Your Voynich comments


NOTE: This page has become very full, thanks to your comments. Quayong.

If you want to search for anything in it, or on my site as a whole, the best way to do it is to go to the Google Advanced search page, put ‘’ into the ‘site or domain’ box’, then put your search item into the ‘Find pages with all these words’  box landsat 8 다운로드.

For general questions about the Voynich manuscript please look first at these sites:

René Zandbergen’s comprehensive website

Yale Library, where it is located 잘난체 다운로드.

Jason Davies’ website where you can see every detail of the whole manuscript 풍아대국 다운로드.

Nick Pelling’s informative blog about it.

Feel free to add your comments on my research below windows 10 정품 다운로드. Thanks







  1. Thomas M. Brymsi im Thurn

    Dear prof,

    I know several languages and speak four fluently. I am also a mathematician with good cryptography skills. I am also very fend of History. However, I know very little on Arabic languages.

    Still, being a scientist, with a very varied knowledge, I have studied several aspects of the manuscript and have gone through your explanations and considerations, especially those identifying Arabic names of plants. However, it has been time since I didn’t put my nose into that and I just wanted to share a really short summary of my thoughts before I forget. Next is my humble guess, and I would like to know both: if it has been examined in some extent and if not, if it could bring any help or hint.

    Fortunately, I have studied a little bit Persian culture and History, and some characters and the writing style strongly reminds me of Avestan or, even more, Pahlavi script, both being ancient Persian scripts. Two considerations arise from there: that the aforementioned scripts go backwards (right to left) and that they do not include consonants. The script direction also depends on support and writing tool. A right-handed scribe writing with an ancient ink pen on paper or parchment will have a lot more ease to write from left to right and will be a lot quicker, while when you are using a pencil or a calligraphy brush, it matters only little, although carving letters into stone for a right-handed person will be a lot easier from right to left. I will explain later on why I mention that.

    Firstly, it is not a problem, with almost any script, to change from left to right to right to left. Secondly, the letters that differ the most from Pahlavi or any other old Persian script seem to ressemble some latin consonants, which could be an explanation for the apparent similarity with latin alphabet of some characters, mixed with some unknown, unlikely latin, characters. That second point is obviously one direction to look into.

    That hypothesis could also explain why we have close to Arabic plant names ; since Persian uses a lot of Arabic terms (and the other way around). I will not go through my careful character analysis and my statistics now, but I could simply note that the character resembling 4 is always followed by a character resembling o, and, if I remember well, always at the beginning of a word. It is often identified as two different characters, however there is never that 4-like character alone somewhere. I suppose you are well aware of it, but I simply mention that, because it could have something to do with what I mention before: not certain what, but it is a strong feeling supported by seemingly concurring traces.

    Now, here is my mythological and historical suggestion, just to complete my idea. There is a mythical tribe in Eastern Europe, around the Black Sea, called the Scyths or Scythians. They have been frequently identified with Persian culture and supposedly the mythical ancestors of ancient Russia, if not of every civilisation in Europe. Yet, what is interesting, is that the History of the foundation of Russia only appear in quasi-mythological chronicles and that they have a very visible lack of documents during this period, leaving all sorts of suggestions open for investigation on what could really have happened in this area and how. Now, is that possible, that some remains of that so-called Scythian culture have gone through the time and given birth to the Voynich manuscript before finally disappearing from that area? That could explain the strange nature of that language and manuscript. That is not a certainty, but an hypothesis that could lead to other ones, and that can claim to bring a lot of noticeable curiosities together.

    I do not have time for that, but I wanted to make sure to share with people who have time and competences to further investigate.

    Thank you for your patience.


  2. Jim Hamman

    Thank you for the wonderful work you have done on this amazing manuscript, Professor Bax. Have you considered the possibility that the manuscript is actually a copy of a much earlier manuscript? This would account for the “error-free” text, and the occasional pauses seen in the writing as the scribe paused mid-stroke to study the original. Also, have you considered that the original author, who either had command of several languages and/or was compiling his manuscript from multiple sources written in various languages, may have signaled within the text which source language he or she was using by prefacing a word, sentence, or paragraph using a specific ‘symbol’ (e.g., P to represent Persian)?

  3. Very interesting! but i think you thinking to far…

    what i think this book could be some sort of a school book or a notebook from someone (analphabetic or handycapt person) who tried to write (maybe fast) with his or her own sort of writing (like kids do, incorrect writing). as my own log from kidsdays is very cryptic too 😉

    i think it may be a person who studied something about herbal medicin and healthcare and wrote the facts about those plants which actualy have some medical healing function. also the drawings of some kind of a woman in some water containers with some kind of tubing around could be some kind of healing bath mechanism??

    medicine is often combined with astronomic knowledge so that would explain the star pictures.

    this is my theory and my guesses don’t judge me for this 😉 it could also be rubish what i’m telling 😉

    Greetings Mike, from Switzerland

  4. Jay L. Stern

    Everyone loves a good mystery, it seems! I did not find anything suggesting that the physical properties of the book have been analyzed. Has the material on which the manuscript was written been identified? That is, is it paper, vellum, parchment, papyrus or some other material? Calling it “paper” for for the sake of simplicity, has the origin been determined, the page thickness measured and any other physical parameters analysed? What about the inks used? Are they known to be from a particular area or region? The ink chemistry can tell you a lot. It might pinpoint the time the manuscript was written and can help identify the region. For example, there is a lot of green in the images shown. What was used to create the color, especially something that has lasted for 600 years? Also, the binding can give information, as can the covers. Any research or informaton on these parts?

    • Stephen Bax

      Answer: a lot

      Please see Rene Zandbergen’s comprehensive site at for answers to all these questions.

    • Mr Stern et Mr Bax
      Here is the explanation by a Rabbi for his students numérical and symbolic singularities
      of the letter lamed twelfth letter of the hébreu alphabet,which comprises 22 Basic letters with in addition five finals
      Nicolas Georges 67ans

    • Page 52 codex voynich.,,,,,,,Letter ….LAMED….alphabet Hébreu

    • Messieurs Bax et Stern
      The text reproduces numérically the exclamation of the singularités of the
      Letter…….LAMED… the Divine structure of This alphabet
      Nicolas Georges
      67ans France

      • Stephen Bax

        Nicolas, this is not a useful explanation. Identifying one letter and saying it looks like a letter in another alphabet does not help us move forward.

  5. Peter Clement

    Dear Dr S Bax,

    Page f67v2: I suggest the information and diagrams on this page are associated with the ancient lunisolar calendar. The main factors of the lunisolar calendar is the moon and the sun, more specifically the Full Moon and Winter Solstice, Summer Solstice, Spring Equinox and Autumn Equinox. However, from the Ancient Egyptian to the Germanic calendar there has been no information on what methods were used to measure and set these factors. I propose that this page f67v2 shows these methods.

    Firstly a quick overview of the page. The circle is made up of an inner and outer border, the inner border shows four flowering plants with stems and a single star shaped flower at its centre. I believe the author has been clever by combining knowledge of flowering plants with the symbols of the sun and moon as a reference to ‘periods of time ’ and will become apparent in the following explanation. On the outer border touching the circumference there are four circles showing the methods used to measure – Full moon, Equinox, Winter Solstice and Summer Solstice.

    Therefore, let me start with:
    1) Full Moon in the NE section. The long stem plant with a face of the moon embedded into its blue flower buds is a clever combination to indicate ‘nighttime’. The moon is an obvious choice but the author has used knowledge of plants to add a further reference. The blue flowered plant resembles a pre bloom Agapanthus, a diurnal flower which closes its flowers at night. The four moons in the circle provides information about the phases of the moon. The face with the stretch neck touching the circumference is the Full Moon.

    2) Equinox in the SE section. Continuing the theme of using the combination of the flower and this time the sun it suggests a Springtime rising sun. The L shaped structure in the circle is similar to a L shaped shadow measure first used by the Ancient Egyptians. The four faces are used as pictorial guides on the function of the structure. The east pillar identified by the two faces with caps indicate that the sun is shining onto the back of their heads and the top face is looking inwards in the direction of the pillar’s shadow onto to the measuring section of the structure. The most noticeable feature on the measuring section is the face with a pointed cap. Its position and hat are important factors providing a visual reference to the mid point along the measuring section and marking the end point Equinox shadow. Only one marker is needed for both the Spring and Autumn Equinox because their sun’s angle are equal.

    3) Winter Solstice in the SW section. Of all the sections this is probably the best known. The open flower and the moon suggest a Winter sunset. The flat U shape structure in the circle is similar to gateways and arches used in Ancient Egyptian temples and bronze age structures (Cairns). The Winter Solstice sunset (21st December) is aligned directly behind these structures focusing the sun and its shadow along a passageway made either by the pillars of the gateway or as in the Cairns an actual passage way. This method is usually measured by a spectacular end point of its shadow. The flat U structure uses the faces and colour to provide a pictorial guide to the function of the structure. The structure is split by the colours, the west pillar blue, the east pillar green and the flat section split equally with red and green. The colour red and blue represent light and darkness respectively. The green is neutral and is not used as a measure. You will note that all the four faces looking outwards indicating the direction of the shadow with again the sunset shining on the back of their heads. What is different is the two faces on the flat are upside down this indicates that a shadow is also produced by the flat section of the structure. The shadow produced by the setting sun on the flat U structure will inevitably be long and will end with a ‘magical end point’. In this case the centre of the Star flower. Why this end point? because of the flower’s name which is the clue – The Star of Bethlehem.

    4) Summer Solstice in the NW section. The flower and the sun suggest a Summer’s morning rising sun. The wedge shaped standing structure is similar to specific standing stones in some bronze age stone circles (Ring of Brodgar Orkney). The method used to measure the Summer Solstice relies on one important factor – the angle of the sun is at its maximum on the 21st June. Matching the sun’s angle to the angle of the the wedge will provide a reference point and no shadow on its edge. An alignment of the structure with the late morning sun rise will provide a good reference position for the structure. The three faces show the direction and track of the sun’s shadow starting from the top and moving clockwise.

    Finally, the key to understanding this page is to confirm the identification of the central flower which I believe to be a ‘Star of Bethlehem’.

    Peter Clement

    • Te Tao Nui Anihana

      Hi I’m not a scientist or genius but Im good at navigating by my view of the moon pie chart given the language. It maybe a coordinate on loop in some type of ancient constilation mapping. Which means the moon is the map and might be how to get there in realtime. I’m no scientist I don’t claim to be right but who ever wrote this was ahead of their time. How I made this conclusion was the eight portions which could represent north south east west and so on why I believe this it sort of looks like a co.pass. I know how bizzare is that but it is a bizzare manuscript by the way have a prosperous journey.

  6. Ian Simons

    I like your approach, working out the proper name of plant first.
    Pl refer to web page.
    Picture at 20/209, I think is wild pansy. And picture at 82/209 is a fern. My choice is maidenhair fern (Adiantum). Best wishes, Ian Simons

  7. Jay

    Stephen Bax,

    I have an idea to help your effort, but I don’t know if it will work or if I can collect enough resources. I became aware of the Voynich manuscript from some YouTube videos related to “unsolved mysteries”. More recently I stumbled upon your video regarding your initial work with the text and became convinced, that if a computer database had all of the necessary resources together, that perhaps some of the mass searching required to solve this puzzle could be done faster. It appears that some of your work relies upon understanding the sounds of the language, and then finding similar sounding words in existing languages that may be related.

    Basically, I’d like to experiment with the possibility of placing the entire transcribed VM (already have that in electronic format) into a database . Then, take the proposed symbol to sound mappings and put them into the database. Then, locate downloadable resources that have words in different languages, their pronunciation (in some common format), and their meanings mapped to English (I’m looking at WordNet related resources for part of this). Then make the database searchable –looking for connections between Voynichese and other languages.

    I was going to do several weeks worth of up-front searching for resources before making my first post, but decided that maybe it would be better to see if known resources already exist. For example, is the IPA (international Phonetic Alphabet) a possible way of mapping the sounds of different languages together? Or have you and others already converged on a standard? Is there a downloadable format that lists words of a language and the phonetics for those words?

    I’m a competent software programmer and as long as the data exists in files that are readable with code, I can figure out ways to enter the data into a database.

    Is this a totally stupid idea? I think I have read where computers were used to attempt to solve the VM but I don’t think they were instructed to take your strategy as a starting point. I don’t know how much time I can devote, but if I can find the right resources, that will help me carve time out.

    I’ll be searching for the resources I described until I get a response.


  8. Mary Smith

    Dear Professor Bax, 16th April 2017 (Easter Sunday)

    I am a housewife and a ‘history’ novice, but am intrigued by ancient history and mystery, particularly Egyptian / Ancient worlds material on Youtube, just for pleasure, and ‘watch’ programmes in preference to reading. Some time ago I saw the Voynich Manuscript on a tv programme and a few days ago Youtube listed it as recommended viewing. Your name and work has been highly recommended by some of the people posting videos relating to the Voynich Manuscript.

    I’ve not seen the whole manuscript and haven’t studied the writing aspect of the manuscript, but I like the pictorial aspect and particularly the stars. Over the past few days I have been watching Zodiac/astrology videos, etc (including Santos Bonacci). One video I saw mentioned number of days between the equinox/solstice, etc, eg 94, 90, 92 and 89 which took my interest.

    I have been looking at one Voynich chart for several days/almost a week and I feel rather elated following a few basic calculations this morning proved pretty amazing. Wow! The chart is not folly but has meaning. I have no clue as to the wording, but the ‘star’ pie chart (plate 68r3 ?) is divided into 8 portions that appears to be seasons (1, 2, 3 and 4 stars), with each alternative slice containing a number of stars, eg 16, 18, 14 and 11, with a combined total of the latter of 59 stars.

    Taking the number of days in a calendar year, 365 days divided by 59 (stars) = 6.1864406

    What is the significance of 6.1864406?
    Appears to be the vital number, although I know not why but it was true of for similar formulas for the reproduction of the Antikythera mechanism.

    What is the significance of 1 star?
    Appears to represent “1 week/7days”

    What is the significance of the central picture?
    Appears to represent the moon (a full moon?)

    Spring 1 star (+ Pleiades) 16 stars x 6.1864406 = 98.983049 days between Equinox
    Summer 2 stars 18 stars x 6.1864406 = 111.35593 days
    Autumn 3 stars 14 stars x 6.1864406 = 86.610168 days
    Winter 4 stars 11 stars x 6.1864406 = 68.050846 days
    TOTAL 364.99997 days / 1 year

    Spring 98.983049 days / 7 (days) = 14.140435 weeks
    Summer 111.35593 days / 7 (days) = 15.90799 weeks
    Autumn 86.610168 days / 7 (days) = 12.372881 weeks
    Winter 68.050846 days / 7 (days) = 9.7215494 weeks
    TOTAL 52.142855 weeks / 1 year

    I’ve just checked the weeks in a year and this corresponds with my sums: = 1 common year = 365 days = (365 days) / (7 days/week) = 52.143 weeks = 52 weeks + 1 day

    This is all my own work and I may be repeating common knowledge but I thought to bring it to your attention should it be relevant and helpful to the Voynich Manuscript studies.

    Very best wishes,
    Yours sincerely, Mary Smith

    • Te Tao Nui Anihana

      Hi I been reading your calculations and my approach was a compass with the moon the center and the 8 portions relating to the calculations you presented and the compass of a moon in orbit of the sun it might be a solar calendar. How I might claim that im no scientist but I deffinetly know a compass when I see one. But I’m no scientist.

  9. Charles Polák

    Dear Professor Bax:

    I have a background (MA 1984 with Philosophy) in theoretical and descriptive Linguistics, under Jan Mulder and Sándor Hervey in St Andrews – but my great love has always been comparative and historical linguistics, including paleography and the history of writing.

    As soon as the first indications appeared that the Voynich language might be Indo-Aryan with an eclectic admixture of Near/Middle Eastern and east-central European elements, I thought immediately of Romany – and indeed this has occurred to several other people, as one might expect.

    What also occurred to me, as an enthralled boyhood reader of Hans Jensen’s Die Schrift, translated by George Unwin as Sign, Symbol and Script, is that central and north India have a large variety of informal Brahmic scripts used by merchants and moneylenders for account-keeping and notes. Could not only the Voynich language be an Indo-Aryan one, as Romany is, but could an otherwise unheard-of thing, an actual traditional script for one of the notoriously unwritten Romany languages and dialects, be transmitted in the Voynich MS – perhaps a bit stylized by a good late-medieval-Renaissance scribe (thus looking somewhat like Beneventan Roman as it does), but otherwise the last dying gasp of an unknown tradition of the Romany diaspora?

    The kind of script I have in mind is exemplified by Mahājanī, for which there exists a 2011 Unicode proposal (, but many others exist in this ‘informal zone’ in the South Asian subcontinent.

    Just something for people’s consideration!

    • Charles Polák

      In particular, look at some of the Mahājanī numbers: if a fair number of these occur, their monotonous repeated strokes would certainly lower the Zipf entropy.

  10. Daniel Briggs


    I would like to bring the following to the attention of Stephen Bax and Derek Vogt especially, as well as anyone who is familiar with any Romani (Romany) languages, old or modern.

    I took Derek’s advice and did an automated search-replace of the EVA text of the Voynich Manuscript. I used Takeshi Takahashi’s version because of its modern understanding and internal consistency.

    I used letters that were in all but two cases identical to Derek’s choices in his videos for those for which he has posited a sound. I just changed $ to ś and the flourished (r), which is m in the EVA, to ŕ. (No phonetic claim in the latter case, I just wanted the difference to be demarcated.)

    Takahashi manages not to use capitals outside of braces, so I used capital letters where no phonetic claim is being made: I, J for EVA i, n, which seem to be numerical or something, show up foremost at the end of words, and in which J is probably just an I with an end-flourish; Q for EVA q, which only shows up at the beginning of a word and is probably phonetic or a stand-in initial; and the rest show up so rarely as to not need to be mentioned, but for completeness, G, V, X, Z for EVA g, v, x, z. EVA c and h (the two halves of what Derek parses as [h]) show up standing alone seemingly only where the text was hard to read, and I use C and H in these rare cases.

    I’ve looked it over briefly and I believe this transcription was error-free, but of course there may be errors. You can get it at

    folia 1–10: dtzj5
    folia 11–20: dtzj0
    folia 21–30: dtzjj
    folia 31–40: dtzjq
    folia 41–50: dtzjm
    folia 51–58: dtzjz
    folia 65–70: dtzj2
    folia 71–80: dtzjn
    folia 81–90: dtzje
    folia 93–100: dtz4s
    folia 101–108: dtz4d
    folia 111–116: dtz4r

    the five-character string being what to put after

    I’ve only looked at it briefly so far, but one thing already stands out to me. In “Botanical Life-Forms in European Romany,”

    F. David Mulcahy does a survey of words and finds the basic words tsar, bor, and kaš referring to general botanical things in Romany. He believes they come from Sanskrit ṭaru, būṭā, and kāṣṭha, respectively. (The second had variously been believed to be of Slavic origin, but that was purely speculative.) The last of these three etyma, kaš, was believed by those with whom Mulcahy was in contact (I know not were they Romany or linguists versed in Romany studies) to be very old indeed, and very general. In Sanskrit, according to Mulcahy, kāṣṭha means “log” or “piece of wood.”

    So I decided to check the document for examples of this. What I found blew me away.

    akaš – 255 occurrences
    nkaš – 24 occurrences
    kaš – 105 without the above two prefixes
    χaš – 283 occurrences
    haš – 598 occurrences
    ǧaš – 80 occurrences

    śar – 82 occurrences

    Mulcahy says that śar is attested by Spanish Kalderasitska (Romany) speakers to mean grass or grerb (little plant, bushy plant); its ancestor ṭaru in Sanskrit seems to only have been used in a verbal sense “to graze.” Let’s look at the first few instances of śar in the document.

    yup, tiny plant, no hard wood
    small plant, no hard wood, word *directly before the plant*
    now this one is interesting. We have śar and χaš *together*, with ǧn and hakaIIJ intervening. Let’s look at the picture
    *It’s a plant with no hard wood growing *out* of a log.*
    You can see the pictures here:

    Derek, when I saw your second video last night it brought a tear to my eyes. Hopefully here I have contributed something of value.

    • Stephen Bax

      Thanks Daniel, interesting, but of course we would need to start to find grammatical elements too, for it to be more convincing. Otherwise it could be chance, sad to say.

      • Daniel Briggs

        Thank you for your reply, Professor Bax.

        Yes, after poring through this transformation of the EVA a little more, I realized that akaš, nkaš, kaš, χaš, haš, ǧaš and the like are repeated rather monotonously in a few places throughout the text. And some basic Romany words (such as baro “big”) fail to show up, in any conceivable form, nearly frequently enough to support a Romany-language hypothesis.

        And even though the “labels” parts of the pages seem to follow what would be phonetic distributions of a natural language, about sixty percent of the “paragraphs” parts of the pages are infused with what must be something else, composed of variants on “Qakoon” (EVA “qokeey”), repetitions of haš, kaš, ǧaš, and similar short words ending in IIJ (EVA iin).

        All these conclusions, I’m sure, are similar to those that anyone who has spent a work-week poring over it has come to.

        I’ve moved on from hoping that Romany can help much without a prior decipherment of the paragraphs sections, even if it does conform to Romany phonology.

        But in the spirit of throwing things at the wall until it sticks, I’d like to offer a few possibilities, if some of these words are Syriac or related, or wanterworter attested in Syriac.

        Eastern Syriac iš’kara: “acacia”
        This label is on a plant that looks just like acacia: buds, furrowing outwards, all on stems stemming from a common stem.

        Eastern Syriac ‘gentša “bud/gemma”
        The picture is of a plant with many little buds, and the word for bud is the first word on the page.

        Eastern Syriac ‘aḥta “stump” (possibly) on a picture that could be a reimagination of, or being compared to, khat

        Eastern Syriac ur’zaga “stamen/pistil/pith” split across a plant?

        Eastern Syriac ‘ukhta “Lilium agreste (Hemerocallis, daylily).” Dubious.

        • Daniel Briggs

          The folio number references got blanked out (probably because of the angle brackets).

          aškar is Folio 101v2, R(2)6
          gntša.hašn.ǧar.arhon.ś.χn.atwIIJ.śwrn-{plant}χar.ǧn- is Folio 9r, P1
          thon.kʰat.thaš.ahtn-{plant} is Folio 24v, P15
          χa.hakʰn.kʰawr-{plant}śagatwJ-{plant} is Folio 56r, P18
          nkooa.wšχon.nkotn.koχ{plant} is Folio 41r, P2.

          I will head over to if I find more (likely or dubious) correspondences.

  11. Bob Wilson

    I have only briefly seen some of the manuscript and it occurs to me that the naked ladies are treading grapes to make wine. No doubt there are recurring words : grape, must, vine, tread, wine, flow, etc ….

  12. Nikolaj

    Good day!
    My name is Nikolai.
    To a question about the key to the Voynich manuscript.
    Today, I have to add on this matter following.
    The manuscript was written no letters, and signs for the letters of the alphabet of one of the ancient languages. Moreover, in the text there are 2 more levels of encryption to virtually eliminate the possibility of computer-assisted translation, even after replacing the signs letters.
    I pick up the key by which the first section I was able to read the following words: hemp, hemp clothing; food, food (sheet of 20 numbering on the Internet); cleaned (intestines), knowledge may wish to drink a sugary drink (nectar), maturation (maturity), to consider, to think (sheet 107); drink; six; flourishing; growing; rich; peas; sweet drink nectar and others. It is only a short word, mark 2-3. To translate words consisting of more than 2.3 characters is necessary to know this ancient language.
    If you are interested, I am ready to send more detailed information, including scans of pages indicating the translated words.
    Sincerely, Nikolai.

  13. Further investigation with my cipher, body sign language and stenography for Queen Elizabeth:
    Evidence is mounting for my cipher as a possible solution to the VMS.

    Quean Lille and “hELD TITLES LILLe”.

    • Stephen Bax

      Hi Edward, did you look at my last comment? If you are suggesting that the manuscript has some relation to Queen Elizabeth I, how do you explain the difference in dates between the carbon dating of the Voynich manuscript (early 15th century) and the fact that Queen Elizabeth lived in the 16th century and died in 1603?

  14. After looking at this again and applying my cipher with greater research regarding Elizabeth. I may have found provenance for the VMS. However I admit its still in the speculative stage. The hidden message reads, “hELD TiTLES LILLe”. My cipher uncovered Lille next to the nymph with the Diadem infinite 8 symbol.

    • Stephen Bax

      Hi Edward, I admire your energy and industry, but unfortunately I feel that your approach, like so many which try to find ciphers, is full of ‘false positives’.

      What I mean by this is that you find rather random sequences such as ‘held titles lille’, which I am sorry to say are meaningless, and then you work them up into large imaginary stories.

      So many other people have done this before you – supposedly finding ciphers in Latin or Italian. How is yours any different or more reliable?

      I am sorry to sound negative, but it just seems that your scheme has no more foundation than the many which have tried the same route before you.

  15. K Kimber

    What if the author was schizophrenic and it doesn’t mean anything?

    • Stephen Bax

      That is too simplistic, in my view. It is like those who simply dismiss it as a hoax. If you look closely at the manuscript it seems to me to be far too complex and structured to be simply explained away in such simple terms.

  16. Sophia

    Hi Stephen
    Have you looked at the Friulan language of northern Italy?

    • Stephen Bax

      I understand that it has been considered……

  17. Akshay

    I am no expert but i am curious dosent’t this plant look like a cannabis plant or a pink thorn weed plant

    • Stephen Bax

      To my mind it has been convincingly identified as the Ricin plant. See here.

  18. Diogenes Collytus

    To me this is just a book of plants with heavy use of symbolism. Colors and cross sections are used in order to make the physical properties of the plants more visible. This is just an idea but I think those women are some type of “mother earth life force energy/souls”. Green color with women = vital energy in plant, blue color with women = vital energy in water.

    I can see the women changing shape, as if in some type of comic book. Some are “sliding down” and some are lying down and look tired and in the picture just below it they look all “vital”. There are some symbols of “crossed pollen”, possibly depicting fertilization. There are birds next to the pollen, possibly depicting the pollen is flying. The resting bird possibly depicts it’s just about to fly etc. There are also some “childish” looking tower-like symbols with the women inside, possibly depicting the stems/branching of the plants and cross-sections.

    Then there are women with different crowns and objects (crosses etc.). It’s possible the crowns depict how strong vital energy is related to other properties (yield etc.). The cross could be a symbol of “becoming whole/holy”. Of course the book is not 100% scientifically accurate (more like childish) but it seems as if the author has tried to understand/communicate some details of plants.

    It would be very interesting to know how these people thought about these things. The text would be fascinating read, if translation succeeds.

  19. Veronica

    Dear Sir,
    The plant in the image ( page 50 in the Voinych Manuscript) is Tarragon .
    Name: Atremisia dracunculus
    in Italian : “dragoncella”- little dragon
    in French : “herbe dragonne”- dragon’s grass
    in modern Arab : tarkhun
    In Iceland this plant is also called ‘’fafnisgras“ (grass of Fafnir – Fafnir was a bad dragon killed by a hero called Sigurth).
    The root of this plant looks like a serpent/dragon. It was believed it can cure snake bites.
    And , finally , dracunculus means “little dragon”!

    Best regards,

    • MarcoP

      Hello Veronica,
      while your proposal is a good explanation for the animal at the bottom, the general aspect of Artemisia Dracunculus does not seem to fit well with what we see in f25v. For comparison, I attach an illustration of tarragon (“dragone”) from Cadamosto’s XV Century herbal, Cod. 5264 Han, Vienna.

      • Veronica

        Dear Marco,
        If you look from above at a plant , you will see the exact disposition of the leaves, as in the V.M picture. I attach a picture of the plant . Please take into account that there were and are a lot of varieties of this plant.
        Have a good day!

      • Veronica

        I found a picture with plants of French Tarragon which can offer you a better image!

        • Peter

          Be careful with animals in the picture.
          Here you see a frog in Moss, this also describes the ground,
          in this case, it is wet.

          • Veronica

            Dear Peter,

            The connection , as I see it , is to point to a certain moss variety. Indeed, there is a frog , near the moss, just to show that is a special moss – called “frog moss”.

            So, as I see it, the author gives a hint, to point the desired plant.
            Don’t you think?
            And, by the way , I have no special training in biology, I just stumbled over the V.M. on this site, and because I grow, for personal purposes, just like a hobby , aromatic plants, I recognize some of them and know their names &history.
            Have a nice day!

  20. Dear Stephen and other Voynich researchers,

    I’m a big fan of this site and its discussions. I’m from the US; I have a Master’s degree in German and have been tinkering with the Voynich script for several months.

    I think the Voynich text may be broken down into 26 or 27 patterns which form the entire manuscript. I posted this image on the Voynich Ninja website (Analysis subforum), so hopefully we can discuss there, if there is not an easy explanation why the script could break down this way. Almost all of the Voynich text (170,000 characters) can be broken into these 26 or 27 patterns that I’ve located.

    I’m curious to hear what the Voynich community thinks, and I’m just posting here to advertise the idea for discussion. My key (and an example of how it works from the May Zodiac) is attached here, and the thread I created on Voynich Ninja is:

    Thomas Coon
    Savannah, GA

  21. Alex not important

    A little remark to reference, that there might be a Georgian language involved. There actually 3 languages in Georgian language group – Main Georgian (east), Megrelian (west) and Svanetian (north). All your references to “Georgian” are for Megrelian language actually, and all of them are inherited from Greek language. Even new year in Megrelian is “Kalenda”, sound/voice is “Xonar”, mint is called “minta” and so on – definitely greek based. So I doubt that any Georgian/Megrelian contributed to that book, it is just greek influence. And also, I’ve seen a website, where a lady decipered a lot of plant names from this book, and they all apear to be named in italian, an ciphered with just letter swap?

    • Stephen Bax

      Thanks for that information. I am not at all a fan of the “letter swap” or anagram ideas you mention, because unless there is a regular and easy way for readers to decode them, they make it impossible to read. Highly unlikely in my humble view.

      • Pawel Bugajski

        Dear Mr Bax. Using this occasion can I just ask if you are aware of any effort put into trying to read the names of 30 women in circles in pages f70-f73? For me, that could provide another significant lead.

        Assuming that we are close to Caucasian/Iranian area with some strong influence of Zoroastrianism, all their days have had distinctive names. Hence we have 30 of them, I would assume these ladies in circles (which in Zoroastranism will be deities responsible for each day) will have names originated from/related to a calendar (Armenian?).

  22. Freda Edis

    I’ve long thought the Voynich manuscript may have had a Caucasian origin, specifically, in relation to the script, with an emphasis on Georgia and Armenia. If so, and it’s a big ‘if’ at present, how did it end up in Prague by the early C17th?

    From the radiocarbon dating, locating its production in the early C15th, it may have had an interesting journey from the Caucasus, if that indeed is where it originated. This was the time of the Muslim conquests of the Near East in their sweep towards Constantinople, so it may have been that whoever had the book, or the book itself, travelled or fled north to Constantinople, as many scholars and works did. When the city fell to the Muslims in 1453, it may have been that the book/author travelled to Italy along with the many Byzantine scholars who made an exodus, or the book was brought there by Byzantine scholars. That event opened up the Italian Renaissance during which more than a few non-Christian, magical, occult, Platonic, Neo-Platonic and alchemical books made their appearance and were studied, later copied and dispersed throughout the continent.

    It wasn’t unknown for Northern European scholars, doctors, alchemists and naturalists to travel to Italy during the Renaissance to study and buy books, so it isn’t a great surprise that it ended up in Prague.

    All highly conjectural, of course, but it may be point of discussion.

  23. Brittany Friedman

    Visually, page one (f1r) of the manuscript appears to be quotes, with the attribution of the quote following.

    The second and third paragraph both end with some slight variation of dwair or daur. Google Translate says that diyar is Kurdish for “said”. This interpretation appears to match up nicely with popular theory at the moment and with this quote attribution idea.

    • Stephen Bax

      Thanks – it is not impossible that the word could be a ‘quotative’ as you suggest, but we would need evidence from throughout the document.

  24. L.S. Dear Mr. Stephen, Just for your information, I would like to point out, that in Spain the “v”is pronounced as “b” and on Bali Indonesia, the “v”is pronounced as “w” and the “f” as a “p”. This might explain some curious endings of words in the “V.M.” Besides this, in my humble opinion, it is not an academic language and indeed put together with words that were known at that time and maybe sometimes borrowed, just like you assumed. I think you are on the right track and the paintings are very amateuristic indeed. Hope you read this sometime and wish you luck with “the Works” With regards Evert B.

  25. Kare F Gold, Ph.D.

    Dear Rene and Bax-

    I have recently taken an interest in the Voynich manuscript and have a few thoughts I have not seen expressed anywhere. I decided it would be interesting to share them with you and hear your opinions on them.

    When reading about the 1400ish carbon dating of the text, and that the manuscript, as a whole, appears to be a recopying of the field notes of an alien explorer documenting new plants and new knowledge, I thought about explorers of the period prior to the dating of the text.

    In addition, I was fascinated by the loopy “gallows” of the script and surveyed the cursive of alphabets up to that period of time.

    I was wondering if anyone had considered whether the manuscript of these field notes were generated by Marco Polo? As we know, Marco Polo spent almost a quarter of a century traveling the Far East, returning to Venice at the close of the 1200s. Polo generated four famous books about his travels. He dictated them to another- suggesting that his literacy was not up to this substantial a task, which is not surprising given the time period. The four books are an extensive documentation of the people, places and current events Polo witnessed and experienced.

    Polo and his family were not native to Venice, but immigrated to Venice from Dalmatia- which is now contained in modern day Croatia. Croatian is not written with Latin letters but rather uses the Glagolitic alphabet. Here is an example of text from the book of Luke in Old Croatian.

    The loopy-ness of the cursive is somewhat akin to the “gallows” letters of the manuscript.

    Is it possible that in addition to the people, places and events Polo found a way to document, he also documented the new “things” and knowledge he found? Could this manuscript complement the content of his four other books?

    As a resourceful but not very literate man, might he have used the Galgolitic alphabet of his childhood bible to write the Italian he spoke as a man of Venice? This might be done in much the way Jews used the Hebrew alphabet to write the German-style language that came to be known as Yiddish. Let’s call Italian from the middle-ages written in the Galgolitic alphabet “Poloish”.

    Clearly a scribe was hired to put the valuable info on velum. He probably made the the cursive Galgolic script simpler and less alien for his transcription, as he probably didn’t understand what he was copying. Further, as Polo’s sketches of the plants contained in his notes were probably not of high artistic quality, a recopying of them probably simplified them and emphasized key features so they were not lost. It explains how other-worldly the illustrations are. It was the result of playing the game “telephone” but by drawing pictures instead of whispering.

    Polo’s geography information did not make it onto maps until the 1450’s. So the idea of a scientific manuscript not being reproduced until this date is quite consistent with the hypothesis suggested.

    I am not a linguist. I was wondering whether either of you know of someone who might be able to get some traction in translating the manuscript, or at least evaluate the possibility that it is indeed in Poloish?


    Karen F Gold, Ph.D.
    Kgoldster at gmail dot com

    PS. I believe the bulbous green things are poppies.

