Home > Features, Folios, Languages > Common Words in Language A that are Rare in Language B

Common Words in Language A that are Rare in Language B

The question was posed: which words are common in Language A but rare in Language B? And vice versa.

For this study I used the Herbal/Balneo folios that are Language A and B respectively (folios 1-25 and 75-84).

There are around 2900 unique words in total, with around 1600 being used in Language A, and 1630 in Language B.

Here are the results. The tables show the words in order of decreasing value of the frequency in A (B) divided by the frequency in B (A), and show the number of occurrences of each word in both Languages.

Common in A, rare in B

Common in A, rare in B

 

Common in B, rare in A

Common in B, rare in A

Conclusion? I have no idea … for now.

Advertisements
Categories: Features, Folios, Languages
  1. March 16, 2013 at 11:17 am

    The bottom-line figures about words here don’t tell the whole story of the individual shapes making up those words. For example, cth occurs in A at 4x the frequency of B, and cph occurs at 3.5x the rate in A than in B: whereas qo occurs in B at roughly 1.5x the frequency of B.

    Similarly, even though I’d always thought of ‘od’ as being very distinctive of A, it actually occurs in A at only about 2x the frequency it appears at in B, far less of a difference than for cth and cph.

    As far as your second list goes, I wonder whether the “lol” words were miscopied “qol”s… “lol” just looks wrong to my eyes, but maybe that’s just me. 🙂

    Incidentally, I accept there are quite big differences in the way e/ee/eee patterns appear in A and B, but I don’t see any mechanism by which e and y might practically be swapped in any sensible way. Just so you know! 🙂

  2. March 21, 2013 at 1:12 pm

    Hi Julian,

    Regarding all the ‘4o’ prefixes in B and not in A –

    A major difference between f.1-25 & f.74-84 is that of line length, 1-25 has on average a much shorter line with about 6.7 ‘words’ per line, whilst 74-84 has about 8.9 ‘words’ per line. More words – more spaces – also the ‘writing’ in 74-85 is smaller than 1-25, and whatever is underlying all this (i.e. ‘c’ & ‘i’ sets) were spaced further apart – I’m sure ‘4’ and probably ‘4o’ are just used as filler. This is also why ‘4’ is virtually absent from the ‘labels.

    Incidentally – back on 49v, look at the end word on the 6th line up from the bottom & the 4th word on the line below it (EVA 8airin & chariin Voy101 8aiyiN & 1ayiiN) the ascender on the EVA r is here just to break up what otherwise would be four ‘i’s in a row – the same happens whenever 4,5 or 6 c’s occur in a row, one or 2 of the ‘c’s has an ascender put on it. One of these occurs on the 4th line up from the bottom – there’s only about forty of these in the whole VM.

    Regards Tony

    • JB
      March 21, 2013 at 2:24 pm

      tony :

      Hi Julian,

      Regarding all the ’4o’ prefixes in B and not in A –

      A major difference between f.1-25 & f.74-84 is that of line length, 1-25 has on average a much shorter line with about 6.7 ‘words’ per line, whilst 74-84 has about 8.9 ‘words’ per line. More words – more spaces – also the ‘writing’ in 74-85 is smaller than 1-25, and whatever is underlying all this (i.e. ‘c’ & ‘i’ sets) were spaced further apart – I’m sure ‘4’ and probably ‘4o’ are just used as filler. This is also why ‘4’ is virtually absent from the ‘labels.

      Incidentally – back on 49v, look at the end word on the 6th line up from the bottom & the 4th word on the line below it (EVA 8airin & chariin Voy101 8aiyiN & 1ayiiN) the ascender on the EVA r is here just to break up what otherwise would be four ‘i’s in a row – the same happens whenever 4,5 or 6 c’s occur in a row, one or 2 of the ‘c’s has an ascender put on it. One of these occurs on the 4th line up from the bottom – there’s only about forty of these in the whole VM.

      Regards Tony

      Hi Tony. Dude, you keep pointing out features I’ve never noticed before! I do see the features you see on f49v. What is going on here? Why would a row iiii be so much worse than a row iii that it requires breaking up or modifying? Same for cccc. And why add the ascender to the second i to make 8aiyiN and to the first i to make 1ayiiN?

      The “lines as functional entities” fits with your 4/4o theory, but then wouldn’t we expect the “words” in A to match with the words in “B” if we simply remove all 4/4o prefixes? (And they don’t.)

      • March 21, 2013 at 2:42 pm

        Hi

        I presume iiiii and cccc etc were broken up because the author thought it would to too obvious for anyone to see how/what he did if he didn’t modify them!!!
        of course it didn’t matter whether the ascender went on 1st, 2nd, 3rd etc. just as long as he broke the sequence.

        I think ‘Words’ in A & B will match if we can get rid of all the disguises, ie. is ‘9’ just a ‘c’ with a tail that’s a disguise like the ascenders, is ‘8’ the same??

    • January 2, 2014 at 1:14 pm

      I know I’m late here, and most of these questions have probably already been asked and answered, but…

      Complex characters/ligatures like ‘cth’ and ‘cph’ are more common in A than B; A has shorter line length than B; and they were probably written by different people.

      Sounds like they could be scribal abbreviations. Are there any words where ‘cth’, ‘cph’, etc. in A match up with certain other characters or (more likely) character sequences in B? That is, words that are equivalent (barring other scribal idiosyncrasies that may exist) other than ‘cth’ in A corresponding to some other sequence in B.

      The disparity in initial qo- is interesting. What do B words with qo- correspond to in A? The frequencies make it look like B was a far more consistent speller than A (assuming the text originates with a natural language), so if, taking that into account, B qo- corresponds to, say, A o-, ‘q’ probably represents some sound like /h/ (or less likely, /w/ or /j/) which is prone to disappearance, or perhaps had already disappeared and was retained irregularly in spelling.

