Reconstruction talk:Proto-Indo-European/h₁édti

@CodeCat What's the evidence for the Narten grade? Just Balto-Slavic? Isn't this one of the prime proposed examples of Winter's Law? I'm not sure I find the lengthened grade compelling. —JohnC5 20:27, 9 April 2016 (UTC)

This is what Ringe gives. But you're probably right. On the other hand, is there any evidence of the zero grade? —CodeCat 20:42, 9 April 2016 (UTC)

De Vaan pointed out that the Narten was unnecessary and also that Hittite preserved the original zero-grade. Looking at Kloekhorst sure shows that to be true. —JohnC5 21:15, 9 April 2016 (UTC)

We should probably have both entries as alternative reconstructions then. The question is which one to use as the main entry, presumably the one with the most support. How many sources can you find that support the regular ablaut, versus sources supporting the Narten ablaut? —CodeCat 21:21, 9 April 2016 (UTC)

So far I find:

• Normal ablaut: De Vaan, Derksen, Kroonen
• Narten ablaut: Ringe
• No ablaut (full grade only): Sihler

—CodeCat 21:41, 9 April 2016 (UTC)

Also:

• Normal ablaut: Beekes, Mallory & Adams, Kloekhorst, Matasović (though mentions Narten as possible)
• Narten ablaut: LIV, Fortson
• No ablaut (full grade only):

—JohnC5 21:44, 9 April 2016 (UTC)

Ok, it looks like the sources for regular ablaut are clearly in the majority. —CodeCat 21:52, 9 April 2016 (UTC)

@JohnC5 Could you provide references for the sources you have on both pages? —CodeCat 21:59, 9 April 2016 (UTC)

@CodeCat Yep! In the mean time (and I hate to be a nudge about this), but what evidence is there for *h₂wḗh₁ti? I haven't looked around at sources yet, but it seems like you'd need ablauts to prove the Narten grade. —JohnC5 22:03, 9 April 2016 (UTC)

The participle-turned-noun has full grade. That implies Narten ablaut, since the active eventive participle has zero grade of the stem normally. —CodeCat 22:04, 9 April 2016 (UTC)

Perfect! I should have figured that out, now that you explain. —JohnC5 22:10, 9 April 2016 (UTC)