Ontology of Theistic Beliefs 9783110565799, 311056579X

Citation preview

Ontology of Theistic Beliefs

Philosophische Analyse/ Philosophical Analysis

Herausgegeben von/Edited by Katherine Dormandy, Rafael Hüntelmann, Christian Kanzian, Uwe Meixner, Richard Schantz, Erwin Tegtmeier

Band/Volume 74

Ontology of Theistic Beliefs Edited by Mirosław Szatkowski

ISBN 978-3-11-056579-9 e-ISBN (PDF) 978-3-11-056651-2 e-ISBN (EPUB) 978-3-11-056589-8 ISSN 2198-2066 Library of Congress Control Number: 2018948484. 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. © 2018 Walter de Gruyter GmbH, Berlin/Boston Printing: CPI books GmbH, Leck www.degruyter.com

Contents Acknowledgements | VII Mirosław Szatkowski Ontology of Theistic Beliefs: A Short Orientation | IX Chris Daly Agnosticism and the Balance of Evidence | 1 Gabriele De Anna Theism and the Ontological Ground of Moral Realism | 19 Michał Głowala Polygeny, Pleiotropy, and Two Kinds of Concurrentist Ontology | 39 Christian Kanzian “Bottom-up” versus “top-down” | 63 Daniel Linford and Jason Megill Cognitive Bias, the Axiological Question and the Epistemic Probability of Theistic Belief | 77 Jason Megill and Daniel Linford On Computable Metaphysics: On the Uses and Limitations of Computational Metaphysics | 93 Uwe Meixner What Evil Must Be in Order to Exist | 113 Elisa Paganini Normative Rules for Indeterminacy | 129 Eleonore Stump The Openness of God: Eternity and Free Will | 137

VI | Contents Mirosław Szatkowski The Recovery of St. Thomas Aquinas. Part I: The Doctrine of St. Thomas Aquinas and its Non-Analytical Versions | 155 William F. Vallicella Does God Exist Because He Ought To Exist? | 205 Peter van Inwagen God’s Being and Ours | 213 Authors of Contributed Papers | 225 Abstracts | 231 Person Index | 237 Subject Index | 241

Acknowledgements This volume could not have been realized without contributions from individuals and institutions which, in many ways, accompanied this effort. First and foremost, I would like to thank heartily all the authors who accepted the invitation to contribute to the volume and showed great understanding and patience during the publication process. I would like to thank the anonymous reviewers for their suggestions and comments. Following their recommendations, authors included several improvements in their texts. Special thanks are due to the National Science Center for the financial support of the publication of this book, within the research grant OPUS 4, DEC2012/07/B/HS1/01978. I would like to acknowledge the Institute of Philosophy and Sociology at the Polish Academy of Sciences (IFiS PAN), for implementing the above research project. Finally, special thanks are also due to Dr. Bartłomiej Skowron for his faithful support of my initiatives, and Mr. Sławomir Szatkowski for his help in preparing this volume for print. I have tried to make the best possible use of all his suggestions and corrections. Mirosław Szatkowski

https://doi.org/9783110566512-201

Mirosław Szatkowski

Ontology of Theistic Beliefs: A Short Orientation* This article is intended as preparation for the relatively challenging reading of this book, in which 12 collected essays are loosely grouped around the theme of an ontology of theistic beliefs. First, the terms ‘ontology’ and ‘theistic beliefs’, as well as the interaction between them, require some explanation. We begin by discussing the meaning of ‘belief’, and then discuss the difference between theistic and non-theistic beliefs.

1 It is difficult to define ‘ontology’. On the one hand, it is a discipline with a long history, but on the other hand, there is no common, widely accepted, definition of the term ‘ontology’. The term remained, until recently, one dealt with by philosophers, but in the second half of the 20th century found use also in computer science, where an ontology is referred to as a special kind of information object or computational artifact (see, Gruber (1993a,b), Guarino and Giaretta (1995), and Guarino et al. (2009)). Following M. Obitko, here we categorize the definitions of ontology into four kinds (Obitko (2001), pp. 4–5): (i). (ii). (iii). (iv).

Ontology is a term in philosophy and its meaning is “theory of existence”. Ontology is an explicit specification of conceptualization. Ontology is a theory of vocabulary or concepts used for building artificial systems. Ontology is a body of knowledge describing some domain, typically a common sense knowledge domain.

After a brief discussion of (ii) - (iv), we will devote more attention to (i). Definition (ii) comes from Gruber (1993b). Its exact meaning depends on the meaning of the terms ‘specification’ and ‘conceptualization’. So, following Genesereth and Nilsson (1987), Gruber defines ‘conceptualization’ as an extensional relational structure, built from a set of entities of a certain area of interest (for example, of objects, concepts, or other entities) and a set of relationships between

* This work has been supported by the National Science Center under the project 2012/07/B/HS1/01978. https://doi.org/9783110566512-001

X | Mirosław Szatkowski them. ‘Conceptualization’, defined in this way, is “an abstract, simplified view of the world that we wish to represent for some purpose” (Gruber (1993b), p. 199). In turn, the specification of a conceptualization are the characteristics of the category, relations, functions, and (possibly) other items occurring in this conceptualization. Definition (iii), ontology as vocabulary, can be found in Chandrasekaran et al. (1999) and Mizoguchi and Ikeda (1996). More precisely, in the words of B. Chandrasekaran and others: [It] is not the vocabulary as such that qualifies as an ontology, but the conceptualizations that the terms in the vocabulary are intended to capture. Thus, translating the terms in an ontology from one language to another, for example from English to French, does not change the ontology conceptually. ..... In other words, the representation vocabulary provides a set of terms with which to describe the facts in some domain, while the body of knowledge using that vocabulary is a collection of facts about a domain. However, this distinction is not as clear as it might first appear. (Chandrasekaran et al. (1999), pp. 20–21)

R. Mizoguchi and M. Ikeda do not give a precise definition of an ontology, but they analyze the depth of the ontology used in eight levels followed by a discussion on what concrete advantages ontology has for real-world problem solving.¹

1 Mizoguchi and Ikeda (1996), p. 4: Level 1: Used as a common vocabulary for communication among distributed agents. Level 2: Used as a conceptual schema of a relational data base. Structural information of concepts and relations among them is used. Conceptualization in a data base is nothing other than conceptual schema. Data retrieval from a data base is easily done when there is an agreement on its conceptual schema. Level 3: Used as a backbone information for a user of a certain knowledge base. Levels higher than this play are ontologies which have something to do with “content". Level 4: Used for answering competence questions. Level 5: Standardization 5.1 Standardization of terminology(at the same level as Level 1). 5.2 Standardization of meaning of concepts. 5.3 Standardization of components of target objects(domain ontology). 5.4 Standardization of components of tasks(task ontology). Level 6: Used for transformation of data bases, considering the differences of the meaning of conceptual schema. This requires not only structural transformation but also semantic transformation. Level 7: Used for reusing knowledge of a knowledge base using DR information. Level 8: Used for reorganizing a knowledge base based on DR information.

Ontology of Theistic Beliefs: A Short Orientation | XI

In the case (iv), an ontology is not only the vocabulary, but the whole knowledge base – the vocabulary is used to describe this knowledge base. This approach recognizes an ontology as an inner body of knowledge, and not as the way to describe the knowledge. The typical example of this definition usage is project CYC (see, Whitten (1997)). CYC is the name of a very large, multi-contextual knowledge base and inference engine, in which knowledge is represented declaratively in the form of logical assertions. That means CYC contains simple statements of fact and rules of two kinds: (i). about what conclusions to draw if certain statements of fact are satisfied (true), and (ii). about how to reason with certain types of facts and rules. Conclusions are derived by the inference engine using deductive reasoning. CYC’s common sense knowledge can be used as the body of a knowledge base for any knowledge intensive system. In this sense, this body of knowledge can be viewed as an ontology of the knowledge base of the system (cf., Obitko (2001), pp. 7–8). And finally, in case (i), ontology is listed as a part of the major branch of philosophy known as metaphysics. But both these philosophical terms: ‘ontology’ and ‘metaphysics’ have hundreds of different systems so there is no single ontology or metaphysics, but many systems in competition. Discrepancies concerning philosophical ontology occur, according to Garbacz and Trypuz (2012), on (at least) three levels: ontological, meta-ontological, and meta-meta-ontological. It is impossible to describe here this extraordinarily rich ontological pluralism – to give an example, we list just three contemporary dictionary definitions of ontology. E. J. Lowe states, Ontology, understood as a branch of metaphysics, is the science of being in general, embracing such issues as the nature of existence and the categorial structure of reality. ... The term ‘ontology’ has some additional special uses in philosophy. In a derivative sense, it is used to refer to the set of things whose existence is acknowledged by a particular theory or system of thought: it is in this sense that one speaks of ‘the’ ontology of a theory, or of a metaphysical system as having such-and-such an ontology (for example, an ontology of events, or of material substances). In a separate, technical sense the term ’ontology’ is the official name of a logistical system created by the Polish logician Stanistaw Lesniewski – a system similar in scope to modern predicate logic and developed by him in conjunction with mereology, the formal theory of part-whole relations. (Lowe (1995), p. 634)

While B. Smith states that, Ontology as a branch of philosophy is the science of what is, of the kinds and structures of objects, properties, events, processes and relations in every area of reality. (Smith (2003), p. 155)

And finally, E. Craig says that,

XII | Mirosław Szatkowski The word ‘ontology’ is used to refer to philosophical investigation of existence, or being. Such investigation may be directed towards the concept of being, asking what ‘being’ means, or what it is for something to exist; it may also (or instead) be concerned with the question ‘what exists?’, or ‘what general sorts of thing are there?’ It is common to speak of a philosopher’s ontology, meaning the kinds of things they take to exist, or the ontology of a theory, meaning the things that would have to exist for that theory to be true. (Craig (1998), p. 117)

Already, it is easy to conclude that the notion of existence stands always in the center of a philosophical ontology. Indeed, many philosophical works are associated with the issues: 1. 2. 3.

What is existence, i.e., what can be said to exist? Can we distinguish the terms such as ‘to exist’ and ‘to have being’? Why does anything exist, rather than nothingness? This question was raised by Leibniz. Are there any shared essential characteristics of existence (being)? Are there many fundamentally different ways to exist (to have being)?

It is worth emphasizing that the concept of existence is ambiguous. U. Meixner, for example, distinguishes semantical and ontological concepts of existence. The semantical concept of existence is the concept of singular reference. Thus, “N exists” means that “ ‘N’ refers to something”. In contrast, the ontological concept of existence is characterizable without mentioning reference. There are two interpretations of “x exists”: 1. “x exists” means that x is [identical to] something, and (2) “x exists” means that x is [identical to something which is] actual. Among many of other fundamental questions, dealt with by ontology or metaontology, are: 4.

5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15.

Is existence a property? What does it mean to say something exists or does not exist? Is existence properly a predicate? Are sentences expressing the existence or nonexistence of something properly called propositions? Is existence an event, flux, or process? Or is it something static, stable, or unchanging? Is existence a genus or general class that is simply divided up by specific differences? How many levels of existence or ontological levels are there? And what constitutes a “level"? How is existence related to time and space? What kind of beings are time and space? Are they beings at all or something else? What is a thing? Into what categories, if any, can we sort existing things? How do things persist and change over time? What are the various modes of being of entities? Are all entities objects? Which entities, if any, are fundamental? What features are the essential, as opposed to merely accidental, attributes of a given object? When does an object go out of existence, as opposed to merely changing? What constitutes the identity of an object?

Ontology of Theistic Beliefs: A Short Orientation |

XIII

16. What is a physical object? Can one give an account of what it means to say that a physical object exists? 17. Can one give an account of what it means to say that a non-physical objects (such as times, numbers, souls, deities, values, imaginative objects) exist? 18. How do the properties of an object relate to the object itself? 19. Do physical properties actually exist? 20. Do beings exist other than in the modes of objectivity and subjectivity, i.e. is the subject/object split of modern philosophy inevitable?

2 What is belief? Or, what is it to believe? Moreover, what is the nature of belief? In the last approximately two and a half centuries, according to Hacker (2004), three main opinions have prevailed in investigations of believing: 1.

2. 3.

Believing that p is a special kind of feeling associated with the idea that p or the proposition that p. The representatives of this view are, for example, D. Hume, W. James, B. Russell and F. P. Ramsey; To believe that p is to be in a certain kind of mental state. The representatives of this view are, for example, D. Davidson, J. Searle and T. Williamson; To believe that p is to have a certain sort of disposition. A. Bain, R. B. Braithwaite, G. Ryle and W. V. O. Quine, for example, are some of the representatives of this idea. (p. 185)

On the question of the difference between believing and not believing something to be so, Hume states that this difference determines the presence of a feeling in the first case and its absence in the second case. He writes: ‘Belief’ consists merely in a certain feeling or sentiment; in something that depends not on the will, but must arise from certain determinate causes and principles, of which we are not masters. When we are convinc’d of any matter of fact, we do nothing but conceive it, along with a certain feeling, different from what attends the mere reveries of the imagination. And when we express our incredulity concerning any matter of fact, we mean, that the arguments for the fact produce not that feeling. (Hume (1976), p. 624)

In turn, he characterizes the feeling as follows: An idea assented to feels different from a fictitious idea, that the fancy alone presents to us: And this different feeling I endeavour to explain by calling it a superior force, or vivacity, or solidity, or firmness, or steadiness. This variety of terms, which may seem so unphilosophical, is intended only to express that act of the mind, which renders realities more present to us than fictions, causes them to weigh more in thought, and gives them a superior influence on the passions and the imagination. Provided we agree about the thing, ’tis needless to

XIV | Mirosław Szatkowski dispute about the terms. ... I confess, that ’tis impossible to explain perfectly this feeling or manner of conception. We make use of words, that express something near it. But its true and proper name is belief, which is a term that everyone sufficiently understands in common life. And in philosophy we can go no further, than assert, that it is something felt by the mind, which distinguishes the ideas of the judgement from the fictions of the imagination. (Hume (1976), p. 629)

Turning to W. James, we quote: As regards the analysis of belief, i.e. what it consists in, we cannot go very far. In its inner nature, belief, or the sense of reality, is a sort of feeling more allied to the emotions than to anything else. ... Belief, the sense of reality, feels like itself — that is about as much as we can say. ... This attitude is a state of consciousness sui generis, about which nothing more can be said in the way of internal analysis. (James (1890), pp. 283–287)

Distinguishing three kinds of belief: memory, expectation and bare assent, B. Russell states that each of them is “a certain feeling or complex of sensations, attached to the content believed” (Russell (1921), p. 233). He adds: “I, personally, do not profess to be able to analyse the sensations constituting respectively memory, expectation and assent, but I am not prepared to say that they cannot be analysed.” (p. 250) And in this same spirit, F. P. Ramsey writes: “The mental factors of ... a belief [are] words spoken aloud or to oneself or merely imagined, connected together and accompanied by a feeling or feelings of belief or disbelief ...” (Ramsey (1931), p. 144). Turning now to the second view, that beliefs are certain kinds of mental states, D. Davidson claims: We know what states of mind are like, and how they are correctly identified; they are just those states whose contents can be discovered in well-known ways. If other people or creatures are in states not discoverable by these methods, it cannot be because our methods fail us, but because those states are not correctly called states of mind—they are not beliefs, desires, wishes, or intentions. (Davidson (2001a), p. 40)

And The only object required for the existence of a belief is a believer. Having a belief is not like having a favorite cat, it is being in a state; and being in a state does not require that there be an entity called a state that one is in. (Davidson (2001b), p. 74)

J. Searle has also endorsed the view that beliefs are mental states. In his words: What exactly is the character of the neurophysiological processes and how exactly do the elements of the neuroanatomy – neurons, synapses, synaptic clefts, receptors, mitochondria, glial cells, transmitter fluids, etc. – produce mental phenomena? And what about the

Ontology of Theistic Beliefs: A Short Orientation |

XV

great variety of our mental life – pains, desires, tickles, thoughts, visual experiences, beliefs, tastes, smalls, anxiety, fear, love, hate, depression, and elation? How does neurophysiology account for the range of our mental phenomena, both conscious and unconscious? Such questions form the subject matter of the neurosciences, ... (Searle (1992), p. 1)

Similarly, T. Williamson claims that beliefs are mental states, but it is worth emphasizing his argument that beliefs are not wholly internal. That is, the internal can not be considered as a complete determinant of mentality. This follows from a simple argument based on natural language semantics. On the one hand, believing is pervaded by language, but on the other hand, the referents of terms within natural languages are determined at least in part by environmental conditions, therefore, believing is determined at least in part by environmental conditions (cf., Williamson (2000), pp. 51–54). Finally, let us begin to discuss the idea that to believe that p is a disposition from a short presentation of the views of A. Bain. He defines belief in terms of behaviour. In his words: Belief has no meaning except in reference to our actions ... no mere conception that does not directly or indirectly implicate our voluntary exertions can ever amount to the state in question. (Bain (1859), p. 568)

And I consider the correct view to be, that belief is a primitive disposition to follow out any sequence that has once been experienced, and to expect the result. It is a fact or incident of out Intellectual nature, although dependent as to its energy upon our Active or Emotional tendencies. (Bain (1872), p. 100)

But, as R. B. Braithwaite notes, Bain partly altered his view: And in the third and fourth editions of The Emotions and the Will (1875 and 1899), though we are told that “Belief is essentially related to Action, that is, volition”, and that “preparedness to act upon what we affirm is admitted on all hands to be the sole, the genuine, the unmistakable criterion of belief” (p. 505), a tendency to action is no longer asserted to be the differentia of belief. (Braithwaite (1932), p. 133)

R. B. Braithwaite argues that the phraze ‘To believe that p’ means the conjunction of the two propositions: (1). To entertain the proposition that p, and (2). To have a disposition to act as if p were true. And similarly, ‘To have a disposition to believe p’ means both: (1). To have a disposition to entertain p, and (2). To have a disposition to act as if p were true. And he adds:

XVI | Mirosław Szatkowski In either case the former proposition is one about my mental experience and the second one about my physical behaviour. The former is subjective or phenomenological, the second objective or behavioristic. It is the latter proposition which in my view is the differentia of actual belief from actual entertainment and of dispositional belief from dispositional entertainment. (Braithwaite (1932), p. 132)

To continue, Ryle (1949), pp. 116–118, compares two verbs: ‘know’ and ‘believe’. Both are dispositional verbs, but of quite disparate types. ‘Know’ is a capacity verb, whereas ‘believe’ is a tendency verb which “does not connote that anything is brought off or got right”. ‘Belief’, as he thinks, can be qualified by adjectives such as ‘obstinate’, ‘wavering’, ‘unswerving’, ‘unconquerable’, ‘stupid’, ‘fanatical’, ‘whole-hearted’, ‘intermittent’, ‘passionate’ and ‘childlike’. For Ryle, beliefs can be inveterate, slipped into and given up – like habits; can be blind and obsessing – like partisanships, devotions and hopes; can be unacknowledged – like aversions and phobias; can be contagious – like fashions and tastes; can be induced by tricks – like loyalties and animosities. One can be urged or entreated not to believe things, and one may try, with or without success, to cease to do so. These dictions or their negatives, however, are not applicable to knowing, since to know is to be equipped to get something right and not to tend to act or react in certain manners, as Ryle says. And finally, W. V. O. Quine also endorses the dispositional account of belief. In his words: A belief, in the best and clearest case, is a bundle of dispositions. It may include the disposition to lip service, a disposition to accept a wager, and various dispositions to take precautions, or to book a passage, or to tidy up the front room, or the like, depending on what particular belief it may be. (Quine (1990b), p. 20)

Some philosophers have joined two or even all three of the views 1. - 3. In Hacker’s states, Others have woven two or more strands together, arguing, for example, that to entertain a thought which one believes to be true, sometimes called “occurrently believing something”, is indeed a feeling, but that non-occurrently believing something is a disposition occurrently to have the belief feeling. Others have suggested that to believe is indeed to be in a certain mental state, but stressed that the state in question is a dispositional state. (Hacker (2004), p. 185)

According to Hacker, ‘believing that p’ is neither a feeling, nor a certain kind of mental state, nor a certain sort of disposition. And importantly, believing that p is not the same as believing the proposition that p. He says: To believe that p is to believe things to be so; to believe the proposotion that p is to believe things to be as the proposition that p describes them as being. (Hacker (2004), p. 185)

Ontology of Theistic Beliefs: A Short Orientation |

XVII

Hacker (2004) gives seven arguments for the thesis that believing that p is not a special kind of feeling, ten arguments for believing for the thesis that believing that p is not a certain mental state, and ten arguments for the thesis that believing that p is not a certain sort of disposition. What, then, is believing that p for Hacker? Unfortunately, he does not give any answer to this question. I doubt if there is any categorial term under which believing can illuminatingly be subsumed. However, one should examine: the needs which the concept of belief satisfies; the purposes which it fulfils; and the contexts of its use.

3 Broadly speaking, theism is defined as the belief in the existence of God or gods. Theists insist that God (or gods) exists (exist) realistically, objectively, and independently of human thought. Following the example of R. Dawkins, one could distinguish de facto theism from strong theism.² De facto theism claims that the existence of God or gods cannot be proven or disproven by scientific methods.³ But this is no problem because faith is not a product of reason and always involves some risk. Strong theism asserts that the existence of God or gods can be proven. However, some theists argue that only some of the arguments for the existence of God have scientific value. The view that there is only one God is called monotheism, and the view that there is more than one God is called polytheism. The denial of theism is called atheism. For example, religious beliefs such as Christianity, Islam, and Judaism are monotheistic, whereas Hinduism is a polytheistic. A comprehensive discussion of all possible interpretations of all these notions would be a hopeless task in this limited space. So we focus here only on monotheistic beliefs. Of course, to say that God exists is to say that there is something that has some divine attributes. And the concept of God is thought of as a cluster or a set of properties that are tied together in some way; more precisely, this entire cluster of divine attributes must be logically consistent. There is a large number of different views on the nature of God, and significantly, even among monotheistic religions there is no consensus in this regard. Therefore, the attributes of God we will discuss here are those which have been developed in the Western theistic tradition. Likely, the most commonplace view of God within this tradition is

2 Cf., Dawkins (2006) 3 Cf., McGrath (2005) and Barackman (2001)

XVIII | Mirosław Szatkowski to identify him as the Supreme Being. St. Anselm expresses this in the definition of God as something greater than which nothing can be conceived. This means that if God has any attributes, he has them to a maximal degree. Usually, God is described with the help of the so called ‘omni-’ atributes: omniscience, omnipotence, omnibenevolence and omnipresence. Omniscience is usually understood as being all-knowing. Thus, to say that God is omniscient is to say that he knows absolutely everything. But what does it mean to say that God knows absolutely everything? In the most radical form, the doctrine of divine omniscience is thought of as comprising the following five claims⁴: (1) God’s knowledge is complete, i.e., God knows all truths (about God himself, creation, logic, and mathematics, etc.); (2) God’s knowledge is sound, i.e., everything that God knows is true; (3) God’s knowledge is tenseless, i.e., God knows the past, present, and future simultaneously; (4) God’s knowledge is necessary, i.e., God necessarily knows whatever he knows; and (5) God’s know-ledge also ranges over unrealized possibilities, i.e., God knows all things that could have come to pass but will not. Omnipotence is generally understood as the power to do anything. But what the word ‘anything’ involves is highly contentious?⁵ Does this word also include a logical impossibility, for example, could God make 2+2 = 5? And there are other dilemmas, such as the paradox of the stone: Could God create a rock so heavy that he cannot lift it?, and the paradox of sovereignity: Could a sovereign God create a law that binds himself? There is a view that the limits of logical possibility or impossibility are no limitations on God’s power. And God cannot do anything that is absurd or self-contradictoty. Omnibenevolence is understood as perfect and infinite goodness, depending on precisely how the word ‘good’ is recognized. Thus, saying that God is omnibenevolent means that he is an absolutely (perfectly) good being. God’s omnibenevolence is, however, problematic. The existence of evil in the world seems to suggest that omnibenevolence, omnipotence and omniscience are incompatible. Because if God knows evil things are happening, why does he allow them to happen if he is omnibenevolent and omnipotent? And finally, omnipresence is understood as the ability to be present everywhere, i.e., in all places at once, at the same time. In this sense, God’s omnipresence means that there is no place and time to which God’s knowledge, power and goodness could be limited.

4 Cf., Szatkowski (2014). 5 Thomas Aquinas writes: All confess that God is omnipotent; but it seems difficult to explain in what His omnipotence precisely consists: for there may be doubt as to the precise meaning of the word ‘all’ when we say that God can do all things. (Aquinas Thomas (1988), I.25.3)

Ontology of Theistic Beliefs: A Short Orientation |

XIX

Among other attributes, it is traditionelly claimed that God is the creator and the sustainer for everything that exists. God is a necessary being, unlike all other beings whose existence is contingent. He depends on nothing, i.e., nothing can bring God into existence nor change anything in him nor end his existence. Many philosophers and theologians argue that God is ‘personal’ rather than a person – he has attributes that are strictly related to being a person, but is not a person because he has no body. In Catholic doctrine, God is one in three persons: the Father, the Son (Jesus), and the Holy Spirit, with which one can take contact in a mystical experience.

4 We can now turn to issues involving the ‘ontology of theistic beliefs’. This ontology can be directed at a believer or what he believes in, but we will limit ourselves below to the first case. Indeed, some form of the believer’s religiosity is bound up with each theistic belief. Depending on how the concept of ‘belief’ is understood, ontological questions of the believer’s religiosity can be divided into three groups. A. Believing that p is a special kind of feeling associated with the idea that p or the proposition that p. The basic questions here include: 1. 2. 3.

4. 5. 6. 7. 8.

What differentiates religious feeling from other types of feeling? Are there necessary and sufficient conditions such that by their simple enumeration religious feelings might be distinguish from other ones? In the case of a negative answer to question 2, does that mean that religious feelings are not governed by logical conditions at all? Or, can one find some discriminants which in some sense count only for or only against the use of a religious feeling? Is the distinction between religious and non-religious feelings metaphysical? Are religious feelings dependent upon non-religious feelings for their existence and nature? Does religiosity grounded in a theistic belief have a distinct ontology from religiosity without a theistic belief? Are all religious feelings subjective? Otherwise, are there also objective religious feelings? Is there a certain type of religious feelings that supercedes others?

B. To believe that p is to be in a certain kind of mental state. Some important issues, for example, are: 1.

The main problem in the philosophy of mind is the so-called mind-body problem: what is the relationship between minds (mental states or processes) and bodies (bodily states

XX | Mirosław Szatkowski

2. 3. 4. 5. 6.

or processes)? Consequently, the question arises: what is the relationship between (theistic) belief’s states or processes, on the one hand, and bodily states or processes, on the other hand. What differentiates belief states or processes from other types of mental states or processes? Are there necessary and sufficient conditions such that by their simple enumeration one might distinguish belief states or processes from other mental states or processes? Is the distinction between belief states or processes, on the one hand, and other mental states or processes, on the second hand, metaphysical? Is it a reasonable assumption that all people have mental states or processes that directly represent specific beliefs? An important issue in the philosophy of mind is the problem of causal exclusion: Given that every physical event that has a cause has a physical cause, how is a mental cause also possible? Consequently, one can also ask about the causes of beliefs – what are the causes of beliefs? Do they have a metaphysical character?

C. To believe that p is to have a certain sort of disposition. Unfortunately, the term ‘disposition’ is vague. It is generally accepted that dispositions are recognizable by their characteristic manifestations under some stimulus conditions. But again, what are the manifestations and the stimulus conditions of a disposition? The answers to these questions determine the answer to the question: What are (theistic) belief dispositions? More specific questions here, for example, are: 1. 2.

3. 4.

Are there necessary and sufficient conditions such that by their simple enumeration one might distinguish (theistic) belief predicates from other dispositional predicates? An important issue was noted by I. Levi and S. Morgenbesser: “Analyses of beliefs as dispositions have sometimes been invoked in attempts to determine what significant relation, if any, is obtained between belief predicates taken in an occurrent and in a non-occurrent sense.” Where, as they explain: The distinction between occurrent and non-occurrent terms is not always drawn in a uniform way. It may contrast between (a) terms applying to entities localizable in time (or space and time) and those which do not so apply : (b) terms which apply entities in time (or space and time) for only brief periods and those which do not; (c) terms which describe changes in the conditions of objects to which they apply and those which do not (“becomes red” vs. “is red”) ; (d) terms which describe evanescent changes and those which do not (“flashes” vs. “beams”). Since disposition terms are usually occurrent in sense (a) and may be so in sense (b) (an iron bar may be magnetic for a brief period) and since most authors treat disposition terms as non-occurrent, we shall usually take the distinction in sense (c) or (d) without distinguishing between them. (Levi and Morgenbesser (1964), p. 221) What metaphysical character do changes in (theistic) belief have? What is the role of (theistic) belief predicates in explanations of human behavior? Does this role also have a metaphysical character?

