SUMMARY
This paper focuses on the asymmetry observed in the realization of the compound marker -(s)I(n). When there is a sub-compound in the possessor position of a larger compound, this morpheme is realized. On the other hand, when the (so-called) sub-compound is in the possessee position of the larger compound, it does not show up. The question that arises at this point is whether it is deleted or never realized. The present paper makes a minimalist analysis of these compounds. It proposes that they have different structures and it is misleading to seek for a parallelism between them. While there are multiple domains for agreement for the former case, there is only one agreement domain for the latter case. It is also maintained that there is no sub-compound of a larger compound in the former case. Rather, there is only one compound in such constructions which has got more than two constituents. It is asserted here that there is no deletion, restriction or ban on the re-occurrence of this morpheme. It simply never re-occurs.