Latin in the Voynich Manuscript

by Justin E. H. Smith

I think I’m finally ready to come out as a Voynich scholar. I’ve been studying hi-res scans of the manuscript off and on for four years or so, and I’ve been reading the so-called secondary literature for about a year. What compels me to come out is the discovery over this past year that for the most part commentators really do not know what they are doing. They divide roughly into two camps: the cryptographers and information scientists, on the one hand (the “quants”, as we say), and on the other hand the ravers and enthusiasts, the people who do not know how to distinguish between gut feelings and real evidence. There seem to be very few proper palaeographers writing about this text: that is, people who know how to attend to handwriting and codicological evidence until plausible patterns of intention begin to emerge. It may be that such people are scared away by the ravers; one need only briefly glance at a list of all the time-travel, Illuminati, and UFOlogical theories the manuscript has inspired to see that it is a real intellectual danger zone. For me it is however a wonderful case study and autoexperiment in the use of abductive inference. I do not yet think I know anything with certainty that no other researcher has established before me. But over time a picture is emerging that leads me to lend significant credence to some explanations over others.

I am, say, 85-90% certain that the manuscript is not a hoax, or, if it is a hoax, it is one that was perpetuated long before the manuscript came into Wilfrid Voynich’s possession in 1912. I am attracted to the idea that, if it is a hoax, this was a hoax perpetuated on Athanasius Kircher, the one-time owner of the manuscript who was known to hastily claim to have cracked other codes (e.g., Egyptian hieroglyphics), and whom his contemporaries may have wanted to expose in his rashness by sending him a nonsense text to interpret. But this is a low-probability explanation. I am, say, 80-90% certain that the text was in the possession of a German, Dutch, or Flemish scholar who knew Latin, but that the manuscript itself is not written in any of these languages. As others have pointed out, the script of the main body of the text gives no indication of any regular repetitions of inflected word endings, of the sort that exist in all Indo-European and Finno-Ugric languages (e.g., the –s and –ed that are attached to plural nouns and passive participles respectively in English). It is plausible, based on quantitative analysis of the script, that it is written in a Sino-Tibetan or other East Asian language, perhaps by a European traveller (e.g., a German Jesuit such as Kircher himself) in Asia. But the lack of any distinctly Asian visual elements in the illustrations weakens this conjecture. The lack of any success so far in identifying any of the numerous botanical illustrations speaks strongly, in fact, in favour of the view that the work is a hoax, or at least a description of fantastical entities coming from nowhere in particular. The calendrical pages however strongly suggest a European provenance, and the illustrations of women bathing strongly suggest that the work is concerned with distinctly European traditions of balearic treatment of illnesses, which, in the early modern period, became a topic of scientific research for iatrochemists and other naturalists working at the intersection of medicine, chemistry, and natural history.

Almost all of the manuscript is written in an unknown script in an unknown language. There are however a few very small exceptions to this: the final page (f116v) features handwriting in Latin and what appears to be German, as well as two words in the “Voynichese” script. The calendrical pages (f70v-f73v) also feature the names of the months, from March to December, in some neo-Latin or Romance variant. The page f66r features, at the bottom, next to an illustration of a man (or perhaps a woman), what appear to be the German words, or partial words: der mus del. Thus:

I do not wish to concentrate on f66r today, other than to note that analysis of the handwriting makes it nearly certain that the same person who wrote these apparently German words is also the author of the Latin/German text on f116v, to which I now turn. The reader will have to look at the text in a hi-res version in order to discern some of the elements I will be discussing, but here is an image of the text sufficiently zoomed-out to see all of the elements on the page:

I am not prepared today to discuss the first line of text. The first word may be pox or vox; the second word may be leben or lebet. I have no conjectures, for now, for the next two words, and would rather move straightaway to the main body of the text. I am using blue to indicate words for which I am very confident in my transcription, green for words for which I am moderately confident, and red for words for which I am not at all confident. The repeated v‘s at the beginning of the third line indicate the two words that are written in Voynichese script, and that therefore have never been reliably transcribed by anyone:

I next list plausible alternative readings, and where appropriate provide further commentary.

