Ingarden vs. Meinong on Ficta’s Generation and Properties

I

Does Sherlock exist? Is he a genius detective? The answer to the latter depends on our answer to the former. In order to say anything about Sherlock, we have first to determine his existential status. At least this is how most non-eliminativist theories of ficta would go about this issue. Alexius Meinong begged to differ. According to him, we can talk about ficta’s properties independently of their existence. In other words, although ficta are non-existent, they have certain properties. In Meinongian terms, the Sosein (being so-and-so) of a non-existent entity is independent of its Sein (existence). This is what Meinong termed the ‘independence principle’. Meinong also speaks of the ‘indifference principle’. According to him, ‘the Object is by nature indifferent to being [außerseiend], although at least one of its two Objectives of being, the Object’s being or non-being, subsist’.¹ In application, the indifference principle states that existence or non-existence are not essential parts of Sherlock. It would be wrong, however, to understand this principle as entailing that Sherlock possesses both existence and non-existence.² These two principles are the key in Meinong’s theory of objects, and it is these two principles on which analytic neo-Meinongians primarily build their reformulations of Meinong’s position.

Richard Routley describes Meinong’s ‘independence principle’ as laying the groundwork for his own characterization postulate. Routley posits that Meinong’s principles pave the way for a view of nonentities (his version of Meinong’s objects), in which they are characterized by their properties irrespective of their existential status. Following Meinong, Routley stresses that ‘existence is not a characterising feature’ and that we cannot determine the existential status of nonentities using characterization alone.³ Indeed, this is what Meinong posited, underlining that ‘existence is not a constitutive nuclear property that qualifies the Sosein of any object’.⁴ Further, Meinong’s independence principle entails the ‘denial of the ontological assumption’ – namely, the assumption ‘according to which no (genuine) statements about what does not exist are true’.⁵ By and large, Routley’s characterization postulate foregrounds Meinong’s emphasis on properties and the role they play in determining the nature of nonentities. However, Routley’s postulate raises numerous issues.

Given that Sherlock is a nonentity, how is it possible that he possesses many properties? Considered prima facie, two hypotheses come to mind as to how non-existent ficta possess properties. We can say that Sherlock possesses the property of being a genius detective due to ‘language’ or ‘thought’.⁶ As Voltolini posits, it can be argued that nonentities have their properties through ‘linguistic ascription’, hence giving language the power of ascription. As Bertrand Russell showed, ascribing this power to language can render the object described contradictory (for example, the existent present King of France would be forced to exist). Another challenge lies in ascribing to language the power of generation. Routley’s postulate suggests that we obtain a different Meinongian object by changing its description. This is the point with which Russell found issue, and which he tried to amend with his theory of definite descriptions.⁷ Meinong himself would not give language either of these powers. According to him, an object is designated by a linguistic term if it outlines a presentation, whose content then presents the object.⁸

What about thought? Can we say that thought ascribes properties to nonentities? We can argue that the thoughts underlying language are responsible for nonentities having properties. This would make nonentities intentional. The generative power of thought can be explained by resorting to Franz Brentano’s intentionality, in which every thought is directed at an object, existing or otherwise.⁹ Moreover, intentional objects are conceived of as having certain properties that distinguish them from other intentional objects. One can argue that Meinong would have gone beyond Brentano to assert thought as having the power of ascription as well. The resultant entities would be intentional objects that possess the properties they are conceived of as having. This formulation is what Voltolini terms the ‘Meinongian phenomenological conception of außerseiende beings’, according to which Meinong’s objects are beyond existence and non-existence, endowed with the properties they have following their being thought of as such. Accordingly, ficta would be a subset of außerseiende intentional objects, introduced into the domain of being by an act of imagination. This account can be labelled a Meinongian phenomenological theory of ficta.¹⁰ A sceptic of außerseiende entities would argue that the problems faced by language possessing generative or ascriptive powers also inflict thought. How can thought bring into being a fictum with properties? Whether it can or not, it is safe to say that Meinong did not adopt a phenomenological view of außerseiende entities. According to him, thought cannot have generative or ascriptive powers either. The intentional acts directed at nonentities only grasp independently constituted objects, and the properties they possess are independent of their grasping intentional acts.¹¹ Following Meinong, many neo-Meinongians have tried to reformulate his position so as to avoid the aforementioned problems.¹²

