Can anybody please explain to me why the Old Slavonic verb рещи (reshchi) "to say, to command" does not reflect PIE: h₃reǵ (related Terms: нарещи (nareshchi), изрещи (izreshchi), пророк (prorok), прорицати (proritsati), нарицати (naritsati), отърицатисѧ (otritsatisya), отърещисѧ (otreshchisya), предърицати (predritsati))?

, where
на/na, из/iz, про/pro, отъ/ot, предъ/pred are the prefixes in the Old Slavonic;

the letter щ (shch) is always derived from the letters к (k) or г (g) + т (t);

the letter ц (ts) is always derived from the letter к (k) when it stands between the letters и (i) and а (a);

  • Do you have a specific semantics in mind or just the sound? Given German rechnen "to calculate" from that root the semantics might fit, analoguously to zählen "to count",erzählen "to tell, recount*, En to tell, count, recount etc. That's what you expected, right? I don't know about the phonetics, though. I'd go as far as considering the prefix er- derived from the same root, comparing to count up, and the root glossed upright, though Ger abzählen would imply ab "down, away*, hence to count down. Now what about rex "king" and op-ressor, (oppressor, excuse the pun). – vectory Apr 30 '19 at 22:15
  • Regarding your question without getting at the heart of it, a) the author of the linked page may be inexperienced, usually glosses and bare stems are given for roots, but that argument doesn't count for the source, Dierksen. The relation may be older than PIE or newer. Go read Dierksen see if he makes an argument against *h3erg-. – vectory Apr 30 '19 at 22:25
  • 2
    @vectory Laryngeals were lost from every single IE language by the thirteenth century BCE when Hittite (which preserved *h₂ in some environments) went extinct. It seems implausible that the ruling class would have kept saying the laryngeals for thirty-three hundred years without ever writing them down, and there's similarly nowhere they could have borrowed from that would retain the laryngeals. (And that's assuming *h₃ would be interpreted as /h/, when its reconstruction is still uncertain…) – Draconis Apr 30 '19 at 23:13
  • 1
    @MichaelStallone It's entirely plausible that *h₃-rg- would end up having a descendant meaning "speak"; compare Latin rogō "to ask" (from something like "to correct yourself", from the original meaning "right, straight"). But assuming the original PIE meaning "speak" makes it very hard to get to Latin meanings like rectus "in a straight line". It's a lot easier to take "right, straight" as the original and go from there. – Draconis Apr 30 '19 at 23:16
  • 4
    @MichaelStallone Sorry, but talking about etymology somewhat requires believing that words can be derived from other words. Citing the Bible isn't quantifiable evidence. – Draconis Apr 30 '19 at 23:28

First, a note: it's often best in etymology to go back to the last form we can be reasonably sure of. As you mention, OCS щ (often written /ť/ because linguists like the Latin alphabet better than the Cyrillic) basically always comes from Proto-Slavic *kt, which sometimes comes from earlier *gt by voicing assimilation. So we can reconstruct the Proto-Slavic form as something like *rekti or *regti.

There's some semantic difficulty in getting from "straight" to "to say", which isn't totally insurmountable: see Latin rogō "to ask" from the same PIE root. Similarly, there's some difficulty in dealing with the PIE "palatovelar" , but Balto-Slavic never satemized fully, and there's plenty of precedent for not satemizing.

So, it seems plausible that it could come from this root. But there's some other difficulty that's harder to get away from. First of all, there's no evidence of *h₃r-ǵ- ever acquiring a meaning of "speak" in Balto-Slavic languages. Instead it's acquired a meaning of "see", like in Latvian redzēt < *regt or Lithuanian regéti. Second, there are a decent number of IE words looking vaguely like *rekti, in widely-spread languages which all have widely different descendants of *h₃r-ǵ-. And absolutely none of them show satemization, which is explainable for Slavic but a bit weird when it happens in all sorts of different satem languages.

So it seems easier to reconstruct a separate root for these, which doesn't have to deal with satemization. Vasmer suggests something like *r-k- plus a suffix.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.