A number that does not count: the twelve-stroke Indus sign as a word, with a pre-registered test and its links to the fish sign, the gaur and the Gulf seals
The Indus sign G55, twelve short strokes in three rows of four, is written like a numeral but is not used like one. Other stroke numerals stand right before the thing they count, while G55 tends to close a group of signs. A pre-registered test on 1,225 texts not used to find the pattern confirmed this: G55 is followed by a closing sign 55% of the time, against 20% for the other stacked numerals (p = 0.033). Exploratory analyses place G55 in a cluster: A fish sign stands right before it (30% in ICIT, 21% in Mahadevan's corpus, against 3% for other numerals). It goes with the gaur on seals, in both corpora. It appears on 3 of 7 round Gulf Type seals found on Bahrain and on none of 14 found elsewhere abroad. It shares texts with the "man" sign on the Gujarat coast and in the Gulf. G55 may be a word for an institution, office or calendar term, such as "the twelve star-months", rather than a quantity. Three predictions for new finds were sealed in advance. Code and data: github.com/RN-Top/Indus-Research-
Authors
- Erin Rae Toppe
Publication Details
- Journal
- Zenodo (CERN European Organization for Nuclear Research)
- Published
- 2026-10-26
- DOI
- https://doi.org/10.5281/zenodo.23680835
- Primary Topic
- Linguistics and language evolution
- Type
- article
- Field-Weighted Citation Impact
- 0.00