When reasoning about modality, it is often useful to think of possibility and necessity in terms of so-called “possible worlds.” A possible world is simply a complete possibility—a way that all of reality could be, or a way that all of reality could have been. (By “all of reality,” I mean everything that ever exists, including everything that ever existed in the past, everything that exists now, and everything that ever will exist in the future, in this universe or beyond: the physical world, God, abstract entities like numbers, and anything else that exists.) In other words, a possible world is a way that absolutely everything—everything everywhere and everything throughout all of time—could be or could have been.
With this concept in mind, we can say that a proposition is possibly true iff it is true in at least one possible world; and we can say that it is necessarily true iff it is true in all possible worlds.
There are different kinds of possible worlds, corresponding to the various types of modality discussed on the previous page. For example, a conceptually possible world is a way that reality conceivably could be. An epistemically possible world, for you, is a way reality could be for all you know. A physically (or nomologically) possible world is a possibility permitted by the laws of physics, and so on.
Besides the various types of modality described previously, there is another important distinction to be drawn when dealing with modes of truth. In the foregoing discussion, the concept of necessity was associated with propositions: a proposition is necessarily true iff it is true in all possible worlds. Necessity that is attributed to a proposition in this way is called necessity de dicto. (The Latin phrase de dicto means “about what is said.”) Sometimes, however, philosophers attribute necessity to things (entities) rather than to propositions. In particular, philosophers often speak of something having a particular property necessarily, which means that the property in question is an essential property—the thing couldn’t possibly exist without that property. Necessity that is attributed to an entity (rather than to a proposition) is known as necessity de re. (The Latin phrase de re means “about the thing.”)
Although de dicto and de re necessity are often conflated in ordinary language, the distinction between them is important. For example, if a theologian says that God is necessarily good, she might mean that the proposition God is good is a necessary truth—i.e., that it is true in all possible worlds. This would imply that God exists in all possible worlds, and thus entails that the proposition God exists is necessarily true as well. On the other hand, the theologian might just mean that God is necessarily good if He exists—in other words, that God (if He exists) has the property of goodness as an essential property. This entails that God is good in all possible worlds in which God exists, but does not entail that God exists in all possible worlds.