This note deals with two related properties of a closed subgroup H of a locally compact group G. The first involves the complete reducibility of a representation ρ of G restricted to H, as compared with that of ρ itself, while the second concerns the Zariski density of H in G, when the latter is linear. The second essentially means that, under appropriate representations, an H-invariant subspace is automatically G-invariant, while the first means that given an H-invariant decomposition into two subspaces, one of which is G-invariant, we can replace the other by some complementary G-subspace. As we shall see, these concepts are related in other ways.We first take up the question of complete reducibility. The least detailed form of Clifford's theorem, to which we shall return later, states that the restriction to a normal subgroup H of a completely reducible representation of a group G remains completely reducible. In Theorem 1 and its first Corollary we take the opposite tack; we ask if the restriction of a representation to H is completely reducible, must the same have been true of the representation itself? Rather than normality of the subgroup, what is needed here is some finiteness property of H with respect to G. Then follows Corollary 3 which is a generalization, to this setting, of the fact that a continuous representation of a compact group on a Hilbert space is completely reducible. Recalling the density theorem of R. Mosak and the author in [6], we then prove two results which guarantee H is Zariski dense in G. Theorem 4 deals with arbitrary connected real linear groups and their cofinite volume subgroups and The research for this article was done while the author was supported by a Collaborative Grant from the Research Foundation of CUNY. The author would like to take this opportunity to thank the RF for its generous support. It is also a pleasure to thank Gopal Prasad for reading an earlier version and making a number of valuable comments.