Fictional discourse and fictional theory

As already stated, the majority of approaches on legal fictionality' have drawn on counterfactuality as criterion, and consequently' tied the question of truth to the criterion of reference. These two interrelated questions also constitute an important issue within fictional theory'. Fictional theory', however, dissociates from the legal treatment of fictionality' by' proposing an alternative to the traditional opinion that fictional discourse needs be understood as a false proposition, by reassessing both the necessity and the consequences of reference as criterion. Regarding reference, two main perspectives are usually' taken: first, that fictional discourse is discourse that refers to fictional entities, and second, that fictional discourse is non-referential discourse. At first sight, the former notion of fictionality seems to be identical with the classical notion, as present in the definition of “fiction” in Ballentines’ Laws Dictionary - which, in its failure to provide a parsimonious definition, mirrors the aforementioned dissociation between legal and extralegal understandings of fictionality:

Fiction. In the sense of a fiction of law, a contrived condition or situation [...] As a literary' work, a novel, a portrayal with imaginary' characters.¹⁶⁸

Since it has been argued that any reference to non-existing objects is necessarily a false proposition, the 19th century' and early 20th century’s treatment of literary fictionality could still closely mirror law’s truth-false dichotomy:¹⁶⁹ Bertrand Russell held that in Shakespeare’s Hamlet “[t]he propositions in the play are false because there was no such man” - and, by doing so, shared Bentham’s account of fictionality as discourse that assumes something to exist which in fact doesn’t - and thus is literally false.¹⁷⁰ In the history of fictional theory, however, the criterion of referentiality has strongly been contested - but nevertheless stayed an issue of debate throughout the whole 20th century, although the referential criterion has consecutively been modified.¹⁷¹ In Of Mind and Other Matters (1984), while agreeing with a modern notion of a conflicting plurality of truth by assuming that there was “no such thing as the actual world”, but that there were “many worlds if any”,¹⁷² Nelson Goodman assumes the classical standpoint: “All fiction is literal, literary falsehood.”¹⁷³ Accordingly, he rejects the idea of fiction as referring to fictional entities, or even to a fictional world:¹⁷⁴

[Mjost fictions seem to be about Active persons, things, and even events or about imaginary and even impossible beings and entities. Works of fiction, we often hear, are about Active worlds. But strictly speaking, fiction cannot be about anything nonactual, since there is nothing nonactual [...]¹⁷⁵

Goodman’s notion is based on a common belief also upheld by Searle: being able to refer to fictional entities would presuppose that fictional entities exist¹⁷⁶ - thus fictional statements, as referring to non-existing entities, are necessarily false statements.

It consequently seems that only the second perspective (the “no assertion view”)¹⁷⁷ could serve as an alternative to the classical understanding of fictionality. Proponents of the no-assertion view usually assume the following:

In fictional discourse the speaker does not claim that his referring expressions have referents. He speaks as if the referring expressions had referents. [...] In fictional discourse, the rule of reference is out of place.¹⁷⁸

Accordingly, for Gottfried Gabriel, fictional discourse can also be defined as being denotationless:

In fictional discourse the speaker does not claim that the predicative expressions used at the subject place have denotations; he only speaks as if these expressions have denotations.¹⁷⁹

While Gabriel, for the sake of a parsimonious definition of fictional discourse, has avoided the referential paradigm,¹⁸⁰ a relevant number of scholars have instead upheld the criterion of reference, while at the same time deviating from the classical understanding that fictional discourse is false discourse. The latter debate seems to partially mirror the medieval debate on the ontological status of universals. Accordingly, while some scholars understand fictional objects as abstract, man-made objects, Platonist ideas have also found their way into the debate. Wolterstorff claims a fictional character is a subtype of persons that differs from other persons only in terms of properties, and since properties are said to exist independently from fictional discourse, fictitious characters are supposed to be rather selected than made.¹⁸¹ Meinongian positions (as those of Terence Parsons and Richard Routley) even assume the possibility of reference without existence in the strict sense of the word - understanding fictional discourse as a discourse involving non-existing but possible entities being referred to.¹⁸² While these objects, from an ontological perspective, are said to be non-existent (“nicht-existierend”), they are, however, understood to subsist (“bestehen”),¹⁸³ thus allowing them to have metaphysical properties, with one categorical exception, though: the property of existence.¹⁸⁴ Terence Parsons has elaborated a Meinongian theory¹⁸⁵ applicable also in the case of literary fiction.¹⁸⁶ Graham Priest propagates a theory of a “supervenience of the fictional on the real”, assuming that “the nonexistent can be meaningfully be talked about, referred to or quantified over in our world ‘supervenes’ on the actions of the relevant existents in that world in particular” once they are created.¹⁸⁷ Marti-nich and Stroll assert that propositions containing a fictitious name or entity are possible in cases where an institutionalized praxis exists that is linked to a fictional name used, allowing the possibility of reference even in the case of non-existence.¹⁸⁸ Certain proponents of possibilism even go further by claiming a certain form of existence for things that do not exist, but that are possible. Erich Rast writes:

According to possibilism in its classical form, the things that have actuality' or existence form only a proper subset of the things that are tout court. What is actual exists, and what is non-actual but only possible does not exist, yet could exist. This is another way of saying that objects that do not actually exist may' still exist in various other ways.¹⁸⁹

A number of scholars even assume that fictional discourse refers not only to non-existent entities but to non-existent but possible worlds.¹⁹⁰ Such idea of fictional worlds, clearly' rejected by Gabriel¹⁹¹ and Goodman (“There are no fictive worlds”),¹⁹² is maintained by' scholars of different provenience, as the wide application of possible world theory on the question of fictional narrative illustrates (Eco,¹⁹³ Dolezel,¹⁹⁴ Pavel,¹⁹⁵ Ryan¹⁹⁶).

Both perspectives, however, allow truth values to be attributed to fictional discourse. While Goodman revives the non-referential Aristotelian notion of a higher truth of literature despite its literal falsity' (“While all fiction is literally' false, some is metaphorically true”),¹⁹⁷ a number of scholars have assessed the truth values of propositions within fictional discourse, partly even maintaining that fictional statements can be true in a referential sense - among them David Lewis, Terence Parsons and Gregory' Currie. Presuming a “fictional operator”,¹⁹⁸ Rast propagates an idea that is already' present in

David Lewis’s theory', one of the first proponents of an elaborated account of fictional worlds,¹⁹⁹ who, as an alternative to Meinongianism, proposed three provisional definitions of fictional truth (termed “Analysis 0”, “Analysis 1” and “Analysis 2”):

• (1) “A sentence of the form ‘In fiction f, 0’ is true iff 0 is true at every world where f is told as known fact’”²⁰⁰
• (2) “A sentence of the form ‘In the fiction f, 0’ is non-vacuously- true iff some world where f is told as a known fact and 0 is true differs less from our actual world, on balance, than does any world where f is told as known fact and 0 is not true. It is vacuously true iff there are no possible worlds where f is told as known fact”²⁰¹
• (3) “A sentence of the form ‘In the fiction f, 0’ is non-vacuously true iff, whenever w is one of the collective believes worlds of the community of origin of f, then some world where f is told as known fact and 0 is true differs less from the world w, on balance, than does any world where f is told as known fact and 0 is not true. It is vacuously true iff there are no possible worlds where f is told as known fact.”²⁰²

Lewis consequently claims that “truth in fiction is the joint product of two sources: the explicit content of the fiction, and a background consisting either of the facts about our world (Analysis 1) or our beliefs overt in the community of origin (Analysis 2)”.²⁰³ According to Lewis, fictional propositions not only can have truth values, but can be true precisely because a fictional proposition is usually a truncated assertion that involves a “fictional operator”. Such assertions are presupposed to always have the logical form of “In fiction f, 0”.

Even non-referential views, as Gabriel’s theory shows, have found a way to attribute truth values to statements containing fictitious elements, by assuming in the semantics of fiction a “non-ontological conception of sense (intension)”.²⁰⁴

For the treatment of legal fictionality, the above-mentioned perspectives are evidently of use. Bentham’s claim of the irreducible fictional nature of any discourse as bound to abstractions (qua fictional entities), which is also found in a modified form in the early ideas of a panfictionality of the legal conceptual system (Tourtoulon, Vaihinger, Lecocq),²⁰⁵ would make much of legal discourse fictional - and thus is comparable to the issue of a supposed panfictionality both in philosophy and literary theory.²⁰⁶ It is evident that a theory which categorizes the majority of propositions in language as fictional might not serve well as a heuristic means - or in the words of Fuller: “If even-thing is ‘fiction’, then the meaning of the word has been lost, and ‘as if has become simply ‘is’”,²⁰⁷ with the consequence that the “utility” of the category- “fiction” will become “utterly' destroyed”.²⁰⁸ Contemporary' approaches in analytic philosophy and literary- theory', which have tackled this particular problem by- establishing more differentiating criteria in order to make the distinction operable again, could also help to shed a more differentiated light on the legal field.

