Content uploaded by Jeffrey Goodman
Author content
All content in this area was uploaded by Jeffrey Goodman on Jul 18, 2022
Content may be subject to copyright.
Fictional Universal Realism
[word count: 5794]
1. The Doctrine of Universal Realism and Some Questions Concerning Fictional Universals
Let’s suppose, for the purposes of this paper, that you subscribe to Universal Realism
(henceforth, UR), which entails the following facts about you.
1
You are a realist about properties and
relations, and you take them to be necessarily existing abstracta of a certain sort, viz., sui generis
universals. You think universals are fairly abundant, and for a fairly wide range of meaningful
predicates, you hold that there is not only the extension – the set of entities to which the predicate
applies – but there is also the universal expressed by the predicate. You further believe that when a
meaningful predicate expresses a universal, that universal is the semantic content of the predicate that
expresses it.
2
Now, when wondering about just how wide a category this is, you of course realize that
‘is non self-instantiating’ expresses no universal.
3
But you also commit yourself to something slightly
stronger: while ‘is a dog’, ‘is a quark’, and ‘has 3 siblings’ all express universals, as does the
uninstantiated ‘is a blue swan’, you deny that ‘is a round square’ expresses a universal.
4
‘Uninstantiable universal’ refers to nothing whatsoever; in order for any predicate to express a
universal, it must be a predicate that possibly has a non-null extension. In other words, you are
ontologically committed to universals that go “beyond science” and correspond to most of our
1
And not just that you believe UR, but you do so rationally. You have arguments or reasons for UR that you
take to be sound and are ready to defend, such as the One Over Many argument, perhaps coupled with
considerations about what you consider the best semantics for predicates.
2
That is, it at least provides the content to the propositions expressed by sincere assertions of sentences
containing those predicates.
3
Absurdly, if such a universal existed, it would entail that something exists that both instantiates itself and fails
to instantiate itself.
4
While ‘is a round square’ would express an uninstantiable universal if it expressed one at all, such a predicate
would be relevantly dissimilar from ‘is non-self instantiable’. If the universal round-square-hood exists, that
does not entail that something exists that is both round and non-round; the universal itself would instantiate
neither round-hood nor square-hood, and any other entity that instantiated one would fail to instantiate the
other. An aside: by my lights, the idea that instantiability is essential to a universal is too restrictive. ‘Is
perfectly round’, e.g., seems to express a universal, even though nothing concrete or abstract could instantiate it.
But, for purposes of the present paper, specifically, to avoid an argument for realism about some fictional
universals that would be trivially sound, I will assume that instantiability is essential.
2
ordinary concepts, even though this means being ontologically committed to universals that are
uninstantiated. A bright line is drawn for you, however, at concepts corresponding to universals that
cannot be instantiated.
There nonetheless may remain for you the following interesting question: does ‘is from the
planet Krypton’ express a universal? Generally, should you think that all such predicates introduced
by works of fiction are relevantly similar to ‘is a round square’? After all, if the only entities that
could instantiate such universals are purely fictional, how could such predicates have non-null
extensions? Or are they relevantly similar to ‘is a blue swan’?
To begin to tackle these questions, let us first focus on a three-way distinction: real (abstract
or concrete) entities, fictional entities, and actual entities. Candidate real, nonfictional, actual entities
are pineapples, coffee cups, the number 7, and the set of sharks. Now, if Fictional Anti-Realists are
correct, then all fictional entities are unreal (and a fortiori non-actual). However, if some Fictional
Realists are correct, then some actual entities are fictional entities, such as Sherlock Holmes,
Superman, and Cinderella.
5
If other Fictional Realists are correct, then all fictional entities are non-
actual entities, that is, mere possibilia or non-existents.
6
At any rate, all such fictional entities are
understood by the theorists just mentioned to be particulars, not shareable universals. So, the
questions of interest from the preceding paragraph can be boiled down to these: What’s the proper
ontology of fictional universals?
7
What should you – believer in fairly abundant, necessarily existing,
abstract universals – say about being a friend of Dr. Watson, having parents from planet Krypton, and
turning into a pumpkin post-midnight? Are these entities real or unreal? If such universals are real,
what, if anything, does this dictate about the extensions of the predicates that express them? Are any
comprised of actual entities? Could they only be comprised of non-actualia?
5
I have in mind here van Inwagen (1977), Thomasson (1999), and Braun (2015).
6
I have in mind here Lewis (1978), Parsons (1980), and Priest (2005). And I say that some fictional realists
accept fictionalia as non-existents or mere possibilia. All such things would be non-actualia. My terminology
in this paper thus presupposes the falsity of Actualism – the thesis that all real things are actual – but nothing of
substance turns on this terminological decision.