  26. Mirel Turmacu

    Professor take a look at this link , and look into the Roma , in the eastern Country called Romania , there are a few elders that still speak a language that is not written anywhere , and they say its from their ancestors … maybe you guys are looking in the wrong place , while the answer was right under your noses , theres another thing you guys are doing wrong , you assume everything is English based , but there are things that the Eastern countries did as well , and their pronunciation and alphabet is a bit different than the english one…. you will never be able to decipher this with the limited alphabet of the english language , try a different alphabet … ( i apologise for my grammar mistakes , english is not my first language)…

    • Stephen Bax

      Thanks – I think you are wrong – no-one is seriously saying that it is ‘English based’. Everyone here is looking at different scripts and languages.

  27. Eduardo de Ceano-Vivas

    Hello Mr. Bax

    I first would like to commend you for your work on the VM, I’m not aware on any other academic who has published any serious work on the subject.

    I wanted to bring to your attention this website: which I ran into by accident. It seems to have many good theories and comments.

    Best of luck to you,


  28. Roland L

    Hello there. I came across VM a long time ago and i´m looking att it from time to time. Now for a few days i´ve been having a strange thought about it which i´m currently checking out. Nothing to say about that at this time.
    After i saw your video and looking through your paper and ideas i really started to be bothered about your Hellebore idea. The name fits, but the picture? I´d rather be comparing page 57r which i think is very very similar to the hellebore plant according to your other pictures on page 35(bottom pic) and 36. These plants are more or less identical to the plant on page f57r. Root, leaves and even the flower. Don´t you agree?
    At the moment i´ve got nothing more but will continue to look through my own idea.
    Oh, and one other thing, page f65r. It makes me think about a chestnut-tree. at least the leaves and fruit.

    • Peter


      • Peter

        f34r Hellbore

        all at work

        • Peter

          all plant from the same area !

          • Peter

            same time more then i like

            • Peter

              is it…or not ?

      • Derek Vogt

        You’re up to 28 plant identifications now, several of which contradict the identifications we/I have been working with so far, which yielded the phonetic system I’ve posted a lot about. So, if your plant identifications are accurate, at least some of the others we’ve had so far were wrong, which takes away some of the basis for the phonetics. But yours could, on their own, yield a completely unrelated phonetic system to replace the current one, which would, once developed enough, have better linguistic results.

        I’m going to try that. I have no idea how long it will take.

        • Peter

          Glad if you continue brings. Of course, not all there. But most are in labor. Am also only in name with T. Unfortunately, even those here where a lot of species are true.
          Hope Mr. Bax this in a single place for a better overview.

  29. Dear professor Bax and people who read Bax’ site,

    I recently finished my paper on Greek imagery in the root-and-leaf section of MS Beinecke 408. The paper turned out rather big, so I decided to chop it up and present it in blog form. It can be found at .

    The blog is up and running since a couple of days. I will update it with one plant analysis daily. I will demonstrate some of the following points:

    – The roort-and-leaf section of MS Beinecke 408 shows evidence of originating in a Hellenistic setting. In this, I follow D.N. O’Donovan.
    – I add my own form of evidence to her conclusions: one particular foldout in the root-and-leaf section contains plants with Greek mythological imagery hidden in their roots.
    – The root-and-leaf section is an illustrated glossary. One of its aims is to teach foreign plant names to Greek speakers.
    – In the foldout I investigated, these foreign, “to be memorized” plant names are Indo-Iranian ones.
    – All plants on this foldout are important ones found on the naval spice trade routes around the Indian Subcontinent. So far I have published mango, millet, saffron and sugar.
    – By analyzing these plants’ labels, I have gradually been able to construct my own understanding of Voynich script. Very broadly speaking, I see it as a double character set, with an ornate and a normal form for many characters. On average there is a “two-to-two” relationship between glyph and sound.

    So in short: the root-and-leaf section is a handbook for “outsiders” (Greek speakers) who want to become active on the Eastern trade routes.

  30. Will Bryant

    I found the ideas in reference to a reversed syriac alphabet interesting. As many of co-workers are Kurds I soon thought of their language as a potential source language, based on history and geography. Due to my limited language skills my best hope for helping will probably be in selecting a single area of the VM to focus on for study and seeking correlations. I hope I can assist in some way.

  31. NicBob

    This also looks like greek.

  32. NicBob

    I don’t know much but I see a word that looks like a Greek name.

    • Təxəllüs

      NicBob- Padania, if you do not already know, is an area in Northern Italy. A good find- one that has already been discovered, as research suggests, though- what section of the manuscript is it in, by chance?

  33. Dan Brown

    The book looks to be a Danish or Dutch rutter used by ship doctors who doubled as navigators and were the general science advisors to the captain. The text looks to be Northern European with Latin a doctor would use for herbs.

    • Ruth McNabb

      That’s interesting, Dan, but how then would you explain the pregnant women?

      • Christy Joy

        I do not think it absolutely depicts pregnant woman, It is possible it just depicts women rather than girls.

  34. Ruth McNabb

    I bumped into the Voynich Manuscript on one of my many Internet rambles a few days ago. I’ve been unaccountably fascinated with it since. Having no expertise in languages or botany or Renaissance anything, I can contribute nothing to the unravelling of the text. However, I am curious about opinions anyone has about the personality of the author. Not who the author was in name but who he or she was in character.

    I toyed around with the ridiculous idea that the author may have been hearing impaired and illiterate (as to his native language) and what type of text such a person may have produced. I didn’t come to any conclusions other than it would have to be symbolic, and most likely have meaning only to himself.

  35. Bryan

    I kept thinking that the script was reminiscent of Georgian so contacted two Caucasian language experts.

    Jost Gippert at Goethe University, Frankfurt, said, “No, I don’t see ANY relationship to Georgian calligraphy or Caucasian scripts in the Voynich manuscript. In my view, this is Latin-based but meaningless (consider the amount of repetitions).”

    Kevin Tuite, Dept of Anthropology, Toronto University said, ” I do not see any resemblance between the Voynich writing and the Caucasus alphabets known to me (Armenian, Caucasus Albanian, and the three Georgian scripts)… My first impression is that this is an artificially-created cypher, concealing a text in some natural language (presumably a language of medieval Europe). ”

    Sounds like the possibility of a Caucasian link can be discounted.

  36. Ladee Jay

    I would like to suggest to you that the Voynich manuscript was not intended for all of “man”kind but possibly just for women or made for a specific woman to use as a way to predict the best times to get pregnant and that it may have in fact been written by a women.

    Most of the humans pictured are women.
    Many of the women appear to be pregnant and most are fully naked.
    Men are rare and usually appear clothed with privates covered.

    Many of the female bathers are wearing a style of hat seen often in artwork and portraits of women in Middle Ages Europe.

    One illustration shows a male holding a crossbow wearing medieval style pantaloons. The crossbow is the astrological symbol for Sagittarius.

    Japanese and Chinese cultures use animals for their 12 zodiac signs. Leaving the strong suggestion that these astrological charts are not of an Asiatic origin.

    On page 134 the water is green in the top pool with the 10 women and in the smaller pool towards the bottom the water is blue. The herbs and flowers could have been infused into the bathing water turning some pools green and some blue.
    Page 210 This appears to be a celestial chart of some kind.
    The 12 red colored rays could represent the sun and the 12 months of the year.
    The yellow moon face in the center and the 12 dark blue rays could represent the moon at night. On the opposite page there are 12 alternating yellow moons and red suns.

    It’s obvious some part of this manuscript are about astrology and some parts herbal (medicines?) but when it comes to the human aspects of the imagery there seems to be one gender represented here far more than the other. I’m sure there is a specific reason for that or it would have been more gender neutral or equally represented in the art work.

    • Christy Joy

      I was wondering if this may be woman-script? Between the subjects of astrology and plausibly botany, and women… I wonder if this crosses between midwifery, astrology, perhaps paganism. Thank you Stephen Bax for what you have studied and shared so far.

  37. Travis McGeehan

    Has any attempt been made to identify active printers/bookmakers in the 15th century who could have produced a work on the scale of the Voynich Manuscript? It seems unlikely that someone who could make a book of that length never made another one. If the manuscript could be linked to a bookmaker then we could identify the underlying language through comparison to their other works.

    • Stephen Bax

      Not that I know of, but maybe others have more information on that?

    • Alex not important

      I believe, in the 15th century there were no print houses or copy shops, where you can go and order some print 🙂

  38. Stuart Nichols

    Why is it all the images I see for this Document appear to be reversed images?

  39. JWalsh93

    Firstly, truly wonderful work sir! In a couple of years you may be able to turn lead into gold!- or perhaps just make tasty cordials! Either way, bravo! Secondly, I wondered if you had made any headway on the manuscript’s authorship? I believe that it could have been the works of a group, two members being the early Renaissance painter Fra Angelico and his wealthy patron Cosimo di Medici. Do give me an email if you think there is some weight to this hypothesis, I do have ‘evidence’. Thanks for your time, and good luck!
    Jake Walsh

    • Stephen Bax

      Thanks – that theory has been advanced before, I believe, but again the question is how to get evidence for it! I prefer to take tiny steps with the language before thinking about the possible author.

  40. Koen G.

    Hi everyone

    I have a question mostly directed to those involved in trying to match Voynichese with old languages. I’m just starting out, and having a hard time finding reliable translations. Greek and Latin are relatively easy, but what about Eastern languages? Using google translate can be a beginning, but won’t work for everything and besides the language you get is too modern.

    Are there any tricks, tools, websites… that you use? Or is it always a case of digging through google and historical sources to find each word? What’s your experience? Any tips are welcome!


    • Stephen Bax

      A big problem with the approach which most people take is to try to match words here and there, often using Google Translate, and then string them together in a sort of pseudo-sentence. It usually results in Gobbledygook, and cannot tell us anything about the underlying grammatical system.

      As for advice, I personally feel that the best way is to try to identify letters and words, starting with possible Proper Nouns, using the context in which they are found – as many of us on these pages have been trying to do. See Derek Vogt’s work here as an example of how this kind of approach can lead to quite detailed patterns.

      • Koen G.

        Thanks Stephen

        I do agree that proper nouns are the way to go. My question was more meant to ask what you do after you’ve identified a proper noun. Say that I’m 100% sure that I’ve found the Voynichese word for “dolphin” or “cup”. Now that would be a major breakthrough. Now I can find the Latin and Greek for of those words, with an added bonus that they are ‘timeless’. But how do I find out how, for example, the Arabs used to call dolphins and cups before 1450? I think in the case of most languages, google translate will be useless because there’s more than half a millennium between that and Voynichese. It’s like using google translate on Middle English.

        Now I guess the answer is “look hard”. But still I wanted to see if maybe someone wanted to share some tips or sources that worked well for them 🙂

        • Derek Vogt

          The speed of change in English in that period has been out of the ordinary, so it’s a bad example of how much change to expect for most others. For most languages, modern and historical words are often the same or at least close enough to be recognizable, and some degree of imprecision is to be expected and accepted as “close enough”, especially for words that also bounce from one language to another.

          Also, more than one answer is often available from general translators like Google’s, and modern sources that specialize in listing names of certain kinds of things in various languages tend to include multiple names per language wherever necessary to account for different versions given by different historical sources. So, for example, Hydra is given as Ὕδρη, ῾Ύδρος, Ὕδρα, Κόραξ, and Δράκων (Húdre, Húdros, Húdra, Kóraks, Drákon; all Greek), and Hidra, Idra, and Ydra (all Latin), while coriander/cilantro is given as cilantro, culantro, and coriandro (all Spanish), and dʰanya, dʰana, धने (dʰane), कोथिंबीर (kotʰimbīr), and kotʰimber (all Marāʈʰī).

        • MarcoP

          Hello Koen, my amateurish experience is that tools and websites are of limited use. Linguistic research on an ancient manuscript requires a certain familiarity with the language in which it is written (or a very close relative). For instance, attempts to read the Voynich script as Italian anagrams are clearly meaningless to an Italian speaker. I am pretty sure that my attempts at languages I don’t know are just as useless (still I often enjoy this kind of pastime). At the current stage of research, I think that a Romany speaker would be in a much better position to contribute then myself.

          Also, when reading and translating Latin manuscripts, I find that most of the difficulty is transcription: making sense of the script and the abbreviations. OCR technology is far from being of any use in this case. I can refer to on-line available resources (e.g. Cappelli), but only because I have a passable knowledge of the language. Greek manuscripts are beyond my reach (but for a few words, with luck and a lot of context). For Hebrew, Persian, Arabic, Sanskrit and all other languages that are not related to the Latin alphabet and language, I am pretty sure that on-line tools cannot help me, since I lack the basic linguistic knowledge. I guess that this is what makes speculation about images so popular among Voynich hobbyists (myself included).

          Anyway, here are some resources you could consider:

          An Arabic Etymological Dictionary (I only recently downloaded this and I am not sure of its actual usefulness).

          These have been mentioned on Stephen’s blog in the past:

          Digital Dictionaries of South Asia at the Chicago University.

          The Global Lexicostatistical Database Search common English words in a large database of languages, including ancient languages.

          CRC World Dictionary of Plant Names by Umberto Quattrocchi.

 Very useful when searching for patterns in the Voynich language.

        • Stephen Bax

          Ah, Koen G, I see what you mean. Well, in the case of ‘centaurea’, the plant which was identified in folio 2r, I did look through many manuscripts in several languages to try to identify the name. So the answer to your question might be to look at manuscripts in those languages (rather than just dictionaries) to see if the word you have found is also found there. You can see that many people on this site do exactly that.

          Not perfect, but probably better than a simple Google Translate search!

  41. Brad B

    I have read a lot of the comments here and have reviewed your work and think it is very intriguing. I have studied the manuscript previously and find myself going back time after time to look again. I have not read all the material published on it so I don’t know if my impressions have been put forth already but I do get an overall sense of the document and your work backs up a lot of what I already thought about it. It appears to be a medicinal journal for a physician of some sort. Not witchcraft but very practical medicine with herbal remedies. And in particular mostly for women. The zodiac charts with women all around were fascinating and a real question as to what they mean. I finally seem to have figured it out and makes sense that fits with the rest of the manuscript. Each chart gives the doctor the ability to tell any woman based on her sign when the best time to conceive and when to not conceive. In the bucket out of the bucket? Also either favorable times for a conception for a particular trait like the strongest or best offspring. Other pages that show multiple women in baths or in a pod container deal with stages of pregnancy and remedies to ease pain etc. Other star charts without women may be for the doctor to advise men the best time for boy or girl of time to conceive. I also noticed most of the plants are exotic to say the least and are probably also medicinal type plants. I haven’t seen this mentioned previously but when looking at the book overall this seems to be a doctors tome.

    • Stephen Bax

      Thanks – yes, the zodiac charts could be for conception, but what would each word beside each figure mean?

      Thanks for your ideas.

  42. Roxanne Olson

    Hi! You should know about this and connect with this info/manuscript/researchers if you haven’t already:

    • Stephen Bax

      Thanks, yes, that is a fascinating case! Rather different from the Voynich manuscript, but still of great interest.

  43. FILIP

    Colleagues. Has anyone have seen in other folios or private correspondence such writing of the letters (Figure 1)?
    Similar transparency occurs in different places in the text when running out of ink. To get the transparency necessary to have a pen sharpened, as shown in Figure 3 or 4. It is even possible to use some kind prototype of a metal pen. This will help determine the “school” of scribes or a specific author.
    Figure 2 – sharpening the pen for left-handers.

    • Neticis

      I don’t agree with your assumption about pen type.
      I have done similar mis-writing with just V-sharpened dip pen with less ink and more force to the paper.

  44. EWilliams

    There are a couple of things I’ve noticed about some of the letters and please excuse me if this has already been discussed.

    1. The letters transcribed in FSG as CT, H, P, K, and F: it is interesting that the letter transcribed as CT can be “combined” with H, P, K, and F to get the letters transcribed as CHT, CPT, CKT, and CFT. It’s almost as though the CT glyph modifies the others in some specific way. As an example to illustrate this, I’ll use lenition from Gaelic. The letter H is inserted after the initial letter to change it from a stop to a fricative in most cases (an approximant in select environments):

    eg. P + h = Ph /f/
    B + h = Bh /v/ or /w/

    This also reminds me of the Tengwar from Tolkien’s Elvish languages. He modifies the initial 4 characters representing P, T, K and Q to add voicing, aspiration, etc. As it has already been suggested in Dr. Bax’s original paper, the script in the VMS may have been created to encode a language that had no writing system. It stands to reason that the individual glyphs may have been created in a systematic pattern in indicate such changes in sound as noted above.

    2. The glyph transcribed in FSG as ET: Again, I’ll use Gaelic to illustrate my point with this one. This glyph appears to be the “CT” glyph with a diacritical mark to create the sound represented by “ET.” In older Irish transcription, such diacritical marks were used to indicate the lenition of the subject sound as noted above; in the case of Irish it was a dot over the letter which indicated it was to be pronounced as a fricative. It is possible that “CT” and “ET” may share such a relationship.


    • Derek Vogt

      According to this page, you seem to be using the Bennett system, not FSG. Most of us here are used to EVA (shown farther down on the same page), which reverses Bennett’s “T” and “H” but is otherwise the same for the “gallows” letters (the tall ones with one or two vertical legs at the bottom and one or two little loops at the top).

      You are right that some other people have thought before that any of the three simple binary options on the gallows letters (one loop or two, straight legs or one bent leg, ligaturized with EVA-ch/Bennett-ct or not) might each have represented separate phonetic choices. But unless we can identify the sounds of at least a couple of those letters and compare them with each other, there’s no way to tell which ones meant what, because there are so many choices. Any of them could have toggled place of articulation, manner of articulation, voicing, length, or the presence or absence of any secondary articulation (effects like pharyngealization, aspiration, palatalization/iotation, retroflexion, labialization, and prenasalization).

      Similar sound differences have been represented by letter shape variations before, in various known real alphabets, including the Latin alphabet forming G by adding a bar to C, J by adding a hook to I, F by tilting & bending the top of Y, V by losing the bottom stroke from Y, U by curving the angle of V, and W by doubling U or V. But sometimes looks can be deceiving. There’s no such connection between X and Y, or between O and either C or Q, or between P and R, or between E and F, or between B and D. I think that, whether it’s real or illusory, the Voynich gallows group’s eight symbols and three dimensions of variability would be the largest and most complex set like this ever discovered outside of India, and would make up the highest faction of its whole alphabet.

      Another clue is that EVA-ch/Bennett-ct shows up disproportionately often right after the gallows letters, as well as superimposed on them, as if that kind of sequence has a separate special meaning, like English’s “sh” and “ng”. So, if visual patterns like this have phonetic meaning instead of just being coincidental, then that letter must represent a sound that can combine with the gallows letters’ sounds in two different ways: one superimposed/simultaneous, and the other sequential.

      The only solution I know of so far (mine) has been that EVA-ch was “h”, the gallows letters were plosives (p-b-k-g), the sequences with that “h” after them were aspirated plosives (pʰ-bʰ-kʰ-gʰ), and the ligatures are fricatives (f-v) or affricates (č-j) which have been known to evolve from those plosives more than once in the real world. And the nature of the gallows plosives themselves is identified by those letters’ shapes alone, even without considering “h”. Also, in that same system, fitting your suggestion, the letter that looks like that “h” with a diacritical addition on top represents what would be called an “emphatic H” in Semitic linguistics: similar to “h” but with more constriction somewhere in the throat or back of the mouth.

      I’m not aware of another complete set of sounds which anyone has suggested to encompass all of the Voynich letters we’re talking about, either with internal patterns like that or without (which is a bit of a disappointment to me; I was hoping to get to compare multiple proposals with each other). There has been a suggestion that EVA-ch/Bennett-ct represented “y”, which would imply that combinations with it could be palatalized/iotated, but that idea hasn’t been associated with any specific sounds for the other letters we’re talking about. Similarly, there has been a suggestion that EVA-T/Bennett-H represented “L”, but that hasn’t been worked into a more complete system with the other letters in this group.

  45. Rashid

    Dear Prof. Bax,
    I am a Ph.D. in natural resources and environmental sciences, from Iran. Recently, I have reviewed the VM book and have seen the comments from the fans in this page.
    I think the book is not an ordinary book, it is in code because it is talking about herbal medicine, poisonous herbs and narcutics. on page 65 (online version) you can find a Papaver somnifera L. plant of opium raw material (it has rooted in humens brain in the page!), you can find the pot (hash) bush in page 31 or castora seeds in page 15, or Passif lora caerulea (clock flower) which is dangerus for pregnant women(for abortion risk):page 47.(however is not in a right scale). but the writer has code the context and the drawings simultaneously by changing the size, the color and the scale of the plants. however he/she was not able to hide his interest to the main organs and desinged them precisely. ( I am also practicing drawing and design). the writer seems to changed the color of petals to different color to not to be distinguishable for commons.
    it dose not seems that the book has an independent language but it seems translated just into a coded text for privacy.
    as you have mentioned above, there are many repeatative words in the context, dosnt seems to be part of the main context may be just to fill out the pages. if we pay attention to the nature of the book, it is a guide line of herbal medicine so we expecting to find the ingradient (the main parts, like petals, roots, stems and …) and amounts like weight or valume. the useful parts has been mentioned in 3 main colors (blue, green and red). it seems blue for poisonus parts.
    in astrological part, i think it is belong to pregnancy time and monthly period of women. it is clearly indicated the women pregnant in the continer, blue and green indicates the effect of plants on pregnency, I think according to the state of moon, stars and zodiac signs.
    I have also found a unique phrase in page 142 end of line 9, look likes “2020209” you can find 02, 09, shape characters every where in important words, in drawings and you con not find them in repeatative strings something like “qoHc89”.
    I aslo would like attract your attention to the first character of the page which is like a snake with 3 dash part on. I have exclude repeatative strings from the context and marked key characters. I found that the first 3 line just is repeatative but the next 3 are unique and the character like “09” repeated in each “02” the other string!
    sorry for a long comment, eagerly would like help in this way.
    thank you and cross my finger for your success.

    p.s. some other interesting findings are available.

  46. I have made a character statistics visualization using a multidimensional scaling algorithm, available here. In short, the algorithm puts contextually similar characters close to each other. The result has some curious features. For natural languages, the algorithm produces clearly visible clusters of vowels and consonants. But for VMS, the clustering is far less prominent. This is consistent with an Abjad, as devoweled English also has no clustering. The only thing resembling a vowel cluster is the i, ii, iii group. Other features: p is very close to f and likewise p is close to t. Could they be merely a writing variations? r, m, n are relatively close, which is consistent with Stephen’s decoding which identifies them as r-based sounds.

    • This is an interesting experiment. I should recommend to repeat it with a different transcription alphabet, e.g. Currier’s or v101. Eva is not the most suitable one for that. One result to be expected is that the characters in Eva: (ch) and (sh) will be close together. These are not single characters in Eva, but they are in the other two mentioned.

  47. Hi,
    I believe this is a medical book for women.
    since you translated “Taurus” then you should be able to get the rest of the zodiac signs. I think it determines the day, month, year, the stars and planetary alignment they were born under, and the age of a women to determine common physical problems and possible cures. I also think the green liquid in the tub with the women could be for fertility.
    It also shows how to grow, harvest, and prepare the plants for medical use based on the months of the year.
    Also, I read somewhere, that the zodiac signs use to represent specific parts of the body.

    • One more comment: you can’t see it in the big picture; but, there are 4 women at the top. I believe they represent a virgin, a women who has had a child, a women in menopause, and old age. But the zodiac sign of Sagittarius, might indicate the month of November in which case there are 30 women, one for each day of the month. Another picture suggest different treatment at different times of the month

  48. My first look after accidentally seeing your Youtube vid. I recognized some of the words by sounds and the writing is familiar. What I believe you have here is an early attempt at the transliteration of the knowledge of the Rromani peoples, the now termed European Gypsies. The time lineage, circumstances, mixtures of languages and ‘codex’ seem to approximate. The main languages by diaspora, Domari, Lomarven, Rromani by westward migration 1200 to 600 BC. The Lyuoli of the eastern migration. You may wish to contact our main linguistic specialist in Texas (Rromanichal). Prof. Ian Hancock. Austin Texas.

    With kind regards.

    John Holt
    Xianggang (Hong Kong)

    Ian Hancock
    E-mail: [email protected]
    Phone: +1-512-232-7684
    Office: CAL 420
    Campus Mail Code: B5100

    • Stephen Bax

      Thank you. I, and others, have certainly considered the Romani possibility – see Derek’s recent post about the linguistic dimension – but as yet there is not enough evidence to make any convincing case for it. Still, as you say, there are lots of interesting aspects there which are worth researching.

    • Claudette, I think it is a mistake to include that stream-of-consciousness, a-historical, rambling bit of modern new-age associative fiction about tarot cards with your efforts to translate the Voynich text. Since that has no historical credibility in terms of medieval thought, your thinking it worth citing here casts doubt on your own judgement.

      • It’s a quotation from Carteri’s Le Imagini Degli Dei. The book was an immense success, it went through numerous editions and was translated into many languages, including English in 1599. It became the iconographic handbook of painters throughout Europe for the next 250 years. The woodcuts are based on those of Bolognino Zalteri printed in the Venice 1571 edition with some revision for modesty’s sake. Catari’s book on the images of ancient gods was first published by Francesco Marolini in 1556, and the first illustrated edition followed in 1571. His work on the gods and their symbols is a key work in the history of symbolism in the Renaissance and the manual for artists as to the proper attributes of the dieties. [See Seznec’s The Survival of the Pagan Gods for a full discussion of the importance of this text.] According to Seznec, Cartari (b. ca. 1500) was probably a protégé of the duke of Ferrara. In his work Imagini colla sposizione degli dei degli antichi (“Images depicting the gods of the ancients”) of 1556, he identifies Lilio Gregorio Diraldi’s De deis gentium varia et multiplex historia of 1548 as one of his sources. Despsite the success of De deis, Cartari’s work competed with it by going for a populist approach, with woodcuts of the ancient gods by Bolognino Zaltieri. Cartari focused on the gods’ iconography and explained their clothing, expressions, poses and attributes. Cartari’s work was an influence on that of Gian Paolo Lomazzo. Unlike the treatises of the other Italian mythographers (Bocaccio, Conti, Giraldi), Cartari’s work was profusely illustrated with captioned images of the pagan gods, and composed in the Italian vernacular. The systematic integration of text and image constituted an original approach to the classical myths, and the use of the vernacular made the text accessible to learned and unlearned alike. Cartari’s iconographical, symbolic interpretation of the images of the pagan gods as they were represented in antiquity and discussed by Renaissance antiquarians proved to be an enormously popular approach to pagan myth. His ‘Images’ was well known to Renaissance artists, poets and critics – “an extremely useful work for historians, poets, painters, sculptors and professors of polite literature.” Nichols is simply where I found the quote.

        • Claudette,
          I’ve just noticed your comment of October 11th.
          I’m aware of the works you mention, and have often cautioned my readers on the weaknesses – while praising the strengths – of Seznec’s work.

          The chief difficulty in applying the content of the Italian renaissance handbooks to interpretations of the Voynich manuscript is, first, that there is no internal evidence in the latter to suggest any connection with the Italian humanist movement or its art.

          I am aware that some writers believe that the hand(s) used for the written part of the text have been influenced by the humanist style, but there remained informed objections to that idea – Barbara Barrett’s among them.

          It is plainly true that the imagery in the Voynich manuscript, and the page layout too, display no evidence whatever of being a product of the Italian renaissance, or even of that period. A number of highly competent experts, without any particular allegiance to other persons or to any “theory-narrative” made that observation very early, and it still holds true.

          Our fifteenth century manuscript shows no evidence of being a fifteenth century artistic creation, but rather a fifteenth century copy, and/or redaction or recension, of matter that originated in a very different milieu.

  49. Tabita Chiriac-Merlea

    Part of plants are there for their pigment (eg f65r – Rubia tinctorum). Thus F102v2 page above is a blue square to bleach or coloring powder. The bottles are of different colors, filled with different powders: blue, white, yellow, red … (I translated from French with google)

  50. Tabita Chiriac-Merlea

    The correspondence letters can be found on the first page (F1R), to the right. The alphabet from A to Z is written vertically in paralel with the signs that are almost extinct. With the blue light we can read them perhaps ? See you.

    • Stephen Bax

      Thanks Tabita. You can see this here:

      I am sure some other researchers have commented on these letters. Can anyone give us a reference to any earlier discussions of them?

      • MarcoP

        Many thanks, Tabita, I knew nothing about this!
        Stephen, I think that people at the Lazarus project could likely make some more letters readable.

        • They were looking at this indeed. The idea was to combine the frequencies that would bring out the iron in the ink, but I was told that this would be a very significant effort. I don’t know what has happened with this idea, but I can ask.

          The alphabets are most likely from a later would-be decipherer. What can be seen of the letter shapes looks to me more like 16th-17th C than 15th C, but this is just my guess.

          In the 2009 documentary Gerhard Strasser has another suggestion about the possible origin/meaning of these tables.

          • MarcoP

            Thank you, Rene! I agree that an annotation by a later “reader” seems the most likely interpretation.

  51. 25r – Mercurialis annua? And what about Ilex aquifolium?

  52. Many of the words beside the women depicted in the Voynich Manuscript’s Libra calendar are names that exist to this day, especially in Finland, Estonia, Sweden, Norway, Latvia, and Lithuania. This may suggest that what we see in this picture is a group of women whose birthdays fall under the zodiac of Libra.

    • Val

      The king you are talking about has been crowned in 1524…

  53. Marco Vidari

    Dear Mr. Bax,
    there are no typing mistakes in the manuscript. I would like to have your opinion about that. Is it humanly possible not to make a single mistake when hand-writing so many words? Thank you, Sir.

    • Stephen Bax

      In fact there are a few places where the writers seem to have made corrections or alterations, but if you look at many mediaeval manuscripts they are also very carefully written. Some say this is evidence of copying – I can’t say. But for me it is not as remarkable as people suggest.

  54. Kyle McCartin

    Greetings, I have looked into the manuscript and your work upon it and through my own analysis (of primarily the illustrations and recorded plants) and come to a few points that may or may not be a new perspective. This doesn’t look at the decoding of the language however may help in gaining context from the (seemingly random) illustrations, which may lead to better identification of themes in the language

    1. the regional context of middle eastern/European influence, as you stated placing it around Turkey or Western Persian influence such as Uzbekistan.
    2. A first examination of the illustration (primarily the later involving women) first gave me a notion of either a berry pressing (for juice extraction) the green colour for the extract and the blue for natural water. however the connections between pictures that seemingly look like pipes alluded to an almost communal infrastructure (possibly a public bathing area for women) . this though has to have connection to the extracts or plants may be for skin cleansing, or for perfume.
    3. this made me look at some of the plants described and analysed by yourself and your colleges. Hellebore is a cooler climate flower common used in Indian, Tibetan cultures as an aphrodisiac for women. Also Cotton Root is also used as an aphrodisiac for women(the root utilisation may be one some plants in the manuscript heavily exaggerate the root section of the plant). furthermore in particular illustration when women are depicted by themselves in a container (presumably a bath of some kind) one illustration shows a motion of a women holding a spoon (assuming consumption of an extract or elixir).
    4. the Astronomical illustrations may be a calendar, instruction of use. It could be a guide to when a women should bath in certain extracts (or consume) to effectively reap certain benefits from the extracts.

    these lead me to believe by looking at the illustrations (without linguistic context) that it may be a guide to the collection and use of plants and their preparation into concoctions and then as to how and when to consume/bath to increase fertility,hygiene or even coincides with human fertility cycles for Women.

    – Thank you for you time and consideration (this is just speculation, i could be completely wrong, but who knows.)

  55. Vladimir Dulov

    Mister Bax . I have two question to you. 1. How do you explain that there is no word 8auU “end” on pages f90r, f53r, f48v, f26v, f17r. In your opinion text on the pages f48v, f90r was written in one sentences and separated into indentions?
    2. Many explorers think that apostrophe changes sound of letters п, 4, d, г. But how explain writing apostrophe in one line with word (,поя) on page f93r?
    I can explain these questions by my method of symbol writing.

    • Stephen Bax

      I’m sorry to say that I can’t explain these details at the moment. I feel we need a lot more evidence about the script and language before we can understand every small feature.

  56. Koen G.

    Dear Mr Bax

    I have discovered VM research through your website, for which I am thankful. Now I, also, have contracted the Voynich Virus, and I lie awake at night pondering the script and its illustrations (which isn’t always a good thing because the nymphs kind of creep me out). I am happy to see a down to earth linguist’s view on this subject, a solid rock in a stormy sea of crackpots.

    I have a couple of questions, forgive me if they aren’t too original. VM research is rather fragmented and hard to keep track of.

    1) Have attempts been made to identify lexical items that only or mostly appear in certain sections? For example, if VM ‘olorom’ (just making this up) mostly appears in the plant section, usually in the second block of text (i.e. next to the plant’s root), it is likely to mean ‘root’? Or if a certain word mostly appears near illustrations of the moon, maybe it means ‘moon’?

    2) The theory I like most, is that the VM author was someone who mastered speaking and writing in an Indo-European language, but only knew how to _speak_ a second (Caucasian?) language. The VM would then be his attempt to create a writing system for this second language. Could the strange linguistic aspects of Voynichese be attributed to the assumption that there was no written tradition yet for this language, or that the author had no knowledge of it? This might result in unusual elements of spoken language infiltrating the document, like the many repetitions. I know of Spanish speakers who repeat the word ‘muy’, ‘very’ many times before finally getting to the adjective they want to intensify (e.g. muy muy muy MUY caliente). Sure, medieval manuscripts were meant to be read out loud to begin with, but still there must have been some degree of formalization. Could this lack of a written tradition explain some strange features of the VM?

    3) Can we really dismiss the “14th century hoax” theory entirely? I also lean towards the natural language approach, and I think you are on the right track, but at the same time I don’t understand why you seem to give the hoax theory so little credit. I think most arguments against it can be countered fairly easily. Am I overlooking something?

    – “It would be too expensive.” There were plenty of excentric rich people, and people backed by them in the 14th century. If you imagine the VM author as someone working at a court, you could also imagine him obtaining the money (or materials) for this project.
    – “What would be the motive?” There could be a number of motives. Maybe the VM is purely artistic. Maybe the author had a way to make money off it, for example as a diviner helping women before, during and after pregnancy. Maybe it was a matter of prestige, to posess an even more powerful spellbook than one’s competition.
    – “The VM is too language-like to be forged”. What if the author used a number of existing books as a basis? He copied the drawings from various works and gave them a mysterious twist (he wasn’t good at drawing to begin with). He transcribed the text “on the fly”, keeping some of the letters (mostly vowels) but changing others to more obscure characters. He used some easy “rules” to obscure everything (add Voynich affixes in certain cases for example). This would explain why Voynichese appears bloated and shows some unnatural regularities.
    He was consistent, but didn’t mind or correct any incosistencies, since it wasn’t supposed to be read anyway. Then he added some incantation-like lines here and there, which he could ‘read’ during his sessions.
    This would result in a natural-looking language, which is still impossible to decipher because it was transcribed on the fly with a fair amount of variation.

    To rephrase my point of view on this third question: I also favor the theory that the VM is a natural language in an unknown script. But I also find it hard to completely let go of the hoax-theory, since I don’t see any decent arguments against it. So what am I overlooking?

    Thanks in advance for any answer!

    • Stephen Bax

      Hello, and thanks for your interest. To answer the first question, yes, this has been tried, using concordancers for example, but with no clear result. As for your second point, it seems very possible to me.

