Home > 8, e, Features, gallows, Genetic Algorithm, y > Language A and B Again

Language A and B Again

A tentative conclusion from comparing Language A and Language B  is that the non-gallows glyphs are used in the same way in both Languages.

That is to say, they appear to mean the same thing. So the “o” in A means the same as the “o” in B.
There is some persistent “mixing” between the e/y glyphs, which is illustrated by the example result below:
ABMixing
There is also some doubt about the “8” glyph, which sometimes seems to mix with the gallows glyphs (e.g. in some cases, the “8” appears in A to function in the same way as a gallows glyph in B and vice versa). This may simply be an error in the comparison method, or it may be that the “8” is a null, or it may be due to some other effect.
The gallows glyphs are different – they don’t appear to mean the same in A and B. I’m focussing on those glyphs now.
Advertisements
  1. tonygdb
    March 13, 2013 at 4:17 pm

    Hi Julian
    The 4 ‘alphabets’ on f57v & the repeating sequence on f49v both intermingle voy101 ‘f’ & ‘g’ in the same positions – I think there is no doubt these two are loosely drawn variations of each other.
    The only difference is the loop in the top left corner – this is the only difference between ‘h’ & ‘k’ – hence are not ‘h’ & ‘k’ the same characters loosely drawn?
    Your transcription does not take into account the spacing between lines –I think this is crucial when comparing them.
    ‘f’ & ‘g’ are predominately on the first line of a paragraph, where (excepting the initial occurrence) they mostly appear after the line above has terminated – they appear more often in the middle of a paragraph in the herbal sections where the line spacing is not compacted, than in the tightly spaced recipes section where they seldom occur in the middle of a paragraph – it simply takes slightly more room to draw/write ‘f’ or ‘g’ than ‘h’ or ‘k’ –
    My inference is that ‘f’ ‘g’ ‘h’ ‘k’ are all the same character. ‘f’ & ‘g’ just being a more elaborate version used when there is space to do so.
    Do you agree with this reasoning?
    Does Python take these factors into account? – Surely, until it does it cannot make a valid comparison.
    Regards Tony

    • JB
      March 13, 2013 at 4:44 pm

      Tony,

      Your logic is hard to challenge 🙂 I can easily change the code so that all f,g,h,k in the transcription are mapped to the same glyph. Presumably you would say that the gallows glyphs embellished with “c”s are also all variants on the same one gallows glyph?

      I will run a test to see what that turns up.

      • tonygdb
        March 13, 2013 at 5:08 pm

        Yes – all the same – look forward to the results.

      • JB
        March 13, 2013 at 5:26 pm

        Not sure that this is exactly what you meant, but here is the replacement table I created:

        seek = [“3”, “5”, “+”, “%”, “#”, “6”, “7”, “A”, “X”,
        “I”, “C”, “z”, “Z”, “j”, “u”, “d”, “U”, “P”,
        “Y”, “$”, “S”, “t”, “q”,
        “m”, “M”, “n”, “Y”, “!”, “)”, “*”, “b”, “J”, “E”,
        “x”, “B”, “D”, “T”, “Q”, “W”, “w”, “V”, “(“, “&”,
        “f”, “h”, “k”, “l”, “F”, “H”, “K”, “L”, “R”]

        repl = [“2”, “2”, “2”, “2”, “2”, “8”, “8”, “a”, “y”,
        “ii”, “cc”, “iy”, “iiy”, “g”, “g”, “ccc”, “G”, “ip”,
        “y”, “s”, “cs”, “s”, “iip”,
        “iiN”, “iiiN”, “iN”, “y”, “2”, “9”, “p”, “y”, “G”, “c”,
        “y”, “cccN”, “ccN”, “s”, “p”, “g”, “g”, “G”, “9”, “8”,
        “g”, “g”, “g”, “r”, “G”, “G”, “G”, “r”, “r”]

        Running with this, a typical result is:

        ngramsB [‘c’, ‘o’, ‘9’, ‘e’, ‘g’, ‘8’, ‘a’, ‘1’, ‘y’, ‘i’, ‘2’, ‘4’, ‘s’, ‘N’, ‘G’, ‘p’, ‘?’, ‘r’, ‘v’]
        Chromosome [‘o’, ‘9’, ‘g’, ‘1’, ‘a’, ‘8’, ‘i’, ‘c’, ‘y’, ‘e’, ‘N’, ‘2’, ‘s’, ‘G’, ‘4’, ‘v’, ‘r’, ‘?’, ‘p’]
        ngramsA [‘o’, ‘9’, ‘g’, ‘1’, ‘a’, ‘8’, ‘i’, ‘c’, ‘e’, ‘y’, ‘N’, ‘2’, ‘s’, ‘G’, ‘4’, ‘p’, ‘?’, ‘v’, ‘r’]

        The gallows “g” is now the third most common glyph in Language A, and the fifth most in B. The best match chromosome shows good agreement, with the usual e/y mixing and some mixing of the rarer glyphs.

  2. tonygdb
    March 13, 2013 at 5:19 pm

    To clarify my last – I do not regard the ‘c’ and picnic table as gallows embellishments – but that the gallows has been placed between the already existing ‘c’s and the crossbar added later or before.

  3. tonygdb
    March 14, 2013 at 6:00 am

    Julian
    In the replacement table ‘)’ should not be replaced with ‘9’

    Treating the fghk glyphs as one is definitely the right way to go and should make ‘word’ comparisons a lot simpler.