7
Note that on my usage here, ‘fictional universal’ functions like ‘brown duck’, not ‘decoy duck.’ A fictional
universal is both fictional (in the sense of essentially being originally mentioned in a work of fiction) and a
genuine universal. A fictional detective, e.g., Sherlock Holmes, is not a genuine detective, but a fictional
universal, if extant, is a genuine universal. This is an important terminological difference from Braun (2015);
see Section 2.
3
2. What Would Anti-Realism about Fictional Universals Look Like?
Recent philosophical literature teems with views regarding the ontology of fictional
particulars – elaborate elaborations of varieties of Fictional Anti-Realisms and Realisms. Braun
(2015) and Sawyer (2015), however, explicitly focus their discussions on the ontology of fictional
non-particulars. Here is Sawyer summarizing her project:
…[T]he … question of whether a fictional predicate refers to a fictional property is, as far as I know,
rarely discussed. … [I]f fictional predicates do not refer to fictional properties, then semantic questions arise
about how to make sense of the apparent phenomena of meaning, reference, and truth. … If, on the other hand,
fictional predicates do refer to fictional properties, then metaphysical questions arise about the nature and scope
of those properties. … I argue that the question of whether a fictional name refers to a fictional character is
inherently bound up with the question of whether a fictional predicate refers to a fictional property.
Consequently, the former, more discussed question (about fictional names and characters) cannot be answered
independently of the latter, generally neglected question (about fictional predicates and properties). Crucially, a
number of semantic theories of fictional names and metaphysical theories of fictional characters presuppose,
either explicitly or implicitly, that there are fictional properties to which fictional predicates refer…. I argue
that this presupposition is inconsistent with antirealist theories of fictional characters and that it cannot be taken
for granted by realist theories of fictional characters either. … [T]he considerations adduced favour … a theory
which is consistently antirealist about both fictional characters and fictional properties. (208-209)
Braun also argues for a kind of anti-realism qua fictional non-particulars. However, he
differs from Sawyer in that he finds no reason in this to be an anti-realist about fictional particulars.
On Braun’s account, ‘is from planet Krypton’ at times expresses a “fictional attribute”, but it never
expresses a fictional universal nor has as its extension a set of particulars that instantiate the property
being from planet Krypton; rather, when the predicate determinately contributes something other than
a gap to propositions expressed by sentences in which it figures, it contributes a created, abstract
particular. And the argument used to establish this is of the very same sort
8
he employs to establish
that the semantic content of ‘Superman’ is also at times a created, abstract particular.
9
So, Braun is a
realist about fictional particulars, and while he is a realist about so-called fictional attributes, he is an
anti-realist about fictional non-particulars, and more generally, fictional universals. The fictional
attributes he believes in are no more genuine universals than Superman is a genuine person.
My project in this paper is as follows: I wish to argue that for you – believer in fairly
abundant, necessarily existing, abstract universals – full-blown antirealism about fictional universals
8
Roughly, it’s the sort of Quinean argument van Inwagen (1977, 41-3) provides.
9
Other times it contributes a gap, other times its contribution is indeterminate.
4
is not supported by the most powerful argument in its favor. In fact, there is nothing additionally
peculiar about fictional universals as compared to ordinary universals that stands in the way of
extending your UR view to encompass Fictional Universal Realism (henceforth, FUR), the view that
some universals are fictional universals. And I wish to further argue, contra Sawyer specifically, that
we can indeed separate questions of the ontology of fictional particulars from questions concerning
the ontology of fictional non-particulars; whether or not fictional names for particulars ever refer, and
regardless of their semantic contribution to the propositions expressed by sentences they figure in,
there are times when fictional predicates express genuine universals. There are some universals
introduced to us by fiction such that it’s possible they are instantiated; some fictional predicates
possibly apply to a non-null set of entities.
In addition to the difference in the way that Braun and I might use terms such as ‘fictional
attribute’ or ‘fictional property’ – (oftentimes) meaning created abstract particular in his mouth, but
meaning abstract sui generis universal in mine – some merely terminological issues regarding
Sawyer’s framing of the issues should also be cleared up. As she points out, we should not, of course,
understand ‘fictional predicate’ to mean predicate that appears in a work of fiction; many ordinary
predicates applying to non-fictional particulars also appear in fictional works. By her lights, a
fictional predicate is an expression that “refers to a fictional property, where a fictional property is to
be understood as a property which is not instantiated by real (non-fictional) individuals” (209). But
note, Sawyer says she is concerned with “whether a fictional predicate refers to a fictional property”
(208, my emphasis). Given that I’ll be understanding the reference of a predicate to be its extension
throughout, a perfectly uncharitable interpretation of her project would take her as endorsing an anti-
realist conclusion qua fictional non-particulars that is trivially true. Even realists (of every stripe)
about fictional particulars such as Superman think that ‘is from planet Krypton’ refers (in my sense)
to nothing actual whatsoever; its extension is actually empty.
