Ontological Commitment Revisited 9783110750041, 9783110749991, 9783110750119, 2021936898

Citation preview

Ontological Commitment Revisited

Aporia / Ἀπορία

Edited by Jesús Padilla Gálvez Advisory Board: Pavo Barišić, Michel Le Du, Miguel García-Baró, Margit Gaffal, Guillermo Hurtado, António Marques, Lorenzo Peña, Nicanor Ursua Lezaun, Nuno Venturinha, and Pablo Quintanilla

Volume 13

Ontological Commitment Revisited Edited by Jesús Padilla Gálvez

ISBN 978-3-11-074999-1 e-ISBN (PDF) 978-3-11-075004-1 e-ISBN (EPUB) 978-3-11-075011-9 ISSN 2197-862X Library of Congress Control Number: 2021936898 Bibliographic information published by the Deutsche Nationalbibliothek The Deutsche Nationalbibliothek lists this publication in the Deutsche Nationalbibliografie; detailed bibliographic data are available on the Internet at http://dnb.dnb.de. © 2021 Walter de Gruyter GmbH, Berlin/Boston Printing and binding: CPI books GmbH, Leck www.degruyter.com

Contents Jesús Padilla Gálvez Ontological Commitment: An Introduction Peter Simons Meinong’s Objectives

1

21

Julian Nida-Rümelin Metaphysical Aspects of Agency

33

Margit Gaffal Representation as Ontological Problem

63

Jesús Padilla Gálvez Ontological Commitment and State of Affairs

79

António Marques “I know” Language-games in Wittgenstein’s Late Philosophy Manuel García-Carpintero Against Propositional Substantivism

99

111

Jonathan Smith Wittgenstein’s Nachlass, a Case for Practical Ontology?

131

Arthur Gibson Paradox in Ontology: Black Holes, Cosmology, Wittgenstein versus Stephen Hawking’s Claim that Philosophy is Dead 139 153

List of Contributors Index of Names Subject Index

155 157

Jesús Padilla Gálvez

Ontological Commitment: An Introduction There are philosophical issues that deal with what is known as ontology. Ontological topics include questions such as whether a thing is real or whether existence can be ascribed to things. A well-known example is the problem of universals, which arose out of the nominalist-realist controversy as to the actual existence of universals or abstract entities. Such entities are generally discussed when we refer to classes or propositions, that is, when we refer to the content of an abstract statement with particular terms. If one reflects on these debates, one notices that philosophers who reject the reality of entities tend to contradict themselves after a certain time. As the history of ontology is complex, we shall approach the matter systematically, beginning with a succinct introduction. What does it mean to ask if an entity exists? Observe that Quine did not focus on the question of whether an entity exists or not, but rather focused on exploring the scope of what it means to ask about an entity.¹ This constitutes a change in approach to examining what exists and what doesn’t. The question invites us to assess what it means to exist, or to be considered an entity. Quine’s research did not provide a catalogue of what exists, but an interpretation of what it means to exist, what the core of a “concept of existence” is as part of a larger “doctrine of being.” In order to argue about the existence of a supposed entity, one must adopt a (general) concept of existence that provides a rational standard for the argument. If we cannot previously characterize what it means to exist, any debate on ontology is reduced to prejudices and arbitrariness. As such, the opening question determines the notion of ontological commitment. It does this by determining two preconditions for ontological debates, namely, the stipulation of a concept of existence and the requirement of coherence. Regardless of how “existence” is determined, all assumptions about “existence” that occur in an ontological debate can only be ascribed if they are ontological commitments given by the previously assumed concept of existence. The notion of ontological commitment is, therefore, a vindication of rationality. Although Aristotle does not use the term “ontology” in his writings, he makes reference to it when he indicates that “there is a science that studies being insofar as it is being – expressed by the expression: “τὸ ὂν ᾗ ὂν””.² How-

 Quine, 1966a, 64– 69.  Aristotle, Metaphysica, 1003a21. https://doi.org/10.1515/9783110750041-001

2

Jesús Padilla Gálvez

ever, Aristotle does not recognize this science as having a “γένος”.³ This discrepancy generated tension within the principle that each science must have a “ὑποκείμενον γένος”, that is to say a fundamental subject and the definition of the science.⁴ Hence, the problem arises that the “γένος”, which is the basis of ontology, is not a genus. Aristotle did not resolve this discrepancy but offered a solution by identifying “γένος” as immobile and immaterial. As such, ontology deals with the highest kind of being, that is, “περὶ τὸ τιμιώτατον γένος”.⁵ Scholars of antiquity argued that being is not simply predicated in an equivocal way, but rather in relation to “the principle of one”, that is, “πρὸς ἓν”.⁶ This “one” is interpreted as the “oὐσία” or the highest class of being. Another fundamental change in how ontology was analysed came from the twelfth-century philosopher Gundisalvo. His proposal was collected by Albert Magnus and Thomas Aquinas. Medieval philosophy distinguishes two different aspects of the “ὑποκείμενον γένος”: on the one hand, the fundamental matter is what is studied, which is the subjectum that represents the ens; on the other hand is genus, which is considered the highest class of being. Gundisalvo argued that species are the characteristics that accompany being insofar as they are considered “consequentia entis”⁷ and that they are subdivided as follows: a being is substance, another, accident. One is general, the other particular. One is cause, another is caused, one is possibility, the other is reality, etc. The aim of ontology has always been to know which entities (things) we consider as existing. For many philosophers, this definition appears redundant since it refers to what is or what exists or may exist. An entity participates in being and has its own properties. We shall leave aside the discussion of what is considered “entity”, but we will rather clarify the ontological point of view from the perspective of enlightenment. This bids us raise the question of our commitment. The first thing that is important to highlight is the primary character of this science – that is, entis scentia –which has always been focused on the study of “speculatur ens”. Ontology seems to be a system that includes the method to be used in the doctrine of being. This is so because it has generally been considered a rational science par excellence since, it uses a demonstrative method that is rational and deductive. Its main concern has always been the examination of the supreme kinds of things from a formal point of view. This point of view should not

 Aristotle, Metaphysica, 998b22. Plato considered the study of “being” as the most important topic, namely “μέγιστα γένη”. Plato, Soph., 253 b8 – c3.  Aristotle, Analytica posteriora, 75a42.  Aristotle, Metaphysica, 1026a21.  Aristotle, Metaphysica, 1003a33.  Gundisalvo, 2007, 100.

Ontological Commitment: An Introduction

3

be confused with logical formalism. It seems that any definition of “ontology” will always be quite imprecise since there remains a certain ambiguity regarding its object of study. On the one hand, ontology can be understood as the science of being in itself, that is, of ultimate and irreducible being in the sense of a primus ens, on which other entities depend. On the other hand, ontology aims to analyse of what entities consist or of what being consists. In line with this view, ontology is considered a science of essences rather than of existences – thus being considered a theory of objects. Additionally, it is a science of pure formalities. This ambiguity remains in contemporary ontology and makes its investigation a complex undertaking. Traditionally, ontology deals with the analysis of being (onta) and analyses everything that we consider to be “there”. This definition is circular since the definiendum “ontology” is defined by use of a term that appears in its nominal composition, “to be”, which is why the definiens must be clarified again. Putting “being” and “what there is” together does not help much because we have to determine what we consider “to be there” before we can say what is there. For instance, if we ask what types of entity exist, the question is what kind of entities we are referring to, abstract or concrete entities? Are these properties or universals or are they rather particular entities? We thereby introduce a new problem because we have to determine the meaning of “existence”. And, in one way, we have to make this determination before we can make this determination. It seems that when we analyse the ontological problem we find ourselves in a self-referential argument. Whenever we aim to define “ontology” we use of the verb “ontos” or “to be” and thereby we introduce synonymous terms such as “be” and “exist”. By stating that we want to know something, we are already affirming that there is something, something that the Greeks called “onta”, meaning “that which exists”. In other words, the person examining a question assumes that the question arises at all. To sum up, without noticing it we introduce a copulative verb into our approach to ontology that determines the very structure of the question. Obviously, this argument is not new. Aristotle had already been confronted with the problem of definiendum and definiens overlapping and with this paradox he introduced a complex problem in an informal way. This approach appears original since it consists of a “formal” version and reading of the argument, taking into account the propositions preferred by classical philosophy. In any case, the presented argument acts as a placeholder that allows for the introduction of abstract entities, such as the verbs “to be”, “to have” and “to exist” as they appear in the discussion. Indeed, the main purpose of classical philosophy was to guarantee a certain “commitment” and to maintain an order of priorities in ontological questions, thereby establishing an ontological thesis from the be-

4

Jesús Padilla Gálvez

ginning of the investigation. This circle in contrahendo is developed in the original definition of the matter we are analysing. However, this definition causes an obstructive error which results from a lacking connection between definiendum and definiens. The result of this error is similar to the one that occurs in a mental caveat that generates a discrepancy between the concept to be defined and the set of terms that explain the definiendum. As such, ontology contains a self-referential argument that implies a prior assumption of some unshakable principles about ontology. Kant was aware of this error and pointed to a lack of progress in the field of ontology since the time of Aristotle.⁸ To break his circle some authors proposed to assume that ontology be expressed exclusively in defined language. That language involves certain commitments inasmuch as we accept the content being expressed in a particular way. We cannot therefore refer to an entity or discuss its existence without clarifying beforehand the meaning of the expressions used and the way in which they are applied. To know the ontological commitment(s) at play we must determine the meaning of the terms, expressions or symbols used and characterize their content. The definition expressed at the beginning, in which definiendum and definiens are self-referencing without actually granting any properties to the referenced object, may be considered the metaphysical program. Ontological commitment is a prerequisite for ontological research insofar as we must study whether the terms used in a given language refer to a specific content and are determined by a specific form. Once we have clarified these elements we can establish the type of entities to which it refers, but not vice versa. At first sight, the notion of ontological commitment appears so simple and informally obvious that no theory concerning it is required. Before we analyse this matter in more detail later, we shall introduce a new problem that is related to this field. The verb “to be” determines a predicate-subject relationship that remained unquestioned until the end of the 19th century. Frege and Russell were the first to introduce what is called the ambiguity thesis. When a speaker states that “John is a man”, we cannot determine exactly whether John has the property of being a man: Does it mean that John is identical to man or that there is some kind of bi-conditional relationship between “John” and “man”? Does it mean that John is an individual who belongs to the class of men or if what is valid for John is valid for some case – which would mean introducing an existential generalization – according to which to be a man is true at least for the given case. Or does it imply a generalisation in the sense that what is valid for John

 Kant Fortschr. d. Metaph. Vorr. (V 3, 84 f.).

Ontological Commitment: An Introduction

5

as an ordinary case is valid for any case? This ambiguity is unintentionally integrated into the usual notation of first-order logic. More specifically, it forms part of the usual notation for quantifiers of any order, since they convey the multiple meanings with which they operate when we apply copulative verbs in natural language to formal structures. Thus, the verb “is” in the sense of existence is expressed through the existential quantifier ($xPx), the “is” of predication through juxtaposition or through a singular term that fills the argument space of a predicative expression, the “is” of identity through the symbol “=”; and the “is” of subsumption by means of a general conditional.⁹ We shall examine the Frege-Russell thesis in terms of what it addresses and what it leaves aside. Any reasonable thinker agrees that the verb “to be” in natural language must be extended to “have” or “exist”, which in ancient Greek is expressed by “ἐστὶν”, but is used in a different way. What the Frege-Russell thesis affirms is that this difference in use is due to the ambiguity of a single word rather than to different contexts in which the verb may be used. What stands out is that neither of the two philosophers has explicitly formulated this principle. However, we owe it to Wittgenstein that he addressed this issue and dealt with possible interpretations.¹⁰ Yet, all its components are unequivocally present in the writings of Frege and Russell and both imputed great importance to the thesis of ambiguity. It was usually referred to in the context of adopting a specific notation in first-order logic including extension to quantifiers of any order, as in Frege’s explicit logic, as well as in Russell’s and Whitehead’s studies. When using first-order logic as a framework for semantic representation, the ambiguity of copulative verbs is revealed by the ambiguous usage that emerges from natural language. Therefore, anyone who deals with ontology will have to clarify what his predilection is among the possible alternatives. In other words, he must state what model he is presupposing in his universe of discourse. We will apply what has been said to an example and examine it. We have argued that the analysis of the concept of identity occupies a central position in the development of the Frege-Russell thesis of ambiguity. The history of such treatment has a surprisingly clear structure. Philosophers have mainly adopted two opposing positions in regard to their theses: most philosophers and linguists accepted Frege’s and Russell’s position, whereas earlier philosophers rejected their distinctions, claiming that ambiguity should be considered a part of the argument. It is therefore pertinent to ask which role the

 The copulative verb is much more ambiguous than the early modern logicians believed. A systematic analysis of the ambiguity thesis is in: Padilla Gálvez, 2017, 12 ff.  Wittgenstein, TLP, 3.323.

6

Jesús Padilla Gálvez

problem of identity plays in modern interpretation and, more specifically, to know what we mean when we use the notion of identity. As we saw above, the identity sign is used in formal logic to express functional dependencies. Thus, when we posit “y = f (x)” we observe an equivalency between the left and the right parts of the formula. In this dependency a numerical identity is expressed as when we perform the operation “4=3+1” and a dependency of one object on another. However, our view of dependency changes if we analyse identity in terms of equitable distribution, for instance, when we put two terms in relation, such as in the statement “John is Mary’s husband”. This view of dependency was mainly used during Antiquity. As the previous example expresses, equitable identity allows us to distinguish two individuals who maintain a balanced marriage relationship. If someone articulates this statement to me and I do not know that Mary and John are married, I have expanded my knowledge and can relate two individuals thanks to the use we make of identity in our everyday language. Frege, however, adopted an exclusively mathematical perspective and developed a formal language as described in his Begriffsschrift. For instance, in the calculus regarding the cube of a number, such as in the example “43=64”, the identity is understood as a relationship between “4”, “3” and “64”. This identity does not generate dependency but rather refers to two names that are defined by: “43” and “64”. When we refer to a mathematical concept of identity, we refer to a quantitative meaning. This quantification is approached in terms of its variables that “extend to” a class of values.¹¹ The existential quantifier merely externalizes a certain nonempty predicate and nothing else. Furthermore, it expresses the dependence or independence of the variables of the quantifiers to which it is linked. Frege ignores that quantifiers also assume a role of dependency indicators, and it is here that the lie takes root.¹² The fact that the logical notation he proposes does not allow the representation of all possible patterns of dependence and independence between quantifiers, and even more between variables, led him to generate the so-called Russell paradoxes, discovered by Frege in his logical proposals. To solve the ambiguity, Russell first tentatively proposed the so-called doctrine of types. A profound analysis of the problems related to such ambiguous term as “being”, as well as the fragmentation of its interpretations, required that ontology be examined from new points of view. This introduction is not intended to provide an overview of ontology. The book title contains the notion of commitment. This is closely related to the idea of “to decide, to stand by something and be inseparable from some-

 Frege, 1966, § 8, pp. 11– 14.  Frege, 1966, § 8, pp. 11– 14.

Ontological Commitment: An Introduction

7

thing.” However, commitment and decision are not equivalent, so we will have to approach these things with great care. In a broad sense, commitment refers to a fundamental constituent of human existence. In a strict sense, however, it refers to concrete actions and their immediate effects. Both aspects co-implicate each other. To be more precise, commitment links theory with praxis. If a speaker utters a proposition, we take it for granted that the content he expresses is closely linked to what is being asserted. The propositions that the speaker emits stand in direct relationship to what is done with them. Otherwise we would hesitate to assume that what is said has nothing to do with the action. Commitment is gauged by taking into account the consistency of what has been expressed, the adequacy of the action carried out and its relation to objective facts. As such, a link is generated between the speaker’s expressions and the actions performed, which reflects a compromise between the subject, the mode of expression and the resulting actions. Socrates provides the paradigm of a committed philosopher; among his antipodes we find Hegel. After a fundamental examination of psychologism and all its digressions, modern philosophy has always pleaded for the primacy of the laws of logic. These laws are purely formal laws that are propositions free of any psychological prejudice. That makes logic a domain of analytical truth. In this context, E. Husserl drew attention to a fundamental distinction relevant to the current understanding of ontology: he defended the objectivity of logical structures. Therefore, philosophical terms and propositions were called “ideal units.” The object of investigation here comprises two separate areas, the logic of categories of meaning and the logic of formal objective categories. The first deals with topics such as truth, the concepts of subject and predicative forms, elementary statements of the forms of logical connection, conclusion, etc. The second deals with the a priori doctrine of the formal constitutive concepts of objects. These formal terms include “object”, “state of affairs”, “unit”, “multiplicity”, “number” and others. This field of study is also understood as a formal ontology. Some of the papers presented in this volume deal with problems in this field. Which ontological commitments are we willing to accept when we study our language and its content? This simple question can only be answered if we elucidate which entities we decide to admit as existing and what links are generated with these entities. Therefore, a precise definition of an ontological compromise criterion is needed, as this is a precondition for any fruitful and rigorous discussion of the problem, for instance a discussion about universals. Any question discussed between nominalist or realistic positions requires a clarification of the ontological question. A simple example shall illustrate the problem we are posing. Someone may love cats; others are indifferent to them; others are allergic, which is why they avoid cats, others have a phobia of felines, some cultures

8

Jesús Padilla Gálvez

have adored them, etc. An orientation towards cats may appear very differently depending upon the ontology the subject is willing to accept. But in addition, people who like cats often keep them as companion pets, others use them to hunt rodents in their gardens, some train them to perform in the circus and in certain Asian cultures their meat is eaten. Thus, depending on the function we allow to the concept of “cat”, very different links will be generated. Each of these contents described generates a link that is often very different from others. For example, people who live with cats will have a completely different posture toward them than persons who react allergically to them. Others may argue that “Felix the Cat” and “Sylvester the Cat” exist, for which reason they are willing to accept certain fictional aspects with anthropomorphic properties as part of their feline ontology. In a similar vein, Egyptologists know the representations of the Egyptian goddess Bastet or Bast, who appears in form of a cat and to whom the protection of mankind and their homes is attributed. In addition, she is considered the goddess of happiness and harmony, and would be offended, surely, to see someone eat cat meat. The question therefore revolves around clarifying or determining what kinds of entities are accepted in a discourse, thus guaranteeing their existence and their exhibiting certain functions. What becomes obvious here is that the term “cat” does not seem an innocuous term referring only to a class of felines called “Felis catus”. To investigate the questions posed we shall examine systematically the history and peculiarity of the concepts mentioned. Quine distinguished two different ways of understanding the question of the existence of entities, which correspond to two opposing concepts of existence. On the one hand, being is associated with meaning, and on the other hand, with reference. Each of these two concepts requires specification of the ontological commitments that go along with the statements we support. First, if signification assures existence, then any proposition must express an ontological commitment to the entity. Second, when we analyse the reference, it seems legitimate to assume that an entity exists when we express a proposition in which it functions as a means of reference. First, when questioning what it means to exist, why should we prioritize reference over meaning? Second, within the doctrine of reference, what are the bases for deciding when a term is a legitimate vehicle of reference? Addressing both issues will be a recurring task. An answer to the first question comes from Quine’s general philosophical conceptions: his naturalism, the rejection of the notions of analyticity, synonymy, and other intentional notions, his principle of ontological parsimony, guided by adherence to Ockham’s razor and standards of ontological admissibility based on criteria of individuation provided by the logical laws of identity. Quine’s answer to the second question, concerning the reasons for deciding whether a term is a legitimate vehicle of reference, is precisely given by his criterion of ontological commit-

Ontological Commitment: An Introduction

9

ment, together with his conceptions of regulation, paraphrase and ontological reduction. In short, the notion of ontological commitment represents the recognition that the assumption of existence in discourses is only legitimate if a conception of existence has been previously adopted. The criterion of an ontological commitment clarifies this bias. The criterion associated with a concept of existence establishes which elements of the discourses confirm a certain assumption of existence. In Quine’s proposal, the concept of existence is equivalent to a referential doctrine of entities, and the elements that confirm the assumption of existence are considered as legitimate means of reference. The criterion of ontological commitment proposed by Quine has dominated the discussion of ontology within analytic philosophy since the mid-20th century. This criterion has two different functions: first, it allows us to measure the ontological cost of theories, as Church put it, in order to decide which theories to accept, and second, once a theory has been established, it helps determine which components of the theory were responsible for its ontological costs.¹³ Ontological assumptions have come to play a controversial role in recent discussions. Quine’s proposal, it turns out, may be used to argue that opposing theories are more costly than their theorists initially admitted.¹⁴ These proposals were used to advance a traditional nominalist direction. Quine indicated that sentences consisting of subject and predicate have no ontological commitment to properties or universals. The locus classicus for Quine’s criterion is his work “What there is”, in which he said the following about theories: A theory is committed to those and only those entities to which the bound variables of the theory must be capable of referring in order that the affirmations made in the theory be true.¹⁵

This makes it necessary that we account for multiple aspects of philosophy in general, and logic in particular. What it affirms, in short, is that a theory should only be committed to those entities to which it can refer, such that when we express affirmations within this theory to be true. Therefore, the core of the discussion revolves around the expression that exclusively links the reference of certain entities to linked variables of the theory. This link has to be carefully scrutinised. Let the statement therefore be:

 Quine, 1960, 270. To study the supposed benefits of such ontological accounting.  Church, 1958, who exercises the criterion polemically against Ayer, Pap and Ryle.  Quine, 1948, 33.

10

Jesús Padilla Gálvez

(E) ($x). x is a cat. x is black. Let T be the theory obtained by adding this statement (E) – as an axiom, or at least as true – to suitable first-order logic. According to Quine’s criterion, theory T should have an ontological commitment according to its presupposition that can be described as follows: … entities of a given sort are assumed by a theory if and only if some of them must be counted among the values of the variables in order that the statements affirmed in the theory be true.¹⁶

According to Quine, his proposal involves a commitment to the existence of any cat and also involves a commitment to the existence of cats in the plural. The relevant criterion is not the reading of the recently cited text in the singular, but rather a general reading that is exposed in another work.¹⁷ In the quote above something different is defined, that is, commitment to any entity (not entities of its type). Such compromise occurs when the truth of a theory depends on some irreplaceable entity. A theory is committed to those and only those entities to which the bound variables of the theory must be able to refer so that the claims made within the theory be true. At first glance, the notion of ontological commitment to theories appears as a trivial matter: theories possess truth conditions.¹⁸ These truth conditions tell us what the world must be like for the theory to be true. Sometimes the conditions demand from the world that certain entities or types of entities exist. The ontological commitments of a theory, then, are just the entities or entity types that must exist for the theory to be true. When Quine refers to the ontological commitments of a theory, his interest is therefore focused on determining the objects and how they are described, rather than on the meaning of the theory. Therefore, his proposed alternative is based on the description of the ontological commitments of a theory within the theory of reference of a meta-theory, whose universe includes that of the object theory. Similarly, the truth for a theory can be described in the reference theory of a metatheory whose notation includes that of the object theory, or an equivalent. On the contrary, A. Church rejected the assertions made by Quine about the criterion of ontological commitment. His refutations are carried out at different levels. The first thing he maintains is that the theory T, according to his criterion

 Quine, 1961, 103.  Quine, 1961, 131.  Quine, 1948, 33.

Ontological Commitment: An Introduction

11

expressed above, lacks an ontological commitment, that is, T has no ontological commitment to any entity.¹⁹ To correct this difficulty, it must be assumed that the ontological commitment should not be carried out with an entity, but with a class of entities. It can be argued that ontological commitment to a class (e. g. black cats) is at least more satisfactory than ontological commitment to an entity (e. g. “Felix the Cat”, “Sylvester James Pussycat, Sr.”, “Bastet “, etc.). Obviously, if the ontological commitment should not be directly with an entity or a class, a predicate in Quine’s sense could be used, instead of a class. This proposal leads to the introduction of a scheme that allows incorporating the criterion of ontological commitment. In fact, a scheme proposed by Church is introduced that unfortunately goes along with some difficulties that shall here be briefly exposed.²⁰ One may describe Quine’s proposal using the following formal scheme: (1) The statement of ($z) M carries an ontological commitment to ẑM. Church proposed that we understand ẑM extensionally as a class name. Let M1 be a particular case of the statement “z is Sir Walter Scott” and let M2 be “z is Sir Walter Scott and z wrote Waverley”. From scheme (1) follows: (2) The assertion of ($z) M1 carries an ontological commitment to zM1. In fact, we can historically corroborate that: (3) M1 ≡ M2. Therefore, when applying the principle of extensionality, it follows: (4) The statement of ($z) M1 carries an ontological commitment to ẑM2. Since (4) is clearly false, since it does not correspond to some reasonable notion of ontological commitment, it follows that ẑM cannot be taken extensively in (1). The first consequence is that it must have, in Frege’s terminology, an oblique occurrence, or in Quine’s terminology it must assume an occurrence that is not purely designative, or is referentially opaque. Indeed, this conclusion emphasizes that the notion of ontological commitment is intensional.²¹ Church argued that in order to deal successfully with ontological commitment, it is necessary to take the phenomenon of oblique occurrence into account

 Quine, 1961, 103.  Church, 1958, 1014.  Church, 1958, 1014, footnote 3.

12

Jesús Padilla Gálvez

– or not purely designative – and to have a detailed logic that is applicable to such occurrences. The way to obtain logic applicable to such an indirect occurrence – which refers to names, substantive clauses, etc. – involves admitting the notion of a class concept and trying to formulate a logic of concepts of class. Of course, this implies ontological commitments that require intensional notions. This means that a viable alternative would be to develop an analysis of the general oblique occurrence of belief statements. The conclusion would be that the notion of ontological commitment belongs to intensional semantics or to the theory of meaning. The same applies to that-clauses. In fact, Quine’s claim in his book ‘From a Logical Point of View’²² that the notion of ontological commitment to an explicitly quantifiable form of language belongs to extensional semantics, is not guaranteed. The argument he presents in favour of his claim turns out to be erroneous, since the consequence of his argument is that if any existential quantification presupposes objects of a given type, then these objects exist in fact. Obviously, the extensionalist proposal is based on his assumption that there exists incompatibility between his aversion to intensional notions and the desire to impose a certain standard of coherence and responsibility in matters of ontological commitment. We previously referred to the concept of “cost” without having analysed it. For instance, let’s consider the question: Do you prefer a “Burger” or a “Boeuf Bourguignon”? From an economic point of view one might choose the cheaper option. However, the consequences of this option could be more expensive in the end, even detrimental to health. Regarding the first choice we free ourselves from the expenses of kitchen-related costs, the unnecessary time of cooking (more time on the sofa having chips) and a company brings the prepared food to us (I am not interested in knowing if there are underpaid foreigners working in miserable conditions). But what exactly are the consequences of the new framework generated by such cost-related “ontological commitment”? At first sight, it was very tempting to save kitchen costs and work, as cooking and everything that our new ontology entailed, meant that kitchen implements are considered redundant. But what does ‘redundant’ imply in this context? The new point of view goes beyond the problems of housing, more free time, unhealthy nutrition, increasing third-world conditions entering the first world, pauperization of society, new cardiovascular diseases, etc. As it were, the new vision of theories provides us with new ontological problems. It eliminates certain problems and creates new ones. How did we get to this point? We tie in with what was said at the beginning. What is presented to us as the pragmatic solution by the pro-

 Cf.: Quine, 1961, 130 f.

Ontological Commitment: An Introduction

13

posals developed in Harvard appears to be based on a naive ontology that generates new ontological problems. The level of complexity of this sentence makes it necessary for us to bear in mind multiple aspects of philosophy in general, and logic in particular. What it affirms is that a theory should only be committed to those entities – whatever they are – to which it can refer, so that when we make statements within this theory they are true. Therefore, the core of the discussion revolves around the expression that links the reference of certain entities to the corresponding variables of the theory. This link has to be examined very carefully. Let us assume the following statement: (1) Abigail eats. Sentence (1) asserts that an individual – a particular subject named “Abigail” – is performing an activity, that is, “___eats”, which is understood as a predicate. By creating an empty space we have eliminated the name of the subject, which can be replaced by “Adrian”, “Anthony”, etc. We can substitute the names by means of the lowercase letter “a” and the predicate by a capital letter so (1) would be formalized by: (1 *) Ma Now, it may be that Abigail is not the only dinner guest at the table, but several diners participate whose names begin with the first letter of the alphabet such as “Adrian, Anthony and Anna.” This group of dining guests who meets once a month has the peculiarity that all its partners are made up of people whose names start with “A”. Therefore, we could express the actions they perform regularly by introducing the existential or universal quantifier in such a way that we affirm: (2) There is an x, such that x eats. Formally expressed: $xMx. (3) For all x, x eats. Formally expressed: "xMx. The variable ‘x’ replaces any individual who performs the action of eating as a dining guest at the table of our example. The variable appears within the scope of the quantifiers in (2) and (3), so that variable is said to be bound. Quine refers to this type of link in his definition of ontological commitment. If we substitute ‘x’ for “Abigail” we affirm that the sentence is true, while if we substituted it for “Maria” we would say that the sentence is false. Without further

14

Jesús Padilla Gálvez

ado, ontology ceases to be analysed as a theory of “being” (onta) and is subjugated to the capitalist criteria of the ontological cost of theories that allow the introduction of a decisive criterion to accept or reject any statement. Suddenly, this procedure allows the evaluation of theories according to their components and thus could determine on which components the responsibility for ontological costs falls.²³ This change has proved important, since philosophical theories may now be treated as if there were an accounting department in charge of studying their details in terms of services, returns, benefits, remuneration and a commission pertaining to what has been stated. By means of this procedure it could be that everything superfluous or generating additional costs in the theories could be discarded. The mercantilist view of ontology has come out of Harvard, and what it demands are profits understood in accounting terms. But complications arise as soon as one tries to specify the truth conditions of a theory: different balances of truth conditions lead to different accounts of ontological commitment. Furthermore, theories formulated in ordinary language do not carry their truth conditions, nor their ontological commitments, up their sleeves. Therefore, the need arises to find an ontological commitment criterion: a test or method that can be applied to theories in a neutral way to determine the ontological commitments of the theory. For example, perhaps the theory should first be translated into a canonical formal language, such as the language of firstorder predicate logic, where truth conditions, and therefore ontological commitments, are easier to specify. But as long as these translations are controversial, the same fate befalls judgments of the ontological compromise. And, even when translated into a formal language, such as the language of first-order predicate logic, different ontological views may disagree as to the existential presuppositions of the theory. The notion of ontological compromise, introduced to help resolve disputes in ontology, becomes hostage to those very disputes. The struggle for the ontological compromises of theories begins to seem inevitable. Quine developed his ontological commitment within the framework of his theory of reference, which directly affects such issues as naming, truth, denotation and the values of variables. From his ontological point of view, a distinction must be made between theory and ideology. Thus, two theories can share the same ontology and differ ideologically.²⁴ His ontological commitment criterion focuses on clarifying which entities are accepted and which are rejected. Thus, proper names can be detailed by descriptions and can be removed. Consequently, anything that we can say with the help of names can be expressed in a lan-

 Quine, 1960, 270.  Quine, 1961, 131.

Ontological Commitment: An Introduction

15

guage that rules out names entirely. Therefore, Quine affirmed that being assumed as an entity is, purely and simply, being considered as the value of a variable. In this context, Quine postulated that a quantified entity is not guaranteed by the fact that every time a variable is linked to existential quantifier. That is why he maintained the thesis that “On what there is” does not depend on language. Quine opposed the central-European tradition, since he considered the Sachverhalt as irrelevant: he only studied the ontological commitments of discourse. Consequently, and contrary to Wittgenstein’s position, what there is does not depend in general on the use of language, although what is said to exist does depend on such use.²⁵ Quine further clarified his proposal and pointed out that a theory assumes an entity if, and only if, it must be counted among the values of the variables in order for the statements stated in the theory to be true. Therefore, we are not facing an imposition of logic upon language but rather the expression of an ontological commitment through language. It could be objected that Quine’s criterion applies exclusively to those who make a claim that something exists. Quine refuted this objection by arguing that the criterion of ontological commitment applies to the discourse and not to the user. On the other hand, it is objected that the logical criterion of ontological commitment leaves us nuances of the notion of existence since the logical unification between “there is …”, “there exists …”, and “there is something that…”, etc seems artificial. To this objection Quine replied that the criterion of ontological commitment applies mainly to scientific language and leaves aside any use in the field of natural language. This volume includes contributions by renowned authors who investigate the topic of “ontological commitment” from different perspectives using different approaches. Their contributions are briefly outlined below. Peter Simons deals with “Meinong’s theory of objectives”, which he considers clearly different from the theory of states of affairs, propositions, or what one calls objects of judgment, assumptions, and other so-called propositional attitudes. Meinong assumed that there is a distinctive category of objects that represent judgments and sentences which he called “Objektiv”, for which the author uses the English equivalent “objective”. The other main class of objects (Gegenstand) for Meinong are “Objekte”, for which the author suggests the English term “things” rather than “objects”, as the latter is the standard translation of “Gegenstand”. Things include anything presentable by an idea and nameable by a term, such as substances, qualities, properties or relationships. For Meinong, things and goals comprise everything relevant to intellection or cognition, as op-

 Quine, 1961, 103.

16

Jesús Padilla Gálvez

posed to appraisal and will. In his work he recognizes the innovative, idiosyncratic, and consistent character of Meinong’s work. The work presents convincing arguments questioning whether nonexistent objects exist. Julian Nida-Rümelin presents his work entitled “Metaphysical Aspects of Agency”, in which he aims to clarify general and abstract aspects of practical rationality that are distant from our immediate experience. Any theory of practical rationality must prove itself in concrete, particular cases and, conversely, we interpret individual, discrete cases on the basis of more abstract presuppositions. Metaphysical reflections attempt to describe the presuppositions we express in our communicative practice. Moreover, they serve to reconstruct the normative and descriptive beliefs that this communicative practice expresses. Metaphysics reflects the discovery of knowledge that we seem to have, but this is not yet explicit, and which we do not usually articulate. Nida-Rümelin proposes a general realistic interpretation of the reasons for these assumptions. He emphasizes that there is no external authority that gives us access to these propositions and rejects any point of view outside of our communication and interaction practice. Looked at that way, metaphysics is considered as immanent or related to an epistemic perspective. This corresponds to the general relation between justification and truth. The author concludes that metaphysics seeks to clarify the implicit conditions and presuppositions on which our judgments and our practice in general are based. The further we distance ourselves from familiar words and ways of speaking, the more indeterminate this project becomes. Margit Gaffal describes the sophisticated approach to an explanation of representation (“Vorstellung”) in Kasimir Twardowski’s philosophy. The Brentano School has propounded a scheme of representation, and Twardowski used the analytic method to refine its philosophical terminology. This contribution explores, first, the parts of representation including act, content and object and examines their role in the formation of judgments. Second, the author delineates the function that Twardowski ascribes to representations when it comes to investigating ethical issues. Likewise, representations are considered useful in the elucidation of aesthetic questions. Jesús Padilla Gálvez’s contribution “Ontological Commitment of State of Affairs” builds on a discussion that goes back to the anti-Kantian proposal developed by Bolzano in his introduction of the “Satz an sich”. Bolzano’s proposal clearly deviated from the idealistic tradition. As such, a gap was opened between representation (Vorstellung) on the one hand and the notion of “objective” (Objektiv) on the other hand, which belongs to the field of assumptions (Annahmen). The aim was to assume an immanent object that is intentionally given. The Brentano School has devoted extensive research to this topic. Aristotelian logic gives way to a new structure whose minor premise is a conditional structure that ex-

Ontological Commitment: An Introduction

17

presses what C. Stumpf coined as “Sachverhalt”, its English translation being “state of affairs.” This may be expressed in two different ways: as a that-clause or as something that “subsists” (bestehen). Padilla Gálvez’s investigation focuses on the content of a judgment when expressed via a state of affairs. Frege clarified that this content is predicative and Husserl went one step further and affirmed that state of affairs is a carrier of properties. Wittgenstein emphasized that when talking about states of affairs we must distinguish between object and carrier. This investigation intends to clarify the question of states of affairs as it has been discussed in detail in the German tradition. António Marques analyses the expression “I know” in language-games of Wittgenstein’s late philosophy. Many philosophical questions regarding the meaning of “ontological commitment” arise from reading Wittgenstein’s late work, not only from Philosophical Investigations (PI), but also from the latest writings on the philosophy of psychology and writings on certainty (Über Gewissheit). The sense in which it was possible to speak of ontological commitment in his late work is a general question that involves unique problems and difficulties in the philosophy of language. In discussing the complex, often ambiguous, and always fascinating exploration of “I know” language games in Wittgenstein’s later writings, one is faced with a sphere of problems that haunted the philosopher in the later years of his life. The objective now is not so much to solve them but simply to make them more precise within the framework of a task that is already an enormous effort to clarify the use of related verbs and expressions related to the sphere of “I know p” or “I’m sure of that p”. As we shall see, this task is not at all easy, given Wittgenstein’s well-known and intricate maieutic style, which seems to accrue more and more subtlety and new insights over time. Focusing mainly on the texts published in On Certainty, this contribution posits that the late Wittgenstein, after dismantling the ontology of the Tractatus, did not intend to establish a new one for different reasons. The claim is that in his later writings, Wittgenstein maintains a unique first-person element in the discussion of existential propositions within the framework of his discussion of the use of the verb “to know.” The singular qualification is based on the fact that the “I know” language games are central to the epistemic and ontological questions that Wittgenstein explores in his latest philosophical exercises. Furthermore, the “I know” language games are also unique because they are enunciated in the first person, without being expressive and, as this essay shows, they generate a space of agreement, which plays a fundamental role in our linguistic system. Manuel García-Carpintero in his contribution entitled “Against Propositional Substantivism” discusses the proposals of Jeff King, Scott Soames, and Peter Hanks regarding advanced substantive theories of propositions. He addresses

18

Jesús Padilla Gálvez

various questions the authors have raised regarding the problem of the unity of propositions. García-Carpintero presents some critical points which problematise recent substantivist proposals. Jonathan Smith asks about the consequences that Wittgenstein’s Nachlass has for an ontology. In his chapter “Wittgenstein’s Nachlass, a case for practical ontology?” he discusses the question of whether we are to take the Nachlass as the central source for the diffusion of that philosophy, either directly or by means of its later literary executors or editors. The author suggests certain questions that might be asked of the Nachlass which should clarify our understanding of it and help to use it as a source. Moreover, these questions are fundamentally ontological. Arthur Gibson critically reflects on the claim made by Stephen Hawking that philosophy is dead. According to Hawking, empirical foundations in science succeed as they enable us to produce research. Philosophy cannot meet this requirement. Consequently, this conclusion undermines any justified foundation that philosophy may provide. Gibson relativizes Hawking’s objection by arguing that the internal properties that can be detected in philosophy are also present in the sciences. Overall, this book tries to analyse a fundamental topic from different perspectives, whereby the authors raise new questions surrounding the common area of ontology. These questions open up new perspectives and highlight new research trajectories, even as they assess and engage with past and current work in the field.

References Aristoteles. 1957: Metaphysica, edited and trans. by W. Jaeger. Oxford: Oxford University Press. Aristoteles. 1982: Analytica Priora et Posteriora, edited and trans. by W. D. Ross. Oxford: Oxford University Press. Church, Alonzo. 1958: “Ontological Commitment.” The Journal of Philosophy 55:1008 – 1014. Frege, Gottlob. 1966: Grundgesetze der Arithmetik. Hildesheim: Georg Olms. Gundisalvo, Domingo. 2007: De divisione philosophiae. Freiburg i. B.: Herder. Kant, Immanuel. 1971: “Preisschrift über die Fortschritte der Metaphysik.” In Handschriftlicher Nachlaß, by I. Kant, AA XX, 253 – 332. Berlin/Boston: Walter De Gruyter. Padilla Gálvez, Jesús. 2017: Verdad. Controversias abiertas. Valencia: Tirant Humanidades. Quine, W.v.O., 1948: “On what there is.” Review of Metaphysics 2.5:21 – 38. Quine, W.v.O. 1960: World and Object. Cambridge, Mass: M.I.T. Press. Quine, W.v.O. 1961: From a Logical Point of View. Cambridge, Mass: Harvard University Press. Quine, W.v.O. 1966: The Ways of Paradox: and Other Essays. New York: Random.

Ontological Commitment: An Introduction

Quine, W.v.O. 1966a: “A Logistical Approach to the Ontological Problem.” In Quine 1966, 64 – 69. Wittgenstein, Ludwig. [1921] 1961: Tractatus Logico-Philosophicus, trans. by D. Pears and B. McGuinness. London: Routledge.

