Six Accounts of Actuality: Taking Stock

In earlier posts I have set out a range of solutions to the problem of accounting for the nature of actual existence: possibilism, actualism, and theories of predication. In this post, I would like to take stock.

We have a total of six viable solutions to the problem of actuality, compatible with a broadly Aristotelian framework:

1. Haecceity theory: actualism with haecceities.
2. Ostrich possibilism: possibilism with a primitive property of actual existence, not reified as an individual accident but treated in an “ostrich nominalist” fashion.
3. Bare Platonistic possibilism: possibilism with a universal of Actual Existence, with a primitive (non-reified) property of instantiation linking some possible substances and accidents to the universal
4. Instantial-tie Platonistic hybrid theory: possibilism with a universal of Actual Existence, and with reified instantial ties linking particular possible substances and accidents with the universal.
5. State-of-affairs Platonic hybrid theory: possibilism with a universal of Actual Existence, with reified states of affairs combining the universal with particular substances and accidents,
6. Thomistic act-of-existence theory: with reified acts of existence associated with certain possible substances and accidents.

Theories 4, 5, and 6 involve a hybrid of possibilism and actualism. They are all possibilists about possible substances and accidents, but actualists about the special entities (instantial ties, existential states of affairs, and acts of existence). What is the justification for such hybrid theories? If actualism is good enough for ties, states of affairs, or acts of existence, why isn’t it good enough for substances and accidents? And, conversely, if possibilism is necessary for substances and accidents, why isn’t is equally necessary for these special entities?

My answer is this: the special entities add nothing to the character of a possible world. They stand outside the possibility as fully determinate possibility. Consequently, there is no need to posit merely possible acts of existential (or existential ties or states of affairs). Acts of existence have no identity of their own, and so there is no need for de re reference to merely possible acts of existence: it suffices to have de re reference to the corresponding substances and accidents. Similarly, merely possible acts of existence are not needed to meet the Parmenidean challenge, since all possible change involves merely possible substances and accidents.

On all six accounts, facts about the actual existence or non-existence of a possible world would be external to that world, not included in the specification of each world as the particular world it is. Consequently, it could be a contingent fact that the actual world is actual, instead of one of the others.

Let’s consider briefly some of the drawbacks of the five alternative accounts, from a broadly Aristotelian perspective. First, there is actualism with haecceity theory. Robert M. Adams (1979, 1981) has argued that the existence of a property of thisness (e.g., the property of being identical to Socrates) must be dependent on the actual existence of the entity in question. If Socrates did not exist, then neither would the property of being identical to Socrates. If that’s right, there could be no uninstantiated haecceities, and haecceity theory would fail to repair the problems with actualism (see also Fine 1985).

We think that it is very plausible that no uninstantiated property could be such that, were it to be instantiated, it would be the property of being identical to some particular thing. An uninstantiated property is necessarily a de dicto property (since there is no corresponding entity that can be named or demonstrated). No de dicto property can entail the numerical distinctness of its possible bearer from all other possible things. De dicto properties would seem to be inherently repeatable, instantiable by different things in different worlds. Haecceity theorists must claim that if haecceity H is instantiated in both world w₁ and in world w₂, then, if w₁ were actual, there would exist something that would also be the unique instantiator of H in w₂, if w₂ were actual (and vice versa). How can a de dicto property in the actual world so coordinate the non-actual members of the domains of two different possible worlds, if (as actualists must suppose) there are no such non-actual members to be coordinated? How does the instantiation of a common, actually uninstantiated property ground the would-be identity of the two instances in the two worlds? That is, what in the actual world can make it the case that H is really a haecceity? It seems that haecceity theory is really a disguised version of possibilism. It only makes sense if there is a fixed domain of possible entities.

There is also a critical problem with all three versions of existential Platonism (bare, instantial-tie, and state-of-affairs). We can ask: does the universal of Actual Existence itself actually exist? It seems that such a universal can be responsible for the actual existence of things only by actually existing. How could a relation to a non-existent thing be the cause of something’s actually existing? But if so, it seems that the existential Platonists’ account is viciously circular. The universal actually exists because it instantiates itself, but its instantiating itself can ground its existence only because it actually exists. So, its existence would ground itself, which is viciously circular.

This is a special problem for Platonism as applied to actual existence. A parallel objection against other Platonic Forms would not be cogent. A Platonic form of Humanity, say, could be self-exemplifying without vicious circularity. There’s no reason why the Form of Humanity could not make things to be human independently of its being human itself. So, the Form of Humanity could make itself to be human without vicious circularity. However, the Form of Actual Existence could not make itself to be actually existent. But this is the only way for the Form to be actually existent, on the Platonist account, and a non-existent Form could not be responsible for making anything to be existent in actuality.

This leaves just two accounts: the existential ostrich-nominalist (ostrich possibilism) and the act-of-existence theorist. The existential ostrich-nominalist must introduce a new, primitive predicate, actually exists. Arguably, the act-of-existence theorists have an advantage here. They can use the same formal-causal relation that stands between substantial forms and packets of prime matter to stand between acts of existence and substantial forms. In Summa Theologiae I, Q4, a1 ad 3 (on perfection), Thomas states that “existence is that which actuates all things, even their forms (ipsum esse est actualitas omnium rerum, et etiam ipsarum formarum).” He states there that acts of existence are formal principles (formale).

Consequently, act-of-existence theorists need no new ideological baggage, which means that an application of a version of Ockham’s razor favors the act-of-existence theory. When adopting theories, we should avoid new primitive vocabulary whenever possible. This kind of ideological simplicity is of more importance than mere quantitative economy (Koons and Pickavance 2017, 140-141; Lewis 1986, 4-5).

In fact, we do not have to presuppose that Thomas’s act-of-existence account of actuality is correct for Thomas’s overall strategy to be successful. We do not have to assume even that simple actualism or possibilism are wrong. Instead, the Thomist can legitimately reason in the opposite direction: from the existence of an uncausable cause of a per se necessary being to the correctness of the act-of-existence theory. We will argue in the next section that only Thomas’s act-of-existence theory provides for the possibility of a truly uncausable, per se necessary being (as required by the Second and Third Ways).