Science certainly does not start with putting label on a problem.
Actually it can.
By putting a label on it you make it easier to group examples of the problem for study.
Look at AIDS. Without the label "AIDS" it would have taken longer to solve the underlying problem because it manifests in so many different ways. Having 10,000 As is easier to analyze than 100 Bs, 100Cs and so on.
Now, sometimes you'll end up grouping disparate problems under one label which can slow things somewhat (you have a treatment that's very good for cause A that makes up only 1% of the group--apply it to the group and you'll see poor results) but such things usually get resolved without too much difficulty--the scientists know they might be dealing with multiple causes.