      Or, for that matter, what if the ‘words’ correspond somehow to syllables? It could be a featural writing system — like Hangul, but for a language that allows consonant clusters and/or diphthongs, making the words of variable length. That might account for the repetition, though I have no idea if the statistical properties match up.

  3. JB
    March 21, 2013 at 1:44 pm

    nickpelling :

    The bottom-line figures about words here don’t tell the whole story of the individual shapes making up those words. For example, cth occurs in A at 4x the frequency of B, and cph occurs at 3.5x the rate in A than in B: whereas qo occurs in B at roughly 1.5x the frequency of B.

    Similarly, even though I’d always thought of ‘od’ as being very distinctive of A, it actually occurs in A at only about 2x the frequency it appears at in B, far less of a difference than for cth and cph.

    As far as your second list goes, I wonder whether the “lol” words were miscopied “qol”s… “lol” just looks wrong to my eyes, but maybe that’s just me. :-)

    Incidentally, I accept there are quite big differences in the way e/ee/eee patterns appear in A and B, but I don’t see any mechanism by which e and y might practically be swapped in any sensible way. Just so you know! :-)

    “eoe” is quite common, and has a slight preference for placing itself at or near the ends or beginnings of lines. I’m not sure how they could be mis-copied “4oe”s, if that’s what you mean?

    The “e”/”y” mixing feature is there, and it may be just that – a swap in meaning. What do you mean by not seeing any mechanism that would swap them in a sensible way? Are you hinting that you know how these Languages work?!

  4. JB
    March 22, 2013 at 10:29 am

    tony :

    Hi

    I presume iiiii and cccc etc were broken up because the author thought it would to too obvious for anyone to see how/what he did if he didn’t modify them!!!
    of course it didn’t matter whether the ascender went on 1st, 2nd, 3rd etc. just as long as he broke the sequence.

    I think ‘Words’ in A & B will match if we can get rid of all the disguises, ie. is ‘9’ just a ‘c’ with a tail that’s a disguise like the ascenders, is ‘8’ the same??

    Yes, but you pointed out similar diacritics over some of the “4o” glyphs, which doesn’t fit with that theory does it? Isn’t it more logical to assume that the diacritics are modifying the value of the glyph, making it into a different glyph?

  5. March 22, 2013 at 1:16 pm

    Dude,

    Best guess is that the diacritics over 4o were a standard abbreviation of the time and are accidental occurrences by the author writing 4o, just as we might use a sign like a question mark and automatically put a dot beneath when we really didn’t mean to.

    “modifying the value of the glyph”. – ciphers just don’t come like that (especially not from around 1500 give or take a century or two) – you find them with marks to modify individual letter substitutes but not whole ‘words/glyphs’.

    Thinking about what you said “Why would a row iiii be so much worse than a row iii that it requires breaking up or modifying?” – the iiN looks like the way the letter m or in used to be written, so are quite acceptable, the same with not too many c’s they just look like old writing, (I’ve got a copy of a letter in German from 1640 next to me, apart from the letters that rise above and below the line everything looks the same – it’s impossible to read) – many have suggested ‘aiiin’ etc are numbers (if you turn the VM upside down and look at ‘aiiin’ in a mirror its almost a dead ringer for the way viiii was written!) – if they are numbers then this would be an excellent reason for the author to disguise 4 or 5 ‘i’s or’c’ in a row.

    It may possibly be just a simple substitution cipher using numbers filled with many nulls to disguise it as writing.

  6. March 24, 2013 at 7:25 am

    Julian
    8 and the gallows – if there are underlying sets of c’s and i’s beneath all the disguise then the lack of words like chedy in A may be explained by the following –
    on the times a single c occurs between 2 other groups of c’s (whatever length) it may end up disguised as below according to the ‘hands’ preference,

    A B
    ccc c ccc ccc c ccc
    ccc ckccc cccdc ccc
    for example –
    ccc ckhey chedy ccc

    A B as above
    ccc c ccc
    ccc s ccc

    The above is in EVA – Voyn101 use 8 for d, 9 for y, K for ckh, others the same.
    So perhaps there is no equivalent of the ‘word’ chedy in A – it’s all to do with spacing and where the ‘letters’ are inserted.
    I did a rough count on the first 1,000 words in both 1-25 and 75-84 – ckh and s are about twice as frequent in A but not enough to compensate for all the dy in B – so some other factor must be coming in to play as well.

    • JB
      March 28, 2013 at 5:35 pm

      Tony,
      The theory that the whole text starts out as little groups of “c”s and “i”s which are then somehow embellished to disguise, is appealing. But I can’t see how some of the statistical properties such as entropy and line positional frequencies would arise from that. If we had a candidate set of rules for the way it would be done, I suppose we could test it.

      • March 30, 2013 at 3:15 am

        Julian

        The VM’s entropy depends upon how it is transcribed – if it looks enough like writing (whatever way it was done), it’s going to have similar entropy to writing. Besides many of the ‘one of’ ‘words’ are just 2 or 3 ‘words’ where the space got filled in combining them together.

        Line positional frequencies – I suspect most of these are attributable to the authors method of drawing – reasoning that the inconsistency in the quire numbering is due to the author utilising unused quires from different ledgers etc. I think he may have been writing/drawing (some of the time) on already assembled quires – being unlike a single folio laid flat on the table, could not the hand coming towards the edge of the page, or even rotation of the page, account for the line positional frequencies.

        A candidate set of rules – I’ll email you a copy of my attempt at programming it.

        Tony

  7. March 24, 2013 at 7:31 am

    That didn’t come out well spaced – the B’s should be over where chedy is.

