content has this particular form, can be considered the explication of the word sense (in linguistics it is called revealing the inner form of a word, or its word-building motivation).
Thirdly, the word sense can imply factual or evaluative associations evoked for this word in the speaker's minds (the so-called associations and connotations).
Since meaning is a content rigidly fixed for a sign, it can be identified and then learned, while sense is something changeable, unregulated, where one has to seek, trace, decipher, find keys to discovery and so forth. A sign possesses meaning as its integral part, while the sense of a sign presupposes something external that temporarily fills a sign. A sign can possibly acquire both meaning and sense, but it can only be filled with sense.
Let us summarize our excursus in the sphere of linguistic semantics [Cobzeva I.M. "Linguistic Semantics", Moscow, 2000]. The differences discussed above receive a unique explanation assuming that there are two close but not identical concepts that correspond in the native speakers' mind to sense and meaning. They can be defined as follows:
Meaning X is the information conventionally related to X, i.e. according to the conventional rules for using X as a means of information transmission.
The sense of X for Y in T is the information connected with X in the Y mind in the time period of T, wherein Y produces or perceives X as a means of information transmission.
No wonder that scientists, who deal with semantics, treat its subject differently. The Russian linguist V.A. Zvegintsev who compared semantics to the "ruling princess", wrote the following concerning this subject: "...our princess possesses a magic quality: everyone who worships her has a different perception of her in the shape that seems most attractive to him. Every one of her warriors protects his image with courage, trying to have others think according to his vision" [Zvegintsev V.A. "Sentence and Its Relation to Language and Speech", Moscow, 1976].
Semantic Analysis makes it possible not only to specify and extend, but also to formalize the definition of such a category as "sense".
Since a state is defined as a certain direction in the semantic space that has two poles, it can be expressed on the basis of complimentary antonymy. Semantic analysis does not only allow to discover the probabilities of this or that state actualization for a given mentality, but also to specify, at what probability rate each of its poles is actualized. This determines the sign of the most probable proposition to a given denotatum as part of interpretation accepted for a studied mentality. Alongside with that, the calculations are made of rigidity (stability) for the revealed interpretations.
Since the sense of an object (word, statement)
is determined by the actual situation and the subject's requirement, it
can be analyzed through the motivational vector
The
choice of an actual/ situation-specified meaning of a given object
Decomposition
of a selected object state (word interpretation) rating
into categories actually determines its sense contents for given conditions,
i.e. it reveals the semes that, being part of significatum, yield the
greatest weight (the sign considered) in the statement motive realization.
That accounts for the third aspect of a word or statement content. Variability
of motivational vector
Semantic Analysis does not only make it possible to gain deep insight into the content of a lexeme, but also to study the processes of nomination as well as generation of new meanings and senses within the "living" language both for different conditions and cultures.
There is one more interesting aspect of Semantic Analysis application in linguistics. It is glottochronology. Glottochronology is a branch of comparative historical linguistics that deals with studying the speed of linguistic changes and establishing on that basis the period of break-up in the related languages as well as the degree of their proximity to one another. Although different levels of related languages (particularly, the phonological and grammatical levels) could be studied for that purpose, as it is done for ascertaining the relative chronology within one language history, the most valid quantitative results for historical linguistics are yielded by statistical research in vocabulary (lexical statistics).
The American scholar Morris Swadesh proposed that lexical changes could have a constant speed, on which basis he conceived his glottochronological theory, sometimes also referred to as lexical statistics. For the speakers of different generations to be able to understand each other, the number of lexical substitutions should not be too large: phonetic and morphological changes can not sufficiently hinder communication, while a great percentage of lexical variation is bound to cause misunderstanding between grandparents and grandchildren, for example. The number of substitutions should be lowest in the basic stock of vocabulary that constitutes the nucleus of any language.
Lexical statistic glottochronology infers the time of branching in relative languages from the assumption of equal speed of the change of the basic vocabulary stock that is essential for application in the most frequent and significant communication situations. According to glottochronology, that basic vocabulary segment would include such stable words as personal and interrogative pronouns, several verbs of movement ("to come"), verbs of elementary physiological functions and perception ("to drink", "to hear", "to see"), adjectives denoting size ("wide", "long"), cosmic phenomena ("the Sun", "the sky"), animals ("a worm", "a snake"), color adjectives ("black"), relationship/kinship terms etc. It is established that within the group of 200 or 100 words that belong to this segment of vocabulary (basic list) about 80% of vocabulary is retained throughout a millenium (for a list of 200 words - 80,5 or 81 %; for the list of 1000 words - 86 %), whenever it is possible to trace a language history back to one or several millennia (for example, in the history of the Egyptian language in its relation to Coptic or Latin in its relation to Romance languages etc.).
Basic postulates of glottochronology would be formulated as follows:
Sum of the above postulates makes it possible to deduce the main mathematical relation in glottochronology:
where the time elapsed
from the development moment to a specific later moment is designated as
The segment of BL words that was retained in the two languages, would constitute, correspondingly:
Knowing the
In most cases the Swadesh theory arrived at the "younger" dates compared to those discovered on the basis of the real language history. All that made the majority of researchers doubt some of the postulates of glottochronology.
There are three most frequently mentioned points:
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
, formed
for the given conditions 
.
to 
. The sign of scalar product
determines the proposition sign for these conditions, while its value
stands for the sense adequacy degree of using this interpretation. 
determines the "flowability"
of object sense, as different from meaning under various pressing conditions.
.
, is constant
(i.e. depends only on the length of period selected, and not on its selection
criteria or what particular words of a particular language are specified
for study).
,
is a "relative
speed of word slip-out" from 
is the
absolute speed); and 
is the number of words from the source
BL retained to the 
being the number of words that had slipped out of the BL.
.
. coefficient
and the segment of words retained in a given language from the BL, we
can calculate the length of the time period elapsed
.
.
. (2)
.
(3)
(red line) practically coincides with the "corrected"
curve (2), assuming 
(blue line). Behind that limit curve (3) slopes more gently
that curves (1) and (2).