anchiton = michiton. This is the most common reading, in fact, to the extent that some people have described this text as being written in “Michitonese”. But the initial part of the conjectured letter m looks much more like the other a‘s in the text (e.g., in ola) than like the other m‘s (e.g., in maria).

ola dabas = oladabas. There certainly does not seem to be any break between the a and the d, but separating the two has the advantage of giving us a meaningful Latin word (dabas = “you gave”). This in turn leaves us with the problem that ola has no meaning in Latin, and certainly is not in the feminine singular accusative, as one would expect if this ola, whatever it is, is the thing that “you gave”. So perhaps oladabas is a proper noun. It is possible, but not likely given the differences in word-length, that anchiton oladabas in line one corresponds to the Voynichese vvv vvvv in line three.

ceve = ??. Note here that the author began by crossing something out before beginning to write the word that remains. We can extract a Latin meaning out of this word, but only with difficulty: it could be the singular imperative form of a verb that means “to swivel” or “to walk lewdly or effeminately”.

fi[x] = si[x]. I’m reading the initial letter as an f only because it preserves the possibility of a concrete meaning in Latin, namely, the singular imperative of “to do” or “to make”.

ubren = obren, uhren. The advantage of my rendering, and of the first alternative, is that it preserves the possibility of referring to “the above”, which presumably in context would be the Latin text above, on which the author is now commenting in German. gas = gar. This latter has been the more common reading in the scholarship, and it is better at conveying a plausible meaning in German (“So take even me” or “So take me quite [fully or completely]”. But what I take to be the long s is simply too much unlike the other r‘s in the text (e.g., in ubren) to be plausibly read as one. Here moreover there are other, holistic considerations that militate in favour of gas, to which I now turn.

One thing to note, first of all, is the overall form of these three lines: the first two are in Latin, and the words are generally concatenated by a + sign; the last is in Voynichese and German, and seems to be an attempt at rendering a meaning in a full sentence of natural language. One thing that other scholars have not noticed is that in the second line in Latin, the x that follows the first four of the six words is not an element of the words themselves, but seems to serve an ordering or registering function in the same way the + sign does. Once we remove it, we have four perfectly meaningful Latin words; as long as it is still there, we have four Latin-seeming nonsense words.

So, as a rough stab at a translation, we have something like this:

+ anchiton ola [you] gave + many + you + * swivel + doors + n + make + maria + move + by force + alka[line] + ma+ria + vvv vvvv false above so gas take me.

Don’t ask me what this means; I’m not working at the level of comprehensive translation, yet. However, a few comments may help us to make sense out of it all. One is that the appearance of the word “false” (valsch, a Low German variant of falsch which strongly suggests a proximity to Flanders or Holland, rather than, say, a Central European origin) makes plausible the interpretation according to which this inscription is itself an effort to crack the code of the manuscript as a whole, and thus that it is superadded at a later date (like the German on f66v as well). If the manuscript does originate in the early 15th century, as carbon-dating indicates, then there could be no mention of “gas”, which is a neologism invented by Francis Mercury van Helmont (of Flemish origin) in his 1648 Ortus medicinae. There could however be mention of alkaline waters, which would have been an important part of balearic therapies and of treatises thereupon already in the Renaissance. So, I conjecture that the first two lines are someone’s notes towards a decipherment of the contents of the manuscript word for word, and the final line is some sort of comment on the meaning suggested by this concatenation of words. I suspect that whoever this was knew more about the contents of the work than we do today, and that he rightly understood the Voynichese text to be dealing with the subject of medicinal spa therapies. I suspect that this person was writing after 1648, with a now-expanded vocabulary for describing the chemical properties of bath waters and, now, “gases” as well. See here for example Nicolaus Steno’s 1660 treatise, composed in Amsterdam, Dissertatio physica de thermis [Physical Dissertation on Thermal Baths], in which the author develops his earlier notion of “chaos” (from which Van Helmont coined the term gas), understood as the maximum unstructuredness of particulate matter, to describe the effects of steam on the human body. I suspect that the author of the Latin/German text has read Van Helmont, and is writing broadly speaking in the intellectual context of northern European iatrochemistry.

Piece crossposted with Jehsmith.com