  8. Diane
    March 25, 2013 at 5:16 am

    Stylistically, f72v belongs with what you call the ‘Balneo’ section. Doesn’t it fit in terms of A/B split?

  9. Diane
    March 25, 2013 at 5:35 am

    I could probably have a fair guess about why these differences exist, and the type of language to expect in the bathy- section – but cipher people don’t listen to art people, do they?

  10. March 27, 2013 at 8:28 am

    A mon avis la transcription des lettres « 4o » devrai être révisée. Sur la page 5r, notamment, il y a au moins deux lettres différentes qui ont été transcrites comme « 4o ». Je crois que l’une d’elles est « d » et l’autre est, probablement, « q ».
    Ruby

    • JB
      March 28, 2013 at 5:27 pm

      Bonjour Ruby, j’ai bien compris votre comment, mais je prefere de repondre en Anglais, si vous permettez 🙂

      On f5r the Voyn_101 transcription has:

      [5r.1]h2o89.g1o9.1ho9.oam.oay.oes9.1o89.8h29.89-
      [5r.2]o1c9.ohc9.4aham.3o.Ho9.Kc9.1c9.oha™os.okoe-
      [5r.3]4oam.okaN.19.8am.okB.1oK9.ok19.4ok1o89-
      [5r.4]okan.2co89.1aN.s.1coy.1oK9=
      [5r.5]k29.2o89.4oam.1oeoes.2o.4ok1co.8Am.2o8am-
      [5r.6]2o.1coy,1c9.4oCc9.4o9hcC9.4o,coy.K9.2ok29.89-
      [5r.7]4okoC9.hcc9.1co.h19.2o89=

      Looking at the hi-res scan of that folio, the only “4o” I find questionable is the first on line 5r.6 … is that one of those you think is not “4o”?

      Julian

      • March 29, 2013 at 2:26 am

        I tried to express my ideas on my blog: I think that from the sixth row all occurrences of “4o” should be read as “d”. Of course, this is a proposal of the moment, it is not definitive, because it seems everything is perfectible.
        Ruby

  11. March 27, 2013 at 8:30 am

    In my opinion the transcription of letters “4o” have to be revised. On page 5r particular, there are at least two different letters were transcribed as “4o”. I think one of them is “d” and the other is probably “q”.
    Ruby

  12. April 21, 2013 at 11:09 pm

    Julian – Purely in terms of area covered, what percentage of the space would you estimate is filled by written text?

    • JB
      April 22, 2013 at 8:36 am

      I have no idea, and don’t see the relevance, I’m afraid.

      • April 22, 2013 at 8:44 am

        Because you have spent so much time looking at each folio, I thought you’d be the best person to ask for an estimate.
        PS
        Relevance to…?

  13. Marten
    May 3, 2013 at 1:53 pm

    A little off topic, But the Author/s of the manuscript. Must have made other works that had to exist or existing up today. Has there been any handwriting similarities or paintings that can be linked to an author? Cause the skill the person/s must have had, makes its kinda doubtful that this is the only material that exists.

  14. Diane
    May 6, 2013 at 11:21 pm

    Tony,
    Another reason why ‘4’ might be all but absent from the labels is that the labels record the names of things in their local language(s), that language lacking the relevant sound which is present in the language used for the commentary. Just a thought, from an amateur.

  15. May 9, 2013 at 6:05 am

    Marten,
    For the imagery there are comparative examples, though they do not appear to belong to the European or medieval Christian tradition.

    Since we don’t know at what stage the written text was enciphered (assuming it is enciphered), the encipherment may be an entirely separate issue and explain why there are no other examples – at least none that are known or recognised.

    Realisation that the manuscript may not be simply the invention of one European auteur has come slowly, and is still not general I think.

  16. May 28, 2013 at 3:57 am

    I’m relieved that the differences are so clear between the vocab. of the herbal and bathy- sections. If they weren’t it would be more of a worry, wouldn’t it, since the subject of one appears so different from the other. If all created at once, but one author, and using randomly generated text, you’d expect more uniformity, wouldn’t you?

  17. July 31, 2013 at 12:06 pm

    Julian.
    It’s very quite in here – have you given it up?
    Tony

    • JB
      August 2, 2013 at 11:19 am

      Hi Tony,

      I’m working (on and off) on a numbers-based idea, but have nothing yet to report.

      I don’t expect to ever give up – it’s too compelling a challenge 🙂

      Julian

      • August 2, 2013 at 3:49 pm

        That’s good to hear.

  18. August 2, 2013 at 3:51 pm

    Hello,

    I wonder if there is an easy way to generate a list of the fifty top words in each hand that are uncommon in the other hand similar to the one posted on March 15th, 2013? I need a larger sample than 10 of each to test my idea. Or could someone tell if there is already such a list on line somewhere?

    I don’t want to put anyone to a lot of trouble, just hoped it might be easily computer generated or already had been.

    Thank you.

    Don of Tallahassee

    • JB
      August 2, 2013 at 4:00 pm

      Sure, Don: here are the lists of top 50 in each.


