Part II also places a heavier influence on the notion of forms of life, and how they affect our relationships with one another. In denying that a dog can feel hope, Wittgenstein is not saying something about the mental capacities of a dog. While anger and joy are raw emotions that can come and go, hope only functions given certain surroundings. We could take a one-second cross-section of a person's life and conclude that she is happy at that moment, but we could not conclude that she is hopeful. Hope requires a temporal context in which we can express desires, speculate about the future, plan ahead, and so on. In saying that a dog cannot feel hopeful, Wittgenstein is not speculating that a dog's mental life is limited, but he is observing that a dog's linguistic life is limited. A dog cannot feel hope because dogs cannot speculate about the future or express desires for anything but immediate gratification. The language-games in which we talk about hope require certain surroundings and certain forms of life which a dog does not share.
This is not so much an observation about dogs as it is about the grammar of the word "hope." It is plausible that our use of that word, or our relationship with dogs, might change so that we could use it in reference to dogs. But as things are, we do not share those forms of life with dogs.
During the discussion of what it means to believe that someone is not an automaton, Wittgenstein asserts: "My attitude towards him is an attitude towards a soul. I am not of the opinion that he has a soul." We might gloss this position by saying that we can talk to people in the second person, but we cannot talk to lampposts in the second person. That is, it makes no sense to tell a lamppost something, to give it an order, to confess a secret, to share a joke, and so on. Calling someone "you" only makes sense if there is a receptive soul who can listen and respond. We may sometimes address a lamppost in the second person, but only as a joke or a sign of insanity. To believe or assert that someone is not an automaton implies that this is a plausible question to ask, and the very raising of the question requires a drastic shift in the way we think about other people. There would have to be hesitancy in our use of the second person, an uncanny feeling of uncertainty as to whether our words have any genuine effect at all. That we treat other people with "an attitude towards a soul" does not mean that we believe or even assume that they are not automata. That they are not automata is simply built into our forms of life in dealing with other people.
This is not to say that it is impossible to ask if someone is an automaton. But it does mean that we cannot simply ask. "He is not an automaton" is a "hinge" proposition (which Wittgenstein discusses in Part I, ¤¤422–570) like "the world has existed for more than five minutes." In bringing it into doubt, we are not questioning an isolated proposition, but a whole attitude that we assume toward the world.