It makes sense only if the reader supplies the explanatory language that is missing.
If I now understand correctly, you are describing a situation in which a reference to an instance of an object A is passed to the constructor for a second object, B. That constructor places the reference to A into a variable private to B. Let’s call that B_ref_to_A.
The claim that this breaks encapsulation is based on the possibility that code outside of B can manipulate A. This can happen, but this does not break encapsulation. If code outside of B could change B_ref_to_A, that would break encapsulation, because B_ref_to_A is a variable private to B. But a change in A’s variables does not break encapsulation, because A’s variables are not B’s variables. If the writer of B needs the state of A to be constant, he/she would indeed need to explicitly clone A during the construction of B; otherwise, he/she would not be justified in assuming that the state of A would remain constant.