In response to the questions posed in the first post of this thread, "the officer" is NOT the subject of the sentence regardless of whether the sentence is active or passive. "The officer" is undoubtedly the doer of the action in both sentences, but it is the subject only in the sentence with the active verb.
In the sentence with the passive verb, it can only appear in the predicate as "agent". It is still the doer of the action but no longer the subject of the sentence.
The subject, in a sentence with a monotransitive verb such as "capture" is always the "receiver" or "undergoer" of the action, and it will change from subject to direct object in an active sentence.
If we have a look at the post before mine, we will see "The officer has been captured by the subjects", which is untrue if we consider the sentences in the first post correct. The subjects have been captured, not the officer.
"The subjects" is not only the grammatical or formal subject of the
passive verb, it is also the notional subject or whatever other name
you might wish to call it. The subject of a sentence is not necessarily
the doer of the action, and that fact doesn't make the subject in
question any less "logical" than it should be. Actually, it'd be
illogical to make "the officer" the subject of the passive verb if what
you mean is precisely the opposite.
The officer, whether as agent
or subject (from a syntactic point of view) will always be the "agent"
of both sentences, the active and the passive, from a different
perspective, from that of thematic roles. Perhaps analysing the
sentence in terms of thematic roles will help you see the diferences
between active and passive constructions more clearly. In this type of
analysis, the agent (not a syntactic function) is the doer of the
action regardless of whether the sentence is active or passive and also regardless of the position the construction occupies in the sentence. The
direct object of an active sentence, and the subject of that sentence
in the passive voice, are called "patient" if a living entity, and
"theme" if it's a non-living entity. Again in this case, it shows that places are not always what define a function.
Leaving that aside and
coming back to transformational grammar, Chomsky and his theories
aren't the easiest to understand. But it is important to remember that
an active sentence and a passive one require different mental processes
and that, according to Chomsky, when you think of a sentence there are
choices you must make before actually coming up with an utterance. One
of those choices has to do with the sentence being active or passive.
If you decide on a passive sentence, that will dictate a number of
sub-processes needed in order to make adjustments to produce a
grammatical sentence. Chomsky says that the choices concerning a
sentence in the passive voice are made at the very beginning, at the
moment you decide your sentence will be in the passive voice instead of
the active. All this usually happens without us being aware of our own
mental processes, but it seems we make decisions such as subject-verb
agreement the very moment we decide what type of sentence we wish to
produce. In his first book (Syntactic Structures, 1957, Chomsky made the rules for the passive voice appear as "optional". Later, in 1965, when he was already closer to becoming a rationalist or mentalist, and farther away from structuralism, he saw that the rule couldn't possibly be optional and that they should appear at the level of the phrase structure rules; in other words, before the "transformational" stage.
I hope this makes sense. I tried to put it in very few words, but I'm not sure it was a good idea. 
Miriam