Furthermore, it is not only the parallelism between fictional theory and the Kelsenian notions of legal language understood as referring to legal concepts as non-actual entities, that might prove the endeavor useful (and can consequently be found in a 2001 publication on legal fictions),²⁰⁹ but also the fact that a number of the aforementioned approaches within fictional theory in general offer an escape to the usual dichotomous treatment of legal fictionality. A consideration of the above should make clear that at least a revision of the classical criterion of falsity is long overdue - and that a number of persisting claims on legal fictionality (e.g. that legal fictions were “unlike literary fictions in that they do not relate to verisimilitude in the same way”, since they had no ‘similitude’ with real things”) are too simplistic in their approach.²¹⁰

The fruitfulness of a reassessment of legal fictionality based on a fictional theory is also supported by the fact that a newer concept of legal fictionality already mentioned (the suggestion that legal fictions are in fact true “within the linguistic jural system within which they originate and are used”)²¹¹ is nothing but a covert paraphrase of Lewis’s account of fictionality (the proposition that a fictional proposition not only can have truth values, but can be true precisely because a fictional proposition is usually a truncated assertion that involves the unsaid fictional operator “In fiction f, 0”).

Even the understanding of fictional discourse as referring to fictional entities or fictional worlds could find a valid application in legal discourse, and this in two regards. On the one hand, both the idea of a realism of legal concepts (e.g. Gray’s question if corporations are “real or fictitious entities”²¹²), which would make legal concepts in general rather found than made, and the ideal of a parallel universe of legal concepts and relations, have been invoked in civil law countries already in early times, the latter both with Platonic and non-Platonic connotations (e.g. Jhering’s metaphor of a “heaven of legal concepts”²¹’’ and Kelsen’s observation that “[i]n the imagination of the lawyer, the person - the physical as well as the corporate - lives as a creature independently existing from the legal order”).²¹⁴ Kuntze in his 1856 monograph even opposes three realms - the sensible world, the legal world and the realm of legal fictions.²¹⁵ Kriickmann (1909/1919) claims, that due to particularities of the concept of causality, in law, an inevitable dissociation between reality and the legal realm necessarily occurs, and that due to this, for example in criminal law, never a “real” committer is to be found, but only the “legal” committer - and that law treats the “legal” committer “as if’ he was the “real” committer.²¹⁶

While newer voices have reiterated the latter topic,²¹⁷ a closer examination of Kelsen would already have provided valuable insights into this matter. Although Kelsen has just received new attention in the English-speaking world due to a translation of his 1919 essay “On a Theory of Legal Fictions”, an idea Kelsen raises in this essay has nevertheless been neglected by this most recent scholarship: Kelsen in fact suggested a “modification”²¹⁸ of Vaihinger’s definition of fictionality, which, in light of fictional theory', could be of interest for a revised approach on legal fictionality in general. Readings of Kelsen’s 1919

essay that assume that Kelsen unconditionally rejected legal fictionality often forget that the essay mainly counters the application of Vaihinger’s epistemological theory' to the field of law, demonstrating that due to the combination of the criteria Vaihinger stipulates (overt counterfactual assumption that is pro-visional/expedient), legal fictions do not qualify as fictions in a Vaihingerian sense. In his 1911 habilitation, however, which appeared as a second, amended edition in 1924 (and thus did not abandon the 1911 position on fictionality, as his 1919 essay' might suggest), Kelsen does not reject legal fictionality' per se, but only a classical notion of fictionality that propagates realist notions by' confusing the realms of nature and of law/ethics:

Not because a [legal] fiction is something feigned, [...] but because, and only insofar as it claims to contain something “being” (“ein Seiendes”), every' fiction needs to be rejected. [...] What needs to be rejected [...] is the methodological error, that in a concept, which has not been abstracted from the world of the “being”, something “being” is nevertheless searched for, [...] though the latter is only' a figment [...] of the wish of an “ought-not to be” (“Nicht-Sein Sollen”).²¹⁹

