Once upon a day the word DINGIR (DIMER) arose... Joannes Richter Abstract This

Once upon a day the word DINGIR (DIMER) arose... Joannes Richter Abstract This essay is written as a chapter in Communications Theory for students in communications and is understood as bad as Tohu wa-bohu or Tohu va-Vohu (תההו ובבההו ṯōhū wā-ḇōhū) by linguists. If you studied linguistics you may as well throw this paper in the paper basket. This paper uses 5 channels and 5 categories. This cannot be mixed with etymology, which is refers to a word or morpheme (e.g., stem[6] or root[7]) from which a later word or morpheme derives. This way one will never know how the first morpheme arose. A word “DIMER” symbolizes the 5 communication channels. In retrospect I am aware of a global turbulence in economy and stability. Probably this moment is a final phase in stability, in which I have a chance to close my books and concentrate as many details in overviews. Probably this manuscript is interpreted as clumsy garbage, but quite nice compared to the papers, which - composed as Nobel-prized theories or copied PhD theses... - are interpreted as the “academical “masterpiece”-Dissertation for the modern politicians (or other 'doctors')”. This essay is concentrated on the observations of the 5 communication channels in the standard alphabets, which may be concentrated in the most important words of the PIE- (Indo-European), Hebrew and probably a few other languages. These 5-letter words are named pentagrams. The paper is based on a large database of detailed documents in the episodes from 1989-2023. From 2023 I reduced my public database to a few papers in Scribd. The most efficient tools to understand the pentagrams are the 2-dimensional tables of the alphabets. The earliest trace of a genuine pentagram is Dingir (Sumerian sky-god) – in emesal pronounced as: DIMER and the latest pentagram is LOUIS. Another Sumerian god is GESTÚ (as Enki, the god of knowledge). The theonyms DIMER and GESTÚ are inserted as the 285th and 286th pentagram in the database at appendix 2, titled: “The (incomplete) Overview of the (~289) Pentagrams”. Fig. 1: Dingir (Sumerian sky-god) – in the emesal dialect pronounced as DIMER (Public Domain) Introduction There once was a time, when I did not understand how the word „Diaus“ had been composed. There were people how claimed they knew the word for the sky deity refers to the daylight and it had been derived from the daylight1. Deriving a sky god's name from “the daylight, what had been created by the daylight” did not satisfy my logic. There should be a number of fundamentals, for which the words or letters had to be created from nothing. I imagined the name „Diaus“ had to be composed from uttering a Word from a standardized mechanism in the human voice. I investigated the human voice and the design of the places of articulation. I inspected the Sefer Yetzirah, the Book of Formation, or Book of Creation, although some early commentators, such as the Kuzari [1], treated it as a treatise on mathematical and linguistic theory as opposed to Kabbalah. Ternary encoding As a student of communication engineering I noticed the number of letters in the alphabet had not be optimized to the law. The author Subhash Kak compares the binary and ternary coding and explained why the ternary code is superior in the comparison to the binary code. Mathematically, ternary coding is more efficient than binary coding. It is little used in computation because technology for binary processing is already established and the implementation of ternary coding is more complicated, but remains relevant in algorithms that use decision trees and in communications 2. The human voice is one of these complex, but most efficient implementation of ternary coding. The places of articulation In the human voice the Latin alphabet however used a 22 bit, 24 bit or 26 bit coding. According to a communication theory the coding of the words with a 22-26 level alphabet looked to be designed with an ugly sub-optimized and “catastrophic” coding system. There had to be a misunderstood mechanism, which is hidden in the background. From the Sefer Yetzirah I learned the human voice had to be interpreted as an optimized 5-channel instrument, each of which channel had been equipped with a 3 bit code or “ternary code”. In fact the theoretical optimum for the levels in communication channels requires e = 2.7 levels for each channel. This value is named Euler's number, e ≈ 2.71828, the base of the natural logarithm. Ternary is the integer base with the lowest radix economy, followed closely by binary and quaternary. This is due to its proximity to the mathematical constant e. It has been used for some computing systems because of this efficiency. It is also used to represent three-option trees, such as phone menu systems, which allow a simple path to any branch. 