19

Peter Simons

Meinong’s Objectives Abstract: The aim of this paper is investigate Alexius Meinong’s theory of objectives. Peter Simons distinguishes this theory from the theory of states of affairs, propositions, objects of judgment, assumptions, and other so-called propositional attitudes. In this context Simons distinguishes between the two categories “objective” and “things”. The term “objective” stands for those objects that, according to Meinong, represent judgments. For the other category, the author suggests the designation “things”, which includes anything presentable by an idea and nameable by a term, such as substances, qualities, properties or relationships. The author highlights the innovative, idiosyncratic, and consistent character of Meinong’s work and presents convincing arguments questioning whether nonexistent objects exist. Keywords: theory of objectives, states of affairs, propositions, objects of judgment, assumptions, propositional attitudes

1 Two Kinds of Cognitive Object Meinong was one of several philosophers who, around the turn of the twentieth century, highlighted the status of a certain category of objects as what judgments intend and what sentences stand for. The names and roles of these objects vary from one philosopher to another, but they have in common the view that there is a distinctive category of objects that stand to judgments and sentences in much the same way that things stand to ideas and names. Among the terms used for such objects – and depending on the philosopher, the role such objects play – are: Satz an sich (Bolzano 1837), Gedanke (Frege, 1918/19), Urteilsinhalt (Brentano, 1869/70 – 1875/77; Marty), Sachverhalt (Stumpf, 1907; Husserl, 1913, 1921; Wittgenstein, 1921), proposition (Moore; Russell), and fact (Russell; Wittgenstein, 1912). Meinong’s name for this category of objects is Objektiv, and I shall use the English equivalent ‘objective’ throughout. The other major class of object (Gegenstand) for Meinong are Objekte, a Latinate term whose English counterpart I cannot use as it is the standard translation for ‘Gegenstand’. I dislike Findlay’s objectum/objecta coinage so, when I need to talk about Meinong’s Objekte, I shall use the all-purpose English term ‘thing’. Things include anything presentable by an idea and namable by a term, such as substances, qualities, properties, relations. Late in his career, Meinong added two further categories of axiological https://doi.org/10.1515/9783110750041-002

22

Peter Simons

object to these, namely the objects of feeling and emotion, dignitatives, and the objects of desire and will, desideratives. I have written about these elsewhere and will not need to refer to them further here. For Meinong, things and objectives comprise all that is relevant for intellection or cognition, as distinct from valuation and volition.

2 Introducing Objectives Objectives were not part of Meinong’s ontology from the beginning. They emerge for his readers in the first edition of Über Annahmen (1902, 21910, AMGA IV), a book Roderick Chisholm described as “without question Meinong’s best book, and […] a landmark in the history of Austrian philosophy.” In the first edition, objectives make a relatively late appearance, in the seventh of nine chapters. In the second edition of 1910 they are brought forward to the third chapter, with extensive discussion in two further chapters. It was indeed in the first decade of the century that Meinong’s gradual turn from a philosopher of psychology to an ontological realist with an unsurpassedly large ontology was accomplished, even though innovations in object theory continued to emerge through his Remington typewriter until around 1916. All philosophers had been agreed since the time of Plato that names or terms stand for things, perhaps of very different kinds, like Aristotle’s ten categories. But the question had and has never been decisively and consensually settled as to what, if anything, is signified by a complete sentence such as ‘Snow is white’ or ‘Trump is orange’. Aristotle toyed with the idea of a pragma which exists if a sentence is true and not if it is false, medievals such as Adam Wodeham and Gregory of Rimini introduced the complexe significabile, that which may be signified by a sentence, as distinct from that which is signified by a term. But it is only with the wave of interest that included Meinong that the objective correlates of sentences, call them states of affairs or what you will, came to the fore in theoretical discussion. In the second edition of Annahmen, Meinong awards himself some responsibility for this upswing, though I think he overestimates his influence.

3 Intentionality Like all Brentano’s students, Meinong accepted the thesis of the intentionality of the mental and its importance. But, beginning in 1890 with the school textbook Logik that he co-authored with Alois Höfler, he held that Brentano’s 1874 account

Meinong’s Objectives

23

in Psychologie vom empirischen Standpunkte was flawed in not distinguishing an internal, immanent object from an external, transcendent object of consciousness. Following Twardowski’s 1894 Habilitationsschrift Zur Lehre vom Inhalt und Gegenstand der Vorstellungen, Meinong adopted the terminology ‘content’ for the immanent object and reserved ‘object’ (Gegenstand) for the transcendent object. It was a disputed question, in the wake of this disambiguation, whether intentionality was to consist in having a content alone, or in having both a content and an object. Whereas Husserl took the former course, and rejected the proposition that every mental act has a transcendent object, Meinong like Twardowski opted for the latter, holding that it was the essence of mental acts to have objects as well as the contents that present them. It was this conception of intentionality, rather than any considerations about language or meaning, that opened the way to Meinong’s theory of objects. Gilbert Ryle famously claimed that Meinong’s large ontology was a result of his adoption of a naïve referential theory of meaning, the ‘Fido’–Fido theory: that the meaning of a word is the object that word stands for. This is the reverse of the truth. Meinong’s theory of meaning is based on his theory of objects and his theory of the mind, and his theory of objects is based on his theory of the mind. Meinong was convinced by Twardowski’s arguments not only that content and object are always different, but that there is never no object. It is for this reason that he rejects the Brentano–Marty term Urteilsinhalt for objectives. Because not all mental acts have really or ideally existing objects, this leads to the conclusion that some objects lack real or ideal existence. These are Meinong’s notorious objects outside being. The final quarter century of Meinong’s life sees him working out this position in its ramifications. The first modification that Meinong made to Twardowski’s theory of objects was to propose that judgements have their own special kind of object. Twardowski’s position on the objects of judgment – sketchily worked out and in any case too late to be influential on Meinong – is that they are simply the existence or non-existence of the object referred to in a sentence’s subject term. This basically Brentanian idea cannot work for compound sentences. Meinong could have adopted Stumpf’s term Sachverhalt (state of affairs), but instead coined the term Objektiv for the objects of judgement. His reasoning is that according to his linguistic intuition, Sachverhalte correspond only to correct judgments and true sentences. In fact Meinong himself contends such connotations could be overridden (as they have for many other philosophers), but Meinong was always very keen on coining his own, precise terminology. Occasionally (AMGA Erg. 220),

24

Peter Simons

he uses the term Urteilsgegenstand (object of judgement).¹ In Über Annahmen these objectives are also adopted as the objects of the episodic propositional attitude Meinong calls assuming, a propositionally-formed entertaining lacking the moment of conviction of a judgement. If I am convinced that the moon landings were a hoax, I will judge so, whereas if I am unsure but surmise that they were, the same objective (that the moon landings were a hoax) is the objective of my assumption. As I obtain further information, assumption can strengthen to judgement, or judgement can weaken to assumption, or in the face of contrary evidence give way to an assumption or judgement to the contrary (that they were not a hoax). The objective itself is untouched by my changing intentional stance towards it, and it obtains or fails to obtain independently of any cognizing subject. The same objectives also serve as objects of the dispositional state of believing. Let us call all mental acts which have objectives as their objects cognitive acts or cognitions, where they are veridical or not. Until his later theory of possibility, of which more later, Meinong divided objectives into two sorts, the true and the false, or rather, since Meinong reserves these terms for actually apprehended objectives, the factual and the unfactual. Factual objectives or facts (tatsächliche Objektive, Tatsachen) have ideal being, they subsist. Unfactual objectives (we may call them unfacts, following Meinong who says Untatsachen) have no being whatever. This is contrary to Russell’s interpretation of Meinong, since Russell confuses objectives with his and Moore’s propositions, which like Bolzano’s Sätze an sich and Frege’s Gedanken are of equal ontological status whether they are true or false. For Meinong, a truth is a factual objective that someone apprehends, a falsehood an apprehended unfactual objective. Russell’s mistake is partly understandable, since Meinong lacks a clear distinction between objectives and propositions, though in discussing Bolzano’s Sätze an sich he does eventually conclude that they are not his own objectives. But while he could have held propositions to be the ideal content of judgements and assumptions, and the senses or intensions of sentences, with objectives as their objects or referents, he does not go down this line, which is rather the one Husserl had adopted in the Logische Untersuchungen. Whereas for Husserl or the Russell and Wittgenstein of logical atomism, there is a correspondence between true propositions and facts or existing states of affairs, Meinong only has a correspondence between cognitions or sentences on the one hand and objectives on the other, not between abstract propositions and objectives. Since the simple intentionality of ideas (Vorstellungen) is different from the polar true/false intentionality of cognitions, Meinong reserves the term vorstellen

 Meinong claims (ibid.) that the term is inexact, but I think it is acceptable.

Meinong’s Objectives

25

for the former. But since they are all intentional, he needs a word for this common relational feature, so he employs the Latinate terms präsentieren and Präsentation. Unfortunately in English both vorstellen and präsentieren are translated as ‘present’, so I shall reserve the English word for the latter and say an idea stands for its object.

4 The Nature of Objectives Objectives are a species of what Meinong calls objects of higher order. These are objects which presuppose or demand other objects, or as Husserl would later say, depend on other objects. Not all higher-order objects are objectives however: some are things, such as a melody. Meinong is here influenced by Ehrenfels’s Gestalt qualities, which presuppose a foundation in simpler things. Meinong calls things which are of higher order complexes. They are complex not in merely having parts, but in presupposing other things, what Meinong calls their inferiora, in determinate ways. A melody is not just a heap or jumble of notes but is structured by relations of succession, relative length and differences of pitch. The complex exists because of the fact that its parts are related thus and so. This constitutive fact Meinong calls the complex’s complexion. Meinong’s theory went on to influence Russell’s theory of facts (as Russell clearly knew) and Wittgenstein’s theory of complexes, facts and states of affairs (as Wittgenstein probably did not know). Objectives need not be relational. The true objective that Julius Caesar exists has only Caesar himself as inferius, and the false objective that Middle Earth exists has only the non-existent object Middle Earth as its inferius. There is no second part called ‘existence’ that enters into these objectives. But most objectives are relational, and Meinong calls the universal relating factor the Relat (relatum), distinguishng it from the relational trope that binds just these things together, and from the fact of their being related, or relation. As often happens in Meinong, while the terminology may appear scholastic, it is in fact perfectly judicious, consistent and clear. A chief distinctive feature of objectives is that, unlike objects, they come in opposed pairs, positive and negative: that numbers exist, that numbers do not exist. It might be tempting to regard the things an objective is about, its subjects, as its parts, that is, to take the relationship between subject and objective as a mereological one. This is indeed how Moore and Russell understood their propositions. But Meinong wisely resists this. Firstly, a factual objective with a real subject, such as the objective that Donald Trump is married, would then be an

26

Peter Simons

ideal or abstract object with a real part, and an unfactual objective such as that Donald Trump is a socialist, would be a non-existing object with a real existing object as a part. Secondly, whereas real objects come into existence and go out of existence in time, objectives, whether factual or unfactual, are not subject to change, which they would have to be if they had real things as their parts. The relationship is a more abstract one of an objective concerning or being about its subjects. Two crucial features of Meinong’s ontology of objectives owe their formulation to Mally. One is the principle of independence, which says that an object’s characterization (how it is, Sosein) is independent of whether it is (its being, Sein). How an object gets to be how it is varies. Actual objects have many of their properties and relations contingently, and they have their existence contingently too. But by the principle of independence they would be the same objects and be the way they in fact are even if they did not exist. Ideal objects and nonexistent objects are as they are essentially, though ‘essential’ is not part of Meinong’s preferred vocabulary. Any description of an object, even an inconsistent and/or incomplete description, necessarily denotes an object which is as the description says. Meinong has a principle of freedom of assumption which guarantees this. This principle is subject to a restriction coming from Mally’s second innovation, made probably in response to Russell’s criticism. This is that there are two kinds of properties of objects: nuclear (konstitutorische) properties which make up that object’s nature, and extra-nuclear (außerkonstitutorische) which do not. Extra-nuclear properties include being, existing, being simple, being consistent, being complete, and their opposites, as well as converse intentions such as being thought about by me, being worshipped by the Jews, being described by J. K. Rowling. Such properties adhere to objects independently of our ability to describe them. That the golden mountain does not exist is outside our ability to stipulate, but that it is golden and a mountain are parts of its nature. Late in his life, Meinong subjected his cognitive object theory to several further refinements. In response to liar-type and other antinomies he excluded certain putative objects, the so-called defective objects, from his domain. He elaborated on the role of incomplete objects as assistant or auxiliary objects of our cognition, enabling us to grasp independently existing complete objects as their ultimate targets. This enabled him to introduce a distinction between a term’s meaning or sense on the one hand and its ultimate referent on the other, a distinction familiar from Frege but ironically abandoned by Meinong’s chief critic Russell. And finally he used this distinction between auxiliary and target objects to uncover a role for objectives in providing an objective foundation for probability, which we shall discuss below.

Meinong’s Objectives

27

5 The Correlation of Content and Object There is a feature of Meinong’s account of the mind and its objects that, like the elephant in the room, is so patent it may go unremarked. This is that there is a remarkable general correlation between the parts and moments of the contents of mental acts and the parts and moments of the objects they present. Thus things, like ideas, lack positive/negative polarity, but objectives have such polarity. The correlation concerns contents, not the acts of which the contents are moments. The moment of conviction in a firm judgement does not correlate with factual objectives have. It lies fairly close to standard German idiom, as in Es besteht einen Unterschied zwischen Hunden und Wölfen –There is a difference between dogs and wolves. So Meinong divides being (Sein) into real (wirkliche) existence (Existenz) and ideal existence of subsistence (Bestand). Objectives can at best have Bestand. Note that unlike Russell I am not following Meinong in restricting the term ‘exist’ to real i. e. spatio-temporal-causal things, but using it in the more general sense of modern logic and philosophy, where it coincides with Meinong’s Sein. Fourthly, whereas things may or may not exist, without being in any qualitative way different (Mally’s Principle of the Independence of Being from BeingSo), objectives have their factuality or unfactuality inherently and primitively (Findlay 78), and Meinong takes factuality to be indefinable. Finally, no objective is temporal or in relation to any specific time. If I judge that Donald Trump is the present President of the USA, that this is so is not tied to a time, but is timelessly so, because the indexical element (‘present’) relates not to the objective but to the time of the judgement, or, if I state it, to the time of the utterance. It is timelessly the case that Trump is President simultaneously with the judgement or utterance, and timelessly not the case that he is not then not President. Finally Findlay teases out of Meinong the sixth point that things are only given for judgement and assertion through objectives. Findlay also firmly and correctly rebuts Russell’s identification of Meinong’s objectives with his own and Moore’s propositions, which exist whether they are true or false. Meinong’s factual and unfactual objectives differ not through a quality of truth or falsity (as in early Russell and Moore), nor through correspondence or lack of correspondence with some fact (as in later Russell, Moore and Wittgenstein), but in ontological status: factual objectives exist (have being, subsist, bestehen), unfactual ones do not. This is a stark existential abyss between facts and unfacts, and nothing on the side of propositions, Gedanken, Sätze, corresponds to it. This means that Meinong’s theory of truth is not a correspondence theory, but an identity theory (Findlay 88): when I judge or state truly, the object of

28

Peter Simons

my judgement or meaning of my statement does not correspond to something that makes it true: it itself is a truth.

6 Modal Complications In mid-Meinong, objectives divide into the factual and the unfactual, tertium non datur. But in the gargantuan late work Über Möglichkeit und Wahrscheinlichkeit (1915) a complication enters. Meinong is intent there on providing a theory of objective and subjective probability. The latter, subjective probability, is what Meinong calls ‘Wahrscheinlichkeit’, whereas he uses the term ‘Möglichkeit’, possibility, for what we would call objective chance or likelihood. Unlike standard modern theories of alethic modality, for which necessity and possibility are dual concepts, for Meinong possibility comes in degrees, and it does so because the things to which possibility applies are what he calls incomplete objects, that is, things which are indeterminate with respect to one or more properties or determinations. Take the expression ‘the next Prime Minister of the United Kingdom after Boris Johnson’. In one sense, this designates a fully determinate individual who is as yet unknown as falling under that predicate. In another sense however, it determines nothing that is not present or implicit in the description. So for example the prediction ‘the next PM of the UK will be a woman’ is neither true nor false, but somewhere in between in truth value if the subject term stands for the incomplete object, but is determinately true or false if it stands for that person, whoever it turns out to be, who will be the next PM, even though we do not now know whether it is true or false. Suppose, on the basis of certain facts about the current British political situation I hold it to be 20 % likely that the next PM will be a woman, and suppose again for the sake of argument that these facts do indeed make it objectively 20 % likely. Then Meinong would say that the possibility of the objective that the next PM will be a woman is 20 %, or 0.2. I may in fact be mistaken about this likelihood, in which case my subjective estimation of the possibility will not coincide with its objective value, but Meinong is adamant that there is an objective value to the likelihood, whether I or indeed anyone knows it or not. The objective then has a degree of factuality properly between objective truth, that is, factuality of degree 1, and objective falsehood, factuality of degree 0. When an objective has factuality p, its opposite objective has factuality 1 – p. The in-between degrees of factuality form a linear series that Meinong calls the factuality line. He calls this graded notion increasable possibility, and it can have in principle any one of infinitely many values. On the other hand there is also an ungraded notion that he calls unincreasable possibility, which is expressed by the bald statement that such and such is possible

Meinong’s Objectives

29

simpliciter – as distinct from being possible to this or that degree. Something is possible in the unincreasable sense if and only if its increasable possibility lies properly between 0 and 1. Notice that there is nothing more possible than plain truth or factuality. For Meinong, there are no necessary truths that have a stronger kind of truth than ordinary truth, and no necessary falsehoods which have a stronger kind of falsehood than ordinary falsehood. In Meinong’s view, we call something necessary when the factuality or unfactuality of an objective is built into its terms, what Meinong calls inhesive factuality or unfactuality. So the objective that the round square has being is inhesively unfactual, while the objective that it does not have being is inhesively factual, as is the objective that 6 = 2 x 3. In thus distinguishing but relating two notions of possibility, a gradable one and an ungradable one, and in denying that the law of excluded middle always holds, Meinong is in fact anticipating views later made the cornerstone of many-valued logic by Jan Łukasiewicz, who said in 1930 that there were only two kinds of many-valued logic that bear intuitive scrutiny, a three-valued one where the intermediate value of possibility is ungraded, and an infinite-valued one where the intermediate values form a continuum. As I have shown elsewhere, this similarity is not accidental, because Łukasiewicz spent time in Graz in 1908 – 9 when both he and Meinong were working on the foundations of probability, as indeed was Ernst Mally. It remains to be seen how objectives with a degree of factuality between 0 and 1 – Meinong calls them ‘subfactual’ (untertatsächlich) get their degree. Meinong’s theory is rather simple, and works only for straightforward cases. It is this. Take the objective that the next card drawn at random from this complete pack will be a spade. Mere combinatorics show this to have the value 0.25, as a quarter of the cards in the pack, 13 out of 52, are spades, and there is no reason to suppose any is more likely to be drawn than any other. This is the old Bernoulli assumption of equiprobability of outcomes, and Meinong never gets far beyond it. His examples include throwing a die, or picking a ball “at random” from an urn, standard introductory examples, which invite frequentistic or collectivistic analyses. Empirical probabilities are given a statistical basis: if 17 % of cars on British roads are German, then the objective that the car on the British road is German, with that incomplete object as subject, has a factuality value of 0.17. The incomplete object is said by Meinong to be implected in (we might say: embedded in) the several complete existing objects which have all its characteristic properties (and more besides). Cases where the degree of possibility depends on many divergent factors, as that of the chance of the next UK PM being a woman, lie beyond Meinong’s theory. It has to be admitted that, for all its object-theoretic ingenuity, Meinong’s theory of probability does not ad-

30

Peter Simons

vance beyond elementary concerns, something which he himself modestly – if verbosely – admits: “Looking back, one cannot be mistaken in thinking that the here briefly summarised main discussions were nowhere able to go beyond the domain of the elementary problems of possibility and probability theory”² – though this admission comes only on the last of 728 pages. The other question we have to consider is whether subfactual objectives, being neither existent like factual ones, nor non-existent like unfactual ones, have intermediate degrees of being. Meinong is happy to talk about degrees of factuality, and appears also to accept that subfactual objectives have what he calls a Seinshöhe, that is more than zero and less than one, but he baulks at calling this a subfactual kind of being, but rather sees it as a property or determination of being (112). Meinong twists and turns on this question and I have considerable difficulty following his tortured discussion, but I fail to see how he can both maintain the intermediate status of subfactual objectives and at the same time maintain that the difference between factual and unfactual is one between being and non-being, tertium non datur. Perhaps his best move would be to say that all objectives except unfactual ones have being, but that the degree of being, or whatever one wants to call Seinshöhe, varies from one down to but not including zero. Meinong gestures briefly at this idea in his posthumously published Selbstdarstellung, when he says that a subfactual objective with possibility p less than 1 (which he calls its chief or main possibility (Hauptmöglichkeit)) also has all the lesser degrees of possibility. He calls this the Law of More (potius), and it applies to all lower degrees except zero, which he calls the worse (deterius). But then it seems arbitrary to deprive unfactual objectives of all being, and if they had being too, then objectives would in effect be propositions, and Meinong’s distinctively ontological distinction between the factual and the unfactual would be gone.

7 Conclusion Meinong’s theory of objectives is clearly different from everyone else’s theory of states of affairs, propositions, or whatever one calls objects of judgement, assumption and other so-called propositional attitudes. As is typical with Meinong, it is innovative, idiosyncratic, and consistent, or at least, not obviously in “Rückblickend ist nicht zu verkennen, daß die hier kurz wiedergegebenen Hauptaufstellungen das Gebiet der elementaren Möglichkeits- und Wahrscheinlichkeitsprobleme nirgends zu überschreiten vermochten.” This is a typical Meinong sentence, for which the best description is the German word umständlich: not unclear, but tortuous.

Meinong’s Objectives

31

consistent. That does not make it correct, of course. As I have made clear in other publications, I do not believe there are non-existent objects, nor do I believe there are states of affairs. But those are other stories.

References Bolzano, Bernard. 1837: Wissenschaftslehre. Versuch einer ausführlichen und größtentheils neuen Darstellung der Logik, edited by J. Ch. A. Heinroth. Vols. I – IV. Sulzbach: Seidelsche Buchhandlung. Bolzano, Bernard. 1973: Theory of Science, edited with an introduction by Jan Berg, trans. by Burnham Terrell. Dordrecht: Reidel. Brentano, Franz. 1869/70 – 1875/77: Logik, Manuskript EL 80. Würzburger Vorlesungen. Wintersemester 1869/70. Revisionen Vorlesungszyklen in Wien in den Sommersemestern 1875 und 1877. [19/09/2017]. Brentano, Franz. 1973: Psychology from an Empirical Standpoint, edited by Linda L. McAlister, Antos C. Rancurello, D.B. Terrell and trans. by Linda L. McAlister. London: Routledge. Frege, Gottlob. 1918/19: “Der Gedanke.” Beiträge zur Philosophie des deutschen Idealismus 1:58 – 77. Husserl, Edmund. [1913]1968: Logische Untersuchungen. Erste Teil: Prolegomena zur Reinen Logik. Tübingen: Max Niemeyer Verlag. Husserl, Edmund. [1921]1968: Logische Untersuchungen. Zweite Teil: Untersuchungen zur Phänomenologie und Theorie der Erkenntnis. Tübingen: Max Niemeyer Verlag. Meinong, Alexius. [1901]1977: Über Annahmen. In Gesamtausgabe, Vol. IV, edited by R. Haller, R. Kindinger and R. M. Chisholm, 1 – 384. Graz: Akademische Druck- und Verlagsanstalt. Meinong, Alexius. [1902]1977: Über Annahmen, Zeitschrift für Psychologie und Physiologie der Sinnesorgane, Ergänzungsband 2, Leipzig 1902, in: Alexius Meinong, Gesamtausgabe, Vol. IV, edited by R. Haller, R. Kindinger and R. M. Chisholm, 385 – 516. Graz: Akademische Druck- und Verlagsanstalt. Meinong, Alexius. 1969 – 1978: Gesamtausgabe, edited by R. Haller, R. Kindinger and R. M. Chisholm. Graz: Akademische Druck- und Verlagsanstalt. Stumpf, Carl. 1907: Erscheinungen und psychische Funktionen, Abhandlungen der Königlich Preußischen Akademie der Wissenschaften, Philosophisch-historische Klasse vom Jahre 1906.4, 1–39. Berlin: Verlag der Königl. Akademie der Wissenschaften. Twardowski, Kasimir. [1894]2016: Zur Lehre vom Inhalt und Gegenstand der Vorstellung. Eine psychologische Untersuchung. Viena: Verlag Alfred Hölder. In Kasimir Twardowski. Gesammelte deutsche Werke, edited by Anna Brozek, Jacek Jadacki and Friedrich Stadler, 39 – 122. Berlin: Springer. Wittgenstein, Ludwig. [1921]1961: Tractatus Logico-Philosophicus, trans. by D. Pears and B. McGuinness. London: Routledge.

Julian Nida-Rümelin

Metaphysical Aspects of Agency Abstract: The aim of this chapter is to clarify general and abstract aspects of practical rationality that are distant from our immediate experience. Any theory of practical rationality must prove itself in concrete, particular cases and, conversely, we interpret individual, discrete cases on the basis of more abstract presuppositions. Metaphysical reflections attempt to describe the presuppositions we express in our communicative practice. Moreover, they serve to reconstruct the normative and descriptive beliefs that this communicative practice expresses. According to this point of view, metaphysics reflects the discovery of implicit knowledge. The author proposes a general realistic interpretation of the reasons for these assumptions. The author concludes that metaphysics seeks to clarify the implicit conditions and presuppositions on which our judgments and our practice in general are based. Keywords: practical rationality, metaphysic, agency, communicative practice, beliefs, implicit knowledge

1 Metaphysics Metaphysics in the sense proposed here is not a foundational discipline. It does not precede other investigations but rather follows them. Metaphysics is tentative because it describes general characteristics of our practice of judgment and action that are mostly not apparent. Metaphysical investigations in our sense are remote from observation but they are not independent of experience. Metaphysics occupies an extreme position in the gradual spectrum ranging from empirically supported individual judgments (including moral judgements based on moral experiences¹) all the way to abstract generalizations that are in need of corroboration within the complete system of beliefs (both descriptive and normative). This emphasis on a nonfoundational role of metaphysics must not be misunderstood as a reversal of the justificatory relation: Even judgments that are close to experience, both descriptive and normative, do not stand for themselves and play no epistemically foundational role, not even in their entirety. Usually, even empirical data are not a given but are the result of interpretations that rest on implicit and explicit background assumptions. This is what we mean  Cf. Nida-Rümelin, 2018; and: Nida-Rümelin, 2016. https://doi.org/10.1515/9783110750041-003

34

Julian Nida-Rümelin

when we refer to the theory-ladenness of observation, which applies to exact natural science such as physics as well. According to this holistic and coherentist perspective, the dispute between particularism and principlism in metaethics – i. e., rationalist theories such as utilitarianism or Kantianism versus the particularist positions, for example by Jonathan Dancy or Bernard Williams – is obsolete. Principles (for example invariance assumptions such as the isotropy of space in physics or the principles of equal treatment in ethics) play no ultimate role for justification, neither immediate empirical or moral experience. In our everyday discourses, and, to some extent, in the individual sciences, we go back and forth and try to make the entirety of our practice of judgments and actions more coherent. We revise individual judgments that are close to our experience if they come into conflict with well-established principles, and we give up principles if they conflict with central individual judgments that are close to experience. We interpret experiences in light of structuring rules and modify the rules in light of experience. This chapter of the book is nothing but the clarification of the more general, more abstract aspects of practical rationality that are remote from experience. Any theory of practical rationality must prove itself in concrete individual cases, and we interpret individual cases on the basis of more abstract presuppositions. The metaphysical reflections in this final section are not aimed at merely describing the presuppositions of our communicative practice, but rather at uncovering and reconstructing the normative and descriptive beliefs that are expressed by this communicative practice. Metaphysics as Socratic maieutikè, as the uncovering of knowledge we already have, but that is not yet explicit, and that we usually do not articulate. The status of this metaphysical opposition is not relational; it is not about identifying specific presuppositions of our linguistic usage. Rather, the interpretation of these presuppositions is realist and in accordance with the general realist interpretation of reasons. At the same time, there is no external authority that provides us access to these propositions, no standpoint outside of our practice of communication and interaction, no fundamentum inconcussum. In this regard, this metaphysics is immanentist, a metaphysics from an epistemic perspective. This status corresponds to the general relation of justification and truth, as I sketched it in the first part (I § 3).

2 Freedom In what follows, human freedom is characterized as a metaphysical proposition. Freedom is a complex of presuppositions, of implicit conditions for our practice of communication and action, which will be gauged against this practice. Free-

Metaphysical Aspects of Agency

35

dom is not merely the existence of alternative possibilities. The availability of alternative possibilities, the option to decide differently, is inextricably interwoven with communicative practice and action, and is itself in turn only an element of a greater complex of concrete and abstract propositions. These have a metaphysical status insofar as they cannot be confirmed or disproved by individual experiences. In other words: human freedom (even in the traditional distinction between freedom of action and freedom of the will) is a structural and implicit characteristic of human practice. In his Critique of Practical Reason Kant is right to say that from a first-person perspective, we have to understand ourselves as free because this is what constitutes our character as rational beings. However, he is wrong in claiming that this first-person perspective is isolated in the sense that from a third-person perspective the principle of causality renders the laws of freedom irrelevant. Indeed, this form of naturalist integration of our experiences of freedom, our self-efficacy, our responsibility, are non-abandonable since without this implicit characteristic feature of our practice, the exchange of reasons would be incomprehensible or be nothing more than a theatre of illusion. Freedom is not merely a feature of the subjective perspective but also of the objective perspective. It cannot be a subjective feature, i. e., be limited to the first-person perspective, to the noumenal I, because the practice of giving and taking reasons refers to (objective) propositions that should be interpreted realistically, and that itself has a truth claim. A realist theory of reasons, as it is developed in this book, is incompatible with a subjectivist interpretation of human freedom. Indeed, such a theory conceives of freedom as nothing other than the capability of being affected by reasons. In other words: Reasons (and their objective and normative character) presuppose human freedom. Reasons that are objective and normative but play no affective role, i. e., do not translate into affections, emotive judgements and guidance for our practice, which in turn means that they become action-guiding and thereby causally effective, would not be real. Their existence would be limited to the realm of ideas, the Third Realm in the sense of Frege or Popper. The close conceptual connection between freedom, responsibility, and reason shows that the phenomenon of human freedom is not merely a pointwise, isolated phenomenon. As a first approximation, we are dealing merely with different aspects of the same phenomenon, namely the capability of being affected by reasons. This, in any event, is the guiding thesis of the following paragraphs. At first blush, this thesis appears somewhat eccentric, possibly counter-intuitive. Does freedom not simply consist in the possibility of acting one way or another, independently of the individual reasons one might have for acting a certain way? Indeed freedom of choice [Willkürfreiheit], as we will call it in the following, i. e., the simple possibility of remaining seated or of getting up, lighting the next cig-

36

Julian Nida-Rümelin

arette or not (for non-addicts), leaving a soiree now or in an hour, once the concert is over, etc. is what seems to be at the center of our experience as free, or perhaps one should here say autarkic, beings. Independently of whether or not I have my reasons for what I do, I can decide how I behave. There is an interesting linguistic parallel between Ancient Greek and English, viz. the characterization of this phenomenon as eph hemin, which corresponds almost exactly to the English phrases up to me, or up to us, it’s up to me, it’s up to us. For the stoics it was of central importance to grasp precisely what eph hemin encompasses, since one’s responsibility extended only as far as the realm of eph hemin. The demanded attitude towards everything else, i. e. the adiaphora, was indifference. It is stoic thinking that initially linked reason to freedom, namely with the idea that humans are only free to the extent that they define what is eph hemin according to rational insight. There is a direct line leading from the Stoics to Kant. Mere freedom of choice [Willkürfreiheit] is not the freedom of rational beings. Those who merely arbitrarily act in one way or another are still part of a heteronomous order. Indeed, arbitrary action is like a random mechanism, it does not have the character of a reasonable practice guided by maxims. This close connection between reason and freedom, both in Stoicism and in Kant, can be criticized with reference to the tradition of understanding human freedom as autarky. The ancient heroes in the Iliad and the Odyssey, the most likely exception being Odysseus, seem to be guided by impulse rather than reason – they act on the spur of the moment, and often regret what they have done. Achilles is considered the greatest hero of Greek mythology and nevertheless it is him, who is unable to give his life practice as a whole a coherent structure. Fully aware of what he is doing, he watches the Greeks rush headlong into disaster, with hundreds and thousands of deaths, until a new, existential experience gives his life a new direction, namely the death of Patroclus, his closest friend. His grief incites his disproportionate revenge against the greatest hero of the opposing side, Hector: he slaughters him like a butcher would kill a beast and desecrates his corpse by dragging it all the way back to the Greek camp. According to the greatest hero, autarky means having the possibility to yield to one’s momentary impulses and of not letting anyone tell him what to do. He is free, but only in the reduced form of the realization of momentary impulses. The whole of this life practice makes no sense: Achilles stumbles from one regret to the next and ultimately pays the price for this lifeform with an untimely death and the dirges in Hades, with which Odysseus is later confronted. Odysseus by contrast, the smart and thoughtful person who is guided by reasons, prefers to spend a long life in the light, although Odysseus’ sophistries and cunningness were considered a problematic character trait in Greek Classics. Indeed, his tactical and strategic moves throughout his odyssey

Metaphysical Aspects of Agency

37

were always self-interested; the motive for his actions was always his own advantage, his cunning way of overcoming unexpected challenges. It is only the embedding of practical reason in the structures of human coexistence and the cosmic order of the logos within Stoicism that breaks up this egocentrism. This applies to the egocentrism of Achilles, who follows his momentary impulses, and the egocentrism of Odysseus, who follows a calculus of self-interest throughout his lifespan. Now, the individual agent has to stand accountable for what she has done, she has to weigh the reasons that speak for or against a certain action; she might even become melancholic or depressive in the process, as did Marcus Aurelius, but the main goal of action remains the embedding of one’s own practice within the greater, rational and insightful context of the logos. The ideal of autarky is superseded by the ideal of autonomy brought forward by the European Enlightenment. Arbitrary freedom of choice [Willkürfreiheit] is pushed from the center to the periphery. We can now systematically characterize the relation between autarky and autonomy: Autarky is the condition for autonomy. It is the possibility of choosing between different options that enables practice to be affected by reasons. The control of my own action has to take place through something that I can identify with, because otherwise it is not I who is acting; my behavior would then merely be the causal consequence of external influences. This is the basic idea that led Kant to juxtapose heteronomy and autonomy. We identify with the reasons that we have for beliefs, actions, and emotive attitudes. Typically, we likewise have reasons for dispensing other reasons. However – and this makes our analysis more complicated – our own decisions belong to the conditionality of our reasons for action. A decision, one could say, generally generates reasons; and as this generator of reasons, the decision affects my practice beyond the respective temporal moment at which I make a particular decision. It generates reasons that reach beyond the temporal moment of the decision and that structure my practice. Let us consider the example of a person being indifferent to a certain outcome after having evaluated the pertinent reasons: Let us further assume that even after careful consideration the reasons in favor of a short trip to Rome are about the same as the reasons in favor of a short trip to Madrid. In order to avoid being forced into inactivity like Buridan’s Ass, the individual makes a decision. My decision to go to Madrid would in turn create a cascade of good reasons for certain actions. Ranking among these might be that I search for a hotel in Madrid and make a reservation, that I book a flight from Munich to Madrid, etc. Once I opt for Madrid, the reasons that previously spoke in favor of Rome as another potential destination for my trip have become obsolete. There is no irrationality at play, but rather, autarky, the freedom of choice [Willkürfreiheit]

38

Julian Nida-Rümelin

to arbitrarily decide between options that I am indifferent about, and this voluntary choice entails a cascade of reasons for action. Freedom of choice [Willkürfreiheit] generates reasons. This example shows that freedom of choice [Willkürfreiheit] is relevant in two regards: On the one hand, it is a condition of autonomy (or in short: autonomy requires autarky), and on the other hand, it is a structuring and reason-generating factor within practice. Without freedom of choice at its center, the conception of autonomy as the freedom or ability to be affected by reasons, as it is defended here, would run into difficulties, at least in connection with a realist interpretation of reasons. If reasons are characterized as normative facts for which subjective features such as preferences, hopes, fears, consideration of others, etc. play a role, but which are not themselves the result of an activity of an individual then individuality, what is unique and what is special about a person, would be dissolved in the objective demands of practical reason. If Kant’s Categorical Imperative always resulted in only one generalizable option, the commandments of morality would be detached from any individuality. And since we do not reduce autonomy to morality, as Kant did, but rather expand it to the entirety of practical reasons, this loss of individuality would be unacceptable. The commitment to certain projects, i. e., the structuring of one’s own life and of one’s own practice through endeavors that I myself have chosen, is merely a special aspect of freedom of choice [Willkürfreiheit], it characterizes the structuring of autonomous practice through autarkic decisions. Now, we can reformulate more precisely and adequately modernity’s idea that there is a categorical difference between theoretical and practical reasons; that ultimately there are either only theoretical reasons, as claimed by David Hume, or that practical reasons can be reduced to the desire-belief scheme, as contemporary Humeans assume; or that practical reasons are nothing but cultural characteristics of a respective established lifeform (in communitarianism) or, perhaps the most prevalent idea, that there are only two kinds of practical reasons, namely those generated by self-interest and those generated by moral obligations, e. g., the utilitarian principle to maximize utility: It is autarky (or freedom of choice [Willkürfreiheit]) which constitutes the difference between theoretical and practical reasons. One could also say, it is merely the phenomenon of freedom of choice [Willkürfreiheit] that marks this difference, and not the mode of justification, not the logic of justification, not the direction of fit, which marks the supposedly categorial difference between theoretical and practical reasons.

Metaphysical Aspects of Agency

39

Kant² distinguishes between laws of freedom and laws of nature. The first are self-imposed, whereas the latter are given by the natural order. Both enable the application of the principle of causality to individual events (to actions or natural processes, respectively). Laws of freedom delimit the space of autonomous action, natural laws the domain of the heteronomous. Both domains are dichotomous; that is to say, where humans do not, out of respect for moral law, act autonomously, they act heteronomously and remain part of the order of nature. This seems to me to be quite obviously false: From a first-person perspective, also those actions not guided by moral motives have reasons speaking in their favor. Even a person who is not morally motivated has reasons for what they do. Their behavior is not merely a causally determined natural process. If that were true, would not have the character of an action; for a behavior for which the affection by reasons does not play any role does not have the character of an action. To the extent that we can interpret some behavior as an action, we assume that a person has reasons by which it is affected. The status of personhood does not begin only with moral motives, as Kant and contemporary Kantianism assume. (Human) freedom manifests itself in actions – in any kind of action. If I am accosted by a person wielding a knife on a lonely street late at night and this person demands that I hand out my wallet, I will probably decide to do this immediately. This behavior doesn’t follow a stimulus-response pattern, it is a well-considered action. I have weighed the pros and cons and have reached the conclusion that a refusal would entail a high risk for my health, perhaps even for my life, and that this possible damage is grossly disproportionate to losing my wallet. Perhaps moral reasons may factor into my decision; for instance, the consideration that it would be irresponsible to risk my life if my family would suffer from this loss. But even if moral motives were completely irrelevant for my decision, handing over my wallet was guided by reasons, it was reasonable and an expression of my freedom to choose. There is an obvious objection to this example: I was not free in my decision since I was forced to hand over my wallet. This type of coercion is part of the specific conditions that shape the situation in which I acted, and which generated reasons for which I acted, in this case reasons that speak in favor of handing over the wallet. This coercion does not belong to the natural order, it is not an expression of a law of nature. Action presupposes freedom. The coercion in this example has no impact on my status as an agent. I am still capable of considering and choosing between different options. The drug addict, whose urges for the next shot are so strong that he loses this ability, the ability to weigh

 Cf. Kant, 1781.

40

Julian Nida-Rümelin

the pros and cons of various options and to be guided by the subjectively best reasons, has lost his freedom and a limine his agency. He appears to us as unaccountable. Human freedom, then, is constitutive for responsibility for own actions in a two-fold sense. Firstly insofar as behavior without freedom of choice, without the possibility of choosing from different options, without being able to weigh reasons for or against one or the other option, cannot be ascribed to the respective person as an action and thus fails to be an object of criteria of responsibility. Behavior that cannot be characterized as action is outside the control of the acting individual and the person can therefore not be made responsible for her behavior. The scope of freedom delimits the scope of what is ascribable to a person and thus the area for which the subject bears responsibility. The person chooses an action from a set of possible actions (the scope of her freedom is determined by what is possible) and is therefore responsible for having preferred one particular action over every other possible action. She can be held accountable for not having chosen different options of action, but she cannot be held accountable for failing to perform actions that are not part of the set of possible options. We are responsible precisely for that which is under our control, i. e., for that which falls in the scope of what we could have realized. We are thus – by derivation – also responsible for all foreseeable consequences that result from the decision for a particular option as well as for all the foreseeable consequences not realized because we refrained from choosing other options for actions. However, when it comes to criteria for criminal responsibility, theoretical freedom becomes relevant as well, i. e., the freedom to weigh reasons for different beliefs, to form own beliefs on the basis of one’s own power of judgment. The criterion one could have known is of great importance in legal contexts and should be interpreted as an expression of theoretical freedom: The person could have gained knowledge about something, and this criterion for responsibility applies, provided it can be considered reasonable that the respective person makes the applicable belief her own. One could have known does not merely mean that the person would have known this under certain circumstances, but rather that given her cognitive abilities and her epistemic conditions she was, in principle, capable of making a correct judgment. One could have known implies firstly that there is an (objectively) undisputable state of affairs and that it was possible and reasonable to expect from the respective person to reach a justified judgment that corresponds to this state of affairs. Thus, there are two claims here: A descriptive one, namely that there is in fact a certain state of affairs, and a normative one, namely that the respective person should have familiarized herself with this state of affairs. These claims contain both an empirical judgment (the respective state of affairs obtains) as well as a normative judgment

Metaphysical Aspects of Agency

41

(the person was obligated to make this belief her own). However, even this theoretical freedom has a gradual character, which is occasionally characterized in everyday language by the expression obviousness: something was “obvious” or “less obvious”. This expression is quite adequate because it takes into account the degree of revision that is necessary in order to integrate the insight of the respective state of affairs. The more work it takes is to integrate the respective judgment at a certain point in time, the less “obvious” it was to do so. Interestingly enough, almost nobody contests the existence of theoretical freedom. Those who would deny the existence of theoretical freedom would on their part expose themselves to the criticism of getting tangled up in a performative contradiction, because arguments are brought forward with the expectation that reasons are relevant for the beliefs a person holds – which is precisely what they would be denying. This speaks in favor of placing theoretical freedom at the center of human freedom.