Ontology of Theistic Beliefs: A Short Orientation |

XXI

The reader may find short descriptions of the articles in this volume in the abstracts placed at the end of the book.

Bibliography Aquinas Thmas (1988), Summa Theologiae, Milan: Editiones Paulinae. Bain, A. (1859), The Emotions and the Will, London: John W. Parker and Son. Bain, A. (1872), Mental and Moral Science, London. Barackman, F. H. (2001), Practical Christian Theology: Examining the Great Doctrines of the Faith, Kregel Academic. Braithwaite, R. B. (1932), “The Nature of Believing”, in Proceedings of the Aristotelian Society: 33, 129–146. Chandrasekaran, B., Josephson, J. R. and Benjamins, R. V. (1999), “What are Ontologies, and Why Do We Need Them?”, in IEEE Transactions on Intelligent Systems: 14(1), 20–26. Craig, E. (1998), “Ontology”, in Routledge Encyclopedia of Philosophy, edited by E. Craig, Routledge, 117–118. Davidson, D. (2001a), “The Myth of the Subjective”, in Subjective, Intersubjective, Objective, by D. Davidson, Oxford: Clarendon Press, 39–52. Davidson, D. (2001b), “Indeterminism and Antirealism”, in Subjective, Intersubjective, Objective, by D. Davidson, Oxford: Clarendon Press, 69–84. Dawkins, R. (2006), The God Delusion, Great Britain: Bantam Press. Garbacz, P. and Trypuz, R. (2012), Ontologie poza Ontologią – Studium Metateoretyczne u Podstaw Informatyki [Ontologies Beyond Ontology - a Metatheoretical Study at the Basis Computer Science], Lublin: Wydawnictwo KUL. Genesereth, M. R. and Nilsson, N. J. (1987), Logical Foundation of Artificial Intelligence, Los Altos, California: Morgan Kauffmann. Gruber, T. R. (1993a), “Toward Principles for the Design of Ontologies Used for Knowledge Sharing”, in Formal Ontology in Conceptual Analysis and Knowledge Representation, edited by N. Guarino and R. Poli, Kluwer Academic Publishers. Later published in International Journal of Human Computer Studies: 43(5-6), 907–928. Gruber, T. R. (1993b), “A Translation Approach to Portable Ontologies”, in Knowledge Acquisition: 5(2), 199–220. Guarino, N. and Giaretta, P. (1995), “Ontologies and Knowledge Bases: Towards a Terminological Clarification”, in Towards Very Large Knowledge Bases: Knowledge Building and Knowledge Sharing, edited by N. Mars, Amsterdam: IOS Press, 25–32. Guarino, N., Oberle, D. and Staab, S. (2009), “What Is an Ontology?”, in Handbook on Ontologies. International Handbooks on Information Systems, edited by S. Staab and R. Studer, Berlin/Heidelberg: Springer-Verlag, 1–17. Hacker, P. M. S. (2004), “Of the Ontology of Belief”, in Semantik und Ontologie, edited by M. Siebel and M. Textor, Frankfurt: Ontos Verlag, 185–222. Hume, D. (1976), Treatise of Human Nature, Oxford: Oxford University Press. James, W. (1890), The Principles of Psychology, New York: Henry Holt and Co. Levi, I. and Morgenbesser, S. (1964), “Belief and Disposition”, in American Philosophical Quarterly: 1(3), 221–232.

XXII | Mirosław Szatkowski Lowe, E. J. (1995), “Ontology”, in The Oxford Companion to Philosophy, edited by T. Honderich, New York: Oxford University Press. McGrath, A. E. (2005), Dawkins’ God: Genes, Memes, and the Meaning of Life, Wiley-Blackwell. Meixner, U. (forthcoming), “No Life without Time”, in God, Time, Infinity, edited by M. Szatkowski, Berlin/Boston: Walter de Gruyter, 101–109. Mizoguchi, R. and Ikeda, M. (1996), “Towards Ontology Engineering”, in Technical Report, AI-TR-96-1, The Institute of Scientific and Industrial Research, Osaka University. Obitko, M. (2001), “Ontologies Description and Applications”, in Gerstner Laboratory for Intelligent Decision Making and Control, Czech Technical University in Prague, Series of Research Reports: GL 126/01. Price, H. H. (1969), Belief , London: Allen and Unwin. Quine, W. V. O. (1990a), The Pursuit of Truth, Cambridge: Harvard University Press. Quine, W. V. O. (1990b), Quiddities, London: Penguin Books. Quine, W. V. O. (1995), From Stimulus to Science, Cambridge: Harvard University Press. Ramsey, F. P. (1931), “Facts and propositions”, in The Foundations of Mathematics and Other Logical Essays, London: Kegan Paul, Trench, Trubner and Co. Ltd. Russell, B. (1921), Analysis of Mind, London: Allen and Unwin. Ryle, G. (1949), The Concept of Mind, London: Hutchinson. Searle, J. R. (1983), Intentionality, Cambridge: Cambridge University Press. Searle, J. R. (1992), The Rediscovery of the Mind, Cambridge: MIT Press. Smith, B. (2003), “Ontology”, in Blackwell Guide to the Philosophy of Computing and Information, edited by L. Floridi, Oxford: Blackwell, 155–166. Szatkowski, M. (2014), “How to Think about the Correctness of Theistic Belief”, in Metaphysica: 15(1), 47–68. Whitten, D. (1997), The Unoflcial, Unauthorized CYC Frequently Asked Questions, http://www.robotwisdom.com/ai/cycfaq.html. Williamson, T. (2000), Knowledge and its Limits, Oxford: Oxford University Press.

Chris Daly

Agnosticism and the Balance of Evidence 1 Introduction The agnostic suspends judgement as to whether there is a God: she neither believes that there is a God nor believes that there is no God.¹ She and the atheist agree about something: they agree that there is no good evidence that there is God. Some atheists think there is good evidence that there is no God – they cite the problem of evil or argue that the concept of God is incoherent. The agnostic disagrees: she thinks that there is no good evidence that there is no God. Some atheists think that if there is no good evidence that there is a God, that fact itself provides good evidence that there is no God. This contention was memorably expressed in Russell’s tale of the cosmic teapot: Many orthodox people speak as though it were the business of sceptics to disprove received dogmas rather than of dogmatists to prove them. This is, of course, a mistake. If I were to suggest that between the Earth and Mars there is a china teapot revolving about the sun in an elliptical orbit, nobody would be able to disprove my assertion provided I were careful to add that the teapot is too small to be revealed even by our most powerful telescopes. But if I were to go on to say that, since my assertion cannot be disproved, it is intolerable presumption on the part of human reason to doubt it, I should rightly be thought to be talking nonsense. (Russell (1952), pp. 547–548)

Russell’s teapot hypothesis at once illustrates a principle and provides an analogy. The principle is that, at least in certain circumstances, the lack of evidence for a hypothesis H provides evidence against H. The analogy is that, just as the hypothesis that there is a cosmic teapot is not supported by any evidence and therefore there is evidence that it is false, so too the hypothesis that there is a God is not supported by evidence and therefore there is evidence that it is false. The challenge to the agnostic is evident. She concedes that there is no all things considered evidence that there is a God yet she does not believe that there is no God. Is she thereby irrational? What relevant difference is there between the cosmic teapot hypothesis and the God hypothesis? In this paper I will explore the agnostic’s position.

1 By ‘God’ is meant here an uncreated free agent who created and sustains the universe and who is omnipotent, omniscient and omnibenevolent. https://doi.org/9783110566512-002

2 | Chris Daly

2 Agnosticism Agnosticism has two components. It consists of a lack of belief and a reason for that lack of belief. An agnostic with respect to the proposition that p neither believes that p nor believes that not-p. Her reason for her lack is belief is that there is (or so she believes) insufficient evidence on which to believe that p or to believe that not-p. We can deepen our understanding of what agnosticism is by asking a series of questions and answering them by means of introducing a number of distinctions. What is the agnostic an agnostic about? Here we can distinguish between local and global forms of agnosticism (following Le Poidevin (2010), pp. 10–13). Someone might be agnostic with respect to the propositions drawn from a particular subject-matter. An agnostic about the existence of God is someone who is agnostic with respect to all propositions which imply that God exists. An agnostic about the future is agnostic with respect to all propositions about the future. An agnostic about morality is agnostic with respect to all propositions about what is morally right or wrong. And so on. Each of these is a species of local agnosticism. One might be agnostic with respect to some subject-matters but not others. By contrast, a global agnostic would be someone who is agnostic with respect to all subject-matters and hence to all propositions. What is the status of the evidence? The agnostic was characterised above as believing that there is insufficient evidence on which to believe that p or on which to believe that not-p. ‘Evidence’ is used in an unrestricted sense: evidence that p is any consideration that provides epistemic reason to believe that p. So understood a piece of evidence may be something physical, such as a fingerprint, something mental, such as a perception, or something propositional, such as an argument. In claiming that there is no evidence for or against p, the agnostic is making a fallible claim. She might have overlooked certain evidence or she might have mistakenly judged that something is not relevant evidence. So denying that there is evidence for or against p may be as controversial as claiming that there is such evidence. There are a number of options which an agnostic might take with respect to the issue of the evidence before her. One option is that she may believe that there is some evidence for p, or for not-p, or for both, but she will further believe that what evidence there is fails to raise her degree of belief so that she thinks that it is more likely than not that p (or so that she thinks that it is more likely than not that not-p) (cf. Draper (2001)). In such a case, either she thinks that the evidence

Agnosticism and the Balance of Evidence | 3

each way is balanced or she is unable to say where the balance of evidence lies.² A second option is that she may believe that there is simply no evidence for p or for not-p. The two preceding options allow that the evidential situation in which the agnostic finds herself may change. Each option allows that evidence might come in which would make her believe that p or which would make her believe that notp. This would be to take it that it is contingent that she lacks sufficient evidence either way. A third option, however, would be for her to believe that there cannot be evidence either way. This option would be to take it that it is necessary that she lacks sufficient evidence either way. I take this to be the view Kant took in his Critique of Pure Reason with respect to the existence of God.³ Call it Kantian agnosticism. (Kant held a further view: that practical reason enjoins us to think and act as if God exists. This further view of his will not concern us). A fourth option is that, although there is a wealth of evidence for and against p, the evidence is too disputed. For instance, many people have had religious experiences in which they are apparently aware of God. The content of such experiences, however, is compatible with many different hypotheses about the source of those experiences, some theistic, some not. We then need to distinguish agreed evidence from disputed evidence. On the one hand, what is agreed evidence between all parties to the dispute is that many people have had experiences with the above content. That evidence, however, seems not to support a theistic hypothesis more than it supports some non-theistic hypothesis. On the other hand, whether there is evidence that many people have experiences which are in fact experiences of God is disputed. The distinction between agreed and disputed evidence does not coincide with the distinction between evidence that is open to all and evidence that is open only to some.⁴ It should be granted that the identification and proper weighing of evidence that is relevant to a given debate very often requires specialised training and competence. Identifying something as an X-ray image and being able to interpret it correctly is a relatively specialised task and not something open to all. Nevertheless, the information it provides is evidence that closely bears on the question of whether you have a broken leg. In the present case, the relevant training and competence is primarily training and competence in philosophy. By ‘parties to the dispute’, then, we mean suitably spe-

2 For simplicity by ‘having evidence’, I assume both that one has evidence and that one has evaluated this evidence. These would come apart, for instance, in a situation in which I have x-rays of a body but I am not sufficiently trained to be able to interpret them. 3 See the Transcendental Dialectic’s Ideal of Reason in Kant (1781). 4 Shalkowski (1989), p. 4, invokes the latter distinction in his discussion of evidence for or against the existence of God.

4 | Chris Daly cialised and informed parties. Professional philosophers who meet the standards of proficiency for the discipline – as indicated by such factors as that they have secured tenure, are employed as journal referees, and so on – persistently disagree about whether the content of people’s religious experiences is evidence that there is a God. For that reason the content of people’s religious experiences is disputed evidence that there is a God. One further matter on the topic of evidence should be noted. Some theists claim that their belief that there is a God is a rational belief because, they claim, God created human beings in such a way that, in suitable circumstances, human beings will acquire (inter alia) a rational belief that there is a God (Plantinga (1983)). It follows from this view that someone’s belief that there is a God can be rational although not based on evidence. The agnostic need not claim that such a view is false. Instead, she suspends judgement about it. Part of her reason is consistency: if she suspends judgement about whether there is a God, a fortiori she suspends judgement about the fashion in which God has created and fashioned us. Another part of her reason is evidential: just as she thinks that there is no evidence that there is a God, so too she thinks that there is no evidence that the above view is true. What is the epistemic standing of agnosticism? The agnostic may believe that her lack of evidence makes it epistemically permissible neither to believe that p nor to believe that not-p. On this option, given what evidence (if any) she has, it is reasonable for her neither to believe that p nor to believe that not-p, but this is not to say that it would be unreasonable for someone to believe that p or for someone to believe that not-p, given the same evidence. Alternatively, the agnostic may believe that her lack of evidence makes it epistemically obligatory for her neither to believe that p nor to believe that not-p. On this option, given what evidence she has, it is reasonable for her neither to believe that p nor to believe that not-p, and it would be unreasonable for someone to believe that p or for someone to believe that not-p, given the same evidence (cf. Oppy (1994)). If the agnostic thought both that there was no evidence that p and that there was no evidence that not-p, or if she thought that there cannot be evidence of either sort, then she would take this second option: she would think that it is epistemically obligatory for her neither to believe that p nor to believe that not-p. If, however, she thought that there was some evidence that p, say, then on some views of reasonable belief (such as Rosen (2001)) there can be cases in which two people have the same body of evidence such that this evidence makes it reasonable for one person to believe that p although not for another. Such views are controversial (Feldman (2007)); I simply note them here for the record. Who is the agnostic speaking for? The agnostic may be speaking for herself or she may be speaking for a larger epistemic community. In the first case, she

Agnosticism and the Balance of Evidence | 5

may be reporting that she is agnostic given the evidence available to her while allowing that other people may have different evidence. If she does not know what evidence they have, she cannot speak for them and say what attitudes it would be reasonable for them to take with respect to p or with respect to not-p. For instance, suppose the agnostic thinks that it is a contingent fact that there is not enough evidence for her to believe that p or to believe that not-p. Still, she might allow that other inquirers are in a better epistemic situation than she is. Alternatively, the agnostic might take herself to be speaking for a larger epistemic community. She might take it that she is not simply talking on the basis of the evidence which is available to her but on the basis of the evidence available to her epistemic community. This epistemic community might be her research team, her intellectual culture, or even the human race as a whole. If, for instance, an agnostic thought that there could be no evidence for or against a given proposition p, then she would presumably be speaking not just about her own epistemic situation but that of every human being. For suppose that someone other than the agnostic had evidence that p. Then that person’s testimony that p could provide the agnostic with some evidence that p.

3 Russell’s teapot hypothesis Russell’s own intention in presenting the teapot hypothesis may have differed from what a proponent of the above principle and analogy takes it to show. Russell’s intention may have only been to argue that in a debate about whether some existential hypothesis H is true, the onus of proof is not automatically on someone who doubts whether H. He may also have intended to argue that in such a debate the onus of proof is on someone who claims that H. At any rate, some philosophers have independently found such claims convincing. Consider the following passage: for any proposition, p, there is an onus of proof on the person who claims that it is known that p is true (or false) which does not rest on the person who denies this. Thus, the person who claims that it is known that there is life on Mars is epistemically obliged to adduce evidence for his claim, even if there is no evidence that there is no life on Mars, while it is surely false that the person who claims that it is not known that there is life on Mars is equally epistemically obliged to adduce evidence for his position, if there is no evidence that there is life there. (Dore (1982), p. 505)

The issue concerning where (if anywhere) the onus of proof lies in a given debate is, however, more complex than this. The issue is also sensitive to the subject-

6 | Chris Daly matter under question. For example, Dore knows what his name is. If someone – a tax officer or some other bureaucrat – were to deny that his name is ‘Dore’, the onus of proof is not on Dore to adduce evidence that he knows his own name. Indeed, the onus rests on the denier for perversely denying that Dore knows what he takes himself to know. The fact that where the onus of proof lies is sensitive to the subject-matter under debate is veiled in the passage quoted above by its choice of subject-matter. In both every day and philosophical contexts, it is controversial whether there is life on Mars and so it is controversial whether anyone knows whether there is life on Mars or not. By contrast, in an everyday or philosophical context it is uncontroversial that Dore knows his own name. We enter such contexts with (what we take to be) a vast stock of knowledge and reasonable belief, some personal and some mutual. Given this presumed basis of knowledge and reasonable belief, the onus of proof falls on someone either when she seeks to augment it (as when she claims that there is life on Mars) or when she seeks to diminish it (as when she denies that Dore knows his own name). In the latter case, in an everyday context the denier might cite evidence of Dore’s apparent amnesia, and in a philosophical context the denier might cite Descartes’ dreaming argument as evidence that we know very little of what we take ourselves to know. Be that as it may, the point is that they would need to cite some such considerations to support their denial.⁵ I assume that Russell is claiming more than that the onus of argument is on anyone who asserts that there is a cosmic teapot. I take him also to be claiming that it is reasonable to believe that that hypothesis is false, that we do not need evidence that it is false, and that we should revise our belief only if we were given evidence that it is true (Le Poidevin (2010), p. 42, and Van Inwagen (2012), p. 22). Just how Russell’s argument is to be spelt out is not obvious. For definiteness, let us work with the following interpretation. Even if it does not exactly reflect the argument Russell had in mind, the interpretation is of interest in its own right and it might capture what many philosophers had in mind when they endorsed what they took to be Russell’s argument. The interpretation takes Russell to be running the following argument that no cosmic teapot exists: (T1) Consider the hypothesis that between the Earth and Mars there is a china teapot in orbit around the sun: call this hypothesis teapot. (T2) There is no evidence that teapot is true. (T3) So the probability of teapot is very close to zero. (T4) So we should believe that teapot is false.

5 For further discussion of the issue of the context-dependent nature of burdens of proof, see Shalkowski (1989) especially pp. 1–9.

Agnosticism and the Balance of Evidence | 7

By analogy, Russell is then arguing in the following way that God does not exist: (G1) Consider the hypothesis that there is a God: call this hypothesis theism. (G2) There is no evidence that theism is true. (G3) So the probability of theism is very close to zero. (G4) So we should believe that theism is false.

4 Van Inwagen’s response Teapot is one of a number of whimsical hypotheses which have achieved currency in academic and popular discussions of the existence of God. The strategy behind their introduction is to claim that there is no more reason to accept theism than there is to accept any of the whimsical hypotheses, and so theism is taken to be no more probable than any of the whimsical hypotheses. Since each of the whimsical hypotheses itself has very low probability, theism will also have a low probability. Accordingly, we have a list of hypotheses such as the following: A B C D E

God exists. The cosmic teapot exists. Santa Claus exists. The Flying Spaghetti Monster exists. The Great Pumpkin exists.

For instance, the following passage forms part of Scriven’s presentation of his case for disbelief that God exists: Why do adults not believe in Santa Claus? Simply because they can now explain the phenomena for which Santa Claus’s existence is invoked without any need for invoking a novel entity. ... As we grow up, no one comes forward to prove that [Santa Claus] does not exist. We just come to see that there is not the least reason to believe he does exist. ... Santa Claus is in the same position as fairy godmothers, wicked witches, the devil, and the ether ... the proper alternative when there is no evidence is not mere suspension of belief [in Santa Claus], it is disbelief. (Scriven (1966), p. 103)

Van Inwagen dismisses C-E on the ground that the entities in question or their supposed activities are physically impossible. (This was the reason given in van Inwagen’s lecture ‘Russell’s Teapot’ given at Syracuse University on 22nd September 2012 and available online.⁶ His print publication (Van Inwagen (2012)) does

6 See http://andrewmbailey.com/pvi/Teapot.pdf at pp. 8–9.

8 | Chris Daly not contain this material). Nothing can travel at sufficient speed to visit every home on Christmas morning, and something composed solely of spaghetti or pumpkin cannot fly and cannot be sentient. Fair enough. But Hume offered various theological hypotheses which are rivals to theism, which each have low prior probability, but which cannot be excluded on the grounds that they are physically impossible (Hume (1779), part 5, paragraph 12): F G

‘some infant Deity, who afterwards abandoned [the world], ashamed of his lame performance’; ‘[the world is] the production of old age and dotage in some superannuated Deity’.

We will return to these examples below. Van Inwagen claims that Russell’s argument against teapot goes through only if a further premise is added: (T2′ ) Teapot has a vanishingly small prior probability of being true.

(T2′ ) says that teapot has a low probability of being true prior to any considerations pertaining to evidence. Why is this? There are countless possible stories about the origin of the teapot and how it came to occupy an orbit between the Earth and Mars. Each of these origin stories, with its tales of aliens manufacturing a teapot and putting it into orbit, has a very low probability. The aggregate probability of all of these origin stories is also very low. The prior probability of teapot is equal to that aggregate probability. So the prior probability of teapot is very low. Since there is no evidence for teapot – as (T2) reports – its probability remains very low; indeed it is close to zero – as (T3) reports (Van Inwagen (2012), pp. 21–22). Van Inwagen thinks that a premise corresponding to (T2′ ) needs to be added to Russell’s presumed argument against theism: (G2′ ) Theism has a vanishingly small prior probability of being true.

But while van Inwagen is prepared to accept (T2′ ) and thereby let (T4), the conclusion of the teapot argument, go through, he claims that we are not in a position to know (G2′ ).⁷ Van Inwagen writes that he sees ‘no way to construct an argument, an argument that employs reasoning that even superficially resembles my reasoning anent the teapot hypothesis, for the conclusion that, prior to the consideration

7 By contrast, see Swinburne (2004), chapter 5, who thinks the prior probability that God exists is not low whereas Mackie (1981), pp. 98–99, thinks that it is ‘pretty low’.

Agnosticism and the Balance of Evidence | 9

of such evidence as there may be for or against the existence of God, we ought to ascribe an extremely low probability to the proposition that God exists’ (Van Inwagen (2012), p. 25). Very often in discussions of the probability of a given hypothesis, by ‘the prior probability of hypothesis H ′ the discussants mean the probability of H prior to the evidence under consideration. In the present context of the debate between Russell and van Inwagen, we are to take it that there is no evidence for or against theism, and that means that there is no evidence under consideration. Accordingly, by ‘the prior probability of theism’ van Inwagen presumably means the probability of theism prior to any evidence at all. We might call this the intrinsic probability of theism. Van Inwagen’s reply implicitly makes a concession. He explicitly claims that there is no argument that shows that theism has a low prior probability. But he implicitly concedes that he has no argument to show that theism has any other prior probability – a middling or high prior probability. Otherwise I take it that he would offer such an argument. In that case he would not merely have no reason to accept (G2′ ) but he would have reason to reject it as false. In short, van Inwagen is agnostic with respect to the prior probability of theism. This has various consequences; here are two of them. First, if the prior probability of theism is unknown, then presumably so too are the prior probabilities of F and G – Hume’s hypotheses about an infant and an infirm god. But then if a teapot-style argument against theism fails for the reason van Inwagen gives, it would also fail against F or G. An agnostic about the existence of God would, I take it, also be agnostic about F or G, but many philosophers would think it unreasonable to believe either F or G and reasonable to reject them. The question facing van Inwagen is on what remaining grounds it would be rational to reject F or G. Richard Swinburne has claimed that examples such as F and G are less simple, and thereby each has a lower prior probability, than theism (Swinburne (2004), chapter 5). Swinburne’s claim is debatable: if F, G and theism are competing hypotheses which might be used to explain some phenomenon – such as the order of the universe – then (1) what F and G each posit is infinitely weaker in respect of intelligence, power and freedom than what theism posits, although (2) each hypothesis otherwise does the same explanatory work as the other. Given these two facts, there is reason to think that F and G are each simpler than theism (cf. Van Inwagen (2002), p. 29). But in any case the avenue Swinburne takes is not open to van Inwagen. Since van Inwagen claims not to know what the prior probability of theism is, he is not in a position to say that it is higher than the prior probability either of F or of G. The puzzle is not confined to theological examples. Consider also:

10 | Chris Daly H I J

Monads exist. Noumena exist. The neutral entities posited by Russell’s neutral monism exist.

There is no evidence that monads, noumena or the neutral entities exist. I take it that the arguments for the existence of such entities fail. Nevertheless, the prior probabilities of H, I and J are unknown. At any rate, I am as confident that the prior probabilities of H, I or J are unknown as I am that the prior probability of theism is unknown. Yet it is reasonable to deny that monads, noumena or neutral entities exist. The puzzle is why this should be so, given the above response to Russell’s teapot argument. Second, if the prior probability of theism is unknown, then even if we find evidence which confirms theism more than it confirms its negation (i.e., atheism), then we will not know whether theism is more probable than not. Let us see how. First of all, what does it take for a piece of evidence to confirm a hypothesis? This is a much debated matter. Still, at least for the purposes of illustration, let’s consider the following principle (E):⁸ (E) When a given piece of evidence e is more probable on H than on H ′ then e confirms H more than H ′ .

Suppose that there is a great deal of evidence e which is more probable on H than on H ′ and so that e greatly confirms H more than H ′ . Suppose too that there is little or no evidence e’ such that e’ is more probable on H ′ than on H. Even so, unless we know what the prior probabilities of H and H ′ are, we are not in a position to say that H’s posteriori probability – H’s probability given evidence e – is greater than the posterior probability of H ′ . For instance, suppose that H’s prior probability is extremely low relative to H ′ . Then, even given e, the posterior probability of H need not exceed that of H ′ . The same conclusion can be drawn using other principles of confirmation besides (E): if we are ignorant of the prior probability of a hypothesis H, we will be ignorant of what the posterior probability of H is, given evidence which confirms H, and so we will be ignorant of whether H is more likely to be true than H ′ . A corresponding point holds with respect to the disconfirmation of hypotheses. The background point is a familiar one. When we talk of confirming a hypothesis H we may mean that evidence e has raised the probability of H as compared with what it was, or would have been, apart from e; or we may mean that e makes H more likely than not to be true (Swinburne (2004), chapter 1). If we are ignorant

8 Schlesinger reports that (E) is the basis of scientific method: Schlesinger (1977), p. 157.

Agnosticism and the Balance of Evidence | 11

of the prior probability of H, we can be in a position to know that e has raised the probability of H as compared to what it was or would have been, but we are not in a position to know that H is more likely than not. Accordingly, even if the phenomena of consciousness, religious experiences, free will or morality each provide more evidence for theism than for atheism (as claimed in Swinburne (2004), if the prior probability of theism is unknown, then it will be unknown whether, given the evidence, theism is more probable than atheism. In sum, van Inwagen’s response to Russell’s teapot argument has a cost: it inadvertently enjoins agnosticism.⁹ What kind of agnosticism is this? That partly depends on the nature of our ignorance of the prior probability of theism. How might this presumed ignorance be alleviated? Would it be by empirical evidence? No, as we have seen, knowledge that there is empirical evidence for the existence of God would be posterior to, and so would not provide, knowledge of the prior probability that God exists. Would it then be by philosophical argument? The difficulty here is seeing how such argument would proceed. In the case of teapot, van Inwagen was able to argue that it had a low prior probability because its probability was equal to the (low) probability of the aggregate of the origin stories. No such parallel can be made in the case of the existence of God, an uncreated creator, since any origin story would have zero probability. But the less confident we are that we can have knowledge of the prior probability of theism, the higher our degree of belief should be that we cannot tell whether theism is more probable than atheism. But that is to say that our epistemic condition is that of Kantian agnosticism.

5 Scriven’s echo of Russell Michael Scriven has argued that, under certain conditions, the absence of evidence for an entity is evidence that there is no such entity: But if we take arguments for the existence of something to include all the evidence which supports the existence claim to any significant degree, i.e., makes it at all probable, then the absence of such evidence means there is no likelihood of the existence of the entity. And this, of course, is a complete justification for the claim that the entity does not exist, provided that the entity is not one which might leave no traces ... and provided that we have

9 It does leave open the option that belief in God is properly basic: reasonable but not based on evidence (see Plantinga (1983), p. 77 and Wolterstorff (1983), pp. 176 ff), but there are independent objections to this option (e.g. Oppy (1994), pp. 157–159).

