Language Log

"Augmenting parametric data synthesis with 3D simulation for OCR on Old Turkic runiform inscriptions: A case study of the Kül Tegin inscription", Mehmet Oğuz Derin and Erdem Uçar, Journal of Old Turkic Studies (7/21/24)

Abstract

Optical character recognition for historical scripts like Old Turkic runiform script poses significant challenges due to the need for abundant annotated data and varying writing styles, materials, and degradations. The paper proposes a novel data synthesis pipeline that augments parametric generation with 3D rendering to build realistic and diverse training data for Old Turkic runiform script grapheme classification. Our approach synthesizes distance field variations of graphemes, applies parametric randomization, and renders them in simulated 3D scenes with varying textures, lighting, and environments. We train a Vision Transformer model on the synthesized data and evaluate its performance on the Kül Tegin inscription photographs. Experimental results demonstrate the effectiveness of our approach, with the model achieving high accuracy without seeing any real-world data during training. We finally discuss avenues for future research. Our work provides a promising direction to overcome data scarcity in Old Turkic runiform script.

[This is a guest post by Brian Pellar]

. . . the consonants are the letters or ciphers which assemble around the vowels to form the words, just as the constellations assemble around the Sun, image of the Divinity, and compose the community of stars over which it presides.                                                                        — Hebreu Primiti

The Consonants of Command

Dear Professor Mair,

In regard to your question, “Is there some sense in which we could think of the 12 aspects/signs/symbols of the alphazodiac as comprising/encompassing the basic sounds of the universe?” I’ve dabbled a bit with some intriguing answers in my papers. For instance, in my very first paper, SPP 196, I placed in the endnotes a very interesting reference from the Gospel of the Egyptians (a Nag Hammadi text) that I feel might bear a relationship to the structure and the underlying “sacred” vowels that comprise the Logos/Word — the breath of God — of the alphazodiac. More specifically,

the “three powers” (the Father, Mother, and Son) give praise to the unnamable Spirit — and the “hidden invisible mystery” that came forth is composed of seven sacred vowels (i.e., the Son “brings forth from the bosom/the seven powers of the great / light of the seven voices, and the word/[is] their completion”), with each of those seven vowels repeated exactly twenty-two times (“iiiiiiiiiiiiiiiiiii[iii]/ ēēēēēēēēēēēēēēēēēēēēēē /oooooooooooooooooooooo/uu[uuu] uuuuuuuuuuuuuuuuu/eeeeeeeeeeeeeeeeeeeeee/aaaaaaa[aaaa]aaaaaaaaaaa/ ōōōōōōōōōōōō ōōōōōōōōōō”) (Robinson 1990: 209–210). [SPP 196, pp. 38-39].

Tangutologists and Old Chinese aficionados take note: the whole 2024 Festschrift for M.V. Sofronov, produced on the occasion of his 90th birthday in 2019, is now available via C. Harbsmeier's academia page: https://t.co/hGUhgrgauE pic.twitter.com/zH7zxY0YjG

— Wolfgang Behr (@behrwolf) June 27, 2024

Sino-Platonic Papers is pleased to announce the publication of its three-hundred-and-forty-ninth issue — Benjamin Porteous, "Reading Genesis 22 and Analects 18 in Late Antiquity":

ABSTRACT: This paper compares modes of scriptural interpretation from two ends of the Eurasian landmass in the late antique period (400–600 CE). Juxtaposing midrashim on Genesis with the Lunyi yishu 論語義疏, a famous expository commentary on the Confucian Analects, the paper argues that the difference between late-antique Confucian and Jewish commentarial practice lies in differing senses of responsibility for the sacred text. The Lunyu yishu curates the full Analects text, while midrashim presuppose a reader who turns elsewhere for the full version of the Hebrew Bible. The paper provides full typologies of commentarial technique in the midrashim and the Lunyu yishu; this is designed to assist comparison and further understanding of the practice of medieval Chinese commentary.

Inscribed sandstone known as the "Singapore Stone", Singapore, 10th–14th century:

Collection of the National Museum of Singapore

(Source; also includes an animated photo that can be rotated 360º in any direction and enlarged or reduced to any size)

A guest post for The Digital Orientalist (4/10/24), under The Magic of Philology and Indexing, Polyglot Asian Medicines (Foundational Resources and Digital Tools), by Michael Stanley-Baker, Christopher S.G. Khoo and Faizah Zakariah (all three are based at academic institutions in Singapore), "Tracking Drug Names Across Language, Time, Space and Knowledge Domains to Produce New Visions of Traditional Medicine".

This is a richly detailed article with many links and citations.  I will not attempt to cover, much less extensively quote, lengthy portions.  Instead, I will begin with the authors' general introduction, note the main sections of the article, refer to the graphs, and quote one typical section to show what the authors' approach can accomplish.

From Tlacuilolli*, the blog about Mesoamerican writing systems, by Alonso Zamora, on March 21, 2024:

*At the top left of the home page of this blog, there is a tiny seated figure (click to embiggen) with a sharp instrument held vertically in his right hand carving a glyph on a square block held in his left hand.  Emitting from his mouth is a blue, cloud-like puff.  Does that signify recognition the basis of what he is writing is speech?

