OK, so just so you know, I took "We can't detect it" as "We sensed no change, or at least none accountable to a mysterious force instead of a force already known."
It's a question of how accurate your model of the force is, and how well you can detect that it works according to that model, and how complete that model is.
We have a fairly complete model of EM in the modern era. We fully understand how radio waves are controlling that R/C car. The only evidence the ancient Romans have that some force is in operation is that the car moves according to commands given by the remote.
At your rate, only theoritizing things, even if there is no reproductible evidence of this thing, I can supputate there is a force of humor in this universe that force any experiment to fail miserably at the most important time... There is even some proof of this force, enough for earthlings to name it. "Murphy's Law".
You can. I would actually - diverging a little here, sorry - qualify that as a legitimate moral force, because the actual result of "If anything can go wrong, it will" is not a statement of probability being perverse, but rather that anything you leave to chance is going to be one more possible failure point, and that potential failure points add up quickly. So the law is better described as, "Minimize the possible points of failure by making sure as many things are in a known-good state as possible before enacting any plan." Why is this a "moral" law? Because it's a law of prescribed behavior which, when disobeyed, has real consequences, but which is not inevitably going to be disastrous every time it's ignored. Phrased another way, it's a variant on, "If a job is worth doing, it's worth doing well."
And to rebound on your example, what's preventing the roman from creating a secret passage based on gears (known at the time from windmills) and putting the R/C car to propel (somewhat slowly) said gears to, say, lift a door. They successfully harnessed the mysterious force causing remote action to influence action on the car. Is that not their technology now ?
They have detected kinetic motion from the car, so as far as they know, there is a mysterious something causing the car to respond to the manipulation of the remote control. This gives them no further information about the nature of this force than they had prior (though it does help them figure out how strong the maximum kinetic energy of the car under its own power is).
They may even incorrectly conclude that this mysterious force is pushing the car directly, or at least directly pushing the car's wheels. They lack the technology to determine that there are, in fact, at least two forces at work: one communicative, and one mechanical. Even dissecting the car and determining that there is a motor that is pushed by "canned lightning" or the like would only give them a clear picture that, somehow, the car was compelled to channel its "canned lightning" into pushing its wheels to move itself. It would take a great deal of study and advancing technology to realize that "canned lightning" is a manifestation of the same force that creates the invisible phenomenon we know as "radio waves."
COULD they eventually figure it out, given the demonstration of the existence of this (to them) "mysterious force?" Sure; such things do spark leaps in technological advancement. But it would take R&D.
If we actually could find equally repeatable demonstrations of human telekinesis, we could begin to study it and trying to figure out how it worked, because we'd HAVE the example of it in action and could start with what we
can measure to try to find where exactly the "mysterious force" is taking over from known forces and properties. That would help us develop hypothetical models, at which point we could start trying to design experiments to precisely detect something that obeys that model.
We say we "know" or "understand" something in science when we have a model that allows us to make reliable predictions with accuracy we consider sufficient. There is some judgment call there as to what is "sufficient," but we are capable of coming to reasonable consensus as to what is sufficient, in general. Generally, it's the point at which we're comfortable making predictions based on it. (This is why we still use Newtonian physics, despite knowing it has inaccuracies; on a human scale, it's accurate
enough that we can make very useful predictions.)
I think that there is some evidence now that space may, in fact, just be not analog at all, but on a quantic grid. A very, very fine grid, but a grid nonetheless. I stumbled upon that when looking around quantic theory.
The interesting bit, to me, about this theory, is that it means there's a mechanism for defining precise location, and, moreover, for moving from one discrete location to another. If this is true, one is given to wonder what that mechanism for transport is, and how it can be used to, perhaps, "skip" one or more of these "quantic grid squares."
Thinking on this, it also opens the question as to the SHAPE of these grid spaces. What is the shape that tessellates the universe?