From: despayre@e... Message-Id: <991227130402CC.22737@w...> Date: Mon, 27 Dec 1999 13:04:02 -0500 (EST) To: elfling@egroups.com List-Help: List-Unsubscribe: List-Archive: Reply-To: elfling@egroups.com Subject: Suggestions for a complete Khuzdul grammar MIME-Version: 1.0 Content-Type: text/plain Content-Transfer-Encoding: 7bit I'm currently trying to construct a complete grammar of Khuzdul, extrapolating from the published material. In the course of this attempt, I have stumbled across some points that might be interesting to think about: 1) Considering the well-attested Dwarvish obsession with the number 7, wouldn't it be likely for their language to have a septal in addition to the singular and plural forms? 2) Another thing about numeri: Albert Posch recently proposed inventing a distinct indefinite plural form to cover the problem of the plural forms that tend to show up with or without circumflex over the second vowel. (For example, we can conjecture that Khazâd is the plural form of the radicals [Kh-Z-D], the singular probably being *Khuzd, as Mr Fauskanger thankfully pointed out in his article on Khuzdul on the Eldalamberon page. But in Khazad-dûm, we find the (obviously plural) form without the circumflex. While this may explained by an obscure rule about the employment of the plural in compounds, the distinction of definite and indefinite plural works okay as well and helps solving a few more problems in addition. 3) While I agree with Mr Fauskangers Construct State theory (ibid., loc. cit.), I have decided while compiling the existing material to view the 'u' in Bundushathûr as an element of the Construct State singular (!) form for a three-consonantal root, thus inventing a pattern 1u23u (1,2,3 being the consonants; in this cas B,N,D) for the singular construct state while sticking to the attested 1a2u3 for the (definite) plural. In effect, if the name were to mean 'heads of clouds' (instead of 'head of clouds'), this would yield Khuzdul banud-shatûr instead of bundu-shatûr. 4) In addition, I decided to stick to three singular patterns of equal meaning, those being 1u23 (yielding such words as Rukhs), 1i2a3 (as in Zirak), and 1a2o3 (as in Gathol) for nouns. One- or two-consonantal roots each have their own singular, septal (?), def. pl., ind. pl. and Construct State forms. 5) The problematic noun uzbad, 'lord' I have rendered as a compound made up of the ind. plural noun uz (def.pl. being ûz, as attested in ûl, streams; with loss of circumflex as in Khazâd/Khazad) meaning 'people' (not, as Khazad, restricted to Dwarves) and an adjectiv bad, denoting 'well-meaning', 'beneficial', thus describing the ruler as 'he who means his people well'. Of course this is all incompetent invention and pure hypothesis, but nonetheless I would like to hear your opinions on it. By the way, I'm new here. My name's Roland Mückstein, but I'm commonly called Carneldó. Hope you don't kick me out already because of this first message. ---------------------------------------------------------------- Get your free email from AltaVista at http://altavista.iname.com