      As for the hoax idea, it just seems so unlikely to me. Apart from the other reasons I have mentioned, why would anyone spend hours writing the last pages, which have no illustrations, and seem to be lists, if they wanted to hoax people? It would simply be a waste of time and effort and resources for nothing. It just doesn’t ring true. Hoaxers tend to be rather limited, greedy folk!

  57. Bradley

    these drawings are not of plants; they are equations and mechanics and blueprint plans of mechanical structural/industrial fabrications painted green and drawn organically, for example the stem of a plant would represent a fuel line for example.
    Picture attached is a multi stage space rocket.
    If it looks like a plant it must be a plant, completely forgetting the cryptic nature of this game. You see what you are shown.

  58. Vladimir Dulov

    HELLO Mr.
    Professor Marcelo Montemurro was interested of my work. Please change with him your opinion.
    In my last diagrams (9 july) the word “fresh” means “juice”.
    There are some symbols in my view.

    • Vladimir Dulov


    • Vladimir Dulov

      decode 2

    • Vladimir Dulov

      decode 2

  59. LGR

    Thank you for your article and your site, which gave me a peek at this extraordinary manuscript.

    Please forgive me if this has been suggested elsewhere. Page f73r of Quire 12, with the several ladies in two concentric circles (and the “lizard” at the center), appears to be a calendar to me. The ladies are each clutching a seven- or eight-pointed star of varying colors. Could these stars be related to the “recipes” later in the book (Quire 10, page f103r)?

    This MS really looks like a medicine/recipe book for women. It would be fascinating if this were a recipe book for getting pregnant (“make this potion on this day of the month and drink…”), and/or for avoiding pregnancy, and so on and so forth along those lines. The Ros page in Quire 14 just screams ovaries to me!

    I’m bookmarking this site and the Voyage the Voynich site.

    Thank you again for your work.


    • Stephen Bax

      Thanks – yes, that idea has been suggested before in various ways, but no harm in getting a different view on it!. The ‘lizard’ you mention is often taken to be Scorpio, as it is part of the set of ‘zodiac rings’. See some discussion here:

      • Richard

        Did you notice that the collection on star sign animals all have the mouth written underneath an that there’s no scorpion if it is a lizard why would that be ?

  60. Vladimir Dulov

    Decoding of mix ingredient out of concrete plant on recipe.

    • Stephen Bax

      Vladimir – I have NO idea of what you mean by these diagrams and numbering. Can you please tell us what each of them means?

      • Vladimir Dulov

        Take from the specified plant.
        1. juice from the top big leaves.
        2. juise from the lowest part of the root and bottom part of the stalk.
        3. juice from the top part of the root , the bottom part of the root and stalk.
        4. juice from the root.

  61. Vladimir Dulov


    • I would guess the language of the Voynich is sophisticated enough to have a way of saying “many” and “very” and “high” without resorting to repeating words or making the letters mimic the drawing to convey a quality. My guess would be that what you are calling a leaf is in fact the word “kei.” Depending on how you sleuth it out, this word could signify gray/hoary or rocky or moving back and forth or hard and narrow like a stick. It could be related to the English word “key” or the Norman word “caillou” or the Old English word “kinan/cinan” meaning to burst into bloom. In other places you combine “kei” with “ei” and call it “juicy leaf.” Again, I’d beg to differ. There is a good probability that this second word “ei” means “not,” so that the word combination may be “not kei,” whatever kei is. As for the symbol of a lack, I’d have to disagree again. I think that’s a “j.” Then your lacking water symbol gives us the word “je,” possibly an affirmative such as the “jo” and “ja” or “io” of many northern languages. In Baltic languages, “jel” is an affirmative or more akin to the start of an imperative to do something, so many of the words in that graphic are probably closer to imperative verbs to make or perform something. Finally, the three words you pick out to signify fertilization are all the same, or at least that’s what I strongly believe. I just don’t think researchers are allowing whoever wrote the Voynich to get sloppy as it’s supposed to be this pristine manuscript with no typos, miraculously. That’s bunk. They got tired. In all three instances, that’s a “p.” In the last one, they sneezed and tried to make it look pretty. The word here may be related to the Finnic word “uupua,” the present active infinitive of “to miss,” perhaps connoting some feature that is lacking. This is all for what it’s worth.

  62. Vladimir Dulov


  63. Vladimir Dulov

    HELLO Mr
    Algoritm №2 & №3
    The first portion of text seems like draft specification. Now it is understood only word-combination.
    Plant don’t have any name.

  64. Mia Y Padrón

    Hi Marco P., some species of Dracaena produce the “dragon’s blood”. It had been used in alchemy. It had medical and magical uses too. There is a legend about the origin of this “blood” (The Eleventh Labor of Hercules: The Apples
    of The Hesperides). Besides, it has been used as a dye.

  65. Mia Y Padrón

    Thanks, MarcoP. The one at Sloane 4016 is for me a Arum dracunculus or Dracunculus vulgaris. The common names it receives are related to dragons or reptiles, like “dragonera” or hierba culebrera”The F25v is probably a young specimen of Dracaena draco, from Canary Islands. Both are considered to have medical or magical uses, specially Dracaena draco. The figures incarnate in a symbolic way the name of those plants, I think.

    • MarcoP

      Hello Mia, I agree about the Sloane 4016 f38r plant being Arum Dracunculus (another one including the flower). I also agree about the function of the animal in Voynich f25v. We have not been able to find a name explaining the presence of the animal yet.

  66. Mia Y Padrón

    On f25v, it remembers me the Yggdrasil legend, but the plant looks more a gen. dracaena, maybe is Cordyline rubra.

    • MarcoP

      Hello Mia, if the strange beast on f25v was meant to be dragon, Dracaena (from the Ancient Greek δράκαινα – drakaina, “female dragon”), would certainly be an interesting candidate. Here is an illustration of another dragon-related plant from manuscript Sloane 4016 (f38r).

  67. Thomas

    I like your approach. I played with the thought, that on similar lines other words could be identified tentatively. So I thought, what words could be relevant in a discourse of plant life, or generally, life? Leaf, stem, root, water, fluid, blood, life… Then, I picked some frequent Voynich words, among them “dam”, supposing it may have to do with life. Then Googled Arabic and I too almost fell off the chair! The same damn chair! “Dm”, “damma” means both blood and life. Just for food for thought from me, a professional amateur.

  68. Bertin

    Dear Mister Bax,
    I’ve seen a part of what you write about Taurus constellation.
    It’s pretty clever but how do you explain the word “doar” in that case?
    It really looks like “doary” you see, the word you translate on “Taurus”.
    Are you 100% sure it’s Taurus?

    • Stephen Bax

      First of all, I’m not 100% sure of anything relating to the Voynich manuscript!

      As for your question, think of the words ‘cat’ and ‘caterpillar’. The first is part of the second, but this is coincidence.

      So, by parallel, we need to examine the word EVA:doar in its own right. We need to look at its Context in particular. Where did you see it exactly? It is no use talking about it in isolation.

      • Bertin

        Thank you for your message. The word EVA:doar is in folio f68v3.
        It’s pretty stange that the word EVA : doary is not on
        it’s found as EVA : doaro. But I agree with you, it looks like a EVA:y, not an EVA:o. What surprise me it’s if I listen to the arabic word “TAURUS” on google translation, it sounds like a “s”: “saura” or “sora”, not like a “t”. It’s a little bit far away from TAURUS or TAURN don’t you think?

        • Bertin

          I thought I add something but as I see it didn’t work out.
          It’s a word with 5 letters of the alphabet and there is just one missing. It’s really different from “cat” and catterpillar” but they have a point in commun, “cat” and “caterpillar” are animal (except the trademark). When you add just one letter at the end, it looks like, for example, plural or verb. But you cannot have plural or verb for the constellation of the Taurus, right?
          About the word EVA:doar, it’s on the same kind of pages as the word EVA : doary/doaro, so I can think that it’s about the same subject and it’s on the center of the stars, so it’s certainly about stars, it would be a real strange coincidence to have to similar words with only the letter of the end that changes without common meaning, don’t you think?

          • Stephen Bax

            Thanks – if you look at my paper and video you will see that my suggestion is that the root word could be ‘Taur’ with the final letter as an additional ending which I read as an /n/ sound, from reading other words. I suggested that this ending could be a grammatical marker of some kind.

            So it is perfectly possible for the same word to occur elsewhere, in other positions in the text, without this ending. In other words, I don’t see any problem with this element occurring on its own elsewhere in the manuscript.

            • Valérie-Anne Bertin

              Thank you for you answer. If it’s a particul of grammar and if we don’t know what kind of prarticul it is, then, i agree with you.

  69. Pedro Váldez

    I stumbled upon a word which seems to be very frequent in the second part of the Manuscript, but which is almost non-existent in the first part of the Voynich Manuscript. I find this pretty weird and it might give us an indication on what kind of word it might be.

    It’s the word ”Otedy” (”oMc89”). If I am correct, Arabic doesn’t have any ‘y’ or ‘i’ at the end of a word. So I went searching for a word named ”oted”. I did find a reference to a language called ‘Sora’ or ‘Savara’ which originates from Eastern India. It must be noted that it is an austroasiatic language. A quick Google search made me find this Sora dictionary:

    Oted and variants of Oted mean: ”To stumble”; ”in a reeling manner, stumbling on account of intoxication”; ”to toddle, to reel”; ”tipsy”.

    I am not sure whether any of the translations of the Sora language might be useful, but I am sure it is useful to note that ‘etody’ is mentioned 155 times and mostly in the second part:

    First I thought it could mean something like ”virgin” or ”women”, because otedy starts to be mentioned when the ”Tub Session” begins. However, that’s a simple guess. ”Tipsy” on the other hand, could match.

    I am not an expert though, I just wanted to contribute.

    Saludos cordiales 🙂

    • Stephen Bax

      When you say the word is ‘otedy’ are you referring to the standard Eva transcription or some other transcription? Can you tell us where the word occurs exactly?

      Also, Arabic has plenty of words ending in an /i/ sound or close to it. For example ‘bayti’ means ‘my house’.

  70. JLS

    Please review PDF attached here:

    These are just some ideas and guesses derived from the manuscript itself and from viewing the text as a collection of sigils, which I compared by using Getting’s Dictionary of the Occult, Hermetic and Alchemical Sigils. You can download the dictionary here for your reference:
    All interpretations and guesses are by no means conclusive. These are just ideas or different angles of observation based on sigils, logic and observation.

  71. D.N. O'Donovan

    I have a question. In what language or languages is the term for the [north] Pole identical to that used for the finial of a tent?

    Hope this isn’t unanswerable.

    • Stephen Bax

      I’m sorry to say I don’t know of one. Did you have any possibilities in mind?

      • D.N. O'Donovan

        I thought it might be a bit of a poser. No, it’s just that a detail on f.86v, in the ‘canopy’ for the quarter North-East, contains something that looks like the finial of a tent/canopy. It might very well represent Mt. Meru, which I noticed as allusion in one of the botanical section’s mnemonics, but that post was written yonks ago.. 2011 or so I think.

        More recently, someone – maybe Marco – included a picture which represented Mt. Meru, rather unusually, as a Pole. Very, very interesting find, I thought. But to envision the ‘Pole’ of the world like the carved finial of a tent would mean use of similar terms in a person’s native language: imagery’s form follows figures in speech, as no doubt you know. That’s how I identified the original makers of the month roundels as conversant with Greek.

        Sorry, I’m off topic now.

    • Linda Snider

      Perhaps it would help to refer to the finials as pole tops? Seems to be a correlation in that sense.

  72. Vladimir Dulov

    to continue my position, here is an addition:

  73. Vladimir Dulov

    Hello , Mr. Bax
    I don‘t speak English, let me give arguments in Russian.
    Thank you!
    for further comments, please visit my VK-page:

  74. Vladimir Dulov

    Hello , Mr. Bax
    I don‘t speak English, let me give arguments in Russian.
    Thank you!

  75. Vladimir Dulov

    Hello , Mr. Bax
    I don‘t speak English, let me give arguments in Russian.

  76. Its not surprising it references Taurus & the Pleyades,since there is the illustration of the birth goddesses who were meant to have brought the first humans to term.(This knowledge was known to Jesuits).
    Check out the research of Wayne Herschel,for example..ufology etc

    • Dear Dr Bax,
      While I was watching your 2014 video, it crossed my mind that the symbol that looks like the number 9 could be a nasalized /a/ [Voynich o], like the Polish diacritical reverse comma attached to the /a/ for the same purpose. You labelled it as an /n/, leaving the vowel sound to be guessed at. Similarly, the reverse C [in Voynich] could be the same letter, applied in a hurry, and some of the other tails might be nasalizations too.

      David Walmsley, teacher emeritus

      My qualifications are one course in Linguistics, 20 years as a teacher of ESL and, of course, curiosity.
      PS It fascinated me that you are able to read Akkadian. I would love to be able to read Zecharia Sitchin in the original.

  77. josh gura
  78. Nijkolaj

    Good day!
    My name is Nikolai.
    To a question about the key to the Voynich manuscript.
    Today, I have to add on this matter following.
    The manuscript was written no letters, and signs for the letters of the alphabet of one of the ancient languages. Moreover, in the text there are 2 more levels of encryption to virtually eliminate the possibility of computer-assisted translation, even after replacing the signs letters.
    I pick up the key by which the first section I was able to read the following words: hemp, hemp clothing; food, food (sheet of 20 numbering on the Internet); cleaned (intestines), knowledge may wish to drink a sugary drink (nectar), maturation (maturity), to consider, to think (sheet 107); drink; six; flourishing; growing; rich; peas; sweet drink nectar and others. It is only a short word, mark 2-3. To translate words consisting of more than 2.3 characters is necessary to know this ancient language.
    If you are interested, I am ready to send more detailed information, including scans of pages indicating the translated words.
    Sincerely, Nicholas.
    [email protected]

  79. Derek Vogt

    The Voynich Manuscript popped up on TV a couple of days ago, in the American show “Elementary”. Watson walked in on Holmes sitting at a big table almost covered with copies of Voynich pages and books lying open, except for a turtle in the middle of the table eating some lettuce…

    WATSON: What’s going on in here?
    HOLMES: I had some time so I thought I’d revisit a mystery that has always vexed me: the true meaning of the Voynich Manuscript.
    WATSON: What language is this?
    HOLMES: Unknown, as are the species of plants that the author drew so meticulously at the dawn of the fifteenth century. Some people believe that the book is extraterrestrial in its origin. I am not one of them.
    WATSON: And Clyde is helping?
    HOLMES: Not really.

    Then they changed to a subject that was actually related to the plot.

    Just imagine, someday those little occasional references will refer to the Voynich Manuscript not as a mystery but as a former mystery, including a brief summary of the solution. 🙂

    • Stephen Bax

      Pity – I thought that my own turtle had nearly solved it.

      However, I myself wonder whether, when this wonderful mystery is ‘solved’, the world will lose interest?

      No sign of that yet – people are still clearly fascinated by it. Even this little website continues to get huge attention, thanks to so many of your contributions. This last month has seen 16,023 page visits, an average of 500 per day. Interestingly, that includes 6,670 new visitors this month alone. Since it began a year ago, more than 130,000 people have visited the site (not counting repeat visitors).

      One IP address is apparently located at 221B Baker Street, London 🙂

      • Darren Worley

        Stephen – the VM could remain a mystery, simply if people ignore the results of any solution. I noticed that in the short movie that you posted a link to, a little while ago – no mention is made of your tentative proposed deciphering, despite the wide media attention this received. This clip was a commentary by Bill Sherman, a Renaissance scholar and curator of the recent exhibition at the Folger Library. He [Bill] seemed to prefer that the VM remained a mystery.

        In fact, outside of this forum, there seems to be very little acknowledgement or references to your paper, or the work that’s been presented here.

  80. Professor Bax,
    The person to whom this belonged was an ancient gyn aka a manager of women’s gestation or a manual for. I found note of menstral cycle (7 days) and probable lists of patient record with birth being signified by a crown on females head. It was obvious that dissections were done on deceased female to get internal information as shown.
    It was not a heart, it was an ovary. This is very advanced knowledge for the time. To
    find you language, look toward a advanced knowledge. Phenicians were the first I thought of. Sadducees is also very possible. I would look to what the inks are composed of. Sir, you maybe dealing with a dead language. Comparing names of the days of the weeks of that time whth hebrew, arabic and arimaic may be your best bet.
    It may well be that region from which it came. And, a very ancient system of gods and
    religion. It may even be a teaching manual of the times. Let me know what you think of all the things I have written of. I am very interested of this since it survived all of
    the burning and loss of such materials of this nature. Please forgive errors in any spelling.

    Thank you for your time and consideration,

    Taylor Morgan

    • Darren Worley

      Hi Taylor – I’d like to know more about your ideas about menstrual cycles. Where in the VM do you think this is described? Do you mean a 27 or 28 day cycle?

      I think this idea is not as strange as it sounds, in Judaism and amongst the Mandaeans, this is closely related to ideas of ritual purity.

      Quote: Of those connected with full ritual immersion; perhaps the quintessential immersion rituals still carried out are those related to nidda, according to which a menstruating woman must avoid contact with her husband, especially avoiding sexual contact, and may only resume contact after she has first immersed herself fully in a mikvah of living water seven days after her menstruation has ceased.

      There are also restrictions on food preparation and access to the temple or areas for worship, for example.

      Ideas about “living water” and “ritual purification” have been mentioned before in connection with the VM.

    • According to a history of obstetrics, “prior to the 18th century, caring for pregnant women in Europe was confined exclusively to women, and rigorously excluded men. The expectant mother would invite close female friends and family members to her home to keep her company. Skilled midwives managed all aspects of the labour and delivery. The presence of physicians and surgeons was very rare and only occurred once a serious complication had taken place and the midwife had exhausted all measures to manage the complication. Calling a surgeon was very much a last resort and having men deliver women in this era whatsoever was seen as offending female modesty.” An informative paper on this can be found in Volume 5 of Health Science Journal: Midwives in early modern Europe (1400-1800) by Maria Kontoyannis and Christos Katsetos.

      Europe has a long tradition of bayna/sauna/spa, in some places even giving birth in the sauna. In the 1500s Klaus Magnus wrote: “Nowhere on earth is the use of the bath so necessary, as it is in the Northern lands.”

      • Darren Worley

        The possibility that the VM may have some application to obstetrics, is something that I have also considered. I would broaden this possibility to cover fertility too. I think it is quite plausible that it may serve some purpose in this area, however, I came to this idea from a different starting point.

        In my case is was from learning more about the Mandaean faith and culture. The Mandaeans celebrate life and fertility and hold Christian belief with distain accusing them of celebrating death (the crucifixion). The contents of the VM seems to fit well with the Mandaean worldview, as I have described elsewhere.

        The “herbal section” could well serve a medicinal purpose, or contain texts for use in amulets. The (suggested) appearance of bodily organs in the “bathing section” could also be explained within this fertility/obstetrics “framework”. The astrological section may also serve some purpose in this area.

        I came across this image below “Trotula Major” that seems to have parallels with other images in the VM zodiacal section.

        Its taken from the “Trotula Major” manuscript – this is credited to Trota of Salerno a female medical practitioner from Salerno, Southern Italy in the 12th century.

        Salerno was widely reputed as “the most important center for the introduction of Arabic medicine into Western Europe” at this period in time.

        Greek was widely spoken in the region of Southern Italy, and nearby Sicily was various governed by the Byzantines, Arabs, Normans and Aragonese during the medieval period – giving a clue to the mix of cultures and
        languages in this area.

        Apparently, these trotula texts were copied and circulated widely in medieval Europe. They also circulated in vernacular languages (rather than just Latin) – for example the first known translation was into Hebrew, made somewhere in southern France in the late 12th century. The next translations, in the 13th century, were into Anglo-Norman and Old French. And in the 14th and 15th centuries, there are translations in Dutch, Middle English, French (again), German, Irish, and Italian.

        Although there are parallels between these images there are also some differences: in the VM the female figures are depicted both clothed and naked, there is also wide variation in the decoration of the birthing tub (?). They are various decorated with vertical, diagonal or horizonal stripes, dots, circles etc.

  81. Nick S

    Not directly related, but a reminder of just how wide the world was, and just how isolated small ethnic groups could be, till not that long ago, and centuries after the VM was created–and in the rough target area of most VM theories, too!

    Things like this are, to me, the best argument for there being a perfectly rational reason for the VM’s creation somewhere in southern Eurasia in the middle ages, and obviating the need to insist that it must have come from the New World or China, etc. Not saying that those things are impossible, but I think a lot of people assume that Western history doesn’t have enough cracks to fall through. I’d beg to differ.

    • When reading about Zana, one might easily be reminded of the treatment of Sarah “Saartjie” Baartman detailed by SJ Gould in his discussion on “scientific racism” in the essay “The Hottentot Venus.” It gives an eye into how people from small, isolated ethnic groups were viewed in Europe during that time. Delving further into Zana’s biography sheds light on what she endured from her captors, who mention her revolt against them while being far less eager to describe precisely what they were doing to her. She was kept on a tether. The village men would get drunk and play a “game” with her that often ended up with one or more of them raping her. The quote calls this violence “sexual relations.” She dunked one of the babies from these rapes into ice cold water, and was from that deed judged to be a brute when she simply may not have wanted the baby to live the life she was living. It indeed appears to be quite likely that she was from a remnant population of runaway Ottoman slaves and that she was merely made out to be some monstrous throwback by people who were themselves monstrous throwbacks, brutal and ignorant. All of this does, I suppose, make the slimmest of cases for the Voynich to be from some exotic isolated population. It’s a stretch but whatever. I personally rather strongly doubt it was isolated at all but rather systematically wiped out so successfully as to leave us baffled by this remnant. The headdresses worn by the women look like headdresses that have been worn for centuries in various European countries. And I have to disagree about any likelihood that it’s from southern Eurasia. Southern Eurasia consists mostly of the steppes–grassland and barrens, not trees. The Voynich is full of depictions of treenware that look most similar to handicrafts from countries like Poland, Norway, Russia, and (during the 1400s at least) the UK, where vast woodlands stretched uninterrupted from coast to coast. Such a ubiquity throughout the manuscript’s pages of treenware would suggest an ease of accessibility not to the steppes but the forest.

      • Johannes Klein

        Dear Claudette,
        I would urge for caution here. The first assumption in your argument is that what is illustrated are some kind of jars. This needs to be clearly outlined and justified otherwise it remains speculative. If this assumption leads you to a prediction that you can test, than it is a valid hypothesis. The second assumption is that these jars are made of wood. Which also needs to be tested.
        From the overall impression of the plant folios, I would argue that very few, almost none indeed, are depictions of trees. This may suggest that trees did not play a prominent role in the landscape known to the artist. But this of course is also speculative.

        • Johannes: I would advise greatest caution in espousing the idea that these are microscopes or telescopes or any other such implement, as I suspect may be the wish at the heart of your objections. A quick look at the state of 15th century scientific technology would annihilate the prospect of these being anything other than containers, but let’s just go with some logic here: why on earth would anyone create a different viewing mechanism for each series of plant parts? Look on the pages. I’m amazed I’m having to point this out, but these things are drawn beside the roots and leaves of herbs, most of them extremely distinct from each other with various patterns, and quite a few of them, in addition, bearing lettering. That is, for those who had literacy and those who didn’t, or perhaps those who used a different language, these containers tend to be clearly differentiated either by labeling or patterning or both. I said they are made of wood from trees. I did not say they contained trees or the parts of trees, although on second look some of the plants depicted could in fact be shrunken pictures of whole trees. Whether or not that is so, nixing the idea that these are treenware merely because they may not CONTAIN trees is, well, goofy. These containers bear an extraordinary resemblance to the treenware that has been made in Europe for the past several centuries. And no wonder. Before the late-17th century, when pewter, silver and ceramic began to filter into use for the gentry, wood was the default material for all. From peasant to king, everyone would have used turned wooden platters and bowls and spoons, wooden jars and sewing necessaires, wooden urns and keepsake boxes… Finally, this is where I’m posting snippets of my research, not the bulk, which resides on my blog, and even that doesn’t contain everything. I would urge you to go look at the over 30 postings that represent the tip of the iceberg of my research. After reading over all of that, you’re certainly welcome to stick with people calling “oxhccS” a word and making up star names and extend to them your words of caution.

          [Claudette later added: låsan – From Old Norse láss, from Proto-Germanic *lamsaz. to lock]

          • Stephen Bax

            Claudette, I love your many explanations of the northern option, but I have real problems with your linguistics.

            It is not a sound approach to take elements from different languages to explain Voynich words (as you did in this posting, with Latin ‘hepar’ and Norse ‘lass’) without a lot of evidence that such a word really existed in some form in a real language. Otherwise you are in danger of getting ‘false positives’, results which look positive but which are in fact without foundation.

            If you can find evidence of a real word resembling ‘eparlasai’ in some real language, perhaps as a portmanteau, then your interpretation might be more believable. If not then I am afraid that your interpretation looks like guesswork.

        • Linda Snider

          I think there are some trees or at least shrubs in the VMS plant list. But at the same time the trees would not have to be shown in the plant list for them to have been the raw material for the containers/mills/x.

  82. Ben Salmon

    I wanted to submit this as a post, but I can’t work out how to register here (probably missing something very obvious), anyway:

    I was thinking about the possible link with Arabic medicine, and went into the humors, and remembered that Avicenna revived Galenism (humorism), and the prominince of his work for many centuries. Looking at f67r2, and using some bad photoshop skills, I came up with this. Perhaps someone can do a better job with it and look closer!

  83. I thought you might be interested in this tidbit from the Voynich Mailing List on 4 September 2000, posted by Gabriel Landini:

    I still think that we need to attack the labels.

    My candidate word is in folio 2r for the plant name.
    It appears twice in that page, as in the first line, and as
    in the 2nd paragraph. It does not appear anywhere else in
    the ms. I think that [Jorge] Stolfi did some logical filtering for finding which
    words occur only in a single folio and nowhere else.

    So let’s assume that the “proper” Latin name of the plant is kydainy
    with let’s say standing for “us” (given that it is already used for
    marking the quire number) and the second is the “normal
    name” in whatever tongue the author uses.

    • Stephen Bax

      Thank you – intriguing! As you no doubt know, that links exactly with my own analysis of that page. The only difference is my analysis of the last sound of the first word, which I suggested might be an /n/ sound, not ‘us’.

      However, I myself am dubious about identifying words which ‘do not occur anywhere else in the manuscript’ as I suspect there are many suffixes, affixes and even infixes which obscure the occurrence of the ‘same’ word on various pages.

      To put it another way, that word EVA:kydainy might also occur with an infix – say ‘kyXXdainy’on another page – which we would not be able to recognise.

  84. MarcoP

    An amazing post by Ellie Velinska about a 2013 auction in Rome: a small book by Athanasius Kircher about the Voynich manuscript was put to sale. It is the only known copy of this work.
    Since apparently the book was bought by a private collector, I am afraid we will never have a chance to read it. Kircher was unable to “bring to light the unviolated secrets” of the manuscript, but I am sure that what he wrote is extremely interesting.

    • Quite amazing. It would be unique, if it were genuine – the only time on record that I know of when Kircher admitted he didn’t know, or couldn’t understand, anything at all! Psychologically, typographically and in every way unlikely.

      I understand, too, that Rene Zandbergen presented the information about typographical inconsistency. I expect the information came from one of the librarians who work in antiquarian and rare books. But I may be in error on that point.

    • MarcoP

      A few months ago, the same auction house Minerva Auctions was about to sell for 150,000 Eur (!) a false manuscript attributed to the Italian poet Giacomo Leopardi:

  85. Stephen,
    I’m thinking solely of the time, and context suggested by items of the imagery – but has anyone ever investigated Chagatai (Chaghatay) as a possibility, do you know?

  86. Melissa R. Schwartzkopf

    I am a simple woman with a few questions. I do not know a lot about the manuscript but as I was watching the documentary, I had a few thoughts. What if all the large letters were points of references instead of an actual letter? (the one that looks like a large H). Could the manuscript be 3 dimensional instead of just two? Instead of looking at the script horizontally, could you look at it vertically using the points of reference that I mentioned in the above? What if you used two mirrors in an “L” shape and used the points of references that I mentioned? I am just throwing some ideas out there. I have no idea if these have already been presented but I would be curious to see if any scholars could look into them. I am very intrigued with this.

  87. K. Sahin

    Dear Mr. Bax,
    I just watched your presentation “Voynich – a provisional, partial decoding of the Voynich script”. I decided to share with you some impressions I have about the Voynich Manuscript.
    I was born in Trabzon (Trebizond) in Northeastern Turkey. As a hobby, I search for old photos and drawings of my hometown and so I often come across with Greek, Georgian and Armenian alphabets. I already had seen the manuscript two years ago and the first thing I intuitively felt when I saw the writings of the Voynich Manuscript was that it had a connection to our region. I don’t have a lot of historical and linguistical knowledge but it appeared to me like a mixture of the alphabets you would find in that time in our region approx. as large as Northern Italia: 1) Greek alphabet used in the Empire of Trebizond 2) Arabic alphabet used by the neighboring Turks (also Persian influence in Caucasus) 3) Georgian alphabet used in the East of the Empire of Trebizond 4) Armenian alphabet used by all Armenians in the whole region. In the end of your presentation you just wrote exactly what I had intuitively felt: That it may be a writing system which took signs from a variety of alphabets.
    And there are three more thoughts I would like to share with you: In history the city of Trebizond had a strong connection to Northern Italy. Venice and Genoa had trade colonies and castles in the city. Until 1950’s there were living Italian Catholics in the city and there is still a Catholic church. In Italian you can find the expression “perdere la Trebisonda” and the first Italian woman to win a gold medal in the Olympic Games was named after the city (Trebisonda Valla). Basilios Bessarion was born in Trebizond and Giorgio da Trebisonda’s family was originated from Trebizond… There are much more examples to a strong connection between Northern Italy and Trebizond.
    Furthermore I would like to call attention to the language diversity in our region as for example the languages Lazuri Nena and Hemshin. It seems to me not unlikely that someone with a previously unwritten mother tongue -as you describe it and I also had thought about- could have come to Northern Italy via the Port of Trebizond. I also would like to pay attention to the plant diversity in our region. Probably you could find all the plants of the manuscript in this region.
    One more thought: Abjad is also used in Turkish. As far as I know the old Turkish writings where written without writing most of the vowels. Unlike in Arabic that’s possible due to the vowel harmony in Turkish…
    I hope these thoughts and impressions may be inspiring for you.
    Best regards

    • orun rubacı

      hey sahin im from bursa see you on facebook


      Here I found, that Khodorjur word ‘gənigən’ means ‘women’ (pages 16, 17). And then, look, please at the first word on page f77r near women’s figure! This is – 8a2cc8a2!

      [SB: Do you mean this word EVA: darchdar ??]

      • Stephen Bax

        Why would this word on that page mean ‘women’?

        • Yes, I meant ‘darchdar’. Because this word is located near women’s figure. And earlier, somewhere in the web I had met opinion that EVA: ‘dar’ is related to the female.

        • And the second word near women at this page is EVA: ‘olkchs’, oxhccS, as I had written earlier, means ‘steam’, ‘dense vapor’, ‘fog’ and can be read as [aeshkhi(v)?] (Georgian “oshch’ivari”).
          So, we have seen armenian word root, georgian word root and persian (indoarian) word root (the word EVA: ‘otol’, oHox, that seems to be ‘atash’ or ‘aetaesh’, written close to fire image, blowing out the tube.) Which language can consist of similar mix? I have found only one. This is Lomavren.
          A realy endangered language. Native speakers: 50 in Armenia (2004)

          • Hello! Seems, I’ve obtained the right pronounce of those 5 elements from f77r.
            1. EVA: ‘olkchs’ is [oshqhep]
            2. EVA: ‘otedy’ is [orega] (here ‘r’ pronounced like portugal back ‘r’)
            3. EVA: ‘otol’ is [orosh] (here ‘r’ pronounced like portugal back ‘r’)
            4. EVA: ‘otork’ is [oronq] (here ‘r’ pronounced like portugal back ‘r’)
            5. EVA: ‘dchdy’ is [ghega]
            Language looks like a mix of albanian, gypsies’s, and probably armenian and georgian. Very strange language.

          • The last word of this row, one after the earth element – (EVA: ‘soral’) seems to be [manush] and means ‘a man’ in a gypsies’s language. So, EVA: ‘S’ seems to be ‘M’. And the first word in this row (EVA: ‘olkchs’) might be ‘oshqhem’ or close to this…

            If this is really Gypsies’s language, there will be a great breakthrough! A forgotten script of gypsies!

            • Sorry for my bad english, I’m from Russia…

            • So, finally:
              EVA: ‘o’ seems to be [ʌ]
              EVA: ‘a’ seems to be [ə]
              EVA: ‘c’ seems to be [e]
              EVA: ‘l’ seems to be [ʃ]
              EVA: ‘k’ seems to be [q] or [k]
              EVA: ‘t’ seems to be [t] or [r]
              EVA: ‘d’ seems to be [g]
              EVA: ‘y’ seems to be [a]
              EVA: ‘s’ seems to be [m]
              EVA: ‘r’ seems to be [n]
              EVA: ‘ch’ seems to be ‘he’ = [ee] (like in word ‘seen’)

              • I’ve just tried to renew the possible transcription of those five elements (air, water, ether, fire, earth). It seems to be more correct for me, than i’ve posted at January 29, 2015 – 10:00 am.
                1. ae-sh-k-e-b = [əʃkeb] (EVA: olkchs)
                2. ae-t-ee-d-a = [əti:da] (EVA: otedy) ( =greek AtlanTIDA, antarkTIDA)
                3. ae-t-ae-r’-k = [ətər’k] (EVA: otork)
                4. ae-t-ae-sh = [ətəʃ] (EVA: otol)
                5. d-e-d-a = [dəda] (EVA: dchdy)
                but then, words for a women and a men isn’t conform that i’ve found earlier in Hodorjur dialect and Gypsy language. As a result – we have d-(u?)-r’-e-d-(u?)-r’ for a women (russian ‘dura’ for a stupid women), and b-ae-r’-(u?)-sh for a men. It isn’t just a gypsy language, it is more complex one.

                • Stephen Bax

                  The danger of bringing in many languages is that it might seem that we are getting ‘false positive’ results.

                  But if you want to put your theory together into a full posting, explaining the sounds you have identified, I am happy to post it?

                  • Yes, I want. Thanks! I will develop something else new theory.

  88. Julie

    Professor Bax
    I have a hypothesis but don’t know who to approach with it. I have translated folio f1 into something that can be read and think I have uncovered the underlying language and maybe explain the minim problem. As I am not a linguist I am finding it difficult dealing with the verb tenses. I have no idea what to do next. I may be wrong here but think that someone ought to have a look, it does however go in a different direction than your work. I would be grateful for any advice and help you could offer, thank you

    • Stephen Bax

      I’m always interested to read interesting new ideas, so feel free to post the translation here, if you want. I’ll also email you so you can send it to me personally if you prefer.

      • Paul Hicks

        Me too!

        Rather than clog up your blog could I send you a document and you can then decide how to move forward?

        I have some process ideas as well.

  89. Oliver Helweg

    Hello. The starting translation of page f2r is “Ha duna at ero L dun oh eru L at erum ehrolsa (aerosa)”

    • Alan Hughes

      Can you please explain this?!

