Up to the present time, there has been no thorough attempt to trace the connections between Minoan and Proto Indo-European (PIE); attempts at decipherment have either worked synchronically (from variations on Semitic, Hittite, Luwian or Greek), or diachronically from a later language such as Lycian. This paper attempts to rectify this situation. Experimental readings are given, along with an elementary grammar and sound rules which connect the language to PIE.

Most Linear A syllable signs can be read sensibly with Linear B values. Two values not included before now are L 85 = dja and L 88 = dju. Evidence for L 85 includes the alternation between da-si-85 (HT 13, 85, 99, 122) and da-si-di-ja (HT 126), both referring to the same town, most probably Lasithi (with d/l alternation also attested in Linear B, as in dapuritojo > labyrinth). Evidence for L 88 = dju includes the alternation between Tetu (HT 7, 13, and 85) and Te.88 (HT 8 and 98), both recurring with a form of the name Teki.

L. R. Palmer showed in 1965 (Palmer 1965, 334-5) that the goddess-title Jasasarame made sense if seen as a variant of Hittite Ishasarasmis, my Lady. (The word also appears several times as Jasasara, the Lady, without the enclitic -me.) Besides figuring out this divine title, he also partially translated the libation table inscription KN Za 10: Ta-nu-a-ti ja-sa-sa-ra-ma / na da-wa-a- du-wa-na i-ja

Palmer pointed out that tanuati is the third person singular of the Luwian verb tanu- , meaning, he erects. Jasasarama is dative, paralleling the old Hittite dative ishasarama. Na is the negative imperative in Luwian. The rest of the second line reads as follows:

Do not take (na dawa-a, negative imperative) the offerings (duwana, which as Palmer noted is related to Hittite duwa- or Luwian tuwa- , to put) from here (i-ja, recognized by Palmer as a probable adverb based on the demonstrative pronoun stem i-.

Palmer did not get further, due to an inability to see the connections between Minoan and PIE directly, or between Minoan and other related languages.

In the Minoan Linear A inscription, there are many aids to help the modern reader. First of all, there are loan words from different languages. For instance, the abbreviation "te," commonly used on the tablets, has convincingly been shown by Jan Best and Fred Woudhuizen to be an abbreviation of an expression "telu" which appears to be related to an Assyrian word meaning delivery (Best and Woudhuizen 1988, 24).

It is also possible to tell approximately what should be on many tablets from context - like commodity signs that are shared with Linear B. One example of these is the short tablet HT 35 which reads, ti-ti-ku GRAIN I-ku-su. Ikusu is known from Philistine times to be a personal name. Amos 7:9 and other verses of scripture note that the Philistines came from Crete. This leaves ti-ti-ku as a form of grain that was issued to Ikusu. There are also other helpful words such as ru-ja (pomegranates - KN W 26), ka-pa (olives - HT 6 etc.) and ma-lu (wool - HT 117) which are shared with Mycenaean or later Greek.

A few sample readings that show the nature of the language follow:

PK Za16 (the corner of a libation table):	to-sa pu2-re-ja
So many (tosa, cf Greek tossos) offerings or things brought
(pu2reja, from the zero grade of PIE *bher- which would yield an o in Greek phoros).

PR Za1 (on a libation table):
Ta-na-su te-ke Se-to-i-ja A-sa-sa-ra-me
Tanasu established (cf. Greek theke) (this table) at Setoia (Sitia), o my Lady.

PK Za11, an incomplete inscription:
A-ta-no-dju-wa-e a-di-ki-te-te [..]
O, Sun Goddess (in the vocative, cf. Luwian astanus=sun, djuwaja = PIE *deiuih2),
you were wronged (cf. adikeomai, although the ending is more Hittite (-tet))
 
Pi-te-ri a-ko-a-ne 
You are persuaded (cf. peitho but once again with a more Hittite ending) by the assembly (cf. agon)

A-sa-sa-ra-me u-na-ru-ka-na-ti
My Lady appears in a dream (cf. onar = dream, gignomai=to be born or appear)

Atana potnia appears to be the Mycenaean attempt to translate Atanodjuwaja, or Athena.   

A Minoan description as osuqare, appears related to her festival, Skira,
or the vine branches or oschai which were waved by young men during that festival.

A stone ladle from Troullos, TL Za1, reads in part as follows:

A-ta-no-dju-wa-ja o-su-qa-re Ja-sa-sa-ra-me u-na-ka [faded] na-ma si-ru[...]
The sun goddess [of the Skira], the Lady [appeared in] a dream....

The last part of this inscription would appear to be Ipinama sirute,
or the One of strong name (Iphi-nama) was dragged away (if sirute comes from
the same root as seira, cord to drag away an enemy).  This and other inscriptions
place Atanodjuwaja in parallel with Jasasara; Athena is "the Lady."

