Without seeing the queries involved, I'm just guessing.
But, it is possible that eve if it's just getting scanned, it's still useful. Let's say your clustered index consists of 1000 pages and your non-clustered index consists of 500. If the optimizer recognizes that either one could satisfy the query, but both are going to be scans, it's going to scan the smaller index because that will have a lower cost for the query.
I'm not saying that's the case here. I don't know anything of the situation to make that kind of assessment.