12 | Chris Daly comprehensively examined the area where such evidence would appear if there were any. (Scriven (1966), pp. 102–103)

Scriven’s argumentative strategy – if an entity x existed, there will be evidence e; there is no evidence e; so x does not exist – is an instance of modus tollens. This reasoning is valid but does Scriven furnish true premises when he applies his strategy to the case of God? Scriven claims that his ‘last proviso [in the above passage] is really superfluous since it is built into the phrase ‘the absence of evidence” (Scriven (1966), p. 103, footnote 7). But the proviso is far from superfluous: the agnostic need not agree that if God existed, there would be evidence that he existed. That aside, even if the agnostic agreed that there would be evidence that God existed, the agnostic need not agree that we would know what that evidence would be. Scriven is concerned solely with an entity ‘that might leave traces’. Now, what traces an entity leaves partly depends upon what activities it engages in, and, in the case of a sentient entity, what activities it engages in partly depends upon what it chooses to do. But, just as the agnostic suspends judgement about whether God exists, she also suspends judgement about what God would choose to do if he existed. Given the omniscience and omnipotence that God would have, and given the finitude of human beings, the agnostic does not presume to say what God would do, even given that God is omnibenevolent. The range of possible actions open to God that is consistent with his presumed nature is too large for the agnostic to judge what God would do. This is a special case of the more general but familiar point that a hypothesis makes predictions (and hence can be confirmed or disconfirmed) only in conjunction with various auxiliary hypotheses (Putnam (1979)). In a situation in which we are ignorant of which auxiliary hypotheses obtain, we would be ignorant of what predictions the hypothesis makes. Furthermore, it would be unreasonable to expect us to have knowledge of God’s purposes without a detailed revelation from God of what those purposes are (Rowe (1988), p.127). Yet the agnostic is precisely someone who suspends judgement as to whether God has made such a revelation or what the content of that revelation is if God has in fact made one.

6 Ockham’s razor The agnostic says she lacks sufficient evidence to believe that there is a God or to believe that there is no God. Now, it might that thought that Ockham’s razor comes into play in such an epistemic situation. But what should we take Ockham’s razor to say? One formulation, an ‘agnostic’ formulation, says: if there is no evidence ei-

Agnosticism and the Balance of Evidence |

13

ther supporting or disconfirming a hypothesis H, then we should neither believe H nor believe not-H. One might wonder what evidence there could be for, or against, the principle expressed by that formulation (Plantinga (1983), p. 60 and Van Inwagen (2009)), but, setting that aside, the agnostic complies with the principle so formulated. A contrasting formulation, an ‘atheistic’ formulation, says: if there is no evidence supporting an existential hypothesis H, then we should believe that H is false. An existential hypothesis is one which posits the existence of some entity or some kind of entity. The negation of an existential hypothesis is not an existential hypothesis but a negative existential hypothesis. Consequently, if there is no evidence for H, the above principle implies that we should believe that H is false. But, although there is no evidence for not-H, the fact that not-H is a negative existential hypothesis removes it from the scope of the above principle. The atheistic formulation of the Razor has been endorsed by Elliott Sober: “The principle of parsimony counsels that we should hypothesise that an entity does not exist, if its postulation is to no explanatory point” (Sober (1981), p. 145). Suppose that H posits the existence of some entity which not some rival hypothesis H ′ does not, but that the converse does not hold. H ′ is then (ontologically) simpler than H. Suppose also that H does not explain anything which H ′ fails to explain. Other things being equal, a simpler hypothesis explains matters better than a more complex one. So, for any evidence which either of these hypotheses explain, H ′ explains it better than H. Furthermore, for any pair of hypotheses, evidence confirms the hypothesis which would better explain the evidence. It follows that no evidence confirms H. Consequently, Sober’s formulation entails that we should hypothesise that an entity or a kind of entity does not exist if there is no evidence for it. Sober claims that scientific practice follows the ‘atheistic’ formulation rather than the ‘agnostic’ one: physics eliminated the aether and evolutionary theory eliminated group selection. In each case the entity was explanatorily superfluous and so was not supported by evidence (Sober (1981), pp.145–146). What should the agnostic make of this? First of all, here is a route which it is tempting to take but which unfortunately takes us full circle. Consider hypothesis K: K

The aether exists.

Since physical bodies such as the Earth would move through the aether, the movement of these bodies should generate interference patterns. The MichelsonMorley experiment failed to detect such patterns and that would seem to count as evidence against K. But that is not the end of the story. Defenders of K, such as Fitzgerald and Lorentz, modified it by claiming that physical bodies contract

14 | Chris Daly along the line of motion according to a certain constant (the Lorentz factor), thereby accommodating these results. Now, it might rightly be said that Fitzgerald and Lorentz’s contraction hypothesis was an ad hoc modification. But then what is wrong with a hypothesis being ad hoc? Perhaps the answer is that the more ad hoc a hypothesis is, the less simple it is, and that, other things being equal, we should accept the simpler of two rival hypotheses. Fair enough, but what attitude should we take to the less simple of the two hypotheses? Specifically, should we reject it as false or should we suspend judgement about it? We are back to the issue with which we started. Here is another way to proceed. Despite the line of thought we examined above, it does not seem to be scientific practice to claim that an entity or a kind of entity does not exist if there is no evidence for it. Prior to July 2012 there was no evidence that the Higgs boson existed, but it was not scientific practice until that time to claim that the Higgs boson did not exist. Instead, scientists suspended judgement. Where does the difference with the case of the aether lie? The aether was characterised by a certain explanatory role: being the medium through which light is propagated. The Higgs boson is also characterised by an explanatory role: it is to explain why the weak force has a shorter range than the electromagnetic force and why the symmetries involving certain fundamental particles requires that those particles lack mass despite the fact that they have mass. The theory of special relativity does not require that anything occupies the first of these explanatory roles, but current physics finds it (at least) promising to take something to occupy the second of these roles. The resulting epistemic situation is as follows. Given our current scientific theories, there is no recorded data and we can expect there to be no data which the aether hypothesis is needed to explain. By contrast, given our current scientific theories, there are recorded data which the Higgs boson hypothesis might explain and for which we have no superior explanation. I suggest that this is why science denied the existence of the aether but did not deny the existence of the Higgs boson (prior to evidence of its existence coming in). As a first approximation, the principles at work here are: (P1) If, given our background knowledge, there are no recorded data which a hypothesis H is needed to explain and we should expect no data which H is needed to explain, we should reject H as false. (P2) If, given our background knowledge, there is data which a hypothesis H is needed to explain, or it is not the case that we should expect no data which H is needed to explain, we should not reject H as false.

To see how (P1) and (P2) apply, consider some earlier examples again: the hypotheses that Santa Claus, the Flying Spaghetti Monster, or the Great Pumpkin exist. Given our background knowledge of the world, there are no recorded data

Agnosticism and the Balance of Evidence | 15

which these hypotheses are needed to explain, and we should expect no data which these hypotheses are needed to explain. By (P1), we should reject these hypotheses as false. By contrast, consider case L: L

Intelligent extra-terrestrial beings within a thousand light-years of the earth exist.

Let us grant that there is no evidence for the truth of L. Nevertheless, that fact does not provide reason to believe that L is false (Van Inwagen (2005), p. 133). The above principles apply to this example in the following way. Our background knowledge of what is to be found within a thousand light-years of the earth is very limited. We might, for instance, find monoliths of the sort portrayed in 2001: A Space Odyssey and they would be good evidence of extra-terrestrial activity. Given our limited knowledge of how the universe is furnished, it is not the case that we should expect no data which the extra-terrestrial hypothesis is needed to explain. By (P2), we should not reject L as false. Where does theism stand? Should the agnostic re-evaluate her position, given an acceptance of (P1) and (P2)? In section 2 we distinguished three kinds of agnostic depending on whether the agnostic thinks that there was (a) very little evidence either way, or (b) no evidence either way, or (c) there could not be evidence either way. An agnostic of kind (a) would think that, although, there is recorded data which theism is needed to explain, there is only very little such data. By (P2), she should not reject theism as false. She remains an agnostic. Agnostic of kinds (b) or (c) would think that there are no recorded data which theism is needed to explain – for otherwise the existence of such data would provide evidence that there is a God. The question then facing them is whether they think that we should expect no data which theism is needed to explain. An agnostic of kind (c), a Kantian agnostic, would claim that, since there cannot be such data, we should expect no such data. According to (P1), she should reject theism. She should become an atheist. The position of an agnostic of kind (b) involves an ambiguity: does she think that there are in a tensed sense – i.e., that there are up until now – no data? Or does she think that there are in a tenseless sense – i.e. that there were, are and will be – no data? If she takes the second way of disambiguating, then she expects that there will be no data which theism is needed to explain. By (P1), she should reject theism. She should become an atheist. If she takes the first way of disambiguating, she is faced with a choice point: from the fact that there currently are no data that need to be explained by theism, should we induce that there will be no such data? If she thinks we should induce this, then, by (P1), she too should

16 | Chris Daly reject theism. If she thinks we should not induce this, then, by (P2), she should not reject theism. She should remain agnostic.¹⁰

Bibliography Dore, C. (1982), “Agnosticism”, in Religious Studies: 18, 503–507. Draper, P. R. (2001), “Seeking but Not Believing: Confessions of a Practicing Agnostic”, in Divine Hiddenness: New Essays, edited by D. Howard-Snyder and P. K. Mose, Cambridge: Cambridge University Press, 197–214. Feldman, R. (2007), “Reasonable Religious Disagreements”, in Philosophers Without Gods, edited by L. Antony, Oxford: Oxford University Press, 194–214. Hume, D. (1779), Dialogues concerning Natural Religion and Other Writings, edited by D. Coleman, Cambridge: Cambridge University Press, 2007. Kant, I. (1781), Critique of Pure Reason, translated by P. Guyer and A. W. Wood, Cambridge: Cambridge University Press, 1998. Le Poidevin, R. (2010), Agnosticism: A Very Short Introduction, Oxford: Oxford University Press. Mackie, J. L. (1981), The Miracle of Theism, Oxford: Oxford University Press. Oppy, G. (1994), “Weak Agnosticism Defended”, in International Journal for the Philosophy of Religion: 36, 147–167. Plantinga, A. (1983), “Reason and Belief in God”, in Faith and Rationality, edited by A. Plantinga and N. Wolterstorff, Notre Dame: University of Notre Dame Press, 16–93. Putnam, H. (1979), “The ‘Corroboration’ of Theories”, in his Mathematics, Matter and Method: Philosophical Papers, vol. I, Cambridge: Cambridge University Press, 250–269. Rosen, G. (2001), “Nominalism, Naturalism, Epistemic Relativism”, in Philosophical Perspectives: 15, 69–91. Rowe, W. L. (1988), “God and Theodicy”, in Philosophical Topics: 16, 119–132. Russell, B. (1952), “Is There A God?”, in The Collected Papers of Bertrand Russell, vol. II: Last Philosophical Testament, London: Routledge, 543–548. Schlesinger, G. (1977), Religion and Scientific Method, Dordrecht: D. Reidel Publishing. Scriven, M. (1966), Primary Philosophy, New York: McGraw-Hill Book Company. Shalkowski, S. A. (1989), “Atheological Apologetics”, in American Philosophical Quarterly: 26, 1–17. Sober, E. (1981), “The Principle of Parsimony”, in British Journal for the Philosophy of Science: 32, 145–156. Swinburne, R. (2004), The Existence of God, Oxford: Oxford University Press, 2nd edition. Van Inwagen, P. (2002), “What is the Problem of the Hiddenness of God?”, in Divine Hiddenness: New Essays, edited by D. Howard-Snyder and P. K. Moser, Cambridge: Cambridge University Press, 24–32.

10 I am very grateful to Eve Garrard and Scott Shalkowski for comments on an earlier draft of this paper. I am also grateful to Robin Le Poidevin for discussion of the topic of agnosticism.

Agnosticism and the Balance of Evidence | 17

Van Inwagen, P. (2005), “Is God An Unnecessary Hypothesis?”, in God and the Ethics of Belief : New Essays in Philosophy of Religion, edited by A. Dole and A. Chignell, Cambridge: Cambridge University Press, 139–151. Van Inwagen, P. (2009), “Listening to Clifford’s Ghost, in Royal Institute of Philosophy, Supplement: 65, 15–35. Van Inwagen, P. (2012), “Russell’s China Teapot?”, in The Right to Believe: Perspectives in Religious Epistemology, edited by A. D. Lukasiewicz and R. Pouivet, Heusenstamm: Ontos Verlag, 11–26. Wolterstorff, N. (1983), “Can Belief in God be Rational If It Has No Foundations?’, in Faith and Rationality, edited by A. Plantinga and N. Wolterstorff, Notre Dame: University of Notre Dame Press, 135–186.

Gabriele De Anna

Theism and the Ontological Ground of Moral Realism 1 Theism and morality According to an old and prominent tradition of Western philosophy, there is a strong relation between human morality and God, or the gods. This tradition dates back to the Greek origins of Western philosophy, and continues to be pervasive in Western culture. Aeneas was ultimately successful, all the misfortunes he had to go through notwithstanding, since he was pious: he did what he was meant to do, on the ground that he feared the gods and respected their will. The idea of a connection between morality and God was taken up by Christianity, although Christianity added some complications which called for a clearer articulation of this view. The idea began to be challenged at the dawn of Modernity, as philosophers began to reason about morality “etsi Deus non daretur” (Grotius is a well-known example, although he was not the first to have taken up this reasoning). The connection between God and morality was subsequently progressively abandoned. The view is still represented today, although it is a minority view (Adams (1999), Craig et al (2009)). If taken in a bold form, a form that none of the contemporary supporters would endorse but which was once common, the view is certainly false. In its bold form, the view purports that moral life implies belief in and respect of God. This implies that atheists could not be moral, and this is certainly false as there are plenty of atheists who are clearly extremely moral. The bold form of the view must be wrong. Its falsity may have eluded the experience of Socrates, of Virgil and even of Christians of the first centuries, given the diffusion of religious belief in those times, but facts speak clearly against it in our world. Does the falsity of the bold form of the view imply that the view should be abandoned in all its forms, and that there is nothing to learn from the speculations located in discussions of the view through the centuries? I think that the thesis that there is a strong relation between human morality and God is still worth discussing: it can be enlightened by contemporary debates and a discussion of it can contribute to our understanding of current issues. Let us then begin by an elucidation of the terms contained in a statement of the view. By ‘morality’, I mean a series of phenomena, which we have experience of, including, for example, the tendency of human communities to share principles of conduct which have a normative impingement on the members of the commuhttps://doi.org/9783110566512-003

20 | Gabriele De Anna nity, the tendency of individual humans to feel a sense of guilt when they perform certain kinds of actions and a sense of well-being when they perform certain other kinds of actions, the tendency of humans to evaluate themselves or other people or the actions performed by themselves or by others as good or bad, and so on. Ways of conceiving God (or the gods) have taken very different forms throughout the course of history, but an important common trait seems to be that such a Being (or such beings) is (or are) taken to be similar to humans in some ways and different in others. It is (they are) similar to humans for its (their) capacity to act, i.e. to responsibly and freely cause events and/or other beings. It is (they are) different from humans in that it is (they are) more perfect than humans, e.g. by having more power, more goodness, and so on. Hereafter, I will use the term ‘God’ to refer to all possible ways of understanding God or the gods, and only towards the end will make some further considerations which will clarify my use of the term. The range of relations which thinkers have claimed to subsist between morality and God (or the gods) is decisively varied: we have at least two of these relations discussed in Plato’s Euthyphro: the good is such since it pleases the gods, or the gods are pleased by the good since it is such. In this essay I will consider members of one family of these relations, namely relations which may support the ontological grounding thesis (OG): OG God is the ontological ground of value.

By ‘value’ I mean whatever features of reality one may take – given one’s ontological and meta-ethical views – to underpin the phenomena that were defined above as constitutive of morality (see also section 2 below). ‘Value’ here does not therefore necessarily refer to a sui generis kind of entity or entities, although this is one possibility that may obtain. With ‘ontological ground’ I refer to the family of relations that have this in common: A is the ontological ground of B if and only if B’s existence depends on A’s existence. Inherence, participation and causation are examples of ontological grounding-relations. Naturally, what kind of relation one takes to constitute the ontological ground, what properties one takes God and value to have, and what one takes their existence to consist in are interconnected issues. OG can be read in at least three ways: from epistemic, ontological or explanatory perspectives. According to the epistemic reading, OG claims that knowledge of the existence of value can only be arrived at via knowledge of the existence of God. On this reading, to claim that God is the ontological ground of value is to say that we could not know that there is value if we did not know that there is God. Given that knowing implies believing, the thesis implies that someone who does not believe in God cannot consistently believe in value (here I use ‘belief’ in the

Theism and the Ontological Ground of Moral Realism | 21

epistemic sense, not in the religious sense of ‘belief in’ which is close to ‘faith’). On the second reading, OG claims that there is a dependency relation between God and value, regardless of how and in what order we come to know or believe facts about the nature or existence of God, value and the relation that holds between them. On this point of view, OG is a claim which does not concern our cognitive capacities at all. Finally, on the explanatory reading, OG claims that the existence of God is the best way to explain the existence of value. The claim does not concern just the order in which we come to know about God and value (like the epistemic reading), neither only the relation holding between God and value independently of our epistemic states (like the ontological reading). Instead, it concerns the rational requirement of believing in God, given the background of one’s epistemic architecture, and given one’s belief – justified by independent evidence – in the existence of value and of some of its features. In this essay, I will suggest that moral realism implies OG interpreted in accordance with the third, explanatory reading. I will take OG in a general form as far as the interpretations of the grounding relation of the properties of the relata, and of the notion of existence implied by them are concerned: what I say could hold for various possible ways of understanding that relation (and, hence, the notions of God, value, and existence), although some constraint in the spectrum of viable possibilities will be suggested by considerations concerning moral realism. In the next paragraph I will consider moral realism and – elaborating on an argument by Thomas Nagel – I will suggest that it implies the existence of a correspondence or an adequacy between values and our practical rationality (cognitive and volitional capacities), which I call normative fitness. I will suggest that such fitness needs explanation. In the following section, I will suggest that teleology is the best explanation of this fitness, and in the section after that I will consider what kind of teleology is needed to get the best explanation. Subsequently, I will consider the relation between teleology and theism: I will contend that the kind of teleology needed to account for the fitness between values and our practical rationality is of an intentional kind. Finally, I will draw some conclusions about OG and the interpretation of it which is supported by moral realism.

2 Moral realism and the problem of normative fitness Moral realism is the view according to which there are values, i.e., features of reality which justify our endorsement of moral judgments. By ‘justify’, here, I mean that values make it necessary for us to make certain judgments, regardless of what

22 | Gabriele De Anna one’s view concerning the content of moral judgments is. A cognitivist will believe that moral judgments can be true or false, and so, if she is also a moral realist, she will think that there must be values to make them so. A non-cognitivist will believe that moral judgments might be valid or invalid (whatever that might mean), and so, if she also is a moral realist, she will think that there must be values to make them so. (It seems to me that a non-cognitivist cannot be moral realist, and I mention that position as a purely academic possibility: you never know how far the creativity of philosophers can go. However, I will set aside this possibility here and I will not consider it further). What are values? Above, I defined them as features of reality one may take – given one’s ontological and meta-ethical views – to underpin moral phenomena. These features could be sui generis objects which can only be recognized by moral agents and expressed in moral judgments, but it does not necessarily need to be so. Values could also be normal, natural features of the world, if these can be seen by agents as giving them reasons in favor of or reasons against certain actions or action types, and thereby as justifying certain moral judgments. If this can be done, then this latter understanding of moral values should be definitely preferred to the former, since it is more metaphysically parsimonious. But are natural features of the world enough to explain our use of moral language? Thomas Nagel has famously and effectively argued in favor of a positive answer (Nagel (2012)). Take my judgment “I should stop the car now”, made silently while I am driving my car and a dog crosses the road right in front of me. Nagel argues that the very fact that a dog is crossing the road, together with the fact that I am driving my car along that road, combined with any other relevant facts, are together enough to give me a reason to stop, thereby making my judgment true (or valid). No further non-natural fact or object is needed. If one wants to press the question further and asks “What makes those facts values for that subject?”, then one is forced either to turn to the subjective states of the agent, and thereby one abandons moral realism in favor of subjective forms of anti- or quasi-realism, or one must point to values seen as further, non-natural entities. According to Nagel, however, we do not need to press the question further: all that needs to be explained about moral judgments is that they are made true or false (or valid or invalid) by facts which constitute reasons for action. It is these very natural facts which constitute reasons and make moral statements true or false (or valid or invalid). Why should we accept moral realism in the first place, however? Natural facts can explain what makes moral statements true (or valid), but this does not explain why there should be something that fulfills this role at all. Why should moral statements be interpreted as having a mind-independent reality? Nagel suggests that the truth of moral realism is supported by some basic aspects of our experience,

Theism and the Ontological Ground of Moral Realism |

23

including the fact that the way in which we deploy moral language and normative judgments purports that our desires should be revised in face of new relevant, objective evidence, when it is given to us. Following this line of thought, in what follows I will understand moral realism as a view according to which moral judgments are made true and false (or valid and invalid), by objective facts, which are normal “natural” facts, rather than employing the existence of sui generis moral objects or properties. The truth of moral realism has some consequences concerning practical reality and practical rationality. If natural facts have to be reasons for action for us at all, they must have the following features: They must include some potentialities such that they can be acted upon and changed in some ways. By ‘potentiality’ I just mean a set of features of a state of affairs, according to which that state of affairs is liable to be changed in certain ways. A fact which gives me a reason for action must include potentialities, in this sense, which my action has to actualize: if there were no scope for action, there would be no reason for action. But in order for some facts to become reasons for an agent, the agent must also have some features. Firstly, the agent must be able to detect and recognize the potentialities inherent in the relevant facts. Secondly, the agent needs to be able to exercise her powers in order to actualize those potentialities. Thirdly, she needs to be aware of the fact that she has that power. Finally, she needs to see the actualization of those potentialities as something worth doing, as an improvement of the status quo, or at least as the avoidance of a worse, future state of affairs. In order to be realism, moral realism must claim that what is valuable is the actualization of some potentialities existing in reality, independently from the attitudes of anyone towards their realization. If this were not so, it would not be reality which makes moral judgments true (or valid), as moral realism contends. On the other hand, if moral realism has to be a practical view at all, humans must be liable to be motivated by the prospect of actualizing potentialities which are valuable in themselves. This means that human practical reason must be inclined or directed toward the realization of value. The latter claim faces a problem connected with evil: if humans are really inclined or directed towards value and its realization, why do they so often act viciously? I cannot address this problem here: it is one of the main challenges for philosophers stressing the importance of inclination in understanding morality, while recognizing an objective grounding for the truth (or the validity) of moral judgments (cf., on this topic Finnis (1983), chapter 2). It suffices to say that practical reason is the faculty with which we choose what to do and justify our actions. In choosing what to do we always aim at what seems good to us, even if we might be mistaken. That reason must seek the good is necessary, if we can judge some bits of reasoning as faulty. (Practical reason stands to practical reasoning in the same relation in which logic stands to theoretical rea-

24 | Gabriele De Anna soning). Mistakes in practical reasoning can depend on our epistemic failures, or on other pro-attitudes, different from reason, that lead us astray. Our practical fallibility, then, shows that we can have epistemic mistakes and volitional mistakes: we are driven by several pro-attitudes and reason does not always prevail. The idea of moral realism, then, is the idea that there is normativity: reality is meant to be realized in certain ways, i.e., by actualizing valuable potentialities present in it, and our agency must be oriented towards that realization. Our practical rationality functions correctly if it recognizes a possible realization of potentialities immanent in reality as worth pursuing if and only if they are really valuable. Our volitional states might not always be spontaneously directed toward the realization of value, but practical reason must still recognize objective value and re-orientate our volitional states accordingly. In order to do this, practical reason might have to frustrate current desires. If our practical rationality can operate in this way, as moral realism seems to presuppose, the volitional architecture of an agent must be such that it can be reshaped in ways that direct her towards real value. The point we have just reached purports that the potentialities existing in reality, the realization of which is valuable, and the ends pursued by our practical reason, when it functions correctly, must fit together: the valuable ways of actualizing potentialities are recognized as such by an agent whose practical reason functions well. In simpler words: value and practical reason fit together. I call this correspondence between value and practical reason “normative fitness”, although in what follows I will speak more simply about “fitness”. The word ‘fitness’ is now popular in evolutionary theory: by using it, I am not committing to an evolutionary account of the facts that I take it to refer to, as I shall explain. The word was previously used in a technical sense by Samuel Clarke, to refer to the correspondence between the will and things in the world. I use it in a similar way, although I do not for that reason commit to all aspects of Clarke’s view, in fact my explanation of the correspondence between the will and things in some ways rejects his account. The claim that there is fitness does not mean that the world is so rational overall that there is one moral solution to each problem. It is consistent with the view I have just laid down that for each state of the world which can be acted upon there are multiple good ways of realizing or actualizing potentialities. For my point, it suffices that we can say that in some cases some of the possible actualizations are good and some are bad. The choice among the viable good realizations will then be open to our creativity. My view is also consistent with the possibility that there are genuine moral dilemmas, dilemmas which resist all our possible efforts to overcome epistemic faults or communication problems: in the case of dilemmas there is still fitness between our faculties and the potentialities of the world.

Theism and the Ontological Ground of Moral Realism | 25

If genuine moral dilemmas are really possible, an explanation of fitness will have to explain also why the cause of fitness brings about the possibility of moral dilemmas, however this is an issue which I will not be able to discuss in this essay. The fitness between value and practical reason calls for an explanation. I am not claiming that value as such requires an explanation: as I mentioned above, Nagel argues that moral realism recognizes value as a fundamental feature of reality, the existence of which does not need to be proven on the basis of something more fundamental. I have agreed with this, and do not take this back. However, I have then claimed that moral realism – even understood as the fundamental view which Nagel takes it to be – implies that there is fitness between value and practical reason. Fitness elicits our wonder and calls for an explanation: how did it come about?

