[Trennmuster] Questions to better understand prefix and suffix
Guenter Milde
milde at users.sf.net
Sa Jan 4 16:30:56 CET 2014
On 4.01.14, Pander wrote:
> Hyphenation priority is:
> 1 -= hyphenation of compound's suffix, multiple = have higher priority
> 2 |= hyphenation of compound's prefix, multiple = have higher priority
> 3 = hyphenation of compound, multiple = have higher priority
> 4 | hyphenation of word's prefix, priority order is from left to right
> 5 - hyphenation of word, multiple - have higher priority
This is a misunderstanding: The priority (Güte) for hyphenation at this point
is generally
"=" (compound parts) > "|" (prefix) > "-" (ordinary)
For hyphenation points of the same category, relative priority is indicated
by repeating the hyphenation character.
For mixed hyphenation signs (|= and -=), the first character indicates
the category of the hyphenation point while a following "=" indicates
that the prefix or suffix is added to a compound word. See the examples in
"wortliste/dokumente/README.wortliste"
(The README uses the "Bindungsstärke" B of a hyphenation point, where
B = 1/hyphenation priority)
un|=wahr=schein-lich un + (wahr + scheinlich)
an-dert=halb-=fach ((an+dert) + halb) + fach
> There is a character to indicate a prefix |
> There is a character combination to indicate a compound prefix |=
> There is a character combination to indicate a compound suffix -=
> 1) Why is there a prefix character in the first place? Is it only to
> prevent too many -'s?
No. In German, there is a distinction of hyphenation points at morpheme
boundaries from hyphenation at "Sprechsilben". This manifests itself at
several places, e.g. the break-up of ligatures:
* Ligature "fl" is present in zweif-le but broken in Auf|lage.
and the spelling of the S-sound when typesetting in Fraktur:
* Weſ-pe (long s for ſp even if the s it at the end of a syllable) vs.
Trans|por-t (round s in sp when s ends a morpheme).
> 2) If there is no reason for |, then |= could be written as =- Correct?
There is indeed no such "hard" criterium for the distinction of | and =,
both indicate a hyphenation at the morpheme boundary.
note: According to the current orthography rules, you may always spell
compound words with hyphen if this improves clarity (i.e. Schiff-Fahrt
instead of Schifffahrt). No such rule exists for prefixes (i.e.
"un-gezogen" would be normally considered a wrong spelling).
And yes, using distinct characters saves many repetitions because
a) "=" is in most cases less binding than "|" (i.e. the binding to a prefix
is tighter and
b) prefixes are right-binding:
This means that instead of
Vor|aus|wahl=lis-te (Vor + (aus + wahl)) + (lis-te)
we whould have to write
Vor==aus=wahl===lis-te
> 3) If there is a reason why is there no character to indicate a suffix?
In contrast to a prefix, a suffix does not force a hyphenation point.
Hyphenation in base+suffix follows the "Sprechsilben".
woh-nung morphems: wohn + ung
However, if a hyphenation point coincides with a morpheme boundary, it is
usually a better choice than otherwise. Therefore we indicate this case with
--.
> Is there no need for it?
> A candidate could be
> http://www.fileformat.info/info/unicode/char/a6/index.htm
I don't know if there is a need.
> 4) If you would introduce a suffix character, then you would also have
> =¦ instead of -= This would make it more consistent and balanced.
This depends. I would always read left to right, the first character
indicates category, i.e. even then it should be "¦=".
> Can you find examples of double suffix usage?
Wirk--lich--keit
Herr--lich--keit
Ver|ein--bar--keit
Ver|ein--fa-ch¦ung # no hyphenation at ¦ !!!!
Hope this helps.
Remember, that the "wortliste" syntax is devised for a special purpose and a
special language by pragmatic non-linguists.
Günter
Mehr Informationen über die Mailingliste Trennmuster