#Sound changes say -mnes that /m/ is a syllabic resonant and should receive an epenthic /u/ if it doesn't have a vowel against it

[xovvo]

English Wordsmithing Pt. 1: From PIE to Proto-Germanic

Ok, so intro post with the basics of PIE is done. Now we can actually get to making these words! Today, I'm feelin' like I want some good words for clergy/pagan priestly roles since pretty much all native English words for them have been Christianized.

There are a few PIE roots we could select (including *weh₂t-, which yielded Latin "vātēs" for an Oracle, prophetess, or seer, Odinn, and several words in English and other Germanic languages meaning "madness", "excitement", "singing", "rage", etc. It's a fun root, but not what I'm looking for right now. Maybe later.), but for this purpose, I want to explore *seh₂k-, which gave us Latin "sacer", "sanctus" and English "sacred" and English/French "saint". It has meanings of making or being holy, as well as making a pact---which is great! Perfect!

A note on Orthography: since my system can't render ḗ correctly, nor é̄, a long vowel bearing accent will be written as e̋.

Now we need to choose endings. I'm going to focus on endings that derive agent nouns from roots or verbs. The first that springs to mind is *-te̋r, which throws the stem into the ø grade. Because it derives nouns from adjectives, I could actually append this not just to the bare *seh₂k- root, but also the various infix-presents, and the factative and causative forms. Unfortunately, I'm not familiar enough yet with PIE word-formation, so I don't have a good idea of what happens when I need to adjust stress/ablaut on more than one syllable.

Wiktionary also claims that *-lós is also a suffix that derives agent nouns from roots/verbal nouns. Great, add it to the pile.

Lastly, I want to explore is actually three endings connected by ablaut. We have *-mén(s) > *-mën, *-mon(s) > *-mō, and *-mn̥. I'm tempted to think *-mn̥ is the original ending since *-mō is its collective/plural, and that was a common path for new words to get coined in PIE (and it's how we got the feminine gender in Post-Anatolian PIE!) and *-me̋n created in analogy, but I don't have data for that. It should be noted that *-mn̥ is neuter and both *-mō and *-me̋n are masculine. Also technically only *-me̋n and *-mō create agent nouns---but in Proto-Germanic two of the endings collapse together, and by the time we get to Old English, they're all the same ending.

as a bonus, I'm also including *seh₂klōys, which only has descendents in Anatolian languages and meant something like ‘custom, customary behavior, rule, law, requirement; rite, ceremony; privilege, right’, according to Dr. Kloekhorst, in his Etymological Dictionary of the Hittite Inherited Lexicon which is great and I love it.

So, our candidate words are:

*sh₂kme̋n ~*séh₂kmō ~ *séh₂kmn̥

*sh₂klōs

*sh₂kte̋r

*seh₂klōys, for funsies

Now, the forms for the genitive and the plurals were distinct, so I'll be listing these words in tables with these four forms of the word (Nominative Singular, Nominative Plural, Genitive Singular, Genitive Plural). Because the oblique cases merge so fast (to the point where we go from PIE's fulll Nominative-Accusative-Genitive-Vocative-Ablative-Allative-Dative-Locative-Instrumental system to Old English's Nominative-Accusative-Genitive-Dative system that was already just a Nominative-Genitive system except for a few rare forms.) and I'm currently not looking to make new words out of oblique forms, we're good leaving them off.

So! our initial tables are:

Image captions coming once I figure out how to trick screanreaders into pronouncing IPA

So, for the purposes of this post, I'll be following a roughly chronological order for the sound changes, but if you're following along at home, the chronological summary of sound changes can be found on page 152 of A Linguistic History of English, Volume I: From Proto-Indo-European to Proto-Germanic by Don Ringe, but it's recommended that you definitely follow along by reading from section 3.2.1.

Now, Immediately, several sound changes are relevant to our words:

Syllabic resonants prepend an epenthic "u"

Word-final bimoric ("long") -ō lengthens to trimoric ("overlong") -ô

Word-initial laryngeals are dropped before consonants, laryngeals that precede a vowel color /e/ and /ē/ and are dropped, laryngeals that follow vowels lengthen them and color /e/ and /ē/ and are dropped, AND because neither Cogwill's nor Osthoff's law apply here, laryngeals between consonants are replaced by epenthic "ǝ"

Giving us:

Once again, this image uncaptioned until I can make a reader read it comprehensibly, I'm so sorry

At this point, the biggest changes we need to handle are Grimm's and Verner's Laws.

Grimm's Law shifts the "voiceless" series of consonants to voiceless fricatives ([ p, t, k ] > [ f, þ, x(orthographic "h") ]), "voiced" obstruents to voiceless obstruents ([ b, d, g ] > [ p, t, k ]), and "voiced-aspirated" obstruents to voiced obstruents (which also had voiced fricative allophones in many positions; [ bʰ, dʰ, gʰ ] > [ b, d, g ]). Now, clusters of obstruents block the shift such that only the first obstruent shifts. Which means for our purposed, only one consonant---the final "k" in the root is affected and nothing else.

Verner's Law and is more complicated. To quote Dr. Ringe in Section 3.2.4: """ After the PIE voiceless stops had become voiceless fricatives by Grimm’s Law, they became voiced by Verner’s Law if they were not word-initial and not adjacent to a voiceless sound and the last preceding syllable nucleus was unaccented; *s was also affected, and became voiced *z under the same conditions """

Also, really only affecting the genitive singular of *-ós: *-ósyo, is apocope, wwhich actually ends up chopped back to *-ós.

So now, at this crossroads we have:

At this point, stress moves to the initial syllable (so I will no longer be marking stress) which strengthens ǝ > a, and then two sequences of changes happen at the same time:

m > n at the end of words, then Vn > V̨ word-final /n/ is lost while nasalizing the preceeding vowel, and then ę̄ > ą̄

unstressed /e/ > /o/ before wC, unstressed /e/ > /i/ everywhere else

after this, the next two big changes happen before the Late contraction of vowels in hiatus wraps everything up:

ji > i, kicking off the general loss of j between vowels except the environments *ijV > *ijV and ǝjV > *jV

After stress moved to the initial syllable the low rounded back vowels unrounded: [ o, ō, ô ] > [ a, ā, â ], then after ę̄ > ą̄ and VjV > VV, the long low vowels re-rounded, regardless of nasalization: [ ā, â ] > [ ō, ô ]

The contraction of vowels in hiatus wraps everything up. For the most part, the contraction meant /o/ and /a/ got lengthened, capping at trimoric length.

So, our words are now in their Final Proto-Germanic state:

It's almost comprehensible to screen readers!

Right?

Well, not quite. Because we have to take into account he morphological changes that were made as native speakers remodeled declensions and shit to suit how they interpreted their language to work.

Here, it's just Nom. sing. "saglas" > "saglaz"

And also the leveling of sag- and merger of -mǫ̂ + -mō endings (with light remodelling.) Now we're ready to head into the next post where we cover the Intermediate stages between Proto-Germanic and Old English, with this set:

Soon, I promise

#linguistics #proto indo european #proto germanic #conlanging #conlang #I still feel like I missed something during the o > a change bc that was more remodelling than expected #Sound changes say -mnes that /m/ is a syllabic resonant and should receive an epenthic /u/ if it doesn't have a vowel against it #yet all the endings have -minis or -miniz