HOW MANY "HANDS" WROTE THE VM?
(A preliminary study to enable the further psychological profile of the VM author)
Jan. B. Hurych

Before we can undertake our task, i.e. the graphological analysis of the psychological profile of the author of the VM script, we have to first check the "2 or more hands versus 2 languages" theory by Captain Prescott H. Currier. The existence of more "hands" would have of course a serious impact on our study.

---

On 30 November, 1976, a one-day seminar entitled "New Research on the Voynich Manuscript" was held in Washington, DC Metropolitan area. Two detailed presentations by Captain Prescott H. Currier provided high points of that occasion. In them, and in the supporting paper printed as Appendix A of the Proceedings, he set forth his theory that there were several different scribes involved in the production of the Voynich Manuscript, and that their individuality was attested not only by characteristic "hands", reliably distinguishable by eye, but also by statistically distinct "languages".

I will start here with his document "Papers on the Voynich Manuscript" , privately circulated typewritten manuscript, dated 30 November 1976, Washington, D.C., as it was edited by M. E. D’Imperio, who served as moderator at the seminar. Jacques Guy and Jim Reeds transcribed Currier’s work into its present form in January 1992 and René Zandbergen presets it in HTML version on his site. It is basically Capt. Currier’s presentation "Some Important New Statistical Findings" and some of his answers during the Q&A period at the above seminar.

Not only Currier claims that several "hands" but also two "languages" appear in the VM. He says "The reason they are important . . . if the manuscript were to be considered a hoax ...it’s much more difficult to explain this if you consider that there was more than one individual involved". He also says that "These findings also make it seem much less likely that the manuscript itself is meaningless".

His first conclusion is of course easily refutable: many hoaxes and frauds were performed by a group of conspirators, not just by one individual. As for his term "language", it is rather vague as he himself admits and his proofs are mainly statistical comparisons of several parts of the VM "texts". The evaluations not being true linguistic, we can of course hardly talk about "languages", i.e. the VM still can be textually meaningless.

He claims he discovered that first 25 folios in herbal section are written in one "hand" (A) while the rest in herbal section (another 25 or so) is in "hand " B and also in different "language", by which he means they are statistically "distinct" , i.e. significantly different. Here is the summary for the whole VM as per Currier/d"Imperio record. the pages are numbered by him, but as René Zandbergen comments: "The page numbers appearing in Currier's Table are inconsistent with the Petersen page numbers". Also, assignment of folios to the pages may be somehow problematic.

- The Herbal Section (pp 1 to 112) see above
- The Mixed Section (pp 113 to 132) no info
- The Astrological Section ( pp 133-146) there seem to be no significant difference in the writing on any of the folios there. The "language" throughout is mostly A but without some of the more pronounced ‘A’ features found in Herbal 'A' "language".
- The Biological Section ( pp 147-166) appears to be the work of a single scribe, all in "language" B. This could conceivably be the result of this section being the product of only one person.
- The Pharmaceutical Section ( pp 167-211, pp 167-173 and two folios (pp 193-198) in the mid-portion of the section are in "language" B; the remaining folios are in "language" A. An interesting point here is the fact that there seemed to be more than the expected two "hands," one for each "language" as in the Herbal Section. In sum, he would venture a guess that there are at least three and perhaps as many as five or six different hands in evidence in this section. On the other hand it may all be an illusion (said by himself) .
- The Recipe Section ( pp 212-234) contains only one folio on which the writing differs noticeably from that on the other folios.The "language" throughout the Section is ‘modified B’ (i.e., con-contain certain ‘A’ characteristics).

Note: Capt. Currier later changed his estimated number of "hands". Also, some VM Internet pages provide slightly different or incomplete list of affected folios, but as I will demonstrate later, we will not need to worry by that problem.