      Words Common in A but rare in B
      ok1oe ratio 40.0253821748 numA 24 numB 1
      Hoe ratio 28.3513123738 numA 17 numB 1
      Koy ratio 23.348139602 numA 42 numB 3
      8aN ratio 23.348139602 numA 14 numB 1
      Koe ratio 20.4296221517 numA 49 numB 4
      81oy ratio 18.3449668301 numA 22 numB 2
      9koe ratio 18.3449668301 numA 11 numB 1
      91oy ratio 18.3449668301 numA 11 numB 1
      h19 ratio 17.5111047015 numA 21 numB 2
      82oy ratio 16.6772425728 numA 10 numB 1
      9k1oy ratio 16.6772425728 numA 10 numB 1
      K9 ratio 16.4919398776 numA 89 numB 9
      h1oy ratio 15.0095183155 numA 18 numB 2
      h1oe ratio 15.0095183155 numA 18 numB 2
      9hoe ratio 15.0095183155 numA 9 numB 1
      91oe ratio 15.0095183155 numA 9 numB 1
      Hc9 ratio 13.3417940583 numA 24 numB 3
      ok1oy ratio 13.3417940583 numA 16 numB 2
      oe8aiiN ratio 13.3417940583 numA 8 numB 1
      Ko89 ratio 12.5079319296 numA 15 numB 2
      829 ratio 11.674069801 numA 7 numB 1
      4o8ay ratio 11.674069801 numA 7 numB 1
      9h19 ratio 11.674069801 numA 14 numB 2
      4oHoe ratio 11.674069801 numA 7 numB 1
      2caiiN ratio 11.674069801 numA 7 numB 1
      k1oy ratio 11.674069801 numA 14 numB 2
      sos ratio 11.674069801 numA 7 numB 1
      8op ratio 10.0063455437 numA 6 numB 1
      Kae ratio 10.0063455437 numA 6 numB 1
      ohoiiN ratio 10.0063455437 numA 6 numB 1
      1h19 ratio 10.0063455437 numA 6 numB 1
      hcoy ratio 10.0063455437 numA 6 numB 1
      19h9 ratio 10.0063455437 numA 6 numB 1
      9k19 ratio 10.0063455437 numA 12 numB 2
      1oy ratio 9.77631461166 numA 170 numB 29
      ? ratio 9.39990035923 numA 62 numB 11
      Kay ratio 8.89452937218 numA 16 numB 3
      1o8 ratio 8.33862128641 numA 5 numB 1
      s1c9 ratio 8.33862128641 numA 5 numB 1
      8? ratio 8.33862128641 numA 5 numB 1
      k2oe ratio 8.33862128641 numA 5 numB 1
      g ratio 8.33862128641 numA 5 numB 1
      4o8ae ratio 8.33862128641 numA 5 numB 1
      ck9 ratio 8.33862128641 numA 5 numB 1
      k1oe ratio 8.33862128641 numA 10 numB 2
      ok1o ratio 8.33862128641 numA 5 numB 1
      h1coe ratio 8.33862128641 numA 5 numB 1
      o8oy ratio 8.33862128641 numA 5 numB 1
      GaiiN ratio 8.33862128641 numA 5 numB 1
      2ok9 ratio 8.33862128641 numA 5 numB 1

      Words Common in B but rare in A
      2c89 ratio 232.052749914 numB 387 numA 1
      4ohcc89 ratio 169.692320996 numB 283 numA 1
      4ohc89 ratio 151.703735732 numB 253 numA 1
      1c89 ratio 93.3407702064 numB 467 numA 3
      4ohaiN ratio 80.0492044275 numB 267 numA 2
      okcc89 ratio 59.3623313732 numB 99 numA 1
      ohcc89 ratio 27.2826876513 numB 91 numA 2
      okaiN ratio 24.2845901072 numB 81 numA 2
      2cc89 ratio 22.1859218264 numB 74 numA 2
      ehcc9 ratio 20.9866828087 numB 35 numA 1
      kc89 ratio 20.3870632999 numB 34 numA 1
      hc89 ratio 20.3870632999 numB 34 numA 1
      haiN ratio 20.3870632999 numB 34 numA 1
      eoe ratio 19.7874437911 numB 33 numA 1
      oehcc9 ratio 19.1878242823 numB 32 numA 1
      82c89 ratio 17.3889657558 numB 29 numA 1
      1c8aiiN ratio 17.3889657558 numB 29 numA 1
      1cK9 ratio 17.3889657558 numB 29 numA 1
      4oe ratio 16.6694223452 numB 139 numA 5
      ae9 ratio 15.5901072293 numB 26 numA 1

      • August 2, 2013 at 6:55 pm

        Thank you very much. I don’t think the second list of 50 completely made it all the way through the internet tubes to me.

        That’s okay – what I have now is enough to work with.

        Thanks again.

        Don of Tallahassee

  19. August 2, 2013 at 8:35 pm

    Can you please tell me where I can find a copy of the code (I assume it is Capatalized EVA???) used in the above lists? If it is Capatalized EVA, it doesn’t seem to be available on the internet. I don’t have a copy of whatever it is.

    Thank you.

    Don of Tallahassee

  20. August 2, 2013 at 9:10 pm

    Even without being able to read the extended list you sent, I can tell the numbers are completely different. Why is this?

    Thank you.

    Don of Tallahassee

    • JB
      August 2, 2013 at 11:04 pm

      donoftallahassee :

      Thank you very much. I don’t think the second list of 50 completely made it all the way through the internet tubes to me.

      That’s okay – what I have now is enough to work with.

      Thanks again.

      Don of Tallahassee

      The transcription is Voynich101 by Glen Caston.
      http://www.deepsky.com/~merovech/voynich/voyn_101.txt

      The numbers I gave are for the whole manuscript A and B, rather than for just herbal/Balneo.

  21. October 10, 2013 at 7:29 am

    I would like to contact you. Do you have email, please ? Thank you.

    • JB
      October 10, 2013 at 8:27 am

      Sure … jjbunn at gmail dot com 🙂

  22. October 10, 2013 at 7:45 am

    https://picasaweb.google.com/104064582393233194631/HerbierPseudoApulee?authuser=0&authkey=Gv1sRgCIeJt9qIhbDH1wE&feat=directlink

    I found these fragments in the book : Pseudo-Apuleius, De medicaminibus herbarum liber, siue Herbarius (Pseudo-Apulée, Herbier)
    http://gallica.bnf.fr/ark:/12148/btv1b84262821

    • JB
      October 10, 2013 at 9:36 am

      Nice! Good find 🙂 The glyphs are quite similar to those in the VMS.

