Re: Szekler Hungarian Rovas (or: Old Hungarian Runes)

From: André Szabolcs Szelp (
Date: Fri Oct 31 2008 - 14:19:48 CST

  • Next message: Michael Everson: "Encoding Old Hungarian script"

    Dear Friends,

    I have to clarify certain issues in reply to Gábor Hosszú's mail.

    >> Old Hungarian was never written in boustrophedon.
    > This is not true; there are examples of this direction in the
    > proposal of the Community of the Hungarian Rovas Writers
    >(, see Fig. 14.10-12).
    > Another example is the "Bologna stick calendar" (see Fig.
    > 14.4-1/a,b,c,d,e,f and 14.4-2). The copy of the original stick
    > calendar remained, but Sándor Forrai (he was one of the most
    > acknowledged Rovas researchers) proved that this stick calendar
    > was originally scratched into the four sides of the stick in
    > boustrophedon. See Sándor Forrai: Sándor Forrai: Az ősi magyar
    > rovásírás az ókortól napjainkig, published by the Antológia Kiadó,
    > Lakitelek, 1994 ("The Old Szekler-Hungarian Rovas Writing from
    > the Ancient Time to Nowadays") ISBN 963-790- 830-7. This is in
    > Hungarian, but on the page 145 you can see on Fig. 67 (in
    > Forrai.s book) where Sándor Forrai described the reconstructed
    > boustrophedon direction of the script of this relic. Nowadays some
    > Hungarian Rovas carver already made the stick calendar again
    > using naturally the boustrophedon directionality.

    Let us put it down, straightforward, as Ádám Joó said it: Old
    Hungarian has never been written in a bustrophaedon manner.
    ad 1)
    Singular examples where a user (who is not even proficient enough in
    it which is demonstrated by a mistaken letter ("e" instead of "ö" in
    the word "jöttél", also not mirroring the lines going backwards)
    writes it in that way in a handwritten note (even if that note is
    reproduced) does not make the script to have bustrophaedon
    directionality. If I write a poem for decorative purposes
    bustrophaedon in latin or greek script does not make these scripts to
    have usual bustrophaedon directionality.
    ad 2)
    The wooden calendar copied by L.F. Marsigli (called "Bologna stick
    calendar" by H.G.) is actually a copy on paper. Each page of paper
    corresponding to roughly one side of the stick, each page containing
    RTL text.
    It was Gyula Sebestyén first, not Sándor Forrai, who suggested that
    the stick might have been carved in a serpentine way, i.e. that when
    you finished carving/reading a side of the stick you would not turn it
    around the main axis and make a "carriage return", but you would
    rather flip the stick and carve again from RTL (as you did just
    before) (but in the three-dimensional structure of the wooden stick
    the beginning of the second line would be closer to the end of the
    first than to the beginning of the first...). Also it was Sebestyén
    who first used the word known from greek paleography, i.e.
    bustrophaedon, to describe this mood of writing. He referred to the
    fact, that if the sides of the stick would be rolled out to a flat
    surface, one would identify what we see as "bustrophaedon". Besides
    the fact, that from both Sebestyén and also Forrai this topology of
    the inscription is only speculated, as we cannot possibly say from the
    manuscript (they tried to get it right from the order of folia of the
    manuscript, but it was proven (Klára Sándor) that they have been mixed
    up before binding, it is in fact even pretty probably that the stick
    was written that way, as we have stick calendars from Scandinavia and
    Siberia which work this way.

    However, a three-dimensional stick, bewritten in a way that if it was
    rolled out would amount for a bustrophaedon (with characters not only
    mirrored, but also standing on their head) is not bustrophaedon
    writing on a flat surface!
    Actually, the stick is logically rather a collection ("folio") of four
    flat surfaces (sides; "sheets"), each written onto in RTL.

