The Germanic word seems to be from *h2ster-no-, i.e. the same formation I fancy to underlie the Celtic 'iron'-word in its unknown source language.
Why couldn't Celtic have borrowed its 'iron'-word from the Pre-Germanic word for 'star', and then have it get borrowed back into Pre/Proto-Germanic with the meaning of 'iron'? Why invoke a mysterious, unknown IE language here?WeepingElf wrote: ↑Thu Jul 20, 2023 1:42 pmThe Germanic word seems to be from *h2ster-no-, i.e. the same formation I fancy to underlie the Celtic 'iron'-word in its unknown source language.
You have completely missed the point. There is no way *h2ster-no- could give *îsarnon in Celtic, or in any other known IE language - none of them has the sound change rules that would be required for this. Hence my speculation about an unknown language.Travis B. wrote: ↑Thu Jul 20, 2023 2:15 pmWhy couldn't Celtic have borrowed its 'iron'-word from the Pre-Germanic word for 'star', and then have it get borrowed back into Pre/Proto-Germanic with the meaning of 'iron'? Why invoke a mysterious, unknown IE language here?WeepingElf wrote: ↑Thu Jul 20, 2023 1:42 pmThe Germanic word seems to be from *h2ster-no-, i.e. the same formation I fancy to underlie the Celtic 'iron'-word in its unknown source language.
You probably know Celtic better than I do (my knowledge of it is, well, very limited). Yes, it seems that PGmc could not be a source of *isarnon, because this would have required *st > *s, and from reading about it seems that PGmc does not seem to allow for *#h2C to become *iC (*h2 is the "a-coloring laryngeal" after all). We would need something which has *#stV > *#istV > *isV. About PGmc, though, I would not rule out the possibility of our unknown language being Para-Germanic.WeepingElf wrote: ↑Thu Jul 20, 2023 4:06 pmYou have completely missed the point. There is no way *h2ster-no- could give *îsarnon in Celtic, or in any other known IE language - none of them has the sound change rules that would be required for this. Hence my speculation about an unknown language.Travis B. wrote: ↑Thu Jul 20, 2023 2:15 pmWhy couldn't Celtic have borrowed its 'iron'-word from the Pre-Germanic word for 'star', and then have it get borrowed back into Pre/Proto-Germanic with the meaning of 'iron'? Why invoke a mysterious, unknown IE language here?WeepingElf wrote: ↑Thu Jul 20, 2023 1:42 pm
The Germanic word seems to be from *h2ster-no-, i.e. the same formation I fancy to underlie the Celtic 'iron'-word in its unknown source language.
You have to know that sound changes in languages are regular; sounds in words don't change willy-nilly, they follow language-specific rules. It is only by this regularity that we can reconstruct languages like Proto-Celtic, Proto-Germanic or Proto-Indo-European. Before linguists had developed the notion of regular sound changes in the 19th century, etymology was a pseudoscience no better than astrology; Voltaire famously characterized it as a field "where consonants count for little and vowels for nothing at all".
And probably, this etymological speculation of mine is just bullfrogs, and *îsarnon has nothing to do with *h2ster-no-. It is just an idea to pursue in a conlang, nothing more
Yes, we need a language with the development you give, and this happened neither in Celtic nor in Germanic. My conjecture is that it was a "Southwest IE" language, as I call a hypothetical branch coordinate to Anatolian and associated with the Bell Beaker culture, which would also be the language family reflected in the Old European Hydronymy (if that is not just a spurious, meaningless pattern falling out of the sheer amount of data). We know very little about its phonological developments, but it seems to have merged *o with *a, and may have lengthened short vowels in stressed open syllables (which may explain why the word for 'linen' has an unexpected long *î in Germanic, Celtic and Italic). The development *#stV > *#istV is similar to what happened in Western Romance and in British Celtic, so this may have happened in some Southwest IE languages as well.Travis B. wrote: ↑Thu Jul 20, 2023 7:13 pm You probably know Celtic better than I do (my knowledge of it is, well, very limited). Yes, it seems that PGmc could not be a source of *isarnon, because this would have required *st > *s, and from reading about it seems that PGmc does not seem to allow for *#h2C to become *iC (*h2 is the "a-coloring laryngeal" after all). We would need something which has *#stV > *#istV > *isV. About PGmc, though, I would not rule out the possibility of our unknown language being Para-Germanic.
The thing here is that inherited *sV became /h/ in Greek, with a few exceptions, such as *sw. This implies that sidêros is probably not an inherited word in Greek.WeepingElf wrote: ↑Fri Jul 21, 2023 12:23 pm And then there is another pair of puzzling words: Latin sîdus, gen. sîderis 'asterism' vs. Greek sidêros 'iron'. They look similar to each other but don't really match, but could be from related substratum languages. And they could be imagined to descend from PIE *h2ster- as well. So the source language may be another unknown IE language. But I should perhaps better stop here.
Right. In fact, sidêros has no accepted etymology, and is probably a loanword from some other language. That language may have had a relative in Italy, which Latin may have borrowed sîdus, gen. sîderis from. Alas, we know nothing about that language. The resemblance to PIE *h2ster- is too slight to draw any conclusions from it. (Of course, the resemblance between *îsarnon and *h2ster- is not really better.)Travis B. wrote: ↑Fri Jul 21, 2023 1:04 pmThe thing here is that inherited *sV became /h/ in Greek, with a few exceptions, such as *sw. This implies that sidêros is probably not an inherited word in Greek.WeepingElf wrote: ↑Fri Jul 21, 2023 12:23 pm And then there is another pair of puzzling words: Latin sîdus, gen. sîderis 'asterism' vs. Greek sidêros 'iron'. They look similar to each other but don't really match, but could be from related substratum languages. And they could be imagined to descend from PIE *h2ster- as well. So the source language may be another unknown IE language. But I should perhaps better stop here.
I thought Lambuzhao meant that geographically (which would be partly right - the core area is rather Northern Mesopotamia and Syria South of Anatolia, but the kingdom extended into Anatolia). In any case, the Mitanni actually spoke Hurrian, a language that forms its own small language family with Urartian; wider relationships have been speculated about, but not been proven. The Indo-Aryan element (personal names, theonyms, terms related to horse training) seems to have been limited to part of the military elite. It's not even clear whether the Indo-Aryans were the ruling class (à la Norman French speaking monarchs and aristocracy in medieval England) or just a significant faction of the aristocracy.
Sorry, I meant Anatolian with respect to geography.
When Indo-Europeanists speak of "Anatolian languages", they usually mean the particular branch consisting of Hittite, Palaic, Luwian, Lycian, Lydian etc. If they talk about languages of Anatolia in general, they say, "languages of Anatolia".
That's right.WeepingElf wrote: ↑Thu Jul 20, 2023 1:07 pmYesterday, an adventurous etymology idea occurred to me. The word in question is the Celtic word (which was also borrowed into Germanic) for 'iron', *îsarnon. I have seen attempts to derive it from PIE *h1esh2er/n- 'blood' (reconstructed from Anatolian and Tocharian data), perhaps due to the reddish colour of rust (iron as the 'bleeding metal'?).
Interestingly enough, Tibetan chak 'iron' has undergone the same semantic drift from 'blood'.That's ལྕགས (lcags) for those playing at home. I have no resources for Tibetan but I do have a copy of Schuessler's etymological dictionary of Old Chinese, which gives 鐵 as being from "black" rather than "blood" (c.f. 驖 "black horse").Talskubilos wrote: ↑Thu Jan 11, 2024 2:36 am Interestingly enough, Tibetan chak 'iron' has undergone the same semantic drift from 'blood'.