  23. October 10, 2013 at 11:05 am

    Yes ! 🙂

    – 37v
    http://gallica.bnf.fr/ark:/12148/btv1b84262821/f78.image
    – 42v
    http://gallica.bnf.fr/ark:/12148/btv1b84262821/f88.image
    – 43r
    http://gallica.bnf.fr/ark:/12148/btv1b84262821/f89.image
    – page de garde recto.
    http://gallica.bnf.fr/ark:/12148/btv1b84262821/f135

    Based on the analysis of the BNF that accompanies the book, many people have annotated the book. Sorry for my english ! 🙂

    ———————————————————————————
    “Titre : Anonyme, De ponderibus medicinalibus (3r). Pseudo-Hippocrates latinus, Epistola ad Maecenatem (3v-5v). Antonius Musa, Epistola de herba vetonica (19-22v). Pseudo-Apuleius, Herbarius (sive Liber de nominibus et virtutibus herbarum) et Pseudo-Dioscorides, De herbis femininis (22v-63v).
    Auteur : Pseudo-Hippocrates. Auteur du texte
    Auteur : Antonius Musa. Auteur du texte
    Auteur : Pseudo-Apuleius. Auteur du texte
    Auteur : Pseudo-Dioscorides. Auteur du texte
    Date d’édition : 800-900
    Type : manuscrit
    Langue : Latin
    Format : Laon ou Reims. – Minuscule caroline de module moyen, une seule main pour l’ensemble, mais dont le niveau d’usure, plus important pour les f. 3-18v que celui des f. 19r-63v, laisse penser qu’un certain laps de temps a pu s’écouler entre ces deux parties ; sous-titres en onciales à l’encre brune ; titres en capitales rubriquées. Les marges sont emplies d’additions anonymes dont la presque totalité est imputable à une main qui semble être du X e s. (début?). Cette main du glossateur principal est responsable d’observations variées, recettes, prescriptions et même d’incantations (voir Joret, 1888, p. 340). Les additions sont la plupart du temps sans rapport direct avec le texte, mais leur contenu constitue des suppléments d’un intérêt certain (aucune source repérée) ; par ex. au f. 18v, à la fin de la seconde table où se trouve la peinture d’Esculape découvrant la bétoine se trouvent deux recettes contre la pleurésie (onguent et cataplasme). De plus, cinq autres mains différentes sont intervenues, dont au moins une est de peu postérieure au travail du copiste. – Peinture d’Esculape découvrant la bétoine (f.18v) : «Scolapius qui vetonicam invenit», légende reprise en marge par le glossateur principal «Esculapius invenit vetonicam» (voir Hubert – Porcher – Volbach, 1968, p. 112 fig. 98). Le frg. de l’Herbier du Pseudo-Apulée (f. 22v-63v) renferme 75 peintures en couleur représentant des plantes ( herbae) et quelques animaux : un scorpion peint (f. 23) et un autre au trait (f. 63v); un aigle (f. 44v), ainsi qu’un volatile non identifié (corbeau?) (f. 46v) ; de nombreux serpents parsèment les marges, certains peints, d’autres au trait (f. 23r, 26r, 33r, 34v, 37v, 41r, 47r et 63v) ; à cette faune s’ajoutent l’illustration en couleur d’un canidé, qui représenterait un «chien enragé» selon Wickersheimer, 1966, p. 68 (f. 22r) et quelques abeilles accompagnant la représentation de l’acore ( achorum, f. 28v); Singer, 1927, p. 35-36 (avec fac-similé des f. 23r et 30v) propose de voir dans ces peintures des influences anglo-saxonnes et italiennes méridionales. – Parch. – 63 f. + I-II à longues lignes. – 285 x 200 mm (just. 140 x 200 mm). – Les f. 1-2 sont des gardes de parch. récents (comme I et II) qui ont été foliotées ; le ms. proprement dit commence au f. 3. ; pour les 61 f. restants du ms., l’ordre de lecture restitué est le suivant: f. 3-8, 10-15, 9, 16-18, 20, 19, 21-22, 24, 23, 25-63 ; ms. mutilé de la fin, avec lacunes entre f. 30-31 et 55-56. – Reliure de maroquin rouge aux armes de Colbert ; titre au dos : « liber botanicus ». – Estampille de la Bibliothèque royale (Ancien régime, avant 1735), modèle identique à Josserand-Bruno, p. 268, type B n° 5
    Droits : domaine public
    Identifiant : ark:/12148/btv1b84262821
    Source : Bibliothèque nationale de France, Département des Manuscrits, Latin 6862
    Description : f. 3. Titre ajouté au XIII e s. : « Liber herbarum medicinalium ». f. 3r. De ponderibus medicinalibus. Frontispice orné d’un cercle inscrit dans un cadre carré, avec décor végétal dans les écoinçons sur fond vermillon ; cadre et cercle de couleur vert / bleu. Il contient au centre un petit texte de métrologie contemporain de la réalisation du ms. Le texte en a été réencré au XIII e s. (de la main qui a ajouté le titre): « De ponderibus. Oportet pondera medicinalia nosse (…) Obolus siliquas tres » (éd. Wickersheimer, 1966,n ° LXI.1, p. 67) ; dans le bas du cercle, une marque de possesseur a été grattée. f. 3r-23v. Medicinalia varia. Additions marginales dans lesquelles se trouvent de nombreuses recettes et diverses observations sur les vertus des simples (voir Wickersheimer, 1966,n ° LXI.5, p. 69-70); saec. X (?), selon Beccaria, 1956, les ajouts en marges seraient