The collective GEAS project called PQN (Paleografia Quantitativa Norditalica) depicts every single graphic shape of the Lepontic/Brembonic/Cisalpine-Celtic, Raetic/Camunic, and Etruscan/Lemnic sub-alphabets…. and the Elder Futhark.
Excerpt from the 40-page long PQN tables with several thousand sign variants in North Italic and Germanic writing. Graphemes which are identical according to the GEAS standards are listed on the same line. From the 125 listed variants of runes, 76 are identical, 49 are similar and none is different.
The decision to include Germanic writing will potentially lead the Rune provenance question to unexplored terrain. Three of the arguments are summarized below:
Evidence No. 1: Low Levenshtein distance demonstrates over-coincidental grapho-phonological congruence
Of the 24 Elder Futhark main types, 21 have identical shapes to graphic variants from North Italic Alphabets, with 13 of them also representing identical sound values, and a further 8 representing similar sound values (i.e. same place of articulation, but different kind of articulation = Phonetic Levenshtein distance of 1 (LDp = 1). Table 2 (from PQN, chapter 3.4) also delivers the combined Levenshtein Distance which cumulates graphical and phonetical similarity (LDg ∧ LDp). The combined Levenshtein Distance between elder Futhark runes and North Italic is so low that it would prove a direct genealogic descendance alone.
First part of Tab. 1 in chapter 3.4.1 of the PQN.
Second part of Tab. 1 in chapter 3.4.1 of the PQN. It lists a total of 44 graphemes, i.e. the 24 main runes and 20 variants that are commonly accepted by runologists. Graphematic and phonetic Levenshtein Distances are given in the columns to the right. The average combined Levenshtein Distance (LDg ∧ LDp) is 4 ∧ 19 ÷ 44 = LDg∧p = 0.52. This is extremely low even if compared to LD’s between writing systems that are related with certainty.
Cumulating the graphematic with the phonetic values, the combined Levenshtein Distance provides an impressive proximity. The Levenshtein Distances given in Tab. 1 (from PQN, chapter 3.4.1) above can be read as follows: Compared to their North Italic predecessors ///, the p-rune has a similar shape (LDg = 1) and an identical sound value (LDp = 0). Thus, its combined Levensthein Distance is 1 (: LDg ∧ LDp = 1). The p-rune is one of only three among the 24 main runes that has no identical predecessor in North Italic alpbabets.
Evidence No. 2: Jaccard Index
Moreover, the PQN shows that the overall Jaccard Index between elder futhark Runes and, for example, Lepontic, is higher than some of the Jaccard Indices among the North Italic Alphabets themselves! This forces us to consider the Germanic runes to be part of the North Italic alphabetic realm. Together with other arguments, we will present an overall rune origin theory at the ISRRI conference in Sankelmark, Germany.
Subset congruence ratio calculated by Jaccard Index. Example: The graphematic correspondence between the Elder Futhark and the Raetic alphabet (J = 0.203) is slightly higher than the one between Etruscan and Raetic (J = 0.200). (From: PQN, Fig. 3).
Evidence No. 3:
The most striking, and actually incontestable evidence for the full grapho-phonological integration of the runes into North Italic alphabets is the presence of so-called “borrowed allpograph groups” – e.g. the parallel transfer of several groups of identical variants with identical or similar phonetic values from one writing system to another. It serves as a probabilistic prove for its own, at least if it includes groups with more than three variants involved. This is the case for the runes: If all existing variants with their (Germanic) readings are considered, several allograph groups have been transferred to Germania: The most extensive one is the k-rune with its variants , , , , , and . All these graphemes are present in North Italic where they represent velar sounds, namely (Cam /k/), (Cam /χ/), (Ven /ks/), (Cam /k/), (Lep /k/; Lep /χ/), (Lep /k/) (Ven /g/), and (Cam /k/). Such a big allograph group makes it mathematically impossible for the sound/shape congruencies to be coincidental. In other words, the PQN table alone is the mathematical proof that the runes and are a subset of the Celtic alphabets.
The paper has been accepted despite its provocative title, «Norditalische und Germanische Runen im Lichte der Drop-in-Entlehnung: Neues aus Archäologie und quantitativer Linguistik».
The forthcoming Runology article that covers parts of the PQN content is to be cited as:
- Mäder, Michael (demnächst): „Der Fura-zara-kara-Merkspruch, der Rahmenquader-Effekt und ein alter Knoblauchkult: Die Übernahme des Futhark aus den norditalischen Alphabetritualen.“ Tagungsband ’Functions of Runic Literacy 200 to 1500 AD. Ninth International Symposium on Runes and Runic Inscriptions (ISRRI), Akademie Sankelmark, Germany, 14-19 June 2022’ (Runrön 26).
The presentation of the Unicode-based “Paleografia Quantitativa Norditalica” at the 2022 Pavia Conference for Celtic Archaeology, announced by GEAS SciDi M. Mäder, at the other hand, has been withdrawn. Unfortunately, private problems (the unexpected death of Mäder’s mother and aunt) have lead to personal and financial troubles. This concerns the Gesellschaft as a whole, because Mäder’s engagement had to be reduced to a minimum. At the same time, we have to regretfully announce the retirement of Sibylle Rogenmoser (office, (c)EndNote data typing). In the mean time, Eva van den BoldDer Fura-zara-kara-Merkspruch will cover all the duties missed by the two, and we hope that Elizabeth Schmutz, (office, (c)EndNote, (c)InkScape Glyph creation) will rejoin us soon!
At any rate, we pre-publish the PQN in the Begleitpapiere Journal (No. 2023/1). For using this most exhaustive collection of graphic variants of the Celtic Alphabets of Northern Italy, feel free and download the GEAS Font collection.