---

So far Currier, but one preliminary observation is already ringing the bell of doubts: he mentions 2 up to 8 "hands", but he found only two different "languages" in the VM . First we drop the "modified hand B", which is apparently some compromise - we have to realize that, in graphological sense, we cannot have a "partial" mixture (it must be either the hand of one person or the other :-). After that, we still have, as stated by Zandbergen, hands 1, 2, 3, 4 and X (hands A and B were apparently renumbered to '1' and '2' to agree with numbering of those two "languages".

Currier bluntly refused to accept any other possibilities. Still, there are many reasons why single person's handwriting would change in time (age, sickness - physical or mental -, injury, influence of some medicinal drugs or even alcohol) and it may even change back and fore again (in the case of alcoholic intoxication). The probability of the change increases for long writings that took a lot of time, as it was with the VM. There are also mechanical reasons that distorts the handwriting: the sharpness of pen, thickness of ink and coarseness of the vellum surface, we have to consider all that as well. Of course, statistical changes in the text are real, but they may show only some coincidence - say the writer wrote some sections with different content when he was older and his handwriting slightly changed as well. The change of the content may also have some other, completely different reasons - say different sections with different subjects require different vocabulary and . Also, if text was coded, the second, different cipher can make the whole difference in statistics. Therefore the considering of connection "different hand = different brain" is especially illogical in the case the different scribes, since Currier claims that the VM was just copied. Why would different scribes need also different authors?

For above reasons, Capt. Currier's classification is rather unfortunate: instead of "hand" we should use more accurate word "handwriting" to account for the possibility that the "hand" could be the one of the author or the one of the scribe (or copyist), since we do not know the real case. Also, instead of "language" we should say rather "content" meaning of course the general variance in data and not any specific dialect (or even language). This would also cover for possible encoding. We kept here Currier's classification, but we will put the words "hand" and "language" always in quotation marks, to stress their true meaning is different.

Also, his claim that all five scribes would participate on one document does not explain why they all used only two "languages". That would indicate we should break the connection between "hands" and "languages" completely and treat them separately. Furthermore, provided that the distinction of "hands" was only visual, such observation had very little to do with the content itself. I took a look at the graphological proofs submitted by Currier and I actually found only the pictures shown at http://www.voynich.nu/writing.html, which I presume were provided by himself (or at least quote exactly the folios that show different "hands"). When one expert objected to Currier's classification, his opinion was discarded by Currier rather quickly: "It’s curious to me that a calligraphic or paleographic expert in one of the writings I have seen says that the writing is consistent throughout, and is obviously the work of one man. Well, it obviously isn't’t, and I don’t see how anyone who had any training could make any such statement, but there it is!" Provided that Currier meant by "writing" really the handwriting, we should be aware that Currier himself actually admitted he had no training in that area. It seems to me that his "hands" are mainly supported by the statistics he did of the text content of different folios. That was logical, since it admits he observed first the difference in writing and the statistics was done later, to prove they are different. Unfortunately, this kind of the "proofs" had nothing to do with handwritings. Note: He was also slightly inaccurate in his statement: the calligrapher is the expert in the field of different scripts, mostly artistic style, while paleographer is the expert in different historical scripts and neither is the expert in personal handwritings. The expert closest to this task would be a Forensic Analyst of Handwriting We will use here the term "graphology", well aware they are two kinds: the scientific category, i.e. one handwriting comparing the specifics of two handwritings and the other "graphology" , trying to create the writer's personal psychological profile. We will try to continue in that direction in the next article.

Of course, the "hands" should be always treated separately (especially since some may not belong to authors but only to the scribes). Besides, there should be enough of purely graphological proofs if Currier observations were correct. Handwriting itself should be really a clear giveaway, since the "hand" is greatly affected by the long time habits of the person that does actual writing. He cannot change them so easily and if faking another handwriting, pretty soon will unconsciously return to his own habits. On the other hand, the content is strongly affected by the author, i.e. his intentions, style, form and vocabulary (and the possible encoding as well). He can of course change all that at will if he wants to, without changing his handwriting.