As Routley’s principle rightly emphasizes, it is properties that define Meinong’s objects. It would not then be off the mark to state that determining the status of ficta properties is tantamount to determining ficta’s status. This is what the Principle of the Freedom of Assumption (PFA) puts forth – namely, a view of non-existent entities that draws from their properties instead of the thoughts underlying their descriptions. We can speak of two versions of this principle: weak and strong. The weak version states that, for every collection of properties, there is a non-existent object that has all of them. In Meinongian terminology, this object-generation principle states that an object corresponds to every being-so.¹³ The strong version states that, for every collection of properties, there is one and only one object that has all and only those properties.¹⁴ This way of putting Meinong’s account turns him into a Platonist. That is, properties are postulated to be out there, and außerseiende objects’ being depends on them. Further, it can be argued that the manner by which the PFA generates Meinongian objects is trivial. ‘A certain Meinongian object is, so to speak, made to belong to the overall ontological domain by the fact that it possesses all the properties involved in any instantiation of the Principle.’¹⁵

So far, two possible readings of Meinong have been laid out. The Platonist reading is more akin to the historical Meinong. The PFA makes Meinongian objects’ generation as having their properties less mysterious but it cannot explain how the overall domain of außerseiende objects can have a subset of ficta – for this reading only explains how außerseiende objects are generated without specifying the generation of subsets within this domain. The phenomenological reading, on the other hand, provides us with a theory of ficta as außerseiende objects of imagination. But this view inexplicably assigns to thought the powers of generation and ascription.¹⁶

Given that Meinong’s historical account does not show how ficta fit within the domain of außerseiende objects, it would be wiser to further pursue the phenomenological theory. To do so, the problems of generation and ascription need to be sorted out. I think what Voltolini is trying to achieve by reading Meinong phenomenologically is similar to what I am trying to defend here – namely, a phenomenological-ontological account of ficta properties. If anything could save Meinong’s account, it would be the phenomenological-ontological reading. But Meinong himself did not take this route. So it seems that we have reached an impasse. We can either explain how Meinong’s entities possess properties and risk ending up with no account of ficta, or we can have an account of ficta and give problematic ontological powers to thought. I choose the latter alternative. In order to give substance to such a view and moreover to avoid the problems identified by Voltolini, I will in the following defend an Ingardenian phenomenological ontology of ficta.

II

In Ingarden’s view, ficta are generated by the author’s creative acts. These acts are imaginative acts, which is a specific kind of intentional act (Phantasie). The generated objects are purely intentional objects.¹⁷ So, imagination is what Ingarden describes as possessing the power of generation. One can raise an objection to this – namely, would not such a view invite too many entities? How can I be sure that the Sherlock I am imagining is the same Sherlock someone else is imagining? Let us say I am imagining ‘Sherlock’. I am picturing in my head a white man dressed elegantly in nineteenth-century clothing. The famous hat is there, too. I am imagining a man as described by Arthur Conan Doyle, the creator of Sherlock, whose imaginative acts first brought him to life. Without the properties ascribed to Sherlock by Conan Doyle, I would not be able to imaginatively distinguish Sherlock from, say, Dr Watson. So, properties are not only needed to determine ficta’s identity criteria within and across written works; they are also needed to determine ficta’s identity criteria within imagination itself. The fact that we cannot imagine, and thereby generate, a fictum without associating it with its properties shows that Meinong was right in foregrounding properties. However, depicting ficta as non-existent entities that are generated independently of thought misses the mark. For one, how can we ground Sherlock’s properties? As I pointed out above, imagining one and the same Sherlock, or at least staying as faithful as possible to the original Sherlock, requires us to refer to his properties as described by Conan Doyle. Since Meinong did not subscribe to creationism, the question of grounding ficta’s properties remains unresolved.

Following Ingarden, grounding ficta’s properties is fairly straightforward. Being products of imagination, ficta lack an essence of their own. In Ingardenian terms, they are ‘heteronomous’ entities. A heteronomous entity is ‘an entity which draws its being and its collective stock of attributes from the enactment [Vollzug] of an intentional conscious experience, which in a specific integrated fashion is endowed with a content, and it would not exist at all without this enactment’.¹⁸ The concept of ‘heteronomy’ is contrasted with ‘autonomy’. ‘An entity […] exists autonomously […] if it has its existential foundation within itself.’¹⁹ The creative acts that bring about ficta are existentially autonomous. Such a formulation, I argue, can avert Voltolini’s criticism of thought as having the unfathomable ontological powers of generation and ascription. To be more explicit, the powers of generation and ascription are problematic because they need to be grounded in something that is not in need of further grounding. Meinong and neo-Meinongians seem to ground generation and ascription in the jungle of non-existence. Apart from the other logical problems associated with it,²⁰ this leads to a circularity problem. Meinong seems to ground generation in ascription and ascription in generation. Put differently, nonentities are generated as possessing their properties, and they possess their properties because they are generated as so. Ingarden, on the other hand, would argue that both the generation and the properties of ficta are grounded in intentional acts. Before saying more about ficta’s properties à la Ingarden, ficta’s generation as purely intentional entities calls for a more concentrated analysis.

Ingarden’s phenomenological-ontological account proves to be superior. This is so primarily because Ingarden grants authors a role in creating ficta. Meinong took a weird turn and raised his ‘principle of the unlimited freedom of assumption’ in defence of ficta’s Platonic origins: ‘In regard to every genuine or, so to speak, ordinary determination of so-being, it is in my power, according to the principle of unlimited freedom of assumption, to pick out – by means of adequate intention – an entity which in fact has that determination of so-being.’²¹ In fact, adhering to the logic of the excerpt above, the concept of ‘generation’ does not fit Meinong’s account at all. If properties only ‘pick out’ ficta from an ever-existing realm, then it is pointless to talk of generation. A problem that concerns genuine Meinongians pertains to the criteria by which ficta are selected, not generated. Put differently, it is unclear how an author selects a certain fictum on the basis of its property set. Sainsbury raises an interesting suggestion by maintaining that, although ficta have no beginning in time, we can postulate a time at which non-existent objects’ Sosein is subjected to ‘enrichment’. This is precisely what the creative process of authors adds to Meinongian objects. Prior to this process, there was already a non-existent object but it only possessed a ‘minimal Sosein’. The creative process equips nonentities with more properties. A major problem with this suggestion is that there are no clear criteria that would allow authors to add properties to a certain object. This brings us again to the ‘selection problem’.²² It is hard, if not hopeless, to treat ficta seriously without associating them with their authors. Sherlock did not just appear ex nihilo, miraculously possessing the properties of the famous Sherlock. It was in 1887 that Conan Doyle published the first story containing Sherlock, thus giving birth to the famous character.²³ In fact, even some neo-Meinongians have conceded authorial creativity. Berto, for instance, argues for a refined Meinongianism, in which ficta are not selected but imagined to be so-and-so.²⁴ Terence Parsons and Edward Zalta, attempting to regulate our discourse about ficta properties, have provided Meinongian accounts of ficta, in which their properties are bound by their authors’ works.²⁵

It can be argued, contra Ingarden, that his account of ficta generation leads to ficta that ontologically appear and disappear. That is, Sherlock only exists as long as an imaginative act is intentionally directed upon him. Once the act fades, Sherlock fades with it.²⁶ To address this argument, we need to take a closer look at Ingarden’s imagination theory. In his ‘Essentiale Fragen’,²⁷ Ingarden delves deeper into objects’ creation (the following is Witold Płotka’s reformulation):