10
So, I think a charitable interpretation
10
For instance, some Fictional Realists are Artifactualists who identify Superman with a created abstractum.
But they still think ‘is from planet Krypton’ has a non-null actual extension because an abstractum is not the sort
of thing that can hail from any planet. Other realists, Meinongians, e.g., identify Superman with a flesh and
blood superhero who genuinely hails from the planet Krypton, but Superman, and anything else hailing from
planet Krypton, are all non-actual in virtue of being non-existent.
5
of Sawyer’s anti-realism is instead the claim that the semantic content of a meaningful fictional
predicate is either nothing whatsoever (not even a gap) or something non-identical to a fictional non-
particular (hence, a fortiori, something non-identical to a fictional universal). In other words, we can
charitably understand (both Braun and) Sawyer as ultimately defending views that entail that for
every meaningful fictional predicate R, when meaningful sentences of the form ‘x is R’ express a
proposition P, the semantic contents of P will fail to include the universal R-hood.
I contend that the most powerful argument (as well as the runner-up) for anti-realism
regarding fictional universals at most shows that a great many, such as the one that would be
expressed by ‘is from planet Krypton’, could not be instantiated, and thus do not exist. With possible
instantiation as the criterion for a predicate expressing a genuine universal, then I surely agree – a
great many fictional predicates don’t live up. But full-blown anti-realism about fictional universals
(or Fictional Universal Anti-Realism; henceforth, FUAR) is unwarranted. The idea that no fictional
predicate can express a fictional universal is made only apparently plausible due to the sorts of
examples typically chosen.
3. Braun’s Argument for Fictional Universal Anti-Realism
The most powerful argument for FUAR does not come from Braun. In fact, he doesn’t offer
any arguments directly in support of FUAR. As I made clear at the outset, instantiability is the
existence condition for the sorts of non-particulars I’m interested in here, but actual instantiation is the
existence condition for non-particulars that Braun employs. His reasoning, if compelling, would thus
directly show that certain uninstantiated non-particulars, those introduced by predicates of fiction, do
not exist. And this conclusion would then entail that all uninstantiable non-particulars introduced by
predicates of fiction thereby do not exist, i.e., his reasoning, if compelling, would indirectly show that
FUAR is true.
What is Braun’s main line of reasoning? Here is a paraphrase:
11
If fictional predicates have
as their semantic content a fictional non-particular, then, if ‘is a witch’ has content, it’s the non-
11
This is a paraphrase of the argument he calls the No Content Acquisition Argument (78). Braun also thinks
the argument applies to predicates introduced to us not only by works of fiction but by false scientific theories.
6
particular being a witch. But the content of ‘is a witch’ is the non-particular being a witch only if ‘is a
witch’ has been applied by at least one speaker to at least one entity that in fact instantiates being a
witch. But that has never occurred; there are no witches. So, fictional predicates do not have fictional
non-particulars as their semantic content.
As Braun himself is quick to point out (80), the believer in fictional non-particulars (a species
of properties he calls “empty properties”) has an immediate reply: reject the principle at the heart of
the argument and maintain that the “semantic content of ‘witch’ is simply […] an empty property that
nothing exemplifies” (80). In FUR-speak, one may the invoke the fictional universal witch-hood as
the semantic content of the predicate ‘is a witch’. The predicate has never in fact been applied by at
least one speaker to an entity that in fact instantiates witch-hood, but this says nothing about whether
this is an impossible scenario. Braun makes clear that while he thinks such a view would be
“problematic”, refuting it is something he has no interest in (80-81). That is why I think Braun’s
argument is not the strongest one in favor of FUAR; he merely sidesteps the response available to the
proponent of UR. That is, he begs the question in favor of FUAR by assuming that non-particulars,
generally conceived, are the sorts of things that must in fact be instantiated in order to exist.
12
4. Sawyer’s Argument for Fictional Universal Anti-Realism
Sawyer’s arguments for FUAR are direct, however, and they are primarily epistemic; if
compelling, they would show that both Fictional Realists and Fictional Anti-Realists (about fictional
particulars) alike lack the epistemic justification for accepting fictional non-particulars as the contents
of fictional predicates. Fictional Anti-realists such as Russell (1905), Currie (1990), and Adams,
Fuller and Stecker (1997) all either implicitly or explicitly help themselves to a FUR ontology when
presenting their views, but none provide the epistemic entitlement that’s needed to do so. There
clearly can be no sensory evidence for the existence of the universal having parents from planet
12
This is not a knock against Braun’s project per se. Many theorists would reject an ontology of abundant non-
particulars that would include uninstantiated non-particulars, and his entire project is aimed at one such theorist
who furthermore wishes to retain the basics of a certain theory of meaning (“The Naïve Theory”) while
providing a plausible semantics of fictional discourse. However, one wonders, then, how such a theorist would
treat ‘is a blue swan’.