---

It is also interesting to clarify here Capt. Currier's attitude by quoting his statements from the Q&A section of the above article:

Currier: "I can prove four ("hands", j.h.) beyond a shadow of a doubt. I’m not a paleographer; I wouldn't’t stand up in court and try to defend this against a paleographer. But I’m positive, particularly in the Herbal Section. I imagine it to have happened something like this: some sixty-five folios were prepared ahead of time with drawings on them. They were placed on a table so. The first twenty-five folios were taken, one at a time, off the top and filled in with writing by one individual. At the end of those twenty-five, he got very tired and he called for help. Another man sat down opposite him at the same table. And they took them off, one at a time: one man took one off and did his thing, in his own ‘‘language,’’ while the other man did his thing with another in his ‘‘language.’’ And they went through the second stack and interleaved them; one man did it one way and the other did it the other way. When they were done, they had the Herbal Section! "

Question: What about the fact that there were no erasures? That makes it look like a copying job.
Currier: It must be a copying job. But how do two people copying from a single source produce material in two different ‘‘languages’’ simultaneously? I can just see them sitting there! I’m absolutely positive this is the way it was done. The folios were prepared in advance by someone else with the drawings on them. Sometimes the writing overlaps the drawings somewhat. The pictures of the Herbal Section look as if they were drawn by a single individual, but this I couldn't’t prove. The writing on folios 1 to 25 was done by one man. On folios 25 to 65, it was done by two men, one who worked a little faster (the man who did the first batch did more of the second batch; he was more experienced).

Question: When a new hand takes over, do you see variations in the mode of writing the symbols?
Currier: Yes, but it’s the overall impression of the writing. In general, for example, in ‘‘Herbal A,’’ the writing is upright, rounded, lines are well-spaced, it looks clean, clear, with no extraneous material. ‘‘Herbal B,’’ in contrast, is uphill, slanted cramped writing. It’s obvious to me. The first thing I noted looking at the manuscript as a whole was this difference in the writing in the Herbal Section, before I had taken a single count. I separated the pages by sight first, then took a ten-page sample in each of the two separate writings, and made separate counts. It stared me in the face — there it was: all my selections were correct. It was a sufficiently controlled procedure to make me think these conclusions are valid. Anyone can see it — just lay the pages out and look. I can’t prove the pages are in the right order, but I just feel that they are. In the Astrological Section, the signs of the zodiac are in the right order.

Comment (j.h.): Currier must have been talking about some other singular folios, since in the samples on the page (the link above) it does not agree: as we can see below, it is the "hand" A that goes in his sample uphill, while "hand" B goes downhill and is not cramped. Those are the folios Currier meant , so to avoid confusion, we will use the very same samples for our investigation (samples marked by black line, They are the same as in the picture quoted, but little bit wider selection to be able to match the same words).

   

Another, additional research was done by Mary D'Imperio in "An Application of Cluster Analysis and Multiple Scaling to the Question of "Hands" and "Languages" in the Voynich Manuscript". She was only concerned with his statistics and used the cluster analysis . Her conclusion is that there are really two different "languages". To speak more accurately, there are only two different statistics, which is of course far from true language differences, that is the grammar and vocabulary evaluation. Also in her book "The Voynich Manuscript - an Elegant Enigma" there are only few sentences about it and again, no details about the "hands" are provided, she accepted Currier's theories without question. However, she mentions four more, unpublished articles by Currier which I had no opportunity to study. Interesting is her comment that Currier gave his computer experts the carefully prepared samples and wanted them to do only specific tasks. That was of course for content analysis only and I have a feeling that true graphological analysis is still long time overdue.