3 Responsibility The close relation between responsibility and freedom outlined results from the role that reasons have, they constitute both responsibility and freedom: I am free insofar as it is my reasons that guide my action (my beliefs, my emotive attitudes). I am responsible for that which is affected by the reasons I make my own and that which is a consequence of my being affected by reasons. This is the indicative wording. In some situations, the grammatical form of the potential mood is more adequate for characterizing (human) freedom and (human) responsibility, respectively: I am free, if it is possible for me to follow the better reasons (the result of an evaluation of reasons). I am responsible insofar as I had the possibility to weigh reasons and follow the result of this evaluation. Responsibility and freedom are equal in extension; those who can be held responsible, are also free and vice versa. Are the predicates free and responsible equal in extension in all possible worlds, which makes them equal in intension as well? I am inclined to answer this question in the affirmative. A specific aspect of responsibility is culpability. I am culpable if I have done something wrong even though I had the opportunity, based on an insight of better reasons, for a different and better action. Interestingly enough, this is true not only with regard to practical reasons but theoretical reasons as well: Ignorance does not relieve me of my responsibility if I could have known what spoke against this wrong action. This capability is normative, not empirical: it is not to be interpreted in the sense that one can imagine a different empirical world in which I would have known this. It was reasonable to expect of me

42

Julian Nida-Rümelin

that I familiarize myself with the momentarily relevant state of affairs, that I inform myself accordingly, that I actualize my power of judgment, etc. The scope of responsibility depends on normative and empirical conditions. These conditions have an objective character, they do not exclusively refer to the subjective, epistemic, or prohairetic state of the respective person. A person walks up to a riverbank at a certain moment and sees a human body in the water. He do not jump into the river and try to save the person, despite being a strong swimmer. Instead, he goes to a nearby intersection and inform a police officer standing there about his observation. The police officer requests help by making an emergency call. The emergency responders determine that the person drowned at least an hour ago, which means that the passerby would not have been able to help when they noticed the body in the water. Nevertheless, he can be subjected to criticism: he should have tried to save the person since he could not have known at this moment whether the person was still fighting for her life or had already drowned. If he did not think he was up to the task, he should have called the emergency hotline right away instead of wasting precious time by walking to the next intersection. He had alternative options for actions available and we blame him for not having made a better choice. He is responsible for what he did. Provided we are talking about an adult and not a child, we will probably not accept the excuse that he was so confused and that is why he did what was wrong. In this case, a rescue attempt or at least an immediate emergency call would have been the right thing to do, and this is what we can expect of a grown-up, fully accountable person who, in addition, is a good swimmer. At the same time, however, we will not hold him responsible for the other person dying, since she were already dead when the passerby noticed the body in the water. It is important to note that this state of affairs determines the scope of responsibility, although the passerby had no epistemic access to it (he could not have known). Nevertheless, this state of affairs limits his responsibility. The limitations of the domain for which we are responsible depend not only on the epistemic state of the acting person and the normative criteria for assuming responsibility, but also on objective possibilities for intervention. It was not possible anymore for the passerby to prevent the death of the person in the water, and we therefore cannot make him responsible for this person’s death. However, if the person in the water had in fact still been fighting for her life, the passerby would have been at least partially responsible for her death, Perhaps other agents are also responsible: the drowning person, who knew of her poor bodily constitution, the local authorities, because they did not provide sufficient information about the dangers of bathing in this stretch of the river, the primary care physician, because he did not inform his patient about risks after a recent cardiovascular checkup, etc.

Metaphysical Aspects of Agency

43

At this point, one might be tempted to divide the concept of responsibility into a subjective and an objective part. Accordingly, the scope of an agents’ objective responsibility would cover their options for intervention, given the empirical conditions. In our example, this would correspond to the limitation of the scope of responsibility by the empirical state of affairs that the respective person was already dead at the moment of observation. She could then – de facto – not have been saved anymore and therefore the passerby cannot be held responsible for the death of the person in the water. Even if the subjective epistemic situation presents itself unchanged, we would have made the passerby (partly) responsible for the death of the drowning person if she had not been dead at the moment of observation. Subjective responsibility would then be relativized to the epistemic and prohairetic states of the acting person, whereas objective responsibility takes into account the factual possibilities of intervening and the factual options for action. This solution, however, would be too simple; it does not lead to a coherent conception of responsibility. We should speak of subjective responsibility only in the sense of an epistemic characterization: The person was subjectively not responsible if he assumed that he had no possibilities at all of intervening in the course of events. Or if he assumed that was not his responsibility or duty to act in a certain way. In both cases it is an opinion that the person holds, an empirical one in the first case and a normative one in the second case. In the first case, he was convinced that there was empirically no possibility of intervening, in the second case he was convinced that he had no duty to intervene. Such opinions can turn out to be false. And we inspect whether they are correct or false based on objective conditions that are either empirical or normative. We verify whether it was actually the case that he could not intervene or that it was not his duty to do so. Subjective responsibility would then be nothing but an opinion of the respective person regarding the scope of his responsibility. Just as subjective facts are not facts, but rather beliefs about facts, subjective responsibility is not responsibility, but rather a belief about responsibility. Both facts as well as responsibilities are objective, even if normative criteria are at play. The same holds for freedom. A subjectively free person thinks that she is free. A person is free in terms of her beliefs, if by evaluating reasons she is, in this sense, able to reach rational convictions. She is unfree in terms of her beliefs if she is unable to do this. We call beliefs pathological if they remain completely unaffected by reasons. These beliefs are resistant to any form of revision and are upheld even if in the face of unequivocal evidence of them being wrong. Actions are pathological if they cannot be affected by reasons. This is what psychologists call “compulsive behavior” or “neurotic disorders”. Persons that perform such actions frequently suffer from pronounced neuroticism. Emotions are patholog-

44

Julian Nida-Rümelin

ical if they cannot be affected by reasons. Resentments form a specific class of such emotions; we can find a convincing description of their genesis in the works of Max Scheler.³ Zeno of Citium, the founder of the Stoic school, did not want to eliminate any form of practice that promoted comfort or pleasure, he rather wanted to distinguish between impulses that can be well justified, for example with reference to human nature, and such impulses that cannot be well justified, for example because they go against nature. For the former, Zeno uses the expression hormai logikai, i. e., rational strivings that express an endorsement on the basis of reasons. It is always the endorsement that constitutes rationality. And this endorsement is the result of an evaluation of reasons and not the optimization of given (reason-resistant) desires. So, at least early Stoicism was not about eradicating all passions, but rather about developing a practice based on evaluative approval. And the criteria of this appraisal are not subjective; their ultimate determinants are neither a pleasant life, nor, in contemporary utilitarian terminology, the optimization of subjective states. That which is rational respects what is natural, but it is not subsumed by it. The stoic apatheia refers to the ability to evaluate reasons sine ira et studio – this seems to me to be the most plausible interpretation. Interestingly, the existence of human freedom has been doubted time and again since antiquity. More precisely, what has been doubted is the thesis that we can choose between alternatives, that alternatives that constitute our freedom of action truly exist. It has been less disputed that we are able to evaluate which beliefs are well justified. This means that, ultimately, theoretical freedom, the freedom to form a judgment, is far less controversial than practical freedom, the freedom of letting oneself be guided by reasons. One possible explanation for the asymmetry could be the materialist conception of causality, which was already prevalent in antiquity. This conception purports that it is the transmission of physical effects or even substances that constitutes causality. This – materialist – causality is hardly reconcilable with the causal role of deliberation, i. e., the weighing of reasons. It seems, then, that the effect of theoretical reasons is compatible with a materialist conception of causality as long as reasons have no effect in or for material events. However, the asymmetry between theoretical freedom, which is considered unproblematic if it is discussed at all, and practical freedom, which is considered problematic because it is incompatible with the materialist principle of causality, rests on a prerequisite that is highly counter-intuitive, namely that beliefs play no role for actions. My conception of practical

 Cf. Scheler, 1978.

Metaphysical Aspects of Agency

45

reason assumes exactly the opposite: It is beliefs that determine our actions – the empirical and normative beliefs of the acting person. Following Stoicism, we even argued that it is normative and empirical beliefs that ultimately guide our action. As the final instance it is this evaluative approval (i. e., the belief that it is reasonable to do X which is the result of deliberation), that determines our action. Without this approval, without the action-guiding belief, actions do not occur, for actions are always the expression of such a normative belief. Behavior that does not represent a normative belief, an endorsement of the respective person, does not have the character of action. However, the categorical distinction between mere behavior and action must not obscure the fact that deliberation, i. e., the weighing of reasons, is decisive for the determination of human practice to a varying degree. In contrast to Stoic radicalism, our theory of structural rationality seeks to grasp this aspect more adequately. Max Frisch has impressively portrayed in his text “Man in the Holocene”⁴ how, even after an extensive loss of comprehensive, structural rationality, a pointwise rational evaluation is still possible. The readers notice only throughout the book, if not only at the very end, once the perspective of the old man is left behind and the strange actions before his death seem in retrospect to make no sense anymore, that the various considerations and activities were the expression of the later stages of dementia. However, even fully accountable grownup individuals commit several structural irrationalities their everyday life: they carefully tuck away their cell phone and their car keys in a jacket that they are not going to use in hot summer weather, preoccupied they carry individual dishes into the kitchen instead of stacking the plates in order to save time, and so forth. Only optimizing certain goals under given current epistemic states at a given point in time is not a guarantee for coherent courses of action – on the contrary, pointwise optimization activities tend to fall apart and do not form a coherent picture of structurally rational practice as a whole. What is missing is the reason-guided approval (what Stoicism calls synkatathesis) that guarantees the structural coherence of an individual’s actions as a whole. The explicit or implicit concession of theoretical freedom, here the phenomenon of forming beliefs on the basis of an evaluation of reasons, of the pros and cons, via an adequate conception of action (behavior that is based on approving, normative beliefs) implies practical freedom. Responsibility and freedom are merely two aspects of the same phenomenon, namely the affection by reasons. As the authors of our life we are free (i. e. the reasons that we adopt are relevant for what we do)

 Cf. Frisch, 2012.

46

Julian Nida-Rümelin

and responsible (i. e. we are able upon request to give reasons for why we did something). I am responsible for everything that is under my control. This conception of responsibility that originated in Stoicism is indispensable. We have characterized the phenomenon of having-something-under-one’s-control as the possibility of it being affected by one’s own reasons. If I adopt the arguments of a person that speak in favor of a certain action, then these arguments become my own reasons. This does not mean that this has any power over me, but rather that I have my reasons for deciding one way and not the other. This undermines the interpretation that has shaped poststructuralist discourses, and which claims that any form of influence is a form of an exertion of power. Pointing out a fact is not a form of an exertion of power, even if the person was not familiar with the fact beforehand. Those who make reality disappear in discourse practices by characterizing knowledge as having no reference to reality, forfeit the possibility of distinguishing. Exertion of power is always about asserting vested interests against other interests. Pointing out states of affairs might sometimes serve one’s own interests, sometimes it might not; in any event, strengthening a person’s power of judgment is eo ipso not a form of an exertion of power, quite the contrary. This is the central postulate of the Enlightenment, i. e., the core of the Enlightenment project: empowering humans to form their own judgment and to act on their own reasons bestows upon them authorship and ego-strength and enables them to come to their own judgements and be responsible agents instead of being driven by forces, dependent and ignorant. Enlightenment is not a form of exertion of power. The project of Enlightenment is never completed. It empowers humans instead of making them powerless. However, enlightenment presupposes that there is something which one can be enlightened about, that is to say: without justified truth claims, the project of the Enlightenment becomes vacuous. If it loses its subject matter, the only thing that remains is the caricature of all discourse as a form of influencing others. The paradigm shifts from knowledge to power. We say that small children cannot take responsibility for their actions and what we mean by that is precisely the fact that their practice as a whole is not yet coherent, because it is not sufficiently affected by the evaluation of reasons. Small children are exposed to a multitude of impulses for actions that they can only coordinate insufficiently. One could say that they react to them in a pointwise manner, that they develop more or less rational strategies, which are devised as short-term measures and which often enough harm their own interests. It is the behavior of the akrates. The akrates is acting with only one specific goal in mind which was evoked by a sensory impression, and, without thinking too much about it, he utilizes all means available in order to reach this goal. The

Metaphysical Aspects of Agency

47

question whether obtaining this goal ends up facilitating or impeding him in reaching other goals that he is currently not aware of does not even enter his mind. It is this specific kind of irrationality that we observe with unrestrained and inconsiderate people who follow their respective momentary impulses which is overcome only by the practice of having one’s own reasons for a lifeform that is consistent overall. This ought is not external to this practice, but rather inherent to the evaluation of reasons, even for those whose practice as a whole is not coherent yet. The reasons themselves are already aimed at ensuring coherence. We do not induce the philosophical idea of a coherent lifeform into the normative discourses of the everyday world, rather, it is constituted by our life-world practice of giving and taking of reasons. This conception of responsibility differs markedly from the two philosophical positions most prevalent today: Humeanism, most dominant in the AngloSaxon debate, considers rationality to be an optimization of given desires. If we take this optimization calculus literally, we must ask whether there can be responsible decision making within Humeanism at all. Indeed, the contemporary decision-theoretical transformation of Humeanism reduces the individual agent to an optimization machine that optimizes any desires, provided that they are sufficiently coherent and can therefore be summarized as a real-valued utility function. Decisions presuppose different options that one can choose from. In the optimization calculus, this is only the case if two options have the same expected utility. In this special case, however, the decision is arbitrary, i. e., it is irrelevant which option is chosen. The only important thing in such indifferencesituations is that some decision is made at all. My position also differs substantially from Kantianism. After all, for Kant and in contemporary Kantianism there are alternatives to choose from (real alternatives that make a difference). This is because the Categorical Imperative only determines the domain of impermissible maxims and thus allows, if interpreted plausibly, some leeway in the decision-making of the respective moral agent. The test of the Categorical Imperative, the test whether certain maxims are suitable candidates for a universal rule of action, excludes certain maxims and allows others and it is up to the individual which maxims to adopt. This means, there is room for individuality, for particularities, for what characterizes the individual person. A decision that is made out of respect for the moral law usually still requires the agent to choose between genuine alternatives. In other words: In Kantianism, the individual is still a genuine agent, whereas in Humeanism, the individual vanishes in the optimization calculus. However, even Kant and contemporary Kantianism are, as it were, somewhat infected by Humeanism. In Kant this takes the form of the so-called pragmatic imperatives which take one’s own well-being as a guideline. Pragmatic imperatives are empirical criteria

48

Julian Nida-Rümelin

to the extent that they determine the conditions that lead to an improvement of my wellbeing, my “happiness”. Kant retraces the increasingly hedonistic subjectivism that had characterized the transition from Greek Classics to Roman Imperialism: He reduces eudaimonia to subjective wellbeing. Indeed, divine retributive justice is supposed to guarantee that moral obligations do not entail disadvantages for the wellbeing of the individual in the long run. The reverent tone of morality, the demand to act out of respect for the moral law, is ultimately tempered by the promise of divine retributive justice, the pragmatic and categorical imperatives converge. It seems, Kant was unable to withstand the smallminded entrepreneurial spirit of the dawning Bourgeois Epoch after all. My own position is distinct from the Kantian position in two ways: Firstly, it is not only about moral reasons, but about reasons in general, and, secondly, the agent is characterized by making certain reasons their own, by letting themselves be affected by reasons and thereby taking responsibility for their practice as a whole. Reasons cannot be divided into two categories, objective (moral) and subjective (pragmatic), but rather, they are all objective, albeit dependent on preceding decisions of the respective agent. The respective practice is an expression of the result of this deliberation. We not only depart from classical and contemporary analytical Humeanism, but also from the semi-Humeanism in Kant. Historically speaking, one could say that we return to Greek Stoicism, whose central thesis is that a reasonable, acting agent is characterized by (evaluative) approval. It is not the hormai, i. e., the respective impulses or desires that constitute rationality but the agent’s evaluative judgment. This is the practice of the weighing of reasons: to develop an evaluative judgment that ultimately determines the concrete practice. The objectivity of reasons does not jeopardize the agent’s individuality because it is the agent who determines through her evaluative judgment which of these reasons determine her action. This comes close to the position of the founder of Stoicism, Zeno – as far as it can be re-traced: He is not concerned with killing off feelings of pleasure in general, but rather with the agent’s control. It is the evaluative endorsement that leads a reasonable agent to act – and not the mere satisfaction of momentary impulses and desires. Desires that correspond to the natural constitution of the human body deserve this endorsement and are therefore not irrational. Stoicism is not about cauterizing these desires, but about the agent being their own master at every given moment. The ethical implications are obvious: The radical restriction of responsibility to that which is under our control, i.e, our attitudes which the Stoics called virtues (aretai). Stoic virtues are precisely not what is considered Aristotelian virtues, viz. behavioral dispositions that are formed through habit and nurture, rather, Stoic virtues are what is at our disposition, i. e., that which we can freely decide over. In my transformation, authorship is determined by normative atti-

Metaphysical Aspects of Agency

49

tudes, the respective endorsement of specific reasons that we believe to have acknowledged and the insight which results from acknowledging these reasons which provides the foundation of our practice. Our practice exhibits our evaluative endorsements. The respective momentary impulses that “seize” us are, for us as reasonable agents, external conditions of our practice, to which we respond as we would to current weather conditions; we decide based on how they are given. They are not the movens of our behavior as reasonable agents. If they were, this means that we are weak-willed, potentially even behaving like non-agents. This slipping into an other-directed order would threaten our status as an agent.

4 Authorship We can summarize the results of the preceding paragraphs as follows: Practical reason, freedom, and responsibility are all different aspects of authorship. Authorship in turn is founded on affection by reasons. The form of this affection by reasons, however, requires clarification. In the Humean tradition the solution seems simple and unambiguous: the evaluation of reasons, the role of rationality, is limited to choosing the means for goals which are ultimately dictated by the acting individual’s desires. According to Hedonism, the final indubitable goal is the optimization of the subjective well-being of the agent. However, if (basic) desires are considered a given, in the sense that they are not themselves the result of the evaluation of reasons, then authorship within naturalistic Humeanism is limited to the choice of the respective optimization strategy. Human action as a whole is no longer an expression of practical reason, but rather the result of an optimization calculus whose utility function is prescribed by basic desires and whose result, except for the case of indifference, is predetermined before any deliberation sets in. If the agent deviates from this predetermined outcome, he is deemed to be irrational, albeit only in the weak sense of having failed as a ratiocinator: He miscalculated. At variance with the classical David Hume, the contemporary radically naturalist version of Humeanism dissolves this already reduced form of theoretical rationality in a causal theory of knowledge.⁵ If the respective empirical judg-

 The causal theory of knowledge is a response to the problem raised by Edward Gettier, according to which truth and justification do not suffice in order to speak of “knowledge.” The Gettier cases show that even if a belief is justified and true at the same time, we do not yet speak of knowledge, viz. when the justification is not connected in an adequate way to the – supposedly – known state of affairs. The causal theory of knowledge then requires an additional, third criterion of knowledge: The respective state of affairs cause empirical judgments. A weaker formu-

50

Julian Nida-Rümelin

ments are merely the result of a causal process and the deliberation of theoretical reasons is reconstructed as a causal process, what remains of the human agent is not even this atrophied form of the empirical power of judgment. Strictly speaking, this framework leaves no room for good reasons, there are only causal processes that can, in principle, be described by science.⁶ The problem we thus face is to come up with a better description of the active role that constitutes authorship. Apparently, the Greek Stoics faced a similar problem. They wanted to capture something philosophically that is in fact obvious, but Greek vernacular lacked the necessary words for expressing it. They reverted to creative word inventions and did not scare away from more and less dramatic reinterpretations of common expressions. The German or English translations of these Stoic neologisms are themselves interpretations that require justification, which means that they are a second-level interpretation of first-level philosophical terminology that itself requires interpretation since these firstlevel concepts are not covered by everyday language. What is more, the majority of the older Anglophone literature leaned on German scholarship and knowledge in order to capture the content of the textual fragments, which in some cases amounts to yet another, at times more and at times less creative, conceptual interference that does not always make it easier to access the insights from Greek Classics. Now, this is not the place to engage with the intricacies of this discussion. Instead, in what follows, I will develop a terminological framework that will allow us to characterize authorship further, and which I believe converges for the most part with the insights of Greek Stoicism. However, whether or not it actually does is ultimately irrelevant for this systematic clarification. Stoic philosophy revolves around the question of what one can ascribe to the agent as their own, i. e., as that for which they are responsible. Stoic ethics

lation of this condition, which can be integrated more easily into the humanist conceptual framework developed here, would merely require that the reasons which one has in favor of a belief refer in an appropriate way to the state of affairs with respect to which we form a belief about (in order to call this belief knowledge). Cf. Gettier, 1963, 121– 123; and Nida-Rümelin, 2018, ch. 1, § 4.  The so-called anomalous monism developed by Davidson tries to uphold this naturalist Humeanism and to simultaneously rid itself of the burden of justification. This is achieved by interpreting the relation between desires, beliefs, and actions as causal, but not as causal in the sense of the usual way that corresponds to the regularity theory of causality. Davidson believes that it is obvious that there are no lawful regularities that would allow a description of these relations. This is what constitutes the “anomaly” in his monism. With this move, the reduction of practical philosophy and social science to physics is excluded; which in turn gives rise to critics who argue that Davidson should either give up the anomaly or confess to a form of dualism. Cf. Davidson, 1963, 685 – 700.

Metaphysical Aspects of Agency

51

is based on the distinction between ta eph hemin and adiaphora, i. e., that which is up to us, those things that we can control and those that we cannot. In late Roman Stoicism, the scope of adiaphora is extended for educational purposes to cover more and more areas of human actions, until the only thing that remains is that which in German translations is usually referred to as Vorstellungen (impressions). Epictetus demands us to understand that it is not the things themselves that unsettle us, but rather the impressions we have of those things, and that it is up to us to detach ourselves of these impressions. It is not the death of a loved person itself, but rather our impression of this death that makes us unhappy. It is not the feared calamity itself that terrifies us, but rather our impression of this disaster, etc. One way of putting it would be to say that Roman Stoicism retreats into an inner citadel: It develops psychotherapeutic strategies that are intended to enable people to be indifferent towards everything they cannot influence in any way. Even the countless everyday exercises that Epictetus recommends, and that the Stoic emperor Marcus Aurelius undergoes, cannot cover up this fundamental problem of this now subjectivist stoic ethics of Roman times. A person who only controls her impressions and who is indifferent towards everything else may retain inner equanimity⁷, but at the same time she retreats to a position of irresponsible, general indifference. Late roman Stoicism inches closer to Epicureanism. However, I believe that deeper philosophical insight of Stoicism lies in a different distinction, namely the characterization of the human agent as somebody who makes normative assessments and has preferences. In classical Stoicism, and as early as in its founder Zeno, having preferences and making normative assessments is an expression of a free (justified) decision. The main idea is that humans take a normative stand before they act a certain way, and this evaluative judgement entails a dissociation from one’s own inclinations, (in Kantian terms), or from one’s desires (in Humean terms). This dissociation does not lead to the extinction of inclinations or desires, they simply lose their practical relevance. Instead of aligning our practice with our desires, we follow our justified assessments. In some Stoic fragments, the expression krisis is used for this phenomenon (prohairesis krisis estin), in others synkatathesis and in still others hormai logikai. ⁸ I am not claiming that these expressions are interchangeable, but they accentuate the same central phenomenon of human authorship within different categorizations. In the first version, reasonable practice is aligned with (theoret-

 “Aequam memento rebus in arduis servare mentem”, Horatio, Odes II, 16.  Cf. asthenês kai pseudês synkatathesis from: Stoicorum Veterum Fragmenta (SVF). Edited by Hans von Arnim. 1903. Part I 67 and III 380.

52

Julian Nida-Rümelin

ical) judgment or at least it is claimed that the former rests on the latter. The decision (or preference) is the expression of an evaluative judgment. The actionguiding decision is not the result of a causal process that leads from given desires (hormai) to a behavior that realizes them. The second version emphasizes the decisional character of attitudes. Attitudes are not something that we simply have, they are subject to our will, something that includes an act of approval. The third version finally traces authorship back to the ability to distinguish reasonable from unreasonable impulses and desires (hormai). We, in a sense, acknowledge, our impulses, desires, inclinations, and our lust, just as we acknowledge other processes in the world, external or internal, i. e., processes that refer to our own mental states or to the empirical world, and we form an attitude towards them. We distinguish those that withstand reasonable critique from those that do not, and we align our practice with the former. It is then the act of endorsement (synkatathesis) or the ability to distinguish reasonable from unreasonable hormai, it is the (evaluative, normative) judgment that constitutes authorship. This holds in general, not just for moral reasons out of respect of the moral law, as it is the case in the reductionist Kantian view. As agents, as authors of our lives, we inextricably linked to this power of judgment and ability to differentiate. We are always judging whether we want to or not. With every decision we make, we pass a judgment whether or not something is valuable, whether or not it is reasonable to act accordingly. It is this power of judgment and decisionmaking that makes us the authors of our life and that makes up practical reason. We find the use of certain expressions of everyday language with altered meanings or the introduction of new terms in order to articulate what is meant already in Ancient practical philosophy. In today’s European languages, “reason” and “cause” are used more or less interchangeably, but a careful distinction between these terms is necessary in order to reach further clarification; and be it that the clarity consists in reasons being causes after all (as Donald Davidson assumes, for example). It is only the conceptual distinction between reasons and causes that makes such a philosophical theory substantial. With Stoicism, we have characterized authorship with recourse to an endorsement that is guided by reasons. We assume that this endorsement is practically relevant, at least in many cases. In other words: Without this endorsement, we would act differently. The making-one’s-own of normative attitudes guided by reasons (this is merely a further possible formulation of this state of affairs) plays a role for our practice. This endorsement, or (normative) attitude, is guided by deliberation and differs from a merely arbitrary act of the will in that it is the result of a practical deliberation. However, there is no categorical difference here between theoretical and practical deliberation: In both cases, there is this act of endorsement, for example the acceptance of a theory after one has convinced oneself

Metaphysical Aspects of Agency

53

of the good arguments speaking in its favor. At this point, the gradualism defended throughout this book comes into tension with the understanding of human authorship. How do we characterize this act of endorsement, the making-one’s-own of a reason, the acceptance of a theory, the decision to do something, within a gradualist framework? Perhaps a metaphor might help: There is, as it were, always more or less; that is to say, accepting a theory does not mean that all doubts have been cast aside. The subjective probability for the truth of a theory is normally smaller than 1. We should not interpret the acceptance of a theory as subjective certitude by identifying the (epistemic) probability as 1, but rather as a suspension of deliberation. In the case of decisions, this seems to me to be obvious: We only speak of a decision if further deliberation is suspended. As soon as we continue deliberating, we retract the decision, even if we reach the same decision at the end of a renewed deliberation. We can understand accepting a theory just the same: Accepting means to assume for now that the theory is true (even if one has doubts regarding its truth) and to refrain from further evaluation of proand con-arguments, to suspend further deliberation. Accepting a theory means bringing theoretical deliberation to a conclusion. In other words: I will reason in the following as if the theory were true. Or: My further theoretical deliberation does not call this theory into question. What is always implied is: for the time being. As soon as this theory becomes once again the object of theoretical deliberation, it is not accepted anymore, even if my subjective probability for its correctness happens to remain unchanged.⁹ Authorship suspends deliberation; it requires the completion of theoretical and practical deliberation (even if only tentatively).¹⁰

 This compatibility of gradualism and authorship (author in the sense of the author of an action, but also in the sense of the author of a belief) can therefore be reconciled with Bayesian terminology, cf. Hartmann, Bovens, 2014. The epistemic probabilities should vary within the open interval (0,1) and not within the closed interval [0,1], since the maximum and minimum would lead to revision-resistance, which would be incompatible with the fallibilist (epistemic) realism that I have argued for in Unaufgeregter Realismus. Cf. Perea, 2012, esp. ch. 8.  The Criticism of decisionism is therefore only justified in this way: Deliberation should play a sufficiently large role in private decisions, and especially in public, political decisions, to guarantee the coherence and comprehensibility of practice. But the decisionist is correct to emphasize that decisions require that deliberation comes to an end.

54

Julian Nida-Rümelin

5 The Status of Reasons As has become clear in the previous paragraphs, we are more and more leaving familiar territory. We consider “metaphysics” to be the endeavor to clarify the implicit conditions, the presuppositions of our practice of judgment and of our practice in general. The further we depart from our familiar life-world, the more underdetermined this project becomes, i. e., the more possible ways there are of determining “metaphysics” of the human lifeform (and -forms).¹¹ The general underdetermination of theory by data finds its continuation in metaphysics in the form of the underdetermination of metaphysical theories by lifeworld practices of communication and interaction. The data here are not collected empirically or documented statistically, instead they are limited to the lifeworld experience shared by participants in this practice. In contrast to, for example, the ethnologist or psychologist, the philosopher is, as it were, in the same boat with the object of the investigation.¹² Authorship presupposes something that contemporary philosophy and science predominantly deny, namely the causal role of one’s own evaluative judgements. This is true for the theoretical and the practical realm. It is me who forms a belief myself; it is not the circumstances that, in the shape of causal processes, entail this belief as a consequence. It is me who decides in a certain way, not the circumstances that lead to this decision as the result of a causal process. This self-efficacy, as we want to call it, cannot be reasonably doubted. We as agents and communicators (have to) assume our own self-efficacy and the self-efficacy of other agents. To doubt this would amount to calling into question the entire practice of communication and interaction. I give the promise that I will come by tomorrow. The addressee of this promise assumes that I am capa This became apparent in the continuing dissent with Charles Larmore, which could not be resolved despite an intensive exchange of ideas and although our philosophical views are otherwise surprisingly close: realism and anti-reductionism with regard to reasons, non-naturalism, pluralism of values… For Larmore, reasons are causes, and their effect would remain a mystery if deliberations were understood as causal phenomena, cf. his Frankfurt Lectures in the context of the Cluster of Excellence Normative Ordnungen: Vernunft und Subjektivität. Cf. Larmore, 2012.  However, even psychological or ethnological research is hard to imagine without a high degree of shared experience (shared with regard to the experiences of the researcher and the research object). The absolutely external standpoint of a pure observer can only be assumed at the price of a loss of interpretational competency. For example, the ascription of actions, i. e. the interpretation of behavior as an action, presupposes an interpretation that refers to mental and special intentional states. The purely behavioristic kind of description would deprive the cultural and social sciences, history, psychology, and even political science of its methodological basis. Cf. Winch, 1958, and Winch, 1987.

Metaphysical Aspects of Agency

55

ble of bringing about my coming tomorrow. They assume the self-efficacy of the person that has entered into this obligation. I bring forward an argument, in the expectation that my conversation partner revises their previous belief. I have this expectation because I assume that my conversation partner has something that one might call theoretical freedom, i. e., the ability to align their judgments with the better reasons she has recognized. If I do not have this expectation, I am implicitly denying the authorship of my interlocutor. In science fiction movies, such a denial might be linked to the fact that I consider the opposite person a humanoid robot whose reactions are governed by algorithms and their authorship only a simulation and not an actual practice. This does not mean that I give up the ‘subjective’ perspective in favor of an ‘objective’ perspective (in Strawson’s sense), it means that I depersonalize my counterpart. And in doing so I question the mutual recognition as agents, as authors of our life, as interaction partners, as an interlocutor, as fellow human beings. What is pivotal here is not any feelings that might be involved but the mutual recognition as authors. This mutual recognition, however, entails certain normative expectations, mutual trust, and expectations. If this trust and these expectations are disappointed, we react with criticism and by discontinuing the interaction or at least by changing our mode of interaction. The mutual expectations then seem inadequate, the other side appears only partially accountable or not accountable at all, which is certainly compatible with a feeling of sympathy, empathy, or antipathy. In this understanding, the alternative is not taking a subjective as opposed to an objective stance regarding my human counterpart, but rather the full recognition as an author of their own life, or, to use a philosophically highly contested but more common expression, as a person. This ability to pass a certain evaluative judgement (normatively and descriptively, theoretically and practically) is the condition for human authorship (the status of personhood). Our practice of communication and interaction presupposes this ability. We can hardly deny this ability – provided our analysis is correct. We cannot deny it as participants in the human practice of communication and interaction. Its denial would have to be limited to a separate realm that would have to be detached from life-world experience. Until most recently, several philosophers postulated such a realm. I am convinced that it does not exist or that if it exists it is at any rate inaccessible to us. There is no such external standpoint. Science is not an authority beyond all life-world practice that evaluates and interprets said practice. It is an extension of our shared practice.¹³ Therefore the actually interesting question is not whether we can take a differ-

 Cf. Nida-Rümelin, 2009, part I.

56

Julian Nida-Rümelin

ent, or objective, standpoint, but rather how we can integrate the inevitable standpoint from the participant-perspective into the scientific worldview. One aspect of this integration pertains to the status of reasons. The normative judgement that is constitutive for authorship has to be embeddable within a structural context that, in a sense makes, the person recognizable. The structural rationality approach describes some elements of this coherence. The deliberative practice, the evaluation of practical and theoretical reasons and their expression in actions establishes an overall coherent life practice that represents the evaluations and beliefs of the respective person. The individual person is more than an aspect of a more comprehensive causal process whose exact description is up to the natural and social sciences, they are an author who intervenes in and harnesses causal processes. It is precisely this characterization of authorship, however, that is met by scientistically motivated skepticism. How can it be that persons intervene in causal events – that persons are something else than merely an aspect of a more comprehensive, causally determined “natural” process? In recent years, there have only been few philosophers who actually bit this radical bullet. Most of them have been philosophers influenced by Christianity who ascribe to persons the feature of being an unmoved mover, which originally was ascribed only to God.¹⁴ Precisely this conception, however, seems to stand in obvious conflict with the scientific worldview, according to which everything that happens has a cause and can be described by laws of nature. Interestingly, of all the sciences it is the most advanced fundamental science, physics, that was forced to give up this principle of all-encompassing causality. What, in the meantime, has been accepted by quantum physics as a matter of course breaks with the so-called universal principle of causality, which is more of a philosophical postulate than scientific heuristics. The notion of causality does not play any role in physics, but it is an important research object of the philosophy of science¹⁵ and popular presentations of results in the natural and social sciences. The fact that of all sciences the most advanced one has considered the universal principle of causality obsolete for almost a century now cannot be rebutted by pointing out that this is only the case for microphysics, whose underdetermination does not apply to the macro-physical world. In fact, this assessment has been proven false, at the latest since the empirical confirmation of Bell’s theorem.¹⁶  Among them the analytic philosopher Roderick Chisholm, 1982.  Cf. The analysis of John Mackie, which has become a classic: Mackie, 1974; or the studies by Spohn, 1988, 105 – 134.  Cf. Bell, 1987, ch. 7.

Metaphysical Aspects of Agency

57

Even classical Newtonian physics, in contrast to its popular version of that circulates among the interested public and in philosophy, is clearly not deterministic. Classical physics presents a potentially all-encompassing model of natural processes, interpreted as point masses whose movement is determined by force fields. Taken together with Maxwell’s electrodynamic equations, the result is a closed physical worldview, which does not however, against the widespread myth, allow the deduction of (deterministic) laws of behavior – not least for the simple reason that this model world contains singularities, i. e., constellations for which it is the case that the following constellation cannot be deduced from the description of the current state together with the laws of classical physics. This holds even if one drops the idealization of point masses, i. e., masses that have no spatial expansion. Pointing out evidence for the indeterministic character of physics is often dismissed by pointing out that indeterminism of course does not guarantee human authorship or freedom of action or autonomy. This is of course true. Actually, however, it works the other way around: If there are good, independent reasons for human authorship, if in a universe without authorship there would be no reason, no freedom, and no responsibility, then we would not need physics as a scientific discipline in order to provide a foundation for human authorship, reason, freedom, and responsibility. Instead, it would only have to be shown that human authorship is compatible with the physical descriptions of the natural world. To put it differently: It would have to be shown that human authorship, or more specifically the causal efficacy of reasons (which are not themselves causes), do not come into irreconcilable conflict with the physical description of natural facts, processes, and events. Such conflict would arise if the popular deterministic interpretation of physics were correct.¹⁷ If moral reasons ultimately only expressed our own desires, then we would not need to worry about the practical efficacy of moral reasons. Scanlon, too, resorted to this move: He considers contractualism to be the continuation of the Categorical Imperative in that an action is morally unacceptable if there is no system (Scanlon offers a less strict interpretation of sets) of principles for which this action is permissible, provided that the rules fulfill the condition that nobody can reasonably reject them.¹⁸ According to Scanlon, the reasons that we invoke in order to justify an action correspond to the rules of such normative systems. But when it comes to the question why we feel motivated to fol-

 Compatibilists deny this, but the reasons they present in favor of a problem-free compatibility of freedom and determinism do not withstand closer scrutiny.  Cf. Scanlon, 1982, 103 – 128.

58

Julian Nida-Rümelin

low moral reasons, Scanlon resorts to some basic desire, namely the desire to be a moral. This move makes moral obligations dependent on the given or nongiven desires of a person, which is obviously inadequate. We reproach a person for acting wrongly even if this person does not have the pronounced desire of acting morally, perhaps especially then. Moral obligations do not vanish just because there are no corresponding desires. Apparently after a long struggle, which was also expressed in the fact that Scanlon published relatively little, he presented a version of contractualism in his 1998 book What We Owe to Each Other, ¹⁹ which is unchanged in its substantive criteria, but is not based on desires anymore, but rather on reasons, which he treats as a fundamental concept. Everybody understands what a reason is. We all agree that reasons tend to motivate us, lead us to change our beliefs, emotive attitudes, or actions. Reasons are normative and refer back to empirical conditions. This does not call into question the supervenience of normative features on empirical features. However, when it comes to the question of the status of reasons, Scanlon seems to shy away from conflict with the naturalistic worldview. He therefore characterizes reasons as empirical states of affairs, which raises the question about the origin of their normativity, and leaves causal relevance unexplained. The common feature of those who either call themselves or are called Kantian Constructivists is precisely that they seek to fulfill the universalistic expectations towards a normative ethics or political philosophy but want to exclude objective moral facts because these are incompatible with the naturalistic worldview. In the case of John Rawls, this is realized in the form of constructing shared rules, which represent the common sense of justice. We invent certain rules that seem obligatory to us because we perceive ourselves as part of a broader social cooperative relation. It is a characteristic of this framework that it fails to legitimize central duties of solidarity towards the most vulnerable, i. e., those whose willingness to cooperate presents no benefit to those who are better off. In order to meet the contractualist framework, however, Scanlon, purports reasons as fundamental, meaning they are not the result of construction, but are presupposed as reasons shared by all. The notion of “a rule that cannot be reasonably rejected” presupposes the availability of normative practical reasons. One could therefore interpret Scanlonian contractualism as a kind of second-order normative ethics that integrates the various normative standpoints into a minimal-ethics. It is the factual acceptance of certain practical reasons that enables the contractualist justification of actions. The fundamental role of

 Cf. Scanlon, 1998.

Metaphysical Aspects of Agency

59

reasons changes nothing about the fact that the ethical justification itself has its foundation in an empirical fact, viz. the shared acceptance or non-acceptance of reasons. According to the realist perspective that I argue for, the making-one’s-own of a reason, the acceptance or rejection of a reason, should be reconstructed as a normative judgment: I believe that something is or is not – in fact – a good reason. The fact of accepting the reason itself is not the ultimate source of legitimation, the existence of a reason is.²⁰ As epistemic optimists, we assume that normative discourses, moral and political communicative practices, facilitate the identification of a factually good reason for an action. However, since we agree with Scanlon that reasons are normative, reasons are not part of the natural scientifically describable world. In this sense, they are not empirical. They can, however, be certainly called empirical in a different sense, namely insofar as moral experience plays an important role in having insights into reasons in the first place. If we perceive a specific action as outrageous, this is an expression of an evaluative judgement. The evaluative judgement does not have to be derivable from a moral codex or an ethical principle, it is not the case that this normative judgement presupposes the acceptance of a codex or of principles. But the rejection of an action or a way of acting as immoral always has the character of a judgment, of a normative belief: I believe that this action is wrong, even if I cannot give a rule that can be applied to this action. In the 1980s, the neo-pragmatist Morton White²¹ proposed the analogy between moral feelings as data for ethical theory building and sensory experience. Quine had objected to this, saying that while observations might not be completely independent from theories and concepts, they have a certain independence which cannot be said of moral feelings, since these results from moral codices. This seems to me to be a mistake in two respects: First, it is simply wrong to think that moral feelings depend on endorsing a moral codex, whatever such a moral codex might be. Even in John Rawls, his reference to comprehensive moral doctrines that the political order should be compatible with in a liberal democracy is difficult to understand. Almost nobody will be able to describe a codex that guides their moral feelings. Above all, however, it seems to me that both the empiricist Quine and the pragmatist White, who are certainly exemplary, exhibit a similar misunderstanding: epistemic systems are systems of beliefs, nor In this respect, there are strictly speaking no subjective reasons, just as there are no subjective facts. “Subjective reasons” means a person’s assumptions about what reasons there are. However, we have also repeatedly made use of this lax, naturalized way of speaking in this passage, even in the formulation “giving reasons and taking reasons”.  Cf. White, 1983; and by the same author: White, 2009.

