The paper explores two issues arising from the extension of the notion of alignment from the domain of monotransitive to ditransitive clauses. The first is, To what extent should other than purely formal patterns of identification be taken into account in determining the ditransitive counterparts of the respective monotransitive alignments, i.e., accusative, ergative, active, etc.? The second issue is, How should conflicts between the formal criteria used in the determination of the alignment of person agreement be resolved? It is argued that only under a purely formal interpretation of alignment is it possible to discern ditransitive counterparts of all the major monotransitive alignments, and that once the notion of neutral alignment is distinguished from absence of agreement, form/order conflicts involving agreement markers can be consistently resolved in favour of phonological form.