3 The 5 channels are the places of articulation, named the tongue, the lips, the teeth, the palatal and the vowel space (gutter). In On the Nature of the Gods Cicero also specifies the categories, although he uses different categories for the speech sounds4. 1 The noun dyaús (when used without the pitṛṛ 'father') refers to the daylight sky (https://en.wikipedia.org/wiki/Dyaus). 2 On Ternary Coding and Three-Valued Logic (from the author Subhash Kak ) 3 https://en.wikipedia.org/wiki/Ternary_numeral_system 4 Notes to Cicero’s Tusculan Disputations, On the Nature of the Gods, On the Commonwealth In fact the human voice is a communication system with 5 independent channels, represented by the 18 (active and passive) places of articulation. 1. Exo-labial, 2. Endo-labial, 3. Dental, 4. Alveolar, 5. Post-alveolar, 6. Pre-palatal, 7. Palatal, 8. Velar, 9. Uvular, 10. Pharyngeal, 11. Glottal, 12. Epiglottal, 13. Radical, 14. Postero-dorsal, 15. Antero-dorsal, 16. Laminal, 17. Apical, 18. Sub- apical. Not all places of articulation are independent. The tongue may be activated for each speech sounds and may always be involved in speaking. Other places of articulation may be distributed in two or three locations, such as 1. Exo-labial, and 2. Endo-labial, or 6. Pre-palatal, 7. Palatal, and 11. Glottal, 12. Epiglottal. Each communication channel is controlled by electric impulses to various muscles, which are activated to generate the speech sounds or phones. The controlling is quite sophisticated and is to be learned in our early phases of the human childhood. The Sanskrit alphabet There might be more than 5 places of articulation. Some languages such as the Old Persian cuneiform alphabet, also include glottal, nasal, sibilants and semivowels. The Sanskrit alphabet was sorted according to the points of articulation (lingual, palatal, guttural, labial, dental), but the 14 vowels (A, Ā, I, Ī, U, Ū, Ṛ, Ṝ, Ḷ, Ḹ, E, AI, O, AU) were more abundant than the Latin A-E-I-O-U set, and are distributed over all 5 categories (and points of articulation).5 The 14 vowels are identified in the second column: The Sanskrit language seem to restrict the vowels and consonants in gutturals, palatals, cerebrals, dentals and labials. Cerebrals (“near or from the brain”) are equivalent to (“Of pronunciation in which the tip of the tongue is raised and bent backwards, so that the underside of the tongue approaches or touches the palate”. The specified vowels (A for gutturals, I for palatals, U for labials) are okay, but some (N and L) of the dentals and others (R, Sh) in the cerebrals seemed to be have exchanged their categories. Also some gutturals (ka, gh) may have missed their correct categories. 5 The Architecture and History of the Eurasian Alphabets Fig. 2 The Sanskrit alphabet (Source: Practical grammar of the Sanskrit language (1864.), by Monier Williams, M.A.) The Old Persian cuneiform alphabet Similar errors seemed to be found in the categories of the Old Persian cuneiform alphabet. The palatals, labials and gutturals seemed to be equivalent of the Sanskrit categories. The Alveolar symbols seemed to be linguals. The dentals were restricted to the R and S. Alveolar consonants are articulated with the tongue against or close to the superior alveolar ridge, which is called that because it contains the alveoli (the sockets) of the upper teeth6. The categorization in the Old Persian cuneiform alphabet defines the triads 1,2,3,4,5,6,7 and additionally 2 nasals N & M , the L and a glottal H: Triad 1 Triad 2 Triad 3 Triad 4 Triad 5 Triad 6 Triad 7 Vowels (Ā), Ī, Ū Velar Guttural K, X, G Palatal C, Ç, J Alveolar T, Θ, D Labial P, F, B Nasal N & M Semi- vowels Y, V, R L Sibilant S, Z, Ś Glottal 1 2 3 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 k- x- g- c- ç- j- t- θ- d- p- f- b- n- m- y- v- r- l- s- z- š- h- K X G C Ç J T Θ D P F B N M Y V R L S Z Š H X C Ç Θ P F B Y L S Z Š H K- G- J- T- D- N- uploads/Geographie/ once-upon-a-day-the-word-dingir-dimer-arose.pdf

  • 12
  • 0
  • 0
Afficher les détails des licences
Licence et utilisation
Gratuit pour un usage personnel Attribution requise
Partager