"New Aztec Codices Discovered: The Codices of San Andrés Tetepilco"

They are beautiful:

Figure 1. Codices of San Andrés Tetepilco: a) Map of the Founding of San Andrés Tetepilco;
b) Inventory of the Church of San Andrés Tetepilco; c) Tira of San Andrés Tetepilco

Sino-Platonic Papers is pleased to announce the publication of its three-hundred-and-thirty-eighth issue:

“Mapping the Language of Spices: A Corpus-Based, Philological Study on the Words of the Spice Domain,” by Gábor Parti.

ABSTRACT

Most of the existing literature on spices is to be found in the areas of gastronomy, botany, and history. This study instead investigates spices on a linguistic level. It aims to be a comprehensive linguistic account of the items of the spice trade. Because of their attractive aroma and medicinal value, at certain points in history these pieces of dried plant matter have been highly desired, and from early on, they were ideal products for trade. Cultural contact and exchange and the introduction of new cultural items beget situations of language contact and linguistic acculturation. In the case of spices, not only do we have a set of items that traveled around the world, but also a set of names. This language domain is very rich in loanwords and Wanderwörter. In addition, it supplies us with myriad cases in which spice names are innovations. Still more interesting is that examples in English, Arabic, and Chinese—languages that represent major powers in the spice trade at different times—are here compared.

[This is the first of two consecutive posts on things Indian.  After reading them, if someone is prompted to send me material for a third, I'll be happy to make it a trifecta.]

Our entry point to the linguistically compelling topic of today's post is this Nikkei Asia (11/29/23) article by Barkha Shah in its "Tea Leaves" section:

Why it's worth learning ancient Sanskrit in the modern world:

India’s classical language is making a comeback via Telegram and YouTube

The author begins with a brief introduction to the language:

The language had its heyday in ancient India. The Vedas, a collection of poems and hymns, were written in Sanskrit between 1500 and 1200 B.C., along with other literary texts now known as the Upanishads, Granths and Vedangas. But while Sanskrit became the foundation for many (though not all) modern Indian languages, including Hindi, it faded away as a living tongue.

No, I'm not talking about the eye parasite called Loa loa (a filarial nematode), which is also called eyeworm.  I'm talking about an image that gets stuck in your brain the same way an earworm (also called brainworm, sticky music, or stuck song syndrome) gets stuck in your head.  We've talked about earworms a lot on Language Log (see "Selected readings" below for a few examples), but I don't think we've ever mentioned eyeworms before.

No, come to think of it, I did use the word "eyeworm" once before (here), but that was in reference to the ubiquitous subtitles of Chinese films, even those intended for Chinese audiences, which — upon first glance — may strike one as unnecessary excrescences crawling around in the viewer's field of vision, except for the reasons I listed in the cited post, which lead Chinese audiences to prefer or even need them to understand the films they are watching.

Beautiful WSJ OpED (6/22/23) by Gerard Gayou, a seminarian of the archdiocese of Washington, who is studying theology at the Pontifical North American College in Rome:

The Guiding Light of Latin Grammar

The language reminds us of what our words mean and of whom we’re called to be.

—–

Nothing bored me more during the summer of 2008 than the prospect of studying Latin grammar. I needed a foreign language as part of my high-school curriculum, and I was loath to choose a dead one. I opted instead for Mandarin Chinese, an adolescent whim that shaped my young adult life. I continued to learn Mandarin in college before working in mainland China after graduation.

Sino-Platonic Papers is pleased to announce the publication of its three-hundred-and-thirty-first issue:

Bettina Zeisler, “Combinatory Sound Alternations in Proto-, Pre-, and Real Tibetan: The Case of the Word Family *Mra(o) ‘Speak,’ ‘Speaker,’ ‘Human,’ ‘Lord’” (free pdf), Sino-Platonic Papers, 331 (March, 2023), 1-165.

Among many other terms, discusses the Eurasian word for "horse" often mentioned on Language Log (see "Selected readings" below for examples).   Gets into IIr and (P)IE.

ABSTRACT

At least four sound alternations apply in Tibetan and its predecessor(s): regressive metathesis, alternation between nasals and oral stops, jotization, and vowel alternations. All except the first are attested widely among the Tibeto-Burman languages, without there being sound laws in the strict sense. This is a threat for any reconstruction of the proto-language. The first sound alternation also shows that reconstructions based on the complex Tibetan syllable structure are misleading, as this complexity is of only a secondary nature. In combination, the four sound alternations may yield large word families. A particular case is the word family centering on the words for speaking and human beings. It will be argued that these words ultimately go back to a loan from Eastern Iranian.

Meant to send this more than a month ago.

Interesting new journal in Iranian Studies

"Berkeley Working Papers in Middle Iranian Philology is a new open access e-journal hosted by UC Berkeley’s Department of Middle Eastern Languages and Cultures and edited by Adam Benkato and Arash Zeini. It publishes short and longer articles or research reports on the philology and epigraphy of the Middle Iranian languages (Middle Persian, Parthian, Bactrian, Sogdian, Chorasmian, Khotanese). Submitted papers will be reviewed by the editors and published on an ongoing basis. The journal promotes a simple and quick publishing process with collective annual volumes published at the end of each year. The editors encourage scholars working on Middle Persian documents in particular to submit their work."