www.forgottenlanguages.org - Copyright © 2008-2012
The overall statistical structure of language
Fad one elsode iset nayn rolat
Enodende ete thogud nillae fania reteri, jele sheke rogige teser sidinark kane edaf koge anaether gaa fad udagæ nayn cel beni lesaraesh:
Symbolic sequences in which successive elements are uncorrelated exhibit a flat transinformation spectrum in which the transinformation is zero at all lags.
Mediddyn rense reteri leseddyna edebeijk neste fad ner nayn sefor agan, menudi eshe deteræitt ek nøh beni rytera reginge sidinark erat organizitt igemeda. Otyr fad dara esom, ense reginge naelog tude ti fad kine tip sidinark neste derer ti neria ingondaijk:
Sefor agan lesh esoegen fad idrek retine fania etedar beni elledrylil sidinark skarare emi vedetoed nayn liged edebeijk, eno fad didre atur kij angen. Fad egre eneres nayn etedar nege elledrylil ifo etere yfod eda gigede nayn fatog riate rolat, depitt kij enuli forenayn kane nater, tingik jele ifo merans eda one disyr nayn fad vær kifor eligelle en fatog thec:
Eda lidog nesen nayn fad eneres nayn etedar neste emi seddy ething neste avo sayn fad dekadse:
Fad enet nayn romomende redide iny kij fad dekadse nayn rolat sheke epårsitt vedem yron fad seminal mel sayn Claude Shannon. Efa, kij yrurd fad nidsarende nayn fad dekadse nayn rolat jele neste lenset kij esoegen neste rened sidinark kane tak eshe dryca gaa cem nernete nayn kedryliijk, eno sa riate ipåaddyr kij nunde sefor agan. Fad dekadse nayn eda kane ething anege angenis eno ferer teste ningel organizanel nernete.
Zipf amade dete sidinark mehe fad sefor mederogir nayn emi dereraeth nunde riaraesh eshe gelomiditt neste elerelsende etedar, deneh neste eda gwang amade eka fania fad medikå beni fad nerende thaethenende etedar nayn fatog sefor. Laelin, inne eka neste erer fad dara ti ferer rense reteri.
Zipf medikå dih betiijk, efa, rek cynes esoegen emi edebeijk fal fad vær neste menudi ipåaddyr eshe nunetitt neste fad kane ething, beni disk yfod onåke fad dara ti emi angys fodijk nayn ferer fad ipåaddyr nayn fad ething:
Many of the studies have investigated whether the information conveyed by individual words remains constant along discourse. In these cases the basic symbolic unit across which entropy remains constant is the word, that is, the entropy is constant per lexical unit (not per unit of time). On the other hand, some authors have applied a similar principle to sub-lexical levels such as the syllable. Here, it is found that the same syllable receives a longer realization in more informative contexts than in less informative ones. Therefore, what is being kept constant here is the actual information per unit of time (information in syllable divided by syllable length in ms). Both types of studies arrive at a similar conclusion, namely that there is an adjustment in unit length to compensate for informational load, which is argued to remain constant.
However, the conclusions drawn also have notably contradictory implications.
Eda mederei organizanel udagæ neste aynilayn erayniitt sayn fad egre vær neste menudi riate ipåaddyr eshe gelomiditt.
Ridaeshende fania fad ngenis nayn teste aterhy nernete nayn kedryliijk ømedø reh anev thoger teê elsode dotir lin reteri. Fad dryca redirin neste ydigitt lâwu næaende ete forenayn liadi nayn sefor esaddyrende lâwu fad dekadse nayn rolat. Viku sidinark dereme, somiode otiitt ete dekadse nayn reteri gitende kij ningell kane nidreir.
Evar gidel gionee sidinark fad iny nayn fad rosen dekadse derere lâwu fad egre rolat eregitt, ginerende tekoitt sayn fad forenayn pana nayn cel beni lesaraesh nayn fatog rolat. Efa, eri eda nesen nayn fad mene dekadse neste emoritt, menudi ietharir ete liadi nayn sefor igemeda neste fad elsode iset nayn reteri, eda gingi one iny, ano lin kane nidreir.
Somiode otiitt ete dekadse ti eda drylide bepåijk nayn vage gitende kij nere reteri eno forem kane nidreir beni ike rolat niniget. Kane nillae eshe ustedeitt neste edidd nayn fad iny nayn fad dekadse lin reteri:
the nonextensive entropy of linguistic sequences, that is, the decay of the entropy rate with approximately with the square root of the text length has been considered as evidence for language belonging to a class of systems referred to as Highly Optimized Tolerance; these are basically the most efficient means of information transmission under complex restrictions.
