Translation Approach
The first challenge was to determine whether the symbols unique to the manuscript made up an alphabetic system (perhaps incorporating digits) or whether they were simply decorative flourishes expressing some aesthetic purpose. While the latter was (and is) a possibility it has always seemed intuitively unlikely. Thus the concentration on the former as the more promising approach, with many of the symbols being similar to, or identical with the letters of the Latin alphabet, and in some cases not greatly dissimilar to letter shapes of other familiar alphabets, such as Cyrillic. Therefore this analysis will refer to all members of the symbol set as letters. However, it may turn out that some of these symbols may double as letters and digits and a small number may even stand for syllables. To be sure, even if the symbols are mainly letters, it does not follow that they combine to form words producing plain as opposed to enciphered text. Their status as parts of meaningful segments of language, that is, lexical or grammatical items, depends on where they fall relative to one another.

Symbols as Letters
The question to be asked is: Do the symbols in the Voynich manuscript, which are candidates for letters, appear to group together repeatedly to form possible words or grammatical elements such as verb tenses or noun plurals in some natural language and in a text written from left to right; or are the symbols simply distributed at random?

The first line of analysis is to accept as likely the fact that the symbols are letters of an alphabet, not some kind of decoration, and that they can be conjoined into words, phrases, and the like to produce text in natural language. Since it is not possible to draw a picture of an adverb or participle, the best approach would be to find recurring relationships held together by grammatical endings, function words in examples such as “and” and the like, in phrases such as “nights and days" and “one or two”.

As noted above many of the symbols do resemble those of known alphabets so therefore the next step has been to establish whether the symbols are distributed in patterns suggestive of plain text in a natural language. In fact the symbols do seem to show patterns that point to base forms of words as well as inflectional endings throughout the manuscript. However in many of the folios which contain great numbers of drawings the words appear in isolation, either as names of objects or their functions, or with some unknown purpose. Thus it seemed most useful to concentrate on the linguistic grouping of folios which not only yields word-like forms but also appears to link them syntactically.

English as a Language Example
In English the letters “s-t-a-r” form a high-frequency English word, “s-t-a-m” forms a potential word (e.g. an acronym of some kind which might become a word), and while “s-t-m-a” is impossible in English as a word or a grammatical item, it could be an abbreviation. These letter combinations are either canonic, representing actual or potential English words or grammatical elements, or they are non-canonic, as in the case of “s-t-m-a” described above.

English of course has many cases in which the writing system diverges widely from the spoken language. Thus the cluster “g-h-t” certainly does not represent today a pronounceable letter set, but it nonetheless occurs frequently in medial and final word position. Examples of this are “lighten”, “thought”, and so on. Similarly, French has two words beginning with the three vowel letters “o-e-u”, oeuf (egg) and oeuvre (work), otherwise the initial combination is very rare in that language. The two words and their derivatives are very frequent, however, so that from the point of view of the writing system the letter combinations are canonic. On the other hand, English words never end with “h-g”, nor do French words begin with “e-o-u”, for such combinations suggest that any text containing them is either full of typographical errors or is an encryption of some kind.

In English for example the isolated word “day” is canonic and very common. As an isolated part of a presumably deciphered message, however, it would count for little. On the other hand if the decipherment included “The days are (?)-ing (?)-er”, the next segment might be: “The days are (grow)-ing (long)-er”.

Syntactic Structure
An emergent syntactic structure, with nouns and verbs seemingly reflecting agreement and government is most evident in the linguistic section although it also is present in a more rudimentary form in the pharmaceutical and astrological sections of the manuscript. The patterning most resembles that of certain members of the Germanic language family.

Germanic Languages
Germanic languages in general are characterized by vowel changes referred to as “Umlaut” and “Ablaut”. These terms, coined by German philologists, apply respectively to sound changes which are brought about either by assimilation, e.g. the slightly archaic English plural “brethren” of figurative "brother”, where “o” of the latter word changes to “e” under the influence of the final “e”, or by the grammatical change in which vowel alternations in the so-called strong verbs signal meaning change as in English “sing”, “sang”, “sung”. As far as has been determined by this research verb Ablaut does not play a role in the language of this manuscript, although regular or so-called weak verbs do show a surprising affinity to the weak verbs of other Germanic languages.

There is a case of two symbols representing one value, that is the pair representing “e”. This digraph is found in great frequency in words in which other, assumed, Germanic cognates show “e”. A similar digraph appears in Gothic words with “ai” which also match “e” in most other Germanic languages.

East Nordic Languages
The Germanic languages from which the accompanying transcriptions and glossings are made appear to belong to the North Germanic family of languages, most likely East Nordic languages. The author or authors regularly cite linguistic forms in comparing and contrasting languages which are referred to as (otei) “Jutish”, (ðei) “Danish”, and (gotei) “Gothic”. As noted in the overall approach to translating this manuscript these terms are tentative but are useful in helping to differentiate relatively close language systems.

One interesting fact emerges, the authorial language is not Gothic because the writer in describing sounds of Gothic uses the form (oðar) "sounds” rather than (hoðar), which is cited in another text as the Gothic form.

Thus it has been convenient to refer to the author's language as the subject language (Jutish/Gothic/other). Also it has seemed best to offer glossings and cited object language forms in a textual presentation as free as possible from question marks and hypothetical notes.