Let’s say I’m an alien, hovering above a major city in my space ship. I make the observation that the green traffic lights make the cars go. What kind of evidence would that be?
I mean, it’s true in a sense that green lights make the cars go, but it’s actually the people inside the cars that make the cars go.
And what would the evidence be called that shows people make the cars go? Because, this statement is actually true.
You are looking at evidence of correlation, but there is a logical fallacy that says that correlation does not imply causation. In statistics, just because two things can be correlated or occur together, does not mean that one causes the other.
Tests for causality are much more complicated and involved, and philosophically, there is an argument that causality cannot be proven, especially outside of questions involving scientific laws. For traffic lights and drivers, there is sociological evidence: we know that humans on Earth are taught that a green light means go, and we can test by asking people what they do when the light turns green.
Casuistry, Case-based-evidence.
Case-based-inference is a form of Causal Inference.
Casuistry is a term of insult. Causal Inference is a philosophical and mathematical question that is getting more attention now that Evidence-Based-Medicine is better understood: Once you realize that aliens can’t do random controlled trials on traffic lights, you want to understand what evidence they can derive without the trials.
I think this gets into the different sorts of causation, of various degrees of directness.
But that’s not the distinction here. The green light does cause cars to go.
Sorry, I shouldn’t have suggested that casuistry is always, or even normally, causal inference. Just in this case.
Casuistry is normally finding a generalization, a rule, or a principle from a case, and is normally considered a faulty form of deductive reasoning. Causes can be considered a form of generalization, rule or principle, and inference is a ‘faulty form of deductive reasoning’.
No, the OP asked what type of evidence the aliens are seeing, and they are seeing evidence of correlation (green lights/cars going). He then distinguished between the green lights (which don’t actually make cars go) and the people driving (who do). As I mentioned, this gets into the philosophical question of causation. But if you accept his premise that it is the drivers who cause the cars to go, rather than the green lights, then this is a case where correlation does not equal causation.
A good test would be to put a bunch of cars in front of traffic lights with no drivers. Although I have not performed the experiment, my hypothesis is that they would not go, even if the lights turned green. Assuming of course, that these are not driverless cars.
Green lights are a necessary but not sufficient factor in cars moving. You would see a strong correlation between traffic lights turning green and cars beginning to move. But if you checked more thoroughly you would see exceptions where the cars didn’t move even after the light turned green. Assuming you had the ability to observe at that scale, you would also see the correlation between the actions of the human in the front left seat and the car moving. And you would see that this correlation is even stronger than the green light correlation. You would observe cases where the light turned green but the human did not take any actions; in those cases the car would not move. (And you would also observe cases where the light was red or yellow but the human was still taking the actions and the car was moving.)
If you had extremely good equipment and could observe mechanical details inside the cars, you would also be able to figure out how it worked. You would observe that there was no sensors in the cars which could detect the changing of the lights or any other means of transmitting or receiving information between the traffic light and the car. But you would observe that the actions the human in the front left seat was taking was operating devices which were connected to the car’s motor and fuel supply.
You might even be able to observe that there are organs on the front of the humans’s heads which are capable to detecting the change from a red light to a green light. And you might note that in many cases where the humans did not initiate the car moving operations they had moved their heads so that their sensory organs were not observing the traffic light. And this point you would have not only deduced the pattern but also figured out the processes that cause the pattern.
Your test implies that drivers are necessary, it does not imply that the light turning green is irrelevant. The OP’s premise is wrong. The green light is part of a causal chain that makes the cars go.
Actually, not even a “necessary” factor technically. If you observed long enough you would see some cars moving without a green light. Some would be turning right on red, and once and awhile, a car would blow through the red light. Probably very confusing to the aliens.
The first time a car blew through a red light and T-boned someone proceeding on green-- I wonder what the aliens would do with that data. Especially if police arrived to direct cars around the accident, and the cars moved or did not move regardless of whether the lights were green.
The Earth Dwellers is based exactly on his concept.
The residents of Uranus study Earth through a telescope, and draw all kinds of erroneous conclusions.
He’s mistaking signaling (the green light signals the drivers that the car can move, and so they move) for direct causation - just like the people who called Gene Roddenberry to ask him for advice on designing automated doors, since the doors on Star Trek open when a person is near - but in fact, the doors open on Star Trek because there was a guy offscreen who saw the actor get close to the door and opened it then.
Your car is controlled by a computer.
For this distinction to be meaningful, you must show that what goes on inside a brain is something other than computation.
I think there’s still a distinction between a computer screen going blank because of a power outage, and a computer screen going blank because the computer has received a signal directing it to blank the screen. In one case there is a physical cause that does not depend on an arbitrary code, while in the other case, the physical cause (receipt of a signal) is not fundamentally associated with the response, but only associated with the response due to an arbitrary choice on the part of the designer.
At a very basic level what you have is an observation and a correlation.
That’s it.
Science begins at the point where you explore that correlation, hypothesise about it, experiment where you can, collect further data and make further observations and further hypotheses, and so on and so on.
When this has been accomplished to the point where you uncover the actual mechanism (or at least a mechanism that explains the observations sufficiently well to be able to make predictions) then you have the “theory of traffic lights” but even that is still subject to future observation and experimentation.
Our Alien is likely to be able to infer that the mechanism in the car that ‘sees’ a green light has some decision-making ability. They would understand that a green light is different from pressing the ‘on’ button on a machine because there is so much variation in the car’s responses.
They might understand that pressing the ‘start’ button on my car does not start the car directly - it merely sets in motion a chain of events that results in the car engine starting. There is an observable difference between that and the green light, because the start button always sets the same chain of events in motion, while the reaction to a green traffic light is variable.
The conclusion must be that cars have at least some limited decision-making capability - unlike the start button on my car.
Your alien is making an observation drawing the wrong conclusion. The alien observes that when the lights go green the cars go. That is a correlation. The alien’s conclusion is that there is causation, which is where it goes wrong.
Well, since it’s based on observation and not experimentation, I would consider it empirical evidence. I observe with my eyes that, when the lights turn green, the vehicles proceed, and when they turn red, they stop.
Except there is causation. When I’m at an intersection and see a red light, I stop my car. The red light caused me to decide to stop. When I see a green light, I start driving again. The green light caused me to do decide to drive again.