Hacking at the Voynich manuscript - Side notes
070 Graphical matching of the SBJ "rooster" recipe

Last edited on 2026-06-04 19:06:23 by stolfi

INTRODUCTION

  In this note we create plots that show some possible detailed matchings of the 
  SBJ "red male chichen" (aka "red rooster", ROOS) recipe and parag f105v.32-38
  of the SPS.
    
SETUP

    ln -s ${HOME}/ttf
    ln -s ../work

    ln -s work/langbank

    ln -s ${HOME}/bin/gawkf.sh

    ln -s work/error_funcs.gawk
    ln -s work/error_funcs.py
    ln -s work/process_funcs.py

    ln -s work/compute_freqs_from_counts.py

DATA FILES

  The main data files are "in/SPSR.txt" (ISO Latin-1) with the contents
  of parag f105v.32 of the SPS in EVA, and "in/ROOS.utf" (Unicode UTF-8)
  with the full text of the "red rooster" SBJ recipe in hanzi. Both
  files have markup to help generate the plots.
  
  The file "in/ROOS.utf" has one line for each hanzi of the "rooster"
  recipe, as extracted from a digital Zhenghe Bencao file with the help
  of Google AI.  There are two fields per line "{CHTYPE} {CHAR}".
  The {CHAR} is one hanzi character from the recipe, surrounded by 
  punctuation (as per the file "work/Notes/077/in/ch/ROOS-b1.4.096.utf")
  padded left and right with ideographic spaces "　" to get the hanzis
  vertically aligned.  The {CHTYPE} identifies significant parts of the 
  recipe:
  
    N (+) Name of the entry ("red male chicken") or sub-entry ("head", "fat", etc.)
    B (-) Flavor and thermal properties field.
    K (+) The 主 ("mainly for") keyword.
    C (-) Patient population tag 女人 ("women");
    A (+) Conditions treated with the medicine.
    F (-) Comment 东门上者尤良。 ("[heads] hung over the east gate are quite powerful").
    E (-) Comment 神物 ("[amber is] a divine substance");
    M (-) Name of 鸡白蠹 of the "chicken white grubs" sub-entry.
    D (-) Indications of the "grubs" sub-entry.
    P (-) Provenance field.
    
  Based on the investigations so far, the items marked '+' above were
  transcribed into the SPS parag, while those marked '-' were omitted
  (not just here but apparently in all entries).
  
  The text of parag f105v.32-38 in the file "in/SPSR.txt" is in the EVA
  transcription alphabet and was taken from our 2025-2026 transcription
  of the SPS, which differs slightly from Rene's latest version.  This
  file has one line for each word of the parag, considering commas and 
  line breaks as word breaks too.  Some words (or parts thereof) are marked 
  "{N}[{STRNG}]" where {N} is a digit and {STRNG} is a string in EVA.
  These are the EVA strings that are being consideered as possible
  translations of the hanzi 主. 
  
  The digit roughly specifies how unlikely that string is to be such a
  translation. These code assignments take as certain that "ir" is a
  scribal error for "iin", so that "dair" is equivalent to "daiin", and
  that "m" is the Scribe's abbreviation for "iin". They also take into
  account that k, l, r, and t may be scribal errors for d, but with
  decreasing probabilities.
  
    0 daiin, dair
    1 daim (assumed to mean daiiin)
    2 dain
    4 kaiin, kair
    5 laiin, lair
    6 raiin, rair
    7 taiin, tair
    
DO IT

  do_note_070.sh