The lemma "o" is noted upos=AUX in Breton-KEB. I wonder if it should not be an ADP.
It is indeed a Tense Aspectual Marker in that it bears aspectual information (/they are "at" working/ > 'they are working'), and it is always before a verb, but it is closer to the preposition ouzh 'on'.
Aspect in the lemma "o" derives from an aspectual reading of the spatial preposition "ouzh" (coïncidence of figure and ground is used for coïncidence of the tense of the action and the tense of the clause). The lemma "o" also appears under the form "ouzh" of the preposition if a proclitic object separates it from its verb ("ouzh e welet" "seing him").
The lemma "o" bears no Tense information, and no agreement. It has no selectional effect on its following verb.
The lemma "o" is noted upos=AUX in Breton-KEB. I wonder if it should not be an ADP. It is indeed a Tense Aspectual Marker in that it bears aspectual information (/they are "at" working/ > 'they are working'), and it is always before a verb, but it is closer to the preposition ouzh 'on'. Aspect in the lemma "o" derives from an aspectual reading of the spatial preposition "ouzh" (coïncidence of figure and ground is used for coïncidence of the tense of the action and the tense of the clause). The lemma "o" also appears under the form "ouzh" of the preposition if a proclitic object separates it from its verb ("ouzh e welet" "seing him"). The lemma "o" bears no Tense information, and no agreement. It has no selectional effect on its following verb.