On Complexity of Persian Orthography: L-Systems Approach
Nassim Taghipour
Pattern Research Center, Tehran, Iran
taghipour@iranprc.org
Hamid Haj Seyyed Javadi
Department of Mathematics and Computer Science, Shahed University, Tehran, Iran
h.s.javadi@aut.ac.ir
Mohammad Mahdi Dehshibi*
Pattern Research Center, Tehran, Iran
dehshibi@iranprc.org
*Corresponding author
Andrew Adamatzky
Unconventional Computing Centre, University of the West of England, Bristol, UK
andrew.adamatzky@uwe.ac.uk
Abstract
To understand how the Persian language developed over time, we uncover the dynamics of complexity of Persian orthography. We represent Persian words by L-systems and calculate complexity measures of these generative systems. The complexity measures include degrees of non-constructability, generative complexity, and morphological richness; the measures are augmented with time series analysis. The measures are used in a comparative analysis of four representative poets: Rudaki (858–940 AD), Rumi (1207–1273), Sohrab (1928–1980), and Yas (1982–present). We find that irregularity of the Persian language, as characterized by the complexity measures of L-systems representing the words, increases over temporal evolution of the language.
