For me, this all began in the early 1980's. I was frequently called upon during graduate school to program. I had been trained in FORTRAN in the mid-1970's, but another language, BASIC, had come into frequent use and I needed to master it. It wasn't a hard language, but there was almost no information for using BASIC for scientific and technical tasks. Then, I found a book called Scientific and Engineering Problem-Solving With The Computer, by William Ralph Bennett, Jr. This was a fascinating book for a number of reasons. It was intended to teach BASIC programming to people who had never programmed, but who had a technical background. This was very unusual for a book written in 1976. It was well-written; it would still be an excellent book to learn programming from today, although BASIC isn't what it used to be and we don't do BASIC on mainframes and teletypes anymore. Sadly, it is out of print, for its greatest value was the strange, but compelling examples that were used to illustrate technical programming.
Among the problems considered are the "Monkeys at the Typewriter" problem and how to make them better at writing Shakespeare, computer identification of authors, highly accurate simulation of moving bodies (with minimal simplifications so common in most texts), random processes (disease diffusion from Columbus' voyages and Monte Carlo simulation), digitizing musical instrument waveforms, extracting meaningful signals from random noise (Raudive voices) and, most importantly, the Voynich Manuscript (VMS for short), the subject of this column.
The Voynich Manuscript was one of the first things I searched for once I got on the Web in late 1995. The Web was still pretty new and I thought that such an obscure subject wouldn't be covered. I found more than sixty high-quality sites, even with the poor search engines of the time. I shouldn't have been surprised; such a troublesome mystery was bound to attract great attention. Most of what I'll tell you in this article will come from Bennett's book and another that I've recently acquired, also written in 1976, called The Voynich Manuscript - An Elegant Enigma by M. E. D'Imperio. At the end, I'll give you links to some good Web resources that you can look at on your own, if you wish.
The modern history of VMS begins in 1912, when Wilfred M. Voynich took possession of a 200+ page manuscript found at a Jesuit college in Italy. The document is a combination of extremely strange drawings that appear to relate to alchemy, botany, physiology, astrology and astronomy and secret writing in a peculiar, unknown alphabet. Many of the drawings show people performing strange actions, while many of the plant drawings show nonexistent plants. The document has never been decoded, although many have tried, including medieval historians and military cryptographers. Indeed, much about the document is unknown. It is suspected by some of being either the original work of Roger Bacon, who died around 1292, or a later copy. There is solid proof that the document existed in 1666 and the evidence is very strong that the document was owned by Rudolph II of Bohemia, who died in 1612. Virtually all researchers believe that it was created by 1550, if not much earlier. It has never been carbon dated, which seems a bit puzzling. Experts can't agree on what language it might have been originally been written in, or even if it is anything but the work of a madman, prankster or swindler; one frequent opinion is that it is a fraud, containing no real information. Here is where Bennett's textbook example made an extremely interesting and valuable contribution to the legacy of VMS, which also enhanced the weird nature of the document and helped to keep researchers from giving up on it.
Bennett applied modern information theory to calculate entropy values for a transliteration of the document. What he found is that the information density of VMS doesn't match any European language, but it also doesn't match gibberish. The relatively low entropy values do approach those of some Polynesian languages, but given the probable date of creation of the document, this finding just added to the VMS mystique. Bennett anticipated the actual meaning of these strange properties of VMS. A recent, interesting paper by Dennis J. Stallings titled, "Understanding the Second-Order Entropies of Voynich Text", which can be found at www2.micro-net.com/~ixohoxi/voy/mbpaper.htm, seems to definitively eliminate the possiblity that it might be a low-entropy language, as well as eliminating other possible explanations, such as it being an elaborate schizophrenic "rant". Bennett suspected and Stallings agrees that VMS may well be a "verbose cipher", where multiple characters of the text represent a single character of real text. While reducing the strangeness of the document, this finding increases the possibility that it might eventually be solved.
Hopefully, I've got your interest. There is no substitute for seeing some of the pages of the document. You can see the document in person at Yale University's Beinecke Rare Book and Manuscript Library, where it has resided since 1969, or you can go to the library's on-line digital collection at http://highway49.library.yale.edu/Scripts/db2www.exe/PHOTONEG.d2w/INPUT. Locate the "Free Text Search" form and type "Voynich" in the search field, hit "Submit WORD(S) Query" and you should get sixty hits, showing many thumbnails of VMS pages. By clicking on those that interest you, you can see a larger image and information about that page.
For an excellent on-line discussion of the document, try, When Words Fail by Lev Grossman in the April, 1999 issue of Linguafranca at www.linguafranca.com/9904/grossman.html.
Perhaps the definitive on-line resource for VMS is the EVMT (European Voynich Manuscript Transcription) Project Home Page at web.bham.ac.uk/G.Landini/evmt/evmt.htm. It is a large, rich site with links to many other sites, and much content of its own, including a Voynich computer font.
Other VMS websites I recommend:
The Voynich Manuscript Mini-FAQ at www2.micro-net.com/~ixohoxi/voy/voymfq.htm,
Jim Reed's Voynich Manuscript page at http://www.research.att.com/~reeds/voynich.html,
Stolfi's Voynich Stuff - Index Page at http://www.dcc.unicamp.br/~stolfi/voynich/,
Gallery (other Yale VMS photos), http://inky.library.yale.edu/voy/voy2.html,
and Solution to the /cryptology/Voynich problem, (this appears to be an abstraction of the VMS section of the USENET rec.puzzles cryptology archive, formatted nicely), http://einstein.et.tudelft.nl/~arlet/puzzles/sol.cgi/cryptology/Voynich.
CATBAR - Brain Candy #38 - Another Unsolved Mystery / Brian Rock / Oct 10 2000