From owner-freebsd-chat Wed May 22 14:54:54 2002 Delivered-To: freebsd-chat@freebsd.org Received: from postfix2-1.free.fr (postfix2-1.free.fr [213.228.0.9]) by hub.freebsd.org (Postfix) with ESMTP id 46ABB37B415 for ; Wed, 22 May 2002 14:54:49 -0700 (PDT) Received: from bluerondo.a.la.turk (nas-cbv-6-62-147-148-251.dial.proxad.net [62.147.148.251]) by postfix2-1.free.fr (Postfix) with ESMTP id 2D74624D for ; Wed, 22 May 2002 23:54:47 +0200 (CEST) Received: (qmail 1685 invoked by uid 1001); 22 May 2002 21:52:36 -0000 Date: Wed, 22 May 2002 23:52:36 +0200 From: Rahul Siddharthan To: cjc26@cornell.edu Cc: Greg 'groggy' Lehey , Brad Knowles , chat@FreeBSD.ORG Subject: Re: Sanskrit numbers (was: French, Flemish and English (was: cvs commit: src/sys/alpha/alpha clock.c)) Message-ID: <20020522215236.GA1640@lpt.ens.fr> References: <20020522192335.P47352@lpt.ens.fr> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: User-Agent: Mutt/1.3.27i X-Operating-System: FreeBSD 4.6-PRERELEASE i386 Sender: owner-freebsd-chat@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org cjc26@cornell.edu said on May 22, 2002 at 16:46:30: [proto-Indo-European] > There's no written evidence for it, as it was spoken well before the > invention of writing, but any claim that it never really existed would > have to explain where all these sound correspondences come from (and > no, "mere coincidence" doesn't count as an explanation :) I suppose that makes sense. There are a few "Hindu nationalists" in India who try to claim that Sanskrit is the mother of all languages and spread from India westwards, but their scholarship is in general quite shoddy. The fact that Sanskrit and Latin have totally different scripts suggests that they both originated from some earlier language which had either no script, or a very inadequate one... > > How do people arrive at > > "Hoi(H)nos" and "h3ekteh3" (how do you pronounce those "3"s?) in PIE? > > Sorry, those are supposed to be "h"s with a subscript "3", which is kind > of an unusual sound pronounced something like "hw". [example of reconstruction of sounds] > (Not all of the reconstructions are this simple, of course) I'm still skeptical about how far you can really go with such techniques. Sounds and pronunciations change over time, and recordings didn't exist until a hundred years ago (in which time span there have already been significant changes in pronunciation), so extrapolating back 10,000 years seems far-fetched. Sanskrit has mostly been passed on by word of mouth by the priestly classes, so has diverged remarkably little between say the north and the south of India, but there are notable exceptions. The first letter in the word for "knowledge" is pronounced roughly "jn" or "gn" (somewhat as in "lasagna") in the south, but "gy" in the north (so, "jnana" versus "gyana"). The former is probably more accurate. The distinction between two different forms of "sh" -- as in "krishna" (more accurately, "krshna") and "sharma" -- is not altogether clear, and in practice most people don't make a distinction. In living languages such as Hindi, the divergence is much greater, as it is in English (even within England there is a huge regional variation in pronunciation, particularly of vowels). Given all this, and given that we can't confidently say what Sanskrit sounded like as recently as the Vedic period (even the written language then was quite different from the later "classical" Sanskrit of 2000 years ago), or for that matter how the Romans spoke Latin or the ancient Greeks spoke Greek, I don't see how sounds of a proto-Indo-European language (for example, the "hw" you cite above) can be reconstructed at all. > Well, we can tell a little about where they lived and what their culture > was like based on which words we can reconstruct in the protolanguage. > So, for example, we can reconstruct the words "sow", "plow", and "cow", so > we know that they knew about agriculture and raising livestock. We can't > reconstruct the word for "chicken", though, so that suggests that they did > not live any farther east than Persia. Also, we can reconstruct a word > for "metal", but not for "iron", so that suggests they lived sometime > during the Bronze Age. That's pretty interesting, and much more believable than the reconstruction of sounds... but not *entirely* believable. The words for "chicken" or "iron" could have changed for some relatively minor reason -- compare "iron" and "steel" in English, whose distinction is not terribly important in practice. Or, they may not have thought a separate word necessary for "iron". The present-day Hindi/Sanskrit word is "loha" but my Sanskrit dictionary suggests "loha" could mean iron, copper or gold; perhaps it's the generic word for "metal" you're thinking of? Iron certainly existed in vedic times, so the lack of a distinct word doesn't mean much. (There are several distinct words for gold, which perhaps show its importance; I'm not sure about copper.) > I really recommend taking a look at Beeke's book if you're at all > interested in the subject...it's a good introduction to historical > linguistics. You could also try googling for "historical linguistics", > "comparative method", "proto-indo-european", etc. Will do -- actually already did the google thing a couple of hours ago. But thanks for your detailed mail, anyway (it's lucky that nothing is off-topic on -chat...) Rahul To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-chat" in the body of the message