The very word “variant” implies a deviation or change from something else taken for a norm. […] So what is commonly done in practice? Some particular text is chosen – often at random – for the norm. Either we use a printed text such as the Textus Receptus, sometimes an edition by Tischendorf, Westcott-Hort, or Nestle; or, we may use the text of a particular MS whose textual affinities are already known, e. g., Vaticanus or Alexandrinus. It is against these norms that other Greek texts are usually compared. And it is this process of comparison which gives rise to a set of variant readings for the text under consideration (Colwell and Tune 1964, 253).
Provided that there is no doubt about the readings in the manuscripts, the collation of manuscripts is a purely mechanical process, consisting of merely identifying places where two or more texts do not match (Hockey 1980, 144-145).
Conosciuti nella loro individualità i testimoni, si procede oltre nella recensio con la collatio, cioè li si mette a confronto parola per parola per quanto riguarda il testo in esame. Si sceglie un punto di riferimento, testo di collazione, rispetto al quale misurare convergenze e divergenze (Stussi 1994, 123).
The first job of all editors is to gather relevant variant forms of the work to be edited and, by collation, establish the relationship they bear to one another. […]
Scholars engaged in critical editing must do character-for-character comparison of all versions that could conceivably have been authoritatively revised or corrected. Collation forms the groundwork needed to determine, first, what changes occurred during composition of the work or between the time the work left the author‘s hand and the time it last rolled off the press while the author was living and, second, to determine whether the author, or an editor or compositor, was responsible for those changes (Shillingsburg 1996, 134).
Collation. Character-by-character comparison of two or more texts. Traditionally, the process by which editors isolate patterns of error that indicate transcriptional descent, to determine whether one or more texts is a copy made from an earlier and thus more reliable copy. Some twentieth-century editors have used the term to describe the process of visual comparison by which transcriptions, galleys, etc. are checked for error. With modern technology, the “compare” function in word-processing software performs this function by comparing the machine-readable versions of galleys and page proofs against each other (Kline 1998, 270).
Collation : Comparaison d’un texte avec un ou plusieurs autres témoins du même texte. Le mot vient du Latin conferre qui signifie poser côte à côte, comparer. – Collation des variantes, collation des témoins : Repérage des leçons variantes en chaque lieu variant des différents témoins du texte. – Collationner : Effectuer la comparaison d’un témoin d’un texte avec un autre (Vielliard 2001, 210).
La seconda operazione che porta all’edizione critica, successiva alla ricognizione dei testimoni, è la loro collazione. Essa consiste nel confronto (così etimologicamente: collatio, dal latino confero) di ciascun testimone con gli altri, al fine di rilevarne le differenze (Chiesa 2002, 48).
collation: Comparison. A collation is either the record of the substantive and accidental differences between two or more texts or the act of comparing two or more texts for the purpose of documenting their differences (MLA 2011).
The editor has normally studied all the extant manuscript and printed material that witnesses the passage of the work through time and has compared at least the early ones word-for-word and comma-for-comma. This usually computer- and machine-assisted process is called “collation” and it forms the basis of an extended report in the edition, usually called the Historical Collation. For practicality, it is often restricted to variant wordings (“substantives”), leaving the now-orphaned commas and other “accidentals” to look after themselves (Eggert 2013, 103).
For the Alexandrians, the gathering of different copies was because they were different, and could therefore be used to prepare lists of variant readings, in a process called collation (from “placing/laying side by side”) for comparison. Such collation, now often done by computer, though requiring some human input and control, is still done today (Greetham 2013, 21).
[In digital philology] transcription refers to the process of replicating the content of a manuscript text into a computer file for later manipulation and display. […] Collation is the process of reconciling these transcribed texts, finding the correspondences and variations beween them (Andrews 2014, 176).