Notes on transcription

The editions in Corpus Grammaticorum Hibernicorum are monoptic (monotypic) and semi-diplomatic.

Abbreviations, even the most common ones, have been expanded and are encoded with the TEI <ex>-element. As this was done in order to facilitate the full collation of the witnesses, suspensions (of e.g. verbal endings) have been expanded as uniformly as feasible (compare Binchy's Corpus Iuris Hibernici), without taking the scribe's language into consideration. It is needless to say that, with some obvious exceptions, forms that have been expanded from contractions or suspensions cannot be used in linguistic research. The use of general (ambiguous) abbreviations is quite common in the codices. The expansion of abbreviations involve an inevitable loss of information. That information may be integrated into the editions at a later stage.

Excepting the substitution of 'h' for the spiritus or punctum delens ('gh' for 'g' with spiritus, etc), no silent changes have been made to the orthography of the witnesses. Capitalisation of proper names etc. follow modern conventions and the researcher must therefore consult the manuscript images if the use of particular allographs or similar is important to the research question. The systematic representation of such detail is a natural addition to the present editions and this is therefore fully supported by the current framework. The punctuation is modern and does not represent that of the manuscripts.

In sum, departures from a strictly diplomatic transcription in the editions presented here are:

• The expansion of abbreviations
• The insertion of modern (syntactical) punctuation
• The capitalisation of proper names, placenames and the removal of capitals foreign to modern practice, such as the frequent use of two consecutive majuscule letters in the beginning of new paragraphs, e.g. ‘INtan’
• The line division of the manuscripts has not been encoded

These editorial choices were motivated by the relative ease of access of photos of the manuscripts themselves (Irish Script on Screen), which furnish a better basis for studies that need to take such features into account. The editions hopefully provide all the information that is relevant to linguistics and stemmatology while facilitating access to the text of the individual manuscripts for the researcher that lacks the palaeographic training required to consult the sources directly.

The editions are typeset in Junicode. Junicode is modeled on the Pica Roman type purchased by Oxford University in 1692. OpenType allows the automatic generation of features such as 'tall s', Insular 'f' and so on (not activated). The document is still searchable.