A further inscription below the rim of a pithos or large storage jar (ZA Zb 3) reads:

Di-di-ka-se a-sa-mu-ne a-se / a-ta-no-dju-de-ka a-re-pi-re-na ti-ti-ku
The annointed (arepirena, cf Gr. aleiphar) sun goddess of the earth
(atano -djudeka(n?), parallel to Hittite taknas estan) judged (cf Greek dikazo)
the grain (titiku) without mark (asamune, cf Gr.asemos) [and] good (ase, cf Hittite assus).
The -e endings on the adjectives asamune and ase show that these are plural,
accusative in this case.  Ti-ti-ku is hiding a plural ending, most likely (i).

Didikase, in technical terms, is a reduplicated s-aorist, familiar to people who have
studied ancient Greek.  There is further evidence for an s-aorist in the participle Amawasi
in the next example, CR Zf 1 (on a gold pin), related to the Greek verb amao (per personal
correspondence from Jean Faucounau) and meaning for the Harvester (in Greek, an epithet
for Demeter).

A-ma-wa-si ka-ni-ja-mi i-ja qa-ki-se-nu-ti a-ta-de

For the Harvester I kill (kaino, in Greek) here / [the pin] will stick into (pegnumi, Greek)
this (ata with allative -de).  This pin was most certainly used during the ceremony of an animal
sacrifice.  It is interesting to note in connection with the Harvest-goddess that two axes
(AR Zf1 and AR Zf2 are inscribed I-da ma-te - mother Ida, the earliest form of Demeter.  

This goddess Ida is also mentioned in inscription KO Za1:

A-ta-no-dju-wa-ja tu-ru-sa du-ra2-re I-da-a / u-na-ka-na-si I-pi-na-ma si-ru-te
The sun goddess, distressed (cf. Greek truo), lamented (Greek duromai, past tense)
and Ida, a dream appears to you; the one of strong name was dragged away.

The last few examples I have come from the Minoan pictographic script, also known as
Cretan Hieroglyphic.  Because this script is hundreds of years older even than Linear A,
it is often difficult to tell the relationship between the characters in Minoan Pictographic
and Linear A.  However, certain signs such as ro . po , ja , pa3 , and a are clearly related
to their descendants.  Others become apparent in context.

HM 1402 is a seal which reads ka-lo-pa3 - kallopia (cf. Greek kallopizo, make up the face)
 - makeup jar.  HT 31 contains a list of vases, among which are included ka-ro-pa3 as well,  

KN He (06) 03 is a medallion with two faces inscribed and two numbers.  One of the faces reads:
Pa-me-ni po-lo   100
The foals (polo(i)) for this year (pameni has the dative ending, but cf. Greek pammenos)
Please also note that the  pictogram for pa looks like a house (cf. Minoan pa-na-so,
place of the house/temple)

Although po-lo and ka-lo-pa3 are both plural as shown by context, the plural ending
is not shown.  This is consistent with the ending being -I  as in Mycenaean Greek,
where the ending of the -oi or -ai diphthong is not written.  Other endings would
have been written.

The following sound change conventions show the relationship of Minoan to PIE, Hittite / Luwian,  and Greek:

   a.  The PIE o-colored laryngeal h3 develops into "u" rather than "o" - as in the
totaling word ku-lo (Greek holon) which occurs on too many Minoan tablets to list.
This also occurs in such words as u-na-ka-na-si (Greek oneiros + (gi)gnasi) -part of
the Libation formula, on KO Z1, and in other spellings on PK Z11 and PK Z12.
A further example is the word pu2-re-ja on PK Za16.  The word for "dream" in PIE
is listed as *h3nrio (Beekes 1995, 268).

  b.  "K" occurs where Greek has an "h" or rough-breathing mark and PIE contains an
e-colored laryngeal h1.  Ku-lo is the best example of this.  Other examples are ki-ro
(for the temple) on HT 1 and elsewhere; the nominative ki-ru appears on MI Z1.
This is opposed to the Greek hiron (temple).  The rough breathing in Greek is caused
by an original e-colored laryngeal, *ish1ros > *iheros > hieros or hiros.  
 
c.  Where Greek or Hittite has an s at the end of a syllable, Minoan appears to do without.
(Atanuwijawa or Atanodjuwaja show *atanu as opposed to Luwian astanus; -ijawa marks 
the first as "city of the sun" much like Hittite Istanuwa.)

d.  Internal laryngeals occurring in a syllable that already has a vowel, are dropped (Asasara as 
opposed to ashasaras/ishasaras).