lineara.eu

A list that does not add up

This is HT 118, a clay tablet from Haghia Triada. A list, like HT 13. This one has a total that does not match its entries, and people have argued about the gap since 1978.

The object

Found in the room the excavators called Casa Room 9. It sits in the Pigorini museum in Rome now, number 83734. Younger gives it to the hand he calls HT Scribe 3.

Photograph of tablet HT 118
Tablet HT 118, the object. Photograph: lineara.xyz mirror. Published here for this document under a permission request. Image rights largely Ecole Francaise d'Athenes.

The tracing

Same rule as before. A person drew the strokes from the object. Every crop below is cut from this drawing, not from a font.

Tracing of HT 118
Tablet HT 118, the tracing. Sign drawings from SigLA: The Signs of Linear A, a palaeographical database, by Ester Salgarella and Simon Castellan (https://sigla.phis.me/), licensed under CC BY-NC-SA 4.0.

The signs, one by one

The first line is a heading: the pig logogram (AB85, SUS in the Linear B convention) and the ligature *516, each with a divider after it. Then names and numbers. Three of the names are followed by a short line of their own: the single sign ki and a smaller number.

  1. Sign AB85 au SUS, HT 118, position 1 AB85 au SUS
  2. Sign A516, HT 118, position 2 A516
  3. Sign AB80 ma, HT 118, position 3 AB80 ma
  4. Sign AB07 di, HT 118, position 4 AB07 di
  5. Sign AB67 ki, HT 118, position 5 AB67 ki
  6. Sign AB16 qa, HT 118, position 6 AB16 qa
  7. Sign AB16 qa, HT 118, position 7 AB16 qa
  8. Sign AB26 ru, HT 118, position 8 AB26 ru
  9. Sign AB67 ki, HT 118, position 9 AB67 ki
  10. Sign AB08 a, HT 118, position 10 AB08 a
  11. Sign AB53 ri, HT 118, position 11 AB53 ri
  12. Sign AB58 su, HT 118, position 12 AB58 su
  13. Sign AB67 ki, HT 118, position 13 AB67 ki
  14. Sign AB53 ri, HT 118, position 14 AB53 ri
  15. Sign AB26 ru, HT 118, position 15 AB26 ru
  16. Sign AB80 ma, HT 118, position 16 AB80 ma
  17. Sign AB81 ku, HT 118, position 17 AB81 ku
  18. Sign AB02 ro, HT 118, position 18 AB02 ro
  19. Sign AB67 ki, HT 118, position 19 AB67 ki
The signs of HT 118 in reading order, each one cropped from the tracing. The label under each crop is the catalogue number and, where one is agreed, the sound value.

The numbers

Four name lines carry 15, 6, 4 and 10. Three ki lines carry 10, 4 and 1. Then ku-ro 30. After the total comes one more ki line, 15 and a damaged sign.

  1. line 1 au_SUS AB85|A516|
  2. line 2 ma AB80di AB0715
  3. line 3 ki AB6710
  4. line 4 qa AB16qa AB16ru AB266
  5. line 5 ki AB674
  6. line 6 a AB08ri AB53su AB584
  7. line 7 ki AB671
  8. line 8 ri AB53ru AB26ma AB8010
  9. line 9 ku AB81ro AB0230
  10. line 10 ki AB6715?
Every line of HT 118 as text. Numerals are added up and shown as digits. A question mark is a damaged sign. A bar is a word divider.

The total

The four name lines add to 35. The ku-ro says 30. Younger's view is that a 5 was left out. Count the ki lines as well and you reach 50, further off still. Our audit calls the total "not terminal", because line 10 carries a number after it. The check below is the audit's own wording.

  1. Sign AB81 ku, HT 118, position 17AB81 ku
  2. Sign AB02 ro, HT 118, position 18AB02 ro

line 9 ku AB81ro AB0230

Arithmetic check: explained. total is not terminal, so the list continues past it: sum of 7 entries 50 vs stated 30, residual 20.

The last line of HT 118: the word ku-ro and a number. The check compares that number with the sum of the entries above it, from the 2026-08-13 audit file.

What everybody agrees on

A list. A heading with the pig sign and *516. Four name-like words, each with a number, and ku-ro 30 on line 9. Schoep's typology files it under specialized commodities.

What is disputed

Whether ki is short for ki-ro, a word Younger's notes list under "owed" (Raison and Pope proposed the link in 1978), or a separate term. Whether the total dropped a 5 (Younger) or covers a different span of lines (our audit). What *516 is, since the mirror record gives no gloss. One data note: our word list for this tablet skips the ki lines, and our sign-by-sign text keeps them. Both are on the record page.

Photograph against tracing

The tracing is somebody's reading of the object. Drag the curtain and judge it yourself.

The comparator needs JavaScript. The layers are the photograph, the facsimile, the tracing and the plate of HT 118, registered into one frame.
Photograph against tracing for HT 118. Drag the curtain. The layers were registered offline; the browser computes no geometry.

Credits

Tracing: SigLA, Ester Salgarella and Simon Castellan, CC BY-NC-SA 4.0. Photograph: as captioned above. Readings and commentary: John G. Younger, Linear A Texts and Inscriptions in phonetic transcription and Commentary (2024), quoted in short. Findspot and museum number: the lineara.xyz record, which draws on GORILA (Godart and Olivier, 1976 to 1985). Typology: Schoep 2002. Arithmetic: our own audit of 2026-08-13.

The full record of HT 118

Four more

The same walk for four other objects from Haghia Triada.