7
Krypton, nor can one’s evidence be grounded in taking such fictional universals to one and all be
complexes of non-fictional universals that we do have evidence for. Having parents is presumably a
universal whose existence we believe in because offspring are directly observable, but no such
evidence exists for being from planet Krypton.
Fictional Realists likewise implicitly or explicitly help themselves to a FUR ontology, but any
reasons they have for believing in fictional particulars will fail to count as reasons for believing in
fictional universals. Quite the opposite, Sawyer argues: our lack of evidence for fictional universals
teaches us that Fictional Realism is itself unwarranted. Neo-Meinongian Fictional Realists such as
Parsons (1980) and Zalta (1983) rely on fictional universals to individuate the non-existent entities
they identify fictional particulars with, but they simply take fictional universals for granted. No
independent evidence is offered for the existence of entities that would be just as metaphysically
queer as the only sorts of particulars that could instantiate them. And Artifactualist Fictional Realists
such as Salmon (1998) and Thomasson (1999) also rely on fictional universals to explain essential
aspects of their views. The truth of metafictional sentences of literary criticism (e.g., ‘According to
the Superman stories, Kent works as a reporter for the Daily Planet’) requires fictional universals that
exist and get ascribed to a character by a work of fiction, and the explanatory work needed to generate
the falsity of certain transfictional sentences
13
(e.g., Superman is from Krypton) also requires the
existence of fictional universals that fail, of necessity, to be instantiated (Superman-cum-created-
abstractum does not, indeed cannot, hail from any planet).
In the end, argues Sawyer, both the Fictional Anti-Realist and the Fictional Realist build their
theories on the idea that there are universals of the sort that could only be instantiated by fictional
entities. But such non-particular entities, given their metaphysical queerness, are epistemically
unwarranted; hence, any ontology of fictional particulars, realist or anti-realist, that either implicitly
13
That is, sentences whose contents do not seem fictional, ones that seemingly are just plain true, not fictionally
true or true according to any fictional work. Transfictional sentences are ones that do not appear, at any rate, to
be elliptical for ‘according to fiction F, …’.
8
or explicitly relies on their existence will be ultimately unwarranted. No evidence points us toward
FUR; hence, FUAR is the better option.
14
I agree that such non-particulars would be metaphysically queer and hence objectionable,
especially if they are conceived not only as uninstantiated, but as uninstantiable, entities. But I don’t
think Sawyer’s argument is compelling, and I intend to explain why in Section 6. I also don’t think
her reasons are the most powerful ones in favor of FUAR either. They are the runner-up. Ultimately,
she attempts to make her case by appealing to either uninstantiated or uninstantiable entities of a
certain sort that we in fact lack evidence for. And she’s right, as far as her examples show. Being
from Krypton
15
is uninstantiable by anything non-fictional. But just because we lack evidence for
many fictional universals doesn’t establish that we should think that all fictional universals would be
relevantly like round square-hood. The defender of UR can make her case for extending her view to
FUR by appeal to instantiable entities that we have do evidence for. Some fictional universals, I will
argue below, are relevantly like blue swan-hood. And if we have reason to accept blue swan-hood,
which we ultimately do, then Sawyer’s case for FUAR is undermined. Whatever positive reasons the
UR defender has for such uninstantiated universals would automatically be the same sorts of reasons
she can employ for accepting FUR.
5. A Kripke-Inspired Argument for Fictional Universal Anti-Realism
There is one sort of argument for FUAR yet to consider, and by my lights, it is the most
powerful argument for rejecting fictional universals. It is an argument that no one in the relevant
literature has previously given (to my knowledge, anyway). However, it is one derived from a line of
reasoning sketched by Kripke (1980) and refined by Zvolenszky (2014), et al.
16
Here is what I have
in mind:
14
Sawyer ultimately favors a thorough-going anti-realism about both fictional particulars and non-particulars
along the lines of Walton’s (1990) pretense theory. Sawyer holds that we talk as if both such kinds of things
exist when playing certain games of make-believe.
15
For consistency, I’ve been sticking with Superman-fiction examples. Sawyer actually uses Pooh-Bear-fiction
examples, such as those that would be expressed by ‘is a tigger’ and ‘is a heffalump’.