    Also, it's not unicode's task, to describe in the standard every
    possible layout of the (writing) line on arbitrary objects. Also it
    does not. It only deals with representation on flat, two-dimensional
    surfaces (screen, paper). Consider what difficulties we'd have with
    toroids ;-)

    >> It is also worth to be clarified that there was no prior character
    >> set encoded for Old Hungarian; Gábor Hosszú regularly refers
    >> to his home
    > Do not mix the "font" and the "character set". The basic of the
    > character set was developed by the Runic Writing Section of the
    > Society of the Friends of the Old Hungarian Culture, mainly by
    > Győző Libisch in the 1980s (see Fig. 14.10-9 in the proposal
    > N3527). In 1995 I accepted this character set. We (with Győző
    > Libisch, the secretary of the above association) extended this
    > character set with the bug symbols, historical ligatures, Rovas
    > numerals and Rovas separator symbol). So it was not my "home
    > character set". This was called "Rovás Szabvány" (=Rovas
    > Standard in Hungarian). I encoded this character set using an
    > 8-bit encoding in 1995 and I made the appropriate True Type
    > fonts. This "Rovas Szabvany" became a de facto standard from
    > 1995.

    So, let's be consequent and let's don't mix terms, as H.G. suggests.
    What Ádám Joó and Gábor Hosszú are disputing above is actually a
    collection of glyphs. It's clearly not a character set, but a
    collection of glyphs. The glyphs were collected into a font with
    Latin-1 pseudo-encoding for ease of use. It's easy to demonstrate that
    we are talking about glyphs not characters: First, each font has
    several glyphs at different code positions, which are merely shape
    variants of the same character. Second, two different official
    versions of the font have glyphs on the same code-point which cannot
    be considered glyph variations of the same character. Third, the font
    contains some of the ligatures. Ligatures are completely optional and
    just a question of taste. They bear no additional meaning to the
    sequence of base characters, so they are glyphs to represent multiple
    So, to put it down, what was "developed" (actually: collected) by H.G.
    and Győző Libisch was a glyph collection. As such, this work was very
    useful, we have to thank for it. Technically, for use, the glyph
    collection was realised in a 256 Latin-1 encoded TT-font which allowed
    access to all glyphs.

    By the way, at the time of the development of this solution, in the
    90'ies this was a perfectly valid process, to have glyphs accessible
    by Latin-1 pseudo encoding. This does not make the glyphs to
    characters however.

    I'd like to point out, that speaking of a "de facto standard" is a
    gross overstatement. What are speaking about is a small private
    club/association, who names one of its products "standard". Only by
    calling something "standard" it does not become one. Real standards
    stand out by the fact that they are authored or sanctioned by an
    authoritative e.g. national or industrial body.

    Now, with unicode we have the chance to take the work of HG collecting
    the glyphs (for which we have to thank him) and some more, and
    implement them in a true standard (Unicode), using the contemporary
    character-glyph model.

    > We never asked Michael Everson to make standard for the
    > Szekler-Hungarian Rovas. However, in July of 2008 Michael
    > Everson asked me to organize a meeting in Budapest for the
    > Hungarian Rovas writers.

    It's not my job to reflect on this. Michael will have the chance
    reading this to comment on that statement.

    >But after this meeting it become clear for us that Everson-Szelp's
    > proposal is based on some relics from the Middle Ages and some
    > extensions for the "revivalists". I refer to the original proposal of
    > Michael Everson and Andre Szabolcs Szelp

    Our proposal coveres the same, not more, not less than HG's proposal.
    It only differes in the character-glyph distinction, i.e. the
    implementation of the given material. Therefore, if our is "based on
    some relics [...] and some extensions", so is HG's.
    And it ought to be:
    the standard is to cover historic and current usage fully. We don't
    want to come adding characters every year, we wanted a complete
    proposal, which we created collecting all the _characters_ from
    historic matieral as well as contemporary usage.
    Actually, any script encoding standard proposal has to consist of
    "relics" i.e. historic characters (especially, if all these historic
    characters are still in use today) and "extensions" i.e. currently
    used ones.
    Even though Mr. H.G. has repeated this above sentence at different
    public forums, I neither see the problem, nor do I understand what he
    wants to put with it. Is it a criticism of our selection of
    characters? Then he criticises his own work as well. Or is it just
    polemic rethoric?
    What we did was indeed giving section headers to the characters. They
    are only describing and factual about some characters having been
    added in the course of the 20th century.

    > (, 2008-08-03).
    > In this proposal there are a lot of sharp political (!) related
    > statements and a strong degrading approach to the "revivalists".

