The word division is the whole problem. Maranatha is preserved in the Greek New Testament manuscripts as an unbroken string of Greek letters transliterating an underlying Aramaic phrase, and Greek manuscripts of this period do not reliably mark word boundaries, so where a reader chooses to split the string changes its grammar entirely. Divided as marana tha, it is an imperative, an urgent invocation, our Lord, come, addressed to Christ and looking forward to his return. Divided instead as maran atha, it becomes a declarative statement in a different tense, our Lord has come or our Lord is coming, a confession rather than a plea. Both divisions are grammatically defensible in Aramaic, and scholars have argued for both readings without reaching a consensus, since Paul, writing to a Greek speaking congregation in 1 Corinthians, gives no gloss and evidently expected the phrase to be understood as it stood, telling us it was already a fixed formula in the earliest Aramaic speaking church.
Its appearance again in the Didache, an early Christian manual usually dated to the first or second century, confirms maranatha functioned as fixed liturgical language rather than one writer's phrase, most plausibly connected there to the eucharistic meal, where an invocation calling on the Lord to come fits the ritual context well. Because Paul left it untranslated for readers who mostly did not speak Aramaic, maranatha stands as unusually direct evidence that this Aramaic liturgical phrase was already circulating as fixed formula among the earliest Christian communities, close enough in time to Jesus's own Aramaic speaking followers that scholars treat it as one of the closest verbal links back to that community preserved anywhere in the New Testament.