The typical case is that you are studying a text which deliberately reuses language from another, earlier text, and that this re-use is most often in the form of two-word phrases.
The “target” text is the text you are studying most closely. It is generally the alluding text, and the more recent.
The target drop-down list offers finer-grained detail, and your results will be sorted by location in the target text by default.
The source text is the earlier, alluded-to text. Selection here is not as fine, as most users are interested in allusions throughout a source work.
Clicking on “Compare Texts” will initiate a default search for parallel language between the selected texts. By default, results will be pairs of lines, one from each text, sharing at least two words; i.e., the classical loci similes.
Words are considered to be “shared” if they partake of a common dictionary headword. Note that parsing is automatic; inflected forms whose stem is ambiguous will match on any possibility. This results in some (to humans) obvious false matches.
Your search may take a few moments, particularly if you have chosen full epics under both drop-down lists. Note that results for such searches often number in the tens of thousands. It is usually wise to choose the target text as specifically as possible.
If you need to search two large texts and your browser is timing out before the results are ready, please feel free to email us and we'll send you your results.