The subtle difference between interaction and causal effect (law of cause and effect)

Question

I was talking with a friend about autonomous cars, he said to me that for these cars is required full interaction with other vehicles. I thought that with "interaction" he was referring to "comunication" between vehicles, i.e. sharing informations, but he told me

Interaction does not mean communicating, but simply knowing that there are other objects and being aware of their movements and actions.

So i replied to him that what he described was not interaction because interaction is a two-way action. The car detects the surrounding space on its own, so there is no interaction.

He then told me

An interaction exists when one object A detects the presence of another object B and A modifies its behaviour according to the movements of B.

and I replied that this is not interaction, but simply the law of cause and effect, and that interaction exists when two objects act on each other.

Here is where the discussion ended.

Since I was not sure about the definition of interaction, I searched for it and find that "Interaction is a kind of action that occur as two or more objects have an effect upon one another. The idea of a two-way effect is essential in the concept of interaction, as opposed to a one-way causal effect." (from wikipedia).

So are we saying the same thing in different ways or one is wrong and the other is right ?

Answer 1

▻ SHIFT OF POSITION

Your friend shifts position on interaction : first it is 'simply knowing that there are other objects and being aware of their movements and actions'. Then it becomes 'detecting the presence of another object and modifying behaviour according to the movements of that object'.

It can't be both : the second involves more than simply knowing. It adds the modification of behaviour in the light of knowledge. Assume that the second formula marks the one your friend's revised position. We'll work with that.

Strictly speaking, interaction involves the action of two or more things on each other. Your friend only refers to the action of one object, the car, not on the other or of the other on it. One car simply maneouvres away from the other. But words have shades of meaning. The situation you describe can be called one of interaction.

Does this formula, 'detecting the presence of another object and modifying behaviour according to the movements of that object', reduce or equate to cause and effect ? In the context you describe, I think it does.

▻ CAUSATION

Causation remains one of the most contentious concepts in philosophy but I am inclined to say that the formula does reduce or equate to cause and effect. But first, a bit of clarification.

I take the view that A is 'the' cause of B if and only if the occurrence of A is sufficient for that of B. If A is not sufficient but only necessary, then it is only 'a' cause of B.

So : if (A) contracting the H7N9 virus is sufficient for (B) catching influenza then (A) is the cause of (B). (If.)

But if dropping a lighted match (A) is not sufficient for (B) the factory's burning down since the fire would not have come about if (C) a worker had not noticed the incipient fire and ignored it and (D) someone had happened once the fire had started, to place inflammatory material near the fire, then the fire would not have happened, dropping the match is a cause of the fire. All else equal the fire would not have happened without it; it was necessary for the fire.

▻ INTERACTION IS A CAUSAL RELATION

For ease of statement, let's assume we're talking about 'the' cause. Is (a) detecting the presence of another object the cause of (b) modified behaviour according to the movements of that object ? An automated car is programmed to behave in certain ways. Given its program, it does not have a choice of behaviour. In defined conditions - detecting an object - it modifies its behaviour, e.g. by moving away from the object. 'It modifies its behaviour' means 'Its behaviour alters' and its behaviour alters because the car has been programmed to do so. The 'because' here is causal : the existence of the active program is the cause of the alteration in the car's behaviour.

▻ FOOTNOTE

My definition of causation can be disputed but it is fairly standard and appears to inform the ordinary ways in which we talk and think of one thing causing another. It is capable of more sophisticated statement; it is capable of rejection by someone who sees causation quite differently or doesn't accept the idea of causation at all. But this is my answer, so I use my own understanding.

It is perfectly possible that AI will produce 'thinking' cars which are not causally controlled by their programs as the car in the example is. But given the car in the example, you are in my view quite right to refer to cause and effect.