In Woit's: Quantum Theory, groups and representations, p109, 110, is proved the "Highest Weight Theorem":
Finite dimensional irreducible representations of $SU(2)$ have weights of the form $$-n,-n+2,\dots,n$$ for a non-negative integer, each with multiplicity 1, with $n$ a highest weight.
There is some part in the proof which I don't understand. One starts with the fact that there exists a highest weight $n$ and pick $v\in V_n$. Next we compute $$v_{n-2j}= \pi'(S_-)v_n\in V_{n-2j},$$
where $\pi'$ is the Lie algebra homomorphim between sl(2,$\mathbb{C}$) and $\mathrm{u}_{\mathbb{C}}(m)$ and $V_n$ is the "weight space". Now the proof says:
Consider the span of the $v_{n-2j}=\pi'(S_-)^jv_n \in V_{n-2j}$. To show that this is a representation one needs to show that the $\pi'(S_3)$ and $\pi'(S_+)$ leave it invariant.
What does it mean the author with "leave it invariant"? The proof goes on to find, as usual, that there is a minimum value of the weight equal to $-n$ but it is not clear what space is left invariant.