Linguistic news

Linguistic news

by Alone Coder

A paper on the Old Pannonian language, "Lingua Pannonica," written with two Moscow co-authors (Mikhail Yastrebov-Pestritsky and Dmitry Rusakov ), was successfully presented at a conference in Gorny and published in a collection. We also prepared a paper on children's counting rhymes for a specialized conference.

I'm currently writing an article on Proto-Greek-Phrygian and a monograph on the order of the breakup of Proto-Iranian (I've managed to identify common Western and common Eastern innovations, but also identified a very early branch of Ossetian). For this book, I spent a long time communicating with speakers of Ormuri and the leading Ormuri specialist in Russia, Artem Trofimov (but he's now busy compiling a new volume of the "Etymological Dictionary of Iranian Languages" ). I tried to contact speakers of Parachi, but without success, and now Afghanistan is cut off from the internet. I worked briefly with Ossetians, but couldn't find a stable partner. Rauf Melikov, author of "Old Persian Inscriptions," shared some literature on the Median language. I was unable to contact the renowned Yagnobi scholar Saifiddin Mirzoyev or obtain his dictionary (I had to use Novak's Yagnobi-Czech dictionary instead). I found a group of students of the Pamir languages, and obtained a couple of books from them. The monograph is mostly organized, except for the tables with examples (they are being polished in Excel). The problem is that I would like to discuss these examples before publication, but I have no one to discuss them with. I would also like to add a large section on phonetic transitions in all the languages discussed, but this is a very complex task.

There are also two unfinished articles: on Celticisms in Proto-Germanic and on Latinisms in Proto-Slavic. Co-authors are needed.

Imagine, for example, a "Proto-Hellenic" (Proto-Greek-Phrygian) dictionary covering some standard set of concepts (e.g., Basic English, AECMA Simplified English, or Oxford-3000 ), linked to a common database of the "central" Proto-Indo-European language. Naturally, the accuracy of the reconstruction will vary for different words, which should be noted and disclosed. This is also evident in the reconstructed Swadesh-207 list. Covering the standard set of concepts is important for viewing the proto-language as a living entity, and not as a random collection of words, as was the case in Porokny, the Leiden series, LIV, NIL, and many other etymological dictionaries. In particular, such a living "Proto-Hellenic" language would help more reliably identify Hellenic onomastics and putative Hellenisms in Linear A and other sources (for example, I am working on this issue with Alexander Solka, although there are other readings of Linear A, for example, Vladimir Livintsev's, based on Italic languages, or Peter G. van Soesbergen's, based on Hurrian).

Similar dictionaries are being compiled in parallel for Proto-Balto-Slavic (last year, a monograph on the basic vocabulary was published and a report on the interim results was presented; approximately 2,000 Proto-Balto-Slavic words have already been collected, with an average of 2.5 words per nest), Proto-Indo-Iranian (see above), and Proto-Celto-Italic. In addition to solving similar problems of filtering onomastics and other records, they can be used to refine the reconstruction of the "central" Proto-Indo-European language and publish a fully coordinated series of etymological dictionaries, built (largely automatically downloaded) from a single database. This database has been compiled since 2004 and, through various experiments, is gradually acquiring its final format.

The main dictionaries (for ancient languages) should be downloaded from four tables (corresponding to the four main branches), approximately 5,000 rows per table, with the following columns for each language:

• form certainty (empty for words recorded in their own script, * for borrowing, ? for reconstruction)
• form (the standard form of a word for a given language - serves as a key for descendant languages, along with meaning )
• oblique form (for dual-stem and ablaut nouns, Nsg|other-, for dual-stem and ablaut verbs, present|infinitive:aorist)
• agreement class (for nouns: m/f/n/mf, etc., for verbs: t/i/ti)
• meaning certainty (empty for words from dictionaries and understandable texts, * for meaning based on context and equal to a sibling/descendant, ? for meaning based on guesswork)
• meaning (second key. One meaning: a word as per Oxford-3000 or other (verbs without "to") - maybe an expression. "Turn oneself," "burn oneself" is not necessary - this information is in the agreement class.
• evidence (original fixation [source], are there more, the same or different; fixation of meaning [source]; if there is an oblique form, then also fixations; if unusual phonetics, then explain (indicate pronunciation here if the language has poor spelling); if borrowed, then indicate from which word in which language. Declension/conjugation class can be written here, but is not required.)
• precision for parent node (number of non-trivial morphological + semantic changes) with + if the simplest affix(es) is added (for nouns, diminutive (-k-, -l-), collective (-w-), comparative degree, or pronominal form; for verbs, the participle suffixes (-m-, -t-, -i-, -n-, -l-), deverbative (-t-, -ak-), single action (-n-), iterative (-sk'-), or reflexivity, reduplication, or a second stem is appended, or the agreement class is changed) (not published, used to automatically find the best reflection of PIE words in a branch)
• other data (other information not for publication, may be an unsorted collection of different versions and other forms)

Dictionaries are being built for various use cases: etymological, reference and index, and translation. The composition of the main volumes has been determined (including the composition of the indexes for convenient work with these indexes). The volume sizes have already been roughly calculated to ensure that each volume fits securely under a single cover and thus does not hinder use. Calculations were based on the known vocabulary of ancient languages, and through experiments with limited sections of vocabulary (Swadesh lists, combined 476-word lists, individual initial phonemes, and lists of individual entries).

Depending on the working group's capabilities, the series should contain 19 to 30 single-volume dictionaries: etymological, reference-index, and translation dictionaries for PIE, PBS, PII, PKI, and the proposed "Proto-Hellenic" language; a number of dictionaries for individual ancient languages, considered both as living and as parts of a common tree (Old Iranian, Continental Celtic, Proto-Baltic, and Proto-Slavic, taking into account the earliest records); additional volumes may contain etymological dictionaries for individual major languages (Sanskrit, Church Slavonic, Lithuanian, Tajik, etc.), linked to a common database through the (form, meaning) of the corresponding parent language. The completeness and consistency of the dictionaries will be verified using a series of specialized scripts.

Thanks to automated downloading, the dictionaries can be periodically republished with additions and corrections.

Such dictionaries would be useful to a wide range of linguists and historians. We invite everyone to join this project.