3 Fitness and teleology Before we can attempt an explanation of fitness, more needs to be said about practical reason. Practical reason is the application of our rationality to practice: its conclusion is an action, or an intention, or a justification for an action. Because of this relation to practice, practical reason is not a mere calculative power, but includes volitional capacities as well. An agent does not decide what to do by deducing a statement from a set of descriptions of facts. Rather, an agent looks for the best option by comparing the impact that the outcomes of different ways of responding to that situation will have on his subjectivity; this comparison relies also on memories about responses to similar situations in the past, and on the outcomes of past actions. Volition is employed in this process of practical reasoning as much as calculation is: through their joint operation, volition and calculation assess facts and compare different possible ways of realizing the potentialities built into a situation. According to moral realism – which we are assuming at this point in the argument – not all volitions, but only those conductive to real value are correct. Hence, a subject might be wrong about what values are open as possibilities ahead of him. In order to reason practically in a correct way, an agent must develop his volitional and cognitive faculties in a way which leads him to act in conformity with the good or value. This seems circular: the good is what a good person would chose (since it cannot be deduced from facts), but a good person is the kind of person who would chose the good. However, this is not a vicious circularity, it is just the condition in which man finds himself according to moral realism: the need to work out objective value with his limited capacities. (I will not insist on this point, which is the object of discussion in virtue theory). If

26 | Gabriele De Anna practical reason and value fit together, then, this can only be in the sense that the architecture of the practical rationality of humans (including both its cognitive and its volitional capacities) is liable to be developed in a way conductive to the recognition and implementation of value. Hence, the fitting together of practical rationality and value is not the result of conceptual necessity, but a metaphysical matter, concerning how humans are and how the reality in which humans have to make their choices structured. The picture of the relation between practical rationality and value that we have now is that reality has potentialities which call for proper ways of realization, human practical rationality has proper ways of developing, and the proper, correct development of practical reason is conducive to the capacity of recognizing the right potentialities in reality and actualizing them in the right ways. This suggests that the fitness between practical rationality and value is the result of a teleological orientation in and of reality. There is a way in which reality is meant to be realized, and there are ways in which human subjects are meant to develop in order to realize reality correctly. Hence, reality is directed toward a telos, since objects included in it have coordinated tele. Teleology offers an answer to the problem of fitness. The problem now is: what sort of teleology must be in play in the solution to the problem of fit? Mark Bedau (1992) has distinguished three degrees of teleology. In teleology of degree one, a process brings about a state which is good for the being which undergoes that process, but that state is not in any way a cause of the process. Think about eating salad: it contributes to the health of a certain person, but that person eats salad only because she likes the taste. Health is a good for that person and it is the end product of eating salad, so there is a form of teleology, but health – even as an object of motivation, not as an evolutionary explanation of the tastes of the agent – did not play any causal role in bringing about the process of eating. This is teleology of grade one. As Bedau recognizes, this can be hardly called teleology, since the end result is not something toward which the process is directed. In grade-two teleology there is an end of a process which is a good for the being undergoing that process and that end has also a role to play in bringing the process about, but it does not play that role qua good. A thermostat keeps the temperature constant by switching a heating system on and off. Keeping the temperature constant is part of the good functioning of a heating system, and so the thermostat has an end which is a good: it is a case of teleology. (This notion of good is only relative to the functioning of the system: it is an open question whether it is good all things considered, i.e. morally good; but I leave the issue of moral good aside here, since I want only to explain the relation between process and good in grade-two teleology. The schema can then be applied to more robust

Theism and the Ontological Ground of Moral Realism | 27

notions of ‘good’). Furthermore, keeping the temperature constant is part of the causal antecedents of the operation of the thermostat: it is for that end that it goes on and off. So it is not grade-one teleology. However the goodness of the end of keeping the temperature constant does not contribute to the causal antecedent of the functioning: the thermostat goes on and off in order to keep the temperature constant, not because it is good that the temperature is constant. (Consider that the relevant process here is the operation of the thermostat which goes on and off, not the actions of those who set it up so that it keeps a certain temperature in the room). There are also cases in which the end plays a contribution to the causal antecedents of the process qua good, and it is for these cases that Bedau speaks of grade-three teleology. Consider the person who installed the thermostat, and suppose that he did it in order to keep his house warm at a temperature which is good for him and his family. In this case the end of the process (keeping the temperature constant) is good (as in grade-one teleology) and – as a content of the states of mind of the agent – a contribution to the causal antecedents of the process (as in grade-two teleology). Here, however, the end plays these roles precisely for its goodness: it is for the sake of the goodness of keeping that temperature constant that the agent installed and set up the thermostat in a certain way. According to Bedau grade-three teleology is typical of intentional action: it only rarely occurs in non-intentional actions, as, for example in cases of features of organisms which are good since they enhance the chances of survival of those organisms, and were selected precisely for the fact that they have that goodness in them. The goodness of these features, in these cases, is among the causes of itself qua goodness. What degree of teleology is needed in order to overcome the problem of fitness? Degree-one teleology will not do. Degree-one teleology concerns cases in which the end plays no role in the causal antecedents of the processes leading to it. The problem of fitness, on the other hand, concerns cases in which the end is a state of affairs the representation of which in the mind of an agent is among the causal antecedents of the processes leading to it. Hence, appeals to degree-one teleology cannot account for the fitness we need to explain. Degree-two teleology has better hopes of success. The end, in this case, plays a role in the causal antecedents of the process leading to it, although it does not play any such role as a good, or for its goodness. This solution can be accepted also by naturalistically-oriented philosophers. Degree-two teleology, indeed, can be accounted for in evolutionary terms. For example, we can claim that biological organisms, but also all other forms of complex systems to be found in nature, have proper ways of functioning, in reference to which we can claim that something can be good or bad for them. At the same time, we can claim that their ways of functioning leading to certain ends were not brought about because of the goodness of

28 | Gabriele De Anna those ends; rather, the ends were selected because they were increasing chances of survival and reproduction. If this can be said also of the valuable potentialities present in reality and about the features of our practical reason that fit with them, then a grade-two teleological account of that fitness can be offered and the implications of moral realism can be explained. Some philosophers seem to take exactly this line (Casebeer (2003)). Many other philosophers, however, tried to debunk morality by criticizing attempts to explain moral realism through evolutionary accounts of teleology (Joyce (2006) and Street (2006)). There are two main arguments for the evolutionary debunking of morality: let us call them the argument from the unreliability of our moral capacities, and the argument from the radical contingency of teleology. Very briefly, the argument from the unreliability of our moral capacities claims that traits are selected for their positive contribution to fitness (in the evolutionary sense), but a capacity to track real moral properties (even if these existed) would have no advantage for evolutionary fitness. Things are different in the case of truth in theoretical matters: knowing the truth makes one evolutionarily fitter. Hence, evolutionism would suggest that our theoretical cognitive faculties are reliable: they track truth, and that is why they were selected. By contrast, on this model, our practical cognitive and volitional faculties have no reason to be reliable: in fact they were selected for advantages they gave independently from what they tracked. For this reason ‘good’ or ‘value’ become explanatorily idle, and there is no need to introduce them in our ontology at all. To put the point in more general terms, we could say that if we can account for grade-two teleology in purely evolutionary terms, the very need to claim that value is real fades away. The existence of value is just an illusion: the illusion might be beneficial to us, since our cognitive and volitional capacities which generate it were selected because they increased our fitness, but it is an illusion nonetheless. The argument from the radical contingency of teleology stresses the extreme contingency of tele brought about by evolutionary processes. Let us suppose that reality has a functional structure as a whole (as the existence of normative fitness seems to suggest), a structure that it is the result of evolutionary processes. Even if this were the case, nothing would support the conclusion that such a functional structure has any normative bearing. This structure would be radically contingent upon casual events of the history of the universe. If something had gone slightly differently, the proper functioning of the universe would also be totally different, and in many ways incompatible with the current state of the structure of the universe. Hence, if some organism has a way of functioning due to which it does not fit in the present conditions of the universe as well as other organisms, this should not have any normative implication concerning the proper functioning of that organism. Suppose that a certain human being, unlike most humans, is not inclined

Theism and the Ontological Ground of Moral Realism |

29

to consider human life a value and to respond in the right manner in problematic situations (e.g., he does not feel compassion for humans who suffer). Given the hypothesis of radical contingency, we should just conclude that the architecture of practical reason of this person is different from that of typical members of the human species. Nothing would ground the claim that he is missing the point, that he is incapable of cognizing values that are objectively there anyway. Moral realism would just be wrong. Nagel sees the strength of debunking arguments and I share his opinion. I also share his view that the strength of debunking arguments should not lead us to reject moral realism. Rather, he claims, the strong intuition that supports moral realism should induce us to reject via modus tollens some of the assumptions from which the debunking of morality follows. Those assumptions include – most importantly, but not exclusively – evolutionary naturalism. In his view, the problem can be solved by rejecting evolutionary naturalism. By ‘evolutionary naturalism’ I mean¹ what Nagel calls “the materialist neoDarwinian conception of nature”, which is not strictly speaking a scientific theory, nor a philosophical position, but “a comprehensive, speculative world picture that is reached by extrapolation from some of the discoveries of biology, chemistry and physics – a particular naturalistic Weltanschauung that postulates a hierarchical relation among the subjects of those sciences, and the completeness in principle of an explanation of everything in the universe through their unification” (Nagel (2012), p. 4). In his view, the impossibility for evolutionary naturalism to account for moral reality is a stumbling block which calls for a revision of that worldview (2012, 105). In other words, he argues that our reasons for believing in moral realism are stronger than our reasons for believing that the world-view of evolutionary naturalism is complete. We should hence recognize that the findings of science do not yet give a full picture of reality, and we should expect that there is more that science must still find out: a completer scientific picture will be able to show also how and why moral realism is true. According to Nagel this is not matter of switching to a transcendent, supernatural outlook of reality, but of enlarging our conception of nature and of the tasks and objects of science. I agree with Nagel’s reasons for resisting the debunking conclusions. I think that debunking arguments have also other problems: e.g., they work on the assumption that morality is an irresistible illusion, and so they interpret our volitional capacities in analogy with our perceptual capacities. As I have argued

1 Independently from Nagel, I gave a fuller characterization of this worldview and of its metaphysically surreptitious presuppositions in De Anna (2012a), Ch. 1.

30 | Gabriele De Anna somewhere else, this misrepresents the actual way our agency functions (De Anna (2012b)). According to Nagel, we cannot really yet imagine what the new image of the world should be like. Nevertheless he thinks that some of its features should be clear. One of the features is the objective existence of value. This implies the view that there is a teleological orientation of the universe, which is part of its natural order. As he makes clear in an argument against Roger White’s (see White (2007)) claim that the teleological explanation of the origin of life must involve intentional teleology, Nagel rejects any concession to theism: the teleology involved in the new worldview he seeks must exclude intentional agents giving orientation to reality. However, as against the suggestion of the current world-view, such teleology should not be radically contingent: some outcomes of evolutionary processes must be preferable over others. One possibility Nagel mentions is that physical laws are not strictly deterministic, so that the world could evolve in more than one way. This would mean that some ways of evolving would be more likely than others because of a probabilistic bias introduced by laws of nature which would be even more fundamental than physical laws. The universe is heading somewhere, on this view, although there is no one leading the universe in that direction (Nagel (2012), pp. 91–93).

4 The best teleological explanation of fitness Nagel’s teleology represents a challenge to White’s contention that what strikes us as marvelous (namely, life) can be teleologically explained only on the ground of intentional teleology. Nagel’s teleology is comparable to Bedau’s grade-three teleology which does not imply intentional action: even if the end of the universe is not chosen by an intelligent agent, the universe is heading towards certain futures in which it can evolve rather than towards others since “those futures are more eligible than others” (Nagel (2012),p. 93). The higher “eligibility” of that subset of possible futures, however, would not be the result of intentional decisions, according to Nagel, but the result of the fact that those futures “have a significantly higher probability than is entailed by the laws of physics alone” (Nagel (2012), p. 93). It must be recognized that Nagel conceives his tentative, probabilistic account of teleology while discussing the challenge to evolutionary naturalism represented by human cognition. Later in his book he goes on to consider the problem of value, which he takes to be an even stronger challenge to evolutionary naturalism than cognition. He hence does not claim that this sort of non-intentional

Theism and the Ontological Ground of Moral Realism | 31

teleology is an adequate explanation of value. Furthermore, he claims that we cannot know yet what an adequate account of teleology will look like: he only believes that it must involve grade-three teleology of a non-intentional kind (although he does not use this terminology). It seems to me that it is contentious that Nagel’s teleology can solve the problem of cognition, but I am not concerned with this problem here. I think that the form of teleology he suggests cannot answer the challenge of value. My point is that the existence of value opens the problem of the fitness between practical reason and values, and a solution to this problem requires the intentional version of grade-three teleology. I will make my point by taking into account Nagel’s tentative, probabilistic account of teleology, even though he does not apply it to value; I hope that it will be clear that the problems I raise would affect all possible versions of non-intentional grade-three teleology. As we have seen above, the claim that there is fitness between the possibilities immanent in reality and our possible responses to the realization of those possibilities is not just a matter of fact, since the very idea that there are values is that there is normativity built into reality: the fit is between the possibilities which are meant to be realized, and what kinds of responses to the possibilities we are meant to have. Teleology is an explanation of this fitness precisely because it claims that there is a way in which reality is meant to be realized and we are meant to respond to the possibilities open in it. Now, is Nagel’s sort of teleology apt to ground this explanation? I would like to argue that it is not, since it leaves something unexplained. What we need to explain is the fact that reality is meant to be realized in certain ways and our agency is oriented towards that realization. As we have seen, our practical rationality needs to recognize a possible realization of potentialities immanent in reality as worth pursuing, and that means that practical rationality needs to recognize that realization as calling – when this is needed – for a response on its part, a reorientation of the agent’s volitional faculties in that direction. This implies, however, that the end must be seen by the agent as an overall good that a perfect or quasi-perfect agent (i.e. an agent with a correct volitional architecture) would choose. The fitness we need to explain is the fitness between what an agent can want (how he can rearrange the volitional architecture of his practical reason) and what values there are, i.e. what realizations of potentialities in reality are worth pursuing. Can Nagel’s teleology explain this? As we have seen, on his view, the fact that there is natural teleology suggests that a certain future of the universe is “eligible” for actualization. This seems to suggest that the answer to the question above is ‘yes’: ‘eligible’ can be understood and ‘worth being elected’, or ‘liable to election’. Thus an eligible future of the universe would be a future worth pursuing. If this were the case, what an agent thinks to be worth pursuing is what is more eligible, and the fitness is explained. How-

32 | Gabriele De Anna ever, Nagel goes on to explain that the fact that one or more possible futures of the universe are “eligible” means only that the universe is biased in that direction, i.e., that some unknown natural laws, deeper than the laws of current physical theory, assign higher probability to those possible futures. Under this interpretation of ‘eligibility’, Nagel’s teleology cannot help with the problem of fitness. The mere fact that a possible outcome is more probable than another possible outcome does not make it more worthy of choice. What is most probable is not necessarily what is worthiest. I can realize that a friend who is discussing a moral dilemma he is facing with me will most probably make the wrong decision, but this is not a reason to stop discussing the issues with him and attempting to persuade him of the right way to act. The probabilistic account of teleology fails to have an account of grade-three teleology, since it does not explain why the good end operates at a causal level qua good. What is the worthiest choice? As we have seen discussing the moral realist conception of value, a good is a possible actualization of potentialities immanent in reality. The worthiest future of the universe, then, is that which actualizes all the potentialities which are meant to be actualized. This means that, although the potentialities were not actual at the beginning of the process, their actualization was already in some way present from the beginning. In other words, the actualities which grounded from the beginning of the process the potentialities which have to be realized must have been what they are precisely in order to ground those potentialities. It seems to me that this is enough to claim that a universe which has this feature must have intentionality. This does not tell anything about possible subjects of intentionality, but it does suggest that an intentional version of grade-three teleology is in play. A supporter of the solution to the problem of fitness via Nagel’s teleology could object that I am confusing the point of view of the agent with the thirdpersonal perspective from which we are discussing the problem of fitness. If we make that distinction we can claim that Nagel’s teleology might explain the fitness between our volitional architecture and the potentialities present in reality in terms of probability, while the first personal perspective sees the directedness of the universe as heading toward goodness. This line of defense, however, does not work: Under this interpretation of the role of teleology in explaining fitness, the goodness of the end of the universe is not the end of the universe, and the conviction of agents that by following the direction of the universe they pursue value is just illusory. The coincidence of the end of the universe and value would be radically contingent just as in the case of evolutionary naturalism, and agents would be under the illusion that they are pursuing value when they in fact are just pursuing higher probabilities. The same reasons that Nagel gives to reject evolutionary naturalism work against this explanation of fitness.

Theism and the Ontological Ground of Moral Realism |

33

As previously mentioned, Nagel does not claim that his probabilistic account of teleology can explain value, and I promised that I would discuss his account of teleology for the sake of reaching conclusions which could then be generalized for all kinds of non-intentional grade-three teleology. I will now briefly do this. A non-intentional account of grade-three teleology must explain how the goodness of the result qua goodness can be present among the causal antecedents of the result itself. Bedau’s example is a survival-enhancing trait of an organism, which is present in the ancestors of the organism and which was selected for its goodness. This works for goods relative to the optimum functioning of an organism. But we need now to account for a teleology directed towards the realization of moral goodness, i.e., the overall goodness of a certain future state of the universe. Since the universe is not an organism and it was not generated by ancestors via reproduction, how can its goodness be among the causal antecedents of its good final state? It cannot have been there from the beginning, since, moral realism claims, the good is the actualization of potentialities. The existence of normative fitness requires an explanation, and given the structure of our practical reason, a teleological explanation of its fit with the potentialities of reality has to show not only that the universe tends toward some end (or at least towards a set of different possible ends), but also that this end is a good, and that it is for its goodness that the universe is heading there. Hence, the goodness of the end must be a part of the causal antecedents of the direction of the universe towards that state of affairs. How can this be done in the case of non-reproducing organisms? Nagel’s probabilistic account offers a possibility, but, as we have seen, it fails since it does not explain the causal role of the good qua good. The reason for this failure seems to be general, however, and therefore applies to all attempts to offer an account of non-intentional grade-three teleology in cases of non-reproducing entities. How can the good be among the causal antecedents if it is not actual yet? “Intentionally”, seems the only possible answer. We can hence conclude that an intentional version of grade-three teleology is needed, if we want to explain teleologically the fitness between our practical reason and values existing in reality, or at least, if we want to give the best available explanation: the evolutionary naturalist explanation and the explanation based on Nagel’s teleology offer partial explanation, but fail to account for the causal role of the good qua good. The explanation based on grade-three teleology is preferable, since it leaves our basic intuitions about morality untouched.

34 | Gabriele De Anna

5 Teleology and theism Is an explanation of the fitness between our practical rationality and values based on intentional grade-three teleology committed to theism? Is the conclusion that there must be an intentional cause of the teleological orientation of the universe enough to claim that that cause is “what we call God”, to put it in Aquinas’s terms? I think it is, if we relax our requirements for the application of the word ‘God’, as I suggested we should do in another context (De Anna (2012c), pp. 195 ff). Often, philosophers start by a definition of ‘God’ taken from religious philosophy, as “OmniGod” (a free, omnipotent, omniscient, perfectly good being). Then they go on to verify whether they have enough philosophical resources to prove that there must be a being that has these characteristics. By contrast, I think that natural theology is a legitimate branch of philosophy which should work independently from religious discourse and dogmatic theology (although of course conceptual comparisons across the boundaries of these disciplines are always possible). Natural theology should work out the meanings of its terms independently from the understanding of those terms coming from religious discourse or dogmatic theology, although it cannot escape the recognition that the meanings of these terms originated in religious and dogmatic traditions. How can that be done? I think that this is possible, if we think of ‘God’ not necessarily as referring to the OmniGod of Western religious traditions, but as a set of possible referents filling a quite wide volume of a semantic space. This space contains other volumes, which are contiguous to the volume of the referents of ‘God’: these contain the referents of ‘deism’, ‘polytheism’, ‘pantheism’, and so on. The literal meaning of each of these terms constitutes the central aggregation point of the corresponding volume. An application of any of these terms is legitimate just in case the intended referent is closer in the semantic space to the central, aggregation point of the volume referred to that term, than it is to the center of any of the other surrounding volumes. So we can use ‘God’ in natural theology not only to refer to the traditional OmniGod, but to hit any of the referents included in the volume surrounding it, or even the entire volume – of which it is the center – as a whole (De Anna (2012c), pp. 200 ff). I do not have the resources to prove that the cause of intentional gradethree teleology of our universe must be the OminiGod of traditional religions, but I also believe that it can be argued that the cause of intentional grade-three teleology of our universe has to be closer to the focal meaning of ‘God’ than it is to the focal meaning of any other possible words expressing a cause of the whole universe. The traditional conception of the OmniGod is that of a Being who is free, conscious, omniscient, omnipotent and perfectly good. A cause of intentional grade-

Theism and the Ontological Ground of Moral Realism | 35

three teleology in the universe must be conscious: the end of the world must be reached for its value and thus it must be chosen by something which is aware of its goodness. Furthermore, choosing a good qua good requires freedom: if the end came about by necessity it would not be chosen qua good, even if it were good. Even if it were not omniscient, the cause of grade-three teleology must know more about our universe than anyone else living in it, since it must be able to control the working of this universe in ways which direct it to its end. Even if it were not omnipotent, it should still have enough power to direct our universe and thus it would be the most powerful being in this universe. Finally, even if it were not perfectly good, it must be the best being in this universe, for choosing to direct this universe toward an end which, although not necessarily the best possible development of it, is at least one of the possible good results: anything which is good in the universe is good relative to this end, and this end is good relative to the cause of this teleological orientation. Having these characteristics, the cause of degree-three teleology in the universe might not be the ominiGod, but must be closer to it than to any other conception of a first cause of the entire universe. Hence, we can call it ‘God’. If the existence of intentional grade-three teleology is the best explanation of the fitness between our practical rationality and the potentialities of reality, as argued in the previous section, theism must be the best explanation of value. OG must be true, under the third interpretation of it given in the first section.

6 The ontological grounding of morality The truth of OG, that I tried to establish above, entails that if there were no God there would be no morality, or morality would be just an illusion. But this entailment does not concern epistemology, since, as I have stressed, my arguments support OG only on the explanatory reading. Therefore, the truth of OG does not purport that the atheist must be immoral, and not even that the atheist can only be moral at the expense of her rationality. As we have seen in section two, moral realism is a view the truth of which both is independent of, and can be accessed independently from, its ontological grounds. That the truth of moral realism is independent of its ontological ground is clear from the fact that moral realism is a view about the goodness intrinsic in normal, natural facts, as mentioned in section two. Furthermore, that the truth of moral realism is accessible independently of its ontological ground is clear from the fact that the natural facts which make moral statements true are accessible by

36 | Gabriele De Anna us directly, without the mediation of any ground or foundation. This means that an atheist can be a moral realist in a fully rational way. In a sense, then, one can reason about morality “etsi Deus non daretur”: moral facts are natural facts and one can access, recognize and reason about them independently of one’s convictions about God. At the same time, however, a committed atheist could consistently maintain moral realism only at the cost of rejecting OG. If a committed atheist is persuaded by my arguments for OG, she may be put off moral realism. This is actually the direction of the argument taken by evolutionary naturalists who attempt to debunk morality (but, in fact, by a much wider spectrum of moral quasi or anti-realists: cf., Vogler (2002), pp. 42 ff): they recognize OG and renounce moral realism since accepting the existence of God would be – in their eyes – too high a price to pay. Nagel found that abandoning moral realism is also a price too high to pay, and looked for a different ontological ground of value which could offer an alternative to OG. I suggested that an alternative is not available, even if he is right – even in the face of the absence of such an alternative – when he claims that abandoning a realist interpretation of morality is too high a price to pay.

Bibliography Adams, R. (1999), Finite and Infinite Goods, New York and Oxford: Oxford University Press. Bedau, M. (1992), “Where’s the Good in Teleology”, in Philosophy and Phenomenological Research: 52, 781–806. Casebeer, W. (2003), Natural Ethical Facts: Evolution, Connectionism, and Moral Cognition, Cambridge (MA): MIT Press. Craig , W. et al (2009), “The debate: is goodness without God good enough?”, in Goodness Without God Enough: A Debate on Faith, Secularism, and Ethics, edited by R. K. Garcia and N. L. King, Plymouth UK: Rowman & Littlefield, 25–48. De Anna, G. (2012a), Scienza, normatività, politica. La natura umana tra l’immagine scientifica e quella manifesta, Milan: FancoAngeli. De Anna, G. (2012b), “Evolutionary Metaethical Scepticism and the Problem of Justification”, in Willing the Good. Empirical Challenges to the Explanation of Human Behaviour, edited by G. De Anna, Newcastle (UK): Cambridge Scholars Publishing, 206–226. De Anna, G. (2012c), “Can There Be Naturalism without Theism? Contra Tooley’s Thesis”, in The Right to Believe. Perspectives in Religious Epistemology, edited by D. Łukasiewicz and R. Pouivet, Berlin: Ontos Verlag, 179–211. Finnis, J. (1983), Foundations of Ethics, Oxford and New York: Oxford University Press. Joyce, R. (2006), The Evolution of Morality, Cambridge (MA): MIT Press. Nagel, T. (2012), Mind and Cosmos. Why the Materialist Neo-Darwinian Conception of Nature is almost Certainly False, Oxford and New York: Oxford University Press.

Theism and the Ontological Ground of Moral Realism |

37

Street, S. (2006), “A Darwinian Dilemma for Realist Theories of Value”, in Philosophical Studies: 127, 106–166. Vogler, C. (2002), Reasonably Vicious, Cambridge (MA): Harvard University Press. White, R. (2007), “Does Origin of Life Research Rest on a Mistake?”, in Nous: 41, 453–477.

Michał Głowala

Polygeny, Pleiotropy, and Two Kinds of Concurrentist Ontology 1 Introduction: concurrentism and polygeny There has been growing interest within the metaphysics of powers in what is called the polygeny of effects, one and the same effect having various powers as its causes, and pleiotropy of powers, one and the same power having various manifestations in concert with various manifestation partners.¹ Power manifestation is a mutual affair, as it is often said.² There is a strikingly rich and largely forgotten tradition of studying numerous forms of polygeny (or concursus causarum, as it used to be called) in scholasticism; and one of the main issues disputed there is concursus divinus – the ways in which God cooperates with created agents in their actions. The idea of concursus divinus is a core idea of theism. The world is full of powers and what is going on in it is clearly a result of their interplay or mutual interference; the theistic claim is that above all these powers of heaven and earth there is the supreme power of God, and that what is going on in the world is not just an interplay of worldly powers, but also a manifestation of the power of God. The power of God is not just one of these worldly interfering powers, just an additional factor in the game; the ways it cooperates or interferes with worldly powers are utterly different from all the ways of mutual cooperation or interference of these powers. There are many forms of concursus divinus that theists talk about: for example, extraordinary divine interventions in the natural course of events (miracles), or supernatural interventions in Sacraments as the sources of grace. Besides these special forms of divine concurrence there is also, however, God’s contribution to the usual course of interplay of powers in this world; it was called concursus generalis and is my main point of interest here. The view that there is such a real divine contribution is usually called concurrentism as opposed to occasionalism (the denial of any real causal contribution of secondary causes) on one hand and

1 This use of ‘polygeny’ and ‘pleiotropy’ was introduced in Molnar (2003), pp. 194–198. 2 See e.g., Williams (2010), pp. 87–89. https://doi.org/9783110566512-004

40 | Michał Głowala to mere conservationism (the denial of any real direct causal divine contribution to the ordinary actions of secondary causes) on the other.³ The most general argument for concurrentism is the following: the effects of actions of created agents as well as their actions themselves are entities in the world; and all entities have their source in creation; so the actions of created agents are performed by these agents on one hand and created by God on the other.⁴ So the occasionalist claim is that since these actions are created by God, they are not performed by the created agents; the mere conservationist claim is that since these actions are performed by the relevant agents, they are not created by God; and the concurrentist claim is that they are both created by God and performed by the relevant created agents. So the concurrentist claim is that created agents’ actions as well as their effects are polygenic at some fundamental level; and it is a special sort of polygeny that is rejected both by occasionalism and by mere conservationism. Concurrentism poses some widely discussed challenges for the theory of natural causality investigated by science on one hand⁵ and for the theory of free human action on the other.⁶ It used to be one of the main sources of the fiercest debates in philosophy and theology in early modern times, and in particular of the de auxiliis debate between Molinists and Thomists.⁷ Here, however, I do not focus on any of these particular challenges or debates. What I am interested in, from a general metaontological point of view, is rather the very issue of polygeny – one and the same effect having various powers as its causes. My main claim is that there is some very general principle accepted, in a weaker or stronger form, in a great number of ontologies of polygeny (I call it the Partial Causation Principle, PCP) which plays a prominent role both in the rejection of concurrentism and in some fierce debates between concurrentists.

3 For a general introduction see Freddoso (1991) and Freddoso (1994); Shanley (1998) and Shanley (2007); and the collection of papers Morris (1988). I discussed the issue of concurrentism in a broader context of scholastic and contemporary metaphysics of powers in my book in Polish Możności i ich akty. Studium z tomizmu analitycznego (Wrocław 2016). In the book I discuss many claims presented here from a broader perspective of key notions of metaphysics of powers. 4 For this argument see e.g., Thomas Aquinas (1929), lib. II, dist. 37, q. 2, art. 2, corp.; Suárez (1861), disp. 22, s. 1, n. 9. I am speaking about creation here in a loose way and do not discuss the issue whether and how it should be distinguished from conservation; I hope this lack of precision is justifiable in the context of my main problem. 5 See e.g., Carroll (2008) and McCann (2004). 6 See e.g., Dvořák (2013); McCann (2012), ch. 5; Shanley (1998). 7 I have profited greatly from discussing these matters with Petr Dvořák; it goes without saying that he is not responsible for the faults of this paper, especially because we tended to embrace opposite standpoints on these matters.

Polygeny, Pleiotropy, and Two Kinds of Concurrentist Ontology |

41

PCP says, roughly, that for each of the concurring causes there is something contributed to the effect just by this cause as opposed to any other, and that the overall effect is a composition, or combination, of distinct contributions of concurring causes; PCP points at some sort of division of causal work. I claim, however, that there are cases of polygeny in which PCP if false: there is some other sort of division of causal work in them. A great number of them do not belong to philosophical theology, but they have been traditionally used as models of divine concurrence. So it is an important question whether PCP should be rejected in case of concursus divinus. I will proceed in three steps. First (Section 2) I present a strong and a weak version of PCP involved in some contemporary ontologies of polygeny, as well as some strictly related theses concerning contributions of causes. Then (Section 3) I point at some cases of polygeny in which both PCP and some of the related theses are false; I also introduce a generalized notion of causal contribution. Finally (Section 4) I sketch the role that PCP and the related theses play in the MolinistThomist debate concerning general divine concurrence. It seems to me that according to the Molinist account (as opposed to the Thomist one) a weak version of PCP is true in case of concursus divinus; that, I think, sheds some light on both standpoints and on the relationship between them.

