Part III (final) of a bit of my new conlang

Here's the text again, for reference. This posting covers "ar laudh..."
all the way to the end ;)

-----------------------------------------------------------------------
English:
"And the Lord came down to see the city and tower that the children of
men builded"

Latin-1:
"Ryf elcwe nau shúla ar sthólegh kyn féagh wriruighenaetha ar laudh
hoin ú dhésar, fém ghaun cheivauvij ar nautyadhel ú chjetyl"

Slightly uglier 7-bit ASCII: (used throughout)
"Ryf elcwe nau shuula ar sthoolegh kyn feeagh wriruighenaetha ar laudh
hoin uu dheesar, feem ghaun cheivauvij ar nautyadhel uu chjetyl"

And the pronunciation in X-SAMPA (A period preceeding two vowels means
a diphthong):

"Ryf elcwe nau  shuula ar sthoolegh kyn feeagh wriruighenaetha ar laudh
/4yf ElkwE n.au Su:lA  A4 sTo:lEG   qyn fe:AG  w4i4.uiGEn.AETA A4 l.auD

hoin  uu dheesar, feem ghaun cheivauvij ar nautyadhel uu chjetyl"
h.oin u: De:sAr   fe:m G.Aun xeiv.auviZ A4 n.aucaDEl  u: xZecl/
------------------------------------------------------------------------



* <ar> = definite article. Not sure if this will inflect much yet.
  However see the form <uu> below. There's no indefinite article.

* <laudh> "city" looks like an ergative form, since it ends in <dh>.
  Actually it's not, it's absolute; this might be a problem.

* <hoin> "and", is only used to join two things. Like "both ... and ..."
  but with the "both" missing but implicit.

* <uu> = definite article. It's ungrammatical to say <ar X ar Y>; the
  alternative form <uu> must be used. No inherent grammatical distinction
  here, but I suppose it will very often indicate that the noun it governs
  is genitive? This is half-stolen from Irish, where 'the' is <an> or <na>
  and it's illegal to say <*an ha'ta an fear> "the man's hat" - you must
  say <ha'ta an fear> (yeah, it might be <fhear> or something).

  This is an example of the lang's avoidence of monotonous sequences of
  similar syllables. There's a good few alternate forms scattered around
  for avoiding clashes.


* <feem>: as with <feegh>, except that we have no <-agh> and we have
  <feem> rather than <fee>. <feem> just specifies that were working
  relative to the tense of the clause two levels up (or maybe just "at the
  top" ? - which is more natural I wonder?)


* <ghaun cheivavij> "they were building"
  * <gh-> is added to <aun> since after <feem> or one of its ilk, we are in
    an urú/eclipsis position.

  * Derivation of verb stem: Primitive root /k_h-b-n`/, inflected to
    /'k_hoban`/. This is stressed on the first syllable, unusually for
    primitve verb forms, and as a result oh the aspirate /k_h/. Hence the
    split into main verb and the auxiliary atavism occurs the other way
    around: /k_hob/+/an`/, rather than */k_ho/+/ban`/. The unstessed syllable
    becomes the atavism, so we have ModLang form <[aun] cauv>, positions
    reversed by analogy with 'ordinary' verbs. As usual this is the
    intransitive (class I) form, and we make the transitive form with the
    suffix <-en> (usually pronounced just /n/) and 'grading' of the last
    stem vowel: <cauv> to <ceiven>.

  * Aspect: <ceivena> /kEivna/ would be the aorist, here we want the
    continuous form <ceivenaun> formed by +an+(repitition of last
    consonant). I seem to have made an error in the text, with the form
    <aun cheivauvij>; I apparantly should be <aun cheivenaunij> /.aun
    xEivn.auniZ/. Happily the corrected form is no longer, since the <e> of
    the transitive affix <-en-> is elided where it has not undergone some
    vowel change (in other words, everywhere except in the perfect aspect).

  * Tense: past, so preceed by that fossilised form, the atavistic coverb
    "aun" and apply seimhiú/aspiration to the main verb: <ceiv-> /kEiv/ to
    <cheiv> /xEiv-/.

  * Person: not really necessary, since we have a noun <nautyadhel>
    supplying the ergative/subject position, and the governing clause
    providing the absolute/object nouns <ar laudh hoin uu dheesar>; but just
    for "biblical effect" I decided to inflect for subject: cf. "which they
    builded, the children of men". Hence the short form <-j> /-Z/ (only
    to be used finally) of the verbal 3rd person plural infixes <-sha->
    <-ssa-> <-nna-> (masc, fem and common). Could also have inflected for
    the object, for a biblical++ effect: "feem ghaun cheivenauninnaj ar ..."
    "... which they builded it (the 'which'), the children ...)

  * In total "-aun cheivenaunij" "they were building (something)" with a
    continual, not habitual sense. I got it wrong in the text, I said
    "-aun cheivauvij", which AFAIK is meaningless ;)


* <ar X uu Y>: alternate forms of the definite article to avoid the ugly
  sound of <ar X ar Y>

* <nautyadhel> /n.aucaDel/: "children", ergative case. Compare:
    <nautya> /n.auca/, "child" absolute case
    <nautyael> /n.aucael/ "children" abs. - the <ae> is NOT a diphthong.
    <nautyadh> /n.aucaD/  "child", ergative.
  I think that the plural used here <-el> is not the regular plural, but a
  "host" plural - a plural of multitude.
  (See also the genitive case in the section on <chjetyl>)

* <chjetyl> /xZEtyl/: "of men", genitive plural of multitude ("of the host
  of men"). The follwing is first in Latin-1 for clarity and then in ASCII:

         singular.               plural of multitude
    abs. cjetú   /kZEtu:/        cjetúël   /kZEtu:El/
    erg. cjetúdh /kZEtu:D/       cjetúdhel /kZEtu:DEl/
    gen. chjetuí /xZEt.ui:/      chjetyl   /xZEtyl/

         singular.               plural of multitude
    abs. cjetuu   /kZEtu:/       cjetuuel   /kZEtu:El/
    erg. cjetuudh /kZEtu:D/      cjetuudhel /kZEtu:DEl/
    gen. chjetuii /xZEt.ui:/     chjetyl    /xZEtyl/

  Note that the genitive is formed by i-affection and seimhiú (aspiration)
  - in the singular, /u:/ becomes /.ui:/ (a long allophone of the [ui]
  diphthong, which (allophone) is notated with a long mark in the
  orthography. It's the second part of the diphthong that's lengthened.)
  and the initial /k/ becomes /x/. In the plural form given, we get from
  abs. <cjetuuel> /kZEtu:El/ to <chjetyl> /xZEtyl/ by the same aspiration
  /k/ to /x/ and i-affecting the cluster <uue> /u:'E/ to */u:'Ei/ > /y/.
  Come to think of it, I think it should be a long "y": <chjetýl> or
  <chjetyyl> /xZEty:l/...


Well, that's it for the moment! Thanks to those who listened and
responded; and thanks to everyone on the list anyway for being generally
interesting and inspiring; For those curious, I may begin work on a
PDF of the grammar after I've sorted out a few things. I know that
elucidating this one sentence has caused quite a few simplifications,
clarifications and of course interesting complications to cross my mind.

Yours with delight,

Stephen