      • Oliver Helweg

        I don´t use the EVA, I have my own ABC.
        And my ABC is working 🙂
        (The “L” is a latin number in this page)

      • Alan, sounds like Strine – a bit.
        (This was supposed to appear a few comments higher)

  90. Mike

    Well done Dr. Bax. I See that your work was chosen by the Daily Telegraph newspaper as one of the most important events in 2014! See this page:

    well done for your interesting research. I know you have not deciphered the manuscript, as they suggest, but still your way of studying it is the best we have!


  91. Nick S

    Another language possibility: that of the Lipka Tatars, part of the great wave of Mongol invasions who eventually made their way to Poland/Lithuania in the 14th century. As part of that wave, they or their recent ancestors could certainly have passed through areas in Asia where the identified Voynich plants grow. And they assimilated fairly quickly, and adopted the Belarusian language in place of their own. Furthermore, they continued using Arabic script up through the 1930s, so a ‘one-off’ transcription attempt into the Roman alphabet by a lone clergyman or diplomat would make more sense than in the case of a language with an already agreed-upon Romanization scheme.

    • Diane

      Nick S
      This is exciting – you speak of “areas in Asia where the identified Voynich plants grow”. Have more people joined this ‘camp’? I only know of Wiart and Mazar, apart from myself, who regards the plants as other than those in western herbals. And Wiart and Mazar differ from me in that they suppose the plants chiefly used in (Ayurvedic) medicine.

      Please provide a link to any other person(s) who identify Asian plants in the Voynich ms. I’d love to compare notes.

  92. Rafaela Rautanen

    In Israel were tree different groups; Pharisees, Sadducee and the Essene: The Essene used herbal treatments and some of this is found in the Dead Sea Scrolls. After Jesus died they traveled far to spread the Gospel, but stayed mostly away from the public eye. They thought their herbal medicines from one generation to the other and obviously the language varied according to their new location and they also passed this knowledge to new converts. Now this explains the similarities to languages such as Aramaic, Hebrew, and Arabic very easily. These people were not necessarily that educated in any particular language of an area, they were early Christians. The manuscript is an attempt to store this knowledge that they had and to keep it. The language of a plant was then carried by word of mouth to the next generation or they might have used a local language at a time. You can check the Essene are online today in Canada and I think France.
    Have a wonderful day,

  93. Marnix Hoekstra

    Professor Bax,
    Congratulations on your achievements. I am confident that your slow and steady approach to this mystery is the right one.
    I have nothing to add to the recent discussions here. I just wanted to share two basic ideas:
    – Firstly, why has the Voynich vellum never been DNA-tested? Even the dead sea scrolls have been preserved well enough to allow DNA-tests. A simple test will reveal what breed of cow provided the vellum. It could be more revealing if the vellum is not calf skin at all.
    – Secondly, I’ve been toying with the idea of an intuitive approach to understanding the VM, inspired by Chomsky’s universal grammar.
    If Voynichese is a natural language, then anyone can learn it as a first language, without a dictionary or grammar book.
    The idea is: Use a transcription system that makes it pronounceable, read one page every day, look at the drawings, and use our natural ability to acquire languages. There is plenty of text with clear illustrations.
    What do you think?

    • Stephen Bax

      Thanks for your confidence. I like your idea of learning Voynichese, though it might drive you crazy!

      Research into language learning shows that learning natural languages usually needs spoken input, and a lot of it over many years. Reading and writing comes later. So I feel the reading approach would not be enough, sadly.

      However I do agree with the suggestion that if it is a natural language then we should be able to analyse its regularities, with some results!

  94. Paul Hicks

    I am unable to find out if you have already worked some things out. Perhaps you need a summary page and discussion for each page of the text? Is this the right place to ask some process questions?

    f2v the water lily. The first word also looks like KAAUR which is what you have for Hellebore in f3v.
    The first word of the second paragraph looks like KSHAR so could be KASHAR which is Turkish for KASHMIR?

    • Stephen Bax

      Thanks Paul. It’s difficult to have a summary page because suggestions keep coming in, and in many cases we are still in the dark. In some cases we do have a specific page e.g this one for f2v, which discusses the first word of that page.

      But I will start to try and give summaries month by month.

  95. giorgetto

    voynich bene io sono in grado di leggere la scrittura,posso per il momento darvi una dimostrazione,partirò dalla mappa la mappa in realtà sono due se guardate bene si noterà che la parte destra è stata applicata al rovescio anche la scrittura è al rovescio,il sole si trova sotto nella parte destra, mentre nella parte sinistra si trova a sinistra in alto.cordiali saluti. giorgetto.

    Google translate gives this: Voynich well I am able to read the writing, I can give you a demonstration for the time being, I’ll start with the map: in the map are actually two if you look carefully you will notice that the right side was applied to the reverse side also the writing is on the reverse, the sun is located under the right side, while the left side is on the left. greetings. giorgetto.

  96. Nick S.

    I’m not a linguist and have no comment on the linguistics other than to say that your theory seems plausible.

    I’m curious about what a careful historical analysis of the period might bring up. I would caution against assuming that the spoken language of the VM necessarily died out–it is just as likely that the political or religious winds blew in directions that shut down any further development of the alphabet. The comments made by other posters about the Persian outposts of the Mongol Empire, and the Genoese colonies on the Black Sea, intrigue me.

    I would look for situations in the correct time period that would involve religious, political, or commercial expansion into an area with an unlettered population–and then a fairly quick retreat from it.

    • Stephen Bax

      Yes, I feel that may be a strong possibility. I do hope the language is still a well-known one, or else we could be searching for years!

  97. H C

    Prof Bax,

    I’m a former student of linguistics and math with an emphasis on probability and codes so naturally the VM is very appealing to me. I’ve been doing a lot of research and I think your approach makes a lot of sense if in fact it is a plaintext of a natural language (I’m not entirely convinced of that yet, I need to do more research).

    I guess a couple things strike me as odd about VM especially when I look at the statistical parts of it, which I know you’re not looking at in this analysis. It apparently looks a lot like a natural language but when you actually look at it there are things that are distinctly odd about it. The two main things I’ve seen noted over and over again are 1) the repetition or almost-repetition of words and 2) that there are apparently no errors/corrections in the manuscript. To my knowledge there aren’t really any language families that the VM is hypothesized to be a part of that have that kind of semantic repetition (correct me if I’m wrong about that). I’m wondering, if in fact the text is in fact plaintext and not encrypted or gibberish, these long chains are not the “errors” that are apparently not there. As in, whoever was writing them did not take the time to correct his mistakes and scrape off the ink in the usual way but instead just kept writing.

    It would be interesting to do a statistical analysis on this (as in, if you come across a chain of 3 similarly spelled words, you throw out the first two in your analysis). However as you’re not doing that, I guess it may be something to consider, that if there are two similar words far apart from each other they may be correct permutations of each other (as in with your ‘centaurea’ versus ‘centaur’) but with a string of similarly-spelled words, maybe it’s only the last one that counts.

    Anyway, good luck with your work!

    • Stephen Bax

      Thanks for your intriguing comments. Yes, the repetitions are odd, and I wouldn’t discount some of the possibilities you suggest. However, languages do sometimes use devices such as republication which look rather similar.

      Re. The lack of corrections, some suggest that this is evidence that it was copied. However, there are numerous contemporary manuscripts which also show no corrections…. some scribes were very careful! But I agree that these are interesting mysteries, set to nag at us for a while to come!

  98. Dear Stephen,

    Do you have a timeline for applying your initial findings on proper nouns to the rest of the manuscript? Are you awaiting verification from other researchers on your initial findings, or rather moving full speed ahead into translation of the rest of the manuscript?
    If your findings are accurate, how long do you think it would take to understand the full document?

    Best wishes,


    • Stephen Bax

      A few of us are working on it step-by-step. It is slow work because we want to be sure of each part – or as sure as we can be – before we move onto the next. That means someone puts up an idea – “it looks like this sound might be X” and then others try to check it to see if it might be true. But that is slow work.

      In this internet age everyone wants things to go quickly, but we should remember that when Linear B from Crete was deciphered, the most promising work started with Alice Kober in the 1930’s, and was brought to fruition as late as 20 years afterwards by Michael Ventris in 1951-2.

      Sorry to say, these things take a long time… and until recently many people dismissed the Voynich script as a fake. But I’m in for the long haul. I’d rather be the tortoise than the hare!

      • Derek Vogt

        It’s even possible to get all the sounds worked out and be able to roughly pronounce the whole thing out loud, but still have no idea what most of those words we’re saying mean or what language we’re speaking!

  99. Mary Tudor

    The Voynich manuscript was written in AD1421 by Catherine De Valois, the wife of Henry V of England. In 1421 Catherine was pregnant, hence the drawings of her pregnant.
    There are Bible codes about the Voynich manuscript, the first in Deuteronnomy 17:15 to Deuteronomy 33:29 “set a stranger over thee which is not thy brother”. “Strangers that are in thy land” – “which are not a people”. The second Bible code in Joshua 18:10 to Judges 4:24 and the third bible code is at Psalm 49:11 to Psalm 89:11

    • Stephen Bax

      What evidence is there in the manuscript itself to support this idea? What is the language?

      • Mary Tudor

        The language altered over time, into what is now called Gaelic. The original was from the time of Phoenius Pharsaid, who was king of Scythia circa 2493BC, he opened a school with his son Niul Namnach in the valley of Shinar near U’Othena.

        Tracing the family history of the Bowes-Lyon family, Queen Elizabeth Bowes-Lyon’s ancestry, you will find that they moved through Europe gradually, to Gotha, Germany -Heber Gluinfionn King of Gothia born 2216BC.
        Spain -Gallamh Milesius, King of Spain 1773BC, his sons Heber and Heremon were the first monarchs of Ireland, his oldest daughter was Eranna born in 1737BC.
        Sir John, The White LYON was born at Glamis in AD1340- Scotland.

        I think that the language was called something like Goidelic. (Berla Tobaide) and was a spoken language which may have evolved into Ogham.

        • Stephen Bax

          I’m sorry to disagree with you, but English could not evolve into Gaelic. They are from different language families, English is Germanic, Gaelic is Celtic. To support your theory you need to show much more evidence in the manuscript itself, not only suggestions from history!

          • Mary Tudor
            • Stephen Bax

              Yes, true, Germanic and Celtic languages (and Goidelic) are all from the large historic Indo-European language family, but one cannot simply evolve into the other over a matter of years.

              I love the idea behind your theory, but I feel it just needs a lot more evidence from the manuscript itself, or else it can be no more than speculation, I’m sorry to say.

          • Mary Tudor


            This is the reference for the bible codes. Maybe you will be the one to decode it. I hope so.

  100. AnoNymous

    I’m immensely impressed. My previous pet theory about this book was Zbigniew Banasik’s Manchu theory, but this method here strikes me as far more relevant.

    In fact, you’ve inspired me to do some Googling. My apologies, but I must share! …

    I decided to look for an identification of the initial word on the water lily page, 2v, which I think would be transliterated “koour” according to the phonetics that you’ve given, noting that the final ‘r’ could be an ‘ur’ or maybe a ‘ro’. I figured I’d take a wild guess and say that the plant depicted was some variety of white Old World water lily; so I navigated to the Wikipedia page (because Internet) for the genus “Nymphaea” (which contains a species called the European white water lily) and then hopped from there to the Azerbaijani version of the same page (because Azerbaijani is a Caucasian language written in the latin alphabet, relevant because the Caucasus was mentioned as a possible origin point for the VM, and because I was too lazy to sound out the Armenian. Also, since Azerbaijani is a Turkic language, I was casting my net just a bit wider that way.)

    The Azerbaijani Wikipedia page for the genus Nymphaea lists “Misir lotus gülü” as a name for the genus, which Google Translates (because Internet) as “Egyptian lotus flower,” and I noticed that gülü and koour are phonetically somewhat similar just like cilantro and coriander (e.g. between g and k is naught but a simple voicing difference, like between d and t). Taken alone, Google said that the Azerbaijani words gül and gülü mean “flowers” (or “flower of”) and “rose” respectively. Gülü also means rose in Turkish; so maybe a more complete translation of Misir lotus gülü is “Egyptian lotus rose,” which could make koour into the name of the plant, a name related to a hypothetical Turkic term for a waterlily “water rose” or “lotus rose.”

    Setting aside the rather unorthodox use of Azerbaijani Wikipedia and Google Translate as a means to get there, “flower of” would seem like a reasonable if pattern-breaking word to start off a paragraph possibly pertaining to Nymphaea alba, whose flowers are the most potent part containing the plant’s sedative and aphrodisiac/anaphrodisiac chemicals. (Wikipedia ascribes this thought to “Nielsen, Giftplanter, 68-69.”)

    So, what do you think of all this? How silly am I?

    • Stephen Bax

      Hello – and thanks for this considered response. I find your research interesting – the gülü and koour link does seem seductive and your suggestion that the first word of some pages might indicate ‘Flower of…’ is also interesting.

      Others have suggested that it might instead indicate a colour, e’g’ ‘Red..’ or ‘Black…’, ut your idea seems equally possible.

      So in my opinion it is certainly worth keeping your thought in mind as we dig on!

    • Derek Vogt

      For the two things that look like [o], the phonetic interpretation so far has been /a/, not /o/, which would make the first three letters “kaa”. But that’s a little technicality that doesn’t really harm your case, because vowels are notoriously fickle. In fact it leads me to something else which could make this connection you’ve suggested even tighter than you thought it was. I’ve seen a handful of cases where that letter seems correlated not just with “a” in other languages, but with “ar” or “al”, as if the consonant sound were implied or had originally been there but then got dropped. It doesn’t always work out that way, but if it did in one or both cases in this word, we’d expect counterparts of Voynichese “kaa” to have an extra sound in there that just spelling the Voynichese element as “kaa” doesn’t account for, as if the Voynichese spelling had been something else like “kaal”, “kala”, or “kalal”.

      Also, the idea that the first word here is either “flower of” or an adjective like “red/black” would imply that the second word should be the plant’s name, which fits perfectly with the fact that the only Nymphaea name I’ve found at that comes close to any word near the beginning of this Voynich page is Hindi “kokaa”, which best fits the second word:

      EVA: ch-e-o-p-ch-o-R
      Baxish: ?1-o-a-?2-?1-a-r
      Hindi: k-o-k-a-a

      So “?1”, EVA-CH, fits where /k/ is found in Hindi. We already have another Voynich letter for /k/, but this doesn’t need to be /k/ for the connection to work; it could be another sound that evolved from or into /k/ in one language or the other, or they could both have evolved from a third original sound. Other Voynich words using this letter which seem related to Hindi words are the fourth one on page 6r, where it looks related to /k/ again, and the first ones on page 24v and in the second paragraph of page 16v, where the Hindi words seem to have nothing corresponding to EVA-CH at all, as if the sound had simply disappeared from those Hindi words or a new one had been added in Voynichese. The fact that this letter has two different apparent relationships in Hindi words is typical of sound shift laws; they always apply only in certain phonetic contexts, depending on what other sounds are before or after them.

  101. dear stephan,
    my opinion about this book is – it bring back (a lot of people use this since a long time) us old knowledge about how the universere and all what is inside, works – we should use this for future – it is healing for all and everybody – and of course it was a group – before atlantis was gone different humans (priest….) took that knowledge and brought it in different forms in other parts of the world – and all cultures and language was and are different – so for me its the reason, at this book a mix of different signs ect. – for example: dr. bach used this knowledge and a lot of other people – it´s very interesting to hear and read about….
    thank you

  102. Sir, as a gardener with an interest in Sanskrit and the Kharoṣṭhī manuscripts, I was viewing this with some interest;whilst fig 15v is almost certainly an orchid, 16r bears a fascinating resemblance to Cannabis Indica, except where the stem joins the root, almost as if the artist has only ever seen a severed stem.
    Paul Causton.

    • Paul, I think 16r is agreed very widely to represent – or at least include – one or more of the Cannabis. So far 100 percent. The botanical section is tiresome, but if you can identify a few more, that section needs as many acute observers as possible.


  103. Chris

    Just to add. Voynich was certainly looking for Alchemical texts and in the VM he thought he’d found one that would reveal Alchemical secrets should he be able to decode it, a matter at the time he would have thought trivial.

    • Stephen Bax

      Thanks Chris. What exactly do you find that is alchemical? Can you see parallels with particular other manuscripts?

  104. Chris

    Whilst I am far from being an expert in Cryptography, and I have spent only a few short minutes looking at the Voynich Manuscript, my immediate reaction is that it is almost certainly Alchemical in its nature, a subject that I do know a little about.

    It is about plant Alchemy and the philosophy thereof. Please write to me if you would wish to discuss the matter further.

  105. What do you make of this YouTube video?

    This person claims to have deciphered the VM by positing a Romani origin deeply rooted in India.

    • Stephen Bax

      He has already contacted me and I discussed his ideas on this page. I think he needs a lot more evidence.

      His Romani idea might well have some foundation, but his way of taking letters from a range of different scripts seems too opportunistic to me. We must follow more rigorous linguistic methods, in my opinion, and also be more cautious!

    • i think that people who are able to recognize words of any language are very, very strong in parts where manifestly only graphisms and patterns are described… next time they could count trees along the road and pretend that’s the same amount as their salary… but who can really believe them ?

  106. Matt Owen

    Hello again! Been masticating the Romani hypothesis vs. others, such as Manidaen, which have popped up: I think the inclusion of vowels in the word-roots points to an Indo-European rather than Semitic language; at least some of the topics would have been of interest to Romanies; they had no other written language at the time; and the VM was written at about the time they came to Europe, fleeing the dissolution of the Byzantine Empire.

    I’ve been pondering the symbology of the VM, and I’ve come up with a common theme: time.

    Plants have a growing season: the analogous period, for animals, is a generation. If we think of the pipes between the pregnant ladies as generations of 20 years or so, what we have is a kind of timeline. This fits in with at least some of the plants being symbolic, and, of course, with the astrological charts.

    An almanac? A kind of “patrin” of history? Bear in mind that Greek learning came to India and comingled with Indian and even Chinese learning long before refugees came west, to the Byzantine Empire…which spoke Greek, and preserved some of the learning Alexander’s phalanxes had brought to Persia, Egypt, India, Afghanistan, etc.

    This hypothesis helped me unblock in order to continue to work on my story, “Ouroboros:” I hope it helps on the nonfiction side, too!

  107. Saskia

    This might be a silly question but I was just wondering if anyone who has done research on this could tell me if it’s possible that the drawings and descriptions were done at different times in different areas. Is it possible the drawings were done in Europe and then brought further east and writing was added?

    • Marnix Hoekstra

      I have suggested that too. See below.

      • Stephen Bax

        Saskia, Marnix it is a fair question, but there is evidence to suggest that the text and pictures are related more closely. For example people have noticed a few places where Voynich text is written inside parts of plant illustrations, perhaps to tell the artist which colours to use, such as these:

        This suggests that the art and the text were done together, though not necessarily by the same person.

        I wonder why you asked the question? Is it because you feel the pictures seem European but the text more oriental?

        For me that is not a problem, since there are several example of texts written in Europe in oriental languages (Arabic in Moorish Spain for example, Hebrew in Italy etc.), not always in their original scripts, which shows that the language could well be non-European in origin, even when the illustrations and other elements olf the text seem very European, or near European.

        • Saskia

          Oh wow, that’s interesting! I didn’t even notice. Does that suggest the art was done first, then the writing, then the coloring? I wouldn’t have thought it to be done in that order.

          • Stephen Bax

            In most places it is clear the drawings were done first, as the writing fits around them, but the evidence I mentioned shows that they were not done completely separately. Maybe it was drawings, then text, then colouring, as a general rule?

            I think the key thing is that we have to treat the writing and the drawings as being from the same ‘culture’. That culture seems to have a strong European base. However, we must also remember that ‘European’ influences did stretch outside the boundaries of geographical Europe, for example into Turkey, Armenia, Georgia and beyond, with Greek settlements (for example) well into Asia.

    • Saskia.
      It’s a non-trivial historical problem. Good luck to anyone working on it; I feel for them.

      • Stephen,
        You say “… That culture seems to have a strong European base”. I have to say that I’ve seen nothing in the imagery to support that idea, though I know it has often been presented as though proven. At best, what we have are some few and often too-late examples of similar motifs in western works, but on deeper examination most of those examples derive from sources beyond Europe. The ‘welkin-band’ motif is one example – derived from eastern art and adopted into western manuscript art long afterwards. I think the habit of adopting the German translation for ‘cloud-band’ mislead any number of people into believing there was something German about that motif. It’s not European at all. Same for the ‘aegis’ or umbrella motifs, as for the representation of Saggitarius as a standing human archer. That began in the eastern side too. And so it goes.

  108. Matt Owen

    I think that the suggestion that the VM might be Romanes has real merit: the proper names which may have been partially decoded would not necessarily be Romani in origin, since the language has many borrowings, especially from the Greek, since, after fleeing the Persian invasion of India, the Romanies settled in the Byzantine Empire, until it, too, began to be threatened with Turkish invasion.

    Not only is the chronology right, but this theory may give us real insights into the motivation behind the manuscript:

    Imagine a Romani scholar who wanted, not only to pass on knowledge, but to alphabetise Romanes in such a way that their wisdom could be kept safe from the Gadje: it makes sense that the author would have done for Romanes what Saejong the Great did for Korean, Sequoyah for Cherokee, etc. The author(s) would have observed the cultural continuity that a written history had given the Jews, who, like the Rom, were living in diaspora, and that Romanes was already beginning to break up into mutually-unintelligible dialects. Wanting to arrest this process, and to preserve what had been learned in the Byzantine Empire, one of the eastern termini of the Great Silk Road, which had been closed, this scholar(s) would want to preserve whatever learning (s)he/they had in such a way as to be accessible to the Rom, but not to the Gadje. Unfortunately, the author(s) died before the alphabet was taught, so no-one, Rom or Gadjo, could read the manuscript.

    One thing which might help: Romani grammar clearly indicates which words are Indo-Romani in origin, and which are foreign borrowings: a system evidently developed from the initial settlement in the Grecophone Byzantine Empire…interestingly enough, this system shows up in the Romani noun for the language and culture, “Romanes:” “Romano” would indicate an Indo-Romani term. Note that, at the time of the Romani exodus from India, Greek had replaced Latin as the “Roman” language of the Byzantines…indeed, Greeks called their language “Roman” until the Hellenic revival which went with Greek independence from the Turks in the 19th Century (Greek was first established as a language of learning as far east as India by Alexander the Great: witness, for example, the Socratic dialogue of Menander, who was called Melinda in Pali, which is one of the earliest Buddhist texts…this was millennia before the Romani diaspora!) The modern Romanians begged to differ, of course…and when the Rom fled to Romania, they found themselves enslaved, which had a profound effect on Romani culture, separating the Vlax Romani from the others of the Romani Diaspora, which had remained fairly cohesive during the Byzantine era.

    So, let me know if any evidence confirms the theory that a Romani scholar wrote the VM: the carbon dating seems to indicate that this may well be the case, and please consider the motivations suggested above. Kushti baxt!

    PS: There is a form of Romanes, “tacho lav” or exact word, which is reserved for the kris, or Romani civil courts: as such, it plays a role similar to Latin in European jurisprudence, Talmudic Hebrew or Koranic Arabic in Jewish or Muslim Sharia law, and would be a good dialect to know.

    Please note also the tradition of the “Patrin,” or road-sign left so that Romanies could mark the “baro drom,” or migration path, for those following: it was traditional made with plants, especially leaves and twigs. Might the VM also contain patrines?

  109. Marnix Hoekstra

    Here is my Voynich hypothesis:
    Gregorios Chioniades, who died in 1320, was a Byzantine scholar who traveled to Tabriz in Persia to study astronomy. At the time, Tabriz was the capital of a Mongolian empire. European diplomats and merchants visited Tabriz. Marco Polo describes it.
    Upon his return to Byzantium Chioniades became a teacher of astronomy, astrology and medicine. He wrote in Greek and in Persian, and translated Persian and Arabic astronomy books into Greek. His translations are peculiar but that is not important.
    With Chioniades we have a direct link between Greek, Persian and Arabic influences, and a link between herbal medicine and astrology.
    Chioniades did not write the Voynich manuscript though. He had no reason to invent a new script.
    But the Mongols did.
    The Uyghur script that they were using did not fit their language well, so they experimented with Persian-Arabic and other scripts. And in Persia alone, they could have encountered a variety of scripts to draw inspiration from.
    The Mongols had conquered a vast territory. Mongolian doctors traveled to every corner of it to learn from local medicinal practices and record it.
    I don’t know if any expert has compared the Voynich-script with the Voynich-illustrations. But I suggest they come from different hands.
    My hypothesis: The Voynich Manuscript was written by a Persian Mongol, in an experimental script, in a Mongolian dialect that was heavily influenced by Persian and Arabic; and illustrated by a European artist. It could be a translation of a book brought to Persia, or written by, Gregorios Chioniades.
    The Mongolian language has a few interesting features:
    – Mongolian words never start with an R. Voynich-words rarely start with EVA-R. The few cases can be explained by omission of vowels.
    – ‘khar’ is the Mongolian word for ‘black’.
    – Apart from the black hellebore, there is also a yellow hellebore. The Mongolian word for ‘yellow’ is ‘shar’. The second word on the Hellebore-page at f3v could be ‘shar’.
    – Word repetitions like ‘khar-khar’ occur in Mongolian.
    I have only been looking at plant names in standard Mongolian and inner Mongolian dialects. The next step is to examine the Persian-Mongolian dialect, which was quite different. The Mongols never left Persia but their dialect died out in the 16th or 17th century.
    I have to focus on my thesis now, so I cannot investigate this any further and I don’t know how plausible this is. But in light of recent developments, I wanted to share it anyway.

    • Derek Vogt

      Interesting… it gives me a new way to interpret the ligaturization scheme with EVA-ch (the one that looks like two of our letter “c” connected by a horizontal bar on top, which appears to be superimposed on any one of five other letters at times to create another five letters). My main thought so far has been that the letter EVA-ch alone was for the sound “h” and the ligatures were for sounds that were somehow combinations of that and the sound of whatever other letter is used. For “s”, that would mean shifting the point of articulation back to something like IPA “x/ç”. For plosives, it would mean converting them to fricatives (p>f, b>v, k>x/X, g>ɣ).

      The Mongolian sound inventory doesn’t seem to fit that very well; there doesn’t seem to be a pattern of plosive-fricative pairing that would cause the language’s speakers to think of fricatization of plosives as a thing that their language does and their alphabet should reflect. But there is another phonetic effect that does create pairs with this one effect switched on & off: palatalization (essentially, combining the base consonant with a “y”, so “ka” becomes “kya”, although it should be thought of as simultaneous if possible instead of subsequent, so “sa” would convert to “ʃa/sha” instead of “sya”). If that’s what EVA-ch ligaturization is about, then EVA_ch alone could just represent our consonantal “y” or “i”.

      This would also solve a dilemma of mine over EVA-t. Based on both my earlier ideas about the gallows letters and my identification of plant f24v as a gaharu/gahara/ghara tree (spelled essentially “ghatwr/gahatwr” on the Voynich page as I was reading it, and its seed pods are the only things I know of for those green things in the drawing to be), I had predicted that that letter’s sound was “g”. But Professor Bax notes that the letter’s appearance as the second letter after a phonetic “a” in a bunch of names on the star pages looks like the Arabic “al” to start star names, and it’s hard to see a way for the same letter to equate to “L” in some cases and “g” in others that isn’t excessively contrived. Mongolian gives a possible link. It has no “L” sound, so any imported name with an “L” would need to be pronounced with the nearest thing it does have: probably either “ŋ/ng” or “ɮ”, which is like a cross between “L” and the sound that most of us Englishers would write as “zh” (although I don’t think it ever appears that way in our words). And those, especially ŋ, are closer to “g” than a true “L” is, making it more plausible for one of them to equate to “g” in cognates.

      Also, Mongolian has no “h”, so an imported name with that sound in it would also need to be pronounced with the nearest Mongolian sound, presumably “x” or palatalized “x”. Since one of those options is already palatalized, it could be the normal sound of a Mongolian palatalizing EVA-ch on its own, in which case it would fit words where I’ve been reading that letter as “h” just about as well as “h” does. (“Kajhor/kajhoar” even already has both a palatal letter and an “h” in that spot together!)

      • Marnix Hoekstra

        Hi Derek,
        If the VM is Persian-Mongolian then you should consider the Persian sound inventory as well. The Persian Mongols adopted the Persian-Arabic script and eventually also the Persian language. The 15th century was a transitional period, where you might expect a mixture of Persian and Mongolian with the phonemes of both. (Such a mixed dialect still exists today: Hazaragi.)
        Anyway, it’s just a guess.
        Several clues point in the direction of Timurid Persia. On the other hand, the VM does not look like Persian/Mongolian art and writing.
        I’ll wait for a Mongolian speaker to look at this and say: ‘Hey, this is interesting’ or, more likely: ‘This is not Mongolian’ 😉

  110. Hello Professor Bax and Everyone!
    I am currently working on folio 5v
    I seem to be getting some interesting results with the mapping
    EVA:’ch’ –> BAX:’JH’ {which was suggested by one of the common names
    of what i call Marsh Mallow (Malva or Althaea officinalis) being
    ‘Kajhor’ in Hindi which seemed to me a close enough mach to the Prof’s proposed
    ‘K A ? O A R’ to think ? might be JH}

    {EVA:’ch’ is the character that looks like ENGLISH:’cc’ but joined
    together at the top. I think the combined character is called a ligature}
    some of them have a funny little mark above them in the middle but
    this one doesn’t}
    So for example the EVA word ‘chor’ becomes ‘jhan’ which i translate
    into english as ‘plant’.{i based this on the result internet searches such as
    ‘jhar means’
    ‘jhar plant’
    ‘jhar herb’

    I’m thinking the language/origin of the VM might be somewhere in the himalaya
    area at the moment.

    Marco’s last post has helped me map EVA:’t’ –> BAX:’KH’ which is looking good so
    cheers Marco and thanks to everyone involved in Voynich research over the years
    especially to Professor Bax for getting the ball rolling and showing it was possible.

    I highly recommend the excellent book:
    “CRC World Dictionary of Medicinal and Poisonous Plants: Common Names, Scientific Names, Eponyms, Synonyms, and Etymology” By Umberto Quattrocchi

    If you would like to check in on my progress i have started a blog which should
    be reachable at:
    *warning – may contain spoilers*

    I’m afraid its only in english at the moment ‘cos that’s the only language i know
    (plus a smattering of Voynichese 🙂

  111. Why don’t you answer to my comment on the picture ?
    i’m waiting for to know your opinion anxiously…

    sincerly , let winds allows you to continue your great job a long time…

    • Stephen Bax

      Hi Eric. I was giving space for other people to comment!

      Personally, I can’t fully see what you mean. Do you think that there is no real message in the manuscript at all, but only graphical symbols?

      • i take the two hypothesis as true… i think it may be have some message simply coded as you show because graphical is not direct , but the presence of graphic code can’t now absolutly denied…
        As graphics are extended on several pages, text can be extended either.
        I love your site on Canterburry, it inspired me that matters in VM can be grids as on your first picture…
        First evidence leads to think there are not only graphic arguments in code, even if they are a lot, then what can it be else than text ?

    • I like that picture
      those are the two plants that kept me going whenever i had any doubts

  112. Luis Henrique Beust

    Dear Professor Bax,
    greetings from Brazil!
    I have watched all the videos of your research available at Youtube, as well as other approaches to deciphering the Voynich manuscript. I have nothing to contribute to your insightful, delicate and respectful work, but I would like to convey some words of congratulation and praise.
    The way you unfailingly express yourself is an example of noble human humility, of intellectual humbleness and of sincerity of soul and purpose. I get as well enchanted by your beautiful research as by your excellent human virtues, which emanate from every word and silence you express. My deepest congratulations.
    Your example fits beautifully in the long noble tradition of British intellectual effort.
    You may or may not be acknowledged as the scholar who has deciphered the Voynich, (I pray you will) but you have already deserved the recognition of all as a brilliant example of how a gifted scholar shuns the traps of egolatry, vain pride and arrogance.
    Congratulations. Congratulations.
    Best wishes,

    • Stephen Bax

      Thank you for your kind words! I’m sure I don’t deserve them, but I’m grateful for your generous comments.

  113. Derek Vogt

    I coincidentally came across this after having recently spent a lot of time looking at various known alphabets and how they had evolved, so I might have been predisposed to look at it the way I am. I started comparing the whole Voynich alphabet, complete with the Bax phonetic interpretations, with other whole alphabets, with the idea that it could have been developed from another one instead of invented independently. If a known alphabet could be identified as the Voynich alphabet’s nearest relative (whether as its ancestor or a cousin), then that could give a basis for inferring the sound values of some more Voynich letters, which could then be used to infer pronunciations of more whole words to try to test, and so on.

    With an independently invented alphabet, double matches (both graphical and phonetic) with the letters of another alphabet would not be expected; letters for similar sounds would usually look different, and similar-looking characters would usually be for different sounds. Because most alphabets are full of examples of known letters that can quickly & easily be seen to fail to double-match the Bax-Voynich letters, most alphabets, such as Glagolitic, Greek, and derivaties of Greek, are fairly quick & easy to dismiss. (Phoenician wasn’t bad graphically but it was much too ancient.)

    Aramaic-derived alphabets in general are, too, but I’ll mention a bit more detail about why here in order to highlight one that stands out as an exception. They all tend to have plausible double matches for the Bax-Voynich letters for /r/ and /n/ in [Res] and [Nun], but, for the other letters, they usually just don’t match up at all. For example, the letter identified by Bax as a sibilant consists of a single stroke that curves back and forth, but Aramaic-derived sibilants are usually either a single big loop or a “pile of sticks” with corners and branching/junction points and more than two loose ends. And Bax identified some vowels that seem like they could be related to [Waw], which started as the consonant /w/ but has been drafted for use for vowels of similar artiuclation in other alphabets… but those Voynich letters are short and round, whereas most versions of [Waw] are tall and skinny.

    But I found that there is an alphabet which yields reasonable double matches for all of the Bax-Voynich letters: Syriac! It was derived from the Aramaic alphabet, but in a unique way which yielded a handful of crucial differences from other Aramaic-derived alphabets which happen to be exactly the changes an Aramaic-based alphabet would need in order to host Bax-Voynich letters. And the distinction is practically perfectly absolute; the Syriac alphabet has all of these traits while other alphabets have none of them:

    1. A short round version of [Waw] which could singlehandedly explain all three Bax vowels (or at least two of them, with the third one having another option I’ll mention below);

    2. A version of [Tet] with two lines crossing each other in the middle and some other lines/curves at the outside connecting the ends of the crossing lines;

    3. A version of [Kap] which can be drawn without lifting the pen, with a vertical stroke on the right, a top piece running to the left from the top of the vertical stroke with a strong upturn at the left end, and, in one form, even a little loop where they meet in the upper right corner;

    4. A version of [Sade] consisting of a single stroke that curves back and forth;

    5. A known allowance for creation of ligatures, which could help explain the more complex Voynich letters;

    6. A lack of any clear mismatches for Bax-Voynich letters, which other alphabets usually have more of than possible matches—the only Bax-Voynich letter left that isn’t already covered by this could be explained as a ligature, so its match/mismatch status is at worst unknown, at least until the components of the ligature are both phoneticized.

