Philological Sciences /  3. Theoretical and methodological problems in the language research.

 

                        Postgraduate Kresin I.A.

                       Belgorod State University, Russia

         New computational methods in cognitive linguistics

 

         Computerization of scientific activity in many areas opens up new opportunities for professionals and has a huge potential. Methodological innovations, based on the use of software, enable programming language to solve research problems and, thus, open up new ways in the linguistic area as well. Below we will have a look at the method of computational text processing in cognitive linguistics, based on statistical analysis.

         As one of the most promising directions in the field of linguistics, cognitive linguistics is based on a number of assumptions, the main of which is the following: "linguistic and cognitive mechanisms are mutually integrated" [3, p. 198], which means that the study of the first one is possible through the study of the second.
         One of the most important concepts of cognitive linguistics is
the so called
recurrence of concepts. The notion of recurrence is inherent not only
in cognitive linguistics, but is used much broader and is understood as a repetition, frequency of any language elements [4, p. 135].  As a "universal linguistic category organizing the process of communication" [4, p. 136] the recurrence is widely studied in the framework of different linguistic areas.

         In cognitive linguistics the notion of recurrence of concepts (or frequency of linguistic representations of concepts in speech) is fundamental and reflects the specific features of cognitive consciousness of the people or of the individual. One of the tenets of cognitive linguistics is that these features can be identified through the analysis of the frequency of language verbalizations [2, p. 148].

         Recurrence of concepts is an important indicator of their relevance in the cognitive consciousness of the people or individual author. It reflects not only the linguistic, but also the cognitive and the social relevance of concepts [5, p. 80]. A significant increase of the units frequency objectifying a particular concept in the text may indicate an increase of its communicative relevance, that is, that it is actively discussed in the society and, consequently, is important for the national conceptosphere [2, p. 149]. Recurrence of the concept can also mean the relevance or irrelevance of the concepts in individual conceptosphere. The usage of various lexical units is traced to the time in which the author lived, and to the characteristics of his personality. In this way the communicative consciousness of the author may be also partially modeled by cognitive interpretation. [2, p. 149]

         Analysis of recurrence of concept allows to some extent to reconstruct the linguistic consciousness of the author, and draw some conclusions about his linguistic and cognitive preferences. It also allows to determine which concepts for each author were more, and which - less relevant, which concepts have been cognitively relevant to one or the other work period.

         In order to adequately investigate the recurrence of concepts in the works of an author, one must use statistical methods. Quantitative methods have long been widely used in historical linguistics and mathematical linguistics [7]. On the interaction of statistics and philology indicates E. Kielich, noting that "statistics is not simply a method of confirming or refuting certain linguistic assumptions, but often reveals patterns of language and the interaction of different language levels" [1, p. 37]. The use of quantitative methods today is one of the features of modern linguistics [6, p. 146]. According to S. Verkhozin, one of the reasons is just "the emergence of applied problems related to the text automatic processing" [6, p. 146].

         To give an example of solving a particular research problem, we can look at how software instrurments work based on the use of quantitative methods in linguistic research.

         One of the research goals was to analyze the linguistic side of the Kant's texts  "Critique of Pure Reason", "Critique of Practical Reason", "Critique of Judgment"  in order to find out which concepts are most relevant for the author at some work periods. Formulated in this way the research objective presupposed the listing of the most frequently used lexical units, taking into account all possible morphological variants.
         To perform this type of analysis, we needed to develop a special software. When creating a program, we used
the Harbour - a universal high level programming language. This choice was dictated by the fact that the Harbour is a free development open source environment, easy to install and to use, and does not require compiling and linking programs, making the way from writing a code to the result as much as possible reduced.

         The program consists of two main modules. The first module works with individual words, analyzing the full text, forming an array of all lexical units occurring in it, and counting the number of occurrences of each of them. The program then analyzes the array by combining different forms of  lexical units according to the prepared and replenished online directory.

         The second module splits the text into sentences and searches at each of them the set phrases in which  the lexical units can take many forms and be separated from each other by other linguistic units.

         The work result, provided by our text processing program, is the creation of special tables illustrating the frequency of use of different lexical units in different texts in relation to other works of the same author.

         We could conclude for example that the "Critique of Pure Reason" features a much more frequent use of the lexical units associated with the theory of knowledge in comparison to other works: Apperzeption (apperception), Wahrheit (truth), Erkenntnis (knowledge), Anschauung (perception), Vorstellung (representation), Erfahrung (experience), empirisch (empirical), Zeit (time), Raum (space), Logik (logic).

         This method of  the computational text processing can serve as a support for the further linguistic research. So, to conduct further analysis of the lexical units in the tables, using the methods of cognitive linguistics, we can model the corresponding conceptual units to rethink the structure of the most relevant concepts for this author and reveal new aspects of his works.

List of references

 

1. Kelih E. History of the use of quantitative methods in Russian Language and Literature // Kritikon Literraum. Slovyanskaya Filologia. 2010. pp. 37-44.

2. Popova Z.D., Sternin I.A. Cognitive Linguistics M .: AST East-West 2007. 314 p.

3. Rezanova Z.I. Cognitive Linguistics in the paradigms of linguistic functionalism and integrated concepts of consciousness // Bulletin of Tomsk State University. 2010. ¹ 334. pp. 195-199.

4. Sokolova, I.B. Recurrence as a universal language category (on the basis of modern English) [Text] / I. Sokolova // News Lugans'ka NHRIs. 2005. ¹ 5 (85). pp. 135-140.

5. Titova O.I. Prospects linguistic research of recurrent units lexicon // Philology. 2003. ¹ 2. pp. 79-86.

6. Verkhozin S.S. On the status of quantitative methods in linguistics //  Journal of Irkutsk State Linguistic University,  Issue number 3 (24). 2013. pp. 145-150.

7. Wang, W.S.-Y., Ke, J.Y., & Minett, J.W. Computational Studies of Language Evolution. Computational linguistics and Beyond // Huang, C.R. & Lenders, W.: eds. Academia Sinica: Institute of Linguistics. 2004. pp. 65-106.