60

Julian Nida-Rümelin

mative and descriptive. Neither feelings nor sensory stimuli are part of epistemic systems. Only the normative beliefs that are connected to moral feelings in one way or another are part of the epistemic system. By analogy: The beliefs that are close to observation associated with sensory stimuli, beliefs that interpret these sensory experiences, form an important justificatory benchmark for empirical theories. Here, the analogy is obvious: If I have a moral feeling, for instance that a certain practice is revolting, then the normative judgment that this practice is morally unacceptable is an interpretation of this moral feeling. However, it can be the case that the sensory stimuli and beliefs that are close to experience diverge just as the moral feelings in the face of a concrete action and its moral assessment do. This is typically the case if I have reasons to mistrust my senses or my moral feelings. Our sensory perception makes a stick immersed into water at an angle appear bent when viewed from the side. This sensory impression is not corrected even if I have convinced myself that the stick is not actually bent, but my empirical judgment takes into account that the stick is not actually bent at the interface between water and air when submerged under water at various positions. The belief that is distilled from other observations is then reflected in individual cases, i. e., I do not integrate the sensory impression of the stick being bent into my epistemic system as a belief that the stick is actually bent. It may be the case that somebody associates homosexual acts with feelings of disgust because of their traditional upbringing but they also have become convinced that discrimination against homosexuals is an injustice and consequently don’t integrate their revulsion into their epistemic system in the form of a negative moral evaluative judgment. We could say: The deliberation of reasons plays an active, in this case particularly visible role in that it blocks a causal theory of knowledge, both for empirical and normative belief formation. The causal role of deliberation becomes manifest in reasonable beings letting themselves be affected by reasons by virtue of their theoretical, practical, and emotive ability to generate insights. Reasons are entities that neither belong to the physical inventory of the world nor are they epistemically constituted, they are real in the following sense: Reasons can be recognized, or not recognized. The fact that reasons are sometimes not recognized does not mean that they do not exist. This, by the way, is our normal language usage regarding reasons. It is in this sense that reasons are objective and non-epistemic, normative (theoretical, practical, and normative reasons always speak for something) and inferential, i. e., they establish a connection between epistemic conditions and a normative assessment (this is equally trued for theoretical, practical, and emotive reasons).

Metaphysical Aspects of Agency

61

The metaphysical doubts concern the fact that we seem to be forced to enrich ontology by “queer”²² objects. Since the world obviously does not only consist of medium-sized, solid objects that are describable by the means of everyday physics, but also of beings that have beliefs and feelings, that think about the world, about how it is and how it should be. We have to accept that, next to objects that can be described by the means of science, there are such objects for which we use the language of mental states, but also the language of inferences, of reasons. I believe it would be philosophically pointless, even a testament to philosophical hybris, if we recognized that (at least) these three forms of language are indispensable in order to grasp the world while simultaneously doubting that the objects these languages refer to actually existed. They are presupposed by our practice of communication and interaction. A comprehensive skepticism regarding these three kinds of objects could not even be articulated without acknowledging that which one was skeptical about.²³

References Bell, John Stewart. 1987: Speakable and Unspeakable in Quantum Mechanics. Collected Papers on Quantum Philosophy. Cambridge: Cambridge University Press. Chisholm, Roderick. 1982: The First Person. An Essay on Reference and Intentionality. Minneapolis: University of Minnesota Press. Davidson, D. 1963: “Actions, Reasons, and Causes.” The Journal of Philosophy 60.23:685 – 700. Frisch, Max. 2012: Der Mensch erscheint im Holozän: Eine Erzählung. Frankfurt am Main: Suhrkamp. Gettier, Edmund. 1963: “Is Justified True Belief Knowledge?” Analysis 23.6:121 – 123. Hartmann, Stephan, and Luc Bovens. 2014: Bayesian Epistemology. Oxford: Oxford University Press. Kant, Immanuel. 1781: Kritik der reinen Vernunft. Hamburg: Meiner. Larmore, Charles. 2012: Ordnungen: Vernunft und Subjektivität. Berlin: Suhrkamp. Mackie, John. 1974: The Cement of the Universe. A Study on Causation. Oxford: Oxford University Press. Mackie, John. 1990: Ethics. Inventing Right and Wrong. London: Penguin Books. Nida-Rümelin, Julian. 2009: Philosophie und Lebensform. Frankfurt am Main: Suhrkamp. Nida-Rümelin, Julian. 2016: Humanistische Reflexionen. Berlin: Suhrkamp. Nida-Rümelin, Julian. 2018: Unaufgeregter Realismus. Eine philosophische Streitschrift. Paderborn: Mentis. Nida-Rümelin, Julian. 2020: Eine Theorie praktischer Vernunft. Berlin/Boston: De Gruyter. Perea, Andres. 2012: Epistemic Game Theory. Cambridge: Cambridge University Press.

 Cf. Mackie 1990, [First edition 1977], ch. 1.  This text is based on ch. VII of my book Eine Theorie praktischer Vernunft.

62

Julian Nida-Rümelin

Scanlon, Thomas. 1982: “Contractualism and Utilitarianism.” In Utilitarianism and Beyond, edited by A. Sen and B. Williams, 103 – 128. Cambridge: Cambridge University Press. Scanlon, Thomas. 1998: What We Owe to Each Other. Cambridge, Mass: Harvard University Press. Scheler, Max. 1978: Das Ressentiment im Aufbau der Moralen. Frankfurt am Main: Klostermann. Spohn, Wolfgang. 1988: “Ordinal Conditional Functions. A Dynamic Theory of Epistemic States.” In Causation in Decision, Belief Change, and Statistics. Proceedings of the Irvine Conference on Probability and Causation, edited by William L. Harper and Brian Skyrms, 105 – 134. Dordrecht: Springer Netherlands. White, Morton. 1983: What Is and What Ought to be Done. Oxford: Oxford University Press. White, Morton. 2009: A Philosophy of Culture: The Scope of Holistic Pragmatism. Princeton: University Press. Winch, Peter. 1958: The Idea of a Social Science and ist Relation to Philosophy. London: Routledge. Winch, Peter. 1987: Trying to Make Sense. Oxford: Basil Blackwell.

Margit Gaffal

Representation as Ontological Problem Abstract: This chapter deals with Kasimir Twardowski’s detailed explanations on representation. After a brief overview of the origins of representation, its constituting parts and their role in the formation of judgments are examined. The article explores the ontology of objects and makes reference to indirect representations. It points to the relevance that Twardowski ascribed to representations for the formation of ethical and aesthetic judgments. Finally, the author analyses the terms function and construct and explains their significance for the creation of meaning. Keywords: representation, ontology, judgments, categorematic term, syncategorematic term, relative truth, function, construct.

1 Introduction What role does representation play within ontology? If we pursue this question, we find ourselves inevitably in a dispute between two opposing epistemological positions: realism on the one hand and anti-realism on the other. It is a question of whether or not representations belong to the field of accepted objects and whether or not they have a function in the perception process. This discussion forms the framework of the present study, in which I will deal with the position of Kasimir Twardowski, who has analysed the concept of representation. Twardowski produced an extensive oeuvre of philosophical writings in German and in Polish. One part he dedicated to an analysis of philosophical terminology, which led him to illuminate its meaning, clarify inconsistencies and subsequently reconstruct its fundamentals. He conceptualized an explanatory model for the structure and composition of philosophical thinking. In order to find a solution to philosophical problems, Twardowski suggested eliminating ambiguities in thinking. It was his aim to solve philosophical problems by clarifying the philosophical terminology. By performing a meticulous analysis of philosophical language he came to set up the logical structure of mental functions. Since language represents our mental functions, it is a matter of examining the objects of these functions. However, when analysing language one is confronted with the ambiguity of concepts (Begriffe). As each term stands for a special kind of representation (Vorstellung), he considered it necessary to carry out a fundamental investigation of representations. He entitled his habilitahttps://doi.org/10.1515/9783110750041-004

64

Margit Gaffal

tion thesis “Zur Lehre vom Inhalt und Gegenstand der Vorstellungen” and pronounced it a psychological investigation. It is in fact a study at the intersection of philosophy, psychology and logic.¹ His interest in philosophy was aroused during his high school years by a book on logic written by his teacher A. Höfler, whose book was inspiration and starting point for further research.² Twardowski’s attitude towards psychology had changed over time. Originally he had seen psychology as basic philosophical science, but abandoned this position after reading Husserl’s Logische Untersuchungen. ³ Later he turned against psychologism, arguing that logic arose before psychology and could therefore not be part of psychology. Furthermore, he maintained that whereas psychological experiments are only probable, logical theses are certain. Psychology is a theory of thinking, while logic tests forms of thinking for their truthfulness. He came to consider psychology a quasi-historic science, as its empirical data are only accessible via someone’s verbally transmitted introspections. Therefore it is a science that combines empiric and historic data. Representations occupy a central position in his work, since he considered them to be the most basic thought movements on which more complex thought processes are based. He explained the origin and elements of representations and how they are logically connected with one another. Thinking takes place through language, in which concepts play a central role. Concepts are representations of clearly defined content. In a meticulous study of the origin of compound terms he explained how more complex thought movements, such as judging or reasoning work. Using examples, he illustrated how we reach a judgment in the areas of ethics and aesthetics. At the heart of his ontology he developed a coherent and comprehensive theory of objects (Gegenstände). Such a theory should, he thought, include physical objects as well as psychic phenomena and their mutual relationships. His aim was to base this all-encompassing theory on logical principles. We will describe his theory of representations and explain their role in the formation of judgments. In this work we will deal with five central elements of his philosophy, including the questions: What do representations consist of and what are their functions? Which ontology of objects does he propose? How do judgments arise? And what are functions and structures?

 Twardowski, 2017, 39 – 122.  Höfler, 1907.  Twardowski, 2017, 14; for Husserl’s antipsychological arguments see Husserl, 1913 (1978), LU/1, 154– 159 and 184– 189.

Representation as Ontological Problem

65

2 The Origins of “Vorstellung”? An analysis of the notion of representation betrays its long history and its numerous shifts in meaning. Despite these modifications, the term apparently always functioned as a link between the person and the world around him. As mediation between the individual and the objects surrounding him, representations appear as reflections of the objects in consciousness. The philosophical roots go back to Plato’s concept of idea (Greek: eidos), and Plato’s theory of ideas appears relevant for understanding the concept of representation. He considered ideas or forms as separate entities which are objectively there and independent of our perception. Ideas are perfect archetypes, immutable and only accessible to thought. They denote the very essence of objects. As perception provides only fleeting impressions and does not bring any certainty, knowledge can only be achieved by comprehending these ideas. They are verbally expressed as concepts. As such, he distinguished between the essence of truth and its deceptive appearance and evaluated paintings as either appearing to be beautiful conceptually or merely creating a deceptive appearance. While for Plato the essence of things was given in the form of ideas independently of their being perceived, the essence of things for Aristotle was immanent in objects themselves. He suggested a dualism of matter and form both of which concur in objects and give rise to development. For him, the essence of an object lies in its potentiality, which could be unwrapped with reference to four types of causes.⁴ The cognitive process proceeds in a similar way, as the receptive mind (substance) perceives something and the creative mind (form) shapes the resulting impressions. Knowledge can only be achieved by grasping the essence of objects through conceptual thinking. Thomas Aquinas considered every object to consist of form and matter, whereby in addition to the immanent form of the object there is also an external form. The external form is viewed as unlimited in time and independent of perception, since it reflects the predefined ordering principles in the form of ideas that the Creator has of all objects. So he says: What is called idea in Greek is called forma in Latin. Hence, ideas are understood to mean the forms of other things – the forms that exist outside of these things.⁵

 The four causes are: the material cause (substance of object), the formal cause (structure), the efficient cause (source) and the final cause (purpose).  Aquinas, Summa theologiae I, q. 15, art. 1. See: Perler / Haag, 2010, 10.

66

Margit Gaffal

However, humans can recognize these forms and use them as patterns for their own creations. Aquinas expanded the concept of the idea and applied it to human experience by viewing it as an ideal pattern or plan that one follows when creating an object. He rejected the assumption of innate ideas, arguing that knowledge can only be acquired through the path of sensory perception of concrete objects. In thinking, sensory impressions and intellect interact through three processes: through the ability to relate to things mentally (intellectus possibilis), through the act of reference (intellectus agens) and through the intellect transforming sensory data into concepts (conversio ad phantasmata).⁶ The late scholastics understood an idea to be an unchangeable, eternally existing entity in the divine spirit, which serves as a template for all unchangeable, temporally existing objects. The term also found its way into everyday language, in which it was mainly used in two ways: on the one hand as a certain form or pattern and on the other as a thought in mind. The two aspects were used in similar contexts and therefore often confused, so that in English and French everyday ideas became ‘mental representations’.⁷ René Descartes took up the connotations concerning ideas and adapted them to the theory of a ‘mental image’ or ‘representation’ of objects, subsequently expanding this notion to include also abstract objects. He called for critical observation and at the same time concluded that a reliable impression of the world is only possible through immediate, clear and distinct perception. Descartes’ proposals regarding the ability of consciousness to obtain a reliable impression of the outside world mark the starting point for Twardowski’s investigations into the functioning of representations.⁸ The linguistic roots of the German expression ‘Vorstellung’⁹ go back to the second half of the 16th century, when the verb ‘fürstellen’ appeared mainly in religious contexts and poems. Its connotations were of putting something in front of one’s eyes or making it known. It was also used in the sense of creating a mental image. Among the entries involving this term in the dictionary of the Brothers Grimm, two stand out, one in a religious context in the sense of ‘placing something in front of the soul’ and elsewhere in the sense of ‘to make appear before the mind’.¹⁰ In a philosophical context the term was introduced by Christian Wolff as a translation of the Latin word ‘repraesentatio’, whereby he distin-

 Heinzmann, 1994, 49 – 50.  Edwards, 1972, 118 – 119.  Twardowski, 2017, 17– 37.  The word “imagination” has the connotation of figment (Einbildung) and lacks reference to an object. See: Kant, KrV, tr. Anal., § 10.  See DWB, 1965 – 2018, s.v. “fürstellen”.

Representation as Ontological Problem

67

guished between the actual process of representing (perceptio) and its underlying content (idea).¹¹ Building on the content of representations, Twardowski described representations as intuitions (Anschauungen) that go along with concepts. These intuitions can either be evoked by actually perceiving something, by recalling one’s memories or simply by thinking of self-generated mental images. He pointed to the fact that we can focus our thoughts on anything. Therefore, he tried to find an explanation for how the activity of representing works when we direct our thoughts to abstract or non-existent objects. He achieved this primarily by breaking down representations into act, content and object.

3 Act, Content and Object of Representations Twardowski viewed representations as thoughts directed towards objects of all kinds. He proposed a detailed description of the logical principles underlying the relationship between representations and objects. Each representation can be approached from the three perspectives: act, content and object. This classification did not come from him but had already been introduced by Bernhard Bolzano¹² and taken up by Franz Brentano¹³ and Robert von Zimmerman.¹⁴ For instance, in Zimmermann’s textbook on Philosophical Propaedeutic for philosophy lessons in grammar schools, the subdivision into content and form of presentation appears. A similar distinction emerges in A. Höfler’s schoolbook ‘Fundamentals of logic’, in which he subdivided representations into the act of thinking, the object of thought and the content of thought.¹⁵ What can be attributed to Twardowski, however, is that he deepened and systematized this subdivision and explained in detail numerous resulting implications. Each thought is intentionally oriented towards an object of reference, be it tangible or intangible. Regardless of what form it has, it is represented in a particular way in every individual. Twardowski’s view is in line with the Brentano School, according to which every psychic phenomenon refers to an immanent

 Wolff denoted an idea as a relevant subject matter or one whose objective is taken into account. The Latin quote reads as follows: “Repraesentatio rei dicitur idea quatenus rem quandam refert, seu quatenus objective consideratur.” Wolff, Psychologia empirica, § 48. For an overview of the evolution of the concept ‘Vorstellung’, see Knüfer, 1999, 15.  Bolzano, 1837; Bolzano 1973.  Brentano, 1995; Brentano, 2008.  Zimmermann, 1867, 47.  Höfler, 1907, 4.

68

Margit Gaffal

object as it presents itself in the person. The emphasis here is on the fact that the way a person represents an object depends on his or her underlying intention. He explained this relation in the following quote: Through every representation something is presented, whether it exists or not, whether it is presented as independent of us and imposes itself on our perception, or is formed by us in our representation; whatever the case, it is, in so far as we represent it, the object of it, in contrast to ourselves and our representational activity.¹⁶

According to this view, in each representation we can distinguish the mental act from its content, both of which refer to an object. In the act someone has an object in mind or is focused on it, whereas the content is the way in which the object is represented in the person’s mind. As such, an object may be referred to by various contents. For instance, one may describe a particular town as “The place where church councils were held under the Visigoths”, or one may equally refer to it as “The town of three cultures situated by the Tajo River”. Although the two descriptions have two different contents, they both refer to the same object, namely the city of Toledo in Spain (and not the one in Ohio). When we speak of an object, this reference has two aspects: the object that one has in mind and the quasi-image that belongs to the content of the representation and which may adopt different forms. Twardowski describes the difference as follows: The three functions of the name are accordingly: First, the announcement of an act of representation that takes place in the speaker. Second, the awakening of a psychological content, the meaning of the name, in what is being addressed. Third, the naming of an object that is represented by the representation evoked by the name.¹⁷

We will illustrate this in a diagram in which act, content and object find their expressions in representation, language and judgment. This threefold division resembles a kind of triangulation through which several interrelated data and proc-

 In German the quote reads as follows: “Durch jede Vorstellung wird etwas vorgestellt, mag es existieren oder nicht, mag es sich als unabhängig von uns darstellen und sich unserer Wahrnehmung aufdrängen, oder von uns selbst in der Phantasie gebildet sein; was immer es auch sei, es ist, insofern wir es vorstellen, im Gegensatz zu uns und unserer vorstellenden Tätigkeit der Gegenstand derselben.” Twardowski, 2017, 64.  In the original text, the quote is as follows: Die drei Funktionen des Namens sind demnach: Erstens, die Kundgabe eines Vorstellungsaktes, der sich im Redenden abspielt. Zweitens, die Erweckung eines psychischen Inhaltes, der Bedeutung des Namens, im Angesprochenen. Drittens, die Nennung eines Gegenstands, der durch die von dem Namen bedeutete Vorstellung vorgestellt wird. Twardowski, 2017, 46.

Representation as Ontological Problem

69

esses can be illustrated. For the sake of clarity the division is used to record all partial references, although in reality they run simultaneously. The scheme shows how an act based on an object and its content is expressed in language in the form of a name (noun, phrase). This verbal expression is evaluated in an act of judgment, in which the content is either accepted or rejected.

Figure 1: Diagram of representation

An example shall elucidate the process. Someone is thinking of a person named X (act), whose name refers to a particular individual whom the speaker knows (object). As such, X’s name has a meaning whereby X can be identified as the object of reference. This object can then be assessed as existing (judgment). This representation does not seem problematic. But how do we represent non-existent objects? Let us take an example from logic. P thinks of the logical term ¬(A=B), which contains the non-existent assumption of A=B. If we break down the representation to act, content and judgment it reads like this: (1) Act: (2) Content: (3) Judgment:

¬ (A=B) (A=B) ¬ (A=B)

P directs thoughts to term, P identifies an equation, P rejects the equation as non-existent.

Any judgment is based on a representation that takes on the status of an assumption. The content of this assumption is then confirmed or rejected. To avoid contradictions, Twardowski claims that objects can only be expressed in categorematic or meaningful terms, whereas syncategorematic terms, such as pronouns or adverbs, have only a modifying character. So the word “nothing” cannot designate any object, as it marks the limit of the conceivable. He under-

70

Margit Gaffal

lined that when someone thinks of “nothing”, one does not represent this with any object. The word “nothing” neither evokes a representation, nor does it have any content, and beyond that it cannot be evaluated. This is explained in the following quote: If “nothing” were an object of representation, it should also be able to be judged, desired or rejected in an appreciative or dismissive manner. This is by no means the case. One can neither say: “Nothing” exists, nor “Nothing” does not exist, one can neither want nor hate “nothing”. (…) So whoever says he is representing “nothing” is not representing at all; whoever represents, represents something, an object.¹⁸

The difference between the content and object of representation is obvious. An existing object has measurable data that can be related to environmental conditions, whereas its content is not directly accessible. For instance, in the representation of a ‘golden mountain’ the colour gold is mentioned, but the colour cannot be set in relation to another colour nor can the height of the mountain be compared to other mountains. The presented content thus lacks the necessary physical points of reference. The difference between the content and the object of a representation becomes apparent by its description. While determinative adjectives complement or extend the representational content (e. g. a reliable person), modifying adjectives occur in examples such as a ‘false friend’ or a ‘painted landscape’, in which the content of the representation is completely changed. Let us imagine a painter who creates a picture of a landscape. By the activity of painting (act) a representation of this landscape is depicted (content) and its outcome is an image painted on canvas (object). Two paintings of the same landscape will never produce exactly the same result. An object can be represented in different ways. In this context Brentano speaks of the object as the ‘primary object of presentation’ and of the content as the ‘secondary object of the imagining activity’. Each representional act has a double object, a primary and a secondary, the object to which the representation is directed and the content through which the object is represented.¹⁹

 Twardowski wrote: “Wäre “Nichts” en Vorstellungsgegenstand, so müsste er auch anerkennend oder verwerfend beurteilt, begehrt oder verabscheut werden können. Dies ist nun keineswegs der Fall. Man kann weder sagen: “Nichts” existiert, noch “Nichts” existiert nicht, man kann auch “Nichts” weder wollen noch es verabscheuen. (…) Wer also sagt, er stelle “Nichts” vor, stellt überhaupt nicht vor; wer vorstellt, stellt etwas, einen Gegenstand vor.” Twardowski, 2017, 64.  Brentano, 1995; Brentano, 2008.

Representation as Ontological Problem

71

4 Ontology of Objects It was Twardowski’s aim to draw up an all-embracing theory that would include tangible and intangible objects. As such, he denoted objects as everything that is. But which objects form part of his ontology? By objects he understood everything that can be thought of or all categories of the conceivable. However, object is not to be confused with thing, since an object refers to abstract elements and things refer to tangible objects. A thing may be a table or a piece of paper, an object includes abstract entities, such as freedom, experiment, idleness or trigonometry. Among the objects he distinguished possible and impossible objects depending on whether they make sense or not. Impossible objects are considered those that contain a negation, a contradiction or are mere phantasy creations. Among the possible objects we find real and non-real ones, real objects are physically measurable (e. g. sound, tree, movement) whereas non-real ones are of abstract nature (e. g. shortage, absence, etc.). Both types of object may in turn be existing or non-existing. Non-existing real objects are those that are within the realms of possibility but in fact do not exist. The following diagram shows the types of objects:

Figure 2: Twardowski’s ontology of objects

Real objects include singular and general objects, and among these we find simple or combined, material and mental ones. Every object forms a unit that consists of concrete and abstract properties. Every representation is directed to an object, irrespective of whether it exists or not. Behind this comes the consideration that even in the case in which one represents an impossible object, at

72

Margit Gaffal

least a representation is awakened. However, representations of impossible objects are a special case as they refer to objects that contain a negation or a negated object, (e. g. nothing, non-smoker), representations that are logically impossible due to a contradiction (e. g. round triangle, square circle), or which contain phantasy objects (e. g. Unicorn, Don Quixote). Twardowski argued that in these cases we nevertheless direct our thoughts to something.²⁰ In his ontoloy we find a detailed description of how complex composite objects tend to be represented. Objects are characterized by material and formal components, both of which are mutually related in that a substance may adopt different forms depending on its shape, colour or size. When describing an object, special attention must be given to its features, since there is a certain ambiguity attached to the notion of feature. In line with the difference between content and object, one must distinguish the features of the content from those of the object itself. Some features occur in many similar objects, so they do not have to be specially represented. For instance, if we represent a table we tend to include its legs without representing them separately in detail. Conversely, although we perceive similar objects, we still recognize the relationship of difference without representing each singular difference expressly. In other words, our thinking is determined by a certain economy of thinking (Denkökonomie) that enables us to comprehend complex objects by highlighting particular features and, at the same time, thinking about their relations to similar objects without affecting the unity of the complex object.²¹

5 Indirect Representations Based on Bolzano’s explanations²², Twardowski described indirect representations as those that contain objects which are related to other objects (e. g. members of a series, Alexander’s father) or those that form part of another object (e. g. the human eye, the gable of a house) or defective objects (e. g. a country without mountains). He calls such relations auxiliary representations because they cause the listener to represent an object through a relation or an imperfection. Twardowski argued that a person is not part of an eye and even the absence of moun-

 Twardowski summarized objects as follows: Allen (Gegenständen) ist gemeinsam, dass sie Objekt (…) sein können oder sind, dass ihre sprachliche Bezeichnung ihr Name (…) ist, und dass sie, als Gattung betrachtet, das summum genus bilden, welches seinen üblichen sprachlichen Ausdruck im “Etwas” findet. Twardowski, 2017, 68.  Twardowski, 2017, 105.  Berg, 1987, 76 ff.

Representation as Ontological Problem

73

tains cannot be part of a country.²³ The characteristic of a defect is particularly relevant with regard to the formation of negative representations (and judgments made on their basis). The way in which general objects (e. g. human being) can be represented takes place in a way similar to indirect representations, through what he called a singular representation (Einzelvorstellung). While general concepts relate indirectly to objects, a single representation relates directly to an object. However, the auxiliary representation has a subordinate position, since the representation is only indirectly aroused. What is the difference between a general and a single representation? The first contains a set of characteristics that apply to many or an infinite number of objects. The object of a general representation combines characteristics of objects of individual representations that are subordinate to it. The way in which general objects are presented corresponds to that of indirect presentations.²⁴ For instance, if one imagines a triangle, one can only do so by involuntarily thinking of a specific triangle, be it right-angled, acute-angled or obtuse-angled. It is an association whose function is to evoke a general representation.

6 Judgments and Relative Truths If we examine the triangulation of act, content and object from the perspective of judgment, then according to Twardowski, we have to deal with so-called ‘relative truth’. He discusses the assumption of a relative truth by examining all its possible implications. First, he analyses the question of whether a judgment of truth depends on given circumstances. According to his view, such relative truth can never be true. He argued that one must distinguish between the statement (enuntiato) by which a judgment is pronounced and the judgment itself in the sense of a mental activity. It is similar to the difference between a linguistic

 He explained the function in the following quote: “Allgemein gesprochen, liegen die Verhältnisse bei indirekten Vorstellungen folgendermassen: Jeder indirekt vorgestellte Inhalt kommt durch Vermittlung von Hilfsvorstellungen zu Stande. Diese sind (…) die Vorstellung des bekannten Gliedes der Relation und zum Teil auch die Vorstellung dieser Relation.” Twardowski, 2017, 113.  In the following quote, Twardowski explains how general representations arise: “Was durch die Allgemeinvorstellung vorgestellt wird, ist ein ihr spezifisch eigentümlicher Gegenstand. Die Gegenstände der dieser Allgemeinvorstellung untergeordneten Vorstellungen werden nicht durch die Allgemeinvorstellung, sondern durch die mit ihr als ihre Hilfsvorstellung auftretenden Einzelvorstellungen vorgestellt, deren Zahl keine bestimmte ist (…).” Twardowski, 2017, 121.

74

Margit Gaffal

sign and the term used to denote it. After all, language serves mutual understanding and we use our language economically with the result that word-combinations are necessarily ambiguous. For instance, the meaning of the adverbs ‘here’ and ‘now’ depends on the situation in which they are used. Thus, everyday communication follows the principle of simplicity and brevity by formulating a question as simple as possible in order to receive an immediate answer. However, the answer implies a number of details, background information and circumstances that we presuppose without reflecting extensively about them. In line with this, representatives of relative truths assume that a pronounced judgment may be true in some circumstances and false in others. Judgments can have a similar content but are expressed in distinct ways depending on the situation; therefore, strictly speaking, they are different judgments. As such, general ethical principles are not incorrect simply because there are counter-examples to issues to which this principle cannot be applied. As long as a principle remains relatively inaccurate, it can be true in one context and wrong in another. To sum up, Twardowski rejected the notion of a relative truth and underlined that a judgment that is clearly and explicitly expressed is always and everywhere true.²⁵

7 Conceptual Representations Concepts are defined as psychic phenomena because they enable us to conceive of objects and use them to form judgments. In line with this view, a word is a symbolic representation of the concept to which it refers. When we think, for instance, of the combination represented by a ‘red blackboard’, we use the memorized image of blackboard and combine it with the representation of the color red; he calls this combination a mental complexion. This complexion may involve the representation of a phantasy character (e. g. blue Pegasus). What happens in the act of representation is that a part of the represented content is changed. He called this partial change a substrate presentation. If a new conceptual representation emerges from modifications made by adding or omitting parts of it, then Twardowski speaks of a mental complexion, such as in the case of one representing an elliptical piece of lawn in the shape of a pentagon.

 He put it like this: Soweit es sich aber um die Urteile selbst handelt, kann man von relativer und absoluter Wahrheit nicht reden; denn jedes Urteil ist entweder wahr, und dann ist es immer und überall wahr, oder es ist nicht wahr, und dann ist es niemals und nirgends wahr. Twardowski, 2017, 143.

Representation as Ontological Problem

75

For Twardowski, to understand a statement means nothing other than being able to assess or judge it. Judgments arise from an analysis of the relation among objects (features, properties, relationships) and of the combination of analytic concepts with substrate representations. What we receive are synthetic terms. Adding or omitting such features to substrate representations enables us to form judgments. Negative synthetic terms are those whose substrate representation is deficient, as in the case of a mathematical point, in the sense of an abstract substrate conception without extension. Concepts such as, for instance, the categorical imperative, law, state, value, etc., are considered complexes of judgments composed of a series of processes of abstraction performed in language. In this sense, knowledge is understood as the ability to reproduce the concepts of judgments. It is about being able to represent judgments about an object and to express them accordingly. In knowledge the substrate representation enters an intimate relation with the represented judgment. Twardowski calls a conceptual definition the sum of judgments made. He finally concluded that his theory and proposals were designed to include all possible objects and that his considerations showed that all terms have their origin in mental phenomena. Later G. Frege challenged this view by introducing a clear distinction between a person’s subjective course of thought and its objective content.²⁶ Frege explained this in the following quote: …it is to sharply separate the psychological from the logical, the subjective from the objective. The meaning of the words must be asked in the context of sentences, not in their isolation; the difference between concept and object must be kept in mind.²⁷

Twardowski’s explanations are combined to an interesting synthesis in the last chapter of his treatise on functions and constructs (Gebilde). Here he connects his ontology with the time component of human activity. According to this view, functions stand for actions, from which constructs emerge as their results.

 Frege wrote: “Man nehme nicht die Beschreibung, wie eine Vorstellung entsteht, für eine Definition und nicht die Angabe der seelischen und leiblichen Bedingungen dafür, daß uns ein Satz zum Bewußtsein kommt, für einen Beweis und verwechsle das Gedachtwerden eines Satzes nicht mit seiner Wahrheit!” Frege, 1892; Frege, 1884, Einleitung, XXII.  The German quote reads as follows: “…es ist das Psychologische von dem Logischen, das Subjektive von dem Objektiven scharf zu trennen; nach der Bedeutung der Wörter muß im Satzzusammenhang, nicht in ihrer Vereinzelung gefragt werden; der Unterschied zwischen Begriff und Gegenstand ist im Auge zu behalten.” Frege, 1884, Einleitung, XXII.

76

Margit Gaffal

8 Functions and Constructs When Twardowski introduced the notions of functions and constructs he focused on the intersection of psychology, grammar and logic. He distinguished between functions or forms of states (Zustände) on the one hand and what he called functional entities or results on the other. Forms of states are in fact activities that are intentionally linked to certain objects. This relation between a state and an object gives rise to a functional entity. According to this view, a painting is considered a functional entity of the artist’s activity of creating a painting. It is a result of the painter’s being intentionally directed towards a sujet. Functions and their resulting entities can be physical or psychical; the latter would be a thought as a result of someone’s reflecting on something. Likewise, functional entities may either be of permanent or temporary nature.²⁸ Twardowski examined the logical difference between word pairs that express functions and constructs, for instance: to talk and a speech, to run and a race. While the verb expresses a dynamic function, the noun stands for an event, whereby two different aspects are part of the same thought which he called ‘figura etymologica’.²⁹ At one time the emphasis is on the functional dynamic side, at another on the phenomenal static side. He distinguished two types of functions and constructs, mental functions and structures (e. g. thinking, judgments) and psychophysical functions and constructs which involve both elements (e. g. write, talk, lie). Words of Latin origin that end in the German suffix “-ung” (e. g. sensation, intuition, conviction) are at the same time functions and entities. The function is the act or completion and its construct is the content. There are aspects of things expressed through functions and constructs, yet which implicitly indicate an important fact when looked at them more closely in the case of the duration of a complex. If one talks about ‘persistent longing’, the term ‘persistent’ does not literally mean that the entity is present all the time; it is rather understood as recurrent and over time. Twardowski specifically pointed out that some psychophysical representations can express something without any special meaning attached to it. Only when the artist presents his represen-

 Twardowski, 2017, 170 ff.  Dieses Verhältnis also zwischen einem Zeitwort und seinem inneren Objekt lässt sich nunmehr dahin charakterisieren, dass das Zeitwort eben das funktionelle (dynamische) Moment, das das innere Objekt bezeichnende Hauptwort mehr das phänomenale (statische) Moment betont. Aber das duch das Haupwort bezeichnete Phänomen (…) ist etwas, das eben dank jener Funktion (…) zustande kommt. Twardowski, 2017, 168 – 169.

Representation as Ontological Problem

77

tation in such a way that the observer can understand it does it acquire meaning. If this does not happen, one can regard the image as an expression of a construct that does not mean anything. Meaning is the ability of a psychophysical entity to convey an impression in the other and to evoke the corresponding mental construct. Meaning is a central criterion in communication, which involves an exchange of the content of messages.

9 Conclusions The achievements of Kasimir Twardowski can be summarized as follows: He aimed to refine the philosophical terminology to increase accuracy in thinking. He presented a detailed description of the dynamics of representations as mental activities. The distinctions concerning the objects that we have examined in this study are ontologically binding. If we characterize ontology as that which Quine summarized in the question “What is there?”, then we have to analyze whether the non-existent objects named by Twardowski should be recognized as objects. Specifically, the question is to what extent non-existent objects have the same status as existing ones. We compared Twardowski’s distinction with the triangulation method. Applied to the concept of representation, however, this leads to a duplication of objects. Seen in this way, one would have to examine the objects of his ontology, such as the aforementioned fantasy structures, before the tribunal of Ockham’s razor in order to see whether it is at all justified to include nonsensical fantasies. The same is true of objects that are actually linguistic constructs with no content. These questions remain unanswered by Twardowski. If we examine the ontological subdivisions, it turns out that not all ontological categories are cleary defined. In fact, Twardowski objected that it is impossible to determine what is exactly understood by what he called real objects.³⁰ Many of his suggestions would have to be checked for the meaningfulness of their content. A related point of criticism is Twardowski’s overemphasis on the role of psychological processes compared to philosophical and logically stringent analyses. Twardowski wrote his work during a time when the position of psychologism was controversial. As can be seen from his work, he leaned towards psychologism in that he explained the dynamics of psychological processes as consistently logical operations.

 Twardowski, 2017, 65.

78

Margit Gaffal

References Bolzano, Bernard. 1837: Wissenschaftslehre. Versuch einer ausführlichen und größtentheils neuen Darstellung der Logik. Ed. J. Ch. A. Heinroth. Vols. I – IV. Sulzbach: Seidelsche Buchhandlung. Bolzano, Bernard. 1973: Theory of Science, edited with an introduction by Jan Berg, trans. by Burnham Terrell. Dordrecht: Reidel. Bolzano, Bernhard. 1987: Wissenschaftslehre. §§ 46 – 90. Band II. Berg, Jan. (ed.). Stuttgart: Friedrich Frommann Verlag. Brentano, Franz. 1995: Descriptive Psychology, trans. by Benito Müller. London – New York: Routledge. Brentano, Franz. [1874]2008: Psychologie vom empirischen Standpunkt. Leipzig: Duncker & Humblot. Psychologie vom empirischen Standpunkt. Von der Klassifikation psychischer Phänomene, edited by Thomas Binder and Arkadiusz Chrudzimski. Frankfurt: Ontos. DWB Deutsches Wörterbuch von Jacob Grimm und Wilhelm Grimm. 1965 – 2018, s.v. fürstellen [29/01/2021] Edwards, Paul (ed.). 1972: The Encyclopaedia of Philosophy. New York: Macmillan Publishing Co. Inc. & The Free Press. Frege, Gottlob. 1884: Die Grundlagen der Arithmetik. Eine logisch-mathematische Untersuchung über den Begriff der Zahl. Breslau: Verlage Wilhelm Koebner. Frege, Gottlob. 1892: “Über Sinn und Bedeutung.” Zeitschrift für Philosophie und Philosophische Kritik 100:25 – 50. Heinzmann, Richard. 1994: Thomas von Aquin. Eine Einführung in sein Denken. Mit lateinisch-deutschen Texten. Stuttgart – Berlin: Kohlhammer. Höfler, Alois. 1907: Grundlehren der Logik. Wien: F. Tempsky. Husserl, Edmund. [1913]1968: Logische Untersuchungen. Erster Teil: Prolegomena zur Reinen Logik (1900). Tübingen: Max Niemeyer Verlag. Kant, Immanuel. [1911]1973: Kritik der Reinen Vernunft, Akademieausgabe. Berlin – Boston: Walter de Gruyter. Knüfer, Karl. [1911]1999: Grundzüge des Begriffs “Vorstellung” von Wolff bis Kant. Hildesheim – Zürich: Georg Holms Verlag. Perler, Dominik, and Johannes Haag (eds.). 2010: Ideen. Repräsentationalismus in der frühen Neuzeit. Band 1: Texte. Berlin – New York: Walter De Gruyter. Twardowski, Kasimir. 2017: “Zur Lehre vom Inhalt und Gegenstand der Vorstellungen. Eine psychologische Untersuchung.” In Kasimir Twardowski. Gesammelte deutsche Werke, edited by A. Brozek, J. Jadacki and F. Stadler, 39 – 122. Cham, Springer. Twarowski, Kasimir. 2017: “Kapitel 2. Idee und Perzeption. Eine erkenntnistheoretische Untersuchung aus Descartes.” In Kasimir Twardowski. Gesammelte deutsche Werke, edited by A. Brozek, J. Jadacki and F. Stadler, 17 – 37. Cham: Springer. Zimmermann, Robert. 1867: Philosophische Propädeutik. Wien: Braumüller.

Jesús Padilla Gálvez

Ontological Commitment and State of Affairs Abstract: This chapter deals with the discussion on the anti-Kantian proposal developed by Bolzano when he introduced the “Satz an sich”. Bolzano’s proposal clearly deviated from the idealistic tradition. As such, a gap was opened between representation on the one hand and the notion of “objective” on the other hand, which belongs to the field of assumptions. The aim was to assume an immanent object that is intentionally given. The Brentano School devoted extensive research to this topic. Aristotelian logic gives way to a new structure whose minor premise is a conditional structure that expresses what C. Stumpf coined as “Sachverhalt”, its English translation being “state of affairs.” This may be expressed in two different ways: as a that-clause or as something that “subsists”. This investigation focuses on the content of judgment when expressed as a state of affairs. Frege considered this content as predicative and Husserl regarded state of affairs is carrier of properties. Wittgenstein proposed a distinction between object and carrier. This investigation aims to clarify the question of states of affairs and its ontology. Keywords: ontological commitment, state of affairs, objective, intentionality, that-clause, judgment, carrier of properties

1 Introduction When we analyze ontological problems, our language becomes blurred and opaque. However, the term “ontological” is strict. It covers two areas that complement each other: on the one hand, “ὄν” – genitive, ὄντος – refers to objects; and on the other hand, “λόγος” is linked to language. Therefore, we must consider the place that objects occupy in our discourse. Let me therefore begin with a brief clarification of the title of this contribution. When I refer to ontological commitments, I detail which objects I give priority to and which objects I reject or disregard because they produce adverse effects. The term “commitment” has a family resemblance (Familienähnlichkeit) or is linked to such terms as “obligation”, “duty”, “responsibility”, etc. A commitment is a promise or a declaration of principles when a judgment is issued. It is said that a person is committed to something when he performs his obligations, fulfills what was proposed or what has been entrusted to him. In other words, that person lives, https://doi.org/10.1515/9783110750041-005

80

Jesús Padilla Gálvez

plans and reacts correctly to achieve a certain aim. For there to be a commitment, there must be knowledge. We refuse to do something if we do not know the obligations which that something entails. If one is really committed to a project one tends to achieve objectives above what is expected. In the end, all this family resemblance indicates that through the use of our language games we choose certain terms that somehow determine our commitment to what we consider to exist. When W. v. O. Quine asks: “What is there?”¹, the answer is monosyllabic: “Everything”. This response implies that no special specifications are necessary in the use of language. For example, I have never used the term “Cthulhu”, a name the writer H. Ph. Lovecraft used for one of his weird figures. I only know this name by my daughter, and except for some reference to this bestiary, I don’t think I would know how to distinguish this figure from other imaginary figures of the common bestiary. I recognize that I could not write a meaningful statement that included the term “Cthulhu.” To do this I would have to read Lovecraft’s fictional novels. I do not intend to raise the ontological question from a point of view in which the ontological superiorities, overlaps or inferiorities are highlighted as they have been neatly discussed in the history of philosophy. Another example may clarify the issue more adequately. I am convinced that ontology does not attend to rigidities imposed by prejudices, assumptions, opinions or mere authority.² As such, for instance, I don’t think anyone doubts that there are traffic rules. All drivers have taken an exam and must know these rules. However, there are drivers who do not consider them important and simply skip them. For some, the rules establish a code of conduct; for others, the rules are systematically violated. We are here looking at two opposite ways of understanding the ontology of traffic rules. I want to focus on the theme named in the title because it has been addressed incompletely. This is the preeminent logical role assumed by certain issues and the commitment we assume in addressing them in that peculiar way. In other words: when Quine’s response to the question is “Everything”, referring to a total number on entities, the function of ontological pronouncement remains unspecified. This unclear response has led to a similar reading. However, such indefinite answers are not necessarily a characteristic feature of ontology; rather, they reveal an inability to provide a solution to the ontological problem. The first problem that is obvious is that the indefinite pronoun “Everything” used by Quine does not refer to a specific person, place or thing. This is because

 Quine, 1961, 1 (Original: Quine, 1948, 21– 38).  Brentano, 1974, 16.