When treating dogmatic and judicial fictions, Kelsen then raises an interesting point about how fictionality' could in fact be thought in the realm of law, leading to an “extended understanding of fiction”²²⁰ - and by doing so anticipates the arguments of some more recent voices in scholarship: legal fiction defined as a contradiction with respect to a referential system alternate to the real world. According to Kelsen, this referential system is the “legal order”, which he also characterizes as an artificial “reality” or “world” that needs to be distinguished from the “natural reality” (the “reality of the senses”),²²¹ and that has its own factuality',²²² with the consequence that legal discourse would have its own truth values, independent from the “real world”.²²³ That in his 1919 essay Kelsen nevertheless claims that legal fictions are not “fictions” seems to be primarily' based on two of Kelsen’s presuppositions. First, Kelsen treats fictionality' in the context of Vaihinger’s definition, requiring counterfactuality' and expediency²²⁴ - and while statutory fictions are found to lack in counterfactuality, both dogmatic and judicial fictions are considered not to be expedient,²²⁵ with the consequence that legal fictions could at the very' most appear as “pseudo-fictions”.²²⁶ Second, Kelsen nevertheless seems to presume a referential contradiction to be a necessary' condition for fictionality', and therefore demands a proposition which is assertive²²⁷ - though it is unclear if Kelsen himself agrees with Vaihinger’s other criteria for “real” fictionality.²²⁸ To decide the latter, however, would be crucial, for in his 1919 essay, though rejecting statutory fictions for the reason of not being assertive (“the law affirmeth nothing, [...] it commands”),²²⁹ some legal fictions are in fact both considered to be assertive and to contain a referential contradiction - and are only rejected by Kelsen for lack of the criterion of expedience Vaihinger prescribed.²³⁰ With fictional theory having in fact fundamentally challenged the classical criterion of counterfactuality, and, as we will see later in the chapter, with certain proponents of speech act theory contesting Kelsen’s premise that fictionality presumes an assertion, these changes would now allow the full scope of legal fictions to be reassessed in light of Kelsen’s aforementioned “extended” definition.

Also, it is not only the idea of an alternate reality that has long been present in a number of legal approaches, thereby establishing a competing reference system for truth claims, but also many of the assumptions of possible world theory. Both in the law of tort and in contract law, counterfactual considerations are commonplace - and could be best described in terms of the construction of an alternate possible universe that differs from reality only on a very' restricted factual basis. “Factual causation” is often considered to be the basis for liability, both criminal and civil,²³¹ and when it comes to determining factual causation, the hypothetical “but for” test is usually' cited as criterion: “[AJn event X causes an event Y if, but for X, Y would not have occurred.”²³² To rethink how the world would now be in absence (or counterfactual presence) of a past event X is also required in the calculation of damages to be paid - for example, in order to account for “lost profits”.²³³ In other branches of law, the questions of hypothetical gains or losses are also commonplace - for example, in tax law.²³⁴ It is, however, procedural law, where a parallel world of legal factuality (which might be substantially deviating from the actual world) is created on a daily basis - with the consequence that the truth value of propositions pertaining to such legal discourse could be assessed in a manner parallel to how possible world theory' is used by fictional theory to assess alternate worlds of facts (Lewis, Beardsley, Reicher). Not only' were a substantial number of the “classical” fictions in fact procedural fictions, but also a number of provisions and principles of nowadays’ procedural law either force the participants to disregard facts or might even require them to assume a factual basis as legally' relevant which deviates considerably from the real world. The doctrine of estoppel, which “prevents a party from asserting a position in one legal proceeding that directly' contradicts a position taken by that same party in an earlier proceeding”,²³⁵ can, for example, lead to a dissociation of factual and legal reality. But not only' in civil procedure might certain evidence be excluded: in criminal cases, a jury' might be asked to also “disregard” certain facts despite them being true. Such deviation from reality can even be cemented, perpetuating an alternate version of reality that is also binding for future recipients: res judicate pro veritate accipitur.²³⁶

But also from another perspective, the question of fictional reference could be valuable for the question of legal fictionality. It has been suggested by Simon Stern (Chapter 5.1 of this volume) that legal fictionality might be comparable to metafiction in literature. Having once established such a hypothesis, fictional theory's treatment of the question of truth and falseness pertaining to metafiction²³⁷ might be apt to provide valuable insights.

Last but not least, the discussion of fictionality within fictionality theory could be fruitful for another aspect of the legal debate: legal and literary fictions both have factual effect in the “real” world in question of knowledge,²³⁸ action and status. Niklas Luhmann has already extensively treated the effect of fictional literature on social practices,²³⁹ and comparably, in Chapter 4.2 of this volume, Sarah Winter investigates the effects of fictional literature within the history of codification of human rights.