    HG regularly cites our _old_ proposal. Equally he could cite N2134
    (1999-10-02) or N1638 (1997-09-18) as well.
    However document given in the link above has been superceded by N3531.
    This is normal procedure. You write something up, you discuss it
    publicly, and then write a corrected/updated proposal containing new
    Citing an out of date document, even with good-will, cannot be called
    fair style.
    Even so, I'm giving a price to anyone spotting "political" statements
    (?). Also, I'm very sorry, if he feels "a strong degrad[ation]" being
    a revivalist, however, this is his personal problem of state of mind,
    as we (Michael Everson and me) have always seen revivalists work as
    something positive and admirable.
    "Revival" is the technically and factually correct term. c.f. e.g.
    Gaelic revival (
    And it's a great thing!

    > When in end of July 2008 the Community of the Hungarian Rovas
    > Writers were formed (as a wide internet-based working group) M.
    > Everson and A. Sz. Szelp modified the text of their proposal,
    > however, they did not change their view. Especially Ádám Joó
    > supports the Everson-Szelp proposal.

    Let me tell you an anecdote about this "Community of the Hungarian
    Rovas Writers"... wait, it isn't an anecdote, it's exactly how it
    First HG created a yahoogroups mailing list to supposedly discuss his proposal.
    Then he basically was not willing to actually discuss it. (He would
    not answer to our queries, and would polemize our proposal to the mass
    of readers of his list in a demagoguic way).
    Second, he declared at one point that this (his) list is from now on
    THE "Community of the Hungarian Rovas Writers", and anyone not
    unsubscribing from it automatically declares that he is supporting his
    version of the proposal.

    Nice story, isn't it?

    > Our view is basically different: our statement is that the Szekler-Hungarian Rovas never became extinct, it was continuously used. It was under development during the whole history. That is the reason why the sources of our proposal are based on a wider range of relics and practice from the historical time to nowadays.

    Even though it absolutely does not matter, what your "view" is, in
    terms of encoding ("It's a technical thing, stupid!"), I'd like to
    contest his view.
    Having something passed down from one generation of scholars to
    another, from one desk-geek to another by scholarly interest or
    because of the curiousity of the matter, is not what we call "never
    fell out of usage". If it was so, we could say that, let's say,
    hieroglyphs never fell out of usage!
    Fact is: the script was discontinued to be used. Then it was passed
    down by a small circle of people interested in it. Then from the late
    19th, early 20th century on one can observe a revival movement, which
    was pretty successful and the number of people _interested_ (not
    "knowing", "proficient" or even "knowledgeable"!) in the matter might
    be even several hundred thousand.
    That's great. But does not change the fact, that the script fell out
    of use at sime time in history.

    I'd like the point HG's attention to the fact, that we never called it
    died out, and also, that it's a mere biological fact, that something
    that's dead cannot be revived. So basically, our choice of word
    "revival" alone already shows that we are not treating the script as
    (ever having been) dead.

    HOWEVER, this does not even matter. It won't change the optimum way of
    encoding the script. Therefore it did not change the way we proposed
    the encoding, and neither would we change the encoding model if it
    were demonstrated by some surprising new evidence that millions of
    people (e.g. martians) used the script continuously.

    > The core of our proposal, the "Rovás Szabvány" was accepted and supported (based on voting) by the World Congress of the World Federation of the Hungarians (August 20th, 2008).

    The "World Federation of Hungarians" is an umbrella organisation of
    emigrant Hungarian organisations/associations.
    In the last years the people who got elected into key positions have a
    questionable ideology, to say the least.
    With all due respect, the "World Congress" of the WFH is a bunch of
    delegates from the emigrant Hungarian associations. They have neither
    deep knowledge about rovás, nor about Unicode or encoding. I'm sure
    that not one read the 106 pages document of HG.
    The statement therefore completely irrelevant.

    > The N3527 proposal (including the whole character set of the "Rovás Szabvány") was accepted and supported (based on voting) by the "Alive Rovas" Symposium in October 4th, 2008.

    Private club's meetings don't qualify as "ultimate arguments" either.

    Thank you for your patience and reading this long explanation,

       A. Sz. Szelp


    This archive was generated by hypermail 2.1.5 : Fri Oct 31 2008 - 14:24:01 CST