16
Kaplan (1973, 505-8), e.g., uses it to show that ‘Pegasus’ cannot possibly refer to anything concrete. See also
Fine (1984, 126-8), Plantinga (1974, 154-5), and Yagisawa (2010, 271-77) for arguments that either names for
fictional particulars cannot refer to any entity that’s merely possible or that fictional particulars cannot be
9
i) If fictional names ever genuinely refer to particulars, what they refer to cannot be
concrete actual nor non-actual particulars.
ii) If what fictional names refer to cannot be concrete actual nor non-actual particulars,
then FUAR is true.
Therefore,
iii) If fictional names ever genuinely refer to particulars, then FUAR is true.
The above is valid by hypothetical syllogism. It is, of course, not strictly an argument for
FUAR, but it is one where it is intended that the conditional conclusion is vacuously true. Whether
Fictional Realism or Fictional Anti-Realism (about fictional particulars) is true or not, FUAR is the
case.
The rationale for premise (i) traces back to some remarks Kripke (1980, 157) makes
regarding unicorns. Consider any non-actual, concrete particular (a Meinongian non-existent, or a
Lewisian mere possibilium, say) having the sorts of features unicorns are said by the legends to have.
No such particular could in the end have just what it takes to be a unicorn. To be a unicorn, a
particular has to instantiate all the properties essential to them, but there are no such properties; there
are no actual, particular beasts which provide them, and our legends incorporating them are utterly
under-specific about what exactly they would be.
Zvolenszky (2014), expanding on some additional Kripkean commentary (1980, 157-8) that
seems to extend this line of reasoning to include the potential referents of fictional names such as
‘Sherlock Holmes’, provides more direct support for (i). She defends the general claim that “the
fiction-writing mode of introducing proper names into the language is unsuited for them to have as
their reference concrete, spatiotemporal objects, whether they be actual or [non-actual]” (462).
To make her case, she first discusses the mythical name ‘Vulcan’. Le Verrier hypothesized
that an intra-Mercurial planet, which he named ‘Vulcan’, explained perturbations in the orbit of
Mercury. But his hypothesis was eventually shown to be false; Einstein’s general theory of relativity
confirmed that the perturbations were produced by the gravitational field of the Sun. ‘Vulcan’, if it
refers to anything at all, does not refer to an actual, concrete planet. But more importantly, ‘Vulcan’
identified with mere possibilia. One of the main lines of reasoning, usually called The Selection Problem –
advanced, e.g., by Sainsbury (2010, 60-3) – for the claim that names for fictional particulars cannot refer to
Meinongian non-existents is relevantly similar.
10
cannot refer to any non-actual, concrete planet either (a Meinongian non-existent planet, or a
Lewisian merely possible planet, say). The only way Le Verrier could have succeeded in naming a
concrete planet is if his original hypothesis had been true, but it’s falsity shows he failed to name
anything concrete, actual or non-actual; to say otherwise is to embrace the absurdity that, by utter
coincidence, and completely contrary to the scientific intentions he had at the time, he succeeded in
naming some non-actual, concrete planet.
We can use this lesson to adopt the following principle about any proper name – ordinary,
mythical, or fictional: If a name doesn’t manage to refer to a concrete particular here in the actual
world, it cannot refer to any concrete, non-actual particular either.
17
Non-actual concrete entities
would, at best, be coincidentally similar to any concrete particular that may in fact be referred to,
given whatever relevant intentions would be used to introduce the name in question. If ‘Sherlock
Holmes’ refers, it cannot refer to a concrete pipe-smoking, deerstalker-cap-wearing, cocaine-using
detective. Any non-actual detective instantiating those characteristics would at best be coincidentally
similar to the particular Conan Doyle describes, and any actual detective would not be fictional, i.e.,
would not be in line with his authorial intentions. (Similar considerations apply to ‘Superman’,
‘Krypton’, ‘Cinderella’, ‘Pooh Bear’, ‘Tigger’, etc.)
Premise (ii), however, is the crux of the argument. The rationale for it is as follows. Suppose
the Kripke/Zvolenszky line of reasoning just rehearsed is correct. And further suppose, for the
purpose of reductio, that FUR is true. If it were, then the only sort of entities that could instantiate
fictional universals would be either concrete, fictional actualia or concrete, fictional non-actualia, that
is, the concrete particulars referred to by our fictional names. Just as only a concrete detective could
smoke a pipe, wear a deerstalker cap, and use cocaine, only a concrete superhero could work at the
Daily Planet, fall in love with Lois Lane, and hail from planet Krypton. But, as was just
demonstrated, there can be no concrete particulars referred to by our fictional names. It thus follows
17
She charmingly names this the Inverse-Sinatra principle for Proper Names (581). Being an Artifactualist
about fictional particulars, Zvolenszky in the end maintains that ‘Superman’ does refer to an actual, created
abstractum, so it does not follow on her view that ‘Superman’ refers nowhere else. Both Fictional Realists and
Anti-Realists, on her view, ought to accept the Inverse-Sinatra Principle.
11
on the Kripke/Zvolenszky line of reasoning just rehearsed that (the above conditional of which)
FUAR (is a consequent) is (vacuously) true.
6. In Defense of Fictional Universal Realism
Various Fictional Realists
18
will be keen to deny premise (i). I do not wish to enter those
philosophical waters here. Rather, I wish to focus on premise (ii); this is where the argument fails.
FUR, recall, is the thesis that some universals are fictional universals. Premise (ii) only seems
reasonable when one narrowly focuses on fictional predicates that are in some way phrased using
fictional names, such as ‘is from Krypton’.
19
It’s true that a predicate like this could only be
instantiated by fictional particulars such as Superman, Jor-El, and General Zod, and so if one
conceives of all fictional universals as being of this sort, one could be misled into thinking (ii) is true,
and even that the argument of Section 5 is sound. But there are other sorts of fictional predicates, and
some of these express fictional universals that are possibly instantiated. The sorts of fictional
universals I have in mind are those that are expressed by fictional predicates that are not phrased
using names for fictional particulars or fictional groups; they’re universals that would be instantiated
by non-fictional entities.
To see this, first consider Scenario 1: Paulina authors a fiction in which, among a great many
sentences about things both fantastical and mundane, the sentence ‘Richard Nixon is 30’ tall’ is
found. This would be a case where a fictional work seemingly ascribes to a non-fictional particular
the non-fictional universal being 30’ tall. Next, Scenario 2: Paulina authors a fiction in which, among
a great many sentences about things both fantastical and mundane, the sentence ‘Richard Nixon is
from the planet Krypton’ is found. This would be a case where a fictional work seemingly ascribes to
a non-fictional particular the fictional universal being from planet Krypton. However, if we assume
the above Kripke/Zvolenszky line of reasoning about fictional particulars is correct, since ‘is from
18
I have in mind here Neo-Meinongians such as Parsons (1980), or defenders of a view relevantly similar to the
one suggested by Lewis (1978).
19
A predicate uses a name for a fictional particular (such as ‘Superman’ or ‘Krypton’) or a name for a group of
individuals (‘is a unicorn’, or ‘is a heffalump’) when that name was introduced via some author’s fictional
intentions, that is, an author’s use of make-belief that ‘Superman’ or ‘heffalump’ is genuinely referential.
12
Krypton’ uses the name for the fictional particular Krypton (something not possibly concrete, so not
possibly a place anything may truly hail from) the universal in question would be uninstantiable.
Hence, ‘is from Krypton’ fails to express any universal at all. Now, Scenario 3: Paulina authors a
fiction in which, among a great many sentences about things both fantastical and mundane, the
sentence ‘Only Superman is flurgish’ is found. According to the details of Paulina’s work of fiction,
we learn that something is flurgish if it is a 30’ tall person. Until this point in history, ‘is flurgish’ had
not been a part of English. Paulina introduces us to this predicate in the authoring of her fiction and
ascribing it only to Superman. Assuming the above Kripke/Zvolenszky line of reasoning about
fictional particulars is correct, Paulina’s fiction is partly about something that is not merely (extra-
fictionally) false, but (extra-fictionally) impossible; ‘Superman’ can’t name a concrete particular, so
can’t name a 30’ tall person, and so can’t name the sort of thing that instantiates (extra-fictionally)
being flurgish. The fictional predicate in question here plausibly expresses no universal at all.
Finally, Scenario 4: Paulina authors a fiction in which, among a great many sentences about things
both fantastical and mundane, the sentence ‘Only Richard Nixon is flurgish’ is found. Until this point
in history, ‘is flurgish’ had not been a part of English. Paulina introduces us to this predicate in the
authoring of her fiction and ascribing it only to Richard Nixon. According to the details of Paulina’s
work of fiction, we learn that something is flurgish if it is a 30’ tall person.
I claim that it is open to the defender of UR to maintain that in Scenario 4, ‘is flurgish’
expresses a fictional universal, one that is (extra-fictionally) instantiable. No (non-imported) fictional
character could be flurgish (outside the fiction), but Nixon could, and so could Aristotle and Michael
Jordan, even though neither is anywhere discussed in Paulina’s fiction. When a fictional predicate is
not phrased using a name for any (non-imported) fictional particular or group, that is, when a
predicate introduced to us via fiction linguistically attaches in that fiction only to non-fictional
imports into the story, we can see that we have no grounds for accepting premise (ii) of the argument
above. In this case, the non-fictional particular Nixon instantiates within the story the fictional
universal being flurgish. But the story describes something that is possibly (extra-fictionally) true;
Nixon could have been 30’ tall. The content of Paulina’s story includes this modal fact, just like some
other works of fiction, such as the Sherlock Holmes stories, involve the non-modal fact that London is
13
in England. Provided the fictional universal in question is not tied by the predicates used in the story
to any names for fictional particulars, instantiation of it (outside the story) is possible.