Ontological Commitment and State of Affairs

81

in English, this pronoun refers to quantification or a distribution which gives the impression that it refers to something defined, since in affirmative sentences, pronouns containing “every” are used to describe a definite quantity. However, Quine’s response refers to a quantification of undefined and incomplete entities. With this language game he presents an elegant solution to a classic problem of ontology. Let us assume the example of a visit to the zoo, in which we find animals that are classified and put in different sections so that the visitors can observe them. I can accompany my daughter to the zoo, show her a tiger and explain this: (1) The tiger is an animal. (2) The tiger is essentially a feline. (3) The tiger eats. In the first statement I link the concept of “tiger” to the most relevant characteristic of a genus. In the second, I state a property, while in the third I express a predicate. All these paradigms have something in common: the object or thing they refer to is in a zoo or preserved in a museum of natural sciences. Certain biologists will claim that this type of ontology allows us to conduct studies of every animal genus and study its behavior. However, this type of ontology situates the “Everything” outside their habitats and does not allow us to link them to their forms of life. What may be called a ‘denatured’ ontology is nothing more than an inoperative count of things, the same as a museum of objects that lack a special function. When we ask about the real commitment which we assume regarding “Everything” as it is represented in said ontology, we move the discussion to a secondary level. One of the common mistakes in these ontological proposals is to commit the fallacy of the wrong concretion by confusing the concept of “tiger” with the generic ‘an animal in captivity’. The great ontological ruse results from the confusion between the tiger as taken out of its natural environment and one that is locked in a cage at the zoo. The language that we use to describe the tiger in natural environment cannot be the same as the language used to describe the caged tiger. The error arises if abstract concepts are mixed up with concrete concepts. This is an example of what could be called the fallacy of misplaced concretion of ontological objects. The tiger is removed from its habitat and loses its close connection to its natural environment. In ontology we repeatedly find the reification of such things. Reification is part of the normal use of natural language where a reified abstraction is intended as a mode of expression. Thus, a discourse is adapted to the prevailing linguistic theory. But the use of reification in logical reasoning is misleading and is generally considered a fallacy. In the following

82

Jesús Padilla Gálvez

pages we will investigate a fallacy that tends to occur in the description and use of the term “Sachverhalt”. The problem of reification and its consequent fallacy arises when we use hypothetical constructions as if they were real events or physical entities. The error is caused by treating something that is not concrete as something which is. In order to resolve this problem I will compare different proposals on states of affairs and discuss its ramifications and consequences.

2 About Assumptions A. Meinong has made important contributions to the development of an object theory (Gegenstandsheorie). By ‘object’ Meinong understood everything to which a psychic act is directed. A psychic act includes any kind of representation, judgment, feeling, and so on. This definition starts from the assumption that one cannot think without thinking something, whereby this “something” refers to some object. It is impossible to know without knowing something; therefore, the object is necessary for the act of knowing, and vice versa, although both are not in the same relationship. The term “object” includes everything that can be expressed in a descriptive and intentional way and must therefore be considered.³. According to Meinong, every object implies a pure object (reiner Gegenstand), and this is a precondition for its being considered an object at all. An object should not be confused with a thing (Ding) or a matter (Sache), or an existing entity, or even an ideal. This “pure object” is involved in any other existing entity or ideal object.⁴ Meinong distinguished between four types of objects (Gegenstände) according to the paradigms of intentional acts: the object (Objekt), the objective (Objektiv), the dignitative (Dignitativ) and the desiderative (Desiderativ). We will focus on the study of the object (Objekt) or object of representation and the objective (Objektiv) as the object of judgments. Objects are considered correlates of representation, which is why existence is attributed to them. On the contrary, objectives are correlates to assumptions (Annahmen) and judgments (Urteile) are positioned against representations. Objectives are attributed subsistence (bestehen). In this paper I shall focus on this last aspect. First, Meinong distinguished between positive and negative objects of being, which are expressed by statements of the type: “… that A is”, “… that A is not.” In line with this he distinguished further between positive and negative objects of being that are ex-

 Cf.: Brentano, 1874[2008], Book 2.  Meinong, 1910[1977].

Ontological Commitment and State of Affairs

83

pressed by: “… that A is B”, “… that A is not B”. Likewise, he determined the objectives that are constituted by objects and what he calls “higher order objectives” that contain additional objectives of the lower order. According to Meinong, truth, possibility and probability are not attributes of objects but rather objectives. Equality and difference are considered objectives since they do not belong to the real. In fact, in this object theory Meinong distinguished between a variety of different ontological elements: On the one hand, there are those objects in which a relationship of friendship is maintained, that are expressed in a sentence such as, for instance, “John loves Mary”. On the other hand, we find such statements as the “loved ones” which he calls existing ones (bestehen). We also discover non-existent statements such as a “round square”, but since the construct of “non-being-square-round” cannot be verified, he called objects of this kind non-subsistent (nicht-bestehen). Propositional attitudes consist of objective elements so that a speaker can believe that a certain state of affairs (Objektiv) subsists or does not subsist. Obviously, the content of a propositional attitude does not share the same properties as the content of a declarative statement. “Objectives” can be positively characterized as “possible” objects although they are usually described by their purely negative form. This negative description has consequences for the formal discussion. “Friend” can be derived from “friendship” as its constituent. The “objective” of the “false friend” has “non-friendship” as a constituent. In the relationship generated with the constituent, the “objective” becomes an object of a higher order. With this step, Meinong proposed a radical restructuring of traditional ontology. The objective element (Objektiv) forms the basis on which the science of logic is applied. This is because objective judgment is always considered an object of higher order.⁵ Meinong considers it the direct object of judgment and it is characteristic for this kind of judgment that it does not “exist” but rather subsists.⁶ This even applies to cases in which the constituent parts have their own existence. Higher order objects are based on lower order objects and objects of lower order form the fundament upon which those of the higher order are based. Higher order objects are characterized by relationship and complexes. For Meinong, an assumption is situated on a spectrum between representation and judgment. He says this:

 Meinong, 1910[1977], 97– 105.  Meinong, 1910[1977], 97 ff. and 507 ff.

84

Jesús Padilla Gálvez

Das Wort “Annahme” soll im folgenden als technischer Ausdruck für alle Erlebnisse gebraucht werden von denen ich zu zeigen hoffe, daß sie jenem Zwischengebiet zwischen Vorstellen und Urteilen angehören.⁷

Assumptions are normally expressed without the speaker expressing any conviction, so it is not required for a subject to create or know what it expresses. Such an assumption could have the following form: (4) I assume that he will be happy. Although the syntactic structure of this assumption is very similar to that of a judgment, we still notice that the speaker is neither convinced of what he says nor has any evidence of the content expressed. Nothing necessary is affirmed in this statement. In fact, this ambiguity enables the speaker to express the opposite of what he knows, believes or conjectures without creating any contradiction. The first thesis that Meinong postulates is that assumptions are not judgments, but they contain forms of judgment as long as we can affirm or deny what is supposed in the utterance. Therefore, he asserts that the essence of the assumption is that he addresses a being that does not need to exist. The solution given to elucidate example (4) is based on a false analogy from mereology. According to this point of view, the objective is considered a kind of complex, that we could call the whole, and its object a kind of component – that is, a part. From this distinction we speak of “being derived from the objective”. The objective can even be characterized as “not being”, which results in a paradox because “the being of the objective” will erroneously refer to the being of its object. Meinong argues that if all the opposition between being and non-being were in the objective and not in the object, then it would become evident that such objects would neither require a being nor a non-being. This point of view is counterintuitive since we started from objects that were considered as “being” yet which later transformed into “not being.” The argument is convincing since the nature of the object will be contingent in any case, regardless of whether the object is or is not. Thus, an absurd object such as “the square circle” is considered both as not existing and at the same time as an ideal object of a similar kind to “identity” and “difference”, with which it shares its non-existence. Therefore, the object is by nature external to being, although of its two objectives, “being” and “not being”, one always occurs.

 Meinong, 1901[1977], 6.

Ontological Commitment and State of Affairs

85

The second thesis is that assumptions (Annahmen) are not strictly limited to representations. This is because we cannot ascribe any attitude to the speaker who utters them, nor can they be shown to express an assertion, so they do not seem to be a form of judgment that can correspond to the class of representations. Meinong indicates that the assumptions neither corroborate that a representation exists, that is, the affirmation or denial of the existence of an object presented, nor do they corroborate its recognition or rejection. Assumptions do not need to be affirmed or denied since they do not contain the representation of their object and do not imply it either. Therefore, assumptions cannot be considered as representations. The third thesis is derived from the previous ones and states that assumptions are neither judgments nor representations but that they belong to an intermediate point between judgments and representations. This opens the dispute about the role and function of propositional attitudes in psychology.⁸ Meinong states that assumptions can be interpreted as fictions that contain certain psychological features. Thus, the example presented in (4) shows that the speaker makes an assumption based on certain subjective conditions that possess an objective characteristic feature. Meinong studies the relationship between the judgment and the object of judgment cases in which judgment is rejected or negated. He argues that affirmation and denial do not affect the beliefs of the individual who makes it. A negative judgment is established not in the form of a judgment but rather as subsistence. Judgments and subsistence do not only refer to an object of which judgment is made but are what he calls “objectively directed”. Therefore, where a negation is expected there is only a representation. In fact, negative judgments do not correspond to the negative representation of an object as such. The use of dissonant terms makes it difficult to understand the proposal developed by Meinong. Therefore, it seems convenient that we approach this program from the perspective of another member of the Brentano School.

3 Stumpf and the Content of Specific Judgment Carl Stumpf coined the concept of “Sachverhalt”, translated into English with the expression “state of affairs”. He states that judgement is determined by specific content, differing from the content of representation. Said content of judgment is

 Meinong, 1901[1977], 48 ff.

86

Jesús Padilla Gálvez

expressed as a sentence containing a subordinate clause or an infinitive in form of a nominalisation. He explains this in the following quotation: “Daß dem Urteil ein spezifischer Urteilsinhalt entspreche, der vom Vorstellungsinhalte (der Materie) zu scheiden sei und sprachlich in” Daß-Sätzen “oder in substantivierten Infinitiven ausgedrückt wird, hat Brentano bereits vor drei Dezennten in logischen Vorlesungen scharf hervorgehoben. Noch früher hatte schon Bernhard Bolzano von dem “Satz an sich” in demselben Sinne gesprochen. Ich gebrauche für diesen spezifischen Urteilsinhalt den Ausdruck Sachverhalt.”⁹

Stumpf considers that the term “objective”¹⁰ used by Meinong to refer to “Sachverhalt” as synonymous with “construct” or “structure” (Gebilde). If constructs or structures are included in psychic functions, then they contain an objective character. In the same sense, Husserl refers to “objectivities” (Objektivitäten)¹¹ of the various acts of conscience.¹² According to Stumpf, the relationship between phenomena and psychic functions described in the problem of intentionality must be resolved.¹³ To resolve this relationship, a third element is required: in addition to phenomena – Erscheinungen – and functions – Funktionen – the specific content of the judgment that appears at the pole opposite to the contents of representation must be analyzed. Bolzano and Brentano thought that a judgement must be approached as a mental act by means of which its content is synthesized by affirmation or rejected by negation.¹⁴ This relationship can lead to a paradox produced by the proposals developed by Brentano. To solve this tension, Stumpf proposes the following solution: every psychic function is constituted by a complex of elements. The function consists of a correlate that is imposed paradigmatically. From mental functions Stumpf distinguishes psychic constructions (Gebilde) that conform to their specific contents. Thus, in summarizing the crux of each matter; in judging, states of affairs; in conceptual thought, conceptual content; in feeling and desire, passive and active value. According to Stumpf, these constructions should not be considered independent realities – against Plato – nor fictions – against W. Kraus (referring to Brentano). This process is named in different ways: Chris-

 Stumpf, 1907, 27.  Meinong, Über Annahmen, Ztschr. f. Psychol., Ergänzungsband II, 1902.  Husserl, 1901[1921], II, 618.  Stumpf, 1907, 27, footnote 17.  C. Stumpf says: “Die Grundfrage […] betrifft das Verhältnis der Erscheinungen zu den psychischen Funktionen” Stumpf, 1907, 3.  Bolzano 1837, § 24, § 25, § 28; Brentano 1874[2008], 241.

Ontological Commitment and State of Affairs

87

tian v. Ehrenfels called it “configuration quality”¹⁵ and considers it a specific result that appears in consciousness in the form of a summary of the essential or the most significant. This summary was called by Stumpf “epitome”¹⁶. E. Husserl designates it “Einheitsmomente”¹⁷. The epitome is a necessary correlation of the synthetic function which occurs in the synopsis. Therefore, when a subject thinks conceptually, he captures the content of the simplest concepts in a way similar to how a mathematical function is applied, so the concepts themselves are understood as correlates. We have indicated that the concept “Sachverhalt” designates the specific content of a judgment. This content has to be separated from the content of a representation – what is in general terms called “the matter”¹⁸ – which is the judgment itself. Such judgment could be something like this: (5) The rose is red. The state of affairs of the judgment issued in (5) is expressed linguistically in two different ways: either by a subordinate clause or by a noun in the infinitive. When we want to know what the specific content expressed in (5) is, the statement of (5) is examined. We would generally answer using a subordinate clause that begins with the conjunction “that”: (S1) “[I think] that the rose is red” (“… daß die Rose rot ist.”) The propositional attitude expressed by a relative statement in (S1) refers to the content specified in (5). The “that”-clause establishes the nominal character of the sentence. Expression (S1) has the same function as a noun although it does not correspond to a noun phrase. As can be seen, the “state of affairs” is expressed by a subordinate subject which takes the position of a direct object by which a cognitive process is presupposed. The structure (S1) gives the state of affairs a temporal dimension and the presence of its own arguments facilitates the interpretation of (5) in a structured way. The nominal character of (S1) implies that the subordinate clause performs functions typically reserved for nominal groups. (S1) designates a state of affairs that refers to a more abstract entity considered to be of a higher order. Therefore, they are less concrete than physical objects. From the moment they lack space-time location, it is impossible to des   

v. Ehrenfels, 1890, 249 f. Stumpf, 1907, 29. Husserl, 1901[1921], II, 230, 274. Stumpf, 1907, 25.

88

Jesús Padilla Gálvez

ignate them deictically, so that the state of affairs embedded in subordinate structures becomes a purely intensional object. The discussion which Stumpf inaugurates with this new grammatical location of the problem in a subordinate structure puts the focus on the question of whether the state of affairs is linked to propositions on the one hand, or to events and situations on the other. As is known, relative statements establish a conditional relationship. The parallelism between construction (S1) and conditional sentences can be specified as follows: relative sentences that normally function as sentence subjects occupy a position equivalent to the protasis of conditional sentences. One can also establish a temporary relationship similar to that between protasis and apodosis as they behave as variables with generic value. This behavior is characteristic of states of affairs, so it requires a formal study. On the other hand, the state of affairs is denoted by substantiation. If we affirm that the content of the judgment expressed in (5) has to do with the following: (S2) “Be-red-of-the-rose” (“Rot-sein-der-Rose”). The difficulty in determining (S2) is that it refers to a syntactically self-determined and self-complemented category. The use attributed to the state of affairs as a name has to do with its specific function of referential argument. We could say that (S2) in determining the state of affairs acts as a name, and this is due to its specific function in the referential argument. (S2) determine the state of affairs as a name in the philosophical lexicon and as a noun phrase, from a syntactic point of view. The reference function of (S2) is of a higher order, so subsistence (bestehen) is required. In general, it can be said that (S2) suspends the referential character in (5) and therefore does not refer to the existence of the red color in the rose. The union of the words automatically excludes its existential character and necessarily transfers it to its subsistent character (bestehen). By proposing this form of expressing a state of affairs, Stumpf generated a significant division whose effects may still be seen today. If we express a “state of affairs” in the form of a propositional attitude as in (S1), then the subject adopts a certain posture or mental attitude to the meaning of the statement. According to Stumpf, an alternative way of expression would be to put a subject’s attitudes in relation to the meaning of the statements he utters. In general, the propositional attitude expressed in “… that the rose is red” (… daß die Rose rot ist) is constructed by means of verbs such as belief, opinion, etc. that establish opaque contexts, making it difficult to replace such statements by another

Ontological Commitment and State of Affairs

89

with the same truth value.¹⁹ According to Stumpf, if a “construct” or “structure” (Gebilde) is conceptualized, then its general concept and the function to which it refers – in this case the judgment – must be taken into account. The expression does not imply that the content really occurs nor must it reflect upon the individual act. Stumpf describes this in the following statement: “Wenn wir einen Gebilde begrifflich denken, etwa einen Sachverhalt beim Aussprechen einen isolierten Dass-Satzes, so muss notwendig die bezügliche Funktion, hier das Urteilen, ihrem Allgemeinbegriffe nach mitgedacht werden; nur wirklich stattzufinden braucht sie nicht, und den individuellen Akt brauchen wir nicht mitzudenken.”²⁰

Thus, when we utter that Cyclops do not exist, a judgment is issued whose content refers to the absence of the predicate. A speaker could express the following judgment: (6) There are no Cyclops. We can mentally conceptualize a “construct” or “structure” even if the corresponding function has no content. Therefore one can think of a state of affairs without a current reference. This is shown in the subordinate clauses that express a propositional attitude since they do not affirm anything but refer to the content of a possible true or false statement. However, the state of affairs cannot only be given independent of an immediate function and in turn be real.²¹ A state of affairs will be real if it is pronounced as a judgment. Otherwise, any false or absurd state of affairs would be considered true and real. Therefore, functions refer to the facts that are immediately recognized, while patterns or structures are made as long as they are considered content of certain functions. On the other hand, if we express the state of affairs using the expression of “the-Being-red-of-the-rose” (Rot-sein-der-Rose) as specified in (S2), then we ascribe it a noun function. In this case, the referent of the state of affairs is determined by classes of fixed entities and not by states of affairs or relationships. If the state of affairs is understood as a function expressed by nominalization, then the entities must be classified as independent concepts. In summary: this distinction is extremely relevant for further discussion and must be carefully elucidated. Stumpf assigns the status of “construct” or “structure” (Gebilde) to the state of affairs. These must be distinguished from “func-

 Frege, 1892, 25 ff.  Stumpf, 1907, 33.  Stumpf, 1907, 32.

90

Jesús Padilla Gálvez

tions”, that is, from our mental acts. But they must also be distinguished from “appearances” in the form of sensory data as classically conceived. Functions are generally considered facts (Tatsachen). Sensory data are the paradigms according to which objects are presented and which are independent of mental acts. According to Stumpf, a state of affairs is determined as the “immanent content” of a judgment which actually takes place. Therefore, it cannot occur directly and be real in itself, regardless of any function. States of affairs, like “constructs” or “structures”, are factual if they are considered to be contained in certain functions. If we mentally conceptualize a state of affairs by expressing a propositional attitude as we did above in (S1), then the function must be thought along with its abstract concept. It doesn’t need to take place and we don’t need to think about the individual act. From this moment on, the philosophical confrontation was clearly revealed. Before addressing the phenomenological proposals in this context, we shall analyze where of notion of states of affairs is situated within logical argument.

4 The Logical Problem Certainly a model does not replace reality. So when we look, for instance on a map, we should not confuse the marks with the territory. Let me make an inversion of the argument. I am not going to discuss the question of what is there, but will focus on the second aspect, namely the λóγος, in which this “Everything” is inscribed in our formal system. As I have already indicated, the problem of “Sachverhalt” will be studied in its formal context. The problem has been discussed in the Austro-Germanic tradition from different points of view and different proposals have been made. As is known, Frege had questioned whether judgments are the product of a combination of mental representations. To refute this approach, he bases his argument on two paradigmatic examples: on the one hand, when a speaker makes a possible proposition. The speaker refrains from omitting a judgment because he has doubts regarding the existence of a law from which his negation could derive. In this case, he would be forced to argue that the negation of the proposition from a universal point of view (generality) should be false. This last conclusion would be inadmissible. On the other hand, he presents a particular affirmative judgment. This could mean propositions of the following kind: (7) It is possible that the earth sometimes collides with another celestial body,

Ontological Commitment and State of Affairs

91

this would be a paradigm of the first case, and: (8) A cold can lead to death. as an example of the second case.²² For all these reasons, current formal logic avoids such judgments because it truly considers them to express beliefs. Such a relationship formally implies that the result of the universe of discourse consists of a relationship between two logical classes and between a predicate and a quantified individual variable. Therefore, logic considers the relationship S and P to be a syntactic relationship that lacks material content, be it a class relationship or a propositional function of predicates.²³ This error was discovered by Brentano and refuted.²⁴ Therefore, current logic transforms deduction into inference. In the logico-syntactic formula the relation between terms and grammar and their possible significance is lost. The syllogism thus loses its categorical character, which would convey a necessary truth, and acquires a hypothetical form. The syllogism is now interpreted within the framework of classical logic and its logical scheme would be of the following type: If class S represents a class with the property of being mankind and class M represents a class with the property of being mortal, then the Barbara²⁵ syllogism is transformed into a rule that is expressed by modus ponens as follows: Major premise: If Napoleon is a man, then Napoleon is mortal. Minor premise: Napoleon is a man. Conclusion: Therefore Napoleon is mortal. If the reference of existential instantiation is applied to individuals, then Aristotelian judgments are formalized as predicate logic of the kind: If all (or some) individuals are ascribed the property “mankind”, then all individuals possesses the property of being “mortal”. In the minor premise it is necessary to prove that the assumed state of affairs is a true fact. Therefore it would be deduced that the individual could assign the final property.

   

Frege, 1879, § 4:5. Frege, 1966, § 14:25 ff. Brentano, 1974, 20 ff. Barbara: P All animals are long-lived. ∀x(Fx → Gx) P Men are animals. Fb C ∴ Men are long-lived. ∴ Gb

92

Jesús Padilla Gálvez

The position of the state of affairs in argumentative structure has been trivialized, although its position is essential. Suppose we argue about a specific issue and this has certain consequences. So if, for instance, one drives drunk and contravenes traffic rules, the driver will certainly be sanctioned. Supposing one has been to a party and has drunk too much. In a subsequent patrol of the roadways the driver is sanctioned. To express these events in logical order we need to postulate different states of affairs (S (Sachverhalt)), on the one hand the “reckless driving” caused by the alcohol and high speed, and on the other hand the “alcohol consumption” itself. The actual facts (H, Tatsachen) are that he was driving at 140 kilometers per hour and had an alcohol level of 0.90 (zero-point-ninety milligrams per liter). Finally, the legal consequence (C) is that the driver is sanctioned with a fine and his driving license is suspended for three months. This is the logic form: (C) Major premise If S is given then there will be a legal consequence C. Minor premise We can prove that the facts H are identified as S. Conclusion Therefore, facts H imply C. Consequently, the syllogism of determining the logical consequence is briefly expressed as follows: (D) Major premise S → C Minor premise H = S Conclusion For H → C If one drives recklessly at a speed higher than is allowed and combines this with alcohol consumption, this is considered a crime. We should know that there are legal consequences by which the driving license can be withdrawn for a period of one to four years. The implication between S and C establishes a legal norm (rule). According to the minor premise we must prove that the facts have occurred and that they are identical to the state of affairs called “reckless driving.” In the minor premise it must be proved that H is a case of S, so a concrete fact is subordinated to a state of affairs that has been mentioned in the major premise. The minor premise is always formulated as a compound hypothetical formulation. Finally, it is derived from the conclusion that C implies H, so that for the specific fact the legal consequence mentioned in the legal norm must be applied. The difficulty lies in understanding the operation of the minor premise, since with this structure the classical syllogistic proposal is rejected and instead replaced by a hypothetical model. To understand the distinction we have to understand the relevance of the states of affairs and their related operations. The

Ontological Commitment and State of Affairs

93

discussion about this in the Austrian tradition goes back to Bolzano.²⁶ Brentano makes interesting contributions.²⁷ The Austrian school develops a whole research program in this regard. The Sachverhalt is always expressed in the form of “Vortäuschung falscher Tatsachen”. This expression contains negation through the notion “false”, which requires us to analyze the impact that the negation of truth has on semantic theories. In the main premise, general rules are expressed; in the minor, individual facts must be proved. The general rules must be interpreted and adapted to the cases to which they apply through a hermeneutical process. Individual facts are first expressed as states of affairs until they can be proven, thus transforming into true facts. Neither adjustment process is definitive, so this procedure is considered fallible. Indeed, both premises can be denied such that the procedure would change its true value, thus obtaining a new and accurate discrimination about what is postulated. In the process itself, infallible propositions are excluded as they are enunciated in classical syllogistic formulae, since the existence of propositions that are given as true without any margin of logical error cannot be accepted in advance. In fact, the states of affairs are statements that express assumptions, so it is assumed that their statement contains the possibility that it is wrong. In the process of reviewing and testing assumptions, this must always be reviewed by new observations, that is, the process must be capable of successive confirmations through certain tests. Keep in mind that the state of affairs postulate that “any assumption may become false if we do not have proof” is not logically equivalent to the proposition that “all we can say is false”, since the latter is a self-contradictory statement.

5 Conclusions As has been shown, we are confronted with a project which has produced more coincidences than disagreements. We have to build on a discussion that goes back to the anti-Kantian proposal developed by Bolzano by introducing the expression “Satz an sich”. His proposal deviated from the idealistic tradition. A gap was opened between representation (Vorstellung) and the field that it calls “objective” (objektiv), which belongs to the field of assumption (Annahme). That which is objective is an immanent object that is given intentionally.²⁸ The Bren-

 Bolzano 1837, § 24.  Brentano 1874e[2008], 241.  Brentano, 1974, 17 f.

94

Jesús Padilla Gálvez

tano School has dedicated much work to this theme. Aristotelian logic gives way to a new structure whose minor premise is a conditional structure that expresses what Stumpf would christen with the name “Sachverhalt”. This scope is expressed in two different ways: as a clause “… that—” or special forms of substantiation that express the “subsistent” (bestehen). Investigation focuses on the content of the judgment when it is expressed through a state of affairs. Subsequently, G. Frege will indicate that said content is predicative. Husserl would go one step further and affirm that the state of affairs is a bearer of properties. However, we must take into account that when the state of affairs is expressed by a name, said name does not bear the meaning of a singular term. Wittgenstein emphasized that in the case of states of affairs we must distinguish between object on the one hand and bearer on the other hand.

References Bolzano, Bernard. 1837: Wissenschaftslehre. Versuch einer ausführlichen und größtentheils neuen Darstellung der Logik, Vols. I – IV, edited by J. Ch. A. Heinroth. Sulzbach: Seidelsche Buchhandlung. Brentano, Franz. 1874[2008]: Psychologie vom empirischen Standpunkt. Leipzig: Duncker & Humblot. Psychologie vom empirischen Standpunkt. Von der Klassifikation psychischer Phänomene, edited by Thomas Binder and Arkadiusz Chrudzimski. Frankfurt: Ontos. Brentano, Franz. 1974: Wahrheit und Evidenz. Hamburg: Felix Meiner Verlag. Ehrenfels, Chr. von. 1890: “Über ‚Gestaltqualitäten‘.” Vierteljahresschrift für wissenschaftliche Philosophie 14:249 – 292. Frege, Gottlob. 1884: Die Grundlagen der Arithmetik. Eine logisch-mathematische Untersuchung über den Begriff der Zahl. Breslau: Verlage Wilhelm Koebner. Frege, Gottlob. 1891. Function und Begriff. Vortrag gehalten in der Sitzung vom 9. Januar 1891 der Jenaischen Gesellschaft für Medicin und Naturwissenschaft. Jena: Verlag Hermann Pohle. Frege, Gottlob. 1892. “Über Sinn und Bedeutung.” Zeitschrift für Philosophie und Philosophische Kritik 100:25 – 50; in: Frege, 1967, 146 – 162. Frege, Gottlob. 1918/19. “Der Gedanke.” Beiträge zur Philosophie des deutschen Idealismus 1:58 – 77; in: Frege, 1967, 342 – 362. Frege, Gottlob. 1923. “Logische Untersuchungen. Dritter Teil: Gedankengefüge.” Beiträge zur Philosophie des deutschen Idealismus 3.1:36 – 51; in: Frege, 1967, 378 – 394. Frege, Gottlob. 1967: Kleine Schriften, edited by J. Angelelli. Darmstadt: Wissenschaftliche Buchgesellschaft. Frege, Gottlob. 1976: Wissenschaftlicher Briefwechsel, edited by G. Gabriel, H. Hermes, F. Kambartel, Ch. Thiel and A. Veraart. Hamburg: Felix Meiner Verlag. Husserl, Edmund. 1913[1968]: Logische Untersuchungen. Erste Teil: Prolegomena zur Reinen Logik (1900). Tübingen: Max Niemeyer Verlag. Husserl, Edmund. 1921[1968]: Logische Untersuchungen. Zweite Teil: Untersuchungen zur Phänomenologie und Theorie der Erkenntnis (1901). Tübingen: Max Niemeyer Verlag.

Ontological Commitment and State of Affairs

95

Husserl, Edmund. 1910/1911: “Philosophie als strenge Wissenschaft.” Logos 1:289 – 341. Husserl, Edmund. 1985: Erfahrung und Urteil. Untersuchung zur Genealogie der Logik, edited by Ludwig Landgrebe. Hamburg: Felix Meiner Verlag. Husserl, Edmund. 1991: Ideen zu einer reinen Phänomenologie und phänomenologischen Philosophie. Zweites Buch: Phänomenologische Untersuchungen zur Konstitution, edited by Marly Biemel. The Hague: Martinus Nijhoff. Husserl, Edmund. 1975: Logische Untersuchungen. Erster Band: Prolegomena zur reinen Logik. Text der 1. und 2. Auflage, edited by Elmar Holenstein. The Hague: Martinus Nijhoff. Husserl, Edmund. 1984: Logische Untersuchungen. Zweiter Band: Untersuchungen zur Phänomenologie und Theorie der Erkenntnis, edited by Ursula Panzer. The Hague: Martinus Nijhoff. Husserl, Edmund. 2002: Logische Untersuchungen. Ergänzungsband. Erster Teil. Entwürfe zur Umarbeitung der VI. Untersuchung und zur Vorrede für die Neuauflage der Logischen Untersuchungen (Sommer 1913), edited by Ullrich Melle. The Hague: Martinus Nijhoff. Husserl, Edmund. 2005: Logische Untersuchungen. Ergänzungsband. Zweiter Teil. Texte für die Neufassung der VI. Untersuchung: Zur Phänomenologie des Ausdrucks und der Erkenntnis (1893/94 – 1921), edited by Ullrich Melle. The Hague: Martinus Nijhoff. Kant, Immanuel. 1983: Kritik der Urteilskraft. Berlin: Akademie Verlag, De Gruyter. Marty, Anton. 1895: “Über subjectlose Sätze und das Verhältnis der Grammatik zu Logik und Psychologie VI – VII.” Vierteljahresschrift für wissenschaftliche Philosophie 19:19–87, 263–334. Marty, Anton. 1905: “Über Annahmen.” Zeitschrift für Psychologie und Physiologie der Sinnesorgane 40:1 – 54. Marty, Anton. 1908: Untersuchungen zur Grundlegung der allgemeinen Grammatik und Sprachphilosophie. Halle: Niemeyer. Meinong, Alexius. 1882[1971]. “Hume Studien II: Zur Relationstheorie.” In Gesamtausgabe, Vol. II., by Alexius Meinong, edited by R. Haller, R. Kindinger and R. M. Chisholm, 1–183. Graz: Akademische Druck- und Verlagsanstalt. Meinong, Alexius. 1901[1977]: “Über Annahmen.” In, Gesamtausgabe, Vol. IV, by Alexius Meinong, edited by R. Haller, R. Kindinger and R. M. Chisholm, 1 – 384. Graz: Akademische Druck- und Verlagsanstalt. Meinong, Alexius. 1902[1977]: “Über Annahmen, Zeitschrift für Psychologie und Physiologie der Sinnesorgane, Ergänzungsband 2 (1902).” In Gesamtausgabe, Vol. IV, by Alexius Meinong, edited by R. Haller, R. Kindinger and R. M. Chisholm, 387 – 489. Graz: Akademische Druck- und Verlagsanstalt. Meinong, Alexius. 1904[1971]: “Über Gegenstandstheorie, Untersuchungen zur Gegenstandstheorie und Psychologie (1904).” In Gesamtausgabe, Vol. II., by Alexius Meinong, edited by R. Haller, R. Kindinger and R. M. Chisholm. Graz: Akademische Druck- und Verlagsanstalt. Meinong, Alexius. 1907[1973]: “Über die Stellung der Gegenstandstheorie im System der Wissenschaften (1907).” In Gesamtausgabe, Vol. V., by Alexius Meinong, edited by Rudolf Haller and Roderick Chisholm, 197 – 365. Graz: Akademische Druck und Verlagsanstalt.

96

Jesús Padilla Gálvez

Meinong, Alexius. 1915: “Über Möglichkeit und Wahrscheinlichkeit.” In Gesamtausgabe, Vol. VI., by Alexius Meinong, edited by R. Haller, R. Kindinger an R. M. Chisholm. Graz: Akademische Druck- und Verlagsanstalt. Meinong, Alexius. 1969 – 1978: Gesamtausgabe, edited by R. Haller, R. Kindinger and R. M. Chisholm. Graz: Akademische Druck- und Verlagsanstalt. Padilla Gálvez, Jesús. 2019: Estado de cosas. Reconstrucción de la polémica sobre el Sachverhalt. Valencia: Tirant Humanidades. Padilla Gálvez, Jesús, and Margit Gaffal (eds.). 2017: Intentionality and Action. Berlin, Boston: Walter de Gruyter. Padilla Gálvez, Jesús. 2017a: “The Paradox of Intentionality.” In Padilla Gálvez / Gaffal, 2017, 67 – 82. Padilla Gálvez, Jesús. 2021: State of Affairs. Reconstructing the Controversy over Sachverhalt. Munich, Viena: Philosophia. Quine, W.v.O. 1948: “On what there is.” Review of Metaphysics 2.5:21 – 38. Quine, W.v.O. 1961: From a Logical Point of View. Cambridge, Mass: Harvard University Press. Reinach, Adolf. 1911: “Zur Theorie des negativen Urteils”. In Münchener Philosophische Abhandlungen. Theodor Lipps zu seinem sechzigsten Geburtstag gewidmet von früheren Schülern, edited by Alexander Pfänder, 196 – 254. Leipzig: Johann Ambrosius Barth. Reprint i: Reinach, 1989, 95 – 140. Reinach, Adolf. 1913: “Die apriorischen Grundlagen des bürgerlichen Rechtes.” Jahrbuch für Philosophie und phänomenologische Forschung 2:685 – 847. Reprint in Reinach, 1989, 141 – 278. Reinach, Adolf. 1989a: Sämtliche Werke, Vol. 1, Textkritische Ausgabe. Vol. 2, Textkritische Ausgabe, Kommentar und Textkritik, edited by Karl Schuhmann and Barry Smith. Munich, Viena: Philosophia. Reinach, Adolf. 1989b: “Zur Theorie des negativen Urteils.” In Reinach, 1989, 95 – 140. Reinach, Adolf. 1989c: “Notwendigkeit und Allgemeineheit im Sachverhalt.” In Reinach, 1989, 351 – 354. Russell, Bertrand. 1904a: “Meinong’s Theory of Complexes and Assumptions (I).” Mind, N. S., 13.50:204 – 219. Russell, Bertrand. 1904b: “Meinong’s Theory of Complexes and Assumptions (II).” Mind, N., 13.51:336 – 354. Russell, Bertrand. 1904c: “Meinong’s Theory of Complexes and Assumptions (III).” Mind, N. S., 13.52:509 – 524. Russell, Bertrand. 1956: Logic and Knowledge, edited by R. C. Marsh. London: Allen and Unwin. Stumpf, Carl. 1873: Über den psychologischen Ursprung der Raumvorstellung. Leipzig: Hirzel. Stumpf, Carl. 1883: Tonpsychologie, Bd. I. Leipzig: Hirzel. Stumpf, Carl. 1890: Tonpsychologie, Bd. II. Leipzig: Hirzel. Stumpf, Carl. 1907. “Erscheinungen und psychische Funktionen.” Abhandlungen der Königlich Preussischen Akademie der Wissenschaften, Philosophisch-historische Klasse 4:1–39. Twardowski, Kasimir. 1894[2016]: Zur Lehre vom Inhalt und Gegenstand der Vorstellung. Eine psychologische Untersuchung. Viena: Verlag Alfred Hölder. Reprint in Kasimir Twardowski. Gesammelte deutsche Werke, edited by Anna Brozek, Jacek Jadacki and Friedrich Stadler, 39 – 122. Berlin, Springer.

Ontological Commitment and State of Affairs

Wittgenstein, Ludwig. 1921[1961]: Tractatus Logico-Philosophicus, trans. by D. Pears and B. McGuinness. London: Routledge. Wittgenstein, Ludwig. 1979: Notebooks 1914 – 1916, edited by Georg Henrik von Wright and Elizabeth Anscombe. Oxford: Wiley-Blackwell.

97

António Marques

“I know” Language-games in Wittgenstein’s Late Philosophy Abstract: This chapter analyses the expression “I know” in language-games of Wittgenstein’s late philosophy. Many philosophical questions regarding the meaning of “ontological commitment” arise from reading Wittgenstein’s late work, not only from Philosophical Investigations, but also from the latest writings on the philosophy of psychology and writings On Certainty. The sense in which it was possible to speak of ontological commitment in his late work is a general question that involves unique problems and difficulties in the philosophy of language. The author investigates language games that contain expressions of the kind “I know p” or “I’m sure that p”. These language games are central to the epistemic and ontological questions that Wittgenstein explored in his late philosophical writings. Keywords: language-games, knowledge, certainty, ontological commitment, Wittgenstein

Children do not learn that the books exist, that armchairs exist, etc., etc., – they learn to fetch books, sit in armchairs, etc. Later, questions about the existence of things do of course arise. “Is there such a thing as a unicorn?” and so on. But such a question is possible only because as a rule no corresponding question presents itself. For how does one know how to set about satisfying oneself of the existence of unicorns? How did one learn the method for determining whether something exists or not? ¹

A lot of philosophical questions related to the meaning of the expression “ontological commitment “ arises out from the reading of Witgenstein’s late work, not only the Philosophical Investigations (PI)² but also from the very last writings on philosophy of psychology and the writings on certainty (Gewissheit). The sense in which it is possible to speak of ontological compromise in his late work is a general question involving singular issues and difficulties in philosophy of language. By discussing the complex, many times ambiguous and always fascinating exploration of “I know” languages games in Wittgenstein’s very last writings, one is confronted with a sphere of problems that haunted the philosopher in the

 Wittgenstein, 1969, OC, 476.  Wittgenstein, 2009. https://doi.org/10.1515/9783110750041-006

100

António Marques

last years of his life³. Our aim now is not so much to solve them but simply to make them more precise in the framework of a task that is already by itself an huge clarification effort of the use of cognate verbs and expressions related to the sphere of “I know p” or “I am sure that p”. As we’ll see this task is not in the least an easy one, given the well-known intricate maieutic style of Wittgenstein, which seems even to gain more and more subtlety and new insights. Focusing mainly on the texts published in On Certainty, we’ll claim in the following that the late Wittgenstein, after dismantling the ontology of the Tractatus, did not pretend to establish a new one on another different grounds. Our claim is that in his latest writings, Wittgenstein maintains a singular first-person element in the discussion of existential propositions in the framework of his discussion of the use of the verb “to know”. The qualification of singular is grounded on the fact that “I know” language-games are central in epistemic and ontological issues that Wittgenstein explores in his last activity. Furthermore “I know” language-games are singular also because they are first-person statements, without being expressive and, as we’ll see, generate a agreement space, which plays a grounding role in our linguistic system. Let us see very briefly the ontological scenario of the Tractatus ⁴. There Wittgenstein declares that the facts are combinations of atomic facts,⁵ which can only be named but not “asserted” (3.221, that to know an object is also to know the possibilities of its occurrence in atomic facts),⁶ objects form the substance of the world and so they cannot be compound,⁷ that substance is what exists independently of what is the case⁸ and is form and content, that our representation of the world, namely empirical knowledge, is a task that presupposes the existence of atomic facts.⁹ At the center of this ontology is the proposition

 As G. H. von Wright remarks about these notes written between 1949 – 51: “Wittgenstein’s writings in the last two years of his life (after May 1949) never advanced to the typescript stage. In these writings three main themes can be clearly distinguished. The one which is treated most fully concerns knowledge and certainty, and what Wittgenstein wrote on this theme was published under the title On Certainty in 1969” (von Wright, 1982, 59). He notes furthermore that in that period “Wittgenstein wrote almost exclusively about knowledge and certainty” (op. cit., 166). One must add that in those two or three last years there was also a dominant pole of interest, that is, the philosophy of psychology.  Wittgenstein, 1922[1981].  Wittgenstein, 1922[1981], TLP 2.  Wittgenstein, 1922[1981], TLP 2.01229.  Wittgenstein, 1922[1981], TLP 2.021.  Wittgenstein, 1922[1981], TLP 2.024.  Wittgenstein, 1922[1981], TLP 2.1.