    So I arranged the Bax-Voynich letters (minus the apparent ligature) in their places next to a list of Syriac letters and started trying to fill some of the gaps that were still left with other Voynich letters, this time just based on appearance alone, now that I was outside the realm of Bax’s phonetic help. I didn’t include in this the five letters that seem to be EVA-ch superimposed with other Voynich letters, because their shapes were obviously too graphically complex to find that many Syriac matches anyway and because their apparent status as ligatures meant that, if that’s what they really are, their proper place in this kind of reconstruction should come second, after their separate primary components had been worked out (if that ever happens!). So I tended to forget that they even exist while I was just working on the rest alone.

    I wrote this all as one thing but I’ll cut off this post here and go through my results in a separate response to myself, to control post length and separate the “Hey look Syriac fits” part from the “Where else can we go from here with Syriac” part; from this point on, there will be a bunch of conjecture (by someone who isn’t a linguist) instead of just observation of facts like above.

    • Derek Vogt

      First, EVA-ch, just for organizational purposes even though it was actually one of the last ones I placed… graphically, it’s a pretty good match for Syriac [He], so, if my Syriac connection idea is right, it should represent a sound close to /h/. Although this was found by looking at independent EVA-ch and not considering its apparent ligatures, it happens to be a good sign for the interpretation of those other letters as ligatures, because /h/ is good for combining with other sounds in various ways. For example, combining it with a voiced consonant could signify devoicing, combining it with a vowel could signify emphasis or a backward or upward shift in vowel articulation, combining it with almost any other consonant could signify whatever vowel sound is normally associated with /h/ in that language, combining it with a stop (especially a voiceless one) could signify aspiration or fricatization, or combining it with a sibilant could signify a sound somewhere between /s/ and /h/, with a point of articulation between their points of articulation, such as either /ç/ or /x/, the two sounds represented by [ch] in German. And exactly that kind of sound is already attributed by Bax to EVA-sh, which just happens to look like a ligature of EVA-s (representing /s/ in Bax’s phonetics) and EVA-ch, without being derived from such a combination.

      Thus, the complete set of Bax-Voynich letters can ALL be linked with Syriac counterparts (instead of just all-but-one of them), and the same conclusion (the same proposed sound value of EVA-sh) has now been reached by two independent lines of reasoning (phonetic placement in names by Bax, and construction from the ligature’s apparent component letters by me). This supports my hypothesis not only that there is a Syriac connection but also that it can be used to find likely sound values by graphical correlation (and that Voynich letters which look like ligatures of EVA-ch with another letter probably really are). So, on to some more letters…

      EVA-x is a dead ringer for Syriac [Sin]. It’s also not too far off graphically from [‘Alap] and [`E], but not as close as it is to [Sin]. If it is actually related to [‘Alap] or [`E], then it wouldn’t be much phonetic help to us because those are likely targets for phonetic reassignment when adopted by other languages.

      EVA-q most resembles Syriac [Taw]; second-best matches are [Tet] and [Sin], but [Tet] already has a phonetic match in EVA-d.

      After the above, EVA-L is troublesome because its best potential match is [Sin] followed by [Tet] and [Taw], which all already have other Voynich matches. One solution to this 4:3 ratio would be that at least one of these Voynich letters actually has no specific Syriac connection or belongs with a more loosely similar-looking Syriac letter outside this group, like [‘Alap] or [`E] for EVA-x, or [Pe] or even [Waw] again for EVA-L. (I’d throw in [Semkat] for EVA-d but the phonetic link to [Tet] there can’t be dismissed.) But another solution is that one Syriac letter really does correspond to two Voynich letters in this group because of a historical split, like between our [I] and [J]. Also, before moving on to other letters, I can’t help but note the similarity in articulation & sound among the letters that have ended up in this “group” (together with [Tet] & EVA-d), and suspect that’s not a coincidence.

      If EVA-e isn’t related to [Waw], it could be related to [`E], as is the similar-looking modern Arabic [`Ayn].

      Best match for EVA-j: Syriac [Semkat].

      EVA-g could be related to [Qop], [Pe], [Semkat], or [Mim]. Incidentally, I can’t rule out [Mim] as the Syriac counterpart to EVA-y instead of [Nun], or that it actually corresponds to both, but the phonetic [Nun] connection for EVA-y is more straightforward and [Mim] does have other options, namely EVA-g and one other which I’ll get to below.

      EVA-i is tempting to connect to Syriac [Yod], the cousin of our [I], and maybe the users of the Voynich alphabet really did derive it from Aramo-Syriac [Yod] or import [I] from the Greek side of the Phoenician family. But EVA-i is graphically no better as a match for that than it is for [‘Alap], [Zayn], or [Nun], and anything more complex than one stroke reduces down to one stroke if you simplify it enough, so this one can be imagined to be related to almost anything. And EVA-v suffers roughly the same problem; it’s a dead ringer for Aramaic [Yod] and a weaker but still tolerable match for the the Syriac form, and its placement over other letters could indicate yotification of adjacent vowel sounds, but the only difference from [Lamad] is rotation, and a handful of other letters would also end up the same with enough reduction/simplification. So even though I do figure these are connected to some Syriac letters, I can’t say anything about which.

      EVA-ii plausibly fits with Syriac [Bet], [Gamal], [Lamad], or [`E], all of which consist of two strokes very close together or touching and tend to lean to the left from vertical.

      You will presumably have noticed that I didn’t address any of those tall two-legged letters beyond the one that had already been assigned a sound value. This is not only because they’re the least like Syriac letters and thus the hardest to find matches for, but also because I believe they are the most influenced by convergence toward each other and (possibly even deliberate) pattern-filling rather than separate “natural” development from original forms, which makes them the least likely to graphically resemble their phonetic counterparts anyway. They also could have originated as ligatures combining earlier forms of letters which later fell out of independent use or evolved into new independent forms by the time we see them in the Voynich Manuscript, even without involving the EVA-ch ligaturization I outlined above. (Cyrillic includes examples of this phenomenon, most prominently [Ю].) However, I will note that several Syriac letters include components which strongly resemble components of these Voynich letters, which can not be said of many other alphabets. And, while a few like [Mim] already do include strokes that could correspond to the Voynich letters’ legs, some others like [Qop] and [Pe] and [Semkat] look almost exactly like what a Voynich letter would look like with its legs cut off, which could imply some significance to having added the legs in the first place.

      Also, you will presumably have noticed that my scheme here has nothing for any of Syriac’s three voiced stops. There could be an explanation for this as a group; it might seem unlikely, though not impossible, that the language just lacked voiced stops, but even if the voiced stops were present in the language, they might have been written the same way as other sounds such as the nearest fricatives or unvoiced stops, as was the case with /g/ in Latin until they invented [G]. Or the voiced stops could be among the Voynich letters that I’ve also left alone here, with the big complex letters just replacing simpler older Syriac ones instead of evolving from them, like what happened to [Samekh] in Arabic. Also, in Syriac itself, [Dalat] and [Res] had already converged so much that they were identical other than having dots added just to distinguish them, so maybe one of the Bax-Voynich letters now associated with /r/ was actually /d/, at least in some settings, or maybe only one letter was needed because the sounds merged as well (as they have for many modern Spanish speakers).

      • Stephen Bax

        Thanks Derek

        I wonder if it is possible to have a single table showing the matches between the Voynich signs and the proposed Syriac equivalents? That would help us to see the argument more clearly. I’d be happy to post it here if you are able to send me one.

  114. Marnix Hoekstra

    Dear Professor Bax,
    Have you thought about which foreigners could have had access to expensive writing materials in Italy in the 15th century?
    There was a community of refugees from the Byzantine empire, who left a lasting mark on Renaissance thought.

    • Stephen Bax

      Thank you. Yes, I have mused on that. I’m not sure, though, that it was actually written in Italy. It could well have been brought from further east, for example from Turkey or even further, by refugees of the kind you mention!

      • Marnix Hoekstra

        Thank you. I had read somewhere that it was probably written in Italy, but that turns out to be uncertain.
        Another question: do you have any ideas about how old the text could be? I mean the content and the script, not this manuscript. The VM could be a copy of an older book.

  115. Carmen

    I would like to ask you some questions about your partial decoding.
    1.- If the graph ‘m’ (EVA) is a flourished final ‘R’, how can the common R be explained when occurring at the end as in OROR?

    2.- Oror (again) on f116v (Michiton page), how would it be connected to that context? How could your transcription/translation as ‘Juniper’ be explained regarding the sentence “So nim geis mi[l]ch”?

    3.- On p.1, you state that “the script was possibly devised to encode an unwritten language or dialect “. However, you have focused on Hebrew, Arabic and even Turkish phonemes. Is it not a bit contradictory? As I see your analysis, it would be then an unwritten language which copied and/or transcribed Arabic, Hebrew and Turkish phonemes into Roman abbreviations and ligatures (See Capelli and other paleographers’ studies. D’Imperio also comments on this).

    4.- If, as you have stated on p.10, the script consists of leaving the vowels out (as in the Abjad ), how could this hypothesis explain the vowels ‘a’ ( /ə/ /u/ /wə/) and ‘o’ (/a/) appearing on the Ms?


    • Stephen Bax

      Thanks Carmen for your considered questions. Let me take them in turn:

      1. the graph ‘m’ (EVA) is a flourished final ‘R’: My suggestion is that this sign, which looks like the ‘R’ sign but with an extra downward flourish, might be used to signal the end of particular sense units. It occurs often at the end of lines, for example at the end of the first line of f16r. When it occurs within a line it might be a way of signalling the end of a sense unit and the start of a new one. So I am not saying it is the final form of ‘R’ everywhere.

      2. On page 116v, the last page of the manuscript, my suggestion is that Juniper berries fit Albus’ idea, stated at the Voynich 100 conference in 2012, that the text on that page is about skin diseases, as juniper berries were used for skin treatments.

      3. My emphasis on Arabic and other languages was part of the attempt to unpick the etymology of what are possibly borrowed words. We have no idea what the language of the manuscript is, but in my view it is highly likely that it borrowed words such as ‘Kentaurun’ from Greek via Arabic as the name of the plant in question. So if we assume we are dealing with an unwritten language which was then encoded in this script, it would naturally include borrowed words. In fact it may be that ALL the words we have so far identified are borrowings.

      You mention ‘Roman abbreviations and ligatures’, but of course we have so far identified no secure Roman abbreviations and ligatures. I’m not saying there are none, but none has yet been interpreted in any systematic pattern with a convincing meaning.

      4. Abjads and vowelliing: I suggested that there was evidence of SOME vowel omission, which happens with abjad scripts. Don’t forget, however, that abjads such as Arabic DO have some vowels in their scripts.
      In short, my view is that this script encodes some vowels, but it also omits them in places where the vowel is understood. By way of comparison, Armenian script is not typically considered an abjad because it has comprehensive vowelling, yet in the 15th century manuscript of Amirdovlat Amasiatsi one spelling of the name ‘Centaury’ exhibits exactly the same type of vowel omission I identified in the Voynich spelling of Centaury on page f2r – see my discussion of that example here. So in my view, we should expect some vowel omission elsewhere in the manuscript.

      I hope I have covered everything? These areas are interesting, so feel free to offer other views, ideas or questions to help us all to untangle them!

      • Cosmo

        Stephen, i think there is strong evidence for your suggestion that the transcribed “m” is a final form of another character – “r” being a good candidate – because of how often it occurs as the last character, particularly in the latter folios.

        When “m” occurs within a line, it’s usually at the end of a word, for example:

        There are a few occurrences of “m” within a word but these are quite infrequent and in some cases unclear.

        IMO the consistent usage of “m” as the last character in a block means that it is either an embellished character or an abbreviation. The only question is why it does not occur more often, given how common “r” terminated words are.

        For example, see this plot of “r” terminated words in yellow and “m” terminated words in blue:

        • Stephen Bax

          Thanks for this. I hadn’t thought to investigate it like that. It does seem to suggest that it is a variant of another character, as you say, and ‘r’ does seem likely.

          I suspect there are other examples of ‘terminal’ characters, such as EVA ‘g’ as a terminal variant of ‘d’, but that needs the kind of investigation you have done for ‘m’, and more besides!

      • Marco

        Hello Stephen and Carmen,
        Albus mentioned Pliny in relation with his translation. Here is a passage about the cure of diseases of the urinary and genital organs, from Pliny’s Natural History (XXVIII,60):

        “The ashes, too, of a horse’s hoof, taken in wine or water, are considered highly useful for this purpose; and the same with the dung of a she-goat—if a wild goat, all the better—taken in honied wine: goats’ hair, too, is used, reduced to ashes.
        For carbuncles upon the generative organs, the brains and blood of a wild boar or swine are highly recommended: and for serpiginous affections of those parts, the liver of those animals is used, burnt upon juniper wood more particularly, and mixed with papyrus and arsenic.”

        I also found this recipe for a skin cosmetic that can be of interest (Girolamo Ruscelli, “La seconda parte de’ secreti”, 1559):
        “Take well or fountain water and let it be distilled. When it will be distilled, for each jar of this water, add an ounce of myrrh, the shells of four raw eggs, half a spoon of juniper’s resin, i.e. writer’s varnish, the crumb of half a fresh bun after letting it a whole night soaking in goat’s milk”.

        Finally (unrelated to “oror”, but supporting Albus’ translation) I have found a reference to goat’s meat as a cure for rottenness, “Chirurgia Universale”, Della Croce, 1583:
        “Capra, la carne di quella abbrusciata con l’acqua manda fuori la putredine, secondo Sesto”.

        “Goat: its meat burnt with water expels rot, according to Sestus”

        I guess the author is referring to Sextus Placitus (370 CE ca), but I was unable to find the original recipe.

  116. Deyan

    I’m offering some food for thought:

    1. Study other already deciphered herbal manuscripts, ideally about the same plants, and see what they say. Chances are there would be a lot of similarities in the content. Words like: a specific color, season, some verbs etc. Then you can use the already deciphered phonetics to try to match words in the manuscript with the names of these words in other languages (especially ones from the prime suspect regions – West India, Iran, Armenia etc.)

    2. I strongly disagree there are that many DIFFERENT words in the manuscript. As a professor of linguistics, surely you’re aware that in many languages only the word’s STEM is what is important. If you have the stem in place, then you can put prefixes, suffixes or even other words next to it to change the meaning of the word. This can be to indicate for example – a gender difference or plurality in nouns, or it could be diminutive. In adjectives it can also be to indicate a comparative or ambiguity among other things. But it is especially the case in VERBS, where a prefix or a suffix can drastically change gender, grammatical tense (past, present, future), plurality, PARTICIPLES(!) and much more.

  117. kathy peterson

    Yes, hello. I was looking at page 115 and noticed that all lines start with a number, not a letter. I think this is consistent throughout the “script” only pages, yes? Just curious if you noticed this.

  118. Marjorie Thompson

    Professor Bax,

    Your research and approach to deciphering this manuscript is amazing. After I viewed the video, I looked up information about the work Leon Battista Alberti did into cryptography.

    I am an artist and have researched different aspects of the Renaissance including individual artists, the history of the plague, witchcraft, alchemy, the Medicis.

    I found a book which ad some interesting information which may or may not be relevant. It goes into the Arabic origins of cryptology.

    The methods of coding outlined in this book include using seemingly meaningless symbols for a coding system as well as leaving out letters like vowels in coding, both of which are present in the way the Arabic plant names are spelled out.
    Published in 2002, it discusses manuscripts which were unknown until fairly recently.

    The Medicis were very interested in manuscripts from ancient civilizations. Could this manuscript have been written in code to mask the Arabic knowledge so that only the owner could read it? As I do not know how Arabic knowledge would have been viewed in Europe, I don’t know why it would have been written in code.
    If the owner could have been accused of heresy for owning it, they might have a reason for a coded manuscript. Most manuscripts are very carefully illustrated. Although I am not familiar with every 15th century manuscript, this one seems to
    be not as carefully painted as most.

    In a recent book about the Pazzi conspiracy, the author wrote about how he decoded a Renaissance letter which goes into the political intrigues behind the Medici assassination plot. Although not related to the Voynich manuscript, cryptology was actively used in this time period.

    I don’t know if these connections would help to gain greater understanding of the Voynich manuscript but I know that many people do not think the manuscript is in code. I wonder if this assumption should be reconsidered.

    • Stephen Bax

      Thanks you very much for these thought-provoking suggestions, certainly worth bearing in mind. Thanks again.

  119. Nick Rowlette

    Regarding this document :

    Ancient Greek Illustrated Dioscoridean Herbals: Origins and Impact of the Juliana Anicia Codex and the Codex Neopolitanus

    Nymphaea alba (JAC) looks almost exactly like the lower part of the plant illustrated on page 27 of the Voynich Manuscript

    see Fig. 7. (F) Nymphaea alba (the plant on the right)

    Reference for VM page numbers :

    The Voynich Manuscript (Internet Archive)

    Having now compared these two illustrations, I retract my previous comment that this plant illustrated in the VM may be Sagittaria latifolia or a related species.

  120. Vittoria Feola

    Dear Stephen,

    I’m an early modernist and have a 15-year long experience on mss from the 15th to the 18th centuries. I know very little about the Voynich. But I’d like to offer a few thoughts.

    I have seen in Paris, among cardinal Richelieu’s mss which were brought back from Venice by his librarian, the orientalist Gaffarel, a 15C ms in Yddish which contains drawings of bathing women that look very much like those in the Voynich. Happy to provide the shelfmark of this unstudied ms, if you want.

    Next, the word ‘arar’ for Juniper continued to be fairly common in early modern herbals kept in England and in the Southern Netherland (Belgium).

    Further, the drawings of women in what really looks like semina makes me immediately think of atomistic matter theories. The 15C was one of those times when atomism came up again. The fact that these ladies in semina are connected to structures resembling anatomical parts might confirm that the text is partly about atomism. I have seen several such drawings of semina with little beings in them to illustrate atomism in mss in the Bodleian, Oxford. Again, if you need shelfmarks, just ask.

    Finally, knowing nothing about the Voynich before looking at it today, and reading your paper afterwards (which I enjoyed a lot) it immediately seemed to me like a natural language for which we need good paleographic skills…What do we know of Hebrew/Yddish/Italian Hebrew from the 15C and atomism in Hebrew mss?

    Anyway, I hope some of this helps. Thank you.

    • Stephen Bax

      Thank you Vittoria, for these fascinating contributions. As I am very keen to follow up your suggestions, I would be grateful for the shelfmarks of the books, if you don’t mind?

      Also any examples of ‘arar’ in early modern herbals kept in England and in the Southern Netherlands would be great!

      On atomism – I agree that we could do more on it.

      On Hebrew – some have argued that it might be Hebrew in some form. I can’t rule it out at this stage, although equally there is no strong evidence in that direction as yet. But I agree that it should be kept on the table!

  121. Dana

    I have some mere non-academic impressions to offer. It seems possible that the book was written as a personal journal – not an instructive text. It seems to be the journal of an explorer – a traveler – who was interested in learning about medicinal plants used by far away cultures and native populations in far away lands. The medicine man could have had just enough education in his own language to be able to form such a journal. All extra letters that preface anything were reminders for himself – to recall a person, a place, or a purpose, etc. from whence he obtained the information on the plant. The author could have come from Europe, but his own use of language could have been “home-schooled” if you will, at the knee of parents (or others) who themselves came from different cultures – thus his written language was a mixture, devised for his own use, just to take notes on medicines. Then while dwelling among people groups that were very different from his home land, he attempted to learn the local words for various plants and their uses in those cultures. A personal journal will present a very different sort of book for translation than would an academic book. – Just a thought.

    • Stephen Bax

      Yes indeed. It does have the appearance of a manual of some sort – though we are pretty sure it was written by more than one hand. Thanks

  122. tom hennessy

    I listened with interest to your coast to coast interview and an idea that might be of use. The manuscript seems to indeed shows a depth of knowledge that is likely profound. Additionally its production is painstaking and altogether beautiful suggesting significant effort went into its writing. I’m wonder if this might be a copy of a much older manuscript, likely falling apart, copied over to preserve the knowledge therein. On a coast to coast show some months ago a guest discussed the possibility that Atlantis or Limeira was located in the Indonesian area. There is an obscure language – proto derimedie (probably spelled wrong) that was presented as the nowadays derivative of the former language of that civilization. Might the manuscript be the older, written, language and if so contacting those few who have knowledge of this older language’s present day successor be of help to you.
    Best regards
    tom hennessy

    • Stephen Bax

      It is certainly not impossible that the manuscript is a copy of another, as you suggest. What is so odd about it is that any previous influences are so hard to identify, except in a general way.

      But you are right – once we make more headway on the script and language we would seek to find a related known language, in the hope that it will help us with the decipherment.

  123. Nick Rowlette

    This webpage shows pics of the general plant form, leaves and fruits which are similar to the plant illustrated on page 27 of the Voynich manuscript :

    Sagittaria latifolia

    Note the arrow-shaped leaves in the VM (Voynich manuscript) illustration which arise from a stylized rhizome and the stylized flower/fruit at the top

    Reference for the genus :

    Sagittaria (genus)

    The page numbers for the Voynich manuscript that I am referencing in this post (and the previous one) are from :

    The Voynich Manuscript online / Internet Archive

    • Stephen Bax

      Thanks, that is interesting – does anyone know of the use of this plant in mediaeval herbals?

  124. Thanks for sharing this book matches my Magi research I began in 2013 like a glove. I got the copy of the manuscript from the University a great read. It’s text is written in Old Avestan , not to be confused with modern avestan which characters were modified. It is a dead language only one book (publicly) exists called the Gathas and its translation is still unfinished. I have been able to decipher the pages with the wheels and the plants. Each of the wheels are used to calculate the “Magi” version of how evolution works. There idea is similar to our modern understanding of astrology. It is showing that plants are also effected by this and can be used to re create species that have gone extinct buy have great medical value. Rest of the text explains that this same idea can be used to alter the humans with the goal of reaching (Superman). Just like the philosophy of German philosopher Friedrich Nietzsche wrote about. Ironic he was also avid Avestan researcher. Hope this helps in your efforts to complete the translation.

    • Stephen Bax

      Thanks – I am interested in the Avestan side of things.

      You say that you deciphered the pages with wheels. Do you mean the Voynich page? If so, can you tell us where your translation is? And how you did it?

  125. Nick Rowlette

    The illustration on page 6 of the Voynich manuscript looks like Lesser Celandine (Ranunculus ficaria)

    These plants are very prolific on mu property – I collected them originally many years ago from the site of an old nursery not far from here (Gladstone, Oregon). I’m still doing research on this plant, but I believe they are not native to Oregon, or the U.S., but came originally from Europe.

    I think the plants that I collected from the site of the old nursery were cultivated as an ornamental and escaped from cultivation. The leaves and the flowers sure look the same as the illustration on page 6 of the Voynich manuscript – a pretty good match

    This webpage shows good close-up pics of the plants :

    Lesser Celandine (Ranunculus ficaria)

    • Stephen Bax

      Thanks for this.

      • Nick Rowlette

        The use of the plants illustrated in the Voynich Manuscript seem to be explained in the undecipherable written script. Assuming that the illustrations are stylistic representations of actual plants that we know today, the uses can be determined by our modern herbals referenced by their botanical names.

        In the Voynich Manuscript it is undoubtably much easier to determine the approximate identity of the plants than to decipher the script, but if the plants can be generally identified, they will be a pathway to understanding the script and the underlying language since they coexist on the pages. That’s my take on it anyway.

        The few examples of the plants that seem identifiable to me (so far) are illustrated in a stylist visual language that can be learned by viewing them several times, one after the other, to get a basic concept of the way the illustrator represented them – which is not the way they are represented in modern herbals. Clues to the approximate time period and geographical location that the VM is focused can possibly be narrowed by looking at examples of some of the existing herbals throughout history.

        List of florilegia and botanical codices


        Ancient Greek Illustrated Dioscoridean Herbals: Origins and Impact of the Juliana Anicia Codex and the Codex Neopolitanus

        Herbarium Vivae Eicones … (1532-36)

        De medicinali materia libri … (1552)

        Buch der Natur (1475)

  126. craig riglin

    At risks making comments about the obvious.I find the last page of the manuscript interesting. Its obvious a different language.I would think that is a language well known. It looks like a list of items punctuated by the plus sign (+). A written commentary about the book perhaps. I’m incline to thing in order for a person to be bold enough to write added commentary to the manuscript. He would of had a good to fair understanding of it. If you know this persons language you know his culture and by knowing this you may better understand about the original author. The significance of the last page is how soon after the manuscript the commentary was added.The early it was the more its significance. Is it possible carbon date the ink alone on the last page to determine this?

    • Stephen Bax

      Hello. In fact that last page has been studied in the way you describe. See my 2014 paper for Albus’ study and discussion.

      • Marco

        Hello Stephen,
        I could not find the complete paper by Albus on the web. Do you please have a link to it?

        An abstract is available here:

        It gives the following Transcription with abbreviations and omissions in square brackets:

        “poxleber umen[do] putriter.
        + an[te] chiton olei dabas + multas + t[un]c + t[an]ta[a](?) cer[a]e + portas + M[ixtura] +
        fix[a] + man[nipulis] IX + mor[sulis] IX + vix + alt[e]ra + matura +
        … … (two ciphered words) pals [ein]en pbrey so nim[m] gei[s]smi[l]ch O”

        And translation:

        “Billy goat´s liver for wet rot
        At the membrane you gave oil, then you bring a lot of the much(?) wax, in a
        fixed mixture: 9 hands full, 9 morsels (from) the only just double mature
        … … (two ciphered words), squash it into a paste, then take goat´s milk.”

        I think there is little doubt that the page is partially Latin: “chiton”, “dabas”, “multos” (or “multas”) “portas”, “vix”, “matura” are all Latin words.
        The German is less certain, but here it is said that Erwin Panofsky judged that the manuscript was German:

        And “geissmilch” (goat’s milk) fits well with the illustration on the left of f116v which seems to represent a goat:

        Panofsky also mentions f66r which at the bottom left corner has a few more German words:
        – “der mus del” which is the same as “der Mussteil”-

        Albus’ transcription is rather in line with the earlier transcriptions presented by D’Imperio at pg. 101 (pdf 103):

        In previous comments you mentioned the fact that Albus’ transcription has been disputed. Could you please provide more details?

        • Stephen Bax

          Thanks Marco, that is a good summary of it as I see it.

          I attended that talk and as far as I know Albus has never put out anything more in print on it.

          I found it generally convincing. The scepticism I mentioned came in personal correspondence with some others who attended, but they didn’t give any more detailed reasons. Maybe it just didn’t fit their own ideas.

  127. Steve Roescg

    Thanks for the C2C show! Why not give each symbol a mathematical value based on frequency? This can be done with each suspected language (ie Latin) , the most common Latin letter inserted for the most common Voynich symbol. The next most common Latin letter for the next most common Voynich symbol, and so on. The frequency can be used for any language for comparison, and if any legible words result- ……..

  128. Mike Binion

    I have just finished looking through the manuscript and one thing kept running through my head. It appears to me the manuscript is describing the flow of energy good and bad from different life sources to another and in general how it impacts humans. It also appears that the colors have a definite meaning. They are not necessarily the colors of the objects but how they impact different conditions.

    Keep up the good work.


  129. ozzy

    After my experience in medical plants, this book is a compilation of cures an effective treatment from different types of plants.

    The book is probably writing in Persia, I have my doubts about.
    Most of the culture have his birth in that erea.Pertian golf.

  130. Cody Chatfield

    Hello Mr. Bax, I just finished listening to you on C2C, very interesting!

    A couple thoughts, though they may be old news:

    Have you considered having all the plants in the book identified by professional botanists, then looking at the areas which they do and do not grow, overlapping these areas to identify a central location, and then using that as to refine the language sources? Also, and I think you may have done this, but if you look at the root words or early names or sources of the names for the all plants in the book, you may be able to connect some of the words to the pictures more easily, or look what these herbs were used to remedy, you may be able to cross link some text to an ailment or type of injury.

    If you can identify an action or process perhaps indicated by the characters in the book (i.e. women in the pool at the spa, maybe there is something else of a more cultural root or traditional action implied in the picture) which you could create words from, to cross identify to the text in the book.

    I think that you may do well by approaching it by making as much of a story or medical recipe from the pictures and ignoring the text, and then take the many translations from the key words for the herbs and remedies/ailments and trying to cross reference them.

    Best of luck and I look forward to hearing about your progress,



    • Stephen Bax

      Thanks Cody for some interesting ideas.
      Regarding the first, we have been working on that angle, but one difficulty is that it is so tough to identify and agreement on the plants.

      I like the second idea, but again it is tough to agree on what the images mean – which is part of the interest, and also the frustration. But any ideas anyone has about them could help.

      Thanks again.

  131. ari mendes

    the manuscript was written by someone proficient in the almost by- then extinct language of Crimean Gothic but the author(s) was writing in a little known distinct Gothic dialect of Crimean Gothic which also became extinct and of which almost no one had spoken at any time. It was probably coded by the author(s) to add to its mystery. The author(s) was of German ancestry. possible?

    • Stephen Bax

      Why do you think it is in that script and language, exactly?

  132. D.S.

    it almost looks like it’s writing Backwards in code ??

  133. Aaron

    Druidic medical manuscript brought to Ireland from Atlantis!

    • Stephen Bax

      Thanks, but why those connections?!

  134. YZA-DORA

    Has anyone come up with the theory about the manuscript being that of ancient pre-druid language or maybe an ancient extraterrestrial herbology journal? Just something to make you think about. The pre-druid culture is as old as the earth itself, and they grew and cultivated many extinct herbs that no longer exist anymore. It could very well be an ancient record as well about their true origins of religion, as we know it.

  135. Barbie

    Several years back – the first time I saw an article about this – I wondered (first thought that flitted through my mind) if this was from another planet which might have been connected to what the Bible calls the Garden of Eden. Life here is no doubt connected to Life that began somewhere else. So let’s call it the Cradle of ALL Civilizations which inhabit the Stars. It could have been anywhere with a fertile land.

  136. Mark Will

    Okay, I hate to come across as another o fthe ‘wierd ones’, but I thought about this and cannot avoid it. You are talking on C2Cam tonight abut the VM. You describe the page with the 7 stars and then say you decoded a word coming out as ‘Ta err in’. This is ridiculously wierd thing for me to say, but I remember reading an ‘interview with a reptilian’ manuscript years ago and now it is on youtube as well, if you care to check this out, but in that interview there was an alleged ‘reptilian’ humanoid who revealed that she was native to earth and actually live inside it. They call themselves Terran and one of their religious symbols is the 7 stars as you describe. Again, sorry to come out as a kook,…but it really rattled my memory when you said that.

    Best of luck!

    • Stephen Bax

      Woops! Thanks for that, but I seriously hope that’s wrong. Otherwise we end up as someone else’s lunch!

  137. john bennett

    Hi i was wondering based on the year and content if this could have been the work of nostradamus.

    • Stephen Bax

      I believe that has been suggested, but evidence is thin!

  138. David Wiffler

    Dear Stephen,

    There is a plant here: that has two human heads attached to the root system. There is a similar thing in an old Italian herbal here: (you will have to go to the page and, using the box in the top right hand corner, jump to page 70).

    Don’t know if you have seen this. Hope it helps.


    • Stephen Bax

      Thanks – I had seen the first but not the second, so thanks for drawing it to my attention. That Italian manuscript is really interesting. The drawings are so reminiscent of the Voynich pictures, with that same naive element, plus it is 15th century. I’m grateful you pointed it out to me – I’ll look through it with interest.
      I wonder if Rene Zanbergen might have referred to that manuscript before? Now I have a vague memory of it….

      By the way – the plant (with two heads in the root) which you mention, which plant is it? I can’t see the name in the Italian text.

      • David Wiffler

        This Italian herbal is the closest match I’ve seen for the Voynich m. If you put the two manuscripts side by side the similarity of the colours is quite remarkable, and the root systems look sometimes as if they are from the same artist. Though of course they are different in many ways as well.

        I can’t help but think that the history of this Italian text may tell us something about the Voynich m.

        Sorry, I have no idea what plant that is. I have a hard enough time with my native tongue let alone medieval Italian.

        • Stephen Bax

          Yes, I agree it is fascinating. I am also interested in the Italian manuscript in the Wellcome collection in London which I refer to in my paper, which was even once owned by Voynich. See my posts on ‘the Minim problem’ which cites that text.
          I’ll look through the one you brought to my attention. It does look intriguing, I agree.

        • Marco

          Dear David and Stephen,
          unluckily the writing of this Italian manuscript is so bad that I find it almost unreadable. But I think the first plant (pg. 70) is “Allium Sativum” (garlic). The first word could be “Alio”:

          In Italian, the bulb is commonly called “testa d’aglio” (garlic’s head).

          The bottom half of the page has been transcribed in a more readable hand: it presents a recipe for the cure of headaches similar in style to Albus’ translation of the last page of the Voynich manuscript.

          • Stephen Bax

            Thanks – very interesting. I find this manuscript so interesting that I have added a separate section on ‘mediaeval herbals‘ and add your and David’s comments to a page devoted to this manuscript

            Many thanks – any more comment son it would be welcome,

  139. I’m curious Dr. Bax. What’s your interpretation of the various ‘ligatures’ in the Voynich? For example, the ones that appear to be like cPc. Some transcribers, like Currier consider that to be a ligature and worthy of only one transcribed letter (a ‘W’ in his case) while other transcribers assign it a full 3 letters.

    Sorry if this has been asked before, I did a search of your site with nothing found.

    • Stephen Bax

      In fact no-one has asked me this on this site. In my view they are possibly ligatures, repressing a combination of sounds as in: /t/ + /s/ = /ts/. That makes it tough to transcribe, as you imply.

      Luckily they are relatively rare, but if we can unpack some of the individual parts maybe we can work out what the two combined might be!

      Do any particular ones catch your eye? If there are any which might be part of a proper name which we can then work on, it might start to give us clues?

      • Actually, I was looking at this from the top down. I didn’t have any specific one in mind. And what I’m seeing as ligatures, and what others have seen, may not be the ones you see. They appear pretty common to me. For example, Currier recognized 2 versions of the double c or the cz as ligatures. One with a caret above it and one without. Dr. Strong, I assume you’re familiar with his work, recognized 4 distinct versions of the double c, the same ones Currier did as well as one with a ‘teardrop’ shape above it and one with an apostrophe shape. On Folio 78R, the one Strong worked on, I can definitely see three. Just browsing the Currier font, he recognized 15 of the 34 characters were possible ligatures or at least very similar letters. The m and n shapes are two others.

        Not only will this present some unique challenges to your theory, depending on how you translate them, it may make computational work challenging. The EVA transcriptions only allows one version of the double c and to computationally determine if anything is a ligature, you have to read ahead 2-3 letters.

        Depending on how you plan to handle ligatures, you may be looking at a complete re-transcribing of the entire manuscript as well as a completely new font to handle it, assuming you want to move beyond pen and paper. That being the case, I can definitely see why you said there was still a lot of work left to be done.

        • A quick clarification. The m and n in Currier are the iin and in in EVA.

          • Stephen Bax

            Thanks for your interesting points. I don’t think it affects my analysis, as I haven’t included any ligatures so far, but your point is a strong one, and shows that we do need to look more closely at various features of the script.

            Thanks in particular for reminding us of Currier’s and Strong’s analyses, which I had forgotten.

          • Then, in your opinion, the sh, in and iin are probably not ligatures?

            • Stephen Bax

              Sorry for the confusion – by ligatures I thought you meant the examples which are clearly a combination of two other characters, e.g. cth and cph in the EVA transcription. Maybe in my paper I used the term ‘ligatures’ too loosely, on reflection.