  4. Knox
    March 14, 2013 at 9:24 am

    For consideration:
    To show dissimilarity of the sections, use nGrams instead of single letters and compare shorter nGrams in A1 with longer nGrams in B2. Rank-frequency tests indicate that a good first trial is to match 3Gs in A1 to 5Gs in B2. But that is from memory; other tests might be better. Include space as a character.

  5. March 14, 2013 at 9:47 am

    I should have written “similarity”. Matching equal length nGrams between A1 and B2 show dissimilarity:
    http://notakrian.pbworks.com/w/page/40475601/Repeated%20n-grams
    In these tests, short nGrams of equal length do not distinguish between the two sections very well. These tests are not amenable to comparison of shorter nGrams of unequal lengths. Also, the results are better presented visually than quantatively. Perhaps GA can overcome these restrictions.

  6. JB
    March 14, 2013 at 3:52 pm

    tonygdb :

    Julian
    In the replacement table ‘)’ should not be replaced with ‘9’

    Treating the fghk glyphs as one is definitely the right way to go and should make ‘word’ comparisons a lot simpler.

    Hi Tony – Thanks for spotting the “(” mistake – I corrected it in the code.

    Do you think grouping the gallows with one vertical with those having two verticals is valid? The lack of space argument doesn’t make sense there does it?

    Julian

  7. JB
    March 14, 2013 at 3:57 pm

    Knox :

    I should have written “similarity”. Matching equal length nGrams between A1 and B2 show dissimilarity:
    http://notakrian.pbworks.com/w/page/40475601/Repeated%20n-grams
    In these tests, short nGrams of equal length do not distinguish between the two sections very well. These tests are not amenable to comparison of shorter nGrams of unequal lengths. Also, the results are better presented visually than quantatively. Perhaps GA can overcome these restrictions.

    Hi Knox – I haven’t really looked much at nGrams > 1 so far. I think that the vocabulary is different between Language A and B, rather than how the “words” are constructed.

    Julian

  8. tonygdb
    March 14, 2013 at 4:10 pm

    Julian
    It’s not “(” that’s the mistake, it’s “)” that’s wrong.

    Absolutely valid.
    All gallows have two verticals – it’s just that with ‘f’ & ‘g’ they’re bent half way down!
    It’s the horizontal lack of space between lines that prevents the writer from using the ‘f’ ‘g’ more frequently in the middle of paragraphs.
    Put your transcription down, hide your computer – pick up a bit of tracing paper, place it over a bit of quire 20, (true scale) & try substituting an ‘f/g’ for a ‘h/k’ in the recipes section.
    I’ll stop here before it gets too technical!!
    Tony

  9. February 28, 2015 at 1:03 am

    Dear Julian,

    I hope this note finds you still a keen Voynichologist.

    I have benefited greatly from your sensible,
    thorough-going & thought-provoking work.

    In fact I have relied on the results of this particular blog post of yours
    for my own work on the Voynich.

    I have made a link to this page here,
    http://blog.voynichology.net/blog1.php/about-blog-a?page=4

    I hope this is ok with you?

    I also quote you here.
    http://blog.voynichology.net/blog1.php/about-blog-a?page=9

    Finally, I have greatly depended on your results
    as one of the basis for my assertion
    that in Currier’s Language ‘A’,
    EVA ‘F’ & ‘P’, which are only used in initial lines,
    are replaced by EVA ‘cth’, which only occurs in 2ndary lines.

    This discussion is found here.
    http://blog.voynichology.net/blog1.php/blog-voynichology-net-post-8?page=6

    I hesitate to try to express my overall results in a few lines.

    Essentially I have taken Brigadier Tiltman’s Paradigm,
    along with the analysis of Captain Currier,
    Mary D’Imperio’s PTAH analysis &
    Jorge Stolfi’s Voynich grammar
    along with your letter frequency analysis
    & demonstrated that virtually everything we know about Voynich grammar
    is matched by the ‘Grammar of Old Turkic’,
    as outlined by Marcel Erdal in his online book of the same name,
    which I quote here,

    http://blog.voynichology.net/blog1.php/about-blog-a?page=9

    & which is found here,

    altaica.ru/LIBRARY/turks/Erdal_OTG.pdf

    or

    http://www.google.ca/url?sa=t&rct=j&q=&esrc=s&source=web&cd=1&cad=rja&uact=8&ved=0CB4QFjAA&url=http%3A%2F%2Faltaica.ru%2FLIBRARY%2Fturks%2FErdal_OTG.pdf&ei=OQfxVNG8MIWVNveagPAB&usg=AFQjCNEkbWRBs6hyP75upD2Lp1JgcsUJXA&sig2=fR63J7IL3pPsCJz-xKDISQ&bvm=bv.87269000,d.eXY

    Combining this identification of Voynichese Grammar as Old Turkic, possibly Kaladj,
    with my identification of the Aramaic/Palmyrene letter values of the 4-fold letter/# sequence,
    (seen here),
    http://blog.voynichology.net/blog1.php/post-9-appendix-c-f57v

    on what I see as a drawing of the rete of an astrolabe on f57v,
    (seen here),
    http://blog.voynichology.net/blog1.php/blog-voynichology-net-post-7

    I have been able to derive letter values for virtually all Voynich letters.

    There is so much more to be said about all of this.

    I look forward to hearing what you think,
    as one of the most level-headed
    & gifted Voynichologists I am aware of.

    Yours sincerely,
    Marius Severin

  1. No trackbacks yet.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: