The bulk of the Plantinga article discusses basic characterizations of modalities, possible worlds, and the like that I mostly discussed when we talked about Kripke. Plantinga acts as if there is only kind of modality that is of interest, something like what I loosely described as metaphysical necessity, which lies between logical necessity and causal necessity:

2+2=4 is, he says, necessary but not logically necessary, while it is possible to travel faster than light though it is causally impossible. He's wrong: all the modalities are, in various contexts, of interest, but he's right that this is the most often employed kind (especially in metaphysics) and so, when a philosopher starts using modal language without specifying, this is probably what you should take to be intended, at least to a first approximation.

In addition, Plantinga emphasizes that none of the relevant concepts have clear descriptions, let alone definitions, and so we have to proceed by example. He seems to think that the notions are, nonetheless, quite specific.

De re modality and de dicto modality

Things as said (de dicto) can be necessary, contingent, possible, or impossible. What is said is truth bearers (the sorts of things that might be true or false, which philosophers take to be, for example, propositions, sentences, statements, utterances, and the like). I'll just take them to be sentences. Thus, it is sentences that can be necessary, contingent, possible, or impossible.

Examples:

• Necessarily, 9 is composite.
• Necessarily, Lumpl is Goliath.

What is said to be necessary is "9 is composite" and "Lumpl is Goliath."

Objects (not truth bearers) have properties, and they may have them essentially or accidentally. If an object has a property essentially, it could not be that object without having that property. If an object has a property accidentally, it might not have had the property.

Examples:

• 9 is essentially composite.
• Lumpl is essentially (the same thing as) Goliath.

It is usually taken to be the case that associated with every nonmodal property there is the corresponding property of having that property essentially. (Similarly, for accidental.) Thus, since 9 has "the property of being composite" essentially, 9 also has the property of being "essentially composite."

Unfortunately, philosophers when formalizing modal talk only use "necessary," not "essential." Thus, the de re modal fact that 9 is essentially composite is almost always expressed as follows: "9 is necessarily composite," or "necessarily, 9 is composite." I just stipulated that I used those turns of phrase to express a de re property of 9, but, unfortunately, the very same sentences can be read as expressing a de dicto truth, namely, that the sentence "9 is composite" is necessarily true. That leads to all sorts of trouble.

For example, if a=b and a has property P, then so does b. But then, since 8 is essentially composite and there are 8 planets, it follows that the number of planets is essentially composite. If that argument is good, it shows that de re modality is incoherent, since otherwise we have an example of an argument about de re modality that is sound, has true premises, and a false conclusion.

The argument is ambiguous:

Version 1:

• 8 has the property "of being composite" essentially.
• The number of planets is 8.

---

• The number of planets has the property "of being composite" essentially.

Version 2: * 8 has the property "of being composite" essentially. * The number of planets is 8.

---

• Necessarily, the number of planets is composite.

All standard arguments against de re modality rely on ordinary intuitions, and so they are phrased in something close to normal English. They are all, as a result, ambiguous.

Nonetheless, there remains a sense that the arguments point toward something deeply problematic about the idea of essential properties. In particular, for any given object, we don't seem to have any clear way of determining what the essential properties are. Was Abraham Lincoln essentially male? essentially human? essentially rational? essentially a political animal?

When you start playing games with the integrity of a person, it very rapidly becomes clear that we have no idea what is essential to the identity of a person.

Quine says that a bicyclist is essentially bipedal but accidentally rational, while a mathematician is essentially rational but accidentally bipedal. Then, he asks, what about a bicycling mathematician?

Gibbard, following Carnap, ducks the kind of problem we are discussing by replacing de re talk about object with de dicto talk about "individual essences."

There is one more kind of argument, though closely related to the above: we eliminate any favored description by using a quantifier:

• Paul is a bicyclist
• Bicyclists are essentially bipedal.

---

• Something is essentially bipedal.

What does that mean?

Either

• %$\square\exists x Px$% de dicto, used in the bad version of the argument.
• %$\exists x\square Px$% ????
• %$\square Bp\rightarrow\square Pp$% de re, used in the good version of the argument.

-- ShaughanLavine - 18 Sep 2008