---

So here we have it: Currier used only his eye to distinguish those two or more handwritings. Let's see, what we can find if we use the professional criteria of the forensic analysis. Since his listing is using the questionable numbering by pages, i.e not the folios, we did not want to use folio nos, provided by Zandbergen since the cross-reference is still incomplete. So we did one step better, we used for our study the very same folios from the composite picture there. We consider that thee picture is listing the folios that Currier had in mind since they belong to he ranges he specified. Further more, the "sub-languages" vary very little so we will disregard them at the beginning and will do first the analysis only first two, that is the "hand A" and "hand B" because there is the difference most pronounced and visible. Of course we use the largest, high quality scans made by Beinecke (see below) since the difference between Currier's pictures could be caused by different contrast or exposure and we want to be sure what Currier actually saw in original folios. They first areas we studied are misplayed below and we can already see there more details and clearer pictures





Since the "hands" A and B are per Currier the most different, we will concentrate on those only. In our enlarged pictures, we can see what apparently raised Currier's curiosity: the script looks different, but we can already see the major reason: the pen was dull (or worn) and the ink on second sample is apparently different color (and consistency) , it soaked in and run more. Also, the second picture by Currier looks much darker than in Beinecke scan, which suggest the vellum had coarser surface. Those differences of course have nothing to do with particular "hand" so we will ignore them and concentrate on real handwriting analysis instead. Now let's first state the graphological criteria we used for our comparison:

1) Slant - is a very typical criterion for individuality, usually kept same for the whole text. However, it may change slightly if the basic line bends, so the angle should be measured from the baseline. For statistical purposes, higher number of measurements should be taken. There is slightly different angel for upstrokes and downstrokes, so only upstrokes are measured for each sample. We draw line through each letter and approximate the upstroke as a straight line. After measuring it for each of consecutive 100 letters, we register the angles in different groups (say 90, 80, 70 degs etc.). The group with highest count is the representative slant angle (sometime called graphotype).
2) Pressure - in our case the thickness of the line will be affected beside the hand pressure also by the sharpness of the pen, the consistency of the ink and the roughness of the vellum. It may indicate not only the stress of the person, but also the illness, injury, and not in the last instance the skill in writing or a "light" or "heavy" hand. The gradation is difficult to measure, but here we only compare the two different writings.
3) Neat versus sloppy - this is also difficult to measure, one can only roughly compare and judge. Also, the the health of the writer, the pen condition and the speed of writing may change the appearance of the writing very much.The grading of legibility often helps the evaluation. All in all, the personal traits will show here as well. For the same person, the neatness of his writing looks different if he copies something instead or writing down his own ideas. The writing of copyists are usually much neater, since there is a minimal mental involvement which may otherwise disturb the writer.
4) Rhythm - we judge if the writing "flows" easily or with interruptions, and there may be various reasons. Also, different persons react differently in otherwise similar conditions.
5) Margins and spaces - meaning spaces between letters, words and lines. It may show personal preferences or habit or it may be even given by available space on the page. Also, professional scribe has much more consistent spacing. Here we have an advantage of direct comparison of two writings.
6) Changes in handwriting - for the same person. Only rarely we can see those changes on a single page. We would have to compare many folios but that would be hindered by the fact that what we may consider the "changes" could be actually some other "hand". My decision therefore was leave this for the profiling in another article. Interestingly enough, the slant usually does not change with the persons's age.
7) Line droop - it can be caused by many factors, but it may be really specific to each person. It is a very good criterion but as we can see, when it happens very often and irregularly, sometimes up or down, we do not have a clear judgement. It is difficult to find what is specific droop for one particular person. Actually, the lines are seldom horizontal, and they swing very often according to space left after the pictures were drawn. Also, no helpful horizontal lines were drawn, as it was usually done in medieval manuscripts.

