[robinzfc]

Isabelle/HOL is still types. The underlying type theory of Isabelle/HOL is not theory of dependent types, but theory of simple types. Isabelle/ZF would be a better example as it encodes Zermelo–Fraenkel set theory.

[jfmc]

Right!