“I know” Language-games in Wittgenstein’s Late Philosophy

101

that and atomic fact is a combination of objects (entities, things),¹⁰ which can only be named but not “asserted”¹¹. Furthermore objects are simple qua substance of the world and this latter guarantees the truth of any proposition. If it were not the case, “If the word had no substance, then whether a proposition had sense would depend on whether another proposition was true”.¹² In fact the Tractatus refuses a mere coherence, non-realistic theory of truth, since a wellgrounded theory of truth requires an ontological commitment. Then a language, e. g. propositional language presents the existence and non-existence of atomic facts¹³ and the “totality of true propositions is the total natural science (or the totality of natural sciences)”.¹⁴ The idea of the presentation of atomic facts by the language of natural sciences is of course quite problematic, since this kind of presentation can only be assured by a strong ontological commitment. Namely one commitment that assures that natural sciences can be true only because there is a substance of the world that philosophy must postulate. Postulating and not researching because, as Wittgenstein notes, “Philosophy is not one of the natural sciences. (The word philosophy’ must mean something which stands above or below, but not beside the natural sciences.)”.¹⁵ The problem remains, how can natural sciences produce truth propositions about something, atomic facts that is postulated as a simple substance by philosophy? In fact, it is this picture (atomic acts qua substance of the world) that is dismantled by the late Wittgenstein and it can be said that this is his main task at the beginning of the PI, associated with the well-known criticism of the so called Augustinian picture of language. PI 46 remembers one of the doctrines about the nature of “knowledge” discussed by Socrates in Theaetetus (it doesn’t correspond to Socrates’ doctrine itself) that simple or primary elements can only be named, not qualified in any way, even with the existential predicate; so simple objects cannot be said to exist or not. This mereological problem is discussed at length in the dialogue, which is led by Socrates in an aporetic style¹⁶. In fact the difficulties related to  Wittgenstein, 1922[1981], TLP 2.01.  “Objects I can only name (nennen). Signs represent them. I can only speak of them. I cannot assert them (sie aussprechen). A proposition can only say how a thing is, not what it is” (Wittgenstein, 1922[1981], 3.221).  Wittgenstein, 1922[1981], TLP 2.0211.  Wittgenstein, 1922[1981], TLP 4.1.  Wittgenstein, 1922[1981], TLP 4.11.  Wittgenstein, 1922[1981], TLP 4.111.  If one doesn’t assume a realistic approach, mereological problems are solved depending on what sort of conceptual framework is used. This is the view of H. Putnam, who discusses different mereological approaches, such as those held by Carnap in the 1950s and those held by the

102

António Marques

the knowledge of the primitive elements and the composed wholes are un-surmountable, but I leave here this fascinating text on the theme of ‘knowledge’. What is relevant is the proximity that Wittgenstein finds between the doctrine discussed by Socrates about simple objects and his own early ontology. So he remarks that “Both Russell’s ‘individuals’ and my ‘objects’ (Tractatus LogicoPhilosophicus) were likewise such primary elements”¹⁷. In the PI the self-criticism of Wittgenstein develops itself in the context of his conception of language as the limitless exercise of language games. Thus “What are the simple constituent parts of which reality is composed? – What are the simple constituent parts of a chair – The bits of wood of which it is made? Or the molecules, or the atoms? – ‘Simple’ means: not composite. And here the point is: in what sense ‘composite’? It makes no sense at all to speak absolutely of the ‘simple parts of a chair’”. What Wittgenstein has in mind is that “asking ‘Is this object composite?’ outside a particular language game is like what a boy once did when he had to say whether the verbs in certain sentences were in the active or passive voice, and who racked his brains over the question whether the verb ‘to sleep’, for example, meant something active or passive.”¹⁸ One can also rack his brains about the simplest parts of a chair, a tree, an object in general, essentially because there are always different kinds of composition and decomposition. A different move from the composite to the simplest parts is determined by the particular language game that is used. At a certain point it is possible that we consider the end of decomposition and to call this or that element “simple”, but it is always a determination in a language–game. Then what in the Tractatus is called ‘naming (benennen) something as simple’ turns out to be in the language-game conception of language a kind of redundant, ineffective act. Naming is not yet a move in a language game – any more than putting a piece in its place on the board is a move in chess. One may say: with the mere naming of a thing, nothing has yet been done. Nor has it a name except in a game.¹⁹

In this ontology declaring the being and non-being of simple elements are equally legitimate positions. That’s why the dismantling of the Tractatus ontology operated in the PI doesn’t generate its absolute opposite, in the sense that in the PI

Polish logicians who argue that for every two particulars there is an object that is their sum (Putnam, 1987, p. 18). One finds a similar view in Teaethetus, where two letters of the syllabe SO (first of the word Socrates is to be considered as a simple whole apart from his letters S and O).  Wittgenstein, 2009, PI, 46.  Wittgenstein, 2009, PI, 47.  Wittgenstein, 2009, PI, 49.

“I know” Language-games in Wittgenstein’s Late Philosophy

103

it is claimed that primitive elements don’t exist. It is curious that in the Teaethetus, Socrates seems to suggest that primitive elements can be eventually named without a supplement of justification, and even that they have not being, yet one must acknowledge that naming and teaching such primitive elements is a necessary exercise as for example in the case of the letters of a syllable.²⁰ The mereology of Wittgenstein’s PI is commanded by the performance of language-games, and thus what counts as the simplest part of a whole and how one constructs it depends on the relevant language-game. Saying that there is one meter in Paris is not to say neither that there is a thing one meter long, nor that there is not. Saying that means only that one marks its peculiar role in the language-game of measuring with a meter-rule, likewise if I keep hermetically sealed a certain color, say ‘sepia’, as one did with the meter in Paris, what I do is using such samples as an instrument of the language-game in ascriptions of length or color.²¹ And of course one can name a thing ‘the Paris meter’, another thing ‘sepia’. At this point we must ask what sort of ontology is that of the PI, which results from the dismantling of the Tractatus substance of the world, namely from the sections of the PI, where the mereological question is discussed? Are we allowed to speak of any kind of ontological commitment in this new context the language-game theory? We suggest to consider the following main issues, which in my view, deserve to be highlighted. First, the conception of the use of language as a countless language-games undermine any realistic ontology, since the existence or non-existence of primitive elements of any compound are determined by the linguistic scheme. Recall the idea expressed by Wittgenstein in PI about the transcendental element of his philosophy,²² where he notes that his investigation is directed not toward phenomena, but to the possibilities of phenomena. Second there is no more a true ontology that corresponds to the whole of propositions of natural sciences as was declared in the Tractatus. Third, a language-game theory is not committed ontologically per se, yet some language-games related to the cognitive activity imply an ontological commitment. Let’s keep in mind this last point, based on the inquiry that Wittgenstein carries out on some of those texts published under the title, Über Gewissheit (On Certainty). If it is true that the conception of language as an activity of language-games undermines any realistic ontology, at least one of the kind of that of the Tracta-

 Platon, Theauthetus, 203a – 205d.  Wittgenstein, 2009, PI, 50.  Wittgenstein, 2009, PI, 90.

104

António Marques

tus-Tractatus, yet it does not erase the inquiry on the “possibility of phenomena” as it was mentioned above.²³ I suggest that an important part of this inquiry takes place and is made explicit in the last Wittgenstein’s work namely in the reflections on the use of cognitive verbs, like the use of “knowing”, “believing”, “doubting”, “being sure”, and so on. The texts edited by G.E.M. Anscombe and G.H. von Wright under the title Über Gewissheit, belong to the last two, three years of the philosopher’s life and represent an exhaustive exploration of that region of language-games. Wittgenstein was stimulated by Moore’s work about the nature of common sense related to the proof of the external world and his defense of the former kind of knowledge was exemplified by certain declarations like “I know with certainty that this is my hand”, “The earth existed long before I was born”. What interests us here is not the Wittgenstein’s refutation of Moore, but how or in what measure the use of cognitive language-games clarify what we referred to as generating the conditions, a kind of transcendental conditions for ontological compromise in general. In the context of On Certainty some main issues must be accessed: 1. “I know” language-games are not cognitive expressions (Äusserungen, Ausdrücke) in the sense of for example “I believe”, “I’m sure”, etc. “Knowing p” exhibits a first- person perspective, what suggests that that all judgments of existence with the knowing clause are first-person declarations, 2. “I know” language-games require an agreement in the same system of references, or in the same form of life, 3. “I know” language games, are rooted in action, in Handlung, 4. though knowing p doesn’t set up an ontological criterion, e. g. of what is and is not, the associated ontological commitment is a necessary, natural rooted mechanism. Let’s begin with the first point. “I know” language-games are singular cognitive language-games, distinct of believing²⁴. The difference lies on the expres-

 Wittgenstein, 2009, PI, 90.  Some authors consider a huge misunderstanding interpreting On Certainty as upholding a form of knowing or epistemic certainty. This is the view of Danièle Moyal-Sharrock, who furthermore states that “it appears to Wittgenstein that our knowing something is not our ultimate way of being sure; it does not constitute our fundamental assurance about our world and ourselves. Underlying knowing is a bedrock, logically solid, objective certainty. A nonepistemic belief, not a knowable one” (Moyal-Sharrock, 2007, 25). It is correct not to consider knowing as our ultimate way of being sure, but knowing language games in the context of linguistic activity are solid and objective pieces that sustain a form of life. Some of these games have a specific role in the regularity of our form of life, but it is doubtful to qualify them as foundational, as well as emptying out them of any epistemic value. Wittgenstein, 1969, OC 96 – 99 refer to experience propositions (Erfahrungssätzen) that acquire a rigid (erstarrt) form and direct other experience propositions. But the picture is not that of a classical foundationalist philosophy. About the problem of the

“I know” Language-games in Wittgenstein’s Late Philosophy

105

sive root of “I believe”, which is not the case of “I know” and this difference is remarkable. In fact One can say ‘He believes it, but it isn’t so’ but not ‘He knows it, but it isn’t so“²⁵. “It would be correct to say: ‘I believe… ‘has a subjective truth; but ‘I know…’ not.²⁶ “Or again ‘I believe…’ is an expression’, but not ‘I know’.²⁷

Despite not being an expression in the above sense, it is crucial to keep in mind that ‘I know’ language games have a first-person characteristic, which seems, at first view, to be only a property of expressive statements, like “I believe”. How to handle this issue? Perhaps one must consider “I know” a special case apart from that dichotomy between expressive and descriptive language games. That is to say, if “I know that …such and such…” is not an expressive statement, it is also not a descriptive one. In War and Peace, Tolstoy is able to describe with tremendous accuracy a huge quantity of details of the battle of Borodin; despite such accuracy, however, it would be correct to say that he did not know many or even most of that huge quantity of such details. At the same time it would be right to say that he knew, through all the historic sources he accessed, not only that the Borodin battle took place in certain space and time coordinates, but also that some of those details that he described were true. He knew some of those things he described (he knew such things with the community of historians) and at the same time he described some others details that he didn’t know, what of course didn’t threatened the incomparable literary quality of his novel. Describing is not the same as knowing. So it is plausible to claim that “I know” language-games are of a different linguistic– epistemic nature of those either of expressive or descriptive games, what justifies our claim that they are singular cognitive ones. In On Certainty, Wittgenstein gives a striking characterization of the singularity of “I know”:

reversibility of the demarcation line between “river-bed propositions and fluid propositions, see Wittgenstein, 1969, OC, 96 – 97. The rigid, for ever separation between these two categories of experience knowledge is a kind of mythology: “The mythology may change back into a state of flux; the river-bed of thoughts may shift. But I distinguish between the movement of the waters on the river-bed and the shift of the bed itself; though there is not a sharp division of the one from the other” (Wittgenstein, 1969, OC, 97). About his specific items see: Bouveresse, 1987, 597– 599.  Wittgenstein, 1969, OC, 42.  Wittgenstein, 1969, OC, 179.  Wittgenstein, 1969, OC, 180.

106

António Marques

I know has a primitive meaning similar to and related to ‘I see’ (‘wissen’, ‘videre’). And ‘I knew he was in the room, but he wasn’t in the room’ is like ‘I saw him in the room, but he wasn’t there’. ‘I know’ is supposed to express a relation, not between me and the sense of a proposition (like ’I believe’) but between me and a fact. So that the fact is taken into my consciousness.²⁸

In the same vein OC 42 reads: “One can say ‘He believes it, but it isn’t so’, but not ‘He knows it, but it isn’t so’. Does this stem from the difference between the mental states of belief and knowledge? No!”²⁹ Regarding the second issue “I know” establishes a relation between the subject and the facts with a singular linguistic-epistemic quality, belonging to another category different of ‘believing’. It would be right to define such quality as a general condition for objective shared experience. ³⁰ Exploring the use of “I know” as Wittgenstein does in his late reflections that compose Über Gewissheit, leads to acknowledge that it (given certain circumstances) creates shared experiences, taken as objective, that is perceived as true by other people. Agreement and objectivity are generated by ‘I know’, not by another cognitive language-games, like ‘I believe’, “I am of the opinion’, for example. The centrality and the power of ‘I know’ is of course a first-person relation to X, but at the same time is non personal, a kind of universal. It must be added, as we’ll see, that ‘I know’ qua language game, takes place in a practical context shared by other people. In Wittgensein’s words: “Even the statement ‘I know that behind this door there is a landing and the stairway down to the ground floor’ only sounds so convincing because everyone takes it for granted that I know it”³¹. One could even say that “There is something universal here; not just something personal”.³² This answer to the second issue we have pointed out above: “I know” language-games is what generates agreement in the same system of references, or one can say, what generates truth. The third main point stressed that “I know” language-games, qua language-games, are rooted in action, in Handlung. The reference to Goethe’s Faust in Wittgenstein’s writings: ‘In the beginning was the deed” expresses the idea that ‘I know’ is not separable from acting. “‘I know all that”. And that will come out in the way I act and in the way  Wittgenstein, 1969, OC, 90.  Wittgenstein, 1969, OC, 42.  This status of “I know” language-games as shared objectivity knowledge is not a central issue in the literature dedicated to On Certainty. It is not the case of Doris Vera Hofmann, who develops an interesting argumentation comparing Kant and Witgenstein’s approaches on truth and certainty against the background of shared external knowledge (Hofmann, 2000).  Wittgenstein, 1969, OC, 439.  Wittgenstein, 1969, OC, 440.

“I know” Language-games in Wittgenstein’s Late Philosophy

107

I speak about the things in question”.³³ The necessary relation knowing–acting is absolutely clear: Giving grounds, however, justifying the evidence, comes to an end; – but the end is not certain propositions striking us immediately as true, i. e. it is not a kind of seeing on our part; it is our acting, which lies at the bottom of the language-game”.³⁴

Of course acting that is on the basis of “I know” points out to an active subject, who generates by this singular language-game the objective- accepted by others truth. In the writings that compose On Certainty the system of knowledge generated in this process presupposes the child learning of a system of primitive language-games and immediately at this primitive stage, “I know” emerges as the linguistic-epistemic condition, of a different category of for example, believing. At this point Wittgenstein suggests regarding man as an animal, “as a primitive being to which one grants instinct but not ratiocination. As a creature in a primitive state. Any logic good enough for a primitive means of communication needs no apology from us. Language did not emerge from some kind of ratiocination”.³⁵ In fact children don’t learn that book exist, that armchairs exist, etc., etc., – they learn to fetch books, sit in armchairs, etc., etc. “Does a child believe that milk exists? Or does it know that milk exists? Does a cat know that a mouse exists’?”.³⁶ “I know” emerges from a natural instinctual ungrounded background and it is from this kind of background that emerges the languagegame “I know this and this”. Now perhaps we have enough elements in order to access the issue of the ontological compromise in the philosophy of the late Witgenstein, after the radical criticism of his previous ontology of the Tractatus. Our initial question was: in which terms is it possible to speak on ontological compromise in his latest work? Any answer to this general question needs to take in account a) the linguistic-epistemic centrality of the “I know” language-game, stressing its non-expressive first-person use, b) its power of generating objective-agreed truth and c) its natural-instinctual primitiveness. A negative answer to that question would stress that a post-Tractatus ontological compromise is not expressive (in the sense of Wittgenstein), so it doesn’t consist in expressive statements, it is not realistic in the sense of a metaphysical realism, it is not a mental state, it is not independent of a linguistic framework, that is separated of the use of lan-

   

Wittgenstein, 1969, Wittgenstein, 1969, Wittgenstein, 1969, Wittgenstein, 1969,

OC, 395. OC, 204. OC, 475. OC, 478.

108

António Marques

guage in the sense of the activity of language-games. In addition, it is not a kind of individual, solipsistic procedure, on the contrary the “I know” language-game requires interpersonal agreement in a same system of references and action. Indeed we can speak of a “we know” as Wittgenstein asserts for example: “‘We are quite sure of it’ does not mean just that every single person is certain of it, but that we belong to a community which is bound together by a science and education.”³⁷ We conclude with the following idea. A (metaphysical) realist will remark with some reason that an ontological compromise like this one, resulting from the mere clarification of the grammar of the use of “I know”, fails to answer to the question about what really exists. In the end what does guarantee our judgments about existence? Yet it looks to be a failure not delivering criteria of being or not being. After all it seems that “I know” adds nothing to our knowledge of the world, and the statement “I know that there is a tree there” has the same content of “there is a tree there”. In some writings on the philosophy of psychology Wittgenstein had already remarked this superfluity of “I know” in a certain kind of nonsense statements like “I know I have a pain” or “I know I believe”, which are expressive first person avowals exclude any cognitive verb or making superfluous any cognitive verbs. In On Certainty, however, it is the use of “I know” that generates the possibility of experiencing phenomena beyond a subjective relation to the world. As Wittgenstein put it, “If someone says he knows something, it must be something that, by general consent, he is in a position to know”.³⁸ As noted above, “I know” language-games possess a collective agreed value. I look at an object and say ‘That is a tree’, or ‘I know that that’s a tree’- Now if I go nearer and it turns out that it isn’t, I may say ‘It wasn’t a tree after all’ or alternatively I say ‘It was a tree but now it isn’t any longer’. But if all the others contradicted me, and said it never been a tree, and if all the other evidences spoke against me – what good would it do me to stick to my ‘I know ?’.³⁹

Wittgenstein’s reflections on the grammar of knowing language-games are a complex field on which epistemic, linguistic and ontological questions intersect in quite intricate ways. What is certain is that what can be called ontological compromises in relation to phenomena in the world require the sharing of

 Wittgenstein, 1969, OC, 298.  Wittgenstein, 1969, OC, 555.  Wittgenstein, 1969, OC, 503.

“I know” Language-games in Wittgenstein’s Late Philosophy

109

“I know” language-games, which are non-expressive, yet first-personal linguistic acts that generate collective agreement.

References Bouveresse, Jacques. 1987: Le Mythe de l’Interiorité. Paris: Minuit. Hofmann, Doris Vera. 2000: Gewiβheit des Fürwahrhaltens. Berlin, New York: Walter de Gruyter. Moyal-Sharrock, Danièle. 2007: Understanding Wittgenstein’s On certainty. London: Palgrave Macmillan. Platon, 1925: “Theaetetus” In Platon: Protagoras, Theaitetos, trans. by Karl Preisendanz. Jena: Diederichs. Putnam, Hilary. 1987: The Many Faces of Realism. La Salle, Illinois: Open Court. Wittgenstein, Ludwig. 1969: On Certainty/Über Gewißheit, edited by G. E. M. Anscombe and G. H. von Wright, trans. by Denis Paul and G. E. M. Anscombe. Oxford: Basil Blackwell. Wittgenstein, Ludwig, 1922[1981]: Tractatus Logico-Philosophicus. London and New York: Routledge. Wittgenstein, Ludwig. 2009: Philosophical Investigations, the German Text, with an English translation by G. E. M. Anscombe, edited by G. E. M. Anscombe, P. M. S. Hacker and Joachim Schulte, revised 4th edition. Oxford: Wiley-Blackwell. Wright, G. H. von. 1982: Wittgenstein. Oxford: Basil Blackwell.

Manuel García-Carpintero

Against Propositional Substantivism Abstract: Jeff King, Scott Soames, and Peter Hanks have advanced substantive theories of propositions, to deal with several issues they have raised in connection with a concern with a long pedigree in philosophy, the problem of the unity of propositions. The qualification ‘substantive’ is meant to contrast with ‘minimal’ or ‘deflationary’ – roughly, views that reject that propositions have a hidden nature, worth investigating. Substantive views, I’ll argue, create spurious problems by characterizing propositions in ways that make them unfit to perform their theoretical jobs. I will present in this light some critical points against recent substantivist proposals. Keywords: propositions; unity of propositions; representational content; representational vehicles

1 Introduction Unlike other concepts that philosophy has used from its earliest beginnings in formulating its proprietary set of questions (What are things ultimately (made of)? What do we know? What is a good deed, or a good work of art?), proposition, like a priori knowledge or possible world, are technical-theoretical notions that philosophers deploy to provide answers to such questions. They don’t figure as such in questions the layman asks. The closest folk pretheoretical conception that philosophical theories of propositions develop is deployed in talk about sameness of content, and in generalizations like ‘what S said/thought’.¹

Note: Financial support was provided by the DGI, Spanish Government, research project FFI2016 – 80588-R, and through the award “ICREA Academia” for excellence in research, 2013, funded by the Generalitat de Catalunya. This work received helpful comments from audiences at the Diaphora workshop on the Nature of Representation, Stockholm, and at LOGOS and MELL, LANCOG seminars. Thanks to John Collins, Richard Gaskin, Peter Pagin, Michele Palmira, Bryan Pickel, Indrek Reiland, François Recanati, Ricardo Santos and Elia Zardini. Thanks also to Michael Maudsley for the grammatical revision.  See Moore, 1999, 3. He convincingly argues that such pretheoretical notion is, as he puts it “individuatively vague”: different contexts involving different explanatory goals for talk of sameness of content or what is said make salient different individuative criteria. This is consistent with the deflationary view endorsed in this paper, see fn. 8 below. https://doi.org/10.1515/9783110750041-007

112

Manuel García-Carpintero

Here are the core features in the philosophical ‘job description’ for propositions – the theoretical roles that they are posited to play, developing that folk notion.² ‒ Objects/contents of attitudes like belief, speech acts like assertion, and perhaps others in each category; ‒ (Partial) meanings of utterances of declarative sentences, and perhaps others; ‒ Referents of ‘that’-clauses; ‒ Bearers of truth and falsity, and the modalities of truth and falsity: necessity, possibility, probability (subjective or epistemic, objective or metaphysical); ‒ What gets assessed in determining the validity of arguments. To properly perform these jobs, propositions should be sufficiently coarsegrained. It should be apt to ascribe the same proposition to an English sentence like “Sophie loves Carl”, and to its translation into an SOV language like Turkish, or an ergative language like Basque or Georgian (Collins 2014, 145). This would afford an easy explanation, say, for why it is rational for a bilingual speaker who has formed a belief in her English inner speech to assert it in its Basque translation for the benefit of her Basque audience. For related reasons, it is desirable to assign the same proposition as content to perceptual experiences (assuming that they have them, with many contemporary views) and to the perceptual beliefs they help to justify.³ Writers including Gaskin (2008), King (2007), Soames (2010) and Hanks (2015) advance different substantive theories of propositions, to deal with issues they have raised, in the vicinity of a concern with a long pedigree in philosophy, the so-called problem of the unity of propositions. ⁴ ‘Substantive’ here is meant to contrast with ‘minimal’ or ‘deflationary’. One might be skeptical of substantive views about propositions on account of a general attitude about the theoretical posits of philosophy. This might be motivated by Yablo’s (2014a) “quizzical” standpoint – the notion that there is no reasonable way that controversies among proponents of prima facie conflicting substantive ontologies can be adjudicated. Or instead by the view that such entities are posits of an “easy ontology” (Thomasson 2015) and, as such, lack a “hidden nature” worth investigating and debating about. Or the skepticism might be rather grounded on a deflationary

 See King’s “What Role do Propositions Play in our Theories?” In King et al., 2014.  I don’t mean to suggest that the explanation of the justification relation would thereby be immediate; cf. below, fn. 17.  Cf. Davidson’s, 2005, 76 – 97 nice presentation of the early history of the debate.

Against Propositional Substantivism

113

attitude specific to the case of propositions; say, one motivated by the measuretheoretic perspective on their explanatory role promoted by Davidson, Perry, Stalnaker and others.⁵ Like numbers measuring quantitative properties, on this view propositions are just convenient tools used to represent through their relations the semantic relations among attitudes and acts in their job description. The latter motivation is of course consistent with the former. Skeptical attitudes about propositions predict that substantive views will create spurious problems, by characterizing propositions in ways that make them unfit to perform their theoretical jobs. In support of a skeptical view, this paper aims to show that recent substantivist proposals confirm the prediction. After briefly outlining the unity debates in the next section, I will rehearse along such lines in the third critical points against King’s view, and against Hanks’ (2015) view in the fourth. On the view I want to promote, the explanations that have been sought by ascribing controversial natures to propositions should be instead given in terms of features of the representational vehicles to which they are ascribed. I intend to provide here indirect support for this view, in so far as it explains the difficulties here presented for substantive accounts.

2 Substantive vs. Deflationary Solutions to Unity Problems A cluster of unity of the proposition problems exercised the founding fathers of Analytic Philosophy, following in the footsteps of a distinguished tradition traceable back to Plato’s later dialogues. What is it that holds the constituents of a proposition together? As this way of framing it shows, the problem presupposes that we can somehow discern ‘constituents’ in propositions’ – the scare quotes will be explained soon. In fact, as it has been pointed out,⁶ there are several problems that have been discussed under that label. A deflationary attitude toward propositions suggests that they are created by the assumption that propositions, like the theoretical entities posited in empirical science, have hidden natures of which we can provide substantive explanatory accounts. King (2019, 1346) makes it explicit. He says that the unity problems concern properties that “seem to call out for further explanation and whose possession seems as though it should be grounded in the possession of ‘more basic’ properties. It may be hard to give a criterion for being  Cf. Green, 1999; Matthews, 2007, 2011; Field, 2016, and Ball, forthcoming.  Cf. King, 2009; Schnieder, 2010; Eklund, 2019, and Collins, 2018.

114

Manuel García-Carpintero

such a property, but properties like being alive, believing that snow is white, and being morally good seem to be examples of such properties”. Against this, philosophers like Bealer (1998) and Merricks (2015) advance a ‘quietist’ or ‘primitivist’ view, opposing the project of providing substantive accounts. As Lewis (1983, 352) protests in a similar context discussing the related Third Man-like regresses: “Not every account is an analysis! A system that takes certain Moorean facts as primitive, as unanalysed, cannot be accused of failing to make a place for them. It neither shirks the compulsory question nor answers it by denial. It does give an account.” Now, against bare deflationism (like perhaps Bealer’s and Merricks’s), the deflationary view I want to promote takes on explanatory concerns in the vicinity of those allegedly addressed by substantive proposals, offering informative answers for them but shunning substantive commitments to the nature of propositions. We can increase our understanding of Moorean facts that are to be taken as primitive by placing them within a wider net of interlocking concepts, some of which receive analyses or definitions. Even though proposition, a priori knowledge and possible world are theoretical kinds, we are acquainted with the phenomena they are intended to categorize, and have intuitions about them. As indicated, there are folk pretheoretical conceptions that philosophical theories of propositions develop, manifest in talk about sameness of content, and generalizations about what S said/thought. In indicating how propositions are supposed to perform their theoretical roles, we can thus call our attention to familiar features of the linguistic and mental representations to which their ‘constituents’ are ascribed, thereby addressing some concerns related to the unity issues. Liebesman’s (2015) account nicely illustrates this point. Elaborating on previous work on Frege’s theory of predication by Burge (2007) and Wright (1998), he distinguishes three views of the semantics of predicates in general, defending one of them. On the entity view associated with Frege, what semantically distinguishes predicates is the (‘unsaturated’) nature of the entities they designate. On a nominalism-friendly matching alternative associated with Davidson, predicates do not need to designate anything. The semantic role of a predicate in a sentence just consists in matching its arguments in the sentence with conditions for it to get a truth-value. On the ascription view that Liebesman defends, predicates do designate properties that can also be the referents of terms. But what semantically defines predicates is the ascription relation that they bear to their designata, distinguishable from the referring relation that terms bear to their referents by the role that the relation plays in matching properties and referents to truth-conditions and truth-values for the sentences in which terms and predicate co-occur (cf. also Johnston, 2006, 681– 687, and Burge, 2007, 600). This may afford some understanding of the relation that puts together the ‘constituents’ of proposi-

Against Propositional Substantivism

115

tions, derived from semantic features of the terms contributing to their expression. However, as Liebesman (2015, 543, 551) grants, when it comes to accounting for the nature of propositions and facts, this is a pretty shallow sort of explanation, which will not satisfy the ambitions that King expresses; for it just assumes as given our previous familiarity with such relation unifying them.⁷ Liebesman’s is thus a proposal that, while noncommittal regarding the natures of propositions, does provide some illumination on the relevant issues, by appealing to properties of the representational states/acts to which those propositions are ascribed. Hartry Field (1992, 322) makes a similar point about propositional truth-minimalism: “on most conceptions of propositions, the question of what it is for a proposition to be true is of little interest, […] what is of interest are the issues of what it is for an utterance or a mental attitude to be true (or, to express a truth or represent a truth)” (cf. also Lewis, 2001, 602– 603). In this paper, I want to recommend the same vehicle-first attitude regarding a related unity of the proposition problem that has been discussed more prominently recently, which Eklund (2019) calls the representation problem. This is the alleged problem of explaining the representational properties of propositions, in particular their truth-aptness and their having truth-conditions. It is related to the more traditional concern: “the judgment or sentence must in some way constitute a unity; its parts must fit together to produce something that can be true or false” (Davidson, 2005, 82). Liebesman (2015, 550) similarly says that “we may reasonably expect that any account of the unifying relations of propositions will help us to understand their truth-aptness”. Predictably, a deflationary attitude has also been advanced toward this version of the problem: “the truth-value aptness of a proposition is at least a good candidate for belonging to the fundamental essence of a thing. But if this is so, the only legitimate reply … may consist not in a direct answer, but instead in a rejection of the question” (Schnieder, 2010, 300). In what follows, I’ll argue for a vehicle-first version of Schnieder’s take,

 Like King, Davidson also expresses high ambitions in his discussion of these topics. He considers a proposal like the one by Burge, Johnston and Liebesman just outlined, and dismisses it with the complaint that “the purported asymmetry does not explain the relation between the thing named and the property” (Davidson, 2005, 86). Earlier he had warned us that, without a substantive account of that relation, the “philosophy of language lacks its most important chapter …; the philosophy of mind is missing a crucial first step if it cannot describe the nature of judgment; and it is woeful if metaphysics cannot say how a substance is related to its attributes” (ibid., 77). Deflationists would dismiss the worry, coming back to Lewis’s point. Tellingly, Burge, 2007, and Liebesman, 2015, provide compelling reasons to doubt that Davidson’s own proposal accomplishes his ambitious goals.

116

Manuel García-Carpintero

by raising worries for substantivist accounts offered by King, Hanks and Soames reproducing a common pattern. I’ll explain now the scare quotes placed so far around ‘constituents’. On an influential account – still the default in contemporary semantics – propositions are sets of worlds, and these in turn are understood by Stalnaker (1976) as given by ways or properties the world might have. I endorse Stalnaker’s view that propositions are just such properties.⁸ As properties, propositions might be finergrained than the sets of worlds at which they are instantiated, and thus not identical to them, thereby dodging well-known difficulties with the identification of propositions with such sets. They can also be taken as properties of entities other than worlds, as in the “truthmaker semantics” of Yablo (2014b) and Fine (2017). This further supports a deflationary standpoint on the representation problem, for no substantive explanation should be expected for why properties apply to things: this is just what properties are, by their very nature.⁹ Propositions understood along these lines do not have constituents in any substantive sense.¹⁰ However, deflationary views might still countenance singular propositions expressed by sentences with names or indexicals. They can explain this in traditional terms, as propositions that are ‘directly about’ objects, though not in virtue of their properties. This is still deflationary because it is taken to merely articulate, without in any way explaining it, the intuitive contrast between the sense in which the propositions expressed by sentences with attributively used descriptions are also ‘about’ the entities satisfying the description, and the one in which those expressed when names or indexicals replace them are ‘about’ their referents. I think Stalnaker assumes this line here: “what is  Cf. Speaks’ “Propositions are Properties of Everything or Nothing”, in King, Soames, & Speaks, 2014; Richard, 2013, and Pautz, 2016. Of course, ‘deflationary’ is a relative term; my take on propositions is deflationary vis-à-vis the ambitions expressed in quotations from King and Davidson above, creating the problems I am about to present. I don’t take the identification of propositions with properties of worlds or situations to be substantive – an ascription to them after all of a “hidden nature” –, because I assume an equally deflationary view of properties. Note that most current views on truth that count as deflationary agree nonetheless that truth is a property in such “abundant” – as opposed to “sparse”, cf. Lewis, 1983c, – understanding. Properties may be understood as explained by resemblance nominalism. Propositions might ultimately be equivalence classes of representational vehicles, as in traditional deflationary views – cf. Buchanan / Grzankowski, 2019; Field, 2016, and Sainsbury, 2018, for related views. As such properties, propositions might be more or less fine-grained for different explanatory purposes, thus ending up “individuatively vague”, cp. Moore’s, 1999.  Ostertag, 2013, 519, reports that Stalnaker himself pointed this out. The properties that I will take propositions to be correspond to the states of affairs that on Matthews’, 2007, 153, presentation of the measure-theoretic account are representatives of the attitudes.  Cf. Keller, 2013, for the problems the contrasting view raises for substantivists.

Against Propositional Substantivism

117

it, on this picture, to believe a singular proposition – a proposition about an individual as it is in itself? One can correctly and aptly ascribe a de re belief whenever one can give a correct and unambiguous characterization (in a given context) of the way the world is according to the believer by referring to a certain individual and saying that, in the world according to the believer, that individual is a certain way” (2009, 243).¹¹ I have been placing the term ‘constituent’ inside scare quotes so far to leave it open for deflationary views to nonetheless contemplate propositions that are, in a distinct sense, about particular objects. Propositions might also be said to have properties as ‘constituents’ in this deflationary sense, if one accepts Liebesman’s compelling reasons for positing entities indicated by predicates in explaining their semantics. Pickel (2017) shows how structured propositions can be made compatible with the standard type-driven functional application that accounts for compositionality in current semantics. He distinguishes for that goal the compositional semantic values of expressions, from the constituents of propositions compositionally assigned to the sentences in which they occur. The deflationary viewpoint may help itself to this proposal, ignoring the structure of Pickel’s propositions by thinking of them as properties. Understood in this way, the proposal has a compelling response to Duncan’s (2018) argument for structured propositions, based on the systematicity and productivity of beliefs. In my terms, compositionality is just a vehicle-level feature. To account for systematicity and productivity we must ascribe structure to representational vehicles, but we cannot derive the structure of contents from the compositional structure of vehicles expressing them, not even their structured character. Duncan’s argument concern mental items, while Pickel’s proposal is about natural language. However, the point extends to them, once we assume representational vehicles there too, as I think we should. Thus, there are good reasons to ascribe to visual experiences a “nonconceptual” map-like character, with a systematicity of its own, and a corresponding compositionality (Giardino / Greenberg, 2014; Camp, 2018; Lande 2018). The differences between the structure that experiences thus get, and that of the beliefs they may generate and justify,

 Cf. Merricks, 2015, 157– 158, Speaks’s “Representation and Structure in the Theory of Propositions” In King et al., 2014, and Keller’s, 2013, 670, “lightweight” constituency. Glick, 2017, 31– 32, questions Stalnaker’s account, correctly pointing out in my view that it is ultimately a form of “vehicle-first” accounts of singularity, which he rejects. Glick longs here, I think, for a more informative account of the singular/general distinction than can be had in terms of propositions. In contrast, here too approaches at the level of “vehicles” or representational acts do allow for a more informative account.

118

Manuel García-Carpintero

are once more vehicle-, not content-level. This leaves open that the experience and the belief fully share representational content (García-Carpintero, 2006).¹²

3 Substantive Accounts and their Discomforts Jeff King, Scott Soames, and Peter Hanks have produced influential work in recent years aiming to provide substantive responses to the representation problem. They find it puzzling that, as King (2019, 1343) puts it, “propositions have truth conditions, and so represent the world as being a certain way, by their very natures and independently of mind and languages”. They thus dismiss primitivism on these matters. In a quote above, King mentions properties like being alive as analogous cases for which we presume that there is a substantive account of their natures. He (2019, 1346) rightly points out that “[s]entences, minds, maps, perceptual experiences all represent. And in each case we feel compelled to explain how/why such things manage to do this. Surely it would seem utterly mysterious to adopt the view that e. g. there is no explanation of how/why perceptual experiences represent things as being a certain why because that they do so is metaphysically basic.” He argues that, analogously, “we should think that same thing about the claim that propositions have truth conditions”. But the analogy doesn’t work. King grants that the primitivist line is justified in the case of propositions understood as sets of possible worlds, or as the properties the world might have determining them: “there need be no explanation of the fact that properties are instantiated or not by certain kinds of things”. But this doesn’t go to the heart of the matter. From a deflationary perspective on propositions, they are only said to have intentional properties (to represent and to be about things) in a derived sense like the metonymical sense in which lion-statues are said to be fierce: one extended from the core sense in which the mental states and linguistic items to which they are ascribed are intentional.¹³ Propositions in general might either be considered representational in the extended sense, or should be denied the label on a more strict understanding. King, Soames and Hanks grant that the truth-conditional properties of propositions are essential to them, but take them to be seriously representational in so far as they obtain their truth Camp’s argument for the claim that her title conveys explicitly assumes her decision to “interpret propositionality in terms of functional structure” (Camp, 2018, 21), something I wouldn’t grant. Even if the argument is valid given the assumption, for the reasons offered here we should take her points to reveal only vehicle-level differences.  Cf. Richard, 2013, 707– 708, 716, and Field, 2016.

Against Propositional Substantivism

119

conditions from the activities of rational beings, in the way that mental states and linguistic items do. This combination of views is not easy to make sense of, as several writers have pointed out.¹⁴ Over the years, King has refined the Tractarian account that he provided in his 2007 book, on which the proposition expressed by “Rebecca swims” is a particular kind of fact, of course not the fact that would obtain if it was true: “that proposition is the fact of there being a context c and there being lexical items a and b in some language L such that a has as its semantic value in c Rebecca and occurs at the left terminal node of the sentential relation R that in L encodes the instantiation function and b occurs at R’s right terminal node and has as its semantic value in c the property of swimming” (King 2007, 51). The proposition being a fact of this sort is supposed to provide a substantive explanation for its representational properties. In a recent presentation (King, 2019), he discusses two objections. The first is that on this account propositions appear to be too fine-grained to properly implement their job description. For instance, the sentential relation R involved in the propositional fact that King posits will surely differ in an English sentence like “Sophie loves Carl”, and in its translation into an SOV language like Turkish, or an ergative language like Basque or Geogian. Now, in coming up with propositional facts like the one given for illustration above, King assumes that, say, particular lexical items are existentially quantified away. His proposal now has it that the specific syntactic relation that encodes ascription of the semantic value of the predicate to that of the arguments is also existentially generalized away in the facts constituting propositions. A similar solution might be provided for a second problem that King confronts, for which I find his own proposal wanting. King has speakers interpreting the syntactic relation R for it to encode ascription. This creates a regress: interpreting appears to be a propositional attitude, requiring propositions with their own unity and representational features to be already around.¹⁵ King’s own response is that the interpreting he posits is a subpersonal affair. But this doesn’t help. The subpersonal states posited by cognitive scientists have propositional contents, which play an essential role in their performing their theoretical jobs. Thus, scientists posit inferential subpersonal processes, sensitive to the

 Cf. Pickel, 2015, for a clear elaboration of these concerns, the most obvious ones of a modal nature. See also Speaks’s “Representational Entities and Representational Facts” in King et al., 2014; Caplan et al., 2013, and Caplan, 2016.  As Pickel, 2015, shows, this might also create a more straightforwardly Bradley-style regress. Once more, as Pickel points out, these difficulties in the King-Soames-Hanks program are evidence for a deflationary view.

120

Manuel García-Carpintero

validity-making roles of the contents of the states involved. These contents appear to generate the same worries that exercise King, but they cannot be solved by appeal to syntax-involving facts. Similar issues arise at the personal level, on the standard view that perceptual experiences have contents which account for their rationalizing role vis-à-vis the beliefs they might give rise to.¹⁶ We may instead extend King’s solution to the first problem to the contents of subpersonal states and perceptual experiences. We just need to assume that, as suggested above, these states also have their own, perhaps iconic, ‘map-like’ quasi-syntactic relations, equally capable of encoding the ascription relation. Propositions would then be highly abstract facts like those King ends up propounding, in which specific linguistic encoding relations have been existentially abstracted away; the same propositional facts of this abstract sort could now be expressed by experiential states, subpersonal states, and linguistic states.¹⁷ This solution, however, raises another problem. The rehearsed difficulties show that King’s original proposal falls prey to its explanatory ambitions, by making it difficult for propositions to play their defining jobs. The suggested solution, unfortunately, has the opposite problem: now the proposal fails by far to meet its explanatory ambitions. One wonders, how does it improve on what deflationists tell us about the unity issues? How does it solve the representation problem in the substantive way that King hoped for? By “generalizing away” components of the relevant fact, we just get that, say, the subject-value and the predicate-value in the proposition expressed by a subject-predicate sentence are somehow or other unified in the proper way, so that the proposition has the truth-condition it does. The abstract facts that the revised King proposal gives us are indistinguishable from, say, Liebesman’s ascriptions. They appear to be just objective “predications of fitting predicables of topics” – as Johnston (2006, 684) puts it –, to be regarded “as existing anyway, and not as just the outcome of judgment” so that “when it comes to predicability, all possibilities are realized;

 Some philosophers of perception dispute that experiences have content; but it would be a cost for King to have to rely on their being right, and in any case he grants in a quotation above that experiences have contents.  This seems to be what King suggests for perceptual experiences, cf. King et al., 2014, 60, 191– 195; Collins 2018, 3543, makes a similar proposal. I should note that explaining why experiences justify beliefs is not so straightforward even on the deflationary picture: inferring a proposition from that very same proposition doesn’t have the justificatory significance that perceptual experiences have for beliefs (Glüer, 2016). I think that these concerns can be addressed by a proper elaboration of the idea that experiences and corresponding perceptual judgments differ in force, even though they need not differ in content (Pryor, 2000, 547, fn. 37).

