A Century-Old Linguistic Mystery: Has an Amateur Cracked Linear A?

Tom Di Mino, a self-taught expert in AI and an amateur linguist, claims to have solved a riddle that has stumped professional scholars for over 120 years: the decipherment of Linear A, a writing system from the Bronze Age Minoan civilization.

Currently, his findings are undergoing rigorous peer review by specialists at Cambridge and Rutgers.

Note: The author of this piece acknowledges a social acquaintance with Di Mino.

Based in the Hudson Valley, Di Mino began his investigation in January of this year. He reports that his primary breakthrough occurred on May 22. If verified, this discovery would be a literal $\text{earthquake}$ in the field of linguistics. For context, when the related script, Linear B, was finally cracked in 1952, it was a headline story for the New York Times.

The Linguistic Theory: A Semitic Connection

Di Mino proposes that Linear A is not Greek, but rather an extinct Semitic language. He suggests it served as a precursor to biblical Hebrew, mirroring the relationship between Latin and Italian.

While this isn't the first time such a link has been proposed—Cyrus Gordon published a similar theory in Antiquity in 1957—Gordon's work never gained mainstream acceptance because it failed to provide the level of translatable detail that Di Mino's current solution offers.

Evolutionary Path of the Language

Understanding the Scripts: Linear A vs. Linear B

To understand the magnitude of this claim, one must understand the relationship between the two Minoan scripts.

Feature	Linear A	Linear B
Era	~1800 BC to 1450 BC	Post-1450 BC
Users	Minoans	Mycenaean Greeks
Nature	Syllabic + Logograms	Syllabic + Logograms
Shared Signs	60 core syllables	60 core syllables
Unique Signs	13 signs not in Linear B	None (derived from A)
Decipherment	~~Undeciphered~~ (Claimed)	Deciphered (1952)

Both systems utilize logograms (symbols representing entire words, such as cow) rather than just sounds. Linear B was deciphered by Michael Ventris—also an amateur linguist and architect—who built upon the statistical and grammatical groundwork laid by Alice Kober of Brooklyn College.

However, Linear A has proven far more difficult because:

There are significantly fewer surviving inscriptions.
Many existing texts are merely ~~narratives~~ trade inventories, which offer little linguistic context.

The "Key" to the Code

The breakthrough happened on May 22 while Di Mino was studying a specific formulaic prayer.

The Formula: IOZa2 (Iouktas): A-TA-I-*301-WA-JA · JA-DI-KI-TU · JA-SA-SA-RA-ME · U-NA-KA-NA-SI · I-PI-NA-MA · SI-RU-TE · TA-NA-RA-TE-U-TI-NU · I

In this sequence, most words were already "known" via their overlap with Linear B. However, the first word—a verb with varying conjugations—remained a mystery. It contained five known signs and one unique Linear A sign: *301.

Di Mino identified *301 as the sound na, which allowed him to isolate the root: $\text{nawaya} \rightarrow \text{"to dwell"}$

In Semitic languages (like Akkadian or Hebrew), roots are often based on a three-consonant system. Here, the root $\text{N-W-Y}$ relates to dwelling or building a pasture. This realization revealed that the prayer was addressed to a Goddess and shared a structure similar to later Hebrew prayers.

Methodology and Technical Execution

Di Mino didn't rely on intuition alone; he leveraged modern AI to scale his research. He used Claude Code to develop a suite of Python scripts to analyze the digitized corpus from the SigLA and GORILA databases.

# Conceptual logic used by Di Mino
def analyze_corpus(database):
    for inscription in database:
        if "formulaic_prayer" in inscription.type:
            cross_reference(inscription, linear_b_values)
            test_hypothesis(root="N-W-Y")

This systematic approach allowed him to test hypotheses at a scale impossible for manual research. Furthermore, his insights into logograms have reportedly helped resolve lingering translation errors in Linear B, providing a form of internal validation.

Tangible Results

Di Mino's research has produced several concrete artifacts:

Sign Readings: Proposed values for 40 of the 102 signs (including all 13 unique Linear A symbols).
Linear B Updates: Resolved the sound values for 5 previously unknown Linear B signs.
Lexicon: A translated dictionary of 408 Linear A terms.
Academic Draft: A 9-page manuscript titled "Ya Diktu: Grammar of the Minoan Peak Sanctuary Libation Formula".

Figure 1: Analysis of the first line of the prayer inscription (Credit: Tom Di Mino, June 2026).