These conditions illustrate that ficta’s pure intentionality differs from Sartre’s and Brentano’s understanding of intentionality. Ingarden’s imagination is an imagination extended in time. The phenomenological complexity of its acts leaves no room for problematic entities that ontologically flit in and out. Another argument in favour of Ingardenian intentionality can be made in reference to his system of dependencies.²⁹ Ficta come to life after their author creates them. This is what Amie Thomasson designates ‘historical dependence’. Sherlock is dependent on the creative acts of Conan Doyle and not, say, those of Virginia Woolf. In Thomassonian terms, Sherlock is ‘rigidly historically’ dependent on Conan Doyle’s creative acts. In addition, ficta enter into a seemingly dependence relation with literary works. Thomasson labels this ‘generic constant dependence’. A fictum depends constantly for its subsistence on the existence of some literary work about it. This dependence is merely generic because any work about it would do.³⁰ According to Ingarden, ficta’s ‘foundation’ is embedded in literary works. He distinguishes between two senses of foundation: ‘immediate’ and ‘derivative’. On the one hand, ficta have their immediate existential foundation in the sentences that carry their descriptions. The literary work’s sentences, on the other hand, have their immediate existential foundation in the creative acts of the work’s author, which makes ficta’s existential foundation derived.³¹ Because they are what stands between the creative acts of an author and the literary work’s life, sentences have a ‘borrowed intentionality’, which gives ficta a relative independence from their generating acts.³²

A clarification of ficta’s persistence is in order. Two main elements can be distinguished within linguistic intentionality. First, a sentence’s intentionality has the ‘source’ of its existence in the intentional acts of the author, whereas the ‘basis’ of its existence lies within ideal concepts/qualities and word signs.³³ What guarantees ficta’s persistence are ideal concepts and their meaning content. According to Ingarden, ‘[i]t is only with reference to the meaning content of ideal concepts that the readers of a literary work can reactualize in an identical manner the meaning content of sentences given to them by the author’.³⁴ But this raises a further challenge: what is the relation between language and consciousness? If we assign linguistic intentionality to subjective consciousness, would we not be stuck again with psychologism? To avoid this turn, Ingarden sought to answer the following: ‘Are the sentences that arise from subjective operations intersubjectively identical? Do they also exist when they are not thought? What is their mode of existence and the ontic basis of their existence if they do exist?’³⁵ In response, he advocated a metaphysical solution, conceding the existence of ‘ontically autonomous ideal concepts’. By disregarding language as the reference point of intersubjectivity, Ingarden shields the literary work’s mode from the threats of psychologism. What sentence-forming operations add to the literary work’s mode is the actualization of the meaning components concerning ideal concepts.³⁶ In short, linguistic intentionality has one basis in authorial intentionality and another in ideal concepts. The latter transcend both subjective sentence-forming operations and the meanings conferred on them. Structured as such, ideal concepts are the ‘regulative principle’ that ensures language’s intersubjectivity.³⁷ It is, therefore, ‘impossible to achieve between two conscious subjects genuine linguistic communication’ without the admission of ideal concepts.³⁸ Through language, these objects become intersubjective, but only under the condition that language itself is not dependent on a subjective consciousness.³⁹

Ingarden also speaks of (Sartrean/Brentanian) imaginary entities but he attributes their existence to a different group of creative acts. He makes a distinction between two different groups of acts.⁴⁰ The first group is responsible for the generation of ‘free’ fantasies, for example reading into a cloud shape a specific entity. This group of acts is satisfied with creating intentional entities that vanish with their generating acts. The acts of the second group take it upon themselves to bring into being purely intentional entities that are over and above their generating acts. To elaborate, two variants within the second group need to be clarified. One variant strives to make the generated intentional entities lasting, relying on an existentially stronger basis that would ensure they outlive their generating acts.⁴¹ This transforms the generated entity from a purely subjective entity into an intersubjective objectivity (Objektivität), which can be accessible by distinct acts of consciousness.⁴² The second variant is comprised of acts that bring about intentional entities serving as ‘models’ (Vorbilder) or ‘blueprints’ for certain autonomous entities to embody.⁴³ By and large, Ingarden’s purely intentional account of ficta differs greatly from Sartre’s and Brentano’s imaginary views. And not just that: Ingarden’s intentionality can be devised to avert the many problems plaguing the view of ficta as imaginary objects.

III

It is time to proceed to the ‘ascription’ problem. According to Voltolini, a phenomenological conception of ficta ascribes to thought an inexplicable ascriptive power, whereas a Platonic view of Meinongianism successfully formulates a view of non-existent objects as being characterized by their properties. In this section, I will show that (1) a phenomenology of properties is not problematic, and (2) a Meinongianism of properties raises many issues that can be remedied with the use of Ingarden’s phenomenological ontology.

Neo-Meinongians striving to save Meinong’s account adopt two different strategies. They are on the same page when it comes to ficta’s characterization by reference to their properties but they disagree over ficta’s status. According to some, ficta are not a subset of außerseiende entities; they are instead abstract entities. That is, unlike Meinong, they believe that ficta subsist (exist as abstracta). One neo-Meinongian approach depicts fictional entities as corresponding to the set of all the properties included in a fiction. This approach can be construed as entailing that a fictum is a set-correlate rather than a pure set.⁴⁴ Other neo-Meinongians consider ficta to be akin to generic objects like ‘forms’ or ‘blueprints’.⁴⁵ To elaborate, in an abstractionist view of ficta that bears some resemblance to Meinong, Zalta suggests we view ficta as Platonic forms. To defend his view, Zalta invokes Plato’s ‘one over the many principle’, according to which ‘if there are two distinct F-things, then there is a Form of F in which they both participate’. Forms can be considered as some sort of A-object (abstract object) that makes the form of an object G any A-object that encodes G. Zalta explicates his view in ‘participation’ terms. It can be stated that an object participates in the form of G if the form encodes a property that the object exemplifies. For instance, every object that exemplifies the colour red participates in the form redness.⁴⁶ In the same vein, Rapaport depicts M-objects (Meinongian objects) as ‘blueprints’:⁴⁷ ‘A blueprint of a house is to an M-object as a house of which it is a blueprint is to an actual object corresponding to the M-object. Just as the house need never be built or many houses may be built from the one blueprint, so there might be no or many actual objects correlated with an M-object.’⁴⁸ In short, neo-Meinongians mostly adopt two approaches to ficta’s characterizing properties: a set-theoretical approach and a Platonic approach. Both approaches follow Meinong’s PFA but differ from him as regards ficta’s mode.⁴⁹

The set-theoretical and Platonic approaches assume that ficta are characterized by their properties. But ficta do not only possess properties that the story assigns to them; otherwise, many properties that ficta do possess would not be admitted. For instance, the property Sherlock is created by Conan Doyle is not included in the fiction about him. It is, nonetheless, a property of Sherlock; a rather important one. How can we, then, determine which properties Sherlock really possesses? To answer this, neo-Meinongians appeal to two distinctions: ‘kinds of property’ and ‘modes of predication’.⁵⁰ The kinds of property distinction is mostly adopted by proponents of ficta as set-theoretical abstracta. The roots of this distinction can be found in Meinong. According to him, a fictum possesses ‘constitutive’ and ‘extraconstitutive’ properties.⁵¹ The first property kind is applicable to the properties that a fiction assigns to the fictum (for example, Sherlock’s being a genius). The second kind concerns the properties that it possesses outside of the fiction (for example, Sherlock’s being fictional).⁵² As elegant as it is, this distinction raises some serious issues. For starters, how is it that an ordinary man shares the same property kind with a fictum? Bearing in mind the latter’s abstract status, how is it that Hamlet is just as much a prince as William, Prince of Wales is?⁵³

Cornered by such concerns, many neo-Meinongians, following Ernst Mally, part ways with the property kinds distinction and adopt instead the ‘modes of predication’. This proposal puts forth a view of properties, in which they are of one kind, regardless of the entity to which they accrue, differing only in their predication mode. Objects are said to possess properties either internally or externally.⁵⁴ Accordingly, ficta possess internally the properties predicated of them inside the fiction, and externally the properties predicated of them outside of the fiction.

Proponents of the property kinds approach argue that a fictum’s property is nuclear iff it belongs to the set that constitutes it. Consider the following:

(1) Sherlock is a fictional entity.

This proposition states an essential property of Sherlock but, adhering to the property kinds, this property cannot be counted as one of Sherlock’s constituent properties. Being fictional is an essential property of Sherlock, even if it is not attributed to the fictum. It is taken for granted that the author ‘unofficially pretends’ (borrowing a Waltonian concept⁵⁵) that fictional characters are ordinary-like, where in effect they are merely fictional. There is no need to officially add the essential property of being a fictum to Sherlock to regard him as a fictum. The property kinds distinction, therefore, does not adequately capture this. Turning to the predication modes distinction, the property of being fictional is essential, and for that it cannot be eliminated as a different kind of property not belonging to the set that constitutes Sherlock. Being fictional is then attributable externally. Following Meinong, it can be argued that the property kinds distinction can accommodate such examples. Meinong would argue that for every extranuclear property there is a watered-down (depotenzierte) nuclear property corresponding to it.⁵⁶ However, as Voltolini indicates, this reasoning seems to be ad hoc. One gets the impression that it is only pursued for the sake of saving the distinction. To highlight the inadequacy of watering-down, consider the following:

(2) Once upon a time there was both a fictional and a watered-down fictional object.