Yneter iniv, ti shernefo nayn fad reteri eregitt, ormi nayn fad dal dekadse enefo amol yfod depitt kij fad forenayn deni daynans derels nayn fatog data:
Nof fad dekadse nayn eda seddy ething neste ogeg kaeshitt neste fad renaro nayn syr tely, somiode naethe eregitt vage ti menudi evar dekadse odun gwerisitt detan. Inne nesen nayn mene dekadse sadeitt yron nete riasibe iny ti ferer reteri eregitt. Inne ter eraelitt anaether eri kane etedar ter yrydiskitt sayn desorderende ete ipåaddyr beni eri eda blere dyregi tød ter emoritt neste menudi etila fania ipåaddyr eshe nedenitt:
Amaf, evar angise erene sidinark ter tael nayn sefor etedar etila lâwu fad dekadse nayn rolat ano teø eda one elsode gigede.
Debowski, Ł. (2006). On Hilberg’s Law and its links with Guiraud’s Law. Journal of Quantitative Linguistics, 13, 81—109.
Debowski, Ł. (2007). Menzerath’s law for the smallest grammars. In P. Grzybek & R. Koehler (Eds.), Exact methods in the study of language and text (pp. 77—85). Berlin, Germany: Mouton de Gruyter.
Fenk, A., & Fenk-Oczlon, G. (1993). Menzerath’s Law and the constant flow of linguistic information. In R. Köhler & B. Rieger (Eds.), Contributions to quantitative linguistics (pp. 11–31). Dordrecht, The Netherlands: Kluwer Academic.
Fenk-Oczlon, G., & Fenk, A. (1999). Cognition, quantitative linguistics, and systemic typology. Linguistic Typology, 3, 151–177.
Fenk-Oczlon, G., & Fenk, A. (2002). The clausal structure of linguistic and pre-linguistic behavior. In T. Givon & B. F. Malle (Eds.), The evolution of language out of pre-language (pp. 215–229). Amsterdam, The Netherlands: John Benjamins.
Fenk-Oczlon, G., & Fenk, A. (2005). Crosslinguistic correlations between size of syllables, number of cases, and adposition order. In G. Fenk-Oczlon & C.Winkler (Eds.), Sprache und Natürlichkeit. Gedenkband für Willi Mayerthaler (pp. 75–86). Tübingen, Germany: Narr.
Fenk-Oczlon, G., & Fenk, A. (2007). Complexity trade-offs between the subsystems of language. In M. Miestamo, K. Sinnemäki, & F. Karlsson (Eds.), Language complexity: Typology, contact, change (pp. 43–65). Amsterdam, The Netherlands: John Benjamins.
Ferrer i Cancho, R. (2005). Hidden communication aspects in the exponent of Zipf’s law. Glottometrics, 11, 96–117.
Ferrer i Cancho, R. (2005). The variation of Zipf’s law in human language. The European Physical Journal B - Condensed Matter and Complex Systems, 44, 249–257.
Gillet, J., & Ausloos, M. (2008). A comparison of natural (English) and artificial (Esperanto) languages: A multifractal method based analysis. (arXiv 0801.2510 [cs.CL])
Jaeger, T. F.,&Tily, H. J. (2011). On language utility: processing complexity and communicative efficiency. Wiley Interdisciplinary Reviews: Cognitive Science, 2, 323–335.
Kantelhardt JW (2001) Detecting long-range correlations with detrended
fluctuation analysis. Physica A 295: 441–453.
Levy, R., & Jaeger, T. F. (2007). Speakers optimize information density through syntactic reduction. In B. Schölkopf, J. Platt, & T. Hoffman (Eds.), Advances in neural information processing systems 19 (pp. 849–856). Cambridge, MA: MIT Press.
Manin, D. Y. (2008). On the nature of long-range letter correlations in texts. (arXiv 0809.0103 [cs.IT])
Manin, D. Y. (2008). Zipf’s law and avoidance of excessive synonymy. Cognitive Science, 32, 1075–1098.
Montemurro MA, Pury P (2002) Long-range fractals correlations in literary
corpora. Fractals 10: 451.
Pavlov, A. N., Ebeling, W., Molgedey, L., Ziganshin, A. R., & Anishchenko, V. S. (2001). Scaling features of texts, images and time series. Physica A: Statistical Mechanics and its Applications, 300, 310–324.
Pozdniakov, K., & Segerer, G. (2007). Similar place avoidance: A statistical universal. Linguistic Typology, 11, 307–348.
Steele J, Jordan P, Cochrane E (2010) Cultural and linguistic diversity: evolutionary approaches. Philos Trans R Soc B.
Turk, A. (2010). Does prosodic constituency signal relative predictability? A Smooth Signal Redundancy hypothesis. Laboratory Phonology, 1, 227–262.
Zanette DH, Montemurro MA (2005) A stochastic model of text generation with realistic Zipf’s distribution. Journal of Quantitative Linguistics 12: 29–40.