              I am unsure about ‘in’ and ‘iin’, as you can see from my discussion of ‘the Minim problem’ elsewhere on this site, though I’m pretty sure they form something like /ir/ and perhaps /ur/ in the words I identify.

              I’m starting to wonder more and more though, whether the sign we transcribe as ‘n’ might in itself represent a ‘ligature’ standing for a syllable, perhaps /ir/ or even /ri/. I hinted at this in my paper, and I am still working on that idea. One reason for this is that the word I transcribe as KNTAIRN would be more satisfactory if it were read ‘KNTAIR I ON’, as that is the form in which it seems to appear in many languages, and is closer to the Arabic. If so, where is the final I in the script? Could it be part of the ‘n’ symbol? That would give the ‘n’ symbol a reading of /ri/.

              As for ‘sh’, the caret sign above the letter is very interesting, as it sometimes appears above other characters too, and does seem to be a diacritic of some sort. I feel it would merit close analysis in itself – how often it occurs and where – but I haven’t had time to do it and I don’t know if anyone else has?

          • I doubt it’s you creating confusion. A more likely reason is my ignorance of proper linguistics.

            Since my skill set in working with the Voynich is very much top-down, I can only give you some rudimentary insight into the iin and sh.

            As a suffix, the iin appears well over 4,000 times evenly distributed through the manuscript, but not once does it appear as a prefix. It’s the most common 3-letter suffix at 11 percent of all words. What’s interesting is that it’s almost exclusively a suffix.

            Of the two complete transcriptions I have, Takahashi found ‘in’ as a prefix twice, Friedman, once. Again, depending on which transcription you use, ‘in’ is the second most common suffix with almost 6,000 occurrences (this includes ‘iin’). Roughly, 15 percent of all words in the Voynich end in ‘in’. Again, almost exclusively a suffix.

            The double c, which I believe is your ‘sh’ and is ‘ch’ in the transcriptions I have, is by far the most common 2 letter prefix, with almost 6,000 occurrences. As a suffix, only 29 times. But, these counts don’t distinguish between those with and without a caret. It, however, occurs frequently in the middle of words. Total occurrences, 10,600+, almost 1/3 of all Voynich words.

            None of them occur as individual words with any real frequency.

            • Stephen Bax

              Thanks – this is interesting. With regard to ‘iin’ and ‘in’, what worries me is that this analysis could hide variation which we have not taken account of in the transcriptions.

              For example, it looks to me as if most of the minim (i elements) in most words are clearly and deliberately not joined, but in some cases they clearly are. My suspicion, based on Latin scribal practice, is that these joined minims might represent a different sign altogether. This might mean that a proportion of the 6,000 occurrences of ‘in’ and iin’ you mention might need to be reclassified.

              I haven’t had time to look through the manuscript to see to what extent this might affect the transcriptions and counts, but I fear we will need to do it. That will of course have implications for all the statistics ever calculated for the manuscript 🙁

  140. Susan Kerr

    Thanks Mr Bax for your video with its amazing introduction to the Voynich manuscript. I have not felt so excited about anything linguistic, and involving such beautiful botanical illustrations, in years. I studied Classical Languages at Exeter back in the sixties. I am only a neophyte in your topic but I will follow your progress with intense interest.

    • Stephen Bax

      Thank you.

  141. Satish Sharma

    I read your paper with interest.
    I can spot something I thinks is a mistake.
    In the end you have listed different “soundings” of the word black, i.e. Kala as it appears in the root sound.

    Your symbol for Kala in Urbu is clearly wrong — I don’t know urdu well .. so I say it with some caution, and I can’t read persian or arabic .. but it appears to me to be arabic and not urdu or persian.

    your’s etc.
    Satish Sharma

    • Satish Sharma

      Sorry I am wrong..
      I see this notation being used on typewritten pages.
      Writing by hand Key and Aliph would be combined and written more like an open top of B ..(or closed bottom) of K

      s sharma

    • Stephen Bax

      Thanks you – I will look at that.

  142. Jens Bilgrav

    Dear Sir,
    F4r, which on Ellie Velinski’s website has been identified by a Steve D as illustrating flax, has as the first word what you would read as:
    I hasten to point out that the Arabic for ‘flax’ is ‘kattãn’ (long a).

    Best regards – and keep up the fascinating work,

    Jens Bilgrav

    • Stephen Bax

      Thank you very much – I had been working on that and I did look for that possibility. But I am stuck to find mediaeval pictures of flax which might help, but I do keep it in mind as a possibility!

      Thanks for being so observant!

  143. Christo Kyossev

    Dear Mr. Bax,
    Congratulations on your fascinating work.
    May I suggest a little clue:

    In Bulgarian and Serbo-Croatian the word КАНТАРИОН (KANTARION) stands for St. John’s Wort – a widely used herb on the Balkan peninsula.
    Sorry, If someone has already mentioned this in the comments above – I have not read them all yet.

    Best regards


    • Stephen Bax

      Thank you very much – yes, I think it has been mentioned that St. John’s Wort has that name, but thanks for reminding us. It is interesting that the mythological Centaur was linked to several plants, and interesting that the related names survive in several languages. Thanks again.

  144. I find that the plant in f31r ressembles carnation of India (Dianthus caryophyllus, see )

    Out of curiosity I looked for the name in Sanskrit, and I found out that it is called “Moonlight beam” or “Moonlight” (candaka; candra), see: (I haven’t checked, but the word “candle” in English might come from the same root).

    Unless I haven’t mixed up some species, what is interesting is that according to other sources, carnation of India in Sanskrit is called “chandani” (“chandni” in Hindi). See: for crape jasmine / carnation of India (botanical name: Tabernaemontana divaricata).

    I pushed my curiosity a bit further and found out that the root of the word jasmine in Sanskrit is interestingly: “g-n-dh”. See:

    Now, when I go back to your analysis of the word “k o o t o n”, I don’t see much similarity with k/g-n-d(h)/t(h)-n root in Sanskrit.

    However, if we accept the possibility that the VM character “8” might be read as “nd/nt” instead of just “d/t”, then we get closer to the Sanskrit and Hindi root k/g-nd/nt-n. English root k-r-n as in “candra” (see above “moonlight”/”moonbeam”) with the inversion of “n(d)” and “r” is also quite close to Sanskrit.

    What is more, I found out that “chand(a)ni” (thus carnation of India) has been used as a medicinal plant against pain. See:

    What do you think?

    • Stephen Bax

      Thank you for this suggestion. I’ll certainly look into it. If you are right, then we would need to see how reading the 8 sign differently would affect the other readings made so far, so it is not an easy matter.

      One different idea I have been working on is that the Voynich language might in places miss out the /n/ sound before a /t/ or /d/, as it might do in the case of Coriander. If that is the case then the word chandani – CH (N) D N might indeed be related to the reading of K OO T O N, so it is certainly worth exploring. But it would of course need a lot more research, including mediaeval pictures of the plant in herbals, before we could push the argument further.

      So thanks – a really interesting suggestion, which I hadn’t thought of before.

  145. Hi Stephen,

    Today when I was looking through the manuscript I came across something that may be of interest. Maybe you’ve already got it on your radar, but I figured it couldn’t hurt to share, in case you don’t yet. Also, I didn’t see it previously mentioned on this page…

    On 70r, where we have Pisces/March, one of the stars in the Pisces drawing is labeled OTOLAL (EVA transcription). To me this must be the star Alpha Piscium (aka Alrescha, Al-Rišā, and Wài Píng), which is the one that connects the two fish.

    It occurred to me that, while the bulk of the text is written in Language X, certain terms may be written in Language Y (or LangX-ified versions of words in Language Y). Language Y here would be a scholarly language, most likely Latin or Arabic. But instead of giving the Latin/Arabic word in their script, perhaps the writer offered a literal translation of the word. In this way, the OTOLAL could be a phonetic rendition of “Alrescha” just as easily as it could be Language X’s word for “cord.”

    I also think it’s worth mentioning that we should assume spelling will be variant. In pre-typographic texts, it’s common to see the same word spelled in multiple ways within a single text. This might explain why there are so many repetitions… Granted, sometimes there are identical repetitions, but often there are near-repetitions with slight variations; perhaps this was the writer’s way of offering multiple renditions of a word.

    I’m totally smitten (and inspired) by your undertaking. I recently finished my MA in Applied Linguistics, in which I specialized in Internet orthography/capitalization and did a lot of research on medieval calligraphy as background… and something like what you’re doing is what I’d like to keep on researching. If you’d ever happen to have a PhD student opening, I’d love to apply.

    Anyway, I look forward to continuing to follow your work and getting myself abreast with all the research that’s been done on Voynich so far. Fascinating!

    Thank you,

    • Marco

      Hello Tim,
      I agree that the identification of the EVA word “otolal” with Al Rescha is a reasonable hypothesis.
      This has been proposed before, see for instance:

      On the other hand, I cannot see how “otolal” could be “a phonetic rendition of Alrescha”.

      Stephen reads EVA “o” as “a”. Assuming that the beginning “ot” is the Arabic article “al”, a possible CVCVCV phonetic reading is “alabib”.

      I find in “A vocabulary, Persian, Arabic, and English” (pg. 61)
      that “alabib” means “the heart string” in Arabic. Other similar words seem to be related to bindings and knots: “abibah” – a cobweb; “talbib” – collaring; “tambib” – producing joints, or “form knots” according to this dictionary by F.Steingass
      But we know that the Arabic name of the star is Alrescha: if the word is Arabic, why not using the common name of the star?

      One can also note that “Abib” was the ancient name of the first month of the Hebrew calendar, roughly corresponding to March/April (and the Pisces Voynich illustration is labeled “Mars”).

  146. David Mihalyfy

    Appendix 1’s proposals for sign-sound relation of consonants reminded me of several features from Coptic, the phonology of which is understudied:

    * t/d confusion could show non-phonemic voicing (

    * combined notation of resonants and vowels (or vocalic resonants?).

    The proposals of KOORATU as “coriander” also has:

    * double vowels in script (though what that represents in Coptic is unclear).

    Many identifications also contain:

    * vowel + glide combinations.

    Richter from Leipzig has suggested Coptic language death among upper classes by around 1300, however, so that throws a wrench in things.

    Just thinking out loud here.

    • Stephen Bax

      Thanks a lot for this. I must say that I have briefly explored the Coptic connection, looking for Coptic literacy beyond 1300. I find the possibility intriguing, so if you have any more thoughts on it, please do let me know.

      (BTW does your name have a Coptic connection?)

      • David Mihalyfy

        One further idea with Coptic:

        On p. 17 of your paper you note that the plant name is prefixed with p sometimes and t sometimes – and the Coptic definite article is p- (masc.) and t- (fem.).

        Could this be gender confusion over a rare noun?

        If this was Coptic, it would probably be recopied by native Arabic speakers (I assume) as part of the body of knowledge belonging to the Coptic Orthodox Church…? From what I understand, that dead language is preserved there like Latin has been among Roman Catholics. If so, the script could be a mess, b/c the language is dead, possibly b/c of a clash between Arabic and a language without phonemic voicing (if that stage of Coptic also lacks phonemic voicing), b/c of the Greek-ish words picked up into Coptic…

        “Mihalyfy” is Hungarian, translates into “Michaelson” (Mihaly is “Michael” like in the first name of sociologists Mihaly Csikszentmihalyi, -fy is “son” like as in Sando Petofi, the Hungarian poet, whose last name translates into “Peterson”).

        • Stephen Bax

          Fascinating! Thanks. Note that it is not that the first sound is actually a P or T, simply that those are the arbitrary letters used for transcribing those symbols. We do not know what their values are.
          But I have often wondered whether those symbols might be some sort of marker of that type, so your suggestion is interesting.

  147. 9 vs 9

    Has anyone discussed the possibility that there are two sorts of “9” characters? In the transcription code I see only one sort, but when you look at the manuscript, I have the impression that there are two sorts. The tail of one looks almost straight, wheras the tail of the other one seems to be written almost horizontally to the left.

    • Stephen Bax

      I must say that I was musing myself on the exact same question. I have noticed some clear differences between the writing of this character, and I think you might well be right. I haven’t seen other discussions of this but I agree that it is worth looking at.

      Have you found any good examples? It would be best to find them on the same page, so that we can be sure it is not just a difference between scribes.

      • For example f2r 1st paragraph the last two words containing “9” character seem to be different. And if we pay attention we can find the same kind of differences practically all over the manuscript.

        • More precisely not the last word in the paragraph, but the two words before the last one.

  148. Hi–

    I’ve now had a chance to read your paper, and I’m most interested in your letter values.

    As I said in the Nick Pelling thread, I find the VMs alphabet is 22 letters: the 20-letter Latin alphabet plus ‘K’, with EVA q a special case. Of those, you give values for 16, and we agree on 5: EVA o, k, i, ii, and y.

    We also agree that im and iim are combined letters. The ‘platform gallows’ are as well. In my value set #1, iiin = IUM.

    You do have too many single-value letters for ‘R’, and it is at this point that rules uncovered by other researchers come into play.

    Many have noticed it is poly-alphabetic. I have found 4 so far; 3 for the labels, and 1 for the text (I’ve been concentrating on labels, and those values don’t work on the text).

    Philip Neal discovered the substitution rules; that certain letters can be replaced by certain others. I found that this applies to the values as well. In EVA:

    1 value s, x, g, i, ii, m, n, d, y
    2 value [o, a], [Ch, Sh], [e, ee], [l, r]
    4 value k, t, f, p
    7 value q

    Mike Roe discovered the VMs alphabet letter order.

    That’s enough for now. If you wish to further discuss this off-site, please do. I find these comment areas too restricting.

    • Stephen Bax

      Thanks Robert – interesting. I’ll have a look at it and work through what you say.

      Best wishes

  149. eric

    does it sound like a real language, i have used your main idea + other arabic word + a few fill up…

    bak hakben to akam entour amben kbar tour amam abkbor kamchen tour taraten hakbat tor toma

    like i said i dont know old language so i cannot guess how it sound… plz leave feed back so i know it does or dosent sound right….

    • Stephen Bax

      Hi Eric. I’m sorry to sound negative, but it is not possible to make progress by filling in the gaps with Arabic and other words. We will need to look more closely at other names and so on to build up a picture of the script step-by-step before we can work out other words and sentences.

      If we don’t step carefully we will most probably fool ourselves into seeing things which are not really there!

  150. Anna Leuchtenberger

    Hi Stephen –

    What’s your interpretation of the drawings of disembodied heads at the base of the plant in Quire 5, f33r?

    Maybe mandrake?
    What with the mythology of those plants containing homunculi in their roots, it would make sense that literal tiny heads were drawn on.

    Or opium perhaps?

  151. Hello Stephen,
    I just wonder if the “astronomical and mythical connections” could be a relevant approach in order to understand the full meanings? I mean, in many Myths of Creations, “seeds” are used in order to describe “creation and recreation as such”. Link:

  152. I read your paper and while I’m not completely convinced of your theory, I am intrigued. As such, I took the work you’ve done and created an online tool which will search the Voynich (using the Friedman or Takahashi transcriptions) and ‘translate’ Voynich words into ‘Bax’ words and sounds. While it’s limited to 6 results (even though hundreds are possible) due to my limited time to work on writing code, it does appear to work. Entering in the EVA ‘shokaiin’ produces what I would believe is the proper result of: CHAK/ə/UR and/kʰ//a//k//ə//ur/

    Feel free to use it at under the topic of Tools and let me know if it’s working according to your theory.

    BTW, this is still a very new website that I’ve just recently opened so expect a few bugs.


    • Oh, and a curiosity for you. While playing around with this tool that uses your hypothesis, I decided to follow your footsteps and tackle another word on f68r3, the label for what is apparently Aldebaran: dchol day. If I’m processing your table of consonants and vowels correctly, this translates roughly into T*CHA* T/ə/N. (asterisk being unknowns).

      Using your scientific technique, I began googling, specifically Richard H. Allen’s book, Star Names – Their Lore and Meaning and I found this passage regarding the name of Aldebaran.

      In all astrology it has been thought eminently fortunate, portending riches and honor; and was one of the four Royal Stars, or Guardians of the Sky, of Persia, 5000 years ago, when it marked the vernal equinox. As such Flammarion quoted its title Taschter, which Lenormant said signified the Creator Spirit that caused rain and deluge…

      I assume that’s François Lenormant, the notied assyriologist.

      And a bit more googling and it’s pretty easy to confirm that name Taschter as Persian and that it’s even been considered before as the possible Voynich name for Aldebaran.

      Coincidence? Perhaps, perhaps not.

      • Stephen Bax

        Thanks – in fact you are not the first who has suggested that name to me, and I do find it worth pursuing. Another possibility is that the word represents ‘Taygeta’, one of the Pleiades. Until we have identified the middle letters with more certainty, it is hard to say, but thanks for your stimulating discussion, and I will keep working on it.

      • It may be takoloutos = folloving (the Pleiades)=aldebaran.

        • Stephen Bax

          Thanks for this idea, which is an interesting one. However, it is not clear to me where you found this possible Greek word for the star. You say on your site:
          “En cherchant dans Bailly et LSJ j’ai trouvé que ce nom serait «(τ)ακολουθος», avec deux voyelles en plus que notre mot.”

          Can you give more exact details of this so others can follow it up?

  153. eric

    hi, i havent read all the comments so maybe what i will say is already here but anyway i havent got that much fun since a long time so i want to participate…
    first about your tought on
    kaaour hara i realy think your right on it black seed, because if you think about the black helleborus (used to tread psychological default)ex:(if you speak french-Nous devons, dit-on, la connoissance des propriétés de l’hellébore, & sur-tout du noir, à un certain Mélampus, qui étoit Médecin ou Berger, & qui inventa la purgation : il guérit avec ce remede les filles de Prœtus, qui étoient devenues furieuses)
    so if kaaour is black i would make sence that the name for hellebore would start with kaaour(black followed by a particularity of hellebore)
    also this is just guessing but the symbole for k is very similar to an other symbole that as 2 loops on top ,i would guess a similar opposite sound like gu.
    an other thing is, you must have a word for flower or plants or roots or tree that will be back in every text. i have tryed to find it but i have no back ground on old language… anyway off to read all the comments
    keep us up to date on your find

    • eric

      all the woman seem to be white and pregnant, the herbs seem to be medical or of importance (not just a botanical book),there is a lot of astrological drawing (which was a common part of the old medecine), the writting is very clean with very few default, but the drawing are very poor, right now i would guess that this book is a copy of an european medical book made by someone who wroted on a regular basis and who wished to bring it back to his poeple. that personne knew how to read, knew at least 2 language and got enuf money to buy a book.

  154. C

    If the first word in f2v is “kain” (assuming that oo is creating the long vowel), then as touched on by a few comments, I think this would make sense to be related to an Indian (Odia?) word for lotus, these are some from a book on Orissa

    Rangkain: red water-lily
    Dhabala-kain: lotus
    Subdikain: blue water-lily

    • Stephen Bax

      Thanks – very interesting. I read it as K A A U R, and I feel that it could be ultimately related to the Sanskrit kairava, which, according to the Monier-Williams dictionary, is “the white lotus-flower (blossoming at night).” However, in this case we would have to explain the loss of the last consonant.

      Do you have any idea of the origin of the element ‘kain’ in the words you mention?

      • Regarding the loss of the last consonant -v, here are some occurrences. -V or -VA might be a suffix or in some cases an interfix. I noticed that the verb to live in Sanskrit is “jiVAti”, in South Slavic languages “zhiVIti” (to live), “zhiVOt” (life), “zhiVOtinja” (animal); whereas in Russian it is “zhit’ ” (to live), “zhiznj” (life), “zhiVOtnoe” (animal). This indicates that the suffix (interfix in this case) VA=VI,VO might be for some reason absent in some words and present in other words (e.g. Russian “zhit” vs “zhiVOtnoe”).

      • C

        Hello, I’m sorry I do not know the origin and I apologize for misreading your chart! In any case, thank you for publishing your work. It’s incredibly exciting to follow.

  155. First I would like to thank Dr. Stephen Bax for his work regarding the VM. I have actually never attempted to decipher it myself, but as a linguist, I’ve been intrigued by it for years.
    After seeing the video and the last word not matching the picture “kooton”, instead of cotton, I was rather thinking of “kacun” /katjun/ which in Bosnian language is a sort of safran. Now, is there any sort of safran ressembling the illustrated plant in the VM?

    Also, for the word “jeer” in Bosnian it is “zir” /3ir/ whereas the word black is “crn” /tsrn/, which is quite close to Sanskrit. In other Slavic languages “crn” is chrn/chern/chorn. /ch/ can be found in form of /k/ in some other Indo-European languages.

    I’ll be glad if I could contribute in any way.

    I speak fluently Bosnian/Croatian/Serbian, French, English. I also have a good knowledge of Russian and Dutch. I can understand some German, and mainly Slavic languages. I also can read some Arabic.

    For info, I studied French, Russian and English linguistics as well as comparative linguistics of Indo-European languages at a higher institute for translators and interpreters in Brussels (= 5-year university degree).

    • Stephen Bax

      Thank you – it is interesting that several people have noted resemblances in Slavic languages. I suspet that many of them come from Arabic and Persian influences – your points about the words ‘seed’ and ‘black’ are particularly appropriate.

      Thanks for your offer – I feel we do need energetic people with linguistics backgrounds to help and comment!

      • I forgot to clarify that “zhir” in Bosnian means “acorn”, but clearly it seems to come form the root “jeer” (seed).

  156. Ian Easson

    Further to my comment about the obvious reference to the Seven Sisters.

    I spent a few moments looking at the manuscript, and here is what I found:
    – There are sections (called Cosmological), in which so-called “nymphs” hold stars. To me, that means a nymph is actually a representation of a star. That fits with the Seven Sisters interpretation I advanced earlier today.
    – There is a so-called “Biological” section, showing similar “nymphs” in water. I suggest that these instead likely refer instead to similar star clusters like the Pleiades. You need to check these against the star clusters that were known by visual observation about 600 years ago.

    The text of these sections needs to be checked for words or phrases corresponding to things like “star”, “constellation”, “sisters”, “princesses”, “Taurus”, “milky way”, etc.

  157. Ian Easson

    I just saw your video, and when I got to the part showing the “odd nymphs” in the funny colored pool, I thought: Yes, of course, a straightforward symbolic representation of the Seven Sisters, the Pleiades, bathing in the Milky Way.

    You should look in the accompanying text for any astronomical words.

    Hope this helps.

  158. davidrn9

    Hi there Stephen,
    This is so basic I hesitate to suggest it, but – Have you taken the Herbal (assuming it is a Herbal), pics to Kew & asked both their archivists & staff if they could identify any more of the plant species/variants? These would give you, perhaps, more names to work with.
    Best, D

    • Stephen Bax

      Thanks – I have thought of that. (For those who don’t know, Kew Gardens is a leading UK centre on plants.) However, a key thing is not only what the plants are, but what the 15th century author thought they were, and what s/he named them at the time. So although you are right that a modern identification is helpful, we still need to do the library research, and that includes linguistic research….

      Also, many serious plant people are fed up with wacky people like me wasting their time! But you are right – it would be good to get more help in this area, it’s true.

  159. Are you ready for a discussion in sci.lang on the Usenet? You can easily google for it
    and find our current discussion. Your approach is tempting but also raises a couple of questions. In any case, good luck with your work.

    • Stephen Bax

      Thanks – I did have a look, and I wish you luck with the discussion!

  160. On the folio showing what you presume is the pleiades cluster of stars in Taurus you are indeed correct as the large star slightly up and to the left is Aldebaran ( the eye of the bull) with I note a Name(?) attached to it. Another clue I hope.

  161. Dave R

    One more comment after downloading and looking through the VM…

    could the plant on page 53r be a Venus Fly Trap?

    I know the problems with that – native to North America/VM is before Columbus… but just a thought.

  162. Dave R

    I very much enjoyed your YouTube video and appreciate your logical approach to the task of translating this enigma. You have moved me away from my previous view on the VM as being a hoax.

    Some ideas…

    Is it possible that instead of the script being a made-up script for some language, that the script could actually be very ancient, and could have evolved into later ancient script(s) with which we are familiar? In this vein, maybe the VM could be a copy of an older manuscript, which was copied by someone – or some group of people – who did not even understand what they were writing, but wanted to preserve something ancient and considered valuable?

  163. I watch the video and, as an entirely amateur linguist, was impressed. One thing that struck me though is that if this is not a European language (and it seems likely that it isn’t), the people pictured in the book certainly appear fairly light-skinned, have light-coloured hair, wear European-style clothes, and use weapons common in Europe such as cross-bows. The pictures at the very least suggest a European origin – but then were the pictures coloured after the manuscript was written?

    • Stephen Bax

      Yes, you are absolutely right, and that needs to be taken into account. That is why any theory about, e.g. Aztecs, India, or Africa, seems to be questionable, or at least has to work hard to explain the pictures you point out.

  164. GS

    Hi, I just watched the whole video you published and i found it amazing and want to congratulate you on your work although I don’t know much about your area of research.
    However one thing strucked me with the word a’ra’r. And with your explanation I think you just explained a word in my language, Portuguese. And I suspect it is of no importance but perhaps with your knowledge, the fact you might be oblivious to it or whatever reason unimaginable I decided to post this in the small chance it might help with context or anything else.

    The word Arar is a verb in Portuguese and means “to plow” as in to work the land. The way you say it is separated like you wrote the arabic on a’ra’r and the way it sounds isn’t much different. A tipical sentence would be “… arar a terra…” – “… to plow the land…’. For some reason i thought you had said it meant the word ‘seed’ but it was my confusion and my point here, but i’ll write this anyway now since i started already.

    Still i find very odd the word in Portuguese. Dictionaries say something like “To work the land with an ‘arado’ (a plow)”. As in the word came from the object. But it makes no sense or it would probably be “aradar” and arar is just too strange of a word in Portuguese anyway. And to call an object with which you perform the verb “arar” would make an “arado(a)”. If you remove the verb ending ‘ar’ you end up with.. ‘ar’ which means air.

    So to me it seems like we completely adopted this word from the Moors which reigned in Iberia all the way to 1492. And it doesn’t seem like other languages use it other than Spanish. It also doesn’t seem latin to me even though it finishes in ‘ar’ like other verbs from latin.

    You said that word was repeated many times over in the Voynich pages, and at first I thought you said it meant seed. Anyway my point was going to be that long ago “Arar” could be seen as “Seed the land” AKA “To seed” or “Ready the land to seed” . That it could be used as the verb ‘seed’ instead of the noun ‘seed’. But that point is useless now and sorry to take you on this nonsense rollercoaster.

    Still and i bet you know this, next go for the common words. As proper nouns and images run out, you’re left with ‘seed’, ‘leaf’, ‘branch’, ‘root’, ‘flower’, ‘petal’, etc; as well as colors (should be easy to find in the many languages the history of the word for each color and similarities – that plant with the strong ‘blue’ allover it, i’d bet all my money the word was used to describe it and maybe by making a list of blue written in 10 languages you suspect were used, you’ll find it); seasons and perhaps months; senses taste-feel-look-smell; names of areas or continents; which should all be pretty common and repeated along the pages.

    Best of luck and can’t wait to be able to buy a translated version of the Manuscript someday in the future.

    • Stephen Bax

      Thank you – as a speaker of Spanish I am familiar with the verb ‘arar’, but unfortunately it is from a completely different etymology, and not at all related to the Arabic word – in fact the first and third Arabic letters represent a guttural sound (‘ayn’) which does not exist in Latinate languages, so the resemblance with ‘a’ is limited. But thanks for the thought.

  165. Serge

    Good day Mr Bax,
    In the last few days, I have been following the news of your partial decipherment of the Voynich Manuscript with great interest. Given the fact that this manuscript was written in early 15th century, and if I remember correctly, the drawing of the castle wall in the manuscript was of a type that was prevalent in Northern Italy only at that time, I was wondering if it has something to do with the medieval Genoese colonies on the Black Sea coast:

    The language of the manuscript, if it is indeed a natural human language, would probably not be Italian, as Italian would be written in Latin script most likely. What would you think about a possibility of the mystery tongue being one of the languages spoken around Black Sea region in 15th century? That could be Circassian, Alan or Turkic Cuman (?)

    Best regards and best of success in your exciting work!

    • Stephen Bax

      By no means impossible, in my view!

  166. bdid1dr

    Have you barred me? I cannot find the post I wrote to you a couple of days ago. So, just in case I’m being paranoid, I’ll try a more brief approach: google Codex Osuna Triple Alliance. jpg

    • Stephen Bax

      No, not deliberately anyway!

  167. Marcos

    Hello Stephen,

    I’m fascinated with your research, trying to check out a bit more I found that this page of the manuscript:

    could refer to some variety of this species:

    The flower color is different but the leaves a really alike:

    there is also one variation that do have blue flowers, so may be it

    Although following your discoveries, the first word of that page should be read somethiing like:

    /k/ /a/ /u/ /a/ /r/

    I guess.

    What do you think ? I know people who knows a lot about plants, if you like I could ask them to find candidates for each plant in the manuscript. Is there an email where you receive information about this research ?


    • Stephen Bax

      Hi, and thanks. I’ll have a look.

      No, No email list, but keep looking here for updates!

  168. T Reis

    I was watching the video and couldn’t help but notice that the /t/ characters in “Tauron”/”Taurus” looks an awful lot like a lowercase delta, which would make sense due to /t/ and /d/ being similar sounds. Also some of the vowels don’t look far away from “o” and “a”. Are there more similarities between known characters and characters in the Voynich Manuscript? And couldn’t this indicate some degree of European proximity?

    I am an absolute leyman in linguistic, so forgive me is this is a very amateurish question.

    Thank you and good luck on your work,

  169. Myrvin Chester

    Sorry, lost the location. It’s: f116R.Q.45.

  170. Myrvin Chester

    Did someone mention Ancient Egyptian? With your subs, it’s amusing to see that, right at the end of VMS, Akhnaton appears:
    [email protected][email protected]@[email protected]@[email protected]

    @ = your schwa; lower case =EVA.

  171. Steffen H

    I have a few questions about one irregularity in the script. On f56r the writer stretches one letter/symbol weirdly to connect (?) the first two words of the page ( ). Does it appear more than once in the VMS (I didn’t find another one after a skim)? What is your hypothesis what that might mean? Does it imply that the same letter/symbol is used at the second place of the first word and the second to last place of the second word? Does this letter-stretching appear in any other known writing system? Why is it used that rarely (if it is used more than once)?

    • Stephen Bax

      Ah! That is still one of the unsolved questions – these are the so-called ‘Gallows’ characters which are still unexplained.

  172. Isomer

    So i saw your video and i have to admit i really liked it. The way you deducted the meaning and so on, is really appealing, but (yeah there’s always a but) there is one “flaw” if you can call it in respect of the vm: f2v starts with the exact same word as f3v: the word which you deciphered as “kaur”.
    i’m fully aware that there is much guessing going on etc. but it just seems strange (i know everything is strange when it comes to the vm) but still… in you’re way of reading it’s out of place.
    maybe i’ll find the time to look for the word throughout the vm, to see if it’s repaeted more often.
    anyhow i just wanted you to know if you haven’t already noticed.

  173. Andrew Miller

    I’ve been trying to simply “get a feel” for the VM and have began to wonder what’s missing …

    Here’s my quick list …


    Historic landmarks, ie known Cities

    Punctuation – especially on the fully written pages ? This might lead us to believe it has been coded ? As I can rarely think of any codes that do contain any punctuation clues or if written as a “from scratch” language in which century did recognizable punctuation first appear ?



    I’m sure other people could add others ?


    The nymphs on Folio 84r appear to be named / titled ? Could deciphering these names give some much needed clues ?

    + this site gives an easy overview of the pages for any newcomers .

  174. Garcol Euphrates

    The Voynich Manuscript bears a striking similarity in style, form and opacity to the 20th century Codex Seriphinianus. Whatever the content of this Codex, some of the methods of analysis via computer image analysis, as documented in the MS Thesis by J C Stanley (, might be fruitful.

  175. Dear Dr. Stephen Bax,

    I have an idea that may help with decoding this manuscript. I am currently an undergraduate student at North Carolina State University (in the USA). During my time in school, I have spent several years studying the German language and tutoring students in this language. My experience as a German tutor has led me to identify a possible solution to identifying the Voynich Manuscript’s meaning.

    Currently it seems like a considerable amount of work has been done to analyze the text and wording of the manuscript, while considerably less has been focused on the actual “sounds” produced by reading it.

    My idea is that, in order to dramatically expand the potential identification of the manuscript’s meaning, have people try to “read” the text out loud in their native dialect. Then, using audio (rather than text) analysis, try to discover a connection to some language. The meaning behind this idea is outlined below;

    In the past, while trying to help American students understand how to correctly pronounce German words, I have written German words with an English pronunciation. Consider, for example, the German word möchten. The trick with this word is the use of the letter ö, which is not found in the English language, and whose sound does not exist in the English language. This word is notoriously difficult for American students to pronounce because of that very fact — that the sound does not exist in English. How, then, can the students make the sound? My solution was to write the word in a way such that its English pronunciation mimics (but does not exactly represent) its German pronunciation. For example, I would often write the word, for the purposes of pronunciation only, as “moorshten.” When students, with a dialect of American English, read the text “moorshten,” they are effectively correctly saying the German word “möchten,” despite the fact that “moorshten” is not a real word in any language that I am aware of.

    In this case this word could be read by an American, without identifying it or knowing its meaning, but when heard by a German speaker, the appropriate meaning could be discerned. The key here is that the word *must be vocalized* and heard by an appropriate audience in order for the meaning to be discovered. In fact, the unusual writing style is not conducted to conceal the meaning, but rather to reveal it, but only in the appropriate context can it be understood.

    Given this theory, one could deduce the purpose of the manuscript. For example, perhaps the manuscript was written to be read by a speaker of one language and heard by a speaker of another (in an environment of education, for example). I could, for example, write a short book to be read in an American English accent which would be understood by German speakers, despite the fact that the American English speaker may not know or understand any German. In a sense, he or she would be speaking German without knowing it!

    This theory accounts for the fact that the manuscript seems to mimic properties of natural language in some cases while still being awkward in other cases. For example, the phrase “ein sichererer Stuhl” (meaning: a more sturdy chair) is a valid statement in German which repeats the sound “er” several times in repetition. This phrase, written to be read by an American English speaker, could be written like this: ine sish air air air shtool. Notice the awkward repetition of the same word “air” three times, which is one of the notably odd characteristics of the Voynich Manuscript. Personally, it is difficult for me to imagine any other possible circumstance where a word could be repeated multiple times yet contain a valid meaning.

    However, this theory does not account for the unusual imagery in the book, although I personally believe the imagery could be simply nothing more than somewhat abstract doodles which accompany the text rather than assist its meaning.

    In summary, I think that the true meaning in the manuscript can only be achieved by studying it audibly.

    Note: It seems that a theory very similar (or perhaps completely equivalent) to mine has already been proposed by Leo Levitov.

    Thank you for your time and I hope the Voynich Manuscript is soon decoded!

    Joe Murray
    Senior, Mathematics
    North Carolina State University

    • Stephen Bax

      Thanks – I hope you have some success with your method! Best wishes

  176. Adriana

    Professor Bax, Botanists at the University of Delaware recently published a paper identifying a few of the plants as native species of the Americas.

