Re: Medievalist ligature character in the PUA

From: verdy_p (
Date: Thu Dec 10 2009 - 15:37:01 CST

  • Next message: Kenneth Whistler: "Re: [OT] New character request"

    If you want to see an example of texts with medieval ligatures requested by ZWJ (instead of using the legacy but too
    limtied preencoded compatibility ligatures, here is an example:

    (This is a very rare book, and one of the oldest about rules and strategies for playing chesses. it is written in
    nearly modern French, but by a Swiss Guard and printed by an editor that still used, apparently variably without
    strictly-defined rules, some of the many medieval ligatures. The text is not completely reencoded but you can
    compare it with the PDF scans by clicking on the tab with the book icon)

    Test your browsers, operating systems and fonts. There are many of them. But they work at least on Windows 7 with
    its native fonts, if using Safari and Chrome. They don't work in XP or with IE8 and Firefox where the ZWJ is
    rendered with a visible (over striking) zero-width glyph.

    Many of these ligatures are formed using simple or double long 's' followed by 't', but there are also case with
    'fi' and double 'f', 'ct'. Some rare ligatures for abbreviations (most of them using overstriking tildes, but also
    some using cedillas/tildes attached to the right of a leter).

    The rules about capitalisation or abbreviation of words is also not completely fixed thoughout the book, and even
    the orthography constantly hesitates without good reasons. It's possible that many of these variations did not come
    from the author, but from the publisher, given the limited number of books that was produced (and probably the small
    priced paid to the publisher to print the book : it's seems that the orthography and even the exact page layout is
    stable between two separate sets of pages, meaning that the books were probably composed on metal types by at least
    two workers using their own skills).

    Note the orthography of the word "échets", now written "échecs" but after fusion of orthographies from another word
    with a very different etymology : this word has a complex history with a branch coming from from Norman and later
    borrowed into English with the word "escheat" used today in legal affairs, or from some Nordic language though some
    Germanic language were it had a completely different meaning. The influence of Persian "shah" (the king) from where
    the game was imported is also suspected.

    In that book, you'll find the two orthographies of this word, with or without the accent (because all accents were
    still unstable in Medieval Oil languages)... The role of printer's ligatures and abbreviation marks seems to have
    influenced a lot the modern orthographies and the introduction of various diacritical accents (before they were
    mostly fixed in modern languages like French, where accents are still not completely fixed, despite of the work made
    by the Academy since the 17th century).

    Sometimes we also see that some accents are changed, or now made optional because they are no longer justified
    phonetically (some old phonological differences are no longer significant in modern languages that have uniformized
    various regional dialects) or because the accents were just remnants of old abbreviations marks, later turned into
    silent letters and then into another diacritical accent for etymological reasons, before their progressive


    > Message du 08/12/09 13:59
    > De : "William_J_G Overington"
    > A : "Doug Ewell" , "Andrew West"
    > Copie à : "Unicode Mailing List"
    > Objet : Re: Medievalist ligature character in the PUA
    > On Monday, 7 December 2009, Andrew West wrote:
    > > I note that many fonts on my system, including quite
    > > a few
    > > fonts from Adobe and Microsoft, include a visible glyph for
    > > ZWJ, ZWNJ
    > > and other format characters, but at least on Windows
    > > operating systems
    > > the user will never see them under normal circumstances.
    > Well, I suppose it depends on what one regards as "normal circumstances", but I find that Alt 8205 in Microsoft
    WordPad running on Windows xp professional using either the Arial font or the Times New Roman font that arrived with
    Windows produces a visible glyph for ZWJ, namely what I think of as the Don Quixote Windmill glyph.
    > Yes, I imagine Don Quixote reading a book with a c, the windmill glyph and a t and imagining a ct ligature glyph
    being there.
    > Some readers might like a short movie in relation to Don Quixote.
    > It is linked as La Mancha in the following blog post.
    > I wonder if Unicode will one day have a ZERO WIDTH ALTERNATE ONE and a ZERO WIDTH ALTERNATE TWO character so that
    one could request an alternate glyph, if available in the font being used, from a plain text file. Indeed if ZWA1
    and ZWA2 are introduced, then cZWJtZWA2 could be the way to request a very swash alternate ct ligature glyph from a
    font using a plain text file.
    > As modern typography has some fonts with many swash alternates and swash alternate ligatures available, it would,
    in my opinion, be good if plain text were to include a way to request such features.
    > William Overington
    > 8 December 2009

    This archive was generated by hypermail 2.1.5 : Thu Dec 10 2009 - 15:38:28 CST