Against Propositional Substantivism

121

predicability guarantees predication” (ibid.).¹⁸ If so, how are the representational powers of propositions still explained by minds or languages?

4 Hank’s Cancellation Predicament I move now to present a structurally similar objection to the alternative acttheoretic account of propositions that Peter Hanks has provided and motivated along King’s lines. Like Soames, Hanks (2015) has developed a proposal to deal with the representational problem that takes propositions to be abstract types whose tokens – from which they are claimed to derive their representational powers – are intentional acts of rational beings. They both call the main such cognitive act predication, but they understand it in different ways. For Soames, it is a very general act, entertaining, lacking the specific illocutionary commitments of, say, asserting, or guessing. For Hanks they are instead assertorically committal. Here I will mostly discuss Hanks’ view, although I think that Soames’ raises parallel issues. Hanks draws inspiration from the near-universal character in natural languages of the three moods (indicative, imperative, interrogative), and the fact that current semantics ascribes sentences in them three different kinds of meanings: assertions, directives, questions (Roberts 2018). He identifies assertoric propositions (predications) with what is indicated by the first mood. Conventionally, however, the indicative mood can be used to perform a great variety of acts, with different commitments: claiming, concluding, conceding, telling, swearing, conjecturing, assuming, and so on. Hanks declares (2019, 1400, fn.) that he wants to put aside “hypothetical predications”, counting them as cancelled (more on this presently). Let us hence assume that it is the commitment to truth that distinguishes predications from other act types. The problem with this is the one that led Frege to endorse what Hanks (2015, 9) calls the constitutive version of the force-content distinction, which he distinguishes from the taxonomic version. The latter is the idea that there is a meaning-component (a truth-conditional component, the proposition proper) common to questions, directives and assertions. The former is the stronger metaphysical view that there are such forceless sentence-level meanings, propositional contents that “are devoid of any feature that has to be characterized using

 As Collins points out, this “combinatorial” view of propositions is compatible with a view on which they are (generically) dependent on the nature of cognition (Collins, 2018, 3543); see below.

122

Manuel García-Carpintero

concepts of force. Concepts of force characterize our actions. Propositional contents are prior to these actions and do not depend on them for their natures or representational features”, ibid., 19). Now, Hanks rejects the force-content distinction in these two versions; however, as Collins (2018, 3538 – 3539) points out, the reasons he gives for rejecting the taxonomic distinction are not compelling. Hanks appeals to the point already granted, that current semantics distinguishes three sorts of meanings for indicatives, imperatives and interrogatives. But this doesn’t establish the taxonomic distinction. For current semantics also distinguishes meanings for NPs, meanings for VPs, and forceless meanings for combinations thereof, which might be common constituents of the full meanings of imperatives, interrogatives and indicatives. As Pagin (2019) notes, Hanks’s formalizations in fact also acknowledge them. Given that Hanks has no qualms with abstract entities per se, the appeal to current semantics legitimizes force-involving sentential meanings as much as their forceless common “parts”. The debate thus hinges on the constitutive version of the force-content distinction. As we have seen, Hanks frames it in terms of priority; at times it just sounds as if he was speaking of temporal priority: “Suppose Obama enters a room, sees Clinton sitting in a chair and judges that Clinton is sitting. There is no neutral act of entertainment that precedes this judgment, whether conscious or unconscious. He just spontaneously makes the judgment and in so doing predicates the property of sitting of Clinton. He does not first contemplate the possibility that she is sitting and then decide to judge it to be the case. Nothing like that has to happen consciously or unconsciously (ibid., 19); “Asking whether Clinton is eloquent is not to predicate eloquence of Clinton and then ask whether this predication is true” (ibid., 24). I take the temporal intimations I have emphasized to be misleading, irrelevant for the real concerns. What matters, I assume, is explanatory priority.¹⁹ Platonism itself is not at stake either; for, as we have seen, both Hanks and Soames assume without reservations propositional types in their ontologies. What is then the disagreement? Hanks opposes the forceless propositions of Johnston’s (2006) or Liebesman’s (2015) views. However, as Collins (2018, § 7) points out, such views do not assert the explanatory priority of forceless propositions over their force-endowed counterparts; they merely reject the priority of the latter that the act-theorists endorses. It is at least consistent with the

 I don’t mean ‘explanation’ in an epistemological sense, but in the objective sense of ‘because’ common in discussions of essences and ontological dependence (cf. Correia, 2008; Rosen, 2010).

Against Propositional Substantivism

123

deflationary attitude towards unity to grant that (in themselves forceless) propositions somehow depend on linguistic or mental acts that express them. I suggest that we state the debate by means of the distinction between generic and specific dependence (Mulligan & Smith (1986), 118); Fine (1995), 288): x specifically depends on y just in case, necessarily given the nature of x, if x exists, y exists; x generically depends on the Fs just in case, necessarily in virtue of the nature of x, if x exists, there are Fs. To illustrate: perceptions specifically depend on perceptual experiences: no perception can exist without a particular perceptual experience constituting it. But even if dependence relations are asymmetric, this is still compatible with perceptual experiences generically depending in their turn on perceptions. This may obtain if it is in the nature of perceptual experiences that for them to have the contents they do, there must be some that constitute perceptions, fixing such contents. Similarly, in previous work (García-Carpintero, 2010, § 3) I have explained in these terms how the Principle of Compositionality and Frege’s Context Principle can state the mutual ontological dependence of the meanings of sentences and that of lexical items in them, consistent with the asymmetry of dependence relations. Compositionality states that the meaning of sentences specifically depends on that of their lexical constituents; the Context Principle, that the meaning of lexical items generically depends on that of some or other sentences in which they occur. The latter can be true if, for lexical items to get their meanings, they must occur as parts of meaningful sentences by means of which representational acts like assertions are performed. Given his identification of propositions with acts, the priority that Hanks asserts in the previous quotes must be the specific dependence of propositions on acts/forces:²⁰ (SD) Each proposition is constituted by (and hence specifically dependent on) a particular force, hence a particular representational act

 Hanks, 2015, 3 – 4, appears to say as much (my emphasis): “Propositions get their truth conditions from particular acts of judgment and assertion which are themselves the original or primary bearers of truth and falsity. The source of truth conditions is to be found in the acts of representation we perform when we make judgments and assertions, not in the propositional contents we use to classify and individuate these actions.” He also tells us that “[p]ropositions are types of actions, and propositional attitude relations are tokening relations”, ibid., 7; thus, it is token acts of representation that ground truth-conditions. Cf. also Hanks, 2019, 1386: “The representational features of these acts are not borrowed from pre-existing propositions. They are generated in the performance of the acts themselves”.

124

Manuel García-Carpintero

Rejecting (SD) is compatible with the view that there is a generic dependence of contents on acts – the claim that, given their nature, propositions cannot exist unless there are representational acts. This can be defended, say, by assuming a deflationary take on propositions as equivalence classes of vehicles, together with the relation outlined above between the Context and Compositionality principles: propositions are ascribed compositionally to propositional vehicles, but for their sentential constituents to get meaning there must be representational acts made with sentences to which they contribute.²¹ Be that as it may, (SD) should be rejected, I think: propositions may exist and have their truth-conditions without being the contents of any act. It is the FregeGeach point that supports its rejection, by showing the need of constitutively forceless contents. It is usually discussed in connection with modus ponens. This is a valid form of inference, and explaining it is thus one of the items in the job-description for propositions. In the minor premise, a content is asserted. In the major premise, this content appears to be merely the unasserted antecedent of a conditional. For such arguments to be valid – as opposed to equivocation fallacies – the same entity should occur first forceless and then endowed with force; that same entity is a traditional proposition. Let us consider a particular case.²² Dummett (1982, 341– 342) helpfully distinguished between conditional bets, and bets with conditional contents. One wins a bet with a conditional content (say, that Peter will come to the party, if Jane comes) if the condition does not obtain, but one neither wins nor loses a conditional bet in that case. Conditional bets detach: if the condition obtains, an unconditional bet is in place. The Frege-Geach point is then that the validity of any such detachment requires that the assertion that the condition obtains be the assertion of the very same thing that was the condition for the bet or the obligation. A view of propositions that rejects SD can deal with this. When one makes a conditional bet, the condition one is thereby putting forward is entirely forceless; it is just the ascription of a property (coming to the party) to a topic (Jane), on the Burge-Johnston-Liebesman understanding of this. Hanks deals with the Frege-Geach problem by appealing to cancellation contexts. The idea is that, when they occur as the condition in our example, propositions still retain their assertoric force (given that they clearly do not have ero-

 Cf. Sainsbury’s discussion of cardinality worries about deflationary views (Sainsbury 2018, 55 – 59).  The problem also arises in many other cases: disjunction, negation (Hom / Schwartz, 2013, 19), the prejacent conditions (sometimes merely contextually implicit, Dowell, 2012) in modals and the open sentences required to account for generality (Collins, 2018, § 6). The case I present is helpful to challenge also Soames’ view, but I cannot develop this point here.

Against Propositional Substantivism

125

tetic or directive force), so that a modus ponens argument like the one envisaged above does not come out an equivocation fallacy. It is just that the assertoric force in such a context is cancelled, so that “the cancelled act of predication does not count as an assertion and does not carry the usual consequences and normative requirements” (Hanks, 2015, 32). Prima facie, it is difficult to make out how this fits consistently together. How could an act of predication, understood as having assertoric force, be part of the condition that Jane comes to the party, without the act thereby having assertoric normative commitments? Hanks (2019) aims to address a concern that Jespersen (2012), Hom & Schwartz (2013), and Reiland (2013) pose, in the form of a dilemma. First horn: cancellation removes the assertoric force. This is bad for Hanks, because the resulting condition still has truth-conditions, in spite of lacking assertoric force (and, by assumption, either erotetic or directive force). Second horn: the remaining act of predication is just a part of the original one, without another component of assertoric commitment. This is also bad, because it appears to collapse into Soames’s view, or the forceless-proposition view; the remaining ‘part’ is undistinguishable from Soames’s acts of entertaining, or is just a plainly forceless content. Hanks (2019) contends in response that cancellation is not a further propositional act, but a (blind) context. He illustrates what he means with the case of actors on stage, or (I assume) fiction-making in general. Presumably Hanks does not want cancellation to result from a further propositional act, because this would have the potential to initiate a vicious regress. This is also why ‘context’ should not be understood in his proposal as in contemporary semantics, namely, as a set of attitudes presumed to be shared. But this creates serious problems, for it seems to follow that, as Hanks admits, “there is nothing the actor can do to make his utterances count as genuine assertions”, aside from taking “himself out of the theatrical context” (Hanks 2015, 93 – 4). However, many philosophers currently writing on fiction endorse a view that ordinary people share, to wit, that fictions make assertoric acts, and may hence be criticized as false or praised as true. Many of these asserted contents are also inseparable parts of their fictional contents.²³ Consider this compelling example from Stock (2017, 24): “Nonhuman animals have gone to court before. Arguably, the first ALF action in the United States was the release of two dolphins in 1977 from the University of Hawaii. The men responsible were charged with grand theft. Their original defense, that dolphins are persons (humans in dolphin suits, one defendant said), was quickly thrown

 Cf. García-Carpintero, 2018, §2, and references there.

126

Manuel García-Carpintero

out by the judge”, in K. J. Fowler, We Are All Completely Beside Ourselves. This proposition is an essential part of the content the fiction-maker is putting forward for us to imagine; Fowler is not “taking herself out” of the fictional context in presenting it. However, given the theme of the novel and the moral seriousness with which she pursues it, together with relevant conventions for the genre, it is natural to take it also as a straightforward assertion, capable of transmitting testimonial knowledge; critics would rightfully object to the novel if the proposition was false. As said, this is correct according to compelling accounts, like Stock’s: we are entitled to acquire knowledge by testimony from reading this piece of fiction. Hanks is committing himself to rejecting these views, without properly engaging with them.²⁴ Hanks (2019) also relies on an analogy with games (football here, chess earlier, cf. Hanks, 2015, 32, 94), but they are not more helpful. Given the constitutive norms of a game, a non-normative act may ‘count as’ a move in the game with normative consequences. There are contexts in which the act does not count as a move in the game, and is thus not beholden to the norms – for instance, when one is showing a learner how chess pieces move. To the extent that I understand the analogy, Hanks assimilates cancelled predication to the non-normative act as taking place in such contexts, deprived of its normative features in the game: “The act of tackling is analogous to the act of predication, and scoring a safety is analogous to performing an assertion” (ibid., 1391). But the analogy doesn’t really work. What is analogous in the linguistic case to the physical act in games (tackling) is the emission of the physical sounds or inscriptions, made (say) just for the sake of practicing elocution, or accidentally produced by a non-intentional agent. But in such cases not just the assertoric force is absent; the representational component is gone as well. It doesn’t help the cancellation account that there are physical events of a kind such that, instantiated under specific circumstances, do not have representational content, with or without force. What Hanks needs to explain by invoking cancellation is the prima

 Cp.: “there is nothing the actor can do to make his lines in the play count as full-fledged assertions. No act of subscription or identification or endorsement can convert his stage utterances into real assertions. The only way for his utterances to count as genuine assertions is for the theatrical conventions to be lifted and the play to end. This drives home the fact that these conventions create a context that overrides the assertive character of the actor’s utterances”, Hanks, 2015, 96, fn. There of course are theatrical examples entirely analogous to the one just offered – moreover, as Ohmann points out, the author of a literary fiction can be seen as an actor impersonating the explicit or implicit teller of the story telling it (Ohmann, 1971, 18). Recanati is also unhappy with this aspect of Hanks’s views, which he otherwise mostly agrees with (Recanati, 2019, 1414).

Against Propositional Substantivism

127

facie appearance that there are events or acts that have truth-conditions, without having assertoric (nor directive or erotetic) force. The games examples are not relevantly analogous.

5 Conclusion In this paper I have endorsed a deflationary view of propositions on which they are properties of “circumstances of evaluation”. Deflationary views predict that substantive accounts like those by Hanks, King, and Soames will create spurious problems. I have shown that this prediction is borne out by the facts, by discussing the problems of fineness of grain for King’s account, and the Frege-Geach problem for Hank’s act-of-predication account. I have suggested that the unity problems about propositions that these accounts aim to solve by ascribing them robust natures should be dismissed, taking the unity facts as primitive at that level. Substantive theories of the representational vehicles to which propositions are ascribed should handle remaining concerns.

References Ball, Derek. Forthcoming: “Semantics as Measurement.” In The Science of Meaning, edited by Brian Rabern and Derek Ball. Oxford: Oxford University Press. Bealer, George. 1998: “Propositions.” Mind 107.425:1 – 32. Buchanan, Ray and Alex Grzankowski. 2019: “Propositions on the Cheap.” Journal Philosophical Studies 176.12:3159 – 3178. Burge, Tyler. 2007: “Predication and Truth.” Journal of Philosophy 104:580 – 608. Camp, Elizabeth. 2018: “Why Maps are Not Propositional.” In Non-Propositional Intentionality, edited by A. Grzankowski & M. Montague, 19 – 45. Oxford: Oxford University Press. Caplan, Ben. 2016: “Soames’ New Conception of Propositions.” Philosophical Studies 173:2533 – 2549. Caplan, B., Tillman, C., McLean, B., and A. Murray. 2013: “Not the Optimistic Type.” Canadian Journal of Philosophy 43.5 – 6:575 – 589. Collins, John. 2014: “Cutting It (Too) Fine.” Philosophical Studies 169.2:143 – 172. Collins, John. 2018: “The Redundancy of the Act.” Synthese 195:3519 – 3545. Correia, Fabrice. 2008: “Ontological Dependence.” Philosophy Compass 3/5:1013 – 1032. Dowell, J. L. 2012: “Contextualist Solutions to Three Puzzles about Practical Conditionals.” Oxford Studies in Metaethics 7:271 – 303. Dummett, Michael. 1981: Frege: Philosophy of Language, 2nd edition. London: Duckworth. Duncan, Matt. 2018: “Propositions are not Simple.” Philosophy and Phenomenological dcwrcResearch 97.2:351 – 366. Davidson, Donald. 2005: Truth and Predication. Cambridge, Mass.: Harvard U.P.

128

Manuel García-Carpintero

Eklund, Matti. 2019: “Regress, unity, facts, and propositions.” Synthese 196:1225 – 1247. [29/03/2021]. Field, Hartry. 1992: “Critical Notice of Horwich’s Truth.” Philosophy of Science 59:321 – 330. Field, Hartry. 2016: “Egocentric Content.” Noûs 51.3:521 – 546. Fine, Kit. 1995: “Ontological Dependence.” Proceedings of the Aristotelian Society 95:269 – 290. Fine, Kit. 2017: “Truthmaker Semantics.” In A Companion to the Philosophy of Language, edited by B. Hale, C. Wright and A. Miller, 2nd ed., 556 – 577. Oxford: Blackwell. García-Carpintero, Manuel. 2007: “Fiction-making as an Illocutionary Act.” Journal of Aesthetics and Art Criticism 65:203 – 216. García-Carpintero, Manuel. 2010: “Gaskin’s Ideal Unity.” Dialectica 64.2:279 – 288. García-Carpintero, Manuel. 2018: “Assertion and Fiction.” In Oxford Handbook of Assertion, edited by S. Goldberg. Oxford: Oxford University Press. [29/03/2021]. Gaskin, Richard. 2008: The Unity of the Proposition. Oxford: Oxford University Press. Giardino, V., and G. Greenberg. 2015: “Varieties of Iconicity.” Review of Philosophy and Psychology 6:1 – 25. Glick, Ephraim. 2017: “What Is A Singular Proposition?” Mind 127.508:1027 – 1067. [29/03/2021]. Glü er, Kathrin. 2016: “Defeating looks.” Synthese 195:2985 – 3012. [29. 03. 2021]. Green, Mitchell. 1999: “Attitude Ascription’s Affinity to Measurement.” International Journal of Philosophical Studies 7:323 – 348. Hanks, Peter. 2015: Propositional Content. Oxford: Oxford University Press. Hanks, Peter. 2019: “On cancellation.” Synthese 196:1385 – 1402. [29. 03. 2021]. Hare, R. M. 1970: “Meaning and speech acts.” Philosophical Review 79:3 – 24. Hom, C., and J. Schwartz. 2013: “Unity and the Frege–Geach problem.” Philosophical Studies 163:15 – 24. Jespersen, Bjørn. 2012: “Recent Work on Structured Meaning and Propositional Unity.” Philosophy Compass 7:620 – 630. Johnston, Mark. 2006: “Hylomorphism.” Journal of Philosophy 103:652 – 698. Keller, Lorraine. 2013: “The Metaphysics of Propositional Constituency.” Canadian Journal of Philosophy 43.5 – 6:655 – 678. King, Jeffrey. 2007: The Nature and Structure of Content. Oxford: Oxford U.P. King, Jeffrey. 2009: “Questions of Unity.” Proceedings of the Aristotelian Society 109.3:257 – 277. King, Jeffrey. 2019: “On propositions and fineness of grain (again!).” Synthese 196: 1343 – 1367. [29. 03. 2021]. King, J., Soames, S., and J. Speaks. 2014: New Thinking about Propositions. Oxford: Oxford University Press. Lande, Kevin. 2018: “The Perspectival Character of Perceptione.” Journal of Philosophy 115.4:187 – 214. Lewis, David. 1983: “New Work for a Theory of Universals.” Australasian Journal of Philosophy 61:343 – 377.

Against Propositional Substantivism

129

Lewis, David. 1978: “Truth in Fiction.” American Philosophical Quarterly 15:37 – 46. Reprinted with postscripts in Lewis, D. 1983: Philosophical Papers, Vol. 1, 261 – 280. Oxford: Oxford University Press (from which I quote). Lewis, David. 1986: On the Plurality of Worlds. Oxford: Basil Blackwell. Lewis, David. 2001: “Truthmaking and Difference-Making.” Noûs 35:602 – 615. Liebesman, David. 2015: “Predication as Ascription.” Mind 124:517 – 569. Matravers, Derek. 1997: “The Paradox of Fiction.” In Emotion and the Arts, edited by M. Hjort and S. Laver. Oxford: Oxford University Press. Matthews, Robert J. 2007: The Measurement of Mind. Oxford: Oxford University Press. Matthews, Robert J. 2011: “Measurement-Theoretic Accounts ofd Propositional Attitudes.” Philosophy Compass 6:828 – 841. Merricks, Trenton. 2015: Propositions. Oxford: Oxford University Press. Moore, Joseph. 1999: “Propositions without Identity.” Noûs 33.1:1 – 29. Mulligan, K., and B. Smith. 1986: “A Relational Theory of the Act.” Topoi 5:115 – 130. Ohmann, Richard. 1971: “Speech Acts and the Definition of Literature.” Philosophy and Rhetoric 4.1:1 – 19. Ostertag, Gary. 2013: “Two aspects of propositional unity.” Canadian Journal of Philosophy 43.5 – 6:518 – 533. Pagin, Peter. 2019: “Review of Peter Hanks: Propositional Content”, Language 95.2: 377 – 380. [29. 03. 2021]. Pautz, Adam. 2016: “Propositions and Properties.” Philosophy and Phenomenological Research 93.2:478 – 486. Pickel, Bryan. 2015: “Are Propositions Essentially Representational?” Pacific Philosophical Quarterly 97.2:470 – 489 [29. 03. 2021]. Pickel, Bryan. 2017: “Structured Propositions in a Generative Grammar.” Mind 128.510:329 – 366. [29. 03. 2021]. Pryor, James. 2000: “The Skeptic and the Dogmatist”. Noûs 34:517 – 549. Recanati, François. 2013: “Content, Mood, and Force.” Philosophy Compass 8/7:622 – 632. Recanati, François. 2019: “Force cancellation.” Synthese 196:1403 – 1424. [29. 03. 2021]. Reiland, Indrek. 2013: “Propositional Attitudes and Mental Acts.” Thought 1:239 – 241. Richard, Mark. 2013: “What are Propositions?” Canadian Journal of Philosophy, 43.5 – 6:702 – 719. Roberts, Craige. 2018: “Speech Acts in Discourse Contexts.” In New Work in Speech Acts, edited by D. Fogal, D. Harris and M. Moss, 317 – 359. Oxford: Oxford University Press. Rosen, Gideon. 2010: “Metaphysical Dependence: Grounding and Reduction.” In Modality: Metaphysics, Logic, and Epistemology, edited by Robert Hale and Aviv Hoffmann, 109 – 136. Oxford: Oxford University Press. Sainsbury, Mark. 2018: Thinking about Things. Oxford: Oxford University Press. Schnieder, Benjamin. 2010: “Propositions United.” Dialectica 64.2:289 – 301. Smith, B., and K. Mulligan. 1982: “Pieces of a Theory.” In Parts and Moments, edited by B. Smith, 15 – 109. Munich: Philosophia Verlag. Soames, Scott. 2010: What Is Meaning? Princeton: Princeton University Press. Stalnaker, Robert. 1976: “Possible Worlds.” Nôus 10.1:65 – 75. Stalnaker, Robert. 2009: “What is De Re Belief.” In The Philosophy of David Kaplan, edited by J. Almog and P. Leonardi, 233 – 245. Oxford: Oxford University Press.

130

Manuel García-Carpintero

Stock, Kathleen. 2017: “Fiction, Testimony, Belief and History.” In Art and Belief, edited by E. Sullivan-Bisset, H. Bradley and P. Noordhof, 19 – 41. Oxford: Oxford University Press. Thomasson, Amie L. 2015: Ontology Made Easy. Oxford: Oxford University Press. Wood, Michael. 1998: The Magician’s Doubts. Princeton: Princeton University Press. Wright, Crispin. 1998: “Why Frege Did Not Deserve His Granum Salis: A Note on the Paradox of “The Concept Horse” and the Ascription of Bedeutungen to Predicates.” Grazer Philosophische Studien 55:239 – 263. Yablo, Stephen. 2014a: “Carnap’s Paradox and Easy Ontology.” Journal of Philosophy 111.9/10:470 – 501. Yablo, Stephen. 2014b: Aboutness. Princeton: Princeton University Press.

Jonathan Smith

Wittgenstein’s Nachlass, a Case for Practical Ontology? Abstract: The aim of this chapter is to discuss the question of whether Wittgenstein’s Nachlass is to be taken as a central source for the diffusion of his philosophy. The author gives several reasons why the Nachlass is a useful source of philosophical research and how its organization by the trustees helps use it as a source. The author comments on various methods of scientific work and discusses them in the context of the Nachlass. Keywords: Nachlass, practical ontology, record, relationships, Wittgenstein

1 Introduction What is Wittgenstein’s Nachlass? This apparently straightforward question is not one I see being discussed with any regularity. Yet it seems to me one which is fundamentally important to our understanding of Wittgenstein’s philosophy if we are to take the Nachlass as the central source for the dissemination of that philosophy, whether directly, or through the mediation of his literary executors or subsequent editors. In scholarly work, no source should be taken at face value. Rather, its evidentiality should be established and the purpose for which it was created and the process of which it forms a part should be investigated, in order to establish its evidential value and to extract the most information from it. Such an investigation is necessarily fluid and evolves over time. As individual instances inform our understanding of the whole source and our knowledge of the whole helps our understanding of individual items within it, the hermeneutic cycle gradually leads to a better appreciation of how we can best make use of it. This is not a paper on ontology. It is not even a paper which uses ontology as its underlying method. What I do hope to do is to is suggest certain questions that might be asked about the Nachlass that will elucidate our understanding of it and help in our use of it as a source. And these questions are fundamentally ontological. There are a number of general factors that make Wittgenstein’s Nachlass particularly interesting. The first is his method of working and his confession in the foreword of the Investigations, penned in 1945 and published in 1953, that he found it impossible to forge the remarks that he produced into continuous narrative text. The second is the struggle to arrange his many philosophical remarks https://doi.org/10.1515/9783110750041-008

132

Jonathan Smith

into an order that he would be happy to see published. Consequent on these is the third factor: that the fruits of his later philosophy were not published during his lifetime, but were entrusted to literary executors, who produced a long series of publications based on material in varying states of completion and supported by little or no editorial apparatus. A fourth is that between Wittgenstein’s death in 1951 and 1969 the manuscripts and typescripts were shared out among the editors and occasionally assistants, while photographic and typescript copies were made of the more important items. All of these factors reinforce the imperative of seeking a greater understanding of the nature of the Nachlass. One answer to the question that begins this piece might be that the Nachlass is what von Wright tells us that it is. There is no doubt that his article ‘The Wittgenstein Papers’, first published in 1969¹, and reissued as further materials came to light, had an influence on the way we perceive the Nachlass. In this paper von Wright defined the essence of the Nachlass as the working papers Wittgenstein produced in ‘doing philosophy’², though he did not define their extent, leaving an opening for the discovery of further materials. He also divided the items that make up the Nachlass into 3 classes, namely manuscripts, typescripts and dictations, though he does take pains in his introduction to hint at greater complexity³. In effect von Wright’s catalogue is a taxonomy, which is almost certainly exactly all he intended it to be. Any description of the Nachlass with ontological pretentions requires a deeper investigation of its various properties. However, before I move on I might ask the question whether there is a danger that von Wright’s catalogue unwittingly presents a simplististic view of what it is to be Wittgenstein’s Nachlass that might affect our understanding of it both as a source for editorial work and of our understanding of the development of Wittgenstein’s philosophy. Indeed we might ask the question whether we should beware of being, as Wittgenstein may have said, bewitched by the catalogue.

 Wright, 1969, 483 – 503.  Von Wright does suggest that the catalogue would later extend to items such as lecture notes and correspondence, but subsequent revisions of the catalogue persist with the original three classes only.  ‘But then one must remember that the distinction between the strata is by no means clear-cut, and that there is no one-to-one correlation between them. Some of the more-finished writings are revisions of earlier material of draft-like character, but others are revisions of material itself classified here as “more finished”. Some of the most finished manuscripts have the nature of “fair copies” of remarks which have been extracted from earlier writings. Others definitely have the character of manuscripts for a planned book.’ Von Wright, 1969, op. cit.

Wittgenstein’s Nachlass, a Case for Practical Ontology?

133

2 The Nachlass as Record So what then is the Nachlass? Before I attempt to answer that question it would be prudent to remember Wittgenstein’s own example of aspect-seeing and recognize that my view of the Nachlass, influenced by my training as an archivist and a lifetime of work with the archives of academics, is one of many possible legitimate views, each forming part of a complex whole. It is, however, a view that I believe has some merit and brings to light a number of considerations that inform our use of the Nachlass. My view of the Nachlass is that it is a record of Wittgenstein’s cognitive process when writing. I believe that there are a number of advantages of viewing the Nachlass this way. First, that it is a record (though an imperfect one) of the attempt by Wittgenstein to bring into being a work or works of philosophy, with all the evidential authority that this entails. Second, this authority supports that authorial intention embedded in the Nachlass. Third it acknowledges the dynamic nature of the authorial process. While of course every author develops different processes in the production of their work, there have been some useful attempts to describe the general aspects of the writing process which underlie their individual methods. The model I want to mention briefly is a fairly general one developed by John R Hayes and colleagues over some 40 years⁴. While it has been used mostly in the area of writing instruction and development, I believe that its analysis of the process of writing helps inform our understanding of the archives produced by that process such as the Nachlass. In brief, the current version of the Hayes model consists of three levels. The Control Level, consists of motivation; goal setting, which as Hayes and Olinghouse point out, has ‘a profound effect on shaping the text that is written’; the plan current at the time of writing, where ‘writers form plans that consist of subgoals for accomplishing the main goals’⁵; and writing schemas ‘which represent the writer’s knowledge of how to create a text’ and which include strategical thinking and an understanding of the properties that a particular text requires and are very personal to the individual. Many of these aspects of the writing process are either addressed directly or hinted at in the Preface to the Philosophical Investigations.

 A discussion of the development of Hayes thinking on the modelling of the writing process can be found in Hayes, 2012, 369 – 388.  Hayes / Olinghouse, 2015, 482.

134

Jonathan Smith

The Process Level includes a number of interlinked sub-processes. The Proposer produces ideas triggered by many factors for inclusion in the text; the Translator takes these ideas and converts them to verbal form; the Transcriber takes the output of the Translator, applies grammatical rules and turns them into written text; the Evaluator examines and evaluates the output of any of the other sub-processes. As a result of the Evaluator’s conclusions, revision may take place using the sub-processes within the Process Level: ‘the processes involved in revision are evaluating (detecting the problem), proposing ideas to fix the problems, translating those ideas into language, and transcribing that language to replace the problematic text’⁶. The third level, the Resource Level, acknowledges Reading and Attention as well as Long Term Memory, where various resources are stored to be retrieved for future use, and Working Memory, a short-term memory resource used to carry out tasks required by the writing process. Crucial to the production of an archive such as the Nachlass is the fact that Working Memory is limited. Studies suggest that most people, even experienced writers, are only able to hold in working memory a surprisingly small number words of a composition before a record of this has to be made, in order to allow it to be cleared to make space for composition of the next phrase or sentence. In this manner, piece-by-piece a work is composed, and the record created. So the record left by Wittgenstein in ’doing philosophy’, while sui generis, follows a pattern dictated by the cognitive process of writing which we can see being played out in the drafting and constant rewriting and rearranging in his Nachlass. One important purpose of this definition of the Nachlass is that it allows us to see it as interrelated items that form an organic whole that developed naturally over many years as part of a process to which it was integral. As such it stands in contrast to a collection of disparate texts gathered together synthetically. This view is important in establishing provenance, the abstract attribute defining its origins, the hands that it passed through and archival history. The importance of provenance for users of the Nachlass as in all archives is its role in establishing authenticity and therefore evidentiality. Defining the class of documents produced by Wittgenstein in the practice of his philosophical work is vital in establishing the strength of the evidence we produce when we use information from those documents in debate. Another is that the model proposed by Hayes and his collaborators allow for the fact that the subprocesses responsible for the genesis of ideas are not record-

 Hayes / Olinghouse, 2015, 480 – 497.

Wittgenstein’s Nachlass, a Case for Practical Ontology?

135

ed. It is only once the author begins to form those ideas into language that recording becomes necessary. And it only becomes necessary because the language produced is intended to produce written text not contemporaneous/instant speech. This fact itself is extremely important in our understanding of the Nachlass and helps us understand that while the Nachlass tells us THAT Wittgenstein made a change or changes to a developing text, it does not tell us WHY he did it. The answer to that has to be approached by other means.

3 Amendments Many of the individual items in the Nachlass are annotated in some way, indeed a crucial part of what makes it such an interesting source are the very many amendments that Wittgenstein makes to his work. These take the form of deletions and replacements of words or words; insertions into existing text, in some cases of a word or two, in other cases a number of paragraphs; and relocation of blocks of text, sometimes the size of several remarks. Usually, where amendments occur all of these types are represented, and in the case of some documents multiple layers of amendments are created as repeated changes are made at different times. Understanding the nature of these various amendments might help us to make decisions in editing Wittgenstein’s writings or to better understand his developing philosophy. To these ends we might look closer at them and ask: What it is to be a change of word or words, what it is to be an insertion, what it is to be the move of a paragraph or section elsewhere. Editors in particular need to consider what are the attributes of these amendments that lead us to bring them together into categories and what attributes allow us to differentiate them, to consider whether one class of amendment is more important than another or whether other considerations, such as the time when an amendment was made is more important than any. For their decisions on these matters are invested with great significance as the decisions that editors make over the various alternatives before them are incorporated in the works by which Wittgenstein’s philosophy is transmitted. Let me now give an example of why I think this is important. Let us look at the German/English edition of the Big Typesctipt edited by Luckhardt and Aue⁷. It is certainly an impressive volume with a parallel text and numerous footnotes on each page. The text, we are told, is an edition the typescript von Wright

 Wittgenstein, The Big Typescript, TS. 213, 2005.

136

Jonathan Smith

named in his catalogue ‘The Big Typescript’ which Wittgenstein produced, probably in the Summer of 1933, with the aid of a typist who supplied a top-copy and at least one carbon, which Wittgenstein began annotating shortly after the typescript was produced. When we compare the edition to the original on which it is based we can see that the amendments Wittgenstein made by altering words in the text are included in the main text – the text we are meant to read – with the words that these alterations replace being relegated to the footnotes. Looking further, however, we see that the examples of another category of alteration that Wittgenstein included in the annotated version of the Big Typescript, that which involved moving sections of text within the manuscript, are not enacted by the editors, though notes that Wittgenstein indicated moving them are included in the footnotes. Given the fact that these alterations appear to have been made together with those of phrasing which ARE enacted in the main text, we might ask the question whether the editors may have presented us with a bastard text and certainly not a representation of Wittgenstein’s intention at any point in time? And not the one that Anthony Kenny was expecting (or so I would hope). The editors do not tell us their reasons for treating these two types of amendment by Wittgenstein in different ways. Fundamentally, however, to make any sense (and I am not saying that it does) the distinction had to be an ontological one. Both types are amendments by Wittgenstein and there is no indication that the majority of each type are made at a different time from the rest. So the differentiation seems to be made based on an attribute that distinguishes one sort of amendment from the other – one is an amendment to existing text, the other is an amendment that moves a section of text elsewhere within the document. But by privileging the attribute of type over the temporal attribute is I think a straightforward category error and one which results in scholars being presented with a text that Wittgenstein would not have recognised.

4 Relationships If the many amendments embedded in each Nachlass item can be seen as concrete attributes of those items, other attributes are rather more abstract, but nonetheless extremely important. One of the most important ontological aspects for the understanding of the Nachlass is the nature of relationships between items within it as these help us understand the progress Wittgenstein was making in developing his thought. Rather than try to group the items that constitute the Nachlass into classes by generalized ‘type’ based on qualities, as in the von Wright catalogue, a more realistic ontology would acknowledge and illuminate

Wittgenstein’s Nachlass, a Case for Practical Ontology?

137

Wittgenstein’s process by recognizing the context of each item and the necessary relationship between items within that process. The fact that manuscripts 105 to 122 form a series of volumes running chronologically from 1929 to 1940 is important, especially given the fact that their intended purpose seems to have been as a repository for those remarks chosen and refined by Wittgenstein, many of which were to be transferred to typescripts at a later stage of development. The fact that the remarks in these volumes are not completely sequentially chronological, but at times jump between volumes, usually where Wittgenstein has travelled from Cambridge to Vienna is also worthy of note. However, perhaps more usefully noteworthy are the relationships between documents that show a development of Wittgenstein’s ideas and how he intended them to be disseminated. The knowledge that typescript 211 is based on remarks from manuscripts 109 to 114, which themselves drew on draft remarks from manuscripts 153 to 155, helps us contextualize his developing ideas, as does the fact that typescript 211 contributes along with 208 and 210 in the production of the Big Typescript (213i) which Wittgenstein heavily annotates (213ii) and tries to further edit in manuscripts 114 and 115 and the Grosses Format (140). But equally, it is important to be aware where there are no close relationships between documents, other than them sharing provenance and being part of the Nachlass. One of the reasons that the literary executors were unwilling to make the whole Nachlass available until a comparatively late date is that they were fearful that academics might interpret it incorrectly. One of the ways in which this can be done is by ‘cherry-picking’ quotations from the Nachlass, possibly to suit a pre-formed opinion of Wittgenstein’s philosophy, without reference to their context, which might reveal a very different interpretation of it, or may provide a subtly altered version of the quotation, or may even show that Wittgenstein discarded it at some point. In order to get this wide perspective of which manuscripts are related to each other by the part they play in Wittgenstein’s writing process, we need to be able to survey the whole Nachlass, not to restrict it as has happened in the past. This wide perspective is also important when we consider that the evidence which the Nachlass gives us goes only so far in that it is a record of what Wittgenstein did when developing his writings not why he did it. We can, however, use the data that we extract from the Nachlass and, by a process of induction, add to our understanding of it and through it of our understanding of Wittgenstein’s work. Take for example the slash and long-s-like marks Wittgenstein used in the margins of many of his manuscripts which indicates whether he wanted to use them in later manuscripts or to reject them. Though he nowhere gives us a key to explain what these symbols mean, we can work out with reasonable certainty their significance using inference. However, this is impossible with access

138

Jonathan Smith

to only one manuscript. To be sure that it is a general aspect of Wittgenstein’s process and to be certain of its meaning it is necessary to compare many manuscripts and to be aware of the relationships between them.

5 Conclusion I stated in my introduction that this was not explicitly an ontological paper, but it is one that formulates and poses questions impacting ontology. It is a plea for greater awareness of and research into the identity of the Nachlass and its constituent parts which, it is important to stress, are not academic exercises for their own sake, but contribute to a greater understanding of Wittgenstein’s creative work. Any investigation into the Nachlass requires scholarly practices of the highest quality to use it to its best effect. In setting out on this investigation we might begin by reflecting deeply on the question ‘What is Wittgenstein’s Nachlass?’

References Hayes, John R. 2012: “Modeling and Rewriting.” Written Communication 29.3:369 – 388. Hayes, John R., and Natalie G. Olinghouse. 2015: “Can cognitive writing models inform the design of the common core state standards.” The Elementary School Journal 115/4:480 – 497. Wittgenstein, Ludwig. 2000: Wittgenstein’s Nachlass: The Bergen Electronic Edition, ed. by The Wittgenstein Archives at the University of Bergen. Oxford: Oxford University Press. Wittgenstein, Ludwig. 2005: The Big Typescript, TS. 213, edited and trans. by C. Grant Luckhardt and Maximilian Aue. Oxford: Wiley. Wright, G.H. von. 1969: “The Wittgenstein Papers.” Philosophical Review 78.4:48 – 503.

Arthur Gibson

Paradox in Ontology: Black Holes, Cosmology, Wittgenstein versus Stephen Hawking’s Claim that Philosophy is Dead Abstract: Stephen Hawking’s and the author’s research interaction together facilitate a backcloth for this engagement between philosophy, reason, observational and mathematical cosmology, which addresses some relations between science and other subjects, so as to draw some original conclusions. Keywords: cosmology, information paradox logic, mathematics, metaphysics, ontology, physics, theism, tightrope walking

1 Introduction In 2013, with typical aplomb and mischief, Stephen Hawking gave his pronouncement to the media that God is dead. Being in the same Maths Faculty, he soon sent the author a message challenging him to reply about the matter. His was not the first nor the last such meeting. This Chapter echoes a sense of some of the matters we disputed. Stephen Hawking claimed that philosophy is dead. The reason for his view is that the empirical grounds for proof in science are successfully offered and justified, which philosophy cannot satisfy. Accordingly, this appears to undermine philosophy as an empirical subject, at least if it is taken to stand in parallel with science as an exemplar of a particular type of justified knowledge. Certainly, Professor Hawking rightly presupposed that certain philosophy also has some baggage that is a dead weight in its claim to conceptual life, though let us leave that aside as a property of some of its fashions, which are

Acknowledgement: It was a tremendous pleasure to be with Professor Jesus Padilla Galvez and Prof. Margit Gaffal, along with their invited colleagues and guests, for is all to engage with them and colleagues. I thank them for being very adaptable to accommodate my pressured timetable. Appreciation and thanks to Jonathan Smith for reading, advising and checking the draft of this chapter. Grateful also to Frank Madsen and Kelsey Gibson for reading and checking the typescript. https://doi.org/10.1515/9783110750041-009

140

Arthur Gibson

not necessarily internal properties essential to its form of life. Even so, on Hawking’s account all philosophy seems dead when aligned with science.