    Perhaps this is a written Nahuatl language long lost/forgotten. We know the Spanish burned the Aztec libraries, this may be one of the only surviving examples of a language that is still spoken today. I’m fascinated with this manuscript, I wish i could dedicate more time to it. I have a degree in Latin American studies and art history so both the mystery and its implications are really important to me. I wish i could help!

    • Stephen Bax

      Thanks – the problem with the theory is the carbon dating, which puts it 100 years before, and also the lack of any convincing link with Nahuatl. But it is intriguing nonetheless.

  177. A.F

    Dear Mr Bax,
    you might well have found something.
    Personally, I’m more in the trade of deciphering Eteocypriot, the Phaistos Disc and Aegean scripts.
    But I once read a statistical Phd dissertation on the Voynich book that made the observation that the text seems to have more textual patterns inside it than European languages when the numbers of letters in the texts are compared.
    This has led me to conclude that the text might indeed be written in some kind of Semitic language, where vowels are usually dispensed with.
    So I’m only half surprised that you seem to be getting positive results, when involving Semitic languages in the decipherment. This is where i would have looked myself. So I would support your research in this direction.
    Personally, I would suggest that you adopt a consonantal approach of the alphabet. Not arar but ! r ! r with ! standing for glottal stop or long a, not just letter a. Look at how Arabic or Hebrew work.
    You probably need somebody well-versed in Semitic and Semitology to make advances in the rest of the text.
    You need to understand full lines before you claim victory. 🙂
    Keep on digging.

    • Stephen Bax

      Thanks – ‘claiming victory’ is a long way off, if ever! Success maybe, if we are lucky.

      As for your kind comments – I do speak Arabic, and also studied Hebrew, and three years of Akkadian, so I do have a Semitics background, and for that very reason I am not now convinced that Semitic languages are the way to go with the VM. The reason you suggest a Semitic language is because of their usual scripts using relatively few vowels – but remember that any invented/human script could do that (e.g. ancient hieroglyphs in Egypt, for a non-Semitic language), and the script could still could encode a non-Semitic language. The Arabic script is used to encode Urdu, Farsi and so on – not Semitic languages – though of course with small modifications.

      So the VM script could be vowel-light, but the language underneath could be from any language family! (Almost)

      P.s. Consonantal approaches have not borne fruit so far.

      • A.F

        What happens when your preliminary values are propagated in the book?
        Do you have pages or paragraphs densely packed with them, that could be worked on more specifically?
        As for Semitic, I was suggesting Semitic because these languages tend to dispense with vowels on a regular basis. I agree that any language including English can be written without the vowels, though this is unusual.

        • Stephen Bax

          If you propagate the values through the document, at the moment you get a mess!

          This is because there are still many gaps, and I might have got one or more wrong, even slightly, plus we haven’t worked out which language it represents… but we’re working on it! Any help gratefully received…

      • A.F

        Which consonantal approaches exist?

        • Stephen Bax

          Sorry, I meant that approaches which have tried to argue that the VM script is all consonants and no vowels have not had any success – no more than that.

          • A.F

            but I think your own preliminary suggestions point at a potentially consonantal system.
            Besides, I’ve just checked that the most frequent signs number 22. Quite an odd coincidence with the Hebrew alphabet!?

  178. Lorxus


    I’m a junior at Princeton interested in linguistics. I just finished watching the video, and I’m interested in helping. I’m going to have a look at the list of proposed plant identifications posted elsewhere and see how your current tentative framework might fit with them.

    Also, as posted elsewhere, you might want to try having the text made computer-searchable and -readable. Possible take advantage of Mechanical Turk for it, or something like that? I agree that it would be tricky, thanks to the minims, among other problems, but a mass transcription to try and average out all the possible errors might well be the way to go.

    • Stephen Bax

      Thanks – that is great. I feel we need more people with a linguistics background helping out.

      You might start at, where you can see a lot of detail about the way the manuscript has already been described and analysed via computers – though, as you say – the ‘minims problem’ is potentially disruptive of the transcriptions we currently have.

      But thanks – the more (young, dynamic and quick-sighted) eyes we have on it the better!

      • Lorxus

        Hold on – are you familiar with Mechanical Turk?

        I heard about it once – apparently, it’s a way to cheaply buy the sorts of tasks only humans are good at – like transcribing these sorts of things. It’s leveraging, in a sense, the same things reCaptchas do.

        I have indeed started at – it will probably take me some time to get properly up to speed.

        • Stephen Bax

          Thanks – best of luck with it, but maybe check out it hasn’t been tried before!

  179. James McDonald


    I wonder if you have considered the possibility that the language might be Romany.

    The timing and some of the details seem to match (Romanies would have arrived in northern Italy a century or so earlier). Most notable is your mention of the the possibility of a connection with Kashmir.

    I tried to find some Romany names for herbs, but it’s not easy. You might have access to more specialised resources.


  180. acce245

    I’ve been interested in this document for some time. I am not a linguist at all. However, this is quite an interesting way to go about it.

    Disclaimer: my Japanese is very limited, so this is just throwing something out there…

    I’d love to be able to help do something, but I’m not sure what someone like myself could contribute. It seems to me, however, that perhaps it has some Japanese influences, in the way it seems to kinda ‘switch’ between two different sets of characters in different contexts (hiragana and katakana, usually for two different purposes, foreign vs native words especially). I’m just guessing wildly here, but perhaps that could somewhat explain the black-black comparison you made, and explain why there appear to be two to three symbols for one approximate sound (lower and upper case letters come to mind as well, or cursive/handwritten or calligraphic/printed styles, say, but in a strange context).

    I think this is most prominent in Corriander. For example:
    VM: K O R A T U
    JP: Ko O Ri A Tu U
    Japanese also has a way to emphasize double vowels in that fashion. Also, if there happened to be two subsets of characters in the VM language, it might explain why there’s two letters that look like ‘R’ or ‘K,’ and could potentially later explain why there are odd doublings in places it might not make sense (I know nothing about Arabic or Hebrew, mind, so if that’s a common thing there too, please forgive me). Those things just jumped out at me as I was listening to your video.

    Honestly, I think your way makes more sense, but I’m just interjecting what I hope is a constructive comment from a mere observer. Great job on your work!

    • acce245

      Well, I’m just gonna put this out there as well. I retract my previous statement.

      I suggest trying Welsh. I decided to pick 1r and run your version of the symbols through google translate with autodetect language to get a start.

      Paragraph two, by that picture that looks like a bird. Taking your idea, I ran ‘adar’ in it, and it came up ‘bird.’ ‘adar un cân’ apparently deals with a song bird, and ‘adar yn cân’ apparently deals with a birdsong. Both are pronounced within your hypothetical sounds. I don’t know Welsh, so I can’t intuit what that next symbol might be.

      Hope that is helpful!

  181. Damon

    Professor Bax,
    Forgive me if I am wasting your time asking a question you have already heard or considered along time ago. I quite ofter look at a problem and possible solutions pop out at me. While I can not implement the suggestion to determine if it is valid or not I have found that when I explain myself correctly my ideas bridge a gap that allows a solution to present itself.

    I watched video and have been to your site and love the approach you have taken. After trying to find the answer on the web and not quite locating what I was looking for I decided I should through it out there so here goes:

    I have seen several places where each character in the manuscript has been input into a table to provide possible translations of the letters. What I haven’t found is weather or not the characters have been digitized into a computer without assigning meaning so that the manuscript can then be typed into a program that would allow searching and replacing features similar to a word document. If the manuscript was entered into a computer so that each character was legible, with a search and replace ability, I feel this may be very useful in the following ways.

    To me a hand written documents are harder to read causing a persons focus to be narrowed to a letter or singe word than to the sentence, paragraph or more importantly the over all meaning. Add to this defects cause by age and wear and tear causes unneeded distractions.

    First let me clarify, I am not suggesting to use a computer to solve the meaning of the manuscript. This method has already failed. What I am suggesting is that it be used for what it is best for, a time saving versatile TOOL only. Don’t use the computer to solve anything, instead use it to highlight patterns, relationships and interesting features that you then could apply your analysis to.

    Second, I work with engineering so I know first hand what happens when you give a task to an computer engineer. The first thing he will do is over think it, expand it, add all kinds of cool features and above all complicate it beyond reason. What you need to find is a preferably a piece of software that is usable as is or can be usable with slight modification. If that fails get a stick and beat the phrase “KEEP IT SIMPLE” into him and then have him create a program that does only what you need it to do. The program should allow the Voynic characters to be typed into a word processor style program. Have search and replace options with the ability to highlight findings and color code changed words. Be able save it, print it, copy and paste and that is pretty much it. Some of the best classes I have ever taken in college have been ones where the homework or lab work had real world applications. With your credentials you should be able to get a group of software engineering students with professor guidance to create a program for you for little to no cost to you and provide an educational experience to them as well.

    With that simple computer program you could replace all the characters in the manual with the characters you feel you have already identified as a translated letter or sound. The computer will let you do this very quickly and can be done on a sentence at a time, a page at a time or the entire manuscript. Now comes the important part… you can use the search feature of the program to look for words that have various levels of complete translations. For words that are fully translated a comparison of that word to the sentence structure, location of use in a sentence, location on multiple locations, and any relationship to the surrounding words not only in one location but through out the manuscript can be conducted. This will provide a kind of proof or sanity check to determine if the suggested translation makes since through out the manuscript.

    For words that are only partially translated a comparison can be made between the relationship of the characters. This could highlight dual purpose letters, (example such as “ch” is meant to modify the sound of the letter c, or things like “ed” are meant to modify the tense of the word).

    The more rules for letter use or word use that are determined should aid to the translation as well as a sort of proof of theory.

    Hope I have provided a useful idea and not wasted you valuable time.

    Thank you for your efforts to enlighten the world.

    • Stephen Bax

      Hello – no, you are not wasting people’s time. However, it is fair to say that many people have used computer techniques, both for big data searches and also in smaller ways such as the one you are suggesting, and I agree that the can be very helpful.

      An important element of this is to have a good transcription, into symbols which a computer can read. A lot of work has gone into this over many decades- see for discussion – but there are reasons to believe that we still need to work on even more accurate transcription systems – see my discussion of minims and how this might upset our idea of our current transcription systems.

  182. Phil

    Could it be a Bogomil dialect?

  183. Doug

    On page 49v of in the upper left margin you can clearly see the symbols 1,2,3,4,5 next to single symbols followed by words. These symbols seem to repeat in roughly the same order (a bit hard to tell due to smudges and resolution level of the digital images).

    Are the 1,2,3,4,5 in the same ink as the text? Are there languages that have similar symbols for their numbers?

    Has this been noticed before?

    • Stephen Bax

      Yes, you can see them better, and zoom in and out, here:

      They have in fact been analysed by a number of people – the Arabic numbers we recognise are assumed to be later additions, unfortunately. The other ones, in Voynich script, as more of a mystery!

      • Doug

        But there does seem to be some consistency between the Arabic numbers and the symbols next to them, no? If so, would that help in understanding something? If the symbols aren’t numbers they could be letters in some order (e.g., we use “bullets,” numbers, and sometimes letters to make a list). Make sense? (If they are numbers, then it seems odd they don’t appear elsewhere alone or in the text for describing something.)

  184. Ian Ison

    Congratulations on a good start. You might want to get a Gujarati speaker to advise you. The L-to-R scripting seems to eliminate Semitic forms & this N-W Indian port region would be ideal for Semitic traders & loan words as well as early European contact which might include the European style of penmanship. Was there a pre-Arabic script related to Gujarati found among the Indus Hindus of Sindh?

    • Andrea

      You’re so on what I think, Ian Ison!
      I totally love Professor Bax’s approach and I love your way of thinking. Has anyone looked at the Voynich in mirror image? If not, it could be worth a try. Whenever I look at the ‘words’ I am convinced I can read/voice them…but it seems to be criss-crossed with other languages. I think Professor Bax should look into the voicing of this script, rather than (what most deciphering attempts did) looking into finding modern words. There won’t be any…the chances of finding the word ‘microwaveoven’ are remote. Not even the Welsh have a Welsh word for ‘microwaveoven’! 😉

    • Teri Nash

      What about a connection to the Romani or Gypsies ? Their origin is Indo-European. The writing in the manuscript reminds me of Sanskrit. They were in the region of Turkey in the correct time frame. They have knowledge of the stars. They are also known healers. Just a theory of mine.

      • Teri Nash

        I forgot to mention Romani codified only in 1971. Sure they had some sort of written language at some point that has been lost.

        • Teri Nash

          I notice the illustrations with the nude women seem to be trying to explain what is happening inside the plants. I think the women are just symbolism. They symbolize the fertility of the plant. What might look like pipes are stems. The women often appear to be inside of pods. Of course all the water in the pictures can be explained by the plants need for water. If you look at the illustrations of the plants then those with the women you will see the same shapes being used. I think the manuscript is like an almanac explaining the use of plants. When to plant them. That is the reason for the astrologic charts. How to care for them, water them etc.. In one of the large drawing with a moon in the center each side has a person holding a plant. I think they are suppose to be in the rows of crops.

          • Teri Nash

            Luca Ghini must of been some how connected to the work in the manuscript. It must of been transcribed in code or in an obscure language by a scribe to keep his studies from being stolen until perfected but some how the manuscript was loss or hidden.

          • Stephen Bax

            Thanks – on the subject of these bathing women, have a look at this interesting article by Taiz and Taiz, which I feel is one of the most insightful discussions of the Voynich in recent years:


          • Teri Nash

            Although, I heard of the manuscript less than 24 hours ago. I agree with Taiz and Taiz article. I can match the drawings of the plants with those of the bathing women. I will send examples later. It is a cross section drawing explaining the inside “workings” of the plant. I speak a bit of Arabic too. I believe you are exactly right with the oror. Also that some letters change as to their place within the word as in Arabic. If you consider the Spice Routes of trade of Europe, the Middle East and India I am certain many plant names were a mixture. I don’t know anything about the languages of India. However, I do know some letters in English cannot be translated to Arabic letters. There is no P for example therefore an Arabic speaker with use a B. A Farsi speaker their soft PA but no hard P sound. I feel still feel there is a Romani connection. Finally regarding the cotton. Perhaps the drawing didn’t included the bowls because they are not what he was most interested in for medicinal purposes. I read about the uses of the seeds in herbal medicine. Many plants are only used medicinally at certain stages of growth. I am a severely disabled Army veteran. I am not a linguist but have a talent for languages. Thank you for sharing this mystery it gives my brain some much needed exercise.

            • Stephen Bax

              Thanks for this [I have edited it as you asked in your other private posting – I hope it’s what you wanted].

              Thanks for your suggestion on cotton – I will look into that. I hadn’t thought of it!

  185. Andrea

    Dear Professor Bax,

    I always had the feeling that the Voynich Manuscript is a medicinal book and not a hoax. Why go through so much trouble to pull future generations legs? That doesn’t make any sense.
    As for it being encoded…hmm…I think you’re on the right track with your thinking of an either forgotten or specialised language. This would explain why it so far could never be ‘decoded’? (No code, no way of decoding). I think of it of a special language used by the medical profession. Given the amount of different languages in Europe and even in the world, it would make sense if all medics/scientists would learn one language in which they could all communicate (think of Esperanto)? I admit that there is a severe lack of examples (none) as the Voynich Manuscript is the only one of its kind. So this theory may not wash. However, it could be the only surviving one, or the prototype. The 15th and the following centuries were bad timing for medics and scientists (alchemists). Think of the Church and the Inquisition. The fact that is was written by more than one person (I read that somewhere) means that a few people must have known about this ‘language’.
    As for the language, it seems to incorporate so many styles/languages, as you have shown in this most fascinating video, that my theory may not be so far off course.

    I’d like to point you into the direction of the German language and perhaps this helps a little in your research. Think of the three German Umlaute: Ae, Oe, Ue (I know they are written as A, O, U with two dots on top of each, but I can’t figure out the Umlaut-coding on my English keyboard. However, I’m sure you know what I mean. The Umlaute are mainly used in plural: Das Haus (The House), whereas ‘das’ is the article and describing an object and is thus neutral of gender. In plural we would say: Die Haeuser (The houses) now you have the Umlaut describing the plural and the article turns into ‘die’ which makes more than one German house female. Sound bizarre, but that’s my native language for you :-). Also think of sounds such as: ‘ch’. It sounds different depending on the vowel in front. If ‘A’, ‘O’ or ‘U’ are in front, like e.g.: Nacht (night), Docht (candle wick) or Bucht (bay) then it is spoken in the throat a bit like ‘hissing’, if you know what I mean. However, if E or I are in front a ‘ch’, as in Licht (light) or ‘Pech’ (tar, bad luck) then the sound is formed with the back of the tongue.

    Also, think of ‘combined nouns’. Where in England you have e.g. a Ministry of Defence in Germany we would knock the two nouns ‘Ministry’ and ‘Defence’ into one word, but moving the noun describing the purpose to the front, so we would have: Verteidigungsministerium (Verteidigung = Defence and ‘ministerium’ is easy to figure out).

    Furthermore, you may want check for certain endings that repeat. In German the ending ‘ang’, ‘ung’ and ‘ing’ are only used in certain nouns and nouns as names are always start with capital letters and every once in while I seem to see a capital letter in the Voynich manuscript. Especially at the start of a sentence (which is normal) but sometimes also in the middle of a sentence. The written German language e.g. can’t survive without the use of capital letters in the middle of sentences, as it would lead to too many double-meanings. Phonetically, this is not a problem as depending on the meaning the words would be accordingly emphasized. Here is an example:

    “Ich habe liebe genossen” (phonetically, depending on what I would like to bring across, I would either emphasize ‘liebe’ or ‘genossen’, so not a problem. However, in written German, I would now have to decide if however wrote this, means to say:
    “Ich habe liebe Genossen” (I have lovely comrades) or
    “Ich habe Liebe genossen” (I have enjoyed making love)

    I am not sure if this is of any to you whatsoever and I probably told you things you already know. But who knows…perhaps this does come in handy for you.

    Thanks for your effort, I hope one day we can all ‘read’ a copy of the Voynich Manuscript.

    Kind Regards,


    • Ae, Oe, Ue (I know they are written as A, O, U with two dots on top of each, but I can’t figure out the Umlaut-coding on my English keyboard. However, I’m sure you know what I mean.
      For future reference you can always copy and paste from some other page containing these characters or you can type:
      alt-132 for ä alt 142 for Ä
      alt-148 for ö alt-153 for Ö
      alt-129 for ü alt-154 for Ü
      alt-137 for ë (hold ALT key while typing 132 on the numerical keypad)

    • Andrea

      For any reference, I am German, I studied Germanistic! Perhaps you really want to stop contributing? I seriously find your input insulting!

  186. Cecilia Martinez Lusarreta

    Mr. Stephen Bax. I have this idea for a long time. The Voynich Manuscript was written by Hildegard Von Bingen.
    Cecilia Martinez Lusarreta

    • Stephen Bax

      I must say I don’t know that theory… but maybe it is worth you researching it?

    • Andrea

      Thank you, Cecilia!
      This would mean that my explaining the German language to Professor Bax, may be of some value to him. Not wanting to critisise, but the ‘von’ in the name would be written in lower case: “Hildegard von Bingen”. The ‘von’ only describes that Hildegard ‘s origin are in the town of Bingen, so no need for upper case 😉
      Omg, I sound as if I lecture you, but I really don’t. The lower case and upper case thing, could be important when it comes to the Voynich Manuscript. So please don’t take my reply the wrong way.

  187. Anna

    Have you considered Vitex Angus Castus for the plant you currently regard as Juniper?

    • Stephen Bax

      No, thanks. I will look into it.

  188. Doug

    Fascinating from a problem solving and pattern searching point of view. But I recall seeing work at the Visionary Art Museum in Baltimore and elsewhere that reminds me of this sort of obsessive writing and drawing. I wonder if this isn’t the work by a very talented obsessive compulsive who is creating a fictional reference book in a fictional language. Of course, it could be deciphered, but perhaps it is not a coherent language. This would explain some of the strange repeating patterns that would indicate a lack of true expression.

    Overall, the analysis I saw in the video made me think that the arguments were extreme cherry picking and apophenia within a huge document, but this is not my field. Nonetheless, the mystery is interesting and the artwork beautiful.

  189. Isto Virolainen


    thank you for publishing this mystery, I love these kind of puzzles!

    To me it seems clear that pages 70-74 (?) are representing calendar dates with some special ordering, with the horoscope symbol in the center to indicate the month, the order is very similar to current one. One exception (among duplicate Ares / Taurus?) is the missing cancer, but from the pdf I have it looks like that page is rolled, between the gemini and lion. (If there is a cancer hiding there, it would proof this theory).

    Also, as the Virgo is holding a star, like all other (nude bathing) women in the circles (and throughout the document), it indicates to me that the circles and drawings represent a fraternity calendar, bathing calendar (of a monastery), spa schedule (to define what herbs are used on each day) or something similar.

    Order in the document with the horoscope symbols in the center: Pisces, Aries, Aries (?), Taurus, Taurus, Gemini, Cancer (Hidden), Leo, Virgo, Equilibrium, Scorpius, Sagittarius.

    Capricorn and Water bearer are missing, unless one interprets the drawing before pisces as water bearer. Perhaps one of the duplicate ares/taurus could be capricorn? Anyways, the drawing order is so close to current horoscope order it cannot be coincidence.

    The sagittarius wields a crossbow, which started to be common in Europe at 1400. So that matches with the age estimates for the document.Also, in medieval times “Some monastic orders supported regular bathing and had baths in the monasteries, while other orders condemned the practice as too self-indulgent. ”

    Other observations:

    – There are 17 letters or symbols used four times in a row in the third circle on the 4 circles picture in page 57v (Question: Are the page numbers added later on to the document? I find it odd to have the numbers similar to western numbering in the document). Old italic language had 17 letter alphabet, for example.

    – In the pictures after women bathing in the water there are jars with a set of the plants. I guess these could be different parfymes or other combinations used for bathing, and those jars as drawn unique could indicate the actual jar containing such a combination of herbs. This makes me think that this document is some sort of fraternity and bathing (spa) manual, possibly used in some old spa to create different kind of spa pools for acquiring certain smell for the skin.

    One of the most common “word” I have seen is the “8and” looking phrase. It could correspond either “with” or “and”, especially if the context of the document is such I am guessing. If the descriptions for each plant are describing it´s usability with other plans as combination, it would be natural to expect the author to write something like: “This plant can be used with Plant A and Plant B and Plant C …” etc. see for example the page after page 25 with the star like plant drawing. The structure of the sentence is having a multitude of “gibberish 8and gibberish2 8and gibberish3 8and”. Problem is that it appears also twice in a row in the near end of the sentence, which I find problematic.

    -Very often the pictures contain 10 nude maidens. Could be symbolic.

    – Page 67 has a definite calendar with 12 “months”, perhaps used to predict/mark the waning waxing status of moon.

    Anyways, good luck with your research! Hopefully these gave you something new to ponder. And if it seems to be a manual used by the (monastery) owner of spa for virgins or nuns (or prostitutes), please let me know!

    • Stephen Bax

      Thanks – very interesting – you will find that a lot of what you say is echoed in research others have done – see if you want to take it forward.

    • Cancer isn’t missing at all. There are various sources to get a copy of the folios. If you look at then yes, 71V only has 3 “horoscope” charts. but if you look at the Jason Davies website : then you can see all 4 signs. for that page. ASame with the Yale University Library website: contains all 4 signs for 71V.

      I see 12 signs which I would describe as:
      70R: pisces
      70V: Goat or ram (aries)
      71R: goat or ram (but should possibly be aquarius)
      (it seems that the first 3 are out of order, could it be that page 70 used to be a larger folded folio, but got separated and reattached before page 71?)
      71V: Goat or ram, goat or ram again(should it be taurus?), gemini and cancer
      72R: libra (scale)
      73R: scorpio
      73V: Sagittarius (archer)

  190. Regarding folio 42v, I would like to suggest that the plant represented is a mandrake or a related plant of the genus mandragora. I say that for 2 reasons.
    1 – because the flower and leaves looks like they can be of a mandrake plant, although the root might be a little thin.
    2 – because of the doodle on the first character of the text. The mandrake has historically been given human-like attributes. When I saw that the writer had been creative and morphed that letter into a woman, I immediately thought of the mandrake because of what it represents.

    I would also like to suggest that the whole book is a alchemy manual. Some argue that it doesn’t look at all like other alchemy manuals of the same time period, but I would argue that if the script is genuine and unlike any other script of the time period it’s because the writer didn’t have any connections with other writers or alchemists outside of his own little cultural circle, and therefore wouldn’t have known the format, language or script of such books.

    • Stephen Bax

      Thank you – I will bear that in mind.

  191. Jarkko Hietaniemi

    Looking at the VM (I’ve got the French 2005 reproduction) some obvious things to try to match are in the astrological chart section: we’ve got something that could be Pisces, Aries, and Sagittarius, in whatever language they might be in the manuscript.

    • Jarkko Hietaniemi

      After Pisces, comes Aries (twice). But even better is a sequence with conceivably Leo, Virgo, Libra, Scorpio, and Sagittarius in their usual astrological order!

  192. Mary Ann

    Since the first time that I looked at the manuscript, I was convinced that it is medicinal treatment journal for women, childbirth and the menstrual cycle.
    Also, I believe that the folio 33v ( sunflower) is an artichoke flower.

    Very interesting!

  193. Myrvin Chester

    PS I should have said the first line of trans is Currier and the second is from Friedman.

  194. Myrvin Chester

    Very interesting. You’ve probably done this already, but I have substituted all(?) your letters into a transliteration of the ms I found on the web. f2r now looks like this:

    ~ is what the trans had as T;
    K N A T R O ch u ur ir are all yours.
    Other letters are original trans.
    I have the rest of the ms done too, but I couldn’t find TAURUS in 68r.

    • Stephen Bax

      Thanks – and now the difficult bit is to make sense of it! But yes, thanks, that is the way forward.

  195. In addition to inviting linguistic experts to the Voynich Convention, may I suggest inviting gardeners, landscapers and botanists? Identifying the stylized botanical illustrations may help identify the general location of the people who created the manuscript.

    Best of luck!

    • Stephen Bax

      Thanks – it is in my mind, but I’m nervous of it getting too unwieldy. Let’s see what interest we get!

  196. Congratulations on the breakthrough. This is very exciting. I hope your method continues to find new matches.

    My only reservation is that it is puzzling that such an elaborate and well-formed book be the only example of this alphabet. Like Jonathan Kuyper says above, how come there are no other books written in this language?

    I used to think Voynich was an elaborate hoax for financial gain, now you’ve changed my mind.

    • Stephen Bax

      Yes, a major conundrum, I agree…. probably the most puzzling aspect of the whole manuscript… give us two years and maybe we can get the answer!

  197. Stephen,
    Your theory certainly does show credibility and as a newcomer to Voynich research it seems a massive and brave leap..

    I have looked at many of the illustrations and became interested in page 71r – such a definite picture, probably a goat or Capricorn, with a clearly designated title must surely warrant investigation ? Especially as we know what we are looking at ?

    My only concern on all the theory is the use of the stylized H symbol (I think you referred to it as the gallows) and such a repetitive use of it on individual pictures on this page

    Do any languages use such a repetative use of possibly a consenent ?

    You have me hooked and as a printer / typesetter this will certainly keep my mind ticking over ..

    – I have to also work very mathematically … How much theory has been made to simple averages ? Surely if we scanned every letter in separately (a massive task I know) and simply count the number of uses for each letter / symbol can we then cross check this with existing alphabets with a law of averages to roughly discern what letters they may now be in current English or other current language ? I looked at Wikipedia and this link was a coincidence !

    As it instantly mentions Linotype !
    Now I’m really hooked !

    • Stephen Bax

      Thanks – a lot of work has been done on letter frequency, in fact – see – but maybe we need a typographical brain to help out! Thanks

  198. Hi Stephen,
    I am a very amateur word lover, professional horticulturist and very free thinker – and am looking at this for the first time after finding out about it a few days ago.

    As I look across all the pages, fairly quickly but looking for ‘big picture’ ideas, this struck me.
    I am seeing a LOT of repetition of words across the pages and within the pages . . . and it almost seems to me that it lends itself to a sort of rhyming cadence or that the style of writing is somewhat poetic.
    Again, in the big picture, it feels to me like this is perhaps a book meant to impart the healing ways of these plants, perhaps in the form of spells (?), which considering the time of ‘publication’ would definitely demand secrecy or coding – and I am totally with you on the notion of this being a one-off, original script that was likely invented solely for the purpose of this publication.
    (wasn’t the ‘Witches Hammer’ text published somewhere around this time? *_*)

    I know this is mostly feeling based, and I will likely continue to go back and forth over the text and try to intuit what may have been said about known plants, as someone with a background in this capacity may be able to see into the words a bit and THEN link them together.

    Has anyone taken words that are exactly the same, counted them up and made a list of repeated words and/or phrases?

    Fun stuff!
    I look forward to seeing if this gets cracked,


  199. Harry Bull

    Congratulations, this is a very interesting paper! I was rather skeptical at first but it certainly seems encouraging.

    A couple things occurred to me that might be worth keeping in mind for the future, and could help to explain some of the more unusual character distributions:

    1) You discuss the idea of the script being more like an abjad or abugida than an alphabet. Could the script also have featural elements? i.e. some of the characters could be encoding features of sounds (such as place and manner of articulation, voicing, aspiration etc.) rather than being complete sounds in and of themselves.

    This wouldn’t necessarily have required modern phonology knowledge – there is at least one real world example of such a writing system in Korean Hangul. Hangul was created in the mid-15th century to represent Korean, which previously had no indigenous writing system. The creator(s) of the Voynichese script could very well have had the same idea as the Korean scribes who created Hangul.

    2) It also occurs to me that some characters might represent suprasegmental features such as tone. As far as I know there are no examples of such a writing system in use in the 15th century, but there are a few modern examples, such as Hmong, which is written in the Latin alphabet:

    Kuv nyiam haus dej
    I like drink water
    “I like to drink water.”

    The v, m, s and j in this example do not represent consonant sounds but instead indicate the tone of the preceding vowel. As with Hangul, this is an innovation that arose from a conscious effort to create a writing system for a language which previously had none.

    If a Voynichese character always occurs in a particular place in a syllable and cannot be explained away otherwise then it might be such a tone marker. We might not expect to see these characters in every syllable but only where necessary to avoid ambiguity. A tonal language might not fit a Middle Eastern, South Asian or Turkic origin but if we’re dealing with a culture that now no longer exists their language may have had features that we wouldn’t expect.

    Admittedly both of these are rather remote possibilities, but I feel with Voynich nothing should be taken as given!

    • Stephen Bax

      Thank you – I was hoping for just such interesting and informative suggestions when I started on this bloggy business! You’ve given me a lot of food for thought. There is a symbol like an inverted V in the script, commonly but not exclusively over one character, which I do suspect is a diacritic of some sort. I hadn’t thought of tone partly because I have no familiarity with tonal languages, but you’ve certainly offered interesting suggestions. Many thanks indeed.

  200. Richard Blumenthal

    Sorry, one more thing that might be of interest. This link shows the evolution of Armenian (can’t vouch for its accuracy). I wonder if the book is a copy of a much older book. The vellum may date differently from the language which may be more ancient.

  201. Richard Blumenthal

    Thanks for the wonderful work on a fascinating mystery. Please have a look at this page containing Armenian (Hayeren) letters. There seems to be a resemblance to the letters you have identified. T = tso, K = key , O = o, CH/X = che and R = reh are all quite close. Perhaps other letters will correspond as well or, at least provide some clues, especially if one can find examples from the time period.
    Incidentally, the Armenian kingdom of Cilicia around that time included the Taurus Mountains.

    • Stephen Bax

      Thanks for this. I’m very grateful for this insight, as I do feel there is potential in the Armenian link. I aim to publish a paper on this soon, so your suggestion is very helpful.

  202. Since you are considering the possibility that the language of the manuscript is Kartvelian, you might be interested in this old Georgian botanical dictionary. It gives dialectical variations of Georgian, as well as the Mingrelian and Svan languages, dictionaries for which are somewhat hard to come by.

    (Of course, this is only useful if you have some facility with the Georgian alphabet. But such facility is quite a bit easier to obtain that facility with the Voynich script!)

    • Stephen Bax

      Thanks – this comment from you is precisely why I set about to publicise my work, so I could hear from people such as you about exactly this sort of book. Many thanks indeed. I’ll investigate, and start to learn Georgian!

  203. Rob Hicks

    Dear Stephen,

    As a researcher of the Voynich manuscript for over 20 years I was naturally thrilled to see the announcement in the press regarding your discovery. Needless to say, I immediately watched your video and then set about the lengthier paper.

    I am sure I will be accused of ‘sour grapes’, but I have to make it clear that I am utterly unconvinced by both your methods and your findings. I am sorry to have to say that so bluntly, but it needs to be said before you fall even further into the trap which has snared so many others before.

    What you have done is, frankly, no different to what has been done by the likes of Strong and Stojko – finding a structure within the text that resembles a word in a known language, then extrapolating to find the meaning of other words from a growing bank of letter to sound associations. Strong found garbled English, Stojko a vowel-less Eukrainian, and you an apparent Asian language polyglot, sans the odd vowel when convenient and using several signs to represent one sound.

    From this you have found a dozen or so tentative words from a text containing hundreds of thousands. Nothing resembling a coherent sentence in any known language, thus in no way a decipherment. Claiming that ‘more is to follow’ is arrogant at best, extremely foolish otherwise. Given any similar text I could easily find a dozen or so seemingly meaningful out-of-context words, indeed I once famously demonstrated how the VMS was written by JRR Tolkein using virtually your method.

    All I can suggest, at this point, is that you consider your alleged findings very carefully. Have you really found meaning in the chaos, or have you found a simulacrum – a Bible Code effect where a large amount of chaos reveals patterns to those who seek them?

    Please feel free to email me if you wish to discuss this further.
    Rob Hicks

    • Stephen Bax

      Thanks – let’s agree to differ! But you say that I followed others in “finding a structure within the text that resembles a word in a known language, then extrapolating to find the meaning of other words from a growing bank of letter to sound associations. ” That is a bit unfair, as I looked at the first word of the pages concerned, not just any word. And I did not make any claims about the language – that would be premature.

      But I feel we will never agree! Thanks anyway.

      • Rob Hicks

        We certainly can agree, when there is something conclusive to agree to. Agreed, you have focused on first words, but that is the only place I can see any consistency. You merrily allow missing vowels, variant spellings, and – from the hunt for matching words using the net – apparently any language that fits closely enough to your crib.

        If it is premature to announce the language of the manuscript, surely it must be doubly premature to claim to have deciphered any of it? How do you account for the use of some of your ‘decoded’ words in other sections of the MS where their meanings, if we judge from the surrounding pictures, are not likely to be the same as the ones you have derived. Have you looked at a concordance of the transliterated text? My own shows most of the words you have ‘decoded’ to be used very regularly, never in the same structure twice.

        This looks like a potato.
        The word next to it therefore means potato.
        I have deciphered a word.
        This looks like a tomato.
        The word next to it does not really match the letters I cribbed from potato, but if I fudge it, leave out the odd vowel, allow other languages and blame any other problems on scribal idiosyncrasies, I can fool myself.
        I have thus deciphered another word.
        I do not know what this is a picture of, but the word next to it, using my crib, is *a*ot.
        A quick look on the Internet, and I find there is a vegetable called a ‘carrot’. I can fudge this. Maybe the scribe left out double letters…

        • Stephen Bax

          Ok, let’s agree to differ and leave it at that!

