{"id":229,"date":"2014-02-19T22:41:41","date_gmt":"2014-02-19T22:41:41","guid":{"rendered":"https:\/\/stephenbax.net\/?p=229"},"modified":"2014-03-15T18:32:46","modified_gmt":"2014-03-15T18:32:46","slug":"voynich-natural-language-or-not-1","status":"publish","type":"post","link":"https:\/\/stephenbax.net\/?p=229","title":{"rendered":"Voynich: natural language or not? (1)"},"content":{"rendered":"

I\u00e2\u20ac\u2122ve had many encouraging comments about my paper on the Voynich script<\/a>, but also some puzzling ones.<\/p>\n

Some people insist that the analysis “couldn’t be possible”, so they don\u00e2\u20ac\u2122t bother to consider it, because the script \u00e2\u20ac\u0153couldn’t represent a 1:1 mapping\u00e2\u20ac\u009d onto real words, or the signs \u00e2\u20ac\u0153couldn’t represent a 1:1 mapping\u00e2\u20ac\u009d onto letters gnuxe \ub2e4\uc6b4\ub85c\ub4dc<\/a>. Therefore they are sure that the script must be some kind of complex code.<\/p>\n

As a linguist, I am perplexed by this, so I thought I\u00e2\u20ac\u2122d explore it in this posting. To start with, this position hides a lot of assumptions. Firstly, it seems to assume that \u00e2\u20ac\u02dcnormal\u00e2\u20ac\u2122 scripts do or should have a 1:1 mapping of letters to sounds burp suite pro<\/a>. Well, English doesn’t for a start, as the word \u00e2\u20ac\u0153thorough\u00e2\u20ac\u009d (‘TH U R A’, roughly) can demonstrate \u00e2\u20ac\u201c 4 sounds, 8 letters. The Arabic verb meaning ‘s\/he wrote’ is written with three Arabic signs (K T B), though it is read with vowels which the reader must supply from memory (roughly\u00c2\u00a0\u00e2\u20ac\u02dcK A T A B A\u00e2\u20ac\u2122 ) \u00e2\u20ac\u201c i.e \uc544\uc774\ud3f0 \uc0b0\ub3cc \uace0\ub515 \ub2e4\uc6b4\ub85c\ub4dc<\/a>. 6 sounds, 3 letters. No sign of a 1:1 mapping there then.<\/p>\n

In terms of words: Arabic \u00e2\u20ac\u02dcana fi\u00e2\u20ac\u2122lbeet\u00e2\u20ac\u2122, written as 2 \u00e2\u20ac\u02dcwords\u00e2\u20ac\u2122 in Arabic script, is semantically 4 words: ana \u00c2\u00a0\u00c2\u00a0fi \u00c2\u00a0\u00c2\u00a0al \u00c2\u00a0\u00c2\u00a0beet\u00e2\u20ac\u2122 = \u00e2\u20ac\u0153I am in the house\u00e2\u20ac\u009d. \u00c2\u00a0So in terms of \u00e2\u20ac\u02dcscript words\u00e2\u20ac\u2122 to \u00e2\u20ac\u02dcsemantic words\u00e2\u20ac\u2122, \u00c2\u00a0no 1:1 mapping there.<\/p>\n

A thoughtful account of the argument concerning 1:1 mapping and the Voynich manuscript has been offered by Elmar Vogt (http:\/\/voynichthoughts.wordpress.com\/<\/a>) as follows:<\/p>\n

\u00e2\u20ac\u0153Any \u00e2\u20ac\u0153brute force\u00e2\u20ac\u009d attack [by computer] would presume that the ciphertext words of the VM are mapped 1:1 from the plaintext words when you want to go crazy<\/a>. And this is extremely unlikely for a number of reasons:<\/em><\/p>\n

    \n
  1. The ciphertext alphabet seems to consist of around 17 frequent letters, plus a large number of rare \u00e2\u20ac\u0153wierdos\u00e2\u20ac\u009d. That maps poorly to a latin alphabet.<\/em><\/li>\n
  2. Some frequent letter groups show up almost exclusively word-initial (\u00e2\u20ac\u0153qo\u00e2\u20ac\u009d) or word-terminal (\u00e2\u20ac\u0153dy\u00e2\u20ac\u009d) \ub9e5 \uc544\uc774\ucf58 \ub2e4\uc6b4\ub85c\ub4dc<\/a>. That\u00e2\u20ac\u2122s unknown for any Central European language.<\/em><\/li>\n
  3. Word-length distribution is odd: There is a shortage of both very short and very long words; words have a comparatively uniform length \u00e2\u20ac\u201d Again, this is unusual for Central European languages.<\/em><\/li>\n
  4. Overall, the words exhibit a very regular structure \u00e2\u20ac\u201d check out\u00c2\u00a0Stolfi\u00e2\u20ac\u2122s \u00e2\u20ac\u0153Core-Mantle-Crust\u00e2\u20ac\u009d paradigm<\/a> \ud30c\uc790\ub9c8 \ud504\ub80c\uc988 \ub2e4\uc6b4\ub85c\ub4dc<\/a>. (Yes, it\u00e2\u20ac\u2122s a tough read, but worth working it through if you want to understand the VM.) they are composed by a fairly rigid \u00e2\u20ac\u0153grammar\u00e2\u20ac\u009d, the like of it is unknown for European languages.<\/em><\/li>\n
  5. Nobody has been able to identify particles and articles (\u00e2\u20ac\u0153a\u00e2\u20ac\u009d, \u00e2\u20ac\u0153and\u00e2\u20ac\u009d, \u00e2\u20ac\u0153with\u00e2\u20ac\u009d\u00e2\u20ac\u00a6) in the VM.<\/em><\/li>\n<\/ol>\n