8) Graphozones - i.e. middle area of the letter, upper loops and lower loops. The dimensions may be effected when samples have different magnification ratio, but fortunately, the relative ratios of each character would not change. The ratios are very individual for each person, it is a very good criterion .
9) Stroke change - the angle may change towards the end of the word of a line. For writing left to right the slant may be progressively more perpendicular or otherwise. Here we could also spot the left-handedness, but in overall trend rather than in individual strokes. The backstrokes are also the individuality traits, as well as braking the character into single strokes, changes of their directions or a specific curvature, they all are important handwriting characteristics for a particular person.
10) Shakiness - may not necessarily come with old age, it could be an illness, both physical or mental, sometime it is however so small shakiness we can see it only when magnified or at close examination and therefore disregard.
11) Capital letters - for the VM, they are only four "higher" symbols, sometimes called "gallows" and they are generally considered as having some special function (which is not that of capital letters] . While we can use them for comparison, there would be some other criteria than those for lower case letters. Some researchers consider more complex lower case symbols as capital letters, not necessarily written with extra heights. Well, author did not bother with dots or commas, why would he bother with capital letters (especially if the text was coded)?
12) Upper loops - there are some specifics for the VM, but here we will do just simple comparison.
13) Lower loops - since there is no other example of the true VM script in existence, we do not know if the proper strokes are (say) round, elliptical, straight and what is the right connection between segments in the symbol, etc. Again, we may compare only the same symbols and search for their peculiarities.
14) Shading - it is believed that the text was written by quill that was cut the usual way (that is cut sharp and with vertical cut in the middle so the ink would run better. The shading can be done only on downstrokes or partial downstrokes. The thickness, start and termination of strokes are all good comparison criteria.
15) Individual characters - may vary from person to person - it is a very complex area and may be also sometimes very misleading. We did only some characters but the results were very convincing.
16) Covering of the page - crowdedness, fullness, also letter size changes, etc.
17) Others - color, ink, pen, etc.

We did the comparison of the whole folios marked on the sectors shown above) and the results are arranged in following table:

Criteria	A	B
Slant	right angle	close to right angle, within one hand range
Pressure	normal	higher, see comments below
Neatness.	neat	less neat but not drastically
Rhythm	normal	little faster
Margins and spaces	vary irregularly	vary irregularly
Changes	minor n the same folio	minor n the same folio
Line droop	variable, irregular	variable, irregular
Graphozones	same value	same value, very close
Stroke changes	minor	minor
Shakiness	no	very little
Capital letters	N/A	N/A
Upper loops	same kind	close
Lower loops	same kind	close
Shading	consistent	consistent
Characters	see below	see below
Others	N/A	N/A

Comments to this table:

1) Slant - the difference is very small, if any, it is not really the different angle to the baseline which also swings, but rather toward the horizontal line which is not proper evaluation. In general, both scripts are almost perpendicular (as was apparently designed by the inventor of the script). The line droop is generally different at the ends of the lines, but not consistently: sometimes line goes up, sometimes down. No helping lines were drawn by scribe.
2) Pressure - "Hand" B shows higher pressure on the pen, but it was needed in the case of the dull pen and rougher vellum. Also the density of the ink may play the role. As much as it is apparent, there is not enough indications to claim that higher pressure is related to the person only.
3) Neat versus sloppy - The sample B handwriting is not so neat as the sample A. But considering that sample A (folio 2) was at the very beginning of writing, the original care to put the text very neatly was probably lost when writing of the folio 26 :-).
4) Rhythm(or the text flow) - it looks the writing B is flowing more easy or without longer pausing which was apparently necessary for sample A since it was written with more care. It may also show the increasing skill in handling the originally not so familiar script and writing quicker.
5) Margins and spaces - In general the width of spaces vary substantially, but for both "hands" to the same degree, well within the same handwriting range. Margins are apparently different, since no measurements from folio to folio were done by author nor any template or guiding lines were used (as it was usually a habit already in medieval times.
6) Changes in handwriting format - within a line almost none, within the page, yes, but very small in both cases. The line droop is generally larger at the end of the line, but not consistent:by sometimes line goes up, sometimes down.
8) Graphozones - for middle area as well as upper loops and lower loops of characters we see pretty consistent ratios in both cases, see below.
9) Stroke change - if we measure angle only to the approximated base line (or rather curve) under each character separately, the changes are within the normal range, in both samples.
10) Shakiness - none for sample one, very little for sample 2, it may be again due to dull pen and coarse vellum or even something else. It is the shakiness of the base line which is more pronounce, letters do not seem to be shaky.
11) Capital letters - We cannot consider higher characters to be capital letters, there are only four. For high loops see below.
12) Upper loops - ratios basically consistent and similar in both cases
13) Lower loops - again, rations similar and as for shapes, see pictures below. This evaluation shows quite consistent brakes in the characters, their number and directions are same for both samples. This is the most important characteristic, being totally individual.
14) Shading - practically same, the shading is generally done by turning the pen downwards and applying more pressure. Of course most of it is enabled by the fact the quill had a slot in the middle, for better run of the ink. Also, the point was never really the point, but more like a very narrow chisel and hardness of both tips wears out with prolonged use.
15) Individual characters - this is what especially distinguishes one person from another, however we have here apparently a rather new script, probably not so automatic even for the writer. We cannot be sure for instance if the curves supposed to be circular or elliptical, neither we know what the true rations should be. Nevertheless, when comparing those two samples, both are pretty consistent in most of the characters. There are several copies on the Net by several researchers done by their own hands, but all are almost immediately recognizable as copies (mine included :-). Nothing like that can be told about original "hands" in the VM.
16) Covering of the page space - varies in both cases. The additional crowdedness happens also in both samples (actually it is worse in sample A:-), but is apparently random and mostly was even "forced" on the writer in order to squeeze the text there. For comparison, we have to almost ignore it, since it does not look as typical trait of the particular "hand". There we can also see for the "hand" A the text writing running "uphill" why on other samples of the same "hand" A it runs down. In many cases, we can see the writer tried to follow the picture line (again, the proof the picture was there first, even if there are several exceptions). One only has to wonder, why had to be text separated by pictures instead of being written under them - that would of course require more vellum.
17) Others (color, ink, pen, etc.) may be different for different folio, but they are not specific for different hand. During the writing, each writer had to change often not only vellum, but also the pen due to wearing out. As for ink, that maybe running out too, however not that often.

As we can see from the table, not a single factor show a clear difference in the "hands". The most critical for the judgement are of course individual characters ( symbols or letters, if you will). For that, we selected several areas within the samples that have similar "words", here we show them in corresponding colored frames.







The detail pictures within the frames are arranged in the matrix form, for better comparison. In general, we were looking for the specifics, e.g. dimensions and their ratios, segments in letters and their direction, etc. We marked here only some dimensions (in black) and some segments with their direction changes (in red). Of course the more detailed study was done on more samples, but for the illustration, we show here the typical ones.

Sample	"hand" A	"hand" B
red
green
violet

As the comparison shows, that ratios of the dimensions (in black) are very close for both "hands", well within the one hand range. Even if the magnification of each sample was originally maybe slightly different, the ratios (say upper part to lower part) will stay the same for each sample. Also, the typical segments (in red) are seen in both the "hand" A and "hand" B. As for thickness and coarseness of the letters, those are not the attributes of the handwriting only, but rather those of pen, ink and vellum (notice that for the sample B the vellum was of darker color and therefore coarser). While both "hands" looks visibly different, the idea that somebody would succeed in simulation of the first hand so closely as we can see in above table is rather remote.

Also, we have to stress here that the Beinecke magnification shows the letters much larger than they are in reality, so differences shown cannot be normally seen that well in the VM by naked eye only. Still, the differences are amazingly small, contrary to those Capt. Currier's say saw with his eyes. They still may exist, but we do not know exactly where to look for them.

CONCLUSION: We are convinced that the "hands" A and B are the same, as well as the other "sub-hands". The differences observed simply do not justify the claim they belong to different persons. I did not consider any further research in the Capt. Currier's controversial theory necessary since it would not have any special use for our purpose. The differences simply do not extent the range of variations for one person's handwriting. Apparently we have now to wait for those, who still believe in many "hands" to provide true graphological proofs to the contrary.

As for different "languages" or "content" or "style" for different folios, that was of course to be logically expected and applies even for one hand only since it was something which the writer had under his full control and that is completely different area of research.