        • Frank Bailey

          It’s unhelpful in the extreme to tear down without building up. Can you offer an alternate approach that might be more likely to bear fruit? Or are you one of the many who has decided that the manuscript will resist any and all attempts at decipherment?

  204. Could there be a relationship of this writing or linguistic style to the Romani people.

    Romani is the only Indo-Aryan language that has been spoken exclusively in Europe since the middle ages. It is part of the phenomenon of Indic diaspora languages spoken by travelling communities of Indian origin outside of India.

    • Stephen Bax

      I confess I find this idea an attractive one. The problem is to research it step by step… but it would not surprise me completely.

    • Teri Nash

      I said the same thing Arlene. Romani has Indian origins with influences from Armenia, the Middle East, and Europe. It has only been written since 1971 but surely the language was written at sometime in history and lost. Also the Romani people were used in Europe as Healers and Fortune Tellers. They would need to know Botany as well as Astronomy. I believe the manuscript was is a “almanac” of the use of healing plants. How to use them. How to grow them, etc… I believe the illustrations of the women in water are symbolism. They are explaining what happens inside a plant. It’s use of water. In most the illustrations you can see the blossom or petals at the top. What looks like water pipes are stems. The women appear in pods and leaves. You can recognize them in the plant drawings. I believe women were used as they are a sign of fertility. On the star charts one shows people in fields around the moon each holding a plant. Perhaps time to harvest.

      • Teri Nash

        In some cultures they have Botanical Horoscopes. They combing Botany, Herbal Medicine, Astronomy and Tarot. I feel the Voynich Manuscript is along these lines.

  205. Thank you for the work and video, its very exciting I hope to have a chance to hear you speak in london. Clearly there is a lot of work still to do, but I think you demonstrate enough evidence for a starting point.

  206. Carol

    Thank you for posting the video. Simply fascinating! I have a observation and questions (which may already been addressed). The writing seems from an experienced hand. The illustrations look rather “inexperienced”. Do you have any ideas about the authorship of the document as to perhaps multiple people, left-handed or right-handed? The inks used? Also, any comments regarding the left to right writing?

    • Stephen Bax

      Good questions – studies have been made on the handwriting, and it is generally thought to be by several people, which rules out any idea of a lone loony hoaxer! Also, others have agreed with you that the illustrations are rather ‘crude’, though in fact some are remarkably accomplished in my humble view. Finally, I reckon it is pretty well established that it is written left-to-right.

      • Carol

        Thank you for your reply. My question about left to right is regarding the origin of the language.

        • Stephen Bax

          Sorry, do you mean the script?

  207. Daniel Raggi

    Excellent theory and practice. Perhaps you would consider a web broadcast of the 2014 conference in order to provide access to those limited by geography.

    • Stephen Bax

      Absolutely – that is definitely part of the plan. I’ll announce it on the website with full details of how to access it.

  208. rdbrewer

    f2v looks like a lily pad. Has that been identified?

    • Stephen Bax

      Yes, thanks, that is a possible identification – if anyone is interested in which plants others have suggested before, the best place is:

      2v is there as an Egyptian Lotus, among others, but of course that still has to be proven, and matched with a word in the text. That’s the tough bit!

  209. Luis Castaño

    Amazing work! Congratulations!

    I could read about it at this Forum:

    I could also view the whole video and it’s very interesting and well explained. As a Philologist I feel immediately the need of thank you for your research. Thank you so much for the moments of pleasure I had with your video.

    Luis Castaño.

    PD: Apologizes for my poor English.

    • Stephen Bax

      Thanks Luis – por casualidad yo hablo castellano, pero seguramente mi espanyol es peor que tu inglés!

  210. Brian C

    Interesting work! I’m not fully convinced (you’d have to show a few more plant matches haha) but I think it’s a good starting point and your methodology is sound.

    I would like to direct you to 57v – the first and third rings (from centre) have isolated glyphs which seems to be the alphabet. If this is correct, it implies that there is no real order, and shows you which letters are distinct (as opposed to simply variants of each other) which would help you in your methodology. The second ring has alternating sets of words and two glyphs, which is weird… I don’t really know what to make of it, but the letter patterns on this page are so different, there seems to be some significance in it. But at the very least, it pokes a hole in the hoax theory that the meaningful language-like statistical pattern was faked, because this page doesn’t even have it. Why spend so much time and effort making a believable statistical distribution and then just forget about it for one page?

    More importantly, I was thinking about the word sticking out of the top left. The placement was odd just like you noticed with the “coriander” example (off placement is very rare in this manuscript so it must be very deliberate, as you’ve noted). It coincides with the line going through the circles. I can only imagine that it’s a marker saying “start reading the circles here”. The other circular pages don’t have it, but notice that 57v is the first circular page, so of course the guidance would be introduced here. After that I used your proposed sound scheme and the word comes out as “tuira[?]” (the last letter is unknown). Here’s where the fun begins! I’m not a trained linguist so I can’t say this with any confidence, but a quick browse through Wiktionary reveals similar words and concepts:

    – Danish tur meaning ride/turn
    – Romanian tur meaning tour/round/lap
    – French tour meaning turn/go/walk/lathe, related to tourner meaning turn/stir/go/work

    Which all derive from the Latin torno meaning turning and Greek tornos meaning lathe – maybe the Voynich tuira[?] is another derivation?

    Another possibility that seems to fit the guidance better:

    – Proto-Turkic *tör meaning root/origins
    – Old Turkic töri meaning birth/arising
    – Kyrgyz töröt/törölüs meaning birth
    – Turkish türemek meaning birth/rising/emerging/originate

    Assuming I’m onto something, I don’t think confirming either possibility really helps us with the last letter, but as turning/root is a common word it would nail down the language family! What a research foundation that will be! What do you think of my ramblings? Helpful or just amateurish speculation? Coincidence or significance?

    • Stephen Bax

      Thanks for this – you have spotted some interesting things, but don’t forget that a lot of other people have also worked on the manuscript and seen these things as well… some of the background is on

      Still, I agree with your basic approach – the question is to find which language it might be!

      Good luck with it!

  211. Sneha Chandrika

    Congratulations Dr.Bax for giving valuable information on the voynich manuscript that has been a mystery since the time it has been discovered. I would suggest you to refer the Vedas of Hindus.In which the Atharva-veda mainly consists of many medicinal healing properties and the ancient plants and methods used for them.It may be of some use in the deciphering some of the words in Voynich script.

    • Stephen Bax

      Thanks – the Indian connection is an interesting one for me. There do seem to be Indian echoes in some of the pictures, but the problem is then to explain the obvious European aspects as well. But it is a useful angle.Thanks

  212. Saffron

    Re: Cotton. That plant looks much more like a variety of milkweed, whose pods produce a silky fibre sometimes compared with cotton. In English it’s known as “wild cotton” or “cotton silk.” A variety of it is native to North Africa and Asia. It’s interesting that the author of the text might have mistaken it for the true cotton plant. Here are some photos:

    • Stephen Bax

      Thanks – interesting possibility. It would be useful to see if that plant was used in mediaeval times, and if so, what its name was in different languages.

    • Elizabeth H

      Good job, Saffron! I was worried by cotton identification, too, as the drawing looks so much like a member of the compositae. As a longtime gardener with basic botanical training, I think your suggestion is a big improvement.

  213. Just a regular nobody

    I’ve been watching your video with immense interested and I want to congratulate you on your progress. This is amazing.

    I did want to ask, for my own satisfaction; did you rule out the possibility of variations of specific “letters” as just a variation of handwriting?

    • Stephen Bax

      Hi – yes, there is a possibility of scribal variation, but the signs have been analysed and studied quite carefully with that in mind – see for more on that. Thanks

  214. Ronald Carter

    Here is a suggested test, and possible opportunity for a letter expansion for you, and one that steps away from “plant name” process, but not too far away.

    Translate “snake” (or “serpent”) into your possible test/target language(s), then that into “Voynichese” via your proposed sign-sign relationships, and test those results against f49r.

    Arabic – (among others) thu’ban
    Hindi – (among others) sannp

    Snakes, when associated with a botanical drawing such as f49r, can have the meaning that the plant in question might be used as a snake bite treatment.

    • Ronald Carter

      Apologies – “sign-sign” should have read: “sign-sound” in my post.

  215. Sam

    I admit to being a bit skeptical on the statistics – you tell a sensible story where a lot of words begin with “K”, but how likely is there to be such a reading by chance? By a quick read of this transcription, you have 26 paragraphs starting with “K”? That would seem to imply a lot of proper nouns starting with “K” – which might I suppose be consistent with a central Asian language, but seems suspiciously low-entropy to me.

    Also – if this find and replace I just did is right – you have long enough words like AKXADOOOR; AKNchAKUIR; XURSNXNDUl; NKOADNAKOOOA that, if this is related to a living language, you should be able to see what it is. Are there living languages that contain a consonant mix that repetitive?

    My main concern remains the statistics – even approximately, how likely is such a reading by chance? Didn’t someone else claim to have figured out the transcription for garlic somewhere in the text and it was clearly wrong?

    Very tough to evaluate the stats. Still, cool stuff.

    • Stephen Bax

      I don’t dispute the difficulties, far from it. As I see it, we need to work more to decode the script in full, with all its subtleties, and then try to identify the language. Many languages have clusters as prefixes and suffixes which might make up part of the long words you identify, but until we have decoded more, and identified possible languages, we can’t explain them.
      To put it another way, we might only recognise the words in a language if we first strip off prefixes and suffixes… and until we can identify the prefixes and suffixes we can’t do that!

  216. Jonny

    Dr Bax,

    Congratulations on your work – the most exciting development in the Voynich for ages! As much as I was persuaded by your argument, I couldn’t help but notice that, in almost all cases, the first word of each folio and paragraph in the VM begins with one of four letters: 1) the sign that you argue is pronounced with a ‘K’ sound; 2) a sign which looks similar to 1), but with an extra loop at the top of the left leg; 3) a sign which looks like a heavily stylised ‘P’ – sometimes extending quite far to the right; and 4) a sign similar to 3) but with an extra loop at the top of the left leg. This pattern continues in the cosmological and recipe sections. If these are indeed signs corresponding to sounds which begin sentences, is it possible that the VM author(s) had only four starting sounds for the beginning of sentences/names of things, including plants? I have no idea whether there are languages which share that characteristic but would love to know your thoughts.

    PS The first word of f2v, if you follow your translation, is Koour. The word that sprung to my mind was ‘coeur’: the Nymphoides (?) leaf shape looks like a heart… That’s probably a totally amateur effort but thought it was worth a try 🙂

    • Stephen Bax

      Thanks for this. Two of the signs you describe have been called the ‘Gallows’ symbols, and have been noted as frequent ‘starter’ signs, and might be discourse markers merely acting to flag the start of a page, or else might be variants of other signs. The jury is still out! Another possibility is that they are ‘determiners’ denoting for example ‘plant’ before the name of the actual plant, as happens in ancient Egyptian and other languages. We need to test these ideas more.

      As for Koour, another suggestion made by a reader today was an Indian word for water lily, which is intriguing too!

      • Eilidh Asselbergs

        Interesting about the starter letters, I looked through most of the plant pages in the script on the Jason Davies website, and noticed that they all start with one of maybe three different signs, which looked to me to be variants of the same symbol, a bit like your analysis of the different “R” symbols in the video lecture (which was very very interesting by the way!). I also thought surely not every word begins with only 3 different letters, unless like you suggested they are ‘sentence/declaration/plant-indicator symbols. But it also brought to mind the names of the Arabic “Moon mansions” which all start with Al/Ar/An.. . . . of course being variants of the determinate article in Arabic. Could it be worth a thought that it means some like “The . . . . .. . .”plant”?
        Fascinating stuff, and I’ve just reserved Simon Singh’s book on codes. Thank you to Lynere Wilson studying at your uni (I think) for alerting me to your work. . . . . . . .

        • Stephen Bax

          Thanks – a lot of discussion has revolved around the starter letters (see My own view is that some of them are decorative, but others (such as the K I identified) are letters. However, there is a possibility, as you suggest, that the decorative ones serve as markers, perhaps as determiners (signalling a plant for example) or as grammatical markers (e.g. for a definite article) or as discourse markers (e.g. to signal starts and ends).

          Your mention of Arabic moon mansions is interesting (see the Wikipedia article: I’m not sure how you are connecting it with the script though?

          You mention the Arabic definite article with its variants: Al/An/Ar, but of course that is standard, and nothing to do with the Lunar Mansions, as I’m sure you are aware.

  217. Zach Victor

    Thank you for publishing your presentation online. It is fascinating.
    I couldn’t help but wonder about the circumstances surrounding the possibility of cultural extinction. Here’s an imaginary “what if” (disclaimer: total conjecture). What if this work was produced by a person who spoke the language but did not write it, either because he never learned its script or because there was no script. He may have been a botanist or a physician: someone who would have had use for this knowledge. He left his homeland. Why? Maybe he converted to Christianity. Maybe his scholarly pursuits compelled him to travel. Let’s say he ended up in Christian Europe and became part of the clergy. He learned to write Latin. He was fascinated by the troves of manuscripts in monastic libraries, particularly the treatises in his own field of interest. Perhaps he worked as a copyist (monastic copyists could be notoriously uneducated, even virtually illiterate). He took it upon himself to commit his learning to writing–and, moreover, to write in his own language. Lacking a script, he devised one based on Latin and what he may have encountered in other manuscripts. Although writing a language that might have been closer to other Semitic languages that were written right-to-left, he made his script go left-to-right, because that is what he learned writing Latin.

    • Mahmud

      Literate people do not unnecessarily invent scripts, when transliteration affords them so much more convenience. Script borrowing is far more common than script invention.

  218. Will

    Dr. Bax,

    Two questions (based on the video, so apologies if this was clarified elsewhere):
    1. Why did you pass on the “Cotton” notion so quickly? According to Sherwood, f31r is identified as Fleabane. A simple search reveals both Fleabane and Cotton thistle are in the same family ( Another possibility is Conyza although this doesn’t seem to match quite as well with your analysis
    2. Have you considered perhaps the plants in the manuscript are sorted alphabetically? (And if it’s almost alphabetical, perhaps some fuzziness introduced through the adjectives eg. Paprika, Parsley, Black Pepper, Peppermint, …)

    • Stephen Bax

      It has in fact been suggested that it is ordered alphabetically, but so far I can’t see how, as we don’t know the order of the Voynich alphabet!

      As for cotton, many thanks for the point you make. Certainly worth considering. I was coming at it from the Arabic angle, where ‘alqtn’ seems to be used only for our cotton, and not for other plants, but I feel it is worth another look.

  219. Mary S. Lynn

    Dear Dr. Bax: I have viewed your presentation on The Voynich Manuscript with with great interest. In my opinion the Voynich Manuscript is a Botanical Field Guide. The guide includes plant drawings, medicinal uses, names of scholars who discovered/were known to use the preparations (such as bathing in a solution of something), astronomical information that records at what time during the year the plant was in flower or came to seed. For instance, if there is a plant mentioned on the same page as the word Taurus and the picture of the moon, does that plant actually bloom in Taurus? During what constellation or phase of the Moon would the plant have been harvested? If this is a field guide then it was created by someone who necessarily had to travel from area to area in order to locate, observe, draw, and harvest the plants. In other words, a healer. It is written in a unique scientific or botanical language by its creator and would of course record common names for plants employed in each region visited as well as in Greek or in that person’s native written language. This is why it is so hard to pin down where it came from. It came from everywhere in the known world at that time, visited one region at a time during a journey of many years duration. Would it be possible to try drawing a map/plotting where the identified plant species may have been located/utilized? This may help to determine where the journey began and ended. Thank you for permitting me to comment on this important work in which you are now engaged. Mary S. Lynn

    • Stephen Bax

      Thanks – an interesting suggestion. I like the idea of a wandering scholar travelling around with the manuscript tucked under the arm!

  220. Dan Thomas

    Great presentation!

    I think the Romani language deserves a look. According to Wikipedia, the Romanis reached south-eastern Europe by the 14th century, their language was heavily influenced by other languages, and their language was “exclusively unwritten.”


    • Stephen Bax

      I can’t disagree with you! In my view the script was possibly invented for a previously unwritten language or dialect, and I suspect that some Romani dialects would have qualified at that time. But of course that would need a lot of exploration.

    • kenif

      Congratulations, an important breakthrough.
      The Taurus page you give as f68R is not the 68R of the pdf and I can’t find it. Any ideas?

      I had also arrived at Romani, and hence probably Anatolia for the location. The manuscript is likely to be a healing treatise. This inference is based on the main components of herbs, anatomy and astrology. The pictures are not ladies in green swimming pools; they are nature spirits (perhaps today we would say ‘angels’ or ‘homunculi’) doing their healing thing in someone’s bile. The healing aspect, Indian, Greek and Middle eastern word similarities, era and approximate location all suggest Romani.

      • kenif

        OK, thanks, found the page as f68R3 on

  221. This is just a bit of fun in regards to all of this. But my rock band in Virginia made what I called a ‘loosely based concept album’ about some imagined intentions of manuscript and all the strange possibilities of its authorship. We called it STELLAR HOAX. Check it out here if you want to have a listen:

    good work!!!!

  222. I noticed something that might be interesting to you while watching your video. Specifically about the Nigella page. The Turkish translation of Nigella Sativa is “çöreotu” (cheu-ray oh-too). “Otu” means “plant of”.

    The common misspelling is “çörekotu”, which would have meant “pastry plant”, (“çörek” being analogous to and also just a more general name for those kinds of pastries) because Nigella seeds are commonly sprinkled on top of bread and pastries in Turkey.

    So hearing the translations from you in other languages makes me wonder if “çöre” really is just “kara”. In any case, “çör” does sound suspiciously similar to the “kaur” / “char” you found, which kind of aligns with your theory.

    The other thing is, I wonder if you’re looking for help. I’ve found some transliterations of the manuscript into computer-readable form, so I’d be happy to help you substitute your new letters in the whole manuscript, and do a computer-based attempt to find the most easily legible words. I do need some help navigating all the different transliterations and what the notation means though. It all seems very convoluted.

    Great work!


    • Stephen Bax

      Thanks a lot – very interesting. I’ll bear it in mind. I suspect you are absolutely right about the Turkish connection.

      As for your kind offer for transliteration, in fact a lot of that has been done – see Réne’s wonderful site for a lot of information about the transliteration and what people have done with it.

      Thanks again for your interest!

      • Cody Raskin

        the frequency of the cluster “kor”, “kur”, or “kar” in the MS is really strange. when looking at the page of month names, for instance, and using your letter substitutions, the month names look like “Xokaon –kur”, “-o- Xr akur”, “nkur nku–“. it seems unlikely that a language could use so many variations on the word “black” for so many things.

        • Stephen Bax

          In some ways your point is right, but it could also be a suffix, as well as a word for ‘black’, just as in English we have a pattern ‘red’ for the colour, which also appears in ‘hatred’, and in ‘kindred’. In other words, the pattern could quite easily mean different things depending on its context.

  223. faraboot

    You’re on Reddit now. Prepare to be boarded by cats, dogs and the humans with opinions about everything.
    As for your work, kudos. Once you’re done, can you please get your hands on the bible, and lets be done with it.

    • Stephen Bax

      Thanks – that must explain why this site had 147 unique visits on Monday and 3391 by 9am today 🙂 That, or the BBC.

      Glad so many people are interested in dusty old manuscripts, as I am!

      • Crytash

        If i may ask, why did you use wikipedia as a source in the video? Was that to “dumb down” it a bit and that people could look for themself?

        • Stephen Bax

          Sorry, where did I mention Wikipedia? For some of the pictures? Actually I reckon Wikipedia is usually pretty good, but of course you are right – it would be better to go to the original sources.

      • Matt

        I think you will find that there are a lot of people who, like myself, are fascinated by such a mystery and a good story/theory to go along with it. Good luck with the translation!

  224. Given your analysis, what do you think the reason is that most “plant names” (assuming that they are always first word in a page) start with either /k/ or a similar looking letter (that resambles a P)?

    At first I thought this was alphabetical but as I progressed I noticed that most of the starting word letters are those two.

    Page 4 of the manuscript also satrts with the word Koour, which is reffered to a plant from the Liliaceae family on the book “A manual of Gardening or Eastern and southern India” (1885, page 101). The image looks like a water lilly leaf (the flower however looks like a white hibiscous).

    • The author of the book is “R. Riddell” and it’s title is “A manual of gardening for western and southern India”. I made a typo originally

    • Stephen Bax

      Thanks Liran, very interesting. I’ll certainly get the book. I had been working on that plant too, and noted that there is a Sanskrit word for white lily related to the word you identified. Many thanks for your lead.

    • kenif

      I’ve been calling that character a ‘thorn’ because of it’s similarity to the Old English runic letter. If it turns out to be pronounced ‘th’ you saw it here first…

  225. Andrew Kozloski

    Hi. I am having difficulty finding a reliable channel for contacting you, despite the available information. Is there really a “Stephen Bax” profile on facebook? Why not offer a direct link?

    On to the substance. I find the provided texts to be somewhat lacking–they are quite faded–a problem which is exacerbated by the imaginary alphabet.

    Does anyone here agree that we would all benefit as a community by first constructing an agreed-upon orthography for the complete text rather than expecting a group effort to move forward on the basis of these poor photocopies?

    I volunteer to help produce this canonical text, if the interest exists.

    • Andrew Kozloski

      I suppose it goes without saying that if we had such a transcription, we could also employ computational methods in the investigation of its meaning…

    • Stephen Bax

      Hi Andrew – yes, I do have a very boring worky site on Facebook, but you can contact me here too.

      Such an orthography does exist – it is called EVA and can be found at Rene Zanbergen’s site

  226. It occurs to me that the decoding might make a good project. The study of the lives of ancient Greeks, digitizing the Oxyrhynchus collection, is an example of how the digitization and decoding of the VM might be crowd sourced, which could bring complete translation in a reasonable amount of time.

  227. Dear Prof. Bax,
    Thank you for all this work. I watched your youtube clip ( and then had a read of your m/s ‘A proposed partial decoding of the Voynich script’. Despite taking some linguistics and semiotics in undergrad, this is not my area of specialty, so please take my comments with a big grain of salt!

    I really have just one query. You explanation of cultural extinction is interesting, but as you note in your work, the Voynich script is the only example of this language. How could this be possible? When one takes a spoken language and creates a written version through an alphabet or semiotics, surely one cannot start with a treatise on nature. It must take much trial and error to ensure that letters/symbols are consistent, that there is not too much overlap between letters/symbols in terms of sound, that the alphabet complies with grammatical rules and conjugation, and so forth. Given this complexity, there must have been many other pieces which began building this process.

    This would comport with the other examples you gave. Glagolitic slavic was developed over (at least) a 20 year span and used Macedonian slavic as the framework for letters. There are many examples of Rongorongo and we know that much more was lost in the 1860s.

    So I realize you cannot know what might or might not have happened to other examples of Voynich. However, does it not stand to reason that many other attempts of writing must have been developed before writing this treatise? Especially given that the Voynich text eventually turned up in Italy, we can assume that this was not some hermetically-sealed experiment. If so, looking for the kind of event that could destroy many other examples of writing/people in the mid-1500s might be a fruitful path forward?

    Thank you again for your exciting and hard-earned work!


    • Stephen Bax

      Thanks Jon…. You are right to highlight these problematic issues. I have no answer now, but my hope is that if we work to get a decoding……..

    • You explanation of cultural extinction is interesting, but as you note in your work, the Voynich script is the only example of this language. How could this be possible?

      There are some scripts that get developed, but don’t get used, and the common pattern (at least in the last few centuries) for pushing a script is to publish a book in it. When the Shavian alphabet for English was first developed, there was only one book published in it. I wouldn’t be surprised if that turned out to be the case for some of the invented Albanian scripts, of which there were six or seven that I can think of.

      It could be that the script was ordinarily written on a material that doesn’t preserve well (that didn’t stop Rongorongo from being preserved, but there’s much less time depth there), or that other material in it just hasn’t been found yet (the first of the Novgorod birch bark documents — dated to the 15th century — was only found in 1951).

      The Elbasan script (one of those Albanian ones) was used in one book and (as far as I can tell) a few minor contexts and nowhere else. It apparently had a variant that’s only attested from one source:

      When one takes a spoken language and creates a written version through an alphabet or semiotics, surely one cannot start with a treatise on nature. It must take much trial and error to ensure that letters/symbols are consistent, that there is not too much overlap between letters/symbols in terms of sound, that the alphabet complies with grammatical rules and conjugation, and so forth. Given this complexity, there must have been many other pieces which began building this process.

      Does this hold for Shavian, Osmanya, those Albanian alphabets, and so on? (I don’t know.)

    • Jeff Woods

      I just viewed the Youtube video, but have not looked at your more formal paper. However, I have a couple of suggestions for consideration.

      – I presume that in the early 15th century Italy had a substantial number of scribes meticulously copying older documents, partly to make additional copies for circulation but also as the only way to preserve the information recorded on deteriorating manuscripts. It is entirely possible that this was a handmade copy of an older document, perhaps much older.

      – The Voynich manuscript may have been created in Italy, but it could have been written elsewhere and was later brought to Italy from anywhere in the world. If the Voynich document is a copy of an older document, the same situation could apply to the older document. Hence, the Voynich document may have been brought to Italy after it was created. Likewise, the hypothetical earlier document may have been relocated before the Voynich copy was made.

      – It is possible (though probably a bit far-fetched) that the hypothetical earlier document may itself have been a copy of an even older document.

      – The original document could very plausibly have been created elsewhere in what is now Eurasia or Africa.

      All we can say with reasonable certainty is that the original information recorded is at least 600 years old. However, it is potentially much older and could have been created across a large swath of place and time (relative to recorded human history, of course).

      It seems likely to me that the original document was created in an area where it’s spoken language was understood, than that it was written by someone who had traveled to another place and learned an alternate language that had a preexisting written form. A displaced traveller (e.g. someone who came to Europe and learned to read and write Latin) would have written the document so it could be read by those around him, not in a newly created alphabet to represent a previously unwritten language that few if anyone around him spoke.

      The theory that this alphabet was created to record a previously unwritten language seems very likely.

      Keep up the great work. It’s very interesting. Thanks!

      • Stephen Bax

        Thanks – I agree that all of what you suggest could well be possible. Thanks again.

  228. Mark Osborne

    My wife is a linguist and I a more former linguist took a look at the manuscript a few years ago. Our interpretation was it was likely a personal natural language borrowing from a middle-eastern language. It was very unlikely to be a hoax and appeared to be from a sort of tome of early Hermeticism. The likely borrowed language being related to Hebrew, Egyptian, or Arabic.

    It’s nice someone of greater ability is finding that we weren’t too off the mark.

    Naturally, the word size, spacing, and near 1:1 ratio, and the often repeated sets of ending characters reminded me of Hebrew just written left to right. Given the historical time period if assuming it was a work of Hermeticism of the 15th century, Hebrew, Sanskrit, and Russian are strong possible root languages.

    Regardless, keep up the good work.

  229. This is a nice try, but I just applied it to two lines on f13r which run like this:

    shkchy. ydal. shy. ykchy. qot*y. daiin. s. ys. y. dchor. shaiin. oeees.

    and it gives me a discouraging result, a mix of alphabet and abjad:


    f13r is identified as a banana plant by Sherwood, but the top left word, “torshor”, translates in your scheme to ?ARKHAR, which does not map to any word for banana I can find.

    f5r, described as wolfsbane (Greek/Romance languages: akoniton, Turkish: kurtbogan, Kashmiri: mohri, Sanskrit: visha, Hindi: bachnag), has in the top left corner “kshody”, which translates in your scheme to KKHATHN, KXADN, KTSATN or something like that. Your scheme could be modified so that this creats KNATN but then your existing decipherment of KH/X/TS loses value.

    Anyway, this is the result of a few minutes of work by someone new to the Voynich decipherment project. I do like the approach to deciphering though.

    • Oh, I just noticed that I was relying on one transcription for f5r “kshody”, but another gives “kchody”, making KNATN slightly more possible. Interesting!

    • Stephen Bax

      Thanks for this. You are right to highlight some of the frustrating issues we now face, and I don’t deny them, but my suspicion is that we are dealing with a language which is not common – PLUS the plant identifications are very tricky. I reckon on around six months research per plant!

    • Marco

      I wonder if F13r could represent an artichoke.

      The first word (EVA “torshor”) could read something like (?)ARCHAR

      According to Quattrocchi’s book that Spud linked in his comment here:
      Arabic words of artichoke include “kharshaf”, “khorchef” (hence the Italian “carciofo”). Indian: “hirshuf”.

      • Stephen Bax

        Thanks Marco. Curiously, I have been looking at another Voynich plant using more or less the same reasoning. I considered plant 93r possibly to be the artichoke, as it looks very much like one to my eyes. The first word reads ‘kodshol’ in EVA, which could be read as KATCHA?. One Hindi word for artichoke is ‘hatichu’, and like you I also thought of a resemblance to Arabic ‘Kharshaf’. I’m still looking for old manuscript references to artichokes which could help us!

        BTW, I am now pretty sure that EVA ‘t’, which you mention in reference to page F13r, can be read with some confidence as /l/. This is because I have been engrossed recently in studying Voynich stars systematically, and one conclusion is that the common prefix ‘ot’ must derive from the Arabic definite ‘al’, with many of the Voynich star names deriving ultimately, though rather distantly, from Arabic star names. I am still working on the research paper, but I aim to release it in the next month.

        In terms of the first word of F13r, I suspect that the first letter (EVA ‘t’ or /l/) is possibly a prefixed discourse marker.

  230. Calvin de Wilde

    I read news about your attempt to decode it. I hope you succeed sir. greeting from New Zealand.

  231. Carol Larkin

    My father, BT Larkin spent a great deal of time during his last year alive working on the Voynich Manuscript. He told me that he had cracked the code and then shredded all his research and notes because “He didn’t want to spoil someone else’s fun.” Dad had an incredible mind! The only clue that I can remember was a symbol he named ‘mickey mouse ears’ and that symbol indicated to move either forward or back two spaces. Perhaps you might enjoy following up on this clue!


    Carol Larkin

    • Stephen Bax

      Thanks Carol – sounds fun! There is a character that looks like that…. I like the idea and will check it out!

      • Stephen,

        You are my hero. You have no idea how impressed I am reading this reply. If everyone could be as thoughtful as you, the world would be a much better place.

    • Sounds like a Turing Machine!

  232. Elena Kuzmanovska

    I read about your most recent findings about the Voynich manuscript in a BBC report, and I was most astonished to read the word Kantairon linked to the picture of the herb Centaury. I would like to point out that in macedonian language (a part of the south slavic languages, with it’s Cyrillic alphabet) the word for the plant Centaurium erythraea is Kantarion (precisely Crven kantarion which means Red Kantarion). There is also the plant Zolt Kantarion (Yellow Kantarion) known by it’s latin name Hypericum perforatum or St John’s wort.
    The word Kantarion is also used for the same plant in Serbian, as well as in Bulgarian language.
    Can this be of any help?

    • Stephen Bax

      Thanks Elena. I am surprised to see how many languages have the form similar to ‘Kantarion’. I suspect that this must come ultimately from the Arabic version of the word, because in the original Greek there is no ‘n’ at the end. So it is indeed helpful to see that it exists also in similar form in Macedonian, Serbian and Bulgarian.

      To my mind this reinforces my analysis, but of course I’m biased, sorry! 🙂

  233. Ilija Pavlic

    Another similar word for Hypericum perforatum (from Centaurea) family is “kantarijon” (or “kantarjon”, “kantarion”) in Croatian and related languages.

    Similarly for ch/x-ar, Croatian has historical and local variants “crn”, “čarn”, “črn”.

    • Stephen Bax

      Thanks Ilija – very interesting indeed… see also Elena’s posting. Looks like the Centaury plant has had a long and well-travelled existence!

  234. SirHubert

    Will you forgive one small observation on methodology? You state on page 8 that “Ventris’ decipherment of Cretan Linear B with the help of Chadwick…made successful use of essentially the same systematic ‘bottom-up’ approach: finding individual proper names in the data and gradually building up from them a set of letter-sound correspondences, then finally identifying the underlying language.” This is correct but slightly misleading as currently expressed.

    The key to Ventris’s decipherment lay in identifying patterns of inflection in Linear B, from which he was able to deduce which characters shared a vowel or a consonant. He then constructed a grid with vowels on the x-axis and consonants on the y<-axis, and was able to position many Linear B characters more-or-less correctly within this framework. So although he didn't know, and made no assumptions about, the values of particular vowels or consonants, he was able to say that a given Linear B character was made up of, say, consonant 6 and vowel B. It was only when he had gone as far as he could with this process that he started experimenting with values derived from place names such as Knossos and Amnisos. Even then, he was trying to avoid making assumptions about the underlying language of the tablets, choosing toponyms because these often survive other linguistic changes.

    Entering even a few values into his grid gave values for vowels and consonants of many more signs, and the rest followed (with some corrections) logically and consistently. Ventris was as surprised as anyone to find that he was looking at recognizable Greek, and it was only at this point that he began collaborating with John Chadwick.

    I suspect you know all this at least as well as I do and have glossed over some details as being unnecessarily technical for an article about the Voynich. And I wouldn't presume to claim that this in itself has any particular implications for your methodology or results. But it does seem to me that your approach is rather different from that which Ventris used, and in particular you're assigning values to symbols at a far earlier stage than he did.

    • Stephen Bax

      Thanks for your insightful observations. In essence I agree with your account, but decided not to give too much detail in my paper because of space. Wikipedia has a good account of Ventris’ procedure:

      “Shortly after Evans died, Alice Kober [whose role we mustn’t understate!] noted that certain words in Linear B inscriptions had changing word endings — perhaps declensions in the manner of Latin or Greek. Using this clue, Ventris constructed a series of grids associating the symbols on the tablets with consonants and vowels. While which consonants and vowels these were remained mysterious, Ventris learned enough about the structure of the underlying language to begin guessing.

      Some Linear B tablets had been discovered on the Greek mainland, and there was reason to believe that some of the chains of symbols he had encountered on the Cretan tablets were names. Noting that certain names appeared only in the Cretan texts, Ventris made the inspired guess that those names applied to cities on the island. This proved to be correct.”

      So yes, Kober and Ventris had made much progress on the grid, but the point I was making in referring to their work was that the crucial leap was Ventris’ linking the text to known proper names in the way which the Wikipedia article describes. That is the leap I was trying to make with the Voynich manuscript.

      Since you are clearly a well-informed critic [flattery!] I’d be interested to hear your views on the later parts of my paper, the part where I move step-by-step from Taurus to Centaurea? Any criticism, positive or critical, is always helpful! Thanks.

  235. Susanna Elcuatro

    Thanks you for your good explanation in the video. It have been worrying about this manuscript for a long time and I think your studies are very good and explain it, but what about the other letters?

    • Stephen Bax

      Thanks Susanna – you are right – we still need to find out what the other signs represent, but that will still need more work – especially as we can’t guess without evidence. That is slow, I’m sorry to say, but I think it’s the only way to do it. Thanks for your comment.

  236. Henry Revell

    Wow, this is impressive. You’ve really cracked it, amazing: -)

  237. Simon Hughes

    Thanks very much for the video which makes it so clear. Brilliant!

Leave Your Comment

Your email will not be published or shared. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>