2010/4/24 Jorge Llambías
<jjllambias@gmail.com>
> In "da brodi drode broda de"
> (the head being "broda"), the relationship between each brivla and each
> sumti seems more intuitive with their distances being optimally short: "da",
> which is the merged x1 of "brodi", "brode", and "broda",
Not necessarily. da is the x1 of broda, but the roles of brode and
brodi can vary.
Yes, that has been my real understanding. For this example I just echoed a term I came across in CLL 5:7: "In fact, ``blanu'' has only one place, and
this is merged, as it were, with the x1 place of ``zdani''." But of course "da brodi brode broda de" is too general an example for such a non-general comment on tanru. I shouldn't have said that.
> On second thought, I realised that a head-modifier tanru would have been a
> reverse of its modifier-head lujvo partner, which I'm not sure would have
> been convenient. But then again why not head-modifier for lujvo components?
> Compounds like this certainly exist in natural languages:
>
> Chinese (SVO): 录音 lùyīn (record-sound --> sound-recording)
> Hebrew (occasionally SVO): בֵּית סֵפֶר bet sefer (house-book --> school)
Spanish (mostly SVO): abrelatas (opens-cans --> can-opener)
Some comparisons:
lantykartci
tcikarlante
snarejgau
gaurveisna
carna'ivinji
vijna'icarna
mitsumvla
valmunmi'u
I briefly thought that it could be a good idea to allow both orders so that we could have more options and possibly shorter / higher score forms to choose for each lujvo. But soon I realised that we would end up being unable to readily predict whether "snarejgau" is likely "recording of sounds" or "sounds of recording".
mu'o mi'e tijlan