The dictionary fascicle doesn't give a very strong content-based treatment of etymologies, and since they're a difficult area, I can understand why. That doesn't stop me from wanting to do content tagging of etymologies, though! Is anyone else doing this? I have some notes that I could gladly share, in trying to come up with a generic set of content-specific tagging sutable for etymologies in dictionaries from the 18th C. and onward. I don't have easy access to any earlier dictionaries that try to explain the origin of words, but I do have a 1727 copy of Nathan Bailey's dictionary, the first English dictionary to mark pronunciation :-).. as well as more modern dictionaries. Is there a forum for discussing such things? I'm particularly interested in ways of expressing the different types of relationships between roots. I'd like to be able to mark that when Bailey says Algebra comes from Al, excellent, and Geber, the name of its supposed inventor (!), that these are parallel, as opposed to where an English word comes from a French word that the lexicographer believed to derive from a Greek word. I'll append a single example; stop reading now if this is numbingly tedious! Lee Justice [_giustizia_, It. _justicia_, Sp. _justissa_, Port. of _justitia_, L.] <Etymology> <Derivation> <word.group> <word><root>giustizia <language>Italian </word> <word><root>justicia <language>Spanish </word> <word><root>justissa <language>Portugese </word> </word.group> <connector>of</> <word> <root>justitia <language>Latin </word> </Derivation> </Etymology> The word.group is a placeholder to group together all the parallel roots; this is a many-to-one relationship, as a DBMS person would say. A one-to-many (which also happens) would have a <word> before the connector and a <word.group> after, and so on.