contemporains : f. 3r « …]nantica et quinantica (…) ad condolomata » ; f. 3v-5r « de pentafilon (…), de petrosilino (…), de piper (…), de petrosilino (…), de costo (…), de peretro » ; f. 4v, le nombre des os du corps humain ( Orus : habet homo ossa CCXXVIIII) et le nombre de maladies, dont celles propres aux femmes ( passiones in omines (sic) sunt II M CCCCLXXXVI ; sunt passiones in mulieres in matrice CCCLXXII ) ; Des recettes pour mettre en fuite les serpents (f. 4r et 8v), mais sans efficacité contre la salamandre (f. 15v, de serpente qui vocatur salamandra …) ; la nature des humeurs et un charmes contre les blessures (éd. Wickersheimer, 1966,n ° LXI.5, p. 70) (f. 12v) ; etc. Il existe une copie de ces annotations (f. 3-18) ayant appartenu à Charles Daremberg (1817-1872), érudit et médecin, dans Paris, Bibliothèque de l’Académie Nationale de Médecine, Collection Daremberg, 420 (1300), de 51 feuillets contenant Hippocratis ad Maecenatem epistola (avec collation sur le ms. CDXXV de Turin) et les Remedia varia. f. 3v-5v. PSEUDO-HIPPOCRATES latinus, Epistola ad Maecenatem ( scil. ad Antiochum), avec lacune de la fin (perte d’un bifolium) : « Incipit epistola medicinalis quatuor temporum, quid in his utendum vel quid vitandum sit. Ippogrates (sic) Mecenati suo salutem. Libellum quam (sic) roganti tibi promisi (…) quibus rebus uti debeat aut abstineri [… » (éd. Corpus medicorum latinorum…, 5, 1968, p. 26-32). Il s’agit d’« une des lettres apocryphes d’Hippocrate précédant le De medicamentis de Marcellus Empiricus » (Wickersheimer, 1966, LXI.2, p. 67) ; cette traduction anonyme en latin serait celle de la lettre à Antiochus de Dioclès de Caryste (s. IV a.c.), selon Opsomer et Halleux, 1985 ; mais aucune des attributions proposées (Antonius Musa, Marcellus Empiricus ou Dioclès) ne fait l’unanimité, voir Kibre, 1979, p. 280-281, col. 1; chap.XX. Epistola n° 4 Ad Maecenatem ; Clavis patrist. pseudepigr. 3A, 2003, n° 951/9a ; voir la Bibliographie des textes médicaux latins. Antiquité et haut Moyen Âge, G. Sabbah, P.-P. Corsetti et K.-D. Fischer [dir.], Saint-Étienne, 1987, p. 136 et n° 335-340). f. 6r-18v. Deux tables : la première, f. 6r-8v, donne une liste des capitula ainsi que le nom et le nombre de vertus de chaque plantes ; la seconde, f. 9r-18v, est une table méthodique qui récapitule les utilisations des simples et indique les n° de capitula où sont mentionnés les maux traités. 1. (f. 6r-8v). Première table. Index des plantes avec n° de chapitre et nombre de vertus (voir Collins, 2000, p. 184) ; les n° de capitula sont rubriqués. En marge de petits ‘d’ onciaux signalent les chapitres en provenance de Dioscoride ; acéphale, le récapitulatif débute au chap. XXII et s’achève au n° CLXVIII : «D.» (in marg.), « XXII. Herba camedris habet uirtutes III ; apud grecos X (…) » (voir f. 40v) ; au f. 8r, « (…) CXXVI. Herba mandragora, masc(ulus), virtutes VIII. usque huc cap(itula) Dioscor(idis) intermixta XXIIII ; dehinc secuntur ordine usque . Herba mandragora, femina (…) » ; f. 8v, « CLXVIII. Herba aparinae, virtutes II. Expliciunt capitula libri medicinalis Ippogratis ( sic), Platonis, Dioscoridis ( p. corr.) et Scolapii de herba[m] vetonica[m] qui ipsam invenerunt. Ex Libro Buta(nico) ( sic) ». 2. (f. 9r-18v) Seconde table. Table méthodique, agencée selon des thèmes [ordre de lecture restitué : f. 10r-15v ; 9r-9v ; 16r-18v]. Il s’agit d’une concordance des maux et maladies avec renvois aux chapitres traitant de la question (avec nombreux ajouts et corrections sur grattages contemporains ou de peu postérieurs) : « Medicamenta huius libri butanici (sic) his capitulis requirendas. Primum autem a capite ostendunt originem deinde secuntur ad cetera (…) » f. 12v, « Actenus causa capitis, sequunturque necessaria (…) » ; f. 13v, « hucusque pectoris morbus ; dehinc sequitur uenter cum subdita (…)» ; f. 14v, « tria originalia preterita ; quartam uisiccam (sic) indigent cum sua (…) » ; f. 15v, « Actenus causas proprie capitis, pectoris, ventris et vissicae. Dehinc secuntur cetera intermixta. De mulieribus (…) » ; f. 9v, « De intermixtis passionibus uel medicinis ( p. corr.) illarum(…) » ; f. 16v, « de febribus (…) » ; f. 17r, « De vulneribus et cetera alia adiuncta (…) » ; f. 18r, « De maleficiis serpentium et ceteris creaturis venenatis et de veneno sumtum ( sic) ». Des espaces ont été laissés à l’origine entre les vedettes pour les enrichissements postérieurs qui n’ont pas manqué (cf. par ex. f. 11). f. 20r-22v. ANTONIUS MUSA, Epistola de herba vetonica [ordre de lecture restitué : 20r-20v ; 19r-19v, 21r-22v]: « Incipit epistula Antoni Muse de herba vetonica quantas virtutes habet. Antonius Musa M. Agrippe salutem. Augusto praestantissimo omnium mortalium (…) Cum volueris uti, sic uteris. Explicit epistola de herba uetonica ». (L’illustration