Closed Under Addition:
Closed Under Scalar Multiplication:
From the hypothesis we know is an element of and via condition 1, and that is an element of from the definition of intersection.
Labeling as , we can say , satisfying condition 1.
In the event that , we can consider it a subspace of , since easily satisfies conditions 1, 2, and 3. If my mathematical parlance is correct, I believe this is the trivial case of being a subspace of .
If , however, then we have additional work to do, to show it satisfies conditions 1, 2, and 3, with regard to the other elements it may contain. We've already demonstrated, however, that contains at least , so we don't need to prove that it satisfies condition 1, in the event that there are additional elements.
Moving forward, assume two elements that are common to , and : from the hypothesis, we know , and , via condition 2, meaning that , for any , in , and , so that satisfies condition 2.
Next taking some element , common to and , we know that for some , we have , and , via condition 3, so that , meaning satisfies condition 3.
Also, since our subspaces, and their union satisfy condition 3, if we choose our to be , and select some , common to and , then , so that has an additive inverse. Since satisfies conditions 1, 2 and 3, of being a subspace, is a subspace of . Additionally, I decided to investigate the case where there are both elements common to , and , and elements that aren't, because I wanted to convince myself whether or not the stated intersection would still be a valid subspace.
Assuming and : The fact that , doesn't violate the conditions for being a subspace, since the definition of closure under addition requires , which would contradict the definition of intersection, and is an intersection. The fact that , for some doesn't violate the conditions because the definition of closure under scalar multiplication again requires , which again contradicts the definition of intersection. This also means that the lack of an additive inverse for in doesn't violate the conditions for being a subspace, because the additive inverse of , , which is just scalar multiplication, and even if we wanted to use addition, say , we'd still need , which we already ruled out.