I thought I'd come back to this, as it is prompt to the rest of this post.
There can be a tendency to discuss RPGing as if what matters is the abstract text of the books we use. But the actual event of RPGing is a concrete that happens among human beings.
So the fact that a rulebook or a module doesn't contain express instructions to the GM who is running it to do railroad-y things (like, eg, blocking or twisting players' action declarations that might undo the railroad) doesn't really tell us whether or not it's a railroad.
What is more important, in my view, is to look at
what sort of play is likely to result if a group tries to use the rulebook or module, following its instructions in good faith. Because assuming that someone in the group has forked out good money for the module, and that someone in the group has spent their time prepping to run the module, then that is what we can expect the group to do.
And if what the module presents is a sequence or bundle of events that the players are expected to work their way through, with certain outcomes being precursors to subsequent events (otherwise they wouldn't be playing the module), then I think its reasonable to characterise the module as a railroad, by which is meant that it has a structure that will tend to produce a pre-scripted experience in play.
Whether this is a good or bad feature of the module is obviously a matter of taste (see eg
@Retreater's post 120 upthread). But I think from the analytical perspective, as opposed to the perspective of preferences, it is a pretty key feature.