I have been going through Shankar's QM book. In Chapter 4 under the topic of "How to Test Quantum Theory", he proposed that first of all we need to have a well known quantum state and for that we start with any arbitrary state and make a measurement (say the momentum) which will make the system collapse in one of the eigenstates of the operator which is something we know completely about.
Then later on in the chapter he claimed that the eigenstates of the momentum operator can't be physically realized because they can't be normalised to unity. Now I am not very much confident with the meaning of the term "Physically realizable" but I am thinking of it as "something which can't exist in reality".
But aren't the two claims inconsistent with each other ? Like if the eigenstates are not physically realizable then what exactly do we mean when we say that the system collapsed in one of the eigenstates of the operator ?
This might be silly but I am not able to go through this concept.
