Schneier on Security

Roger • February 23, 2006 5:46 AM

(Sorry folks, totally off topic, but hey, I find it interesting!)

@antibozo:

Gnoscere is not just similar to the classical Greek verb “to know”, which is gnoskein, the stem is literally a transliteration.
I suggest you consult an etymological dictionary.

I did consult an etymological dictionary, several in fact, and that was where I found “gnoskein” (I don’t speak Greek myself, classical or modern). One dictionary suggested for etymology that (g)noscere was “akin” to gnoskein (rather hedging its bets on what “akin” means), and I couldn’t help but notice that it wasn’t so much “akin” as “identical”:

“gnoscere” is not a transliteration of a Greek word;

But they — the stems, that is — are a transliteration, regardless of ones theories; it is easily demonstrated by simply doing it! Take the stem of gnoskein (gnoske-) and compare to the stem of gnoscere (gnosce-). The only difference is that Greek K has become Roman C. (The Latin alphabet did have a K, which in early Latin was used for many words, but later in most of these the K was turned into C (c.f. Kalendae/Calendae) while the letter K was used almost exclusively for transliterating Greek loan words.)

it is a Latin word descended from the same origin as the Greek word.

The other etymological dictionary offered the PIE theory, like this:
gnoscere “to come to know,” from PIE base *gno- “know.”
I am no expert in the matter but do know enough to understand that the asterisk before “gno-” means “hypothetical reconstruction”. To put it another way, it is unproven that there actually existed any such thing as a PIE word “gno” which meant “know”, but if there did — if we accept that there was some ancestral language from which nearly all European languages were derived — then gnoscere is probably derived from it.

However even if we go this far, that in no way precludes the possibility that it came via Greek! It is well known that there are Greek loan words in Latin (and presumably the other way round, too) — the language even includes letters that were used exclusively for transliterating Greek loan words. Presumably PIE adherents believe that these all had a PIE origin, whether they arrived in Latin via Greek or some other route. (Hmm, interestingly, most of the ones that I know either seem to be related to philosophy or fishing, perhaps because they had to cross the Adriatic, and hence travelled either with seamen or those wealthy enough to travel).

Do you really think Latin had to borrow “to know” from Greek?

No, I think they had their own perfectly good word already, “scire”, and with it a family of related words (discere, sciens, scientia, conscius, praescius, etc.) I think (or rather, hypothesise on the strength of 15 minutes scanning) that scire words originally meant “to know” without any specific nuance, whereas “(g)noscere” words were borrowed from Greek and had a philosophical nuance, but these nuances seem to have reversed much later, perhaps in mediaeval Latin.

The Greek verb you are referring to has infinitive “gignoskein”. I am not familiar with the form “gnoskein”. “gnoo” is another form of this verb.

I don’t speak Greek, Classical or otherwise, I got “gnoskein” from an etymological dictionary. If I understand the rather dense abbreviations correctly (which I quite likely don’t!), “gignoske-” comes from early Attic Greek (east coast and Aegean), while “gnoske-” comes from Doric Greek (west coast, and closer to Italy), the Attic form evolved into a “gnosoma-” form in Koine Greek, and earlier there was also an Archaic GINOSKO form. But at any rate, this is an evolutionary thing and it took various forms in various dialects and at different times.

I am not a linguist, but I was rather under the impression that large chunks of the PIE theory … had been seriously thrown into doubt by genetic drift studies.

I don’t know about “large chunks”; regardless, it should be clear from the English words of Anglo-Saxon descent that GNO is a root in whatever theory you prefer, and that it expresses in many different language families. See also Sanskrit “janati”, for example.

By “large chunks”, I mean the whole PIE skeleton or shape of European languages evolving in the form of a tree, with a common trunk and spreading branches progressively less closely related, and never crossing or twining back over, each only having a single ancestor. It seems that PIE theorists long ago realised that arguing from similar words leads to blind alleys, but they thought that “morphological” features were reliable and stable over very long periods of time. By morphological features it is meant things like a language being inflected, or agglutinative, or possessing a dual as well as singular and plural. Anyway on the basis of this morphological analysis they established these great tree-like patterns of supposed ancestry between different European languages, and hence presumably the people who spoke those languages. But genetic studies showed that the results of those theories were largely wrong, and hence seriously called into question the stable morphology idea and the whole tree picture.

It may not be total rubbish — at the beginning of the present interglacial about 12,000 – 10,000 years ago, parts of Europe which had been buried under ice became inhabitable again, and there was a great wave of migration into the continent. It is quite possible that, as PIE proponents argue, many or most of these people spoke a common language or a few related languages.

However it is unlikely that it proceeded from there in the manner of a tree, but rather in a far more complex and intertwined manner, continually exchanging artifacts, genes and ideas (the latter including language, technology and all the other aspects of culture). More interchange would have occurred between physically closer groups, but some even among the most distant (it doesn’t take a lifetime to cross Europe, and then as some now some folk would have itchy feet). Not so much a tree as an ecosystem.