2 Partial causation principle (PCP) 2.1 Polygeny as a composition of forces: A strong version of PCP Molnar gives a simple example of polygeny: two horses moving along two sides of a canal pull a barge. One pulls it north-west and the other south-west, and the result is that the barge moves westwards (although no single horse pulls it precisely westwards). The forces exerted by the horses are contributions of concurring powers, and the joint effect is a combination of the contributions. In this context he asks the general question of what happens when a number of powers combine to yield a single effect (Molnar (2003), pp. 195–196). There is a suggestion in Molnar, developed into a systematic theory of causality in Mumford and Anjum (2012), that combination of powers is something like vector addition or the addition of

42 | Michał Głowala component forces.⁸ Now I think there are four fundamental theses specific to this sort of ontology of polygeny, and it is a good point of departure for a more general discussion to spell them out. (i) The theory introduces contributions of concurring powers as something distinct from their (typically polygenic) effects: distinct concurring powers have distinct contributions, although they have a common effect (Molnar (2003), p. 195; Mumford (2009), p. 104).⁹ So we have Contribution Thesis: Contributions of distinct concurring powers are always mutually distinct (and distinct from the joint effect).¹⁰

(ii) The distinction of effects and contributions is closely connected with the idea that contributions, as opposed to effects, are not polygenic; to identify a contribution of a concurring power is to find something that comes exclusively from that power itself and not from any other concurring power (Molnar (2003), pp. 194–195). So we have Monogeny Thesis: Contributions of concurring powers, as opposed to the overall effects of concurrence, are not polygenic.

(iii) The same distinction is also closely connected with the idea that a single power always makes “exactly the same contribution to the production of an event, irrespective of other accompanying powers” (Mumford (2009), p. 104). Powers may have various effects in various contexts (are pleiotropic), but they always make the same contribution to these effects, and the difference of effects is due to the influence of other powers. So we have Monotropy Thesis: One and the same power always makes exactly the same contribution (powers are not pleiotropic in their contributions).

8 It is remarkable that in the work of Bernard Lonergan SJ vector addition (additio vectoralis) is mentioned as something analogous to just one of possible sorts of concurring, as opposed to other sorts (Lonergan (2011), p. 250). What I am trying to show below is that the component forces model is relevant only for some class of cases of polygeny. 9 It is worth noting that this use of ‘contribution’ corresponds to one of the main uses of ‘concursus’ in scholasticism; on one hand ‘concursus’ means just concurrence of many causes in producing a single effect; on the other hand, ‘concursus of F’ means just the contribution specific to the cause F. 10 It should be noted that contributions are not intermediate effects and are not caused by concurring powers; but it may be said that in some sense they are produced by the powers (there are some problems in the theory which I cannot discuss here; see e.g., Marmodoro et al. (2011)).

Polygeny, Pleiotropy, and Two Kinds of Concurrentist Ontology |

43

(iv) Then the fundamental idea of the theory is that the contributions of concurring powers are composed or combined in some way; the composition of forces is a simple example of it. There are, however (as Mumford and Anjum make clear) other kinds of composition which are non-linear or non-additive (Mumford and Anjum (2012), pp. 96–99). In such cases there is also a sort of composition, but a more complicated one. In general, this suggests that the main problem of the ontology of polygeny is to explain, as Molnar says, “what it is for the contributions to combine” (Molnar (2003), p. 196). So we have Composition Thesis: A resultant effect is a composition or combination of contributions of concurring powers.

An ontology of polygeny accepting these four theses accepts a version of PCP: the idea that contributions of concurring powers combine to yield the joint effect. More precisely: the first three theses introduce the contribution of a concurring cause as something coming exclusively and immediately from the cause. Then according to the Composition Thesis contributions of concurring powers are somehow combined or composed in the effect. All this describes some special sort of division of causal work in cases of polygeny. PCP may be treated also as a metaontological principle that shows what is required for a sound ontology of polygeny: on one hand it is an identification of something that is produced immediately and exclusively by each of the concurring causes; on the other hand, it is a description of the way in which these things are composed or combined in the effect showing, to use Molnar’s words, “what is it for the contributions to combine”.¹¹ I have said that what we have here is a relatively strong version of PCP. It is only relatively strong because component forces are not parts of the resulting force, and, more generally, vectors that are added are not parts of their sum. It is especially clear in the case of equilibrium of forces, when there is no resulting force.¹² There are even some reasons to doubt the reality of component forces.¹³ By contrast, there are cases of polygeny in which there are distinct parts of the joint effect which are produced by distinct causes; and a very strong version of

11 So Creary (1981) distinguishes “laws of causal influence” identifying forces or causal influences (like Coulomb’s law or the law of gravity) and “laws of causal action” telling how these influences are composed or combined (like vector addition law). For a brief review of Creary’s position see also Cartwright (1983), pp. 62–69. 12 See Massin (2010); Cartwright (1983), pp. 59–73; Mumford and Anjum (2012), pp. 38–44. 13 See e.g., Cartwright (1983), pp. 56–73.

44 | Michał Głowala PCP would claim that basically any concurring cause of a joint effect produces a distinct part of this effect.¹⁴ On the other hand what we have in this sort of ontology of polygeny is a strong version of PCP. On one hand, the Monogeny Thesis and the Monotropy Thesis make the distinction of contributions a strong one; on the other hand vector addition is a composition or combination in a pretty strong sense.

2.2 Polygeny and features of the effect: A weak version of PCP There is also a considerably weaker version of PCP involved in Alfred Freddoso’s sketch of an ontology of polygeny presented in his defence of concurrentism; it echoes some scholastic ideas concerning polygeny. The key idea here is that, roughly speaking, various features of the joint effect are primarily derived from, or primarily traceable back to, various concurring causes. Freddoso gives a clear example: suppose one draws a square on a blackboard with a piece of blue chalk. Then both the chalk and the person who uses it “count as immediate causes of a single effect”, namely the blue square appearing on the blackboard. In such a context, Freddoso suggests, the colour of the line (blue rather than some other colour) “is traced back primarily to the causal properties of the chalk”, whereas its shape (square rather than a circle or some other shape) is traced back primarily to the influence of the one who uses the chalk.¹⁵ Giving this rough sketch Freddoso says that “what remains is to find a philosophically rigorous characterisation of the idea that features of a unitary joint effect can be traced back primarily to one or another of the cooperating causes without destroying the unity of the effect” (ibidem, pp. 150–151). The application of this idea to the issue of concursus generalis would rest on the suggestion that “the fact that the unitary effect is something rather than nothing” is traceable primarily to God, whereas “the fact that the unitary effect is of one determinate kind rather than another” is traceable primarily to its secondary causes (this is at least close to the Molinist standpoint which I discuss in 4.1). On this account, to identify the contributions of concurring powers one has to identify the features of the joint effect that are primarily traceable to these powers, and the suggestion is that there is a distinct contribution for each of the concur-

14 As is well known, Mill was greatly impressed by the reality of “partial effects”, at least in some cases of “composition of causes” (see Mill (1893), Book III, ch. 6); Cartwright says: “Mill thinks that in cases of the composition of causes each separate effect does exist – it exists as a part of the resultant effect” (Cartwright (1983), p. 60). 15 Freddoso (1994), pp. 148–150; similarly Dvořák (2013), pp. 622–624.

Polygeny, Pleiotropy, and Two Kinds of Concurrentist Ontology |

45

ring powers. This corresponds to Molnar’s distinction of effects and contributions (although one may wonder whether Freddoso would accept the Monogeny Thesis and the Monotropy Thesis). Then the contributions are in some way combined in the effect as many features of a single subject. This corresponds to the Composition Thesis, although combination of features in a single subject is in many respects unlike vector addition. Freddoso thinks that such an analysis of polygeny should be applicable also in cases such as when a car is lifted by two people, one holding the left rear bumper, and the other the right rear one (ibidem: 149). Here he differs significantly from the scholastic proponents of this sort of analysis of polygeny; they usually think that this sort of analysis would be applicable only to some special sorts of concurrence, and would rather not be applicable to the case of lifting a heavy object by a number of persons, for here, they think, there is nothing identifiable in the effect which comes exclusively from one of the concurring causes.¹⁶ At the same time Freddoso rejects some ideas closely related to PCP – and especially the idea of splitting any cooperative action into the plurality of distinct actions specific to the concurring agents; so it is possible according to him that distinct agents do cause a single effect by a single action (ibidem, p. 154). So it seems to me that although Freddoso’s sketch of the ontology of polygeny does involve some version of PCP, it is a considerably weaker version of it than the one accepted within the component forces analogy.

3 A rejection of PCP My claim, however, is there are cases of polygeny without any sort of combination or composition of contributions (in Molnar’s sense) of concurring causes whatsoever; so there are good reasons to reject PCP in its general form. To show this, I will proceed in three steps; first, I will argue that PCP is false in some cases, secondly,

16 For a standard formulation of this thesis about the people lifting a heavy object see e.g., Vázquez (1614), disp. 83, n. 9 and Arriaga (1647), disp. 32, s. 6, n. 38; for other cases of polygeny in which contributions of distinct concurring causes are not identifiable via the distinct features of the object see e.g., Duns Scotus (1981), dist. 17, p. 1, q. 2, n. 78–79 and John of St. Thomas (1949), disp. 13, art. 4, n. 494–499. There is at least one scholastic author, however, who takes Freddoso’s standpoint and thinks that even in the case of two people carrying a heavy object there must be distinct features of the effect corresponding to the different causes: it is Luis Montesinos OP (Montesinos (1621), disp. 30, q. 2, n. 47).

46 | Michał Głowala I will introduce a generalised notion of contribution, and finally, I will add some remarks on the notion of partial cause.

3.1 An argument against PCP It is well known that PCP poses some problems for concurrentism and so is rejected by prominent concurrentists. For example, Aquinas maintains that one and the same effect is not caused partly (partim) by God and partly by the natural cause; it is rather, he claims, caused wholly by God, and at the same time, caused wholly by the natural cause.¹⁷ Suarez repeats this doctrine.¹⁸ It seems that if these statements are taken in a strong sense, they involve a rejection of PCP in the case of God’s concurrence.¹⁹ There are also some other contemporary formulations that seem close to the rejection of PCP; for example, McCann says that “God is not an inhabitant of the created world interacting with entities within it” (McCann (2004), p. 33); and Carroll warns that to conceive God’s causality like a physical force “is to make God a kind of competing cause in the world” or even “just one more cause in the world, although considerably more powerful than any other” (Carroll (2008), p. 593). One may wonder, however, whether divine supernatural causality is the only case that may falsify PCP. I think it is not: there are many purely natural cases of polygeny in which PCP is false. This is also clearly the standpoint of Aquinas and some other scholastic concurrentists. In other words: it is the very analysis of natural polygeny that is enough to falsify PCP in its general form and to establish some principles of a sound general ontology of polygeny that may be used to study concursus divinus generalis. To see this let us again consider Freddoso’s example of drawing a square with a piece of blue chalk. Freddoso says that the shape of the line is traced back primarily to the drawing person and not to the chalk (as opposed to the colour of the line which is traced back to the chalk having some relevant colour properties and not to the person). I do not want to reject this analysis of polygeny and this sense of being tracked back to some cause; but I think another sort of analysis would be pretty plausible here, and another sense of being tracked back is relevant here, too. It is clear that the movement of the chalk is causally relevant for the shape of

17 Thomas Aquinas (1961), lib. III, c. 70, n. 7. 18 Suárez (1861), Disp. 22, s. 1, n. 22. 19 In fact, however, they are taken in a rather weak sense by Molina himself as well as by Freddoso and Dvořák.

Polygeny, Pleiotropy, and Two Kinds of Concurrentist Ontology |

47

the line. Surely the movement of the chalk is not a constant property of the chalk, as opposed to its colour; this difference, however, does not undermine the causal connection between the moving chalk and the shape of the line. So I think that the shape of the line is traceable both to the chalk and to the drawer; both are responsible for the shape of the line and both produce it.²⁰ The chalk produces it by moving in a given way, and the drawer produces it by moving the chalk in a given way. This suggests the following picture: one and the same trait of the joint effect, namely the shape of the line, is traced back to both concurring causes. One of the causes produces it doing something different from the other; this, however, does not mean that there is something in the shape of the line that comes exclusively from the chalk or exclusively from the drawing person: everything in the shape of the line comes from both, although in different ways. Moreover, there is no combination or composition of things coming from one cause with things coming from another. And we do not have to say that what the causes do in order to cause the effect, namely moving in a given way (in the case of the chalk) and moving the chalk in this way (in the case of the drawing person) are somehow combined or composed. One may wonder what is precisely the relationship between the person’s moving the chalk and the chalk’s moving; the Aristotelian ontology of transeunt action (like moving something) offers some interesting answer here.²¹ At any rate, however, it seems clear that the person’s moving the chalk and the chalk’s moving are not things combined or composed in the production of the effect. Notice further that it is basically possible that each of the traits of the effect comes from both causes: there is nothing in the effect that comes exclusively from one of the causes as opposed to the other. The difference of roles between these causes does not consist in the fact that there is anything in the effect that comes from one of them as opposed to the other. Both causes produce the same traits, but one of them produces them by doing something different from the other. In other

20 It is worth noting that Aquinas in his analysis of the action of instrumental causes (unlike Freddoso) does not distinguish aspects of the effect coming from the instrumental cause from the ones coming from the principal cause; he insists rather that, for example, both cutting and making a piece of furniture are operations of an axe, but whereas one comes from it due to its proper form, the other comes from it due to its being moved by the craftsman (“securis operatio secundum propriam formam est incisio, secundum autem quod movetur ab artifice, operatio eius est facere scamnum” (Thomas Aquinas (1903), III, q. 19, art. 1, corp.)) For various (both Jesuit and Dominican) possibilities of interpreting some key Aquinas’s formulations here, however, see the whole paper by Dvořák (2013). 21 I cannot discuss it here; see e.g., Freddoso (1991), pp. 569–572, and Marmodoro (2007).

48 | Michał Głowala words: there are sorts of division of causal work which are utterly different from the one specific to PCP. This involves the idea that the difference of ways of causing the effect is ontologically more fundamental and epistemologically more primitive than the difference of things contributed to the effect. I think this is a sound idea: things contributed by causes are often postulated entities of some sorts:²² but ways of causing are usually more readily accessible. This is quite clear in Freddoso’s example of two men lifting a heavy object: it is difficult to say what precisely each of them contributes, but it is easy to describe different ways in which they lift or carry it. This, I think, reveals a wide group of examples of polygeny (within natural limits) that falsify PCP. A wide range of instrumental causes, for example, belong here. Aquinas, in fact, gives the example of instrumental and principal causes as an illustration of one and the same effect being produced wholly by each of the two concurring causes.²³ Alternatively, we may consider the case of teaching someone to draw some special sort of shape by holding and moving his hand: in that case the line is produced by the chalk and by two drawers, and each of the causes is responsible for all the traits of the produced line.²⁴ These cases belong to the created realm and are in no way specific to theology. It seems significant, however, that they are traditionally used as models or pictures of divine causation and concurrence. This shows that although PCP has a strong intuitive appeal and proves sound across a wide range of cases of polygeny, there are cases in which PCP is false. In this regard, there exist deeply different sorts of polygeny, and it would be interesting to inquire, for example, into which sorts of polygeny PCP holds true for.

3.2 A generalised notion of causal contribution One may wonder, however, how much one should be prepared to reject if one rejects PCP. Do we have to reject the very notion of the contribution of a concurring cause, and the distinction between contributions of distinct causes? I think the answer is negative, but we will have to introduce a more general notion of causal

22 For doubts concerning “things contributed to the effect”, see e.g., McKitrick (2010), pp. 80–83. 23 Thomas Aquinas (1961), lib. III, c. 70, n. 7. 24 Another interesting example of concurrence without distinct correlates of distinct causes in the effect occurs in the scholastic debate of causal relevance of human habits: according to John of St. Thomas, the contributions of the habit and the relevant power of the soul cannot be distinguished in a way postulated by PCP (see e.g., Głowala (2012)).

Polygeny, Pleiotropy, and Two Kinds of Concurrentist Ontology |

49

contribution. Then we will be able to distinguish the causal roles of distinct concurring causes without introducing contributions of the sort postulated by PCP. It seems specific to PCP that a given cause’s contribution is conceived as something contributed to the effect. It is possible, however, to think of the contribution of a given cause in terms of the very act of contributing to the effect. Now in some cases of polygeny contributing to the effect is just producing a part of the effect: what such a cause contributes is a part of the whole effect. In other cases, contributing to the effect is exerting some force, as in Molnar’s case of horses pulling a barge; what such a cause contributes is a component force. In still other cases contributing to the effect is just determining one of its features, and what such a cause contributes to the effect is a feature traced back to it, as in Freddoso’s example. But there are also cases in which contributing to the effect is producing it in some way, and there is just nothing in the effect that is contributed by one of the contributing causes as opposed to the others. For example, the chalk’s contributing to the production of the blue square on the blackboard is moving along some line and leaving its parts on the blackboard: the drawer’s contributing to the production of it is drawing it with the chalk. There may be nothing in the blue square that is contributed by one of the causes as opposed to the other, yet there are contributions of concurring causes in the sense of acts of contributing to the effect – that is to say, producing the effect in the ways specific to the chalk on the one hand and for the drawer on the other. From this point of view the ontologies accepting PCP are applicable to some special cases of contributions in the most general sense. It is true that distinct concurring causes do have distinct contributions in this most general sense: their acts of contributing to the effects are distinct. But the relevant criteria of identity and distinction for these acts need to be further investigated. For example, in the Aristotelian ontology of transeunt action the person’s moving the chalk and the chalk’s moving are not two distinct causally connected acts, but rather one and the same act belonging to two causes. It is, on the Aristotelian account, one and the same motion which is, on the one hand, the chalk’s moving (in so far as it belongs to the chalk), and, on the other hand, the person’s moving the chalk (in so far as it belongs to the drawer). So within such an ontology of transeunt action the man’s and the chalk’s contributions are distinct ones, but the distinction between them is of a very special sort. Hence I think we can accept the Contribution Thesis in just this most general sense, corresponding to the generalized sense of contribution. There are also, however, two theses characterizing contributions, namely the Monogeny Thesis and the Monotropy Thesis. As for the former one, I think it should be rejected in its general form. In some important sense the chalk’s contribution to the production of the blue square line does depend on the man’s contribution – namely, on

50 | Michał Głowala the way he moves the chalk. More generally, it seems specific to some variety of polygeny cases that the contribution (in the generalised sense) of one of the concurring causes depends in some of its aspects on the contribution of some other cause: the latter cause directs or steers the action of the former one. There are some cases in which the contribution of one of the concurring causes just is that of determining or directing the contributions of other causes. So I think there are good reasons to reject the Monogeny Thesis in its general form. Moreover, if you reject it, it seems that you have to reject the Monotropy Thesis, too; if, in a given case, the contribution of a cause is steered by the contribution of some other cause, one and the same causal power does not always make the same contribution, independently of the context pertaining to other concurring causes. This poses some problems for the ontology of concursus divinus which I sketch in Section 4 below.

3.3 What is a partial cause? It might seem, however, that PCP is just trivially true: providing there are many concurring causes, each of them, by definition, causes the overall effect partially, and providing they really are concurring, their contributions must be combined in some sense. Indeed, there are some definitions of a partial cause on which a concurring cause is, as such, a partial cause: for example, Leo Elders says, with reference to his distinction of causa totalis and causa partialis, that “while the former brings about the whole effect, the latter produces it together with other causes’ (Elders (1993), p. 297). So he is not only saying that any concurring cause is, as such, a partial cause, but is also suggesting that a partial cause does not, as such, cause the whole effect. It seems, moreover, that when the expression ‘a partial cause’ is used these days without any special qualification, what the author typically means is just a concurring cause. Now we sometimes encounter this sense of ‘causa partialis’ being used in the scholastic debates; it is, however, often distinguished from another one (both in some Jesuits including Molina and in some Dominicans).²⁵ In the latter sense, a partial cause does need a concurrence of another cause the causal role of which belongs to the same rank or sort (causa eiusdem ordinis), while a total cause, causa totalis, may well need a concurrence of other sorts of causes, but their roles will

25 See e.g., Molina (1588), q. 14, art.13, disp. 27, p. 174, who speaks of causa integra in suo gradu; Toledo (1615), lib. III, c. 3, q. 7, p. 59 vb; Suárez (1861), Disp. 22, s. 1, n. 22; Billuart (1839), diss. 8, art. 4, p. 272.

Polygeny, Pleiotropy, and Two Kinds of Concurrentist Ontology |

51

have to belong to another rank or sort (causa alterius ordinis). For example, both the chalk and the drawing person would count as total causes of the blue line; more generally, both the instrument and the principal agent who uses it are said to be total causes of the effect they jointly produce.²⁶ The chalk draws the blue line of a given shape only if it is moved by the drawing person, but the causality of the chalk and the causality of the drawing man are not of the same rank or sort. (By contrast, a piece of blue chalk would be a partial cause of a multicoloured picture). I think in some cases the idea of ranks or sorts of causality – of causes belonging to the same/another rank or sort – is intuitively pretty clear. It is also clearly related to the Composition Thesis: concurring causes are composed or combined only when they belong to the same rank. There is, moreover, some connection between this and the Monogeny Thesis: it seems that when you have concurring causes of different ranks the contribution of one of them is directed or determined by the contribution of the other, so that the Monogeny Thesis holds only for concurring causes of the same rank at most. It would therefore seem that PCP is intuitively convincing mainly in cases of concurrence of causes belonging to the same rank.²⁷ But, of course, much work has to be done to offer a more rigorous description of rank identity and distinction for concurring causes. At any rate it is interesting that there is some important sense of ‘causa partialis’ in which not all concurring causes are, as such, partial causes – and that this sense, although it used to be well known in some ontologies of polygeny, is rather forgotten today.

4 PCP and God’s general concurrence In this section I would like to sketch briefly the two main forms of God’s general concurrence fiercely debated in the de auxiliis discussions between the Dominicans and the Jesuits: namely, the concursus simultaneus (accepted by both sides) and the concursus praevius (accepted by the Dominicans only). I focus only on the

26 See, e.g., Thomas Aquinas (1961), lib. III, c. 70, n. 7; Billuart (1839), diss. 8, art. 4, p. 272. 27 The Jesuit Leonardo Penafiel distinguishes between two sorts of partial causes: one of them has a contribution distinct from other concurring causes (concursus distinctus) and it contributes to the part of the effect (concurrit ad partem effectus, non ad totum effectum). In the case of another sort of partial causes they produce the whole joint effect in an indivisible way (totum effectum indivisibiliter attingunt). (Penafiel (1666), disp. 10, s. 3, n. 6, p. 283). It is clear that in the latter case PCP would be false.

52 | Michał Głowala issue of God’s concurrence within the ordinary course of action of created agents (and not on issues of grace). Moreover, I will be interested here only in those aspects which are closely related to embracing or rejecting PCP.

4.1 Concursus simultaneus and the indifference of God’s contribution The most accessible way to think about concursus simultaneus seems to be the following: an action (whether voluntary or not) of a created agent is some sort of entity, and as some sort of entity or reality it must be created by God. So one and the same action is performed by the created agent and created by God, if the action in question is decision making, you can say that one and the same decision is made by the relevant created agent and created by God. Making a decision and creating it are clearly two distinct ways of producing it or being its cause in the most general sense. One could even think that these ways are mutually exclusive in the following sense: if an agent creates a decision, he does not make it in the usual sense, and vice versa (by the way, God creates neither His acts of will nor His acts of thought). More generally, performing the action and creating it are two clearly distinct ways of producing this action (and it is plausible that they are mutually exclusive in the sense indicated above).²⁸ This, of course, does not necessarily mean that what is contributed by performing an action is distinct from what is contributed by creating that action: the distinction between performing an action and creating it is a distinction with respect to ways of causing the action, and it need not be derived from any distinction that pertains to what is caused or contributed in these ways. On the other hand, it is an open question whether all the features of the action are reducible to both causes of the action; there may be some special reasons for a negative answer. (Aquinas maintains that the very moral evil of the action is not caused by God as opposed to the created agent.)²⁹ Molina, however, does think that each created action has some features traceable back to the divine cause, as opposed to features traceable back to the created agent: he thinks that God’s contribution provides only the most general features

28 It should be noted again that I leave aside here the important issue of the difference between creation and conservation; I think some issues concerning PCP may be discussed without entering the debates about creation and conservation. 29 See e.g., Thomas Aquinas (1929), lib. II, dist. 37, q. 2, art.2, corp.; Thomas Aquinas (1961), lib. III, c. 71, n. 2.

Polygeny, Pleiotropy, and Two Kinds of Concurrentist Ontology |

53

of the action, whereas all the more concrete features of it are determined by the created agent only – by its nature in the case of non-voluntary agents, and by its will in the case of free agents. More precisely, he maintains that one and the same created action depends both on God and on the created agent, and that this action has some specific properties (e.g. being heating as opposed to cooling) only inasmuch as it depends on the created agent, and not on God. Its only features issuing from its dependence on God are its most general ones, like being an entity or something real. In this sense, God’s contribution to the action is indeterminate (concursus indifferens), and does not determine any of the specific properties of the action: they are determined only by the created agents.³⁰ This seems especially important in the case of free created agents, which determine the specific properties of their actions not by their natures, but rather by their free choice. So in this sense, on the Molinist account, God’s general concurrence fulfils PCP: there is, according to that account, precisely the kind of division of causal work in divine concurrence postulated by PCP. On the other hand, Molina says pretty clearly that the whole effect comes both from God and from the created agent: totus quippe effectus et a Deo est, et a causis secundis. ³¹ So there is only a weak form of PCP involved in the Molinist account of divine concurrence. Now some concurrentists (mainly Dominican Thomists) think that something like the concursus simultaneus and indifferens as sketched above cannot be the only form of general divine concurrence. There are two main objections here, which shall present respectively in 4.2 (a) and 4.3 (a). On the other hand, accepting some stronger form of divine concurrence is sometimes seen as hard to reconcile with created freedom. I shall therefore briefly address these difficulties, in both in 4.2 (b) and 4.3 (b), focusing again only on some issues connected with embracing or rejecting PCP.

4.2 Concursus indifferens and the Monotropy Thesis (a) The first objection is that divine general concurrence cannot only determine the most general features of created actions. The general line of argument is the following: on the Molinist account the created cause somehow determines what God is going to produce in a given case of concurrence.³² More precisely,

30 Molina (1588), q. 14, art. 13, disp. 26, pp. 170–172. 31 Molina (1588), q. 14, art. 13, disp. 26, p. 174. So Molina repeats Aquinas’s thesis that created actions are not caused partly by God and partly by created agents, but interprets this thesis in a weak way. 32 See e.g. Alvarez (1620), lib. III, disp. 21, pp. 85–86.

54 | Michał Głowala what is eventually caused by God and by His contribution is, in a sense, determined by some secondary agent. Now in a way, this is what happens when forces are composed, and it is also the fundamental idea of the Monotropy Thesis.³³ The problem is, however, that this sort of influence or determination of contributions seems specific to imperfect created causes, in that specific to them that the overall results of their contributions depend on the contributions of other concurring causes. Sunlight passing through colourful stained glass and producing colourful patches on an opposite wall might be a good example of concursus indifferens, with the brightness of the patches being traced back to the light, but their colours to the coloured pieces of stained glass. But once again, it seems specific to the light as a created and imperfect cause that its final effect can be thus modified by the concurrence of other causes. The Molinist theory of God’s will and middle knowledge ensures, of course, that the determination exerted by secondary causes on God’s contribution will not imply any change in God Himself.³⁴ The point of the argument against concursus indifferens, though, is that the very determination of a concurring cause’s contribution by the contributions of other causes is something specific to imperfect and created causes. On the other hand, it seems that, paradoxically, the Molinist account also weakens the created causality itself by stating that it is not due to the created agent that the overall effect is something real. This corresponds roughly to the idea (of Gilson and some other authors - not to mention Aquinas) that no created agent can be a cause of esse. But, as Wippel has shown, Aquinas says rather that created agents do cause esse, but only insofar as they act in virtue of God.³⁵ At any rate, it may seem controversial that the created agent as such is not responsible, in terms of his contribution, for the reality of his actions or their effects. In the case of the decision of a created agent that would mean that such an agent, as such, cannot make any real difference, for the very reality of it is beyond the scope of

33 I am not saying that Molina, for example, would accept the Monotropy Thesis in general; for he makes clear that God’s power may have different manifestations and some of these differences do not depend on the differences between secondary agents at all (Molina (1588), q. 14, art. 13, disp. 26, p. 172). I am only saying that on Molina’s account here, God’s general concurrence fulfills the Monotropy Thesis. 34 Molina presents a sketch of the theory in his explanation of concursus indifferens (Molina (1588), q. 14, art. 13, disp. 26, pp. 172–174). 35 See Wippel (2007). It is worth noting, however, that McCann, who basically does not embrace the Molinist standpoint, repeatedly says that created agents are not causes of esse, and that, for example, by acting freely we do not confer existence upon our acts of will (McCann (2004), p. 33; McCann (2012), p. 101).

