Author: Glen Gordon (http://paleoglot.blogspot.com/)
Email: glengordon01@yahoo.com
Last updated: July 02 2008

The purpose of this pdf is to attempt to outline what grammatical changes, sound changes, and prehistoric events
took place to shape the eventual language known to linguists as Proto-Indo-European (PIE) which has been
dated to approximately 4000 BCE and was probably centered in the region of modernday Ukraine. I don ot
consider this a final theory or the only possible answer to the question of Pre-PIE stages. However, my desire
here is to inform and inspire others to look at this intriguing problem for themselves as well. Linguistics lags
behind another, comparatively more \u201ctangible\u201d science dealing with the past, namely archaeology. By
contributing online and by encouraging hundreds of other busy minds to share their knowledge and ideas as
well, I believe that much progress can be made to finally piece together the linguistic past further back than a
mere 6000 years or so.

Minoan belong. In particular, I believe that a parent language dated to about 7000 BCE at the latest which we
may nameProt o- I ndo- Ae ge an may be reconstructed in the finalmost stage with pronouns (*m\u0259i/*m\u0259/*m\u0259n\u0259
'I/me/my',*tau/*t\u02b7\u0259 'thou/thy',*\u0294\u0259i/*\u0294\u0259 'he/him',*w \u0259i/*m\u0259s 'we/our',*t\u0259i/*t\u0259s 'you/yours'), demonstratives
(*ka 'this',*ta 'that',*h a 'that yonder'), case & number markers (*-(a)m [acc.],*-(a)se [gen.],*-(a)la [dat.],

there, Pre-Proto-Indo-European would have slowly evolved away from the core and settled in the area of the
Danube and Balkans, developing its own idiosyncracies for the next 3000 years. As of the date above, I currently
believe that the following changes have taken place in the stages prior to commonly reconstructed Proto-Indo-
European:

Through agglutination, the case system was first formed from pre-existing postpositions with the vowel*a intervening between consonant-ending noun stems and consonant-initial case suffixes when necessary. Some postpositions such as*bi \u201cby\u201d and*di \u201cin\u201d nonetheless

1)*si >*-(a)si [genitive]
2)*la >*-(a)la [dative]
3)*t\u2019a >*-(a)t\u2019 a [ablative]
4)*\u0294i >*-i [locative]

Due to influence of neighbouring Pre-Proto-Abkhaz-Adyghe to the east, the original 4-vowel
system of*a,*\u0259,*i and*u collapsed into a centralized system of only*a and*\u0259. In open
syllables*i and*u were diphthongized to*\u0259i and*au respectively while in closed syllables,
they became*\u0259 and*a respectively. Note that at this stage, while*i and*u may have
continued to exist on the phonetic level as allophones of diphthongs*\u0259i and*au, they were
now only variants of*y and*w on the phonemic level.

1)*kui >*k\u02b7\u0259i \u201cwho?\u201d
2)*t u >*t \u02b7\u0259 \u201cyou\u201d
3)*k ah un i >*k ah \u02b7an \u0259 \u201cdog\u201d

This is caused simply by the process of lenition and is a sound change common to many other
languages around the world. Word-final*-s is subsequently retained in Proto-IE while
rhotacizing to*-r in Proto-Aegean.

1)*-\u0259t [plural] >*- \u0259 s
2)*-\u0259t [2ps. active] >*- \u0259 s
3)*-at [stative noun] >*-as (remaining*-at - in oblique case forms)
4)*-ak [dual] >*-a\u0294

stative conjugations. Presumably, the active conjugation which used personal endings
exclusively with*\u0259-vocalism (i.e. 1ps*-\u0259m, 2ps*-\u0259s and 3ps*-\u0259) and the stative
conjugation which used personal endings exclusively with*a-vocalism (i.e. 1ps*-ax, 2ps*-

Penultimate Accent Shift (PAS)
Accent shifts such that the former accent fixed on the initial syllable is placed instead on the
penultimate (second-to-last) syllable of all words.

Quasi-Penultimate Accent Rule (QAR)
Word-final Rhotacization (*-n# > *-r#)
Mediofinal Lenition of Glottal Stop (*-\u0294- >*-h-)
Lenition of Ejectives to Creaky Stops (*k\u2019/*t\u2019 >\ue000g\ue001 /\ue000\ue002\ue001)

Schwa Reduction (unaccented*\u0259 >*\u1d4a by default)
Suffix Resistance Exception (*\u0259 remains)
Paradigmatic Resistance Exception (*\u0259 remains)

The optional regressive shift of accent is due to the development of rising tone caused by
voiceless syllable-coda laryngeals which we may presume eventually steals the primary
accent away from the original syllable. Following this change in tone, voicing and omission
of the laryngeal occurs.

(e.g. [*p\u00e1rx\u1d4a'n\u00e0s\u1d4a] > [*'p\u00e1rx\u1d4anas\u1d4a] > [*'p\u00e0r\u1d4anas\u1d4a])
Laryngeal Vocalization (*-h- >*-e-,*-x- >*-a-,*-h\u02b7- >*-w-)
Laryngealization of Geminates (*-ss- >*-z-,*-tt- >\ue000\ue014 \ue002\ue001 \ue014)

This rule causes compensatory lengthening in nouns declined in the nominative singular and verbs conjugated in the 3ps past due to the unexpected syncope of MIE deictics*-sa and*-ta to*-z and\ue000\ue006\ue002\ue001 (via*-z\u1d4a and\ue000\ue006 \ue002\ue001 \u1d4a) despite the Suffix Resistance exception of

1)*b er\ue00b\ue002\ue001 \ue00b \u201che carried (3ps.)\u201d >*b\u0113r \ue002\ue001 (> PIE*b\u0113rt)
2)*g\ue001 \ue00b\ue016 \ue010\ue008 z\u1d4a \u201csister-in-law (nom.sg.)\u201d >*g\ue001 \ue016\ue010:\ue008 -z (> PIE*g\ue001 \ue016\ue017u s)
3)*k \u02b7\u1d4aiz\u1d4a \u201cwho?\u201d /*k\u02b7\u1d4a i\ue002\ue001 \ue00b\ue015\u201cwhat?\u201d >*k \u02b7iz /*k \u02b7i\ue002\ue001 (> PIE*k\u02b7is /*k \u02b7i\ue002\ue001

Compensatory lengthening was also caused in newly created monosyllabic wordforms
which had once been di- or polysyllabic in the Mid IE period through the syncopation of
word-final supershort schwas.