The paper starts with a presentation of the pure happiness theory, i.e. the idea that the quality of a person?s life is dependent on one thing only, viz. how happy that person is. To find out whether this type of theory is plausible or not, I examine the standard arguments for and against this theory, including Nozick?s experience machine argument. I then investigate how the theory can be modified in order to avoid the most serious objections. I first examine different types of epistemic modifications of the theory (e.g. the idea that a person?s happiness is more valuable for her if it is based on a correct perception of her own life), and then turn to a number of modifications which all make the value of a person?s happiness depend on whether the evaluative standard on which her happiness is based satisfies certain requirements. In connection with this, I present and defend my own modified version of the happiness theory.