I maintain that a universal like being flurgish is relevantly like blue swan-hood. It may not be
instantiated, but it’s instantiable. Blue swan-hood was not introduced to us via a fictional predicate,
so it’s not a fictional universal. But being flurgish was, we are supposing, introduced to us via
Paulina’s fiction (in Scenario 4) and the predicate she coined therein. Even if the universal is never
instantiated, we understand perfectly well what it would take for it to be instantiated. With blue
swan-hood, we use what we know of blue-hood and swan-hood to understand what’s required for
instantiation, and with being flurgish, we use the contents of Paulina’s fiction. What’s fictionally true
here spills over into “real life”; her fiction tells us just what it would take for a certain kind of
universal to be instantiated. The kind of specification Kripke/Zvolenszky demand for telling us when
we have a unicorn or Sherlock Holmes is irrelevant here. Paulina’s story tells us everything we need
to know concerning when something is flurgish.
To further drive the point home, suppose, contrary to fact, that Juvenal in 100 CE authors a
satirical poem according to which cross-global communication occurs nearly instantaneously between
two people by way of holding up a blackish rectangular device to one’s ear. Use of such a device is
not attributed to anyone in his poem, and the device itself is not given a name. His poem merely uses
predicates novel to Latin to explain such a device. In that case, communicating by way of… would be
a fictional universal, one first expressed by the Latin predicates coined by Juvenal, and one that is
instantiated nowadays by mobile phones.
Of course, if any of us were to meet a pipe-smoking, deerstalker-wearing, cocaine-using
detective, we would not be meeting Sherlock Holmes. If we met someone sincerely claiming to hail
from planet Krypton, we may justifiably think him insane. But upon seeing someone using a mobile
phone in 2021, we should, in the counterfactual scenario just imagined, judge it to be the case that the
fictional predicate Juvenal introduced to the Roman world via his poem has a non-null extension. We
should judge the fictional universal here to be instantiated.
While premise (i) of the argument presented in Section 5 may be true regardless of the
ontological status of fictional particulars, I maintain that premise (ii) is false regardless of the
14
ontological status of fictional particulars. Contra Sawyer, questions of fictional universals are indeed
separable from one’s ontology of fictional particulars. If the only thing that could instantiate a
fictional universal were a fictional particular or a fictional group, then I would agree that her
complaints about the metaphysical queerness of such entities would apply. But when predicates are
not expressed in fictions in ways that use names for fictional particulars or groups, they may express
fictional universals. The case that Sawyer, Braun, and the proponent of the argument from Section 5
each makes for FUAR is only apparently compelling due to the particular fictional predicates chosen.
7. A Response and Some Concluding Remarks
I have critically assessed three lines of reasoning for FUAR. The most powerful one (by my
lights, anyway) is intended to show that fictional universals fail to be real regardless of one’s views
about fictional particulars. However, not being able to deliver an appropriate particular is no knock
against the view espoused here by the defender of FUR about some fictional predicates. Fictional
predicates can, on some occasions, deliver a set of entities as an extension, thereby meeting the
criterion for the existence of a corresponding universal.
I imagine a natural sort of reply to this defense of FUR would be the following. The
reasoning I have provided here on behalf of an adherent of UR fails to employ a proper understanding
of the nature of fictional entities. ‘Being flurgish’, if it expresses a universal at all, expresses a non-
fictional universal, one that gets imported into Paulina’s story in the same way that the non-fictional
particular Richard Nixon gets imported. According to UR, all universals are necessarily existing, sui
generis abstracta, so it trivially follows that being flurgish, if it exists, is a non-fictional import, not
one that Paulina used her authorial activities to genuinely bring about. She was being quasi-creative
when she coined ‘is flurgish’, but that was simply employing a bit of new terminology to express a
universal that we already were able to express using other well-known predicates. And Paulina may
not have intended to import being flurgish into her story in the way she intended to import Nixon, but
she accidentally succeeded in doing so. On the other hand, if being flurgish fails to be identical with
being a 30’ tall person (or any other instantiable non-particular), then Paulina’s fictional predicate
15
does not, indeed cannot, have a non-null extension. Her fictional intentions would ensure that the
predicate could not apply to anything.
This reply misses the mark. The defender of UR will assert that not all entities imported into
a fiction are, of necessity, non-fictional. Many influential Fictional Realist theories (of fictional
particulars), viz., Neo-Meinongianism and Possibilism, defend the claim that all fictional particulars
are strictly discovered by our authorial activities and are hence strictly imported into our works of
fiction. Authorial creativity is thus understood not as bringing into existence but rather
“fictionalizing”.
