Martin's law

from Wikipedia, the free encyclopedia

The Martinsche law is one of the language laws, the Quantitative Linguistics has developed. It makes a statement about the hierarchical structure of the vocabulary in the lexicon of a language.

Elaboration of the required data

If you look up a word in a lexicon to understand its meaning, that word is explained by other words. That's what it says in Duden. German universal dictionary (²1989) for armchairs as an essential explanation: seating furniture . If you ca n't do anything with the explanatory word seating furniture , you look it up again and learn about it: furniture . For furniture you get: Furnishings and, as an explanation, object . In this way, "sequences of definitions" of the type armchair - seating furniture - furniture - furnishings - objects can be created for any keywords , which are characterized by the fact that more specific expressions are replaced by more general ones. Carrying out this procedure for many words results in levels that begin with the specific keywords (in the example shown, armchairs ) and end with very general expressions (here: subject ). These levels are occupied by fewer and fewer expressions, from the particular to the general.

Such observations were apparently made for the first time by Martin (1974) for the French lexicon, which is why further investigations speak of Martin's law (s) . Martin's law meant that between the levels created by the chains of definitions, certain proportions exist, which in their simplest form take the form of a geometric distribution.

An example

A sample prepared by Schierholz is presented as an example; Bagheri was able to fit the 1-shifted mixed Poisson distribution to this sample :

x observed calculated
1 1482 1478.97
2 1110 1118.78
3 642 629.46
4th 334 340.78
5 160 165.14
6th 74 67.82
7th 23 23.59
8th 7th 7.06
8th 2 2.39

(Here x is the rank of the words, starting with the keywords in the lexicon and progressing with the increasingly abstract words that serve to explain the words of the previous rank; on rank x = 1 - based on the example in the previous section - among other things Armchairs appear; seating furniture on rank x = 2, furniture on rank x = 3, and so on. "Observed" indicates the number of words of the corresponding rank; "calculated" indicates the number of words of the corresponding rank that can be expected if the 1-shifted mixed Poisson distribution is a suitable model for the observed data Result: the 1-shifted mixed Poisson distribution is a good model for this sample with the test criterion P = 0.94, where P is considered good if it is greater than / equal to 0.05. For more detailed explanations, please refer to the literature.)

Bagheri presents further results in the same place for French and Polish; In addition to the sample given, a survey on a frequency dictionary was also tested for German. In all four cases the same distribution proves to be suitable.

literature

  • Dariusch Bagheri: Definition sequences and lexeme networks. In: Gabriel Altmann, Dariusch Bagheri, Hans Goebl, Reinhard Köhler, Claudia Prün: Introduction to quantitative lexicology. Peust & Gutschmidt, Göttingen 2002. pp. 94-133. ISBN 3-933043-09-3 .
  • Karl-Heinz Best : Quantitative Linguistics. An approximation . 3rd, heavily revised and expanded edition. Peust & Gutschmidt, Göttingen 2006, p. 92f. ISBN 3-933043-17-4 .
  • Robert Martin: Syntaxe de la définition lexicographique: étude quantitative des définissants dans le "Dictionnaire fondamental de la langue française". In: J. David, R. Martin (eds.): Statistique et Linguistique . Klincksieck, Paris 1974, p. 60-71.
  • Jadwiga Sambor: Lexical networks . In: Reinhard Köhler, Gabriel Altmann, & Rajmund G. Piotrowski (eds.): Quantitative Linguistics - Quantitative Linguistics. An international manual . de Gruyter, Berlin / New York 2005, pp. 447–458. ISBN 3-11-015578-8 .
  • Jadwiga Sambor, Rolf Hammerl (Ed.): Consequences of definitions and lexeme networks. Vol. 1. Richter-Altmann Medienverlag, Lüdenscheid 1991. ISBN 3-9802659-0-0 .
  • Stefan J. Schierholz: Lexicological analyzes of the abstractness, frequency and polysemy of German nouns. Niemeyer, Tübingen 1991. ISBN 3-484-30269-0 .

Web links

Individual evidence

  1. ↑ Jadwiga Sambor, Rolf Hammerl (ed.): Definition sequences and lexeme networks . Vol. 1 1991
  2. ↑ Archived copy ( Memento from May 18, 2016 in the Internet Archive )
  3. ↑ Stefan J. Schierholz: Lexicological analyzes on the abstractness, frequency and polysemy of German nouns , p. 33, sample I.
  4. ^ A b Dariusch Bagheri: Definitions and Lexeme Networks. In: Gabriel Altmann, Dariusch Bagheri, Hans Goebl, Reinhard Köhler, Claudia Prün: Introduction to quantitative lexicology. , P. 124