Well, it should be said, first of all, that in Latin the verb in an indirect question is invariably in the subjunctive mood. This use of the subjunctive simply indicates the subordination of the questioning clause and has no distinct meaning.
The only issue you raise that seems to need explanation is when, in certain adjectival or adverbial clauses (not indirect questions), the subjunctive is used instead of the indicative. Such a construction is known as virtual oratio obliqua and is used when there is no specific verb of ‘saying’ in the passage yet the context clearly demands it.
This substitution of the indicative for the subjunctive shows the writer representing the reported statement of someone else, as opposed to his own.
For instance:
supplicatio decreta est, quod Italiam bello liberassem. A thanksgiving was decreed because (as they said) I had saved Italy from war.
It is necessary to add in a phrase such as “as he said” whenever one spots VOO. Such a construction is often used when the speaker wants to devolve responsibility from himself and place it on, perhaps, the accusers. e.g.
Socrates accusatus est quod corrumperet iuvuntatem.
Socrates was accused of corrupting the youth (so the accusers said).
The use of ‘corrumpebat’ would imply that the writer agreed with the given charge.
The reverse of this can exist, too, such as in a subordinate qui-clause which, as you know, would generally be in the subjunctive in oratio obliqua, the writer can use the indicative mood to show that what is said there is his view, not that which was reported, and it is thus almost an aside.
So that’s that.
All question words introduce verbs, when in indirect questions, in the subjunctive mood. The forms of interrogative particles do not change between direct and indirect questions, except that num when used indirectly does not imply a negative answer and nonne can only be used after quaero.
Surely you know words like quantus, qualis, quotiens, quam, quo, qui…etc?
~dave