Polygeny, Pleiotropy, and Two Kinds of Concurrentist Ontology |

55

his causal contribution – or that I can just join in an action that only counts as real thanks to some other cause. As Aquinas says, any agent, as such, is a cause of esse: omne enim operans est aliquo modo causa essendi.³⁶ Of course, it is a fundamental theistic claim that any created agent causes esse only thanks to God’s concurrence, but to say that esse is totally beyond the scope of the contribution of a created agent (i.e. to admit this sort of division of causal work) is to make created causality pretty weak. To sum up, it seems that the application of PCP in the Molinist concursus indifferens in some ways weakens both divine and created causality (including that which is created and free). By admitting the sort of division of causal work postulated by PCP one makes the causality of both God and the created agent considerably weaker. (b) It may seem, however, that in rejecting the Molinist standpoint here one is rejecting created freedom, or even embracing occasionalism.³⁷ It might be tempting to say that only if, in the (free) action, there is something determined exclusively by the created agent (and not by God, or anybody or anything else), is the action freely chosen by the created agent; otherwise nothing is left to be determined by the created agent itself. As far as causality in general and the threat of occasionalism is concerned, I think, however, that the plausibility of this idea stems (at least to some extent) just from the PCP and the Monogeny Thesis. In other words, it may stem from the presumption that the only possible form of division of causal work is the one postulated by PCP. So it could be stressed again that God the creator does not determine anything in the created action whatsoever in the way in which a performing agent determines it – and, on the other hand, that a performing agent does not determine anything in the performed action in the way in which God the creator determines it. The ways of determining anything which are specific to these concurring causes are, then, mutually irreducible. As far as freedom is concerned, it is important to note a key difference between physical determinism and the case of concursus divinus considered here. According to the incompatibilist thesis relating to physical determinism (the thesis which I accept), if my action is determined in every respect by physical causes, it cannot be freely determined by me: there is no room for my free choice. Now there is no room for free choice there because these two ways of determining or causing are

36 Thomas Aquinas (1961), lib. III, c. 67, n. 1. It is important to note, however, that even a Molinist might say in a weak sense that the created agent causes the esse of his action, because a concurring cause might be said to cause (in a weak sense) something that is beyond the scope of its contribution. 37 That is the standpoint of Dvořák (2013), p. 630.

56 | Michał Głowala incompatible. In the case of concursus simultaneous it is the very reality of free choice itself, or the very entity of free choice itself, that has to be explained by creation.

4.3 Concursus praevius, the Monogeny Thesis and the intertwining of contributions (a) The second objection is that still another form of concurrence – one called concursus praevius – has to be allowed, in order to grasp the causal role of God in the ordinary courses of action of created beings. The difference between these forms of concurrence is precisely this: whereas the concursus simultaneus is concurring with a created agent in causing the action or its effect (influxus cum causa secunda in eius effectum), the concursus praevius is a sort of action exerted upon the created agent (influxus in ipsam causam secundam). Within concursus simultaneus it is not the case that one of the causes makes another contribute to the effect: there are rather two parallel contributions.³⁸ As a matter of fact, Molina claims that there might be a coordination of causes and mutual dependence of their contributions (one of them being real only when the other is real) without any causal influence of one of them on the other. In this sense, the Molinist concursus simultaneus conforms to the Monogeny Thesis: contributions of concurring causes, as opposed to a common effect, are not polygenic. In this sense, concursus simultaneus may be said to be a parallelism of contributions, whereas concursus praevius looks like a sort of intertwining of them. Freddoso’s example of drawing the blue square is a clear case of concursus praevius, and in this case the Monogeny Thesis is false. The man makes the chalk move and the chalk’s own contribution, namely moving along some line, depends upon the man’s contribution, namely moving the chalk along that line. But there may be significantly different ways in which concursus praevius could occur. If there is a divine concursus praevius in the case of decisions and free actions, then not only are my decision or action causally dependent on God, but also my contributions in the generalized sense – namely, my making the decision and my performing the action do causally depend on Him. In such a case, the Monogeny Thesis is false, for my contribution is in a way itself polygenic. We may argue for concursus praevius thus: if my making the decision confers reality upon the decision, and my performing the action makes my action real – that is, if the reality of the overall effect is also within my contribution – then

38 See e.g., Molina (1588), q. 14, art. 13, disp. 26, pp. 167–170; Alvarez (1620), lib. III, disp. 18.

Polygeny, Pleiotropy, and Two Kinds of Concurrentist Ontology |

57

God, as the first source of esse, also must be the source of my contributions. To accept the antecedent one has to reject the Molinist doctrine of concursus indifferens, but, as I have noted in 4.2 (a), it is precisely at this point that the doctrine may well seem controversial. It is natural to think that a created cause makes its actions real just by contributing to them: that is, just by performing them – that it is, inter alia, my own contribution to my actions that makes them something real, or that I do not just join in some action which is real thanks to some other cause. And then, according to the general theistic claim, we may argue that this is possible only because my contributions themselves in fact have their source in God’s contribution – these contributions being intertwined. In other words, for various reasons it may well seem controversial to say that the reality of the created action is due to the contribution of one of the concurring causes as opposed to the other; we therefore have some reasons to reject such an application of PCP to divine concurrence here (i.e. to reject this sort of division of causal work here). But because one has rejected it, one is obliged to then say that the contributions of created agents are themselves caused by God. One thus has to reject the Monogeny Thesis in the case of concursus divinus. But, on the other hand, there are some cases of polygeny, and even some simple ones beyond the advanced ontology of theism, in which both PCP and the Monogeny Thesis are false. So it seems reasonable to accept concursus praevius precisely in the sense of allowing that the very contributions of created agents are somehow caused by God. In particular, you have to accept concursus praevius, the intertwining of contributions, if you want to make created agents responsible (in a stronger sense) for the very reality of what they do.³⁹ This all, however, leaves open the question of what, precisely, concursus praevius is; what I have said presumes only that it is a sort of intertwining of contributions. Diego Alvarez devotes the whole of one of his discussions to that issue after having established that there must be concursus praevius in respect of God’s general concurrence.⁴⁰ He considers four different answers put forward by various Thomists (ibidem, pp. 77–79); they are, so to say, different forms of intertwining of

39 Dvořák suggests that the reasons for rejecting concursus indifferens and allowing for concursus praevius are related to the need to account for providence and God’s foreknowledge. I think, however, that there are some very important reasons relating to the reality of created agents as such, and accounting for their powers to do something real. 40 See Alvarez (1620), lib. III, disp. 19. This shows that for Alvarez the general arguments for concursus praevius are arguments just for the intertwining of contributions, and it is another issue what, precisely, the form of this intertwining is. I think this is very important for understanding the Thomistic standpoint.

58 | Michał Głowala contributions. Here I would like to focus on one of them only, and on Alvarez’s rejection of it. Note first that both on Molina’s and on Alvarez’s account the very action of a created agent is that agent’s contribution insofar as it comes from that agent. It is, at the same time, God’s contribution (his concursus simultaneus) insofar as it comes from God. Now the idea is that the same action is also God’s concursus praevius, insofar as it comes from God and determines the agent (for it is clear that the action an agent actually performs does determine the agent himself). So on this sort of account of concursus praevius, this kind of contribution on the part of God is distinct from God’s concursus simultaneus only in some special sense. And the intertwining of contributions which is specific to concursus praevius takes place within the very created action itself. Admittedly, this sort of answer greatly simplifies the issue of concursus praevius. Alvarez, however, rejects this sort of account of concursus praevius. One of his arguments is especially interesting: he points out that the very action itself is produced both by the created agent and by God, and he claims that God’s concurrence within concursus praevius must be something produced by God alone, and not by the created agent.⁴¹ So this argument employs a version of the Monogeny Thesis: there must be something within concursus praevius which comes exclusively from God and not from the created agent. This shows that in spite of the rejection of the Monogeny Thesis, and even allowing for the intertwining of contributions, the Thomist account of God’s concurrence admits, for various particular reasons, things coming exclusively from one of the concurring causes. For example, moral evil, as such, comes exclusively from created agents, and God’s contribution in concursus praevius is something that comes exclusively from God. But it goes without saying that while rejecting the Monogeny Thesis in its general form – rejecting, that is, at a general level this idea of a division of causal work – one may well still have a number of particular reasons for admitting that something comes exclusively from one of the concurring causes. (b) It may seem again, however, that allowing for concursus praevius necessarily undermines created freedom – the reasons possibly being similar to those pointed out in 4.2 (b). Here, however, I think it could be helpful to remember that in accepting concursus praevius one is accepting just the intertwining of contributions, and such an intertwining of contributions still calls for very careful examination. There are examples of intertwining contributions in cases of using tools,

41 Alvarez (1620), lib. III, disp. 19, pp. 80–81.

Polygeny, Pleiotropy, and Two Kinds of Concurrentist Ontology |

59

as with the chalk in Freddoso’s example, but there may be other examples with other sorts of intertwining. One might say that using something as a tool leaves no room for freedom in respect of that which is used, but there is no freedom in the chalk’s action not because of the intertwining of contributions, but because the chalk as such cannot act freely; it is a question of the chalk’s contribution itself, and not that of any intertwining. Further investigation is required, moreover, into what sorts of intertwining of contributions are possible in the case of free agents.

5 Conclusions My immediate conclusion drawn from the above discussion is to endorse polygeny pluralism: (i) there is a wide range of deeply different sorts of polygeny and sorts of division of causal work; it might be added that, historically speaking, concurrentists have been particularly sensitive to this sort of pluralism. In particular, (ii) there are cases in which PCP has a very strong intuitive appeal and proves heuristically fertile; however, there are also cases, and also beyond philosophical theology, in which PCP and some related theses are clearly false. As for the various ontologies pertaining to concursus divinus generalis, (iii) Molinism seems to accept a weak version of PCP, whereas Thomism rejects it. More precisely, the Molinist doctrine of concursus indifferens involves PCP and the Monotropy Thesis, the Thomistic rejection of concursus indifferens involves rejection of PCP and the Monotropy Thesis, and the Thomistic doctrine of concursus praevius involves a rejection of the Monogeny Thesis while allowing for the intertwining of contributions. Since PCP has very strong intuitive appeal, (iv) the connection between Molinism and PCP may be (at least to some extent) responsible for the intuitive appeal of Molinism itself. On the other hand (v) the rejection of PCP and some related theses in Thomism may be (at least to some extent) responsible for the alleged oddity of Thomistic doctrine with respect to concursus divinus. In general, cases in which PCP is false are still not so well investigated, and some ideas concerning them may well seem counterintuitive. (vi) The intertwining of contributions surely merits further investigation: the debates concerning concursus praevius are part of such an investigation, and such an investigation may shed some important light on the issue of freedom.

60 | Michał Głowala Finally, (vii) there is some sense, pointed to in 4.2 (a), in which Molinism, as opposed to Thomism, weakens the causality of created agents (including free ones) by putting the reality of effects beyond the scope of their contributions.

Bibliography Alvarez, D. (1620), De auxiliis divinae gratiae et humani arbitrii viribus, et libertate, et legitima eius cum eflcacia eorumdem auxiliorum concordia, Lugduni: Sumptibus Iacobi Cardon et Petri Cavellat. Aquinas, Thomas (1903), Tertia pars Summae theologiae, vol. 11, editio Leonina, Romae: Ex Typographia Polyglotta S. C. de Propaganda Fide. Aquinas, Thomas (1929), Scriptum super libros Sententiarum magistri Petri Lombardi episcopi Parisiensis, t. 2, edited by P. Mandonnet, Parisiis: P. Lethielleux. Aquinas, Thomas (1961), Liber de veritate catholicae Fidei contra errores infidelium seu Summa contra Gentiles, t. 2-3, edited by P. Marc, C. Pera, P. Caramello Taurini, Romae: Marietti. Arriaga, R. (1647), Disputationes theologicae in Primam Secundae Divi Thomae, Lugduni. Billuart, C. (1839), Summa S. Thomae hodiernis academiarum moribus acomodat, vol. 1., Lugduni: Apud Pelegaud et Lesne. Carroll, W. E. (2008), “Divine Agency, Contemporary Physics, and the Autonomy of Nature”, in The Heythrop Journal: 49, 582–602. Cartwright, N. (1983), How the Laws of Physics Lie, Oxford: Oxford University Press. Creary, L. (1981), “Causal explanation and the reality of natural component forces”, in Pacific Philosophical Quarterly: 62, 148–157. Duns Scotus, J. (1981), Ordinatio I, dist. 11-25, (vol. V of Doctoris Subtilis et Mariani Ioannis Duns Scoti, ordinis Fratrum Minorum, Opera omnia. Studio et cura commissionis scotisticae praeside P. C. Balić), Romae: Typis polyglotis Vaticanis. Dvořák, P. (2013), “The Concurrentism of Thomas Aquinas. Divine Causation and Human Freedom”, in Philosophia: 41, 617–634. Elders, L. J. (1993), The Metaphysics of Being of St. Thomas Aquinas in a Historical Perspective, Leiden: Brill. Freddoso, A. J. (1988), “Medieval Aristotelianism and the Case against Secondary Causation in Nature”, in Divine and Human Action. Essays in the Metaphysics of Theism, edited by T. V. Morris, Ithaca, New York: Cornell University Press. Freddoso, A. J. (1991), “God’s General Concurrence with Secondary Causes: Why Conservation is not Enough”, in Philosophical Perspectives: 5, 553–586. Freddoso, A. J. (1994), “God’s General Concurrence with Secondary Causes: Pitfalls and Prospects”, in American Catholic Philosophical Quarterly: 67, 131–156. Głowala, M. (2012), “What Kind of Power is Virtue? John of St. Thomas OP on Causality of Virtues and Vices”, in Studia Neoaristotelica: 9(1), 25–57. John of St. Thomas (1949), “Cursus philosophicus”, in Iam-IIae, De habitibus, edited by A. Mathieu and H. Gagné, Quebeci: [s. n.]. Lonergan, B. (2011), Early Latin Theology (Collected Works of Bernard Lonergan, vol. 19, Toronto: University of Toronto Press.

Polygeny, Pleiotropy, and Two Kinds of Concurrentist Ontology |

61

Marmodoro, A. (2007), “The Union of Cause and Effect in Aristotle: Physics 3.3”, in Oxford Studies in Ancient Philosophy: 32, 205–232. Marmodoro, A., McKitrick, J., Mumford, S. and Anjum, R. L. (2011), “Causes as Powers”, in Getting Causes from Powers, edited by S. Mumford and R. L. Anjum, Oxford: Oxford University Press. Massin, O. (2010),“When Forces Meet Each Other”, A draft available at http://www.philosophie.ch/journals/index.php/spps/article/view/85/94; accessed 14.12.2014. McCann, H. J. (2004), “Divine power and action”, in The Blackwell Guide to the Philosophy of Religion, edited by W. E. Mann, Oxford: Blackwell, 26–48. McCann, H. J. (2012), Creation and the Sovereignty of God, Bloomington: Indiana University Press. McKitrick, J. (2010), “Manifestations as Effects”, in The Metaphysics of Powers. Their Grounding and Their Manifestations, edited by A. Marmodoro, New York: Routledge, 73–83. Mill, J. S. (1893), A System of Logic, London: Longmans, Green & Co. Molina, L. (1588), Concordia liberi arbitrii cum gratiae donis, divina praescientia, providentia, praedestinatione et reprobatione, Olyssipone: Apud Antonium Riberium typographum regium. Molnar, G. (2003), Powers. A Study in Metaphysics, Oxford: Oxford University Press. Montesinos, L. (1621), Commentaria in Primam Secundae Divi Thomae Aquinatis, Compluti: Apud viduam Ioannis Gratiani de Antisco. Morris, T. V. (ed.) (1988), Divine and Human Action: Essays in the Metaphysics of Theism, Ithaca: Cornell University Press. Mumford, S. (2009), “Passing Powers Around”, in The Monist: 92(1), 94–111. Mumford, S. and Anjum, R. L. (2012), Getting Causes from Powers, Oxford: Oxford University Press. Penafiel, L. (1666), Tractatus et disputationes in Primam Partem D. Thomae, Boissat. Shanley, B. J. (1998), “Divine causation and human freedom in Aquinas”, in American Catholic Philosophical Quarterly: 72(1), 115–116. Shanley, B. J. (2007), “Beyond libertarianism and compatibilism: Thomas Aquinas on Created Freedom”, in Freedom and the Human Person, edited by R. Velkey, Washington: The Catholic University of America Press. Suárez, F. (1861), “Disputationes metaphysicae”, in Opera omnia, vol. 25, Parisiis: L. Vivés (Reprint: Hildesheim: Olms 1965). Suárez, F. (2002), On creation, conservation and concurrence. Metaphysical Disputations 20– 22, translated by A. J. Freddoso, South Bend: St. Augustine’s Press. Toledo, F. (1615), Commentaria una cum quaestionibus in octo libros Aristotelis de physica ausctatione, Coloniae: Ex oflcina Birckmannica, sumptibus Hermanni Mylii (Reprint: Hildesheim: Olms 1985). Vázquez, G. (1614), Commentaria et disputationes in Primam Secundae sancti Thomae, vol. 1 (qq. 1–89), Compluti: Ex Oflcina Ioannis Gratiani apud viduam. Williams, N. E. (2010), “Puzzling Powers: the Problem of Fit”, in The Metaphysics of Powers. Their Grounding and Their Manifestations, edited by A. Marmodoro, New York: Routledge, 84–105. Wippel, J. F. (2007), “Thomas Aquinas on Creatures as Causes of esse”, in Metaphysical Themes in Thomas Aquinas II, Washington: Catholic University of America Press.

Christian Kanzian

“Bottom-up” versus “top-down” 1 Introduction Important currents in the mainstream of ontology seem to be guided by a specific world-view which posits that there are different levels of reality. The supreme example that seems to legitimate this insight is taken from the seeming divide between micro-physics and the macro-level of our world. Basically, it is assumed that atom-like entities (whatever they are called) constitute the bottom level of reality, and that, moving from the bottom up, we come upon higher levels, including the level of the macro-things of our everyday life-world. This bottom-up world-view has its roots in the corpuscular theories of the 18th century (and earlier), the ultimate model being given by Sir Isaac Newton, who starts in his speculative considerations with “atoms” and ends up with “macro”bodies, which are considered as compositions of the former: “the smallest particles of matter cohere to compose bigger particles, which in turn compose still bigger particles, until the biggest particles, which in turn compose bodies of a sensible magnitude”.¹ Evidently, this modern theory could be traced, in part, to some earlier atomistic theories, but this is not the purpose of this paper. What I intend to do here is to show how this bottom-up world-view influences and determines contemporary ontological theories. To put my cards on the table, I think that such a bottomup approach is the wrong way to go about constructing a reasonable ontological frame. Thus, my angle of approach, here, will be primarily critical. I want to question the presupposed premises of contemporary ontological theories (which are normally left implicit) inasmuch as they give theoretical expression to this (pretheoretical) bottom-up world-view. My work in this paper, then, will be firstly to point out some typical, and usually hidden, bottom-up premises in ontology today (section 2), and secondly, to critique these premises and those ontological theories like supervenience-, emergence-, or constitution-theories which depend on them (section 3 and 4). If I accomplish this, my critical work is done. But I have, as well, certain positive suggestions for which I want to argue: namely that a top-down world-view is a worthy alternative to the bottom-up view, and that we can get to one by putting

1 Newton (1704/1952), p. 394; quoted from Schaffer (2003), p. 499. https://doi.org/9783110566512-005

64 | Christian Kanzian together an appropriate ontological frame (section 5). This, I argue, will bring us to a theistic top-down perspective on our world (section 6). The general topic to which this volume is dedicated is meta-ontology: the exploration and critique of the principles of building ontological theories, bringing to light both their presuppositions and the consistency with which ontologies are built on those presuppositions. I regard the bottom-up versus top-down controversy to be eminently metaontological. In ontology we discuss “what there is”, accounting, on some level of abstraction, for the main characteristics that make it legitimate to say that our assumed entities exist. But the bottom-up/top-down decision pertains to the way in which certain presuppositions have organized the systematic approaches to the theory of what entities exist and what their main characteristics are. Let me illustrate this point by an example: If an ontologist starts from a bottom-up position, presupposing the micro-world as the basis of reality, she/he will consequently deny substances (understood as three-dimensional “endurers” in the sense of Lewis (1986), p. 202) as the basic entities. Processes or metaphysical simples, like tropes (see Campbell (1990), or Simons (1998)), will instead gain this status; the reason is that processes and tropes are much more suited for an ontological interpretation of micro-physical models than substances. His/her ontology will in consequence present itself, implicitly or explicitly, as an anti-substance ontology, since substances would significantly contradict the bottom-up worldview. In short: the denial of substances is an ontological thesis, but the bottom-up presuppositions which lead to this denial, are meta-ontological. I would say that this denial of substances is consistent, since the relevant premises of the chosen meta-ontology commit us to it. The question of whether it is not only consistent but also true depends, and this brings us back to the task of this article, on whether the relevant premises can be legitimated or not. In the case of bottom-up programs, I would contend that they cannot. This observation immediately transitions us to section 2, where I undertake a critical investigation of some of the main or representative premises of bottom-up strategies.

2 Premises of bottom-up programs It is not the task of this article to give a taxonomy of the several versions of bottomup strategies and their differences, but instead to focus on the common premises they share. Without claiming to have a complete list of these premises, I will mention three of them that are critical.

“Bottom-up” versus “top-down” |

65

The first is where we started this paper: the assumption of the ontological priority of the bottom or the basis. Of course, there may be differences in the emphasis put on this claim and the interpretation of the relation between the derived and basis levels. Also the understanding of the basis may differ. But, and this is what I want to assert, for every bottom-up meta-ontology, the priority-thesis is crucial. In other words: For every ontological theory which is committed to a bottom-upmeta-ontology, this assumption is essential. The second premise, inseparable from the first, is that all bottom-up strategies must suppose that we can successfully reconstruct higher levels from the basic one. We must acknowledge and account for the possibility of reconstructing the inhabitants of our macro-world from the assumed bottom- or basis-entities. Otherwise we must either cease to think of ontology as a theory about everything, or we must deny that the inhabitants of our ordinary macro-world belong to the scope of “everything”, in effect denying their existence. It must be possible either to explain how Susan, the sheep on the grass, fits into the ontological framework that begins with the atoms or the processes, or this sheep must be, as it were, ontologically butchered – thrown out as an existent or an entity. The third premise is that the supposed base level is a domain explored at first by scientists, normally physicists or quantum mechanicians, who give us the evidence upon which we proceed to theorize. Ontologists, then, have to look first at the results of quantum mechanics as the preliminary findings of their own theorizing. Ontology thus becomes an a posteriori-discipline.² Quantum mechanics informs us about the basic elements of reality, technically spoken of entities. The third premise has strong naturalistic or physicalistic implications. So it is legitimate to state, that standard bottom-up strategies are expressions of naturalistic or physicalistic attitudes in philosophy.

3 Against premise three: natural sciences exhibit entities In this article I do not intend to discuss fundamental questions concerning naturalism or physicalism. It is also not the place to ask whether there may be nonnaturalistic strategies that still defer to the third premise. I rather want to illustrate the methodological or meta-ontological difficulty of the mentioned premise three of bottom-up programs. To restate the premise, ontology has no choice but

2 Simons (1998), p. 251.

66 | Christian Kanzian to start with the results of quantum-mechanics. However, there remains the question: what these results that are so dear to the holders of a premise three programm should be? Certainly, they are not the uninterpreted bare empirical data. No ontologist finds the data provided by a scientist operating with an electronic microscope particularly pertinent. And even if he were interested in them, he does not, as an ontologist, have the training to understand what he sees. What philosophers need are interpretations. And the first level of interpretation contains models physicists use to come to theoretical explanations of the given data. Simple material elements for instance (and it does not matter if they call them atoms, or electrons, or quarks) are such models. And it seems to be the case that successful physical theorizing relies on such models. The problem is that philosophers who assume premise three and are eager to start their ontological projects with the results of physics must consciously or unconsciously import such models, taking them – and this is the decisive point – as the hard ontological facts. To remain with this example we can state that they regard material simples (some actually call them “atoms”, some other “simples”, some interpret them as qualities and call them “tropes”) as the basic units of reality and give them genuine ontological characteristics like primitiveness or indivisibility. The result is an atomistic ontology in the style of Newton’s speculation: material simples, called “atoms”, “simples”, or “tropes” are taken as primitive and undividable basic units of reality. They are regarded as the basic category of entities, from which all the macroscopic phenomena can be reconstructed. The case of material simples is just one example, but a very interesting and influential one; because it has given rise to leading theories within the project of bottom-up oriented ontology. To repeat the point: what is going on here, it seems, is ontologizing or hypostasizing physical models without a philosophical argument as to why we should do so. In fact there do not exist, in a proper sense, simple, primitive, undividable material units (whatever you call them). “Atoms do not exist” seems to be a rather dangerous thesis. I am aware of that and, therefore, must clarify here that I, of course, do not deny the usefulness of models for interpreting the empirical data we receive from micro-reality. However, my point is that we have no reason to think that we should convert models into entities. In so far as bottom-up strategies rely in their meta-ontology on premise three, and import, as a posteriori ontologies, entities from science, they can be criticized for giving us hypostatisation without presenting any reason for accepting it. If we argue, on the contrary, that there are no entities here on the microphysical level as constructed by science, we then also collapse premise one, that is, the claim that ontological priority must be given to the micro-physical basis. In other words: in so far as premise one, as a genuine expression of the theoret-

“Bottom-up” versus “top-down” |

67

ical part of bottom-up meta-ontologies, depends on premise three, it fails when premise three is shown to be ontologically invalid. With these results we can proceed to our next step:

4 Against premise two: there is a successful reconstruction-procedure In my criticism I want to address and call into question two standard reconstructionprocedures. The first consists of the family of procedures grouped under the master concept of “supervenience”. My meta-ontological criticism is that standard supervenience-theories conflate the description of a phenomenon with its explanation. The second prominent bottom-up strategy for ontological reconstruction consists in those oriented towards “constitution”, or the attempt to show painstakingly how lower levels constitute higher ones. This requires a refutation on an ontological level.

4.1 Supervenience The introduction of “supervenience” into the philosophical vocabulary took place at the beginning of the 20th century, and was first used in ethical theories: a specific correlation between moral and natural predicates was described as being supervenient in so far as once the natural or non-evaluative predicates of an object or act are determined, its moral or evaluative predicates are also thereby determined; and in so far as two objects or acts cannot differ in their moral or other evaluative predicates unless they also differ in some non-evaluative ones. The assertion of supervenience is a descriptive claim concerning the covariation of different groups of predicates, and as such may have its theoretical merits. But during the last 30 years supervenience has inflated beyond its modest beginnings, taking on an immense role not only in ethics, but also in semantics (the truth of sentences supervenes on being (Bigelow)), social theory (the social sphere supervenes on individuals (Campbell)), and philosophy of mind (mental events supervene on physical ones (Davidson)).³ In this march of supervenience imperialism, what has been left behind is its original usefulness as a descriptive

3 See e.g., Bigelow (1988), Campbell (1990) (here esp. chapter 7.8), Davidson (1980) (here esp. Mental Events).

68 | Christian Kanzian claim that there are correlations between truth and beings (“truthmakers”), social entities and individual ones, mental and physical states, and it has assumed the dignity of an explanatory claim that provides an understanding of the relation between the supervenient and its basis. To put it another way, the description of a correlation was transformed into an explanation of the relation of the correlated, as for instance in philosophy of mind, where it is used to explain the relation between mental and physical events. Given the triumphant march of supervenience through these branches of philosophy, it is no surprise that it would turn up as a trump card for bottom-up ontologists trying to provide a reconstructionprocedure of “higher levels of reality” from lower ones.⁴ I do not intend to refute all applications of “supervenience” in ontology. Jaegwon Kim, the “father” of the ontological connotations of the term, explicitly states and defends the view that supervenience occurs not only between groups of predicates or other linguistic expressions, but also between properties and events which are, according to Kim’s understanding, entities.⁵ It was also Kim’s merit to distinguish the modal strength of different supervenience-correlations, by differentiating between weak, strong, and global supervenience.⁶ But Jaegwon Kim himself seems to be aware of the original meta-theoretical status of supervenience. He regards “supervenience” as part of “a phenomenological claim”⁷ that something, a correlation between groups of entities, is the case and can be described in a (modally) differentiating way. According to him, “supervenience” never can be the solution of a problem, particularly not of the problem of how the relation between correlating groups of entities can be explained. John Heil has also made this point: “If As supervene on Bs, then the question is: what is it in virtue of which this is so? [...] If all we know is that As supervene on Bs, we know only that As covary with Bs”.⁸ The assertion of supervenience is a claim concerning co-variation, not an explanation why the co-variation occurs. We actually do not know in virtue of what the asserted co-variation occurs.

