Re: SURVEY: Idiomatic Expressions In Your ConLang Or ConCulture

Sorry - I seem to have got behind a bit with reading this list.

Henrik Theiling wrote:
> Hi!
[snip]
> Ah, shoot, I gave the wrong verb -- a school is a 'reading-place'
> (atuarfik).
>
> The meaning *is* more special than what can be deduced.
Yes, I agree. This I would consider idiomatic since it suggests (a) that
the place set aside for reading is a school, and (b) that the main
activity in a school is reading. Both are incorrect.

One would expect "reading-place" to mean something like a reading-room
or maybe a (public) library.

[snip]
> Maybe more examples: eat-place = table.  sleep-place = bed.
> Deducible, yes, but they are more special than what they express
> literally.
Yes - 'twould be OK is all occurrences of VERB+place denoted the item of
furniture at which the activity (if indeed 'sleep' can be so described)
normally takes place. But the example of "reading-place" shows that this
is not so.

FWIW the same nouns in Speedwords are:
ryu <-- ry (building) + u (FAVORABLE) = school
tab [monomorphemic] = table
dorm

>>...
>>What I would call 'idiomatic derivation' are things like "evue":
>>  e-   v-        u-  e
>>be-ASSOCIATION-one-AUGMENT = "corporation"
>
> Ok, I fully understand! :-O
==================================

Paul Bennett wrote:
[snip]
> Idiomatic, for sure, but no more so than the Latinate incorporate, which
IMO it is a *lot* more idiomatic. The vast majority of English speakers
do not, I am certain, think in terms of the languages from which each
English words is derived, in some cases from languages spoken some two
millennia or more ago!

In _synchronic_ terms 'evue' is four morphemes and is IMHO idiomatic. It
is as tho in English we were to say something like "Mega-one
beingish[ness]" (Speedwords compounds are head-modifier sequences).

> I  parse as something like into-body-become, and indeed corporation,
> which I  think is something like that_which_is_made_into-a_body.
In _English_ 'corporation' is, at best, trimorphemic. I think it is
arguable whether the adjective 'corporate' is bimorphemic or not, but
assuming for the sake of argument that it is, we have:
corpor- bound morpheme = 'body'*
-at(e) - bound morpheme forming adjective denoting a perfect passive state.
corporate = united into a body (so as to act as a single unit)

 From 'corporate' we derive quite unidiomatically IMO the words:
corporation, corporatism, corporatist, corporative.

I see nothing idiomatic in this.

*body has several meanings in English, as did 'coprus' in Latin, one in
both languages being "a community of people (united by some common
tie)". This might well be described as 'metaphor' which IMO is not the
same as 'idiom'.
==============================================

But, strewth, why this obsession with finding so-called idioms in the
_diachronic_ derivation of words? By these arguments we might as well
claim that 'lord' and 'lady' are both idioms as lords do not really
guard loaves or ladies (for the most part) no longer knead bread.
===============================================

FWIW I now, after some thinking around the matter, adhere to Trask's
definition of 'idiom':
"An expression consisting of two or more words whose meaning cannot be
simply predicted from the meanings of its constituent parts."

Thus a single word cannot, by this definition, itself be an idiom.

I think, however, we can talk in terms of whether morphemes used in some
compound or derived word are used idiomatically or not. Thus I am in
perfect agreement with Rick Harrison when he says of Speedwords:
"Unfortunately, the definitions of most of these affixes are vague, and
their uses are very unpredictable and idiomatic."

--
Ray
==================================
ray@carolandray.plus.com
http://www.carolandray.plus.com
==================================
MAKE POVERTY HISTORY

Reply