Proponents of property kinds would argue that being fictional is the watered-down property corresponding to the extranuclear property of being a fictum. This, however, would mean that the other property, being a watered-down fictional object, cannot be identical with the watered-down nuclear property. Consequently, it must be another watered-down nuclear property. This form of reasoning commits us to an infinite regress.⁵⁷

Ingarden’s approach to the aforementioned examples would be relatively straightforward. Ingarden acknowledges that a fictum is endowed with a double-sidedness. A fictum is a purely intentional object and, as such, it has two sides: a ‘content’ (Gehalt)⁵⁸ and an ‘intentional structure’:

As regards (1), being fictional is an essential property of Sherlock. Ficta’s pure intentionality entails that being fictional belongs to Sherlock’s intentional structure. Concerning (2), Ingarden would argue that being a fictum is included in both the fictum’s content and intentional structure, with both structures reporting back to authorial intentionality. Being a watered-down fictum would be admitted to the entity’s content. Put briefly, add as many watered-down properties as you like; that would not change anything in ficta’s double structure. We cannot have an infinite regress of properties in an Ingardenian framework, for both sides of the purely intentional object are reducible to their generative intentional acts.

Parsons, too, advocates a view of ficta as being doubly structured. However, his version of the view lacks Ingarden’s sophistication. Parsons posits a view of properties categorized into two classes: set-theoretical properties (corresponding to the metafictional level) and properties determined by the sentences of the fictional work. Parsons also holds real objects to conform to this double structure, ignoring the borderline separating real and fictional objects. Ingarden, by contrast, defends a view in which Parsons’s set-theoretical properties consist of formal and intentional properties that ficta possess as such, without entailing their existence. In addition, Ingarden reinforces the line separating real and fictional objects, and specifies the mode in which ficta and real objects possess properties.⁶⁰ Indeed, Ingarden distinguishes between ‘strict’ and ‘intended’ properties.⁶¹ Real objects can only possess properties strictly. Ficta possess both strict and intended properties. This distinction is particularly helpful with determining the truth-value of fictional propositions. Consider Sherlock is a genius detective. Adhering to Parsons’s set-theoretical view, Sherlock’s property would be the same property possessed by a flesh-and-blood detective. Ingarden would rather attribute to Sherlock the intended property of being a genius detective, for only existing detectives can be strictly said to possess this property.

A similar approach is found in the modes of predication distinction. In Zalta’s terminology, Sherlock encodes the property of being a genius detective, and a flesh-and-blood detective exemplifies the same property. I do not think Zalta’s distinction is inspired by Ingarden but I will nonetheless claim that the two distinctions are reducible to one another. Of course, this is not without textual evidence. Ingarden, Uemura explicates, distinguishes between ‘standard’ and ‘non-standard’ property instantiation. Ficta instantiate properties in both modes. They instantiate the properties connected with their content non-standardly and the properties connected with their intentional structure standardly, that is, in the same way real objects do.⁶² But there are many problems with Zalta’s distinction. If Ingarden’s distinction is the same, then it is also subject to the same difficulties. A major problem raised against Zalta’s distinction has to do with its ability to account for reference to ficta. Zalta’s strategy to capture a fictional name’s unique reference, as developed with Otávio Bueno, is embedded in Object Theory (OT). Holmes’s denotation is:

Put informally, Holmes is ‘the abstract object encoding exactly the properties F such that, in the Conan Doyle novels, Holmes is F’. A story (situation), likewise, is an ‘abstract object that encodes only properties of the form [λyp] (being such that p), for some proposition p’. The truth of a proposition p in a situation s is determined by whether it encodes [λyp]:

Consequently, ‘In the Conan Doyle novels, Holmes is F’ becomes:⁶³

Francesco Berto et al. find fault with what Bueno and Zalta postulate to facilitate their model, that is, ‘a determinate group of features F such that, in the relevant stories, the fictional object is F’. Even if we assume that at a specific time there is such a content, it is not certain that we can have a determinate story content across different time spans. Storytelling is a temporally extended, revisable process. In Star Trek: Discovery we were introduced to new descriptions of Spock. For example, we learned that Spock has an adopted sister. Therefore, it can be argued that ‘In the Star Trek saga, Spock has an adopted sister’ was false until 2016. By Bueno and Zalta’s identification criterion, it appears that the referents of ‘Spock’ in 2016 (Spock2016) and ‘Spock’ in 2017 (Spock2017) are distinct. Spock2017 encodes a property that Spock2016 does not: that of having an adopted sister. Given OT’s identity criterion, which states that abstracta a and b are identical iff for any property F a encodes F iff b encodes F, Spock2016 and Spock2017 are two distinct objects. Consequently, the referent of the name ‘Spock’ is different from the one we had in mind in 2016.⁶⁴ In Zalta’s defence, it can be argued that storytelling is an ‘extended baptism’. Before the author is done with their storytelling, it is illegitimate to raise questions about their referring.⁶⁵ But this is not how fiction is written. As Sainsbury argues, when Conan Doyle added more descriptions to Holmes, those details were added to a character that we had already encountered. Authors think about the character first and then add to it embellishments, a past, and so on. The character itself is ‘fixed’ early in the process. Through the ongoing process of narration, Holmes acquires more properties. Conan Doyle was free in describing Holmes as he pleased. He could have described Holmes differently. ‘If Holmes is individuated_M by the properties he is in fact ascribed,’ Sainsbury proceeds, ‘this is impossible: being ascribed different properties would amount to being a different individual’.⁶⁶

Ingarden’s predication modes distinction, albeit equivalent to Zalta’s, is not undermined by the above charges. This is because Ingarden does not treat ficta as abstract objects encoding certain properties. At the risk of sounding repetitive, Ingarden treats ficta as products of creativity. It is the latter that contains ficta’s essence and supports their persistence. Imaginative acts are extended in time. It is therefore natural to keep ficta’s pool of properties open for additions. This does not mean that we cannot refer to a fictum until it is completed. Sometimes, there is no way of knowing when or if a character has been completed. As long as a fictum’s origins are grounded in an author’s creative acts, there should be no problem with identifying it as one and the same across different intervals. Another way of approaching this is connected with the overarching problem weakening Meinongianism. Zalta, although admitting ficta as created entities, ontologically prioritizes properties. In other words, he starts from the presupposition that there are bundles of properties, and Sherlock is an abstract object that encodes all of his properties. It is a natural consequence that his predication distinction cannot account for reference to ficta across different times.⁶⁷

Another difficulty that the modes of predication distinction stumbles upon is fixed by resorting to Ingarden. Proponents of this distinction have it that ‘a property is possessed internally by a fictum iff it belongs to the property set that constitutes that fictum’.⁶⁸ This definition neglects implicit truths in fiction. Consider the following example borrowed from Lewis:

(3) Holmes lived nearer to Paddington Station than to Waterloo Station.

This statement indicates an implicit fact in the Sherlock story. It has never been explicitly stated in the story but it is nonetheless true. Taking a quick glimpse at London’s map suffices to see that 221B Baker Street is indeed closer to Paddington Station than it is to Waterloo Station.⁶⁹ Neo-Meinongians would read out (3) from the set that constitutes Sherlock because it is not internally (explicitly) predicated of it. This practice, I argue, limits our reading of fiction to the literal, which goes against our tendencies to read into fiction hidden aspects that we pick out while reading.⁷⁰ Ingarden, by contrast, would treat (3) as a ‘spot of indeterminacy’, which can be concretized by readers.⁷¹ Namely, (3) is a lacuna not covered by the author, which invites competent readers to fill it in, deploying previously concretized experiences.⁷² Sherlock’s world is a represented objectivity. As such, it is endowed with ‘schematized aspects’. There are many aspects that are unfulfilled in Sherlock’s world; for example, we do not know which is closer to his residency, Paddington Station or Waterloo Station. Competent readers who can decipher the meaning units comprising Sherlock’s story fulfil such aspects, by making use of previously concretized experiences – namely, checking London’s map.⁷³ Ingarden’s approach to fiction, therefore, is compatible with how we ordinarily read fiction. Unlike neo-Meinongianism, it allows readers to play a role in reading into the story implicit facts that, taken together, constitute its identity.