2 Resurrection for Philosophy? My reaction to Stephen’s challenge has this kernel: the internal properties that he detected in philosophy are also present in some of his science, and other sciences.¹ Elsewhere I noted Berto and Plebani² argue that “the border between ontology and metaphysics in the works of contemporary philosophers is fuzzy”. This is also true of some science borders. The astrophysical scientist Dallaporta³ assigns a metaphysical status to empirical phenomena that are of a smaller scale, below the Planck constant, So, in that sense, a justified empirical domain of science in Hawking’s sphere is metaphysics. Its proof criteria are quite different from laboratory science. No doubt there will be dispute about what are the borderlands of how to stipulate the identity of such metaphysics. Nevertheless, it is easy to detect some domains in what is regarded as empirical science that exemplify the metaphysical status of some research areas. For example, dark energy is an explicit case in which there are masses of observations, theoretical conjectures coupled with a complete failure to formulate a scientifically coherent or adequate theory or framework that fits the standard model of physics and quantum mechanics. What is more, it is taken as evident that there is no current available theoretical or observational grounding to indicate what would be an empirically and conceptually adequate prospective route to a solution. The situation concerning dark energy gives further substance to concerns that the cosmological paradox generates support for the calculation that there are errors in current relations between mathematical and observational cosmology of approximately 120 orders of magnitude, which pertains to over 70 % of the universe in relation to the rate of its expansion. Notoriously, Stephen Hawking himself observed that this arena is witness to the most spectacular failure of a physical theory in the history of science and its current situation. This highlights a state of affairs that in applied science, and in mathematical cosmology, use models as analogies of terrestrial empirical sci Challenge was responded to in the Cambridge University Mathematics Faculty, Institute of Cosmology, Potter Room, 6 May 2013, as encapsulated in Plate 1.  Berto and Plebani, 2015, 3.  Cf. Gibson God and the Universe, p. X.

Paradox in Ontology

141

ence. Such conjectural fits, properly understood, occupy conceptual spaces that perforce advertise their at-least-partial metaphysical status. There are all manners of rich ad hoc recipes bolting on all such observations and theorization that are aimed at rendering such states of affairs in science as facilitating a fit with the standard model of physics, for example with dark energy modelling. It is worth warning ourselves of retrospective mission-creep that can infect not only the general reader, but also following a scientist’s or philosopher’s retrospective equivocation or conflation.

3 The Information Paradox Is the Information Paradox true or false? A whole subject of logic has been constructed on this side of the debate, only largely to dry up as arid after years of certainty. So much for rationality by itself. As Malcolm Perry (2021) observes, Stephen Hawking’s initial view, that quantum mechanics breaks down, is a logical possibility. So as to logic this is rationality. What this fully fledged logic would be, is well formulated; yet this logic in various forms has not yielded evidence of the Information Paradox being true. Furthermore, in relation to its empirical experimental base, it is less secure in empirical and conceptual rationality than an amateur walking on a tightrope. Additionally, at this stage and in the current state of knowledge, we cannot appropriately calculate blackhole evaporation. And accordingly, there is no suitable theory as vehicle for quantum gravity. So, we cannot furnish nor finish the treatment. Classical theory in General Relativity is not going to yield an account treating Hawking radiation. A semi-classical explanation fails. The intricacy and deep complexity of such an understanding however is outside the technical boundary even of many physicists’ expertise and technical mathematical cognizance. Hawking’s initial view that there is an Information Paradox appears to be incorrect. Its logic does not map into empirical theoretical ontology. Yet conversely, in such a balancing act, after much other research, Perry (2021: 5) offers a calculation that “allows for the possibility of a resolution of the information paradox”. This changes the shape of some rationality and accordingly credulity. There appears to be a deeper counter-intuitive rationality in Black Holes.

142

Arthur Gibson

4 Counter-Intuition in Science Counter-intuition cuts across, and is contrary to, the notion that science’s logic and proofs are self-evident. Stephen Hawking predicted that when two black holes collide, they would ‘gently’ merge. He used the area theorem to calculate this, though some scholars were sure he was wrong. When confirmation came, it turned out he was right. It illustrates that even where the basis for a calculation seems intuitively and inductively obvious, intuition may prove to be opposite to what a specialist assumes is correct. This itself counters the somewhat naïve rational assumption that science in principle is the sort of rational enterprise whose rationality is self-evident and surveyable as such. As, for example, the great pure mathematician Mihalis Dafermos stated and proved that General Relativity has the reputation of being notoriously complicated and counter-intuitive. Other grounds for this truth, also applying to other areas of science and mathematics, have been published elsewhere.⁴ Dafermos also expanded this by proving the foregoing. For our purposes, it is sufficient to note and remark on the following: General Relativity generalises Newtonian gravity, for as with electrodynamics, General Relativity is a field theory, where the field now describes the non-trivial “geometry of spacetime”. A feature of this, following on from Einstein,⁵ is that the equations which are hyperbolic partial differential equations, are relevantly similar to Maxwell’s equations, but nonlinear as with the Euler equations.⁶ The geometry of spacetime, as with that of Riemann, does not arise intuitively from Euclid’s first theorems, which Euclid treated as self-evident to the rational person. Rationality built on such premises ends up as a cul de sac where counter-intuitive rationality is concerned. Despite this, what we might term normative scientific intuition has assumed the role as a sort of canonical marker for what is rational, wrongly so.

5 Credulity in Science and Religion I conclude from the foregoing and the attendant references, that these are a basis for exploring and possibly sustaining the concept that rationality in science is

 See Gibson’s Introduction, in Wittgenstein, 2020, and Gibson 2020.  That is, Einstein’s Die Feldgleichungen der Gravitation.  I.e., expressed as noninear hyperbolic partial differential equations. Cf. Dafermos and Luk, 2017.

Paradox in Ontology

143

not in principle impervious to, nor autonomous from, other subjects’ relevant fields of reasoning in certain domains of the humanities, of which semantic fields of theism are candidates for such investigation in these pertinent respects. If so, philosophy has much to re-learn concerning its triumphalist assumptions about some reason. We can benefit from going back to consider some of Wittgenstein’s uncertainty about a bunch of issues about logic, though this is not to presuppose he was free of bias, as with all of us. We are in the thick of dense assumptions and disputes with philosophy’s differing stance towards science and religion. Pichler has made an original start on a batch of concerns.⁷ He addresses Wittgenstein comparing the honest religious thinker to a tightrope walker.⁸ With Alois Pichler, we notice Wittgenstein’s formulation: The honest religious thinker is like a tightrope walker. It almost looks as though he were walking on nothing but air. His support || footing is the slenderest imaginable. And yet it really is possible to walk on it.⁹

Pichler observes that Wittgenstein may well be tracking Newman’s Catholic thinking here: “Recollect, to write theology is like dancing on the tight rope some hundred feet above the ground. It is hard to keep from falling, and the fall is great… . The questions are so subtle, the distinctions so fine, and critical jealous eyes so many.”¹⁰ So Severin Schroeder: “On Wittgenstein’s account, the respectable theist is the one who is knowingly not reasonable in his religious beliefs (LC 58 f.). But how can one believe what, at the same time, one believes is not likely to be true? – This is the unresolved tension in Wittgenstein’s philosophy of religion. The tightrope walker’s carefully limited abeyance of reason may not be objectionable; but is it possible? Hume famously claimed that ‘whoever is moved by Faith to assent to [the Christian Religion], is conscious of a continued miracle in his own person, which subverts all the principles of his understanding’ (131). With regard to the majority of religious people that is certainly not true; but it

 Pichler, 2021, 2.  Wittgenstein, 1998, 84.  See the Nachlass Wittgenstein, 2015; CV p. 84e; written on July 5, 1948, in Ms-137,67b; for the tightrope walker-simile, and Pichler (2021) and his diplomatic transcription at http://wittgen steinsource.org/Ms-137,67b[4]_d.  Pichler, 2021, 9, informs us of a letter by J.H. Newman to Emily Bowles of April 16, 1866, and John T. Ford’s “’Dancing on the Tight Rope’: Newman’s View of Theology” (p. 127 ff.) Pichler’s reference, https://ejournals.bc.edu/ojs/index.php/ctsa/article/view/3302/2915.

144

Arthur Gibson

may be true of the consciously not reasonable believer Wittgenstein envisaged.”¹¹ As Schroeder rightly considers aspects: “To be rational is to do what is appropriate in the light of one’s aims and objectives. In general, to be well-informed is likely to be useful in the pursuit of one’s goals, to be misinformed is likely to be a hindrance to success. Therefore, it is rational to be critical. Credulity, a lack of care about the correctness of one’s beliefs is, in general, irrational. But there are exceptions. For one thing, there may be areas of belief so far removed from anything we need to know in order to pursue our goals efficiently that an error is of no consequence.”¹² One of my strategies is that parallel species of counter-intuition in the forgoing scientific domains can be found in some theism. This prospect applies to select spheres of deeper physics and features rationality in mathematical and logical patterns. It is about time that scientists and philosophers recognised more than is customary that this credulity, together with the lack of counter-intuitive attention to some borders between science and our ignorance, have failed to recognise similarities with certain grounds of language in religious belief, as well as metaphysical properties, and limits, in some scientific languages. This pertains to grasping some such metaphysical spheres in areas of black holes mathematical physics. Consider the case of dark energy.¹³ Such ignorance and credulity can be found in some other more standard domains of physics, marked by uncertainty and dogmatism. For example, think of range of examples where prizes are offered for theorems that do not yet exist. What ‘credulity’ is present, sometimes scientific or religious, is problematic, as contrasted with or compared to the criteria presupposed in the supposed selfevidence in explicit specification of assumptions. Many physicists would fall off their tightrope, and maybe not be able to stand on it. This is not solely lack of technical expertise. Rather, paradoxical dispute about conflict-resolution that has unstable differing formulations. It is also where formulations can fall off the tightrope and crash to the ground. So, even where there is expertise, there

 Schroeder, 2007, 462.  Schroeder, 2007, 458.  On dark energy, see Gibson, 2020. Martin Rees’s note to me speaks of a way scientific intuition can be violated by dark energy research: “As regards ‘dark energy’ most physicists would agree that we won’t understand it until there is a tested unified theory. Most such theories (e. g. string theory or quantum loop gravity) suggest that space should have ’graininess’ or structure, but on the Planck Scale (10 – 33 cm). So I’d guess it will be a long time before we have such a theory. And it wouldn’t be surprising if it violated our everyday intuitions even more than quantum theory does.”

Paradox in Ontology

145

is disagreement about the locations and nature of the tightrope, on which they reportedly stand, even if they were (to think of how Wittgenstein also framed some balancing) to be in perfect dietary health on the tightrope. In sum, tightrope walking is not the sole prerogative of dream-religious language, but also of scientific rationality aspiring to cosmic universality.

6 Rationality & Credulity in Ontology In this perspective of realist emerging cosmology, we would do well to draw on Jonathan Smith’s way of formulating exposure of “evidential authority support[ing] authorial intention embedded in the Nachlass”.¹⁴ That is, by drawing attention to some newly published evidence of the “record of Wittgenstein’s cognitive process when writing”, for example in dispute with Russell’s and indeed Euclid’s assumptions. In Wittgenstein (2020),¹⁵ this manuscript argues that non-self-evidence occurs in areas of logic and maths, also maintaining that our capacity to have knowledge is often bewitched by our perceptions, even in some mathematics and science. Here are some such uses in a manuscript that manifests Wittgenstein’s (2020) authorial cognitive processes in semantic engagement, with what is tantamount to encountering borders of counter-intuition relating to credulity: One of the great difficulties which is in logic was this – what is the criterion for a proposition being a proposition of logic? By some it was taken to be self-evidence. This was what both Russell and Frege said. At certain points where this is a difficulty one tends to bring in psychology. The word ‘self-evident’ is taken from psychology. It is what everybody is in his right senses would say is true. It is the highest degree of plausibility. But it is clear that what is self-evident may not be to another. People did not mean to say these propositions were self-evident to Europeans. The self-evidence seemed to be an objective one, not a subjective one. It seemed to lie in the proposition.

The manuscript’s conclusion is: This shows a logical proposition has nothing to do with self-evidence.¹⁶

 Smith, 2021, 153.  Wittgenstein, 2020, xviii  Wittgenstein, 2020, 273.

146

Arthur Gibson

We might here draw attention to Severin Schroeder’s statement: Credulity, a lack of care about the correctness of one’s beliefs is, in general, irrational. But there are exceptions.¹⁷ But how can one believe what, at the same time, one believes is not likely to be true? – This is the unresolved tension in Wittgenstein’s philosophy of religion. The tightrope walker’s carefully limited abeyance of reason may not be objectionable; but is it possible?¹⁸

This is exemplifying, especially, the case for care in the domain of counter-intuition, where examples may incorrectly be merged in some cases with lack of credulity, when it is counter-intuition. In assessing what it is to be “what is likely” helpfully characterised by Severin, we should re-weigh some typical philosophical standardised dogma too readily presupposes and implements questionable notions of what are the boundaries deployed to police reason. The epistemology of ignorance illustrated by Hawking’s genuine changing of mind (quite properly so). We should especially re-consider how such fine speculation runs almost entirely free of laboratory’s science criteria of experimentally justified knowledge, as well as allow that even with such closely tracked experimental science as the emergent family of Higg’s boson discoveries, how new knowledge relates to generalisation about the universe’s history and consequently its present, is not as stable as to eliminate apt application of tightrope walking. Although the process and empirical status of such science and discussion about theism are vastly different, the identity of ignorance, and mistaken balancing have parallels. We can extend Smith’s¹⁹ enquiry into this perspective: by asking the question of whether we should beware of being, as Wittgenstein could have said, bewitched by modernist’s science overblown, which is sometime misused as the catalogue of contrast between itself – as the alleged sole arbiter of what is admissible to move over into the boundary of objective science – wrongly supposing itself have competence to recognise the worth and identity of all emerging deep original machinery intricate transcendent tightrope walking as if it were an entirely contentless subjective enterprise. We should recall that for some time Einstein was deemed by such reactionary authority to be unfit for a university post. Sure, we all know the validity and strength of the science we know and use. Rather, the question here is how to with total disinterest to dispose our-

 Schroeder, 2007, 458.  Schroeder, 2007, 462.  Smith, 2021, 152: “Indeed we might ask the question whether we should beware of being, as Wittgenstein may have said, bewitched by the catalogue” (of his Nachlass formulated by Von Wright).

Paradox in Ontology

147

selves objectively to originality may be uncovering science that requires a new symbolic language to depict hitherto unknown empirical reality.

7 Counter-Intuitive Similarity between Science, Humanities, and Theism In the case of very-difficult original advanced science, scientists and mathematicians experience difficult challenges at the very limits of knowledge, which also has to do with the depth of unexpected complexity – sometimes beyond previous experience. Previous research attempts to answer the question of whether this shows our knowing a concept is not in itself a recognitional capacity to understand instance of its scope.²⁰ This particularly obtains for the present context. For example, would educated people have a rational capacity, if informed, to recognise any instance of a pattern we think we know in say maths or science, if it occurred in some humanities context, for example in theism? And vice versa? Mathematics is richly filled with a history of non-empirical discoveries that are totally surprising, also in directions asymmetric to what is previously known, whilst being entirely consistent with prior knowledge pure mathematics. This is counter-intuition. Consider, only retrospectively, a rather obvious, yet previously unrecognised instance of this – a case that I have introduced elsewhere.²¹ Sachs²² did not see nor consider any link between his symmetry group theorem and relevant black hole mathematics as considered by Hawking. He considered his theorem in effect locked off from further contribution to certain black hole problems below their surfaces, which came to be seen as involving soft hair. That was until Perry and co-authors, using what previously seemed an elementary isolated point of Poincare´’s. They developed a radically new solution for black hole entropy.²³ This creative probing of the active surfaces of black holes yielded a new route to furnish a route, which with another contribution appears lead to solving Hawking’s information paradox – the paradox discovered by Hawking in his earlier research in 1974.²⁴

    

Gibson, 2020. Wittgenstein, 2020, xix. Sachs, 1962, 103 ff. Haco, S., Hawking, S. W., Perry, M. J., & Strominger, A., 2017 and 2018. Perry, 2021 [forthcoming].

148

Arthur Gibson

The deployment of the term ‘paradox’ by Hawking, bears on its face, the need for a rather rare sort of scientific humility. Hawking’s proposal and its withdrawal ,or at least transforming the paradox, into another family resemblance, is a sort of mathematical physics tightrope walking within the deeper atmospheric surface within black holes, for which there is no direct verification. This nicely exemplifies how the epistemology of ignorance within Hawking’s cosmology relies on the sort of science that has family resemblance to tight-rope walking discussion of a deeper sort in some metaphysics – admittedly very rare in theology or philosophy of religion. Such a move as made in this present chapter is only the beginning of a counter-intuitive renaissance of research that needs to engage in original ways with mathematical issues of infinity, and their relations to and contrasts with matters of theism and their relation to counter-intuitive ontology. This nudges exploration into a fresh future counter-intuitive arena for research, that of pure mathematics, associated with cosmology, introduced into metaphysics, to research questions of the prospect of new exploration of empirical infinity, and its relation to divine cosmology, which we might designate as divining cosmology. This presupposes we have a grasp of mathematical infinity in certain realms, which of course we do not adequately understand, as Woodin has shown.²⁵ Woodin reveals our ignorance of infinity. So we would do well to place investigation of our epistemology of ignorance at the centre of research in ways that Stephen Hawking did not, pioneering though his capacity to raise profound new questions was. Such extremes present difficult complexity. Particularly is this the case in some areas of mathematics. This displays asymmetry between what is known, and highly counter-intuitive states of affairs. The question of formulating the most suitable language adequately to expose the inner depths of these structures is vast, even sometimes when the area becomes known. Consider for example those domains covered by known equations for which a theorem still remains unknown or elusive. And vast problems for those naturally disposed to a programmatic sense of self-evident intuition. Rather like a ballerina having to spin on her tip-toes – on a large spinning ball lubricated by thin oil. Postmodernism and its children need to make space for entirely new counter-intuitive ontological research. Not to revive the somewhat gross models of past theism that attempted to eliminate or compromise modernism; nor to be replaced by an enfeebled theistic ontology; neither weak formulation of infinity that suffocates bold exploration.

 Woodin, 2017.

Paradox in Ontology

149

Accordingly, is there not some similarity in the case of certain areas of the humanities, which present equally deep conceptual, albeit distinct, difficulties for comprehension? Cases at extreme point from the known, are familiar yet enormously difficult. Even akin to a forever-receding horizon of the universe, which beyond it, as Martin Rees has stated, may be unknown to us have a further region of say 1010 domains beyond observation; yet at which cosmology connives measurements, or leaves the matter aside, though they may bear on our origins and remote future? Certainly, there are enormous differences between science and humanities. Nevertheless, consider the stage at which numbers in higher mathematical abstraction seem to be unstable in our hands, or too counter-intuitive for us yet to know what they amount to, as with infinity. Reflect also on how such spheres as these are grasped partially by all those distinguished professional scientists and learned mathematicians who comprise and guide institutional consensus, for whom original contribution to these research worlds is not an option, even though appreciated, be it in science and some humanities. Disinterested humility needs not to be a rare commodity in the light of our ignorance. In the rarefied yet vast deep different identities in the science and in the humanities, entirely fresh investigation is needed on certain prospects of parallels between them in relation to what counter-intuitive analogy is. Such a project realm is too easily missed, shrunk, belittled, ignored only admired from a distance or easily missed. This project can be overstated or understated. Despite this, it would do all parties some good to look with research sympathy on the others. All will be having difficulties of mastering the unknown with languages which seem to be quite unstably elusive, as with devising new languages in science and humanities, or counter-intuitively exploiting radical extension of new ones. Counter-intuitive family resemblance, unbewitched by intuitive language, could confer, adorn, expose fresh subtle identities between the known and the unknown, as well as between science and humanities, or mutually spawn some original recognition in the other. This also applies to that special case, theism. Just as science had to rid itself of alchemy, whilst finding that stars may practice its changing elements. So also, theism needs to rid itself of its equivalent. But not by ejecting ontology. Rather, by dry-cleaning it with counter-intuition.

150

Arthur Gibson

8 Conclusion In Stephen Hawking’s 1980 Cambridge Inaugural Lecture as Professor, for the Chair that had been held by Newton, he appeared to expect that the end of physics was in sight, perhaps hoping it may be brought by his contribution. Even so, there was his undertone of irony and humour, which were there till the end of his life. He was preoccupied with himself, other people, and humour and the like, in engagement with his colleagues and friends. We should recall this now, for this is how sometimes he approached the matter of God. Was he ever serious assessing questions about God? Yes and No. Like many of us, he was more interested in his autobiography and intellect. We can attend his disposition to engage with the Universe as if it were his own autobiography. This is one way of viewing the film about him: The Theory of Everything, not least because it was actually based on his first wife’s book about him, quite different from his own take on his life. In private conversation I asked him what he thought about his wife having her view of him presented in the film, as if it were him, since his view was quite sometimes different from her vision of him. He replied, “Well, she has the right to have the opportunity to have her say.” So has God, I mused. He seemed to smile, mischievously with a tear there. A good way to engage with him on cosmology was to begin with humour. So, in such familiar vein, having chatted with him earlier in the day, aware of his eagerness to know of the Oscar outcome, I later emailed him as soon as the Oscars were announced. I mockingly congratulated him for ‘gaining’ an Oscar for his role in the The Theory of Everything. Here is my email, with his reply: At 18:23 27/02/2015: Dear Stephen, I just wanted to congratulate you on your splendid portrayal of Eddie Redmayne in “The Theory of Everything”. You deserve the Oscar. Arthur.

Stephen modestly replied by email: Stephen Hawking To: Professor Arthur Gibson Date: Sat, 28 Feb 2015 11:56:58 +0000 Subject: Re: Stephen and Eddie Arthur: Thank you. I played Eddie well. Same appearance, same voice and same sense of humor. Stephen.

Paradox in Ontology

151

When I talked to Eddie Redmayne about my exchange with Stephen, and later showed this correspondence to him, he replied to me, appropriately on 1st April 2018: “It made me laugh. A lot. I only met Stephen a handful of times but it was always his humour that left me reeling.” We are left with the tragedy of wonder at his gifts, his suffering amid humour, reflecting if he had spacetime to allow for counter-intuitive thinking about God – beyond the confines of his suffering, as he did about the Universe.

References Berto, Franseco and Matteo Plebani, 2015: Ontology and Metaontology. Bloomsbury, London Dafermos, Mihalis and Jonathan Luk. 2017: The interior of dynamical vacuum black holes I: The Co-stability of the Kerr Cauchy horizon, 2017; online at http://arxiv.org/abs/1710. 01772 Gibson, Arthur, 2000: God and the Universe. Routledge, London. Gibson, Arthur, 2020: “Intuition, Counter-Intuition, & the Absence of Ontology for Dark Energy”. Margit Gaffal (Ed.); Language, Truth and Democracy: Essays in Honour of Jesús Padilla Gálvez, De Gruyter: Berlin/Boston: 124 – 135. Haco, S., Hawking, S. W., Perry, M. J., & Strominger, A. (2017): The Conformal BMS Group. arXiv: 1701.08110. Haco, S., Hawking, S. W., Perry, M. J., & Strominger, A. (2018): Black hole entropy and soft hair. arXiv: 1810.01847v4. Perry, Malcolm J. 2021: “ No Future in Black Holes. [For JHEP, Forthcoming] Pichler, Alois. 2021: “Religion without beliefs?” Sachs R. K. 1962: Gravitational waves in general relativity VIII. Waves in asymptotically flat space-time, Proc. R. Soc. London, 270, 1340, 103 – 126. Schroeder, Severin. 2007: “The tightrope walker”, Ratio, 20:4, 442 – 463. Smith, J. 2021: “Wittgenstein’s Nachlass, a Case for Practical Ontology?”, in the current volume. Wittgenstein, L. 1998: Culture and Value, ed. by Georg Henrik von Wright, rev. ed. London, Wiley-Blackwell. Wittgenstein, Ludwig. 2015: Nachlass Wittgenstein 2015; diplomatic transcription at http:// wittgensteinsource.org/Ms-137,67b[4]_d. Wittgenstein, Ludwig: 2020: Ludwig Wittgenstein: Dictating Philosophy To Francis Skinner – The Wittgenstein-Skinner Manuscripts. Transcribed and edited with Introductory Chapters and Notes by Arthur Gibson; Editors: Arthur Gibson and Niamh A. O’Mahony. A Special Edition of the Nordic Wittgenstein Studies Springer: Cham. Woodin, W. H. 2017: Foundations of Mathematics. Logic at Harvard. Essays in Honor of W. Hugh Woodin’s 60th Birthday. Contemporary Mathematics, 690. Eds. A. Eduardo Caicedo, J. Cummings, P. Koellner and P. B. Larson. American Mathematical Society, Cambridge, MA., Providence.

List of Contributors Prof. Dr. Peter Simons Professor Emeritus Trinity College Dublin College Green, Dublin 2 (Ireland) [email protected] Prof. Dr. Julian Nida-Rümelin Ludwig-Maximilians-Universität München Lehrstuhl für Philosophie IV Geschwister-Scholl-Platz 1 80539 München (Germany) [email protected] Prof. Dr. Margit Gaffal Departamento de Estudios Ingleses: Lingüística y Literatura Universidad Complutense de Madrid Avenida de Filipinas, 3 28003 Madrid (Spain) [email protected] Prof. Dr. Jesús Padilla Gálvez Faculty of Law and Social Sciences University of Castilla-La Mancha San Pedro Mártir s/n. 45071 Toledo (Spain) [email protected]

Prof. Dr. Antonio Marques Departamento de Filosofia (DEF) Instituto de Filosofia da NOVA (IFILNOVA) Universidade Nova de Lisboa Av. Berna, 26-C Lisboa (Portugal) [email protected] Prof. Dr. Manuel García-Carpintero Departament de Lògica, Història i Filosofia de la Ciència Universitat de Barcelona Montalegre, 6 – 8, 4ª planta 08001 Barcelona (Spain) [email protected] Jonathan Smith B.A., M.Ar.Ad. Trinity Street Cambridge, CB2 1TQ (United Kingdom) [email protected] Prof. Dr. Arthur Gibson Department of Pure Mathematics and Mathematical Statistics Centre for Mathematical Sciences Wilberforce Road Cambridge CB3 0WB (United Kingdom) [email protected]

Index of Names Aristoteles 1 – 4, 16, 18, 22, 48, 65, 79, 91, 94 Ball, Derek 113 Bealer, George 114 Bell, John Stewart 56 Berto, Franseco 140 Bolzano, Bernard 16, 21, 24, 67, 72, 79, 86, 93 Bouveresse, Jacques 105 Bovens, Luc 53 Brentano, Franz 21 – 23, 67, 70, 80, 82, 86, 91, 93 Buchanan, Ray 116 Burge, Tyler 12, 114 f., 124 Camp, Elizabeth 117 f. Caplan, Ben 119 Carnap, Rudolf 101 Chisholm, Roderick M. 22, 56 Church, Alonzo 9 – 11 Collins, John 111 – 113, 120 – 122, 124 Correia, Fabrice 122 Dafermos, Mihalis 142 Davidson, Donald 50, 52, 112 – 116 Descartes, René 66 Dowell, Janice L. 124 Dummett, Michael 124 Duncan, Matt 117 Edwards, Paul 66 Ehrenfels, Chr. von 87 Eklund, Matti 113, 115 Field, Hartry 113, 115 f., 118 Fine, Kit 116, 123 Frege, Gottlob 4 – 6, 11, 17, 21, 24, 26, 35, 75, 79, 89 – 91, 94, 114, 121, 123 f., 127, 145 Frisch, Max 45 Gaffal, Margit

16, 63, 139

https://doi.org/10.1515/9783110750041-011

García-Carpintero, Manuel 17 f., 111, 118, 123, 125 Gaskin, Richard 111 f. Gettier, Edmund 49 f. Giardino, Valeria 117 Gibson, Arthur 18, 139 f., 142, 144, 147 Glick, Ephraim 117 Green, Mitchell 113 Greenberg, Gabriel 117 Grimm, Jacob 66 Grimm, Wilhelm 66 Grzankowski, Alex 116 Gundisalvo, Domingo 2 Haag, Johannes 65 Haco, Sasha 147 Hanks, Peter 17, 111 – 113, 116, 118 f., 121 – 127 Hartmann, Stephan 53 Hawking, Stephen W. 18, 139 – 142, 146 – 150 Hayes, John R. 133 f. Hegel, Georg Wilhelm Friedrich 7 Heinzmann, Richard 66 Höfler, Alois 22, 64, 67 Hofmann, Doris Vera 106 Hom, Christopher 124 f. Horatio 51 Husserl, Edmund 7, 17, 21, 23 – 25, 64, 79, 86 f., 94 Jespersen, Bjørn 125 Johnston, Mark 114 f., 120, 122, 124 Kant, Immanuel 4, 35 – 39, 47 f., 66, 106 Keller, Lorraine 116 f. King, Jeffrey 17, 111 – 113, 115 – 121, 127 Knüfer, Karl 67 Lande, Kevin 117 Larmore, Charles 54 Lewis, David 114 – 116 Liebesman, David 114 f., 117, 120, 122, 124

156

Index of Names

Mackie, John 56, 61 Marques, Antonio 17, 99 Marty, Anton 21, 23 Matthews, Robert J. 113, 116 Meinong, Alexius 15 f., 21 – 30, 82 – 86 Merricks, Trenton 114, 117 Moore, Joseph 21, 24 f., 27, 104, 111, 116 Moyal-Sharrock, Danièle 104 Mulligan, Kevin 123 Nida-Rümelin, Julian

16, 33, 50, 55

Ohmann, Richard 126 Olinghouse, Natalie G., Ostertag, Gary 116

133 f.

Padilla Gálvez, Jesús 1, 5, 16 f., 79 Pagin, Peter 111, 122 Pautz, Adam 116 Perea, Andres 53 Perler, Dominik 65 Perry, Malcolm J. 113, 141, 147 Pichler, Alois 143 Pickel, Bryan 111, 117, 119 Platon 2, 22, 65, 86, 103, 113, 122 Plebani, Matteo 140 Pryor, James 120 Putnam, Hilary 101 f. Quine, W. v. O.

1, 8 – 15, 59, 77, 80 f.

Recanati, François 111, 126 Reiland, Indrek 111, 125 Richard, Mark 116, 118 Roberts, Craige 121 Rosen, Gideon 122

Russell, Bertrand 4 – 6, 21, 24 – 27, 102, 145 Ryle, Gilbert 9, 23 Sainsbury, Mark 116, 124 Scanlon, Thomas 57 – 59 Scheler, Max 44 Schnieder, Benjamin 113, 115 Schroeder, Severin 143 f., 146 Schwatz, Jeremy 124 f. Simons, Peter M. 15, 21 Smith, Barry 123 Smith Jonathan 18, 131, 139, 145 f. Soames, Scott 17, 111, 112, 116, 118 f., 121 f., 124 f., 127 Speaks, Jeff 116 f., 119 Spohn, Wolfgang 56 Stalnaker, Robert 113, 116 f. Stock, Kathleen 125 f. Strominger, Andrew 147 Stumpf, Carl 17, 21, 23, 79, 85 – 90, 94 Thomasson, Amie L. Twardowski, Kasimir

112 16, 23, 63 f., 66 – 77

White, Morton 59 Winch, Peter 54 Wittgenstein, Ludwig 5, 15, 17 f., 21, 24 f., 27, 79, 94, 99 – 108, 131 – 139, 142 – 147 Woodin, W. Hugh 148 Wright, Crispin 114 Wright, G. H. von 100, 104, 132, 135 f., 146 Yablo, Stephen

112, 116

Zeno of Citium 44 Zimmermann, Robert

67

Subject Index act

3, 16, 23 f., 27, 36, 39, 43, 46, 48, 51 – 53, 60, 66 – 70, 73 f., 76, 82, 86, 88 – 90, 101 f., 106, 109, 112 f., 115, 117, 121 – 127, 141 action 7, 13, 33 – 48, 50, 52 – 54, 56 – 60, 75, 104, 106, 108, 122 f., 125 agency 16, 33, 40 akrates 46 Anschauungen 67 anti-Kantian 16, 79, 93 a priori 7 a priori knowledge 111, 114 argument 1, 3 – 5, 12, 16, 21, 23, 28, 41, 46, 53, 55, 64, 84, 87 f., 90, 112, 114, 117 – 119, 124 f. Aristotelian logic 16, 79, 94 aspect 2, 7 – 9, 13, 16, 33 – 35, 38, 41, 45, 49, 56, 66, 68, 76, 82, 90, 126, 133, 136, 138, 144 assumption 1, 4, 9, 12, 15 f., 21, 24, 26, 29 f., 33 f., 59, 66, 69, 73, 79 f., 82 – 85, 93, 113, 118, 125, 142 – 145 authorship 46, 48 – 57 behavior 37, 39 f., 43, 45 f., 49, 52, 54, 57, 81, 88 being 1 – 4, 6 – 8, 14 f., 17, 23 – 27, 29 f., 35 – 37, 40 – 43, 46 – 49, 52 f., 55 f., 60 f., 65, 68, 73, 75 f., 79, 82 – 84, 89, 91, 100, 102 – 105, 107 f., 113 f., 118 – 121, 124, 131 – 134, 136 f., 139, 141 f., 145 – 147 belief 12, 16, 33 f., 37 f., 40 f., 43 – 45, 49 f., 53 – 56, 58 – 61, 85, 88, 91, 104, 106, 112, 117 f., 120, 143 f., 146 Brentano School 16, 67, 79, 85, 94 carrier 17, 79 carrier of properties 17, 79 categorematic 69 categorematic term 63 categorical imperative 38, 47 f., 57, 75 causal theory of knowledge 49, 60 https://doi.org/10.1515/9783110750041-012

certainty 17, 65, 99 f., 103 – 108, 137, 141 cognition 15, 22, 24, 26, 121 communicative practice 16, 33 – 35, 59 community 105, 108 comprehensive moral doctrines 59 concepts, psychological 7 f., 12, 28, 50, 59, 63 – 67, 73 – 75, 81, 87, 89, 111, 114, 122 conceptual 35, 50, 52, 65, 74 f., 86, 101, 139, 141, 149 conceptual representations 74 conditional structure 16, 79, 94 consciousness 23, 65 f., 87, 106 construct 63, 75 – 77, 83, 86, 89 f., 103 content 1, 4, 7 f., 16 f., 23 f., 27, 50, 64, 67 – 70, 72 – 77, 79, 83 – 91, 94, 100, 108, 111 f., 114, 117 – 126 conversio ad phantasmata 66 epistemic states 45 eudaimonia 48 experience 16, 33 – 36, 54 f., 59 f., 66, 104 – 106, 112, 117 f., 120, 123, 147 explanation 16, 44, 63, 67, 72, 75, 112 f., 115 f., 118 f., 122, 141 fact

6 f., 11 f., 15, 17, 21, 23 – 29, 38, 40, 42 f., 45 f., 50, 55 – 61, 64, 67 f., 71, 76 f., 83 – 85, 89 – 93, 100 f., 105 – 107, 113 – 115, 118 – 122, 126 f., 134 – 137 figura etymologica 76 freedom 26, 34 – 36, 38 – 41, 43 – 45, 49, 55, 57, 71 freedom of choice 35 – 38, 40 function 8 f., 16, 47, 49, 63 f., 68, 73, 75 f., 80 f., 85 – 91, 119 fürstellen 66 game 17, 81, 99 f., 102 – 109, 126 f. grammar 67, 76, 91, 108 history 1, 5, 8, 22, 54, 65, 80, 112, 134, 140, 146 f.

158

Subject Index

hope 38, 131, 136 hormai 44, 48, 51 f. human action 49, 51 Humeanism 47 – 50 implicit knowledge 33 indirect representations 63, 72 f. intellectus agens 66 intellectus possibilis 66 intentionality 22 – 24, 79, 86 judgment 14 – 17, 21, 23, 33 f., 40 – 42, 44, 46, 48 – 50, 52, 54 f., 59 f., 63 f., 68 f., 73 – 76, 79, 82 – 91, 94, 104, 108, 115, 120, 122 f. know 1 – 4, 6, 8, 17, 25, 28, 40 – 42, 49, 66, 69, 80, 82, 84, 87 f., 90, 92, 99 – 101, 104 – 109, 111, 116, 144, 146 – 150 knowledge 6, 16, 34, 40, 46, 49 f., 65 f., 75, 80, 99 – 102, 104 – 108, 126, 131, 133, 137, 139, 141, 145 – 147 language 4, 6 f., 12, 14 f., 17, 23, 41, 50, 52, 60 f., 63 f., 66, 68 f., 74 f., 79 – 81, 99 – 109, 112, 118 f., 121, 134 f., 144 f., 147 – 149 language games 17, 80, 99, 102, 104 f. life-world 54 logic 5 – 7, 9 f., 12 – 15, 27, 29, 38, 64, 67, 69, 76, 83, 91 f., 107, 139, 141 – 143, 145 maieutikè 34 mathematics 139 f., 142, 145, 147 f. meaning 3 – 8, 10, 12, 17, 23, 26, 28, 52, 58, 63, 65, 68 f., 74 – 77, 88, 94, 99, 106, 112, 121 – 124, 138 meaningful 69, 80, 123 mental acts 23 f., 27, 90, 123 metaethics 34 metaphysic 16, 33 f., 54, 115, 139 f., 148 modalities 112 mythology 36, 105 Nachlass 18, 131 – 138, 143, 145 f. name 6, 11 – 15, 21 f., 68 f., 72, 80, 88, 94, 101 – 103, 116

natural Language 5, 15, 81, 117, 121 necessary truths 29 necessity 28, 112 normative 16, 33 – 35, 38, 40 – 43, 45, 47 f., 51 f., 54 – 60, 125 f., 142 object 3 f., 6 f., 10, 12, 15 – 17, 21 – 31, 40, 53 f., 56, 61, 63 – 77, 79, 81 – 85, 87 f., 90, 93 f., 100 – 102, 108, 112, 116 f., 126 objective 7, 15 – 17, 21 – 30, 35, 38, 42 f., 48, 55 f., 58, 60, 67, 75, 79 f., 82 – 86, 93, 104, 106 f., 112, 120, 122, 144 – 146 objects of judgment 15, 21, 23 ontological commitment 1, 4, 7 – 17, 79, 99, 101, 103 f. ontological questions 3, 17, 99, 108 ontology 1 – 9, 12 – 14, 17 f., 22 f., 26, 61, 63 f., 71, 75, 77, 79 – 81, 83, 100, 102 f., 107, 112, 131, 136, 138 – 141, 145, 148 f. ought 47 perceptio 60, 63, 65 – 68, 120, 123 philosophy 2 f., 7, 9, 13, 16 – 18, 22, 27, 50, 52, 54, 56 – 58, 64, 67, 80, 99, 101, 103 f., 107, 111 – 113, 115, 131 – 135, 137, 139 f., 143, 146, 148 philosophy of language 17, 99, 115 philosophy of psychology 17, 99 f., 108 possible worlds 41, 118 practical ontology 18, 131 practical rationality 16, 33 f. practical reason 35, 37 f., 41, 45, 49, 52, 58 predicative 5, 7, 17, 79, 94 pretheoretical conception 111, 114 prohairetic states 43 proof 93, 104, 139 f., 142 proposition 1, 3, 7 f., 15 – 18, 21, 23 – 25, 27, 30, 34 f., 88, 90, 93, 100 f., 103 – 107, 111 – 127, 145 propositional attitudes 15, 21, 30, 83, 85 psychic phenomena 64, 74 psychology 22, 54, 64, 76, 85, 145 Quine’s Criterion

9 f., 15

Subject Index

reason, philosophical 8, 16 f., 23, 29, 33 – 41, 43 – 50, 52 – 61, 91, 108, 112, 115, 117 f., 122, 131, 136 f., 139, 143, 146 record 69, 131, 133 f., 137, 145 regularity 50, 104, 131 relationships 15, 21, 64, 75, 89, 131, 136 – 138 relative truth 63, 73 f. repraesentatio 66 f. representation 5 f., 8, 16, 63 – 77, 79, 82 f., 85 – 87, 90, 93, 100, 111, 114 – 118, 120, 123, 136 representational content 70, 111, 118, 126 representational vehicles 111, 113, 116 f., 127 responsibility 12, 14, 22, 35 f., 40 – 43, 45 – 49, 57, 79

159

Sachverhalt 15, 17, 21, 23, 79, 82, 85 – 87, 89 f., 92 – 94 Satz an sich 16, 21, 79, 86, 93 self-efficacy 35, 54 f. self-evident 142, 145, 148 semantic 5, 12, 93, 113 – 117, 119, 121 f., 125, 143, 145 sine ira et studio 44 skepticism 56, 61, 112 state of affairs 7, 15 – 17, 21 – 25, 30 f., 40 – 43, 46, 49 f., 52, 58, 79, 82 f., 85 – 94, 116, 140 f., 148 status of reasons 54, 56, 58 subsists 17, 79, 83