Why does ObjectOutputStream.readObject() return object of type Object and not object of type I wrote it in?

Say I have class Car:

class Car implements Serializable{
    String color;
    
    public Car(String color) {
        this.color = color;
    }
}

Now I want to write object of class Car into ObjectStream. So it looks like this:

ObjectOutputStream oos = new ObjectOutputStream(new FileOutputStream("obj.dat"));
oos.writeObject(new Car("yellow"));

And then of course I would like to read object! So I would need to do this:

ObjectInputStream ois = new ObjectInputStream(new FileInputStream("obj.dat"));
Car car = (Car)ois.readObject();

My question is following: Why does method readObject() return Object and not Car? Doesn't my file obj.dat store all information about objects it holds? So no doubt it knows object I want to read is of type Car. So why it doesn't simply return Car, so that I don't have to do any casting?

Because it has no way of knowing what type of object is in the file/serialized data.

It does return a Car in this case.

But the compiler can't know that at compile time, because it depends on the data in obj.dat.

So the compiler has to assume the general case. The only thing it knows for sure is that readObject() will return some Object.

You can cast it to a more specific type if you know it will always be that, just as you did.

Or think about it the other way around: if you serialized a Car object and a Bike object after each other in the same file, then calling readObject() twice should return those two types.

Now the return type of that method is defined statically, so there's no way in the Java type system for the first call to indicate a different return type than the second call.