Another problem with the modes of predication concerns ficta’s individuation. As Voltolini argues, just because we have a set of properties does not guarantee that a fictum will spring out of it, hence risking ending up with a bunch of properties and no fictum: (a) the no-ficta problem. Further, we can have one property set with many corresponding ficta: (b) the many-ficta problem. This is the problem of ficta generation all over again – namely, how does a fictum come out of a set of properties? (a) is exemplified by Saul Kripke’s ‘Moloch’, where we have a bunch of properties and no corresponding entity, and (b) is expressed by Jorge Luis Borges’s thought-provoking invitation to imagine a text written by Menard, which so happens to be identical with Miguel de Cervantes’s Don Quixote. What we have then are two distinct ficta, Menard’s Don Quixote and Cervantes’s Don Quixote, corresponding to the same property set. Therefore, the individuation of ficta must include something over and above mere sets.⁷⁴ This ‘something’, as Ingarden would argue, lies in recognizing ficta as created pure intentionalia. Ficta are not just bundles of properties. They are intentionally created entities that are brought to life at a certain temporal point. As regards (a), this problem is permissible when we rule out authors as creators (or ontologically prioritize properties, as Zalta does). When an author intends to create a fictum, they mobilize their creative acts to create a fictum and its properties, simultaneously, which they then make intersubjectively accessible through literary works. Readers acquaint themselves with the created fictum as intended by the author. This allows them to individuate the fictum when confronted with its properties. When an author creates a fictum as having a certain property set in a fiction, anyone familiar with the latter will be able to single out the former.

The same reply is applicable to (b). Neo-Meinongians risk running into multiple ficta that correspond to one property set because they do not acknowledge the role of authors as creators and/or give ontological precedence to properties. Revisiting the case of Don Quixote, two distinct creative acts went into the creation of Don Quixote, meaning that the resultant ficta are distinct. The generated ficta and their works may be identical but the creative acts that led to their generation are not. To bolster its role as regards ficta’s individuation, it is useful to visit Thomasson’s ‘intentional object theory of intentionality’. This theory demonstrates how conceiving of Ingardenian intentionality as distinguishing between object and content can be helpful in overcoming many problems. For instance, it helps us cognize how two contents can have one object as their target (for example, my thoughts about Cordelia’s father and Goneril’s father are about King Lear). This theory also shows how two objects can have the same content. We accomplish all this just by treating intentionality as a ‘mediated relation’, without the need to postulate a special type of relation adhering to intentionality alone.⁷⁵ Back to the Menard/Cervantes case, although they use the same content to describe Don Quixote, the authors’ intentional acts are not directed at the same object, which makes the ficta in question distinct. The reason for this is that the two entities are founded in two different works and mediated by two distinct acts.⁷⁶ In short, ficta properties are grounded in the creative acts of an author, and it is these acts that determine everything, ficta-wise, from conception to birth.

IV

In this paper, I have explored the problems of ficta ‘generation’ and ‘properties’ in light of Meinong and Ingarden. These problems were analysed in the context of Voltolini’s phenomenological reading of Meinong. I have illustrated that a phenomenology of properties need not be problematic if we pick the right parameters. These parameters are found in Ingarden, whose phenomenological ontology surpasses Meinong’s in many respects. To accomplish this task, I have tackled many problems that are raised against the historical Meinong and his followers. On my view, these problems are not insurmountable. They can be accounted for if ficta’s generation and properties are taken to be ontologically grounded in the creative acts of their authors.

Funding Information

This work is supported with a grant from the National Science Centre, Poland (PRELUDIUM 20, grant nr. 2021/41/N/HS1/00813). For the purpose of Open Access, the author has applied a CC-BY public copyright licence to any Author Accepted Manuscript (AAM) version arising from this submission.

Competing Interests

The author has no competing interests to declare.