er to XYZ; likewise, if there was H2O in the oceans, 'water' would refer to H2O. So the primary intension can really just be thought of as picking out what Chalmers refers to as "watery stuff" in whatever possible world one is in; therefore, the primary intension of a concept is a priori. The secondary intension of a concept is not a priori; rather, its reference depends upon how our world is. The secondary intension is discovered by applying the primary intension of a concept in our world. For example, we discover that what we call 'water' actually has the chemical composition H2O, and this is its secondary intension. We can then apply the same reference across all possible worlds, so that in a world in which XYZ is found in the oceans, the term 'water' still only refers to H2O, not what's in the oceans. Given that the primary intension of water (i.e., watery stuff) picks out H2O in our world, the concept WATER picks out H2O in all possible worlds according to its secondary intension. Once we understand this structure, we can determine the difference between logical (a priori) and metaphysical (a posteriori) necessity. A statement is logically necessary if it is true in all possible worlds when it is evaluated according to its primary intensions. Conversely, a statement is metaphysically necessary if it is true in all possible worlds when it is evaluated according to its secondary intensions. Chalmers' strongest attack on materialists who wish to employ a Kripkean metaphysical identity move in the case of consciousness stems from the natural idea that having phenomenal experiences is all that it takes to be conscious in any possible world. According to this idea, if human consciousness consisted of physical processes and Martian consciousness, for example, consisted of immaterial processes, as long as Martians experienced phenomenal feelings, their consciousness would be as much a form of consciousness as ours is. But this argument is ...
