Three Laws Rebuttal

Anderson's new Three Laws of Robotics seem to be thorough, but they have their own problems, just like Asimov's original Three Laws. Anderson's new Three Laws are basically rewrites of the original Three Laws with a few qualifiers thrown in to make it seem more like his laws. The assignment was to create Three Original Laws of Robotics, not to reuse the old ones. Anderson's Second Law, while it protects the robot from deliberate harm, does not protect it from accidents or unintentional harm. For example, a robot could be trying to help someone in danger, but in the process of saving them, accidentally destroy itself beyond repair. The robot could be charged with saving someone from a volcano. There is a high likelihood that if the robot stays there too long, it will start to melt and lose some of its functions. Or, in another example, a robot could be rescuing a person from the top of a very tall bridge, slip, and fall into the water, accidentally destroying itself. In both cases there was no aggressor involved. The robots were destroyed because of their bad luck.
The terms "mental harm" and "emotional harm" are difficult to understand in the context of the Laws. If, according to Anderson's First Law, the robot wishes to avoid causing mental and emotional harm to a person, the robot must analyze the person's reaction to what they did. If it does turn out to cause mental or emotional harm, the robot would have broken the First Law. How is a robot supposed to measure emotional or mental harm? These are not easily measured because everyone reacts in different ways to different situations. For a robot, measuring emotional and mental harm is especially difficult. The robot does not have the advantage of being a human to interpret body language for signs of mental or emotional harm. It would be impossible for a robot to determine what constitutes as mental or emotional harm. As a result, robots would live in a constant state of confusion and not be able to function properly.