Re: Why blackletter letters?

From: Philippe Verdy <>
Date: Wed, 11 Sep 2013 00:06:32 +0200

Ad aburdum ? Not really.

IPA is a well-cntrained environment which does not attempt to reproduce an
orthography or grammatical rules of the language, but only its phonology at
best (using conventional "perceived" equivalences between relized phonemes,
even if there are many exceptions probably in all languages, depending on
ontext of use where disambiguation is still needed). Phonetic notations are
completely orthogonal to writing systems, so they don't interact and in
fact they are typesetted in compeltely distinct environments which are
easily separable (except if looking at a basic OCR or only a plain-text
representation without conventional syntaxic rules suchas delimiting

On theopposite the Fraktur script has its own distinctive behavior, with
lot of complex and historic conventions, sometimes contraditing with each
other, but more frequently contradicting with the modern use (or Classical
Roman use) of the Latin script (many modern extensions of the script are
simply impossible to reproduce clearly in the Fraktur script, and some
features of Fraktur are unified where they are not in the Latin script).

There are so many caveats that most medieval texts stillcannot be reliably
rendered if the modern extensions of the Latin script are applied (and it's
not reasonnable tothink that these extensions will be disabled, their
current use everywhere is too much frequent, including in documents that
frequently mix now Fraktur texts and Latin texts).

In fact what we call the modern Latin script is effectively a separate
branch from the Classic Roman script, and this branch is not really
compatible with the Fraktur branch, they are really distinctive scripts, as
much as between Classical Latin and Classical Greek, and much more than
between Greek and Coptic ! These incompatiblities are so huge that even
writing modern German in Fraktur requires a orthography than in modern
Latin (some examples include the modern umlauts which were plain letters in
Fraktur and that have been unified with diaeresis, modern sharp S as a
single letter where it was a ligature of two letters in Fraktur, and the
many ligatures and abreviations inFraktur that simply no longer exist in
the modern Latin alphabet (even with its extensions). Other examples
include the desunifications of many accents in the modern Latin script.

The modern Latin script (still compatible with Classical Roman Latin
script, even if it was initially unicameral) obeys different rules for many
other cases as well, such as word breaks online wraps, nonstrictly
horizontal placement of letters, ligatures, and abreviation symbols.

The desunification of the two scripts would not cause a major problem, if
we use transliteration schemes, because these schemes will just continue
using hueristics that are applied without any indication in texts currently
encoded with a single script. So nobofy is really satisfied and we still
have to live with interoperable OpenType fonts and rendere which only know
the modern Latin rules but ignores compeltely the Fraktur rules, and with
incompatible fonts specifically tweaked for Fraktur but that miserably fail
to render the modern script correctly if Fraktur features are enabled.

For now the inly solution requires enabling specific OpenType features, but
this requires document formats where these features can be selected and
enabled/disabled on demand. Fraktur texts still cannot be rendered
correctly from a pure plain-text representation. And any attempt to
introduce additional letters for Fraktur in the unified Latin script just
makes the handling of the Latin script even more complex than what it is

Who said that the Latin script was simpler and natural, and Hangul was
complex ? In fact it is even more complex than Arabic or Hebrew, and even
Hangul is MUCH simpler (ven if we consider the few historic jamos).

Of course there will be people attempting to render and handle the modern
Latin script with fonts that *look like* Fraktur, but they'll do it using
the modern Latin rules, not the correct rules for Fraktur. IF they use then
a transliteration scheme, this will work correctly most of the time but not
always. Transliterators sometime need to perform arbitrary choices between
options, even though the original Fraktur texts had theur own orthographic
rules (or historic exceptions).

Finally the Fraktur script is honestly very hard to read by most readers of
the modern Latin script, and even more difficult to write, as people often
don't know how it should really look like (Using an uniform sans-serif
Swiss/Helvetica/Arial/Verdana style for rendering strokes of the Fraktur
script is non-sense,but even if you do use a serifed font with regular
stroke widths, the Fraktur text will still be difficult to read by good
readers of modern Latin script AND by good readers of the Fraktur script as

The differences are much more important than between cursive and
non-cursive presentations, or between various cursive presentations (even
if most of them full of personalisationsor if people don't always use a
cursive style for handwriting, notably because the use of pens is
decreasing fast now, with the use of computing devices everywhere for
written communications, or the use of paper to fill administrative forms
that request using only non-cursive capitals).

2013/9/10 Gerrit Ansmann <>

> On Tue, 10 Sep 2013 22:55:43 +0200, Markus Scherer <>
> wrote:
> For use in IPA etc., there are in fact small caps letters:
>> %5B%3Aname%3D%2FSMALL+CAPITAL%**2F%3A%5D&g=<>
>> For small caps as a style, you would use markup.
> I am fully aware of all this. I was trying to lead Philippe Verdy’s
> arguments for encoding blackletter as a separate script ad absurdum, since
> one would also have to encode small caps as a separate script (or similar)
> then.
Received on Tue Sep 10 2013 - 17:09:03 CDT

This archive was generated by hypermail 2.2.0 : Tue Sep 10 2013 - 17:09:04 CDT