Neven Jovanović: Atomic philology and parallel philology – some implications of CITE architecture ------------------------------------------------------------------------ [Examples of CTS queries][1] -- [Literature and Links][2] -- [Slides][3] -- [Handout][4] The CITE architecture, developed from 2006 by Neel Smith and Christopher Blackwell, is an editorial concept with a potential to transform digital philology, primarily because it is able to solve two difficult problems - it deals with overlapping textual hierarchies, and it connects annotations with the relevant textual segments. The CITE framework also impels us to think more precisely about what we are doing as philologists. We realize that our work begins at the "atomic" level of a text, and that interpretative processes happen "in parallel". CITE architecture gives the editor full control over (and responsibility for) the level of segmentation of the text. Each segment, regardless of what it consists of, is marked by a hierarchically structured Uniform resource identifier. The problem of overlapping hierarchies - for example, the standard Stephanus page divisions which may clash with a logical division of Plato's text into sentences and paragraphs - is thus solved by proposing a compound digital edition. Such edition must not be limited to one text containing all of the editor's theories about it; it may comprise multiple "analytical exemplars" - in the case of Plato, one such exemplar may be divided into Stephanus pages, the other into paragraphs and sentences, the third into individual words, the fourth may consist only of "key words" expressing key philosophical contexts. Because each segment in each of these parallel, single-function exemplars has a unique URI, it is possible (and necessary) to construct a concordance table, saying that this passage in one exemplar corresponds to that one in the other exemplar. The architecture also extends the concept of unique URIs to the annotations - each note bears a URI pointing to the relevant textual segment (in the relevant exemplar). We have tested the CITE architecture by applying it to texts of Croatian Latin, most of them early modern works (the earliest text is from the 976, the latest from 1984) written by Croatian and other authors connected with people and regions of today's Croatia. We have also used the CITE annotations to describe lexical and grammatical features of selected words (place names and place references), but also to describe realia - actual (or sometimes imaginary) places and periods signified by these words. The experience leads me to propose a new vision of philology - the discipline as an ability to manipulate texts at their most basic, "atomic" level, with such precision that interpretations arise from studying different configurations of numerous simple "philological atoms". Once we are able to follow a certain combination - for example, a real place in an imaginary time, denoted by different words in different grammatical forms - through a large number of texts, the interpretation happens almost by itself. [1]: https://osf.io/dbf73/wiki/Catullus%20CTS%20Queries/ "Examples of CTS queries" [2]: https://osf.io/dbf73/wiki/Literature%20and%20Links/ [3]: http://croala.ffzg.unizg.hr/atomic-parallel/ [4]: https://github.com/nevenjovanovic/catullus-cts/blob/master/docs/jovanovic-atomic-parallel-handout.pdf