de la bétoine occupe le reste du f. 20v) ; f. 19r-v. « I. Herba vetonica. Omoeos : cestros. Alii : cyroe (…) et in ore detineat dolorem discutit », (la fin est copié par la même main, mais sur grattage des 7 dernières lignes) ; « Item herba uetonica (…) ieiunus edat » ; f. 21r : « ad facioneris dolorem (…) trita et inpositra mire dolores linire experti sumus [… » (éd. Howald – Sigerist, Corpus medicorum latinorum… 3, 1927, p. 3-11). f. 22v-63v. PSEUDO-APULEIUS, Herbarius (sive liber de nominibus et virtutibus herbarum), avec Epist et interpolations en provenance du PSEUDO-DIOSCORIDES, De herbis femininis [ordre de lecture restitué : 22v, 24r-v, 23r-v, 25r-63v ; lacunes entre f. 30-31 et 55-56 ; manque la fin]: « Vires herbarum et herbas. Incipiamus alium herbarium Apulei Platonis. Apuleius Platonicus (p. corr.) ad cives suos. Ex pluribus paucas vires herbarum et curationes corporis (…) invitis etiam medicis profuisse videatur (…) alii paeonia vocantur [… » ( Herbarius, ms. de classe β : éd E. Howald et H. E. Sigerist, 1927, Corpus medicorum latinorum… 4, Leipzig ; De herbis femininis : éd. Kästner, Hermes 31 (1896), p. 578-636, qui a utilisé ce ms.) Chaque herba occupe un chapitre numéroté de II à LXI, le De herba vetonica d’Antonius Musa tient lieu de chap. I. Sur l’ensemble décrit dans la table initiale, la présence de soixante-et-une rubriques conservées avec lacunes (XLVI herba grias et XLVII herba astula regia [index, f. 6v] + lacune du début de la rubrique traitant du Lapatium) permet d’estimer la perte à plus de 110 f. Les extraits du ps.-Dioscoride qui ont été ajoutés à l’ Herbarius (ci-dessous indiqués entre crochets droits, avec les références aux pages de l’éd. Kästner) sont signalés par des sous-titres en onciales dans les chapitres concernés ; ils ont été systématiquement revus par le premier glossateur peu après la copie, probablement dans le même scriptorium (écriture proche de celle du copiste; il se pourrait que ce soit la même main), mais avant qu’intervienne le glossateur principal (qui travailla dans un autre centre d’étude?). Il ressort de ces révisions que toutes les mentions de Dioscoride dans les sous-titres sont de la main du premier glossateur qui a, soit gratté et réécrit une partie de l’inscription primitive, soit l’a simplement complétée. Ainsi, il a remplacé en plusieurs endroits la locution « secundum grecos » par « secundum Dioscorides » ( sic, ut semper), sans se soucier de l’accord grammatical et auquel il a parfois ajouté le nom de l’oeuvre ( De herbis femininis), voir f. 33v, 37r, etc. Parfois une seule illustration s’applique aux deux sources (ps.-Apulée et ps.-Dioscoride); d’autres fois, l’article a donné lieu à plusieurs illustrations (voir Wickersheimer, 1966, p. 68 ; Catalogus Translationum … IV, 1980, p. 131 ; Munk Olsen, T. 1, 1982, p. 30, n° B.90). Plus qu’une simple version interpolée, il s’agit véritablement d’un essai intelligent de synthèse entre l’ Herbarius et le De herbis femininis (voir Collins, 2000, p. 186) : f. 24r-23v [Plantin]. « II. Herba plantagine (…) ex parte figura altera plantagine auctoritatem grecorum (…) apud ipsos arnoglossa. Continet autem virtutes XXXII » (inscription en partie effacée) [p. 624-626, arnoglossa] (fig. 1, f. 24r ; fig. 2, f. 23v) ; f. 26v-27r [Verveine]. « IIII. Nomen: Herba Verminatia » [p. 629, hiera] ; (…) f. 30r [Armoise]. « X. Nomen: Herba Artemisia monoglonosus (…) secunda Artemisia Tagantes dicitur grece (…) » (fig. 1, f. 30 ; fig. 2, f. 30v) ; f. 31r-v [Rumex]. « [lacune du début, perte de la première variété du Lapatium] ; item alteram figuram secundum Dioscorides de herba lapatium capitulum XLVIII ; virtutes XVII (…) Item lapatium genus tertii, nomen: oxy (…) item lapatium genus quarti ; deest figura cum virtutibus suis » [p. 623] (fig. 1, lacunaire ; fig. 2, f. 31r ; fig. 3, f. 31v ; fig. 4, f. 32r, remplacée par anticipation par l’ herba dracontea, qui se trouve illustrée de nouveau au f. 32v) ; f. 33r-v [Serpentaire]. « XII. Herba dracontea (…) Item alia figura secundum Dioscorides ( sic), tit. XL » [p. 619-620] (fig. 1, f. 32v et cf. 32 ; fig. 2, f. 33r) ; f. 36v-37r [Renouée]. « XVI. Herba proserpinatia (…) Item altera figura in libro Dioscorides de Herbis femininis cap. VIIII et cum ipsas ( p. corr. in margine) de herba proserpinatia quam ipsi poligonos vocant » [p. 595-596, poligonos] (fig. 1, f. 36r ; fig. 2, f. 36v) ; f. 37v-38r [Aristoloche]. « XVII. Herba aristolocia (…) Item altera figura secundum Dioscorides, cap. XII de herba aristolocia » [p. 597-598] (fig. 1, f. 37r ; fig. 2, f. 37v) ; f. 40v-41r [Germandrée]. « XXII. Herba camedris (…) Item de herba camedris in libro Dioscorides, tit. VIII » [p. 595] (une seule fig., f. 40v) ; f. 41r-42v [Cardère]. « XXIII. Herba camellea (…) Item altera figura secundum