20
(This aspect of these versions of Fictional Realism is precisely what has turned
some instead toward ontologies that involve authors genuinely bringing into existence fictional
particulars via story composition.)
But accepting that Paulina’s authorial activities fail to be appropriately creative is not on the
table for the defender of UR. According to UR, universals are indeed abstract and exist necessarily,
and so Paulina is being creative with respect to authorship of her fiction involving being flurgish even
though she is, perhaps unwittingly, importing the universal into her story. If an author intends to “cut
a fictional character from whole cloth”, it’s highly likely she’ll be able to succeed, regardless of the
correct ontology of fictional particulars. Similarly, if one intends to “cut a fictional universal from
whole cloth”, it’s highly likely she will succeed. In this context of assuming the truth of UR, an
author cuts a universal from whole cloth and is appropriately creative when she incorporates into her
story some universal via the coining of novel, instantiable predicates, that is, ones introduced to us by
her fiction that express that universal.
I conclude that not all fictional predicates are on a par. Some genuinely express fictional
universals, some may not. If a fictional predicate is expressed using a name for a fictional particular
or group, then it may,
21
of necessity, have a null extension, and hence, according to UR, would
express no universal whatsoever. Braun, Sawyer, and the defender of the Kripke/Zvolenszky-inspired
argument in Section 5 make their case for FUAR by focusing solely on fictional predicates that
20
See, e.g., Parsons’ (1980, 188) discussion of authorial creativity.
21
That is, if the Kripke/Zvolenszky line of reasoning in Section 5 is sound when restricting predicates in this
way.
16
plausibly cannot have an extension. One would thus be forgiven for thinking along with them that
any corresponding universal would fail to exist. But according to the thesis I have defended here,
while ‘is Superman’ is analogous to ‘is a round square’, there are some fictional predicates that can
refer, hence there are some fictional predicates that express fictional universals. They are no more
metaphysically queer than blue swan-hood. And if I’m correct, we have reason to further reject
Sawyer’s contention that universal realism necessarily tethers you to various claims about the nature
of fictional particulars.
I say that accepting FUR dictates nothing about which ontology of fictional particulars is
most plausible. For you – adherent of UR – nothing stands in your way of adopting the further idea
that some fictional predicates are meaningful in virtue of expressing a fictional universal. That is,
there is nothing additionally peculiar about such universals that prevents you from extending your
view to FUR.
References
Adams, F., Fuller, G., and Stecker, R. 1997. The semantics of fictional names. Pacific Philosophical
Quarterly 78: 128-48.
Braun, D. 2015. Wondering about witches. In Brock, S. and Everett, A. (eds). Fictional objects.
Oxford:
Oxford University Press, 71-113.
Currie, G. 1990. The nature of fiction. Cambridge: Cambridge University Press.
Fine, K. 1984. Critical review of Parsons’ Non-existent objects. Philosophical Studies 45:
95–142.
Kaplan, D. 1973. Bob and Carol and Ted and Alice. In K.J.I. Hintikka, J.M.E. Moravcsik, and P.
Suppes
(eds.) Approaches to natural language. Dordrecht: D. Reidel, 490–518.
Kripke, S. 1980. Naming and necessity. Cambridge, MA: Harvard University Press.
Lewis, D. 1978. Truth in fiction. American Philosophical Quarterly 15: 37-46.
Parsons, T. 1980. Non-existent objects. New Haven, CT: Yale University Press.
Plantinga, A. 1974. The nature of necessity. Oxford: Clarendon Press.
Priest, G. 2005. Towards non-being. the logic and metaphysics of intentionality. Oxford: Clarendon.
Russell, B. 1905. On denoting. Mind 14: 479-93.
17
Sainsbury, M. 2010. Fiction and fictionalism. New York, NY: Routledge.
Sawyer, S. The importance of fictional properties. In Brock, S. and Everett, A. (eds). Fictional
objects.
Oxford: Oxford University Press, 208-29.
Salmon, N. Non-existence. Nous 32: 277-319.
Thomasson, A. 1999. Fiction and metaphysics. Cambridge: Cambridge University Press.
van Inwagen, P. 1977. Creatures of fiction. American Philosophical Quarterly 14: 299-308.
Walton, K. Mimesis as make-believe. Cambridge, MA: Harvard University Press.
Yagisawa, T. 2010. Worlds and individuals, possible and otherwise. Oxford: Oxford University Press.
Zalta, E. Abstract objects: an introduction to axiomatic metaphysics. Dordrecht: D. Reidel.
Zvolenszky, Z. 2014. Artifactualism and Authorial Creation. Proceedings of the European Society for
Aesthetics 6: 457–469.
