In much recent literature, it is proposed that sluicing may not always hide a regular embedded question, but instead, a copular Wh-question (often a cleft). Such claims raise questions about the degree to which identity conditions on ellipsis are sensitive to syntax. This paper reviews the literature on non-isomorphic sluices, presents new evidence for such cases, and concludes that identity conditions in sluicing must, at least, be blind to syntactic differences between copular Wh-questions and their non-copular antecedents. An important challenge to such a view comes from case-matching effects in sluicing (Ross 1969). Copular Wh-arguments in sluices, in many languages and contexts, differ in case from their antecedent correlates. Given case-matching, we expect such sluices to be unacceptable. I show that an empirically motivated characterization of case-matching is compatible with a view that allows (non-isomorphic) sluiced copular clauses. In short, abstract Case is irrelevant in sluicing, and instead, morphological case is what matters. In languages that mark case on remnants, case must match with the correlate (e.g., German, Russian, Greek), but in languages that do not (e.g. Brazilian Portuguese, English, Spanish), abstract Case need not match. One consequence, is that ellipsis of copular clauses is available in many instances, and not others, with cross-linguistic consequences about when and where we expect non-isomorphic copular sluices to be available.