Contents of Glottometrics 5, 2002 (including abstracts)

To honor G. K. Zipf

 

Kromer, Victor
Zipf´s law and its modification possibilities 1-13
Abstract. In this paper we consider the possibilities of known Zipf-Mandelbrot canonical law modifications. The proposed modifications explain the behavior of the right tail of the distribution  and the presence of a deflection in the central part of the distribution (a crater). It is shown that the average word information load is invariant to the sample heterogeneity and that the proposed usage measure "places" the words more correctly with regard to their "importance".
Li, Wentian
Zipf´s Law everywhere 14-21
Abstract. At the 100th anniversary of the birth of George Kingsley Zipf, one striking fact about the statistical regularity that bears his name, Zipf's law, is that it seems to appear everywhere. We may ask these questions related to the ubiquity of Zipf's law: Is there a rigorous test in fitting real data to Zipf's law? In how many forms does  Zipf's law appear? In which fields are the data sets claiming  to exhibit Zipf's law?
Fenk-Oczlon, Gertraud & Fenk, August
Zipf´s tool analogy and word order 22-28
Abstract. This article starts with Zipf’s (1949) “Tool Analogy”, where the artisan arranges and re-designs his tools in a way minimizing his total work; as a result, more frequently used tools tend to be nearer to him (better accessible), smaller and multifunctional. We then argue that short distance, small size and multifunctionality reflect not only a high overall relative frequency of usage, but in particular a high frequency of usage in the first steps of a variety of complex working procedures. Tool order – word order? This extended Tool Analogy fits to the tendency of more frequent words to obtain initial positions in frozen binomials (Fenk-Oczlon 1989) and the new finding (Fenk & Fenk-Oczlon 2002a,b) that the short, frequent and multifunctional function words tend to concentrate in the first part of  sentences.
Hilberg, Wolfgang

The unexpected fundamental influence of mathematics upon language

29-50

Abstract. The functional structure of human language networks in the brain could be revealed in an indirect way by measurements in the abstraction level of words. The result is a very large deter­ministic graph or network, respectively, which was unknown in mathematics up to now. The whole network can only be represented in a matrix. Following Shannon's theory, it displays optimum properties for information processing (maximum entropy). The structure of the network can be subdivided by introducing word classes with increasing magnitudes which could contribute to an understanding of the biological generation of networks. The hypothesis is that such facts are valid for all natural languages. Differences will exist only in the individual distribution of matrix dots. That means, speaking precisely, that every language has a distinct individual network structure of its own. Surprisingly it can be shown that the superior general type of the universal network structure can be generated by statistical experiments. The properties of these networks were compared with those of natural language networks which are definitely not statistical. Finally the enigma of the famous Zipf-diagram can be disclosed by observing networks and text paths which run inside of them along existing connections from node to node. A staircase curve emerges, which is a better description of reality than a smoothed power law. All this can be repeated by experiments, which means that eventually we found a transition from descriptive to constructive science. Therefore the new ideas could be applied immediately also in technology. Certainly the basic biological language structure arose a long time ago. Later on the typical patterns of the network connections for different language families should have evolved separately and were almost certainly accompanied by optimization processes for maximum entropy. Nowadays the details of the connection patterns for any language have to be learned anew by every child, and in this process, unusual alterations are not allowed by its language community.

Köhler, Reinhard   
Power law models in linguistics: Hungarian 51-61
Abstract. First, the status of Zipf(-Menzerath)’s Law and its criticisms are discussed, and the application of power law models, particularly in linguistics, is supported from a general point of view. The following sections, empirical studies on dependencies are conducted which test the Zipf-Mandelbrot Law, other power law models (Menzerath-Altmann’s Law, the length-frequency dependency), and the word length distribution on data from Hungarian (a text and a dictionary).
Meyer, Peter
Laws and theories in quantitative linguistics    62-80
Abstract. According to a widespread conception, quantitative linguistics will eventually be able to explain empirical quantitative findings (such as Zipf’s Law) by deriving them from highly general stochastic linguistic ‘laws’ that are assumed to be part of a general theory of human language (cf. Best (1999) for a summary of possible theoretical positions). Due to their formal proximity to methods used in the so-called exact sciences, theoretical explanations of this kind are assumed to be superior to the supposedly descriptive-only approaches of linguistic structuralism and its successors. In this paper I shall try to argue that on close inspection such claims turn out to be highly problematic, both on linguistic and on science-theoretical grounds.
Robbins, Jeff

Technology, ease, and entropy: a testimonial to Zipf´s Principle of Least Effort

81-96
Abstract. Evidence for the truth in George Kingsley Zipf’s Principle of Least Effort can be found in the deep attraction we have for anything promising us an easier route. The selling point of virtually all technology is the promise of new means to ease. But, beneath the vast glittering surface in the sea of hype runs a dissipative current.  With so many things capitalizing on our aversion to effort, individually and collectively, as a species we're losing it because we're not using it. Seeding that awareness is the first step in reversing the flow.
Grzybek, Peter & Altmann, Gabriel
Oscillation in the frequency-length relationship 97-107
Abstract. The analysis shows that there is no intrinsic oscillation in the relation between frequency and length of words. The rise of oscillation is caused by using moving averages for smoothing the extremely dispersed data.

Glottometrics ist eine unregelmäßig erscheinende Zeitschrift für die quantitative Erforschung von Sprache und Text.

 

Glottometrics is a scientific journal for the quantitative research on language and text published at irregular intervals

Beiträge in Deutsch oder Englisch sollten an einen der Herausgeber in einem gängigen Textverarbeitssystem (vorrangig WORD) geschickt werden.

 

Contributions in English or German written with a common text processing system (preferably WORD) should be sent to one of the editors
Glottometrics kann aus dem Internet heruntergeladen, auf CD-ROM (PDF-Format) oder in Buchform bestellt werden. Glottometrics can be downloaded from the Internet, obtained on CD-ROM (in PDF) or in form of printed copies

Herausgeber/Editors:

G. Altmann 02351973070-0001@t-online.de
K.-H. Best kbest@gwdg.de
A. Hardie a.hardie@lancester.ac.uk 
L. Hrebicek hrebicek@orient.cas.cz
R. Köhler koehler@uni-trier.de
V. Kromer kromer@newmail.ru 
O. Rottmann otto.rottmann@t-online.de
A. Schulz reuter.schulz@t-online.de 
G. Wimmer wimmer@mat.savba.sk
A. Ziegler arneziegler@compuserve.de

Herunterladen/ Downloading: http://www.ram-verlag.de

ISSN 1617-8351                                                                                     back