Dioscorides, tit. XXXVIIII in libro femininarum herbarum ; virtutes V » [p. 615] (fig. 1, f. 41v ; fig. 2, f. 42r) ; f. 43r-44r [Parelle]. « XXVI. Herba brittanica (…) Item secundum Dioscorid(en), de herba brittanica in libro femininarum herbarum, tit. XXIIII » [p. 605] (une seule fig., f. 43v) ; f. 49v-50r [Prêle]. « XXXV. Herba ippirum (…) Item de herba ippirum secundum Dioscorid(en), tit. XXXII, alii anabasia, a quibusdam anabasis nuncupatur » [p. 610-611, hyppiris] (les deux figures sont sur le même registre, f. 49v, avec comme légende pour la première mascul(us) et pour la seconde femina, Dioscorides) ; f. 51r. Un espace laissé vacant pour une illustration qui a été oubliée (?) ; f. 55v [Coloquinte]. « XLV. Herba colocintios agria ; Dioscorides, titulus XLVII (ajouté dans la marge sup.) » [p. 621-622] (une seule fig., f. 55v) ; [Lacune des cap. XLVI et XLVII (respectivement, herba grias et herba astula regia de l’index, f. 6v)] f. 56r-57r [Pavot]. « XLVIII. Herba papaveris silvatici (…) Item altera figura ex parte de herba moecon, latine papaver agreste nuncupant. Dioscorides, tit. XLVI » [p. 620-621] (fig. 1, f. 56r ; fig. 2, f. 56v) ; f. 57r-58r [Narcisse]. « XLVIIII. Herba narcissus (…) Item altera figura de Narcisso quam Aegyptii Bulbo semeticon et alii Bulbo rufum vocant. Figura ex parte altera et stomcho convenit. Dioscorides, titulus XLIIII » [p. 618-619, bullus rufus] (fig. 1, f. 57v ; fig. 2, f. 58r) ; f. 59v [Cétérach]. « LII. Herba splenion (…) Item de herba splenios siue colopendarios. Dioscorides tit. XLI » [p. 616] (fig. 1, f. 59r ; fig. 2, 59v en marge) ; f. 60v [Consoude]. « LV. Herba confirmam », où a été ajouté autour de l’illustration par le premier glossateur : « Haec herba duplicem habet figuram, istam presentem confirmam et alteram qui vocatur simfitum album siue confirmam ; figura illius require capitulo CXXI (…) ».
    Description : Peut-être copié dans les alentours de Reims (Hubert – Porcher – Volbach, 1968, p. 112), à Hautvillers, vers le milieu du IX e s. (cf. Barberi, 2000, p. 144) ou à Laon (exposition BnF, 1982, n° 57 et selon Contreni, 1990), sous l’épiscopat de Pardule (845-856). Le style pictural n’est pas sans rappeler celui d’autres réalisations, dont le Physiologus de Berne (Bern, Burgerbibliothek, Codex Bongarsianus 318 ; voir Homburger, 1962, p. 101-117), auquel a été assignée une même origine rémoise, quoique aussi incertaine. Toutefois, les critères retenus (stylistiques et paléographiques) pour attribuer ce ms. à un scriptorium rémois demeurent peu sûrs. Il est probable qu’il ait effectué un long séjour à Chartres en raison du titre ajouté au XIII e s. (f. 1r) qui se retrouve textuellement dans l’inventaire de la bibliothèque de St-Pierre de Chartres rédigé en 1367 (Omont, 1890, p. XXXI, n° 102). Bien que le ms. ne soit pas mentionné dans le catalogue du XI e s. de cette même institution (voir Omont, 1890, p. XXI-XXIV), l’hypothèse a été émise qu’il puisse y avoir été apporté dès la fin du X e s. « sans doute par les soins de l’évêque médecin Fulbert » (selon la notice de l’exposition BNF, 1982, p. 71 ; Beccaria, 1956). C’est peut-être à Reims au cours de ses études dans la cité champenoise que Fulbert (960-†1028 ; évêque à partir de 1006) entra en possession du ms., toutefois sans qu’aucun fait ne vienne étayer cette hypothèse basée uniquement sur la mention au XIV e s. du repertorium librorum de St-Pierre de Chartres. On ignore les circonstances qui conduisent le ms. à Paris à la fin du XVII e s., où il apparaît dans un lot de ms. acheté en 1678 (de provenance inconnue) par Etienne Baluze (†1718) pour la bibliothèque de Jean-Baptiste Colbert († 1683). Il est mentionné dans l’inventaire des « Livres achettez en diverses rencontres pendant l’année 1678 » (Paris, Bnf, Baluze, 100, f. 110r-v), sous le titre « Liber botanicus cum figuris, folio, ancien » qui a été porté au dos lors de sa reliure dans la bibliothèque colbertine (l’appartenance à la collection de M. Hardy est une erreur de la notice de l’exposition BNF, 1982, p. 71). Enfin, le ms. est rapporté dans le procès-verbal de la prisée des livres et médailles de Colbert : « Liber botanicus, vélin, 20 s.» (Mélange de Colbert, vol. 77 ; extraits cités par Delisle, Cab. des mss., I, p. 479, sous le n° 966 du catalogue des mss. de Colbert établi par Baluze) et figure ainsi parmi les mss. de Colbert acquis pour la bibliothèque du roi en 1732.
    Provenance : bnf.fr
    Date de mise en ligne : 13/06/2011”

  24. George Turner
    February 6, 2014 at 1:09 pm

    Well, it seems you now need to run some comparisons to several dialects of Nahuatl, to see if perhaps the manuscript was written by an Aztec brought over from the Americas, or so goes a new theory.

  1. No trackbacks yet.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: