Eystein Glattre, VF-NMBU, Oslo, Norway
Havard R. Glattre, Codestitcher AS, Oslo, Norway

Abstract: In this paper we demonstrate that the meaning of words is fractal, the consequences of which are far-reaching. We also address the problem whether the complexity of sentences dominate that of words. In a previous publication we demonstrated that there exist word-based Complexity Connections in Corpuses (CCIC) and therefore that corpus texts contain a typical, hidden, word-connecting structure which we called the Network of Correlation Connections (NWCC). In this article we show that CCIC is valid not only for words, but also for the phonemes of corpus texts. NWCC makes it possible to demonstrate that the time of speech is fractal, non-Euclidean and two-dimensional. To study this in greater detail we use Vrobel s Now-theory and develop a verbal Now-hypothesis by which we manage to demonstrate that human speech has both temporal depth and temporal length. This Now-hypothesis facilitates the shaping of the Message Production Hypothesis or MPH by which CCIC and the NWCC-network become consequences of the way the human, cerebral machinery produces words. They are marks of how humans verbally present their thoughts. The NWCC network and its mathematical model opens for new ways to study the activity of neuronal networks.

Keywords: fractals, phonemes, verbal-Now, CCIC, speaking formulas, 2D-time, network model