    \u00c2\u00a0All of these differences between natural languages and the VM make it highly unlikely that the enciphering mechanism simply always turned plaintext word \u00e2\u20ac\u0153A\u00e2\u20ac\u009d into ciphertext word \u00e2\u20ac\u0153X\u00e2\u20ac\u009d, and \u00e2\u20ac\u0153B\u00e2\u20ac\u009d into \u00e2\u20ac\u0153Y\u00e2\u20ac\u009d.*) I\u00e2\u20ac\u2122m convinced that one VM word is\u00c2\u00a0not\u00c2\u00a0equivalent to a plaintext word, but rather that it only represents a few letters.\u00e2\u20ac\u009d<\/em><\/p>\n

    <\/em>It is interesting that Elmar mentions European <\/i>languages four times, then says \u00e2\u20ac\u0153All of these differences between natural languages and the VM……..\u00e2\u20ac\u009d \ub354\uc704\uccd03 \ub2e4\uc6b4\ub85c\ub4dc<\/a>. I have two issues with this:<\/span><\/span><\/p>\n

    a)\u00c2\u00a0\u00c2\u00a0\u00c2\u00a0\u00c2\u00a0\u00c2\u00a0 There\u00e2\u20ac\u2122s a big confusion here between language<\/i> and script<\/i>. In theory just about any language could be written in any script \u00e2\u20ac\u201c the two terms are not synonymous. So it is wrong to mix them up, making a point about a script and then using it to make a deduction about a language in these words \ub2e4\uc6b4\ub85c\ub4dc<\/a>.<\/p>\n

    b)\u00c2\u00a0\u00c2\u00a0\u00c2\u00a0\u00c2\u00a0\u00c2\u00a0 We need to consider non-European <\/i>scripts and languages. It is perfectly possible that the underlying language of the VM script is non-European. And a non-European script \u00e2\u20ac\u201c for example an \u00e2\u20ac\u02dcabjad\u00e2\u20ac\u2122 like Arabic, or a script which in other ways does not represent all vowel sounds \u00e2\u20ac\u201c could well explain most of the points which Elmar mentions Monmon<\/a>.<\/p>\n

    As for his final \u00c2\u00a0point: \u00e2\u20ac\u0153I\u00e2\u20ac\u2122m convinced that one VM word is\u00c2\u00a0not\u00c2\u00a0equivalent to a plaintext word, but rather that it only represents a few letters\u00e2\u20ac\u009d<\/em>, well, in an abjad like Arabic, with vowels omitted, that is exactly what we get \u00e2\u20ac\u201c \u00e2\u20ac\u02dca word represented by only a few letters\u00e2\u20ac\u2122 because the reader must fill in the missing vowels herself.<\/p>\n

    Why do we Voynich people torture ourselves and insist on complicating things, when what we need to do is work out the script step-by-step, NOT closing off options by assuming that it must be European, or that it cannot be a 1:1 matching Idol Mubi<\/a>?<\/p>\n

    In fact in my paper I conclude that the words I have identified also do not represent a plain 1:1 match. For example the word for the plant (genus) Centaury is identified as KNTAIRN, with no written vowel after the first consonant, nor before the last consonant. No 1:1 match there of sign to sound, but why on earth should there be?<\/p>\n

    More to follow on this methinks\u00e2\u20ac\u00a6<\/p>\n","protected":false},"excerpt":{"rendered":"

    I\u00e2\u20ac\u2122ve had many encouraging comments about my paper on the Voynich script, but also some puzzling ones. Some people insist that the analysis “couldn’t be possible”, so they don\u00e2\u20ac\u2122t bother to consider it, because the script \u00e2\u20ac\u0153couldn’t represent a 1:1 mapping\u00e2\u20ac\u009d onto real words, or the signs \u00e2\u20ac\u0153couldn’t represent a 1:1 mapping\u00e2\u20ac\u009d onto letters gnuxe … <\/p>\n

    Read more »<\/a><\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[5,7],"tags":[],"class_list":["post-229","post","type-post","status-publish","format-standard","hentry","category-voynich","category-voynich_script_and_language"],"_links":{"self":[{"href":"https:\/\/stephenbax.net\/index.php?rest_route=\/wp\/v2\/posts\/229","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/stephenbax.net\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/stephenbax.net\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/stephenbax.net\/index.php?rest_route=\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/stephenbax.net\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=229"}],"version-history":[{"count":10,"href":"https:\/\/stephenbax.net\/index.php?rest_route=\/wp\/v2\/posts\/229\/revisions"}],"predecessor-version":[{"id":576,"href":"https:\/\/stephenbax.net\/index.php?rest_route=\/wp\/v2\/posts\/229\/revisions\/576"}],"wp:attachment":[{"href":"https:\/\/stephenbax.net\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=229"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/stephenbax.net\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=229"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/stephenbax.net\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=229"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}