Re: [tied] Day and dies, deus and theos

----- Original Message -----

From: Håkan Lindgren

To: Cybalist

Sent: Saturday, January 27, 2001 11:05 PM

Subject: [tied] Day and dies, deus and theos

>When I studied Latin, I took it for granted that these words were related, but, like Piotr said, their origin is different. The OED says that the 'OTeut' (this is what we would call Proto-Germanic if I recall correctly, the OED terminology was recently mentioned on this list) form of 'day' was '*dago-z'. This word "is usually referred to an Aryan (Indo-European?) verb" 'dhagh-' "to burn". This verb also lies behind Old Prussian 'dagis' "summer".

 

There are some problems here: the "burn" root is really *dHegWH- (cf. Greek tephra 'ashes' < *dHegWH-ro-), and *dHogWHos (with its o-grade) should have given PGmc *dagwaz rather than *dagaz; secondly, the semantic connection between "day" and "burn, heat" (also "ashes, tar") is a bit strained. Many linguists prefer to link Germanic *dagas to Old Indic ahar (Gen. ahnas) 'day' < h2ag^H-r, *h2ag^H(e)n-. The initial *d in Germanic has been explained as "stolen" from a preceding demonstrative pronoun (*tod h2ag^Hr 'that day' > *..d-h2ag^Hr > *dHagHr > Germanic *dag-).

>Latin 'dies', on the other hand, is connected to the PIE 'sky' and 'god' words. According to Alois Walde (Lateinisches etym. Wörterbuch, Heidelberg 1930) the PIE form was '*d(i)e:us', "sky", "day". The next pair of false friends, deus-theos, made me wonder what 'day' looked like in Greek; I looked it up and found 'he:mera'. Getting into a little deeper water: if 'day' and 'dies' did come from the same PIE word, would they look as similar as they do now? Not all PIE words with d- in Latin have d- also in Germanic. Compare 'decem' - 'zehn'.

 

We normally have Latin initial d- : English--Dutch--Scandinavian t- : High German z- from PIE *d via regular development (word-internal developments are slightly different in High German). PIE *d > PGmc *t is part of "Grimm's Law":

 

duo -- two -- zwei

decem -- ten -- zehn

dens -- tooth -- Zahn

OLat dingua -- tongue -- Zunge

domo: -- tame -- zähmen

duco: -- tug -- ziehen

 

You are right that whenever we have Latin d- : English d-, true cognacy is ruled out (cf. diabolus : devil; they come from a common Greek source but are loans both in Latin and in English).

>'Theós' is "nicht sicher erklärt" says Friske's Griechisches etym. Wörterbuch (Heidelberg 1970). But it cannot be connected with 'deus' - such a connection is "lautlich umöglich", phonetically impossible. I guess this means that some phonetical law is operating here, but he doesn't say which one. He connects 'theós' with Armenian 'di-k' "gods" and a reconstructed PIE word '*dhe:s-es'. This PIE word lies behind some Latin words as well, but these words do not look anything like 'deus'; two of his examples are 'fanum', "temple" and 'feriae', "holidays". Perhaps '*dh-' gives Latin 'f-'?

 

It does, via a Proto-Italic voiced dental fricative. The commonly accepted etymology of theos derives it from PIE *dHeh1s- (*dHe:s-)/*dHh1so- (*dH@...); the plural *dHeh1ses underlies Armenian dikh, and Latin fa:num < *fasnom < *dHh1s-no-m. I wonder, however, if a connection with *dHwes- 'breathe; breath; ghost, spirit' could not be defended. If so, theos would be derived from *dHwes-o-s and related to English deer and Germman Tier < *dHeus-o-m 'breathing, i.e. living thing; animal' and dizzy *dHus-ik-ó- 'breathing hard'. But it's a private idea on which I don't insist.

When I looked up '*dhagh' in Pokorny's etymology I found some examples that gave me this idea. There's no '*dhagh' in Pokorny, he lists the reconstruction as '*dheguh-' and connects it to a lot of Latin words beginning with 'f-': 'foveo', "to warm", 'fo:culum', "Feuerpfanne" (I guess this last word is connected to French 'feu'). And a newbie question - what exactly is the sound '*dh-'? Is this the same sound as in English 'there'?

 

No, it was certainly a stop, not a fricative, in PIE. The precise articulation cannot be reconstructed with absolute confidence; the traditional notation *dH suggests a "breathy" (a.k.a. "voiced aspirated") stop like Indic <dh>. It did produce fricative reflexes e.g. in Italic and Germanic (in some positions). Word-initially, PIE *dH- > Indic dh-, Latin f-, Old Greek th- (aspirated [tH], a fricative [T] in Modern Greek), and d- in most other branches (Germanic, Celtic, Slavic, Iranian, ...)

>'Deus', old Latin 'deivos', comes from PIE '*deiuos', which is said to be an ablaut of  the verb '*deieu(o)-', "shine". It is a part of the names Iuppiter and Diana, and in Greek it appears, not as 'theós', but as Zeus. But just to make matters a little more complicated, there were Greek dialects where Zeus was called Deus (boiotian and "lak." = lakedaimonian?). And we've got the 'dios-kouroi', the twin brothers Castor and Pollux.

 

The sky god's name was *d(i)je:us (Acc. *die:m, Gen. diwós; it was also a common noun meaning 'daylight, clear sky'); the derivative *deiwos developed the general meaning 'god, deity'. The English cognate of the root *djeu- is the first syllable of Tuesday (< OE Ti:wes daeg = Tiu's/Tyr's day).

 
>To round things off, I looked up a third pair of old false friends. 'God' and 'good' look very similar in many Germanic languages; as I suspected, they had nothing to do with each other. 'God', Proto-Germanic '*go:dho-', did not become an important word until it was picked up by Christianity (and changed from neutrum to masculinum). The main gods of the pre-christian pantheon were called '*ansu-z' (in modern Swedish we call them 'asar'); '*go:dho-' doesn't correspond to our conception of 'god', its meaning was closer to the Latin 'numen' says the OED.

You've got something wrong here. Good derives from PGmc *go:da- (of uncertain origin), but God < *gudan, very likely from *g^Hu(h)-tó- 'invoked' (of gods and other supernatural forces), derived from a root that is found mainly in Indo-Iranian and Slavic, but with isolated cognates popping up elsewhere.

 

For those who check etymologies in the OED: OTeut. ("Old Teutonic") means Proto-Germanic, and pre-Teut. should be parsed as (Proto-)Indo-European.

 

Piotr