Philosophical problems never sit still. As they pass from person to person, they shift about. Different thinkers have different views on what, precisely, is relevant to the problem; arguments are refuted, salvaged, defended, refuted again; new arguments are proposed; new evidences are introduced into the discussion, sometimes as new additions, sometimes as replacements for old evidences; people forget what their predecessors had already done, or misinterpret them, or never learn about them; and so forth. If you pick a philosophical problem - the mind-body problem, say, or the problem of evil - determining exactly what it is, is something of a feat.
I once knew someone who argued in this way: "What use is it to know Descartes's attempt at a solution to the mind-body problem, or any other historical solution? I know enough about the mind-body problem that I can start working on a solution to it without dealing with the historical attempts at all. At most they could save me time, to the extent that they would already have covered the ground I need to cover." And this is entirely true; if you already know the character of the mind-body problem, you don't need to know much else in order to start working on the solution. But without looking at Descartes, Aquinas, et al., you don't know the natuer of the problem; or, at least, without knowing them, you could only know the nature of the problem by a completely lucky guess. For the mind-body problem is not a unified thing; it has shifted, continues to shift, and will continue to shift, and if you try to formulate it as a monolithic thing, you will inevitably make assumptions about the problem that are not necessary. It happens again and again.
History of philosophy, as a discipline, is in part concerned with precisely this sort of issue. It is not an easy thing to determine the nature of the problem of evil, for instance; it consists of a myriad of variations on a legion of themes. Generalizations about the problem itself (and it is virtually impossible not to make generalizations about it) can only be warranted by a study of the actual forms the problem has taken. The actual forms the problem has taken are part of the data for any such generalizations; and, while they are data that need to be dissected and heavily analyzed in order to draw any conclusions, we have very little else on which to say anything about what the problem of evil actually is. Thus HoP uses basic historical tools - historical tools insofar as they are relevant to the study of texts - in order to give us some hold on the problem of the philosophical problem. This is not the only thing it does, I think; but it does occupy a great part of the end for which HoPers work.