4 See Campbell (1990), e.g., p. 37. Campbell holds a supervenience-reconstruction-view which implies a strong version of the “priority of the bottom”-thesis, since he assumes the supervenient levels as “pseudo-additions” to reality. This marks the difference to emergence or emergentistic versions of bottom-up strategies. Emergent levels are no pseudo-additions. They are real, in an ontologically relevant sense. Entities do emerge from other entities. But the meta-ontological problem of mixing up description with explanation pertains, to my opinion, also to emergentism. 5 See also Kim (1993), pp. 175 ff. 6 Cf. Kim (1993), pp. 58 and 64 (weak supervenience), p. 65 (strong supervenience), p. 68 (global supervenience). 7 Kim (1993), p. 168; see also Kim (1996), p. 237. 8 Heil (2003), p. 67.

“Bottom-up” versus “top-down” |

69

Here is where my critical view of the way supervenience has been used in bottom-up ontological accounts gains its traction: If, as I contend, Kim, Heil, and others are right that “supervenience” is a description that only becomes an explanation via a transparently faulty hypostasis, ontologists who use “supervenience” in explanations of the (assumed) relation between different levels of reality, beginning at the micro-physical basis and ending, bottom up, at the macro-world (and I fear they must use supervenience as an explanatory concept to say anything interesting in order to solve their problem, the reconstruction-problem), have fallen victim to the supervenience “hype”. And their failure is at first meta-theoretical, because it pertains to an inconsistency between the real range of a theory (description of co-variation between elements of different groups) and the intended theoretical purpose (the explanation of the relation between these elements). From which I conclude that we can safely discount supervenience-theories in the context of reconstruction-procedures of bottom-up programs.

4.2 Constitution Not all bottom-up programs rely on supervenience- or emergence-⁹ theories, and thus not all of them are susceptible to the meta-ontological critique that they have turned a description into an explanans. The strategies upon which we now focus try to reconstruct the macro-world from the micro-world by offering a veritable explanation how this reconstruction can take place: higher levels are constituted by lower ones. Here, I want to use, as a teaching example of this kind of theory-making and its problems, Lynn Rudder-Baker’s book The Metaphysics of Everyday Life.¹⁰ Baker would not regard herself as physicalistic or (in a narrow sense) naturalistic philosopher. Nevertheless her theory is (essentially) committed to premise two, and thus can stand, in this narrow sense, as a typical version of this bottom-up strategy. The central concept of Baker’s reconstruction of “Everyday Life” is “constitution”. Her agenda is realized through a detailed (and technical) interpretation of the “constituting” relation; in turn, this interpretation is supposed to explain how the assumed multi-layer picture of the world works (ibid, pp. 237 ff). This distinguishes her theory from standard supervenience-theories.

9 See footnote 4. 10 Here: Baker (2007).

70 | Christian Kanzian According to Baker, “[t]he fundamental idea of constitution is this: when a thing of one primary kind is in certain circumstances, a thing of another primary kind – a new thing, with new causal powers, comes to exist”.¹¹ The basic idea of constitution, according to Baker, is that constitution is a relation which occurs between things, taking “thing” as a technical ontological term. Things are examples of primary kinds. It is essential for things to belong to a kind or to a species, which determines the thing’s identity.¹² By constitution a thing of a primary kind F becomes a thing of a primary kind G (which is different from F). If things depend in their identity on their kind or species, constitution causes an ontological difference. The constituted thing (of kind G), x, cannot be the same as the constituting one (of kind F), y. Nevertheless, between x G and y F (which are related by constitution) there is a proper unity, a unity which is regarded as being contingent and temporally limited. Baker’s illustrating example for constitution, understood in this sense, is the relation between a statue and its stuff or its material. According to Baker a lump of bronze constitutes, under some suitable circumstances, the statue. Their unity is consequently interpreted as accidental or contingent: One and the same statue may be constituted in the course of its history by different parcels of material matter. And it may be the case that one and the same lump of bronze constitutes, at different times, different statues. Baker also applies such considerations to other examples in our everyday lifeworld, e.g., to the relation between a piece of plastic and a traffic sign. But she also explicitly urges that the same relation of constitution occurs on “deeper” levels of realities, which are different from the level of our macro-world, even between the levels of atoms and molecules and their aggregates.¹³ If science wants to teach us about sub-atomic spheres, we have to postulate things as the basis of constitution even there. This argument seems to be a proper opening to introduce my critique of Baker’s theory. It is clear that she presupposes that on all assumed levels of reality we can find entities, which are supposed to be instances of the category of things. I, on the other hand, see no reason to believe this supposition. Let me argue with reference to the first level, which is represented by the statue/bronze or the traffic-sign/plastic example. There are good reasons why we do not take the material-aspects of things, e.g., a lump of bronze or a portion of plastic, as things themselves, since we ordinarily don’t classify lumps and portions of matter as

11 Baker (2007), p. 32. 12 Cf. Baker (2007), pp. 33 ff. 13 Baker (2007), pp. 35 ff.

“Bottom-up” versus “top-down” | 71

members or instances of sorts or kinds of things that provide criteria for the identity of their instances and thus, e.g., for their definite countability. But individuals of the category of things need such a sortally determined identity and must be countable in a proper or, as Lowe puts it,¹⁴ “determinate” sense. This becomes especially clear in the case of living-beings, whose identity is obviously determined by their sort or their species, and can thus be counted in a determinate way: definitely, we would say, there are three and not four sheep on the grass. This kind of identification is also applied to artefacts, like statues or traffic signs, even if their sorts function in their identity-determining role in another way than non-artificial things. But when we come to lumps of matter, we immediately differentiate them from things for which sort-dependent identity-criteria allow determinate counting. Thus lumps of matter are better categorized as quasi-individuals,¹⁵ even if they may become – as suitable circumstances determine – the material-aspect of things. But if lumps of bronze and pieces of plastic are quasi-individuals, and thus no members of the category of things, Baker’s notion of constitution in the context of a bottom-up program is blocked. Until now I have considered quasi-individuals of our macro-world. Derivatively, we can argue that at the assumed micro-physical basis there do not exist things. In section 3 above, I have argued that, instead of entities, what microphysics deals with are models that are set up to allow for the interpretation of pertinent empirical data. But if micro-physics is not a matter of entities, then it surely cannot be a matter of things, with their fixed identities, which could be the basis of Baker’s constitution. Given the insurmountable problems with both the supervenience agenda and the constitution agenda, the explanatory gap of bottom-up programs seems to constitute a sufficient reason to reject them. This isn’t to say that I have dealt with all possible alternatives in the family of supervenience (emergence) and constitution ontological theories, but only that these pseudo- or highly problematic explananda seem to assert themselves in any bottom-up ontological schema. Our everyday life-world cannot be considered as supervening (emerging) or being constituted by inhabitants of the assumed micro-world. What does this mean? I would point to the bottom-up direction as the villain in this story of meta-ontological malfunction. If we are looking for a truly explanatory account, we should move to the contrary position, and explore topdown strategies. In the following section 5 I try to give an outline of such a meta-

14 Lowe (1998), p. 76. 15 Ibid.

72 | Christian Kanzian ontological alternative, which, to say it in advance, can at best be realized in macro-world ontologies by taking things or substances (instead of atoms or other simples) as the primary entities.

5 Top-down integration as the alternative to bottom-up reconstruction To present top-down-ontology as the alternative to bottom-up strategies, we can start with the top-down supposition that it is our macro- or everyday life-world where we find the basic elements of reality. Such a supposition may sound harmless, but it has very broad implications. Actually it is a rather radical thesis. It is not the micro-sphere where we find the basis of reality, as it is assumed by the bottom-up premises. It is the macro-world, the world, in which you and I, our cats and dogs, cars and computers, occur. This implies a quite different world-view from that animating the bottom-up paradigm. What also matters is that the task of ontology changes relative to bottom-up meta-theories: Ontology describes the basic structures of our macro-world according to its own terminological and methodological tools. Ontology does not explain the macro-world, in the sense of a bottom-up approach to reality. What are suitable candidates for elements building up the basic structure of our everyday life-world? As we have already indicated, we can endow things or substances with this role.¹⁶ It cannot be the aim of this article to provide a full interpretation of this classical ontological category.¹⁷ Standardly substances are regarded as entities, which exist in a properly independent way, which are – due to their three-dimensionality – identical with themselves not only at each moment of their existence, but also throughout their history, and which are (thus) bearers of properties, proper capacities, and changes. Important for the context of this article is to give an idea of how we could analyze substances in an ontological way, committed to top-down meta-ontology. One possibility would be to describe substances as primary units with a complex inner structure. “Complex” means that substances are compositions of different

16 The claim that substances are the primary macro-entities is not only accepted by their ontological friends but, also, by their ontological opponents, especially those who (due to bottom-up intuitions) deny the ontological importance of our macro-world, and therefore the relevance of substances. See e.g., Simons (1998). 17 See my Ding – Substanz – Person (here Kanzian (2009), in German).

“Bottom-up” versus “top-down” | 73

elements or aspects: an individual material- and an individual form-aspect: What a thing is made of and how the components are built into the unity of the whole substance. Both the “what” and the “how” aspects are irreducible to each other in their functions for the composition of the whole. That substances are nevertheless “primary units” implies that material and form are not entities on their own right; that is, they are not entities which would constitute bottom-up the substance at stake and provide a secondary status to her. The primary-unit-status of substances allows for material and form to be aspects which can be derived from some topdown analysis. The division between material and form is not the only possible result of an analysis of substances. There may be a wider range of further methods of topdown analysis of a complex macro-substance. One of them would be to focus on the details of the ontological functions of material and of form, that e.g., the material aspect may be understood as the ontological basis of space and spatial relations, or that the form aspect is a thing’s principle of individuation.¹⁸ A physical investigation into the material aspect of any thing may subsume it in the framework of very informative physical theories, and may legitimately use illustrative models, such as our above-mentioned cases of atoms, protons, and – if you want – quarks. But in the top-down case, we cannot cede primacy to this necessary scientific model-making. If we go from the top down according to the particular methodological tools of a specific science, we leave the level of ontological reasoning. We enter the sphere e.g., of (micro-)physics, which of course, can be integrated into the general and universal frame of an ontological theory. This remark leads us to a further top-down-characteristic task of ontology: ontology integrates the results of scientific investigations, instead of importing scientific models as the primary given in their ontological scheme.¹⁹ Ontological concepts like “individual forms” or “individual material” may be understood as parts of a general conceptual frame, in which less general terms, e.g., physical or biological ones, can be brought into a systematic context. This presupposes that there are also e.g. biological interpretations of what we call in ontology “individual form”, or physical explanations of the “individual matter” of a thing; and that ontology can interpret scientific interpretations and explanations due to its own methodology. For example, if we understand biology as the science that regards life or the sum of life-functions as the “how”-aspect of living beings, ontology, in

18 See Baker (2007), where forms are called “unifying principles” (ibid., p. 145) which brings the material to the unity of “robust objects” (ibid.). 19 Cf. Smith and Klagges (2008), p. 21 ff, who focus on this “integrative function” of ontology.

74 | Christian Kanzian its top-down guise, would interpret life as the individual form and as the principle of individuation of a certain kind of substance. This paper would have to extend to an unwieldy length if I were to deal in an integrative way with all the conceptual tools of science, e.g., “power” or “capacity”. My aim is just to make the reader aware of the integrative function of a top-down-ontology. One of the distinct advantages of the top-down program is that it avoids those problems of bottom-up procedures to which we have called attention. We do not need to account, on the primary level, for atoms or other micro-world“entities”, since we assume that the direction we are moving in starts from basic elements of reality that are found in our macro-world. We don’t, therefore, need a reconstruction-strategy, since there is nothing which we have to reconstruct. Our cars, dogs, and last but not least, we ourselves are given as primary entities. With this result let me switch to the final step of this article: the question for the compatibility of (meta-) ontological programs with the theistic perspective.

6 The theistic perspective A theistic perspective on reality may be characterized by the convictions that there is a personal God, who has not only created the world, but keeps it in its existence permanently, and who gives an ultimate aim to all individual creatures as well as to the world or cosmos as a whole. The question we are dealing with in this final section is: how does a top-down ontological approach apply to the theistic vision? Before I give an answer, I want to repeat that the field we have been working in is meta-ontological, and deals with the assumptions of ontological theories, rather than properly ontological. We have also distinguished between the theoretical and the pre-theoretical dimensions of meta-ontology. The former provides premises for ontological theory-building; the latter views such assumptions as constituting a specific world-view.²⁰ Religious convictions, like the acceptance or

20 Just for the sake of terminological clarification let me add that the proper “locus” of the bottom-up/top-down distinction is (as I hinted in the introduction) the pre-theoretical dimension of meta-ontology. Bottom-up or top-down approaches are at first distinctions concerning world-views. In an analogous sense we can also regard theoretical premises as bottom-up or topdown: inasmuch as they are consistent with or expressions of pre-theoretical attitudes. In such an analogous sense whole meta-ontological programs or strategies can be labelled as bottomup or top-down: that is, they rely on particular premises that express particular pre-theoretical attitudes.

“Bottom-up” versus “top-down” | 75

the neglecting of a theistic perspective, I would regard as belonging to the pretheoretical part of a meta-ontology; even more: religious convictions are the coreelements of world-views. On the one hand, we cannot convict all advocates of bottom-up programs as having atheistic or anti-theistic world-views – nor are top-down programs only consistent with “theistic” world-views. On the other hand, it can be assumed that world-views which are inspirited by theistic assumptions are coherent with theoretical meta-ontological premises, which are significant for top-town programs. And that pre-theoretical anti-theistic attitudes, like materialistic or physicalistic ones, are standardly regarded as being compatible with premises which make up the theoretical part of bottom-up meta-ontologies. In short: Standard bottom-up programs are largely determined (in their pretheoretic part) by anti-theistic perspectives – that is, perspectives which exclude the hypothesis of a divine creator. Top-down stances more typically include the hypothesis that there exists such a creator. So are substance-ontologists theists and atomists and tropists atheists? This is a too simple statement of the case. The connection between an ontological theory and a religious or anti-religious conviction is never a conceptual or a priori matter. But, nevertheless, we also cannot deny that a substance-ontology, insofar as it is formulated as, eventually, an expression of theistic pre-theoretical attitudes or a religious world-view, is evidently ontology constructed from a theistic perspective. And those philosophers who argue for some kind of atomistic ontological approach to reality are mostly guided by meta-ontological principles which are consistent with, or at least do not conflict with, materialistic, and thus antireligious, pre-theoretical attitudes, Among this group, we are much more likely to find ontologists who deny the theistic perspective. I am aware of theistic oriented materialists, or atheistic writers who sympathize with substances. That is why I concede that the bottom-up versus top-down controversy does not break down absolutely into a theism versus atheism debate, conceptually. Nevertheless this controversy is a promising forum for bringing these different perspectives on the world into a philosophical dispute. This is perhaps one of the most interesting reasons for taking it seriously.

Bibliography Baker, L. R. (2007), The Metaphysics of Everyday Life, Cambridge: Cambridge University Press. Bigelow, J. (1988), The Reality of Numbers: A Physicalist’s Philosophy of Mathematics, Oxford: Oxford University Press. Britton, T. (2012), “The Limits of Hylomorphism”, in Metaphysica: 13, 145–153.

76 | Christian Kanzian Campbell, K. (1990), Abstract Particulars, Oxford: Blackwell. Davidson, D. (1980), Essays on Actions and Event, Oxford: Clarendon Press. Heil, J. (2003), From an Ontological Point of View, Oxford: Clarendon Press. Kanzian, Ch. (2009), Ding – Substanz – Person, Frankfurt: Ontos. Kim, J. (1993), Supervenience and Mind. Selected Philosophical Essays Jaegwon Kim, Cambridge: Cambridge University Press. Kim, J. (1996), Philosophy of Mind, Boulder: Westview Press. Lewis, D. (1986), On the Plurality of Worlds, Oxford: Blackwell. Lowe, E. J. (1998), The Possibility of Metaphysics, Oxford: Clarendon Press. Newton, I. (1704/1952), Opticks: Or a Treatise on the Reflections, Refractions, Inflections, and Colours of Light, New York: Dover. Schaffer, J. (2003), “Is There a Fundamental Level?”, in Nous: 37, 498–517. Simons, P. (1998), “Farewell To Substance: A Differentiated Leave-Taking”, in Ratio: 11, 235– 252. Smith, B. and Klagges, B. (2008), “Philosophie und biomedizinische Forschung”, in Biomedizinische Ontologie, edited by L. Jansen and B. Smith, Zürich: vdf Hochschulverlag, 17–30.

Daniel Linford and Jason Megill

Cognitive Bias, the Axiological Question and the Epistemic Probability of Theistic Belief 1 The axiological question and pro-theism Analytic philosophy of religion has generally focused on a handful of questions. Of course, much of the literature concerns the question, “does God exist?”; some attempt to defend or refute the traditional arguments for or against Theism (the Cosmological argument, the Ontological argument, the Problem of Evil and so on), while others attempt to devise new arguments for or against the existence of God. Some of the literature concerns the divine attributes; e.g., what properties does God have if God exists? And some of the literature focuses on other questions. But one question that has only recently started to generate discussion is: “would it be good or bad if God exists?”¹ What difference would the existence or nonexistence of God make to the value of the actual world? That is, would the actual world be a better or a worse place if God exists? Call this the “axiological question”. The axiological question is distinct from the question of whether God exists: for example, one might think (and some have thought) that even though God does not exist, it would be better if God did exist.² One possible response to the axiological question is “Anti-Theism”.³ Antitheism claims that the world would be significantly worse if God existed than if God did not exist. There are not many extant anti-theistic arguments, though some can be found in Nagel (1997), Hitchens (2007), and Kahane (2011). Hitchens

1 For work on this issue, see Kahane (2011); Kahane (2012); Kraay and Dragos (2013); and Mawson (2012). 2 Russell and Camus are two Non-Theists who suggested that the world might be better at least in certain respects if God existed. Kahane (2014), p. 476, writes, “Russell thought that, in the absence of God, we must build our lives on ‘a foundation of unyielding despair’.” And “Albert Camus is famous for expressing this kind of perspective, suggesting that the lack of an afterlife and of a rational, divinely ordered universe undercuts the possibility of meaning (Camus (1955))” (Metz (2013), Section 4). A more recent example is Tooley (2009), p. 311: “I also think that it would be very good if it turned out, contrary to all probability, that God did exist, for while the existence of such a deity would not entail that this was the best of all possible worlds, it would ensure that the world was very good indeed.” 3 See Kraay and Dragos (2013) for a taxonomy of the different positions one might take on the axiological question. https://doi.org/9783110566512-006

78 | Daniel Linford and Jason Megill described his view as “anti-theism”, by which he meant that: "we ought to be glad that none of the religious myths has any truth to it" (Hitchens (2007), p. 102). Likewise, Nagel (1997), p. 130, remarks, “I hope there is no God! I don’t want there to be a God; I don’t want the universe to be like that”. Kahane (2011), p. 8, elaborates: I suspect Nagel has in mind something like the following. A world in which God exists is a world where human beings stand in a distinctive and inescapable relation to another person. It is a world where we are the subordinates of a moral superior, a superior that deserves our allegiance and worship ... The idea is that God’s existence is logically incompatible with the full realization of certain values. Thus a world in which God exists is a world where we would not be the moral equals of all other rational beings – equal members of a kingdom of ends that has no ruler. Such a world seems incompatible with complete independence, or with complete privacy and genuine solitude ...

See Kraay and Dragos (2013) for a discussion of these claims. Another possible response to the axiological question is “Pro-Theism”. ProTheism claims that whether God exists or not, God’s existence would make the actual world significantly better. Regardless of one’s position on the existence of God, i.e., whether one is a Theist, Atheist or Agnostic, Pro-Theism is not obviously false. Indeed, one can develop various arguments for it. For example, (1) If God exists, then God is all-good, all-knowing, all-powerful etc.

This is simply the traditional conception of God from classical theism. Furthermore, (2) If God is all-good, all-knowing, all-powerful, etc., then our lives are not meaningless, or absurd, or pointless.

Presumably, a perfectly good God would not create beings that have a meaningless existence. And (3) follows from (1) and (2) with Hypothetical Syllogism, (3) If God exists, then our lives are not meaningless etc.

So, the existence of God would guarantee that our lives have meaning, whatever that meaning might be. This is not to suggest that God’s existence is the only way our lives could have meaning; rather, it is simply that God’s existence would ensure that our lives have meaning. God’s existence might be sufficient, though not

Cognitive Bias, the Axiological Question and the Epistemic Probability of Theistic Belief |

79

necessary, for our lives to have meaning.⁴ So, if God exists, then we know for sure that our lives are not meaningless; this seems preferable to not knowing whether our lives have meaning, and so seems to be at least one way that the existence of God would make the actual world better than it might otherwise be. Here is another argument for Pro-Theism. As Kraay and Kraay and Dragos (2013) point out, it is a logical truth that God cannot allow any gratuitous evil. An all-powerful, all-knowing, all-good creator would not and could not create a world with senseless or avoidable suffering. So, (1) If God exists, then gratuitous suffering does not exist.

But also note that, (2) If gratuitous suffering does not exist, then the actual world is better (in at least one respect) than it might otherwise have been.

The fundamental idea behind (2) is that suffering that serves some purpose or other, whatever that purpose might be, is somehow better than pointless or gratuitous suffering. Pain that exists in the wake of life-saving surgery is better than pain that exists for no reason at all. Soreness that exists because of exercise is better than some random, pointless soreness. So it would be better if there was no gratuitous suffering; but then if gratuitous suffering does not exist, the world is better – in at least one respect – than it might have otherwise been. Finally, we can infer with hypothetical syllogism and (1) and (2) that, (3) If God exists, then the actual world is better (in at least one respect) than it might otherwise have been.

Again, the idea is that if God exists, then our suffering must serve some purpose, whatever it might be. And this would make the world better – in at least one respect – than it otherwise might have been. One could also defend Pro-Theism with what can be called “The Argument from Justice”. Premise (1) is, (1) If God exists, then God is perfectly just.

4 To be clear, some have defended the view that God is necessary for life to have a meaning; see, e.g., Craig (1994), Haber (1997), and Gordon (1983). But we do not endorse that view. We are simply saying that God’s existence might be sufficient for life to have meaning; and as we’ll see below, our main argument in this paper does not even depend upon this claim.

80 | Daniel Linford and Jason Megill Given classical theism, this premise is true by definition; God is thought to be a being that is perfectly just, whatever the nature of justice might be. Moreover, (2) If God is perfectly just, then after death people will get the reward or punishment that they deserve.

We don’t know if anyone deserves to be punished in hell for eternity, or if no one does. We don’t know if anyone deserves to be rewarded in heaven with eternal life or not. For all we know, it might be that no one deserves an afterlife of any kind. But whatever it is that people deserve, that is what they will get if God exists and God is indeed perfectly just. Furthermore, (3) If people get the reward or punishment that they deserve, then the actual world is better (in at least one respect) than it might otherwise have been.

The idea is that it is somehow bad when someone deserves something – whether it is a reward of some sort or a punishment of some sort – and they do not get it. It is bad, e.g., if an innocent man spends his life in jail. It is also bad if someone gets away with some heinous act. But if (1), (2) and (3) are true, then so is, (4) If God exists, then the actual world is better (in at least one respect) than it might otherwise have been.

If (1) P → Q, (2) Q → R, and (3) R → S are true, then so is (4) P → S; so the argument is valid. And the premises appear plausible. Indeed, Kahane (2011) suggests that one reason many Theists are Pro-Theists is that if God exists, then the virtuous and the vicious will receive the proper reward or punishment, and that is thought by some to be a very positive outcome of God’s existence. A common theme in these arguments is that the nature of God puts constraints on the sort of world that God can create; by definition, God must create worlds in which certain goods obtain; e.g., life has meaning, suffering has purpose, justice occurs, and there are probably other examples too. So, if God exists, then the world will contain these goods, and so will be better in at least some respects than it might have otherwise been. We suggest, again, that Pro-Theism is not obviously false. It is not too difficult to think of ways the actual world would be better if God exists than if God does not exist. And it is not difficult to see why some might endorse Pro-Theism, whether they are Theists or not. And again, Pro-Theism and Theism might very well have different truth-values; e.g., an Atheist could consistently endorse Pro-Theism; the Atheist should not simply unreflectively reject Pro-Theism.

Cognitive Bias, the Axiological Question and the Epistemic Probability of Theistic Belief |

81

However, our argument below does not depend upon the claim that Pro-Theism is true. Rather, we appeal to a much weaker claim: whether Pro-Theism is true or false, many Theists think that it is true. That is, many – including Theists who are not particularly philosophically oriented – think that it would be much better if God existed than not. Theists think that if God exists, various good outcomes would occur that would not or at least might not happen otherwise; for example, many think that it would be better if God existed because then there would be an afterlife, which would be a good thing. In short, regardless of the truth-value of Pro-Theism, many Theists think that it is true. We think it is obvious that many Theists are Pro-Theists, even if implicitly. For example, churches are filled with people who think that God exists and this is a good thing; the universe is better because God exists. It is not as if your average churchgoer, or at least the average churchgoer who is there through their own volition, wishes there was no God because then the universe would be improved in certain respects. It is not as if Theists think God’s existence is a bad thing and only begrudgingly worship God. That is, we agree with Kahane (2011) that the vast majority of Theists not only believe that God exists but also want God to exist; they want God to exist (at least in part) because they think the universe is better in various respects because God exists, and furthermore, since God is perfectly good, simply “adding” God to any universe would increase the universe’s value.

2 Cognitive bias and religious belief Over the last several decades, psychologists and others have empirically studied various cognitive biases that plague human cognition; these biases often lead to irrational and/or false beliefs. In this section, we first discuss this literature in general terms. We then discuss two previous attempts to explain religious belief specifically as being (at least in part) a product of cognitive bias; Hume’s famous discussion of miracles and a recent paper by Draper and Nichols (Draper and Nichols (2013)). Finally, we discuss several specific cognitive biases (e.g., the valence effect) that play a role in our main argument. The cognitive psychologists Kahneman and Tversky did groundbreaking work on cognitive bias beginning in the early 1970s (see, e.g., Tversky and Kahneman (1971), Kahneman and Tversky (1972) and Kahneman and Tversky (1973)). They demonstrated that humans often reason in ways that violate logic, rational choice theory and basic mathematics. The most famous case discussed by them involves “Linda” (Tversky and Kahneman (1983)). Linda was described to participants in their study as someone concerned with social justice. Participants

82 | Daniel Linford and Jason Megill were then asked which was more likely: (i) Linda is a bank teller or (ii) Linda is a feminist and a bank teller? Most claimed it was more likely that she was a feminist and a bank teller. Of course, this answer violates basic probability theory; given possibilities (A) and (B), the probability of (A) and (B) both holding cannot be greater than the probability of just (A) holding. So it appears people violate basic probability theory when reasoning. Kahneman and Tversky, and other researchers who subsequently started studying cognitive bias, have since discovered dozens of cognitive biases: some infect decision making; some affect our judgment of causal relations; some affect our memory; others involve mistakes in reasoning that we make when risk is involved; others involve biases that arise while making plans; and still others affect some other aspect of human cognition. To offer some examples of particular cognitive biases from the literature, some of which have since become well known: “confirmation bias” is the tendency to only pay attention to evidence that confirms one’s previous beliefs. The “anchoring effect” is the tendency to overemphasize one bit of information to the exclusion of other pieces of information. “Bias blind spot” is the tendency to see oneself as being less prone to bias than others. “Hind-sight bias” is our tendency to see past events as being more predictable than they were. The “framing effect” is our tendency to draw different conclusions from the same information because of variations in how that information is presented. We have a tendency to assign a higher probability to claims that rhyme (the rhyme as reason effect); we overestimate the degree of control that we have over situations (illusion of control); and we evaluate the worth of a decision based on its outcome rather than by examining the available information at the time (outcome bias). Moreover, we systematically underestimate the time a task will take to complete (planning fallacy); we ignore negative situations (Ostrich effect); and we like things simply because we are familiar with them (mere exposure effect). One cognitive bias that has become well known is the “Dunning-Kruger effect”: those who are not skilled at something tend to overestimate their competence, while those who are skilled at something tend to underestimate their competence (see, e.g., Kruger and Dunning (1999)). But these are only some examples; again, cognitive psychologists have uncovered dozens of biases; for a discussion of many of these cognitive biases, see, e.g., Pohl (2004). Beliefs that are formed as a result of these biases are epistemically suspect; these beliefs cannot be trusted and are often more likely false than true. Moreover, one can – and some have – examined religious belief specifically in light of cognitive bias. Hume (1977) might be the first that discussed the relationship between human cognitive bias and religious belief. In section VIII of Enquiry Concerning Human Understanding, Hume (1977), pp. 55–56, imagines a traveler who has re-

Cognitive Bias, the Axiological Question and the Epistemic Probability of Theistic Belief |

83

turned from a “far country”. If the traveler returns with fanciful stories about the people in that land, stories which make the people sound too good to be true (perhaps they “knew no pleasure but friendship, generosity, and public spirit”), then we would rightfully be skeptical. For Hume, part of what makes other humans intelligible is that there are lawlike generalizations which we form from our particular interactions with others; the traveler’s stories are just as fanciful as if someone had “stuffed [their] narration with stories of centaurs and dragons, miracles and prodigies” (Hume (1977), p. 56) because they contradict these generalizations. We have developed particular expectations for what humans are like based upon our prior experiences; we are justly skeptical when those expectations are violated. The degree to which we are skeptical is modulated by how large of a deviation from our expectations the accounts are, just as we are more skeptical when we hear of a severed limb re-growing than when we hear of a person who can jump exceptionally high. Indeed, one purpose of section VIII is to argue that we already accept determinism (and reject the doctrine of “liberty”), because we already have antecedent expectations for how others will behave without which their actions would be unintelligible. Hume returns to this theme in section X during his famous discussion of miracles. He argues that we should not believe any of the miracle claims that have ever been produced. Hume’s arguments against miracles are tied to his conception of natural law. According to Hume, our beliefs about “matters of fact” should always be proportioned to the evidence we gather from experience. Moreover, miracles contradict our everyday experience. Thus, Hume concludes, we should only believe that a miracle has occurred when the alternative would be more miraculous. This has never occurred, so we should not believe that there has ever been a miracle. To provide an example, Hume again considers the traveler (Hume (1977), pp. 78–79). Hume (1977), p. 78, asks rhetorically, “[w]ith what greediness are the miraculous accounts of travelers received, their descriptions of sea and land monsters, their relations of wonderful adventures, strange men, and uncouth manners?” Given previous claims, this rhetorical question might seem strange; the “miraculous accounts” of travelers are met with skepticism, not credulity. In that light, Hume’s previous statements take on a normative quality: people are credulous far too often. However, Hume (1977), p. 79, thinks that when religion is involved, people are even more overly credulous than usual: “[...] if the spirit of religion join itself to the love of wonder, there is an end of common sense; and human testimony, in these circumstances, loses all pretensions to authority.” Worse, those who examine religious accounts often do not possess “sufficient judgment” to examine all of the evidence; indeed, the fideist might even refuse to examine the evidence at all. And when they do examine the evidence, this effort is thwarted by the psy-

84 | Daniel Linford and Jason Megill chological biases of the examiner: “if they were ever so willing to employ [judgment], passion and a heated imagination disturb the regularity of its operations. Their credulity increases [the religious person’s] impudence: And his impudence overpowers their credulity” (Hume (1977), p. 79). Furthermore, there are multiple examples of people believing in extraordinary events (“miracles, prophecies, and supernatural events”) when they should not have because there was later determined to be strong countervailing evidence. The psychological temptation to believe that such events occurred is strong; this is reason to be particularly skeptical when any such occurrence is proposed. Hume (1977), p. 79, writes that there is a “strong propensity of mankind [sic] to the extraordinary and the marvelous, and ought reasonably to beget suspicion against all relations of this kind”. Hume considers an example from everyday life. Due to the propensity that people have towards spreading and believing gossip, regardless of whether the gossip is true, we should be more skeptical concerning information that we hear in gossip. Therefore, whatever probability we assign to some juicy piece of gossip, most likely, the probability of the gossip being true is lower. We should likewise be skeptical of the extraordinary events claimed by travelers: given our desire to believe in mermaids, etc, whatever probability we give to their stories is likely greater than what it should be. In sum, Hume claims that – as a matter of human psychology – humans have certain cognitive biases; e.g., in some circumstances, humans have a tendency or a temptation to believe that extraordinary events have occurred. These cognitive biases lead people into believing irrational things; e.g., they believe some extraordinary event occurred despite a lack of evidence or even perhaps despite countervailing evidence. Our cognitive biases inflate the epistemic probability that we assign to some claims; so when faced with these sorts of claims, we should be skeptical of them. Moreover, Hume applies this idea to religious belief and specifically, to the case of miracles. In a recent and provocative paper, Draper and Nichols (2013) also examine the possibility that cognitive biases cause some to assign a higher probability to Theistic beliefs than they should. Specifically, Draper and Nichols argue that cognitive bias (specifically, biases involved with affect and biases that tend to arise in groups) adversely effects contemporary philosophy of religion. They identify four symptoms of poor health: philosophy of religion is “too partisan, too polemical, too narrow in its focus, and too often evaluated using criteria that are theological or religious instead of philosophical” (Draper and Nichols (2013), p. 420). Based on their reading of the psychological, cognitive science, and sociological literatures, Draper and Nichols conclude that religious discussion triggers deep emotional reactions that lead people astray. Although cognitive biases affect all areas of inquiry, Draper and Nichols argue that philosophy of religion is particularly

Cognitive Bias, the Axiological Question and the Epistemic Probability of Theistic Belief |

85

susceptible to bias. They conclude their paper with several recommendations for philosophers of religion. They suggest, for instance, that philosophers of religion distance themselves from apologetics (be it Theistic or Atheistic). They also suggest that philosophers of religion should at least sometimes use argumentation as a way of testing their beliefs, rather than merely as a way of establishing the truth of their beliefs. They also suggest that philosophers of religion take a certain amount of risk, meaning that they seriously accept the possibility that their particular beliefs might be mistaken. We conclude this section by discussing a few more particular cognitive biases from the literature; these biases will play a role in our argument below. Some cognitive biases cause individuals to overestimate the probability that a belief is true. These include wishful thinking, the valence effect, and optimism bias. Wishful thinking is when individuals overestimate the epistemic probability of a hypothesis because they want the hypothesis to be true. In one study (Bastardi et al. (2011)), researchers examined parents who planned to leave their children at daycare despite an initial belief that the children would be better off at home. The parents had a strong desire to believe that their children would be at least as well off in daycare as they would be at home. When presented with ambiguous scientific studies as to where children are better off, the parents changed their minds and affirmed a strong conviction that children are equally well off in day care as they are at home. A control group, consisting of parents who did not intend to use daycare, showed only a very slight shift in their beliefs on the matter. Similarly, the valence effect of prediction occurs when individuals estimate the likelihood for favorable outcomes to be higher than the likelihood for disfavorable outcomes; see, e.g., Rosenhan and Messick (1966). Optimism bias involves the systematic overestimation of the frequency of positive events and the underestimation of the frequency of negative events; psychologist Tali Sharot (2011), R941, writes “it is one of the most consistent, prevalent and robust biases documented in psychology and behavioral economics”. Common among these biases is the overestimation of the probability that a belief is true precisely because individuals associate positive outcomes with the truth of the belief. If we think a belief will lead to positive outcomes, we tend, as a matter of our psychology, to overestimate the epistemic probability that the belief is true. Why we suffer from this cognitive defect is a question for evolutionary psychologists and others; we simply point out that according to various scientific experiments, the defect is there.

86 | Daniel Linford and Jason Megill

3 The argument We now explicitly formulate our argument, which concludes that Theists – and also perhaps many Non-Theists (be they Agnostics or Atheists) – should lower the epistemic probability that they assign to Theism. We first give the argument in general terms, before expanding upon our claims. The first premise is, (1) If A thinks that the truth of P would entail good consequences, then A should lower the epistemic probability that one assigns to P.

In other words, an agent (or group of agents) A should lower the epistemic probability they assign to any claim that they think would entail good consequences. The reason for this was discussed above: if one thinks that P would lead to positive consequences, there are a host of cognitive biases that lead one to inflate the likelihood that P is true. Given that, it is prudent to lower the epistemic probability one assigns to any view that one thinks is associated with good outcomes. The evidence for premise (1) is empirical; as discussed above, cognitive psychology has shown that these biases do exist, and they do indeed cause people to erroneously inflate the probability that they assign to certain claims. The second premise is, (2) If Theists think that the truth of theism would entail good consequences, then Theists should lower the epistemic probability that they assign to Theism.

Premise (2) is merely premise (1) but with two substitutions: we have replaced “Theists” for A and “Theism” for P. That is, (2) is merely a substitution instance of (1); (1) and universal elimination entail (2), so if (1) is true, so is (2). Premise (3) is, (3) Many Theists think that the truth of Theism would entail good consequences.

This is merely the claim – discussed in section one – that many Theists are also Pro-Theists. But then, given (2) and (3), with Modus Ponens we can infer, (4) Theists should lower the epistemic probability that they assign to Theism.

In sum, the (often implicit) endorsement of Pro-theism by many Theists is problematic. If indeed many Theists are Pro-Theists, and so think that it would be better if God existed than not because various good outcomes would arise if God exists, then this makes Theists particularly susceptible to various cognitive biases (those involved when positive outcomes are at stake) and so to irrationality; specif-

Cognitive Bias, the Axiological Question and the Epistemic Probability of Theistic Belief |

87

ically, they are likely to inflate the epistemic probability they assign to Theism. Therefore, they should lower that probability. Insofar as the argument concludes that we – or at least some of us – should lower the epistemic probability that we assign to Theism, it functions in a similar manner to the evidential argument from evil. Here is a different, though related, argument. Define “Intellectual Theists” (ITheists for short) as those Theistic philosophers of religion who are well versed in the arguments for and against the existence of God. Similarly, define “Intellectual Atheists” (I-Atheists for short) as those Atheistic philosophers of religion who are well versed in the arguments for and against the existence of God. Notice that both I-Theists and I-Atheists are aware of the best evidence for and against God’s existence, yet come to different conclusions. Note that: 1. 2. 3. 4. 5.

Many I-Atheists think that the evidence is strongest for Atheism; Many I-Theists think that the evidence is strongest for Theism; For the impartial observer, it is difficult to decide who is right about the state of the evidence; Most I-Theists think that it would be better if God exists than otherwise; Some I-Atheists agree that it would be better if God exists than otherwise.

We want to know who is right. Given that both I-Atheists and I-Theists have the same evidence available to them, yet both sides often feel very confident in their conclusion, we may wonder if one side or the other is systematically misestimating the strength of the evidence for their position. Now consider the following fairly plausible epistemic principle: S.

All things being equal, we should be suspicious if we find the evidence for a position h to be convincing if we also think that it would be better if h were true. Given the evidence from cognitive psychology, in such a case, it is probable that we are overestimating the evidence: the probability of h given the evidence – P(h|e) – is probably lower than we think.

Thus, we should be suspicious of the I-Theists who, as in (2), find the evidence for God’s existence to be compelling because of (4). Indeed: this is certainly many I-Theists. S provides reason to think that Theists overestimate the evidence for God’s existence. This does not mean that the evidence against God’s existence is

88 | Daniel Linford and Jason Megill not compelling.⁵ Perhaps the probability of God’s existence, given the evidence, is not as high as Theists often take it to be, but it might still be very high indeed? But now consider the I-Atheists who, as in (1), find the evidence for God’s non-existence to be compelling, yet, as in (5), think that things would be better if God existed. Call this group the “Pro-Theistic I-Atheists”. Certainly, S would not tell us that Pro-Theistic I-Atheists are overestimating the evidence. So this gives us reason to think that Pro-Theistic I-Atheists are correctly assessing the evidence that is available to them. Nonetheless, we might think that there are still biases at work for the Pro-Theistic Atheist: S′ .

All things being equal, we should be suspicious if we find the evidence for a position h to not be convincing if we also think that it would be better if h were true. In such a case, it is probable that we underestimating the evidence for the negation: the probability of h given the evidence – P( h|e) – is probably higher than we think it to be.

That is, Pro-Theistic I-Atheists are likely to underestimate the evidence for Atheism because, in light of S, they are likely to overestimate the evidence for Theism. Thus the evidence for Atheism is probably stronger than Pro-Theistic I-Atheists estimate it to be. So we’ve arrived at this collection of conclusions: 6. 7.

The evidence for Theism is probably not as strong as I-Theists estimate it to be; and The evidence for Atheism is probably stronger than Pro-Theistic I-Atheists estimate it to be.

Notice that (6) and (7) would be highly surprising if Theism were probably true and not at all surprising if Theism is probably false. Thus, (6) and (7) are evidence for the conclusion that Theism is probably false. Finally, there is one interesting group – including Nagel, Hitchens, and Kahane – that might overestimate the probability of Atheism. Those who endorse Anti-Theism, the idea that it would be better if God did not exist, should be susceptible to the same biases that Pro-Theists are; although in this case, they think that Atheism will lead to good outcomes, and so they will inflate the probability of Atheism.

5 Indeed, one of the present authors has defended some of the traditional arguments for God’s existence in the past (see Megill and Mitchell (2008); Megill and Reagor (2012); and Megill (2012)); so we are not saying that the evidence for Theism is necessarily poor or even non-existent. Rather, we suggest that, as a consequence of human psychology, perhaps this evidence is not as good as some Theists believe?

Cognitive Bias, the Axiological Question and the Epistemic Probability of Theistic Belief |

89

4 Concluding remarks: An additional implication and some further questions Consider Pascal’s wager. Roughly, Pascal argues that, since believers are rewarded in heaven with infinite happiness, belief in God has infinite positive expected utility. Atheists, however, are punished in Hell for eternity; so not believing in God has infinite negative expected utility. When faced with a decision problem, it is rational to choose the option with the highest expected utility. Therefore, it is rational to believe in God. Moreover, belief in God’s existence is rational even if we place a very low probability on the existence of God; if we assign a non-zero probability to existence of God, no matter how low that probability is, belief still has an infinite positive expected utility. In effect, Pascal’s wager separates the rationality of belief in God from the epistemic probability that we assign to God’s existence; even if the epistemic probability of God’s existence is very low, it can still be rational to believe in God. However, we suggest that Pascal’s wager is problematic in ways not previously noticed. Recall that when we think that a claim’s truth would have a positive outcome, we should lower the epistemic probability that we assign to that claim. So if we think that a belief in God’s existence will lead to an infinite positive expected utility, then we should lower the epistemic probability of belief in God. So Pascal’s wager, in arguing that it is rational to believe in God because of that belief’s positive expected utility, simultaneously – though inadvertently – suggests that we should also lower the epistemic probability of belief in God. This renders Pascal’s wager problematic for the Theist: she can only accept the wager at the cost of lowering the epistemic probability of her belief in God’s existence. But things might be even more dire for Pascal’s wager. This is ultimately an empirical question, but it could be that the greater the utility that we assign to a belief, the greater we inflate the epistemic probability of that belief. If so, then the more we value the putative outcomes of that belief, the more we should lower the epistemic probability of that belief. That is, perhaps, to the extent that there is a known cognitive bias towards a particular belief, the epistemic probability we assign to that belief should be adjusted inversely to the strength of the cognitive bias? And since Pascal’s wager assigns an infinite positive expected utility to Theistic belief, then in order to adjust for the cognitive bias, we should set the epistemic probability of God’s existence to zero. So, Pascal’s wager tells us to believe something for which we should set the epistemic probability to zero, which is absurd. The examination of religious belief in light of the realization that humans suffer from many cognitive biases is still in its infancy; a number of interesting questions remain unanswered. Given that Theists, for example, suffer from bias, can

90 | Daniel Linford and Jason Megill Theists overcome this bias? If so, then how? One might suggest that a sound proof of God’s existence that was convincing to everyone (or nearly everyone) in the debate – even Atheists – would likely not be the result of bias; of course, no such argument has yet been discovered; though perhaps considerations involving bias makes the need for such a proof more pressing for the Theist? Moreover, one might wonder if there are biases that affect (many or even all) Atheists as well? If so, then what are these biases? Indeed, perhaps Agnostics are also affected by bias (perhaps those biases involved in making us overly risk averse)? Perhaps, as a result of cognitive bias, religious beliefs – Theistic or otherwise – are untrustworthy?

Bibliography Bastardi, A., Uhlmann, E. L., and Ross, L. (2011), “Wishful thinking: Beliefs, Desire, and the Motivated Evaluation of Scientific Evidence”, in Psychological Science: 22, 731–732. Camus, A. (1955), The Myth of Sisyphus, translated by J. O’Brian, London: H. Hamilton. Craig, W. (1994), “The Absurdity of Life Without God”, reprinted in The Meaning of Life, 2nd Ed., edited by E. D. Klemke, New York: Oxford University Press, 2000, 40–56. Draper, P. and Nichols, R. (2013), “Diagnosing Bias in Philosophy of Religion”, in The Monist: 96(3), 420–446. Gordon, J. (1983), “Is the Existence of God Relevant to the Meaning of Life?”, in The Modern Schoolman: 60, 227–46. Haber, J. (1997), “Contingency and the Meaning of Life”, in Philosophical Writings: 5, 32–44. Hitchens, C. (2007), God Is Not Great: How Religion Poisons Everything, New York: Twelve Books. Hume, D. (1977), Enquiry Concerning Human Understanding, Clarendon Press. Kahane, G. (2011), “Should We Want God to Exist?”, in Philosophy and Phenomenological Research: 82, 674–696. Kahane, G. (2012), “The Value Question in Metaphysics”, in Philosophy and Phenomenological Research: 85, 27–55. Kahane, G. (2014), “Our Cosmic Insignificance”, in Nous: 48(4), 745–772. Kahneman, D. and Tversky, A. (1972), “Subjective Probability: A Judgment of Representativeness”, in Cognitive Psychology: 3(3), 430–454. Kahneman, D. and Tversky, A. (1973), “On the Psychology of Prediction”, in Psychological Review: 80(4), 237–251. Kraay, K. and Dragos, C. (2013), “On Preferring God’s Non-Existence”, in Canadian Journal of Philosophy: 43, 157–178. Kruger, J. and Dunning, D. (1999), “Unskilled and Unaware of It: How Diflculties in Recognizing One’s Own Incompetence Lead to Inflated Self-Assessments”, in Journal of Personality and Social Psychology: 77(6), 1121–1134. Mawson, T. (2012), “On Determining How Important it is Whether or Not there is a God”, in European Journal for Philosophy of Religion: 4, 95–105. Megill, J. (2012), “Two Ontological Arguments for the Existence of an Omniscient Being”, in Ontological Proofs Today, edited by Miroslaw Szatkowski, Frankfurt: Ontos Verlag, 77–88.

Cognitive Bias, the Axiological Question and the Epistemic Probability of Theistic Belief |

Megill, J. and Mitchell, J. (2008), “A Modest Modal Ontological Argument”, in Ratio: 22(3), 338–349. Megill, J. and Reagor, A. (2012), “A Modal Theistic Argument”, in Ontological Proofs Today, edited by Miroslaw Szatkowski, Frankfurt: Ontos Verlag, 89–112. Metz, T. (2013), “The Meaning of Life”, in The Stanford Encyclopedia of Philosophy, Summer 2013 Edition, edited by E. N. Zalta. URL = . Nagel, T. (1997), The Last Word, Oxford: Oxford University Press. Pohl, R. F. (2004), Cognitive Illusions: A Handbook on Fallacies and Biases in Thinking, Judgment and Memory, Hove, UK: Psychology Press. Rosenhan, D. and Messick, S. (1966), “Affect and Expectation”, in Journal of Personality and Social Psychology: 3, 38–44. Russell, B. (1903), “A Free Man’s Worship”, reprinted in The Collected Papers of Bertrand Russell, vol. 12, Routledge, 1985. Sharot, T. (2011), “The Optimism Bias”, in Cur. Biol.: 21, R941–R945. Tooley, M. (2009), “Helping People to Think Critically about their Religious Beliefs”, in 50 Voices of Disbelief , edited by R. Blackford and U. Schüklenk Malden: Wiley-Blackwell. Tversky, A. and Kahneman, D. (1971), “Belief in the law of small numbers”, in Psychological Bulletin: 76(2), 105–110. Tversky, A. and Kahneman, D. (1983), “Extension versus intuitive reasoning: The conjunction fallacy in probability judgment”, in Psychological Review: 90(4), 293–315.

91

Jason Megill and Daniel Linford

On Computable Metaphysics: On the Uses and Limitations of Computational Metaphysics 1 The theme of this book is the ontology of theistic beliefs. In this paper, we discuss one possible – though certainly not the only possible – way to view a set of metaphysical or theological beliefs: any actual or possible metaphysical or theological belief system is a set of sentences that could be the output of a Turing machine. Viewing metaphysical or theological belief systems as the possible output of a Turing machine opens up novel possibilities, or so we argue. Call a set of characters (e.g., an alphabet) along with a set of rules (e.g., a natural language’s grammar) for combining these characters into well-formed strings (e.g., sentences) a “symbolic language”. Examples of symbolic languages include the natural languages like Polish or Chinese; first and second-order logic; the symbolic systems used in mathematics (e.g., the notational systems used in Calculus or Set theory); and the symbolic systems used in various sciences such as physics or chemistry. And consider any random moment in human history; this moment might be in the past, it might be the present moment, or it might be far in the future. For any random moment, there will be a set that consists of all of the character strings in symbolic languages that humans have produced up to and including that moment. In other words, for any moment in human history, there will be a set that consists of the sum total of human symbolic output up to that moment. Call the set for a given moment an “ASO”, for “actual symbolic output”. A string of characters will be a member of the ASO for a given moment if and only if at least one person has produced the given string of characters up to that moment. So, for example, the ASO for the current moment will contain all well-formed character strings in symbolic languages that have been produced up until now; the ASO for some moment a hundred years ago will contain the strings produced up until then but will not contain the strings produced after that moment; and the ASO for some

https://doi.org/9783110566512-007

94 | Jason Megill and Daniel Linford moment a hundred years from now will contain the current ASO along with all of the character strings that will be produced from now until then.¹ The ASO for any given moment will have various characteristics. Of course, the cardinality of the ASO continuously grows as we produce new strings of characters in symbolic languages, so the ASO for one moment will likely be larger than the ASO for the previous moment. And once a string of characters is in the ASO, it will be in all future iterations of the ASO; the ASO consists of all strings produced up to a given moment, and once a string is produced, it cannot be unproduced. The ASO is, by definition, not closed under logical consequence; it contains only those strings that have actually been produced as output by someone up to the relevant moment, but it does not necessarily contain all of the logical entailments of those strings. Some of the strings in the ASO will correspond to true claims, some to false claims, and some (e.g., exclamations, questions) will lack a truth-value altogether. Since people sometimes produce strings that contradict other produced strings, the ASO will be inconsistent. But the following is the most important property of the ASO for our purposes: the ASO is, has always been, and will always be finite. The ASO for any random moment throughout human history, even if that moment is far in the distant future, will contain only a finite number of strings of characters. Consider the following argument: even though there are an infinite number of natural numbers, any particular natural number is finite; this is a basic mathematical fact. Furthermore, one can – at least in principle – assign one and only one natural number to every person that has existed throughout human history up to any particular moment. Call the first person “1”, the second “2”, and so on, until every person that has existed up to the relevant moment is assigned a particular natural number. Suppose that these numbers are assigned to individuals in the same order as the natural numbers; as each person is born, they are assigned a number one higher than the previous person born. If so, then for any particular moment, there will be a person who is associated with a number higher than any other person; this will be the most recent person born. And this number will also correspond to the total number of people that have existed up to that moment. Since this highest number is a particular natural number, and since all particular natural numbers are

1 Some of the claims in this section were previously discussed in Megill et al. (2013) and Megill and Melvin (2014); and although the discussion here is self-contained, sometimes the discussions in these previous papers go into more detail than the discussions here. Also, additional arguments for the claims in this section are offered in these previous works. This paper is an attempt to apply these previous ideas to issues in metaphysics, e.g., its methodology; this was not attempted in these previous papers.

On Computable Metaphysics ... | 95

finite, and since this number is the number of people who have ever existed up to that moment, there will only be a finite number of people who have existed up to that moment. Moreover, note that each of the people who have existed up to a given moment will produce a set of symbolic strings; e.g., Turing produced a set of symbolic strings that contains “On Computable Numbers” (Turing (1939)) and “Computing Machinery and Intelligence” (Turing (1950)), along with small talk with his colleagues and so on. Call the set of strings that a particular individual produces their “Individual Set”. One can, for any Individual Set, assign a particular natural number to each string produced by that individual. For example, the first string that Turing produced is “1”, the second “2”, and so on all of the way up to the last string he produced before his untimely death. The number associated with the last string will be a particular – and so finite – number. That is, each individual will produce only a finite number of symbolic strings.² By definition, the ASO for a given moment contains all of the symbolic strings produced by everyone who has existed up to that moment; that is, the ASO for a given moment will be the union of all of the Individual Sets produced up to that moment; and so will be the union of a finite number (again, there will only be a finite number of people who have existed up to any moment) of finite sets (again, each individual will only produce a finite number of strings). Basic set theory states that the union of a finite number of finite sets is finite, so the ASO for any particular moment throughout human history will be finite. Given that the ASO for any moment is finite, fundamental and uncontroversial mathematical results in the theory of computation entail that ASO sets will have additional properties. All finite sets are recursive, so any particular ASO will be recursive. All recursive sets are recursively enumerable (see, e.g., Wang (1981), p. 242), so any particular ASO will be recursively enumerable. All recursively enumerable sets are axiomatizable (see Craig (1953)), so any particular ASO will be axiomatizable. And all recursive sets are Turing computable (Wang (1981), p. 242),

2 Consider, e.g., Boolos and Jeffrey (1999), p. 19): No human being can write fast enough, or long enough, or small enough to list all members of an enumerably infinite set by writing out their names, one after another, in some notation ... The problem will remain, that for all but a finite number of values of n it will be physically impossible for any human or machine to actually carry out the computation, due to limitations on the time available for computation, and on the speed with which single steps of computation can be carried out, and on the amount of matter in the universe which is available for forming the symbols. That is, any particular individual will only be able to produce a finite number of symbolic strings.

96 | Jason Megill and Daniel Linford so any particular ASO is Turing computable. In other words, for any given moment throughout human history, there is in theory a Turing machine that could produce as output all of the symbolic output of humanity up to that moment.³ But crucially, a much stronger claim can also be proven. Presumably, for any given moment, the symbolic output that we have in fact produced up to that moment could have been different than it is. For example, consider the ASO as it exists at the present moment. It contains “On Computable Numbers”, Turing’s Dissertation (1939), and the rest of Turing’s published work. But suppose that history had turned out differently; perhaps, for instance, there is a possible alternative timeline in which Turing is left alone. Indeed, Turing lives another 40 years after he died in our timeline and produces a dozen important papers during this time. In this timeline, the ASO for the current moment is different; it contains strings of characters (the strings that form the missing Turing papers) that our ASO does not. Denote a possible ASO “3ASO”. Note, however, that no matter what path human symbolic output takes, or no matter what alternative timeline we are considering, the symbolic output for any given moment in that path will still be the product of a finite number of people producing a finite number of strings. That is, 3ASOs, like ASOs, will be finite sets of character strings, and so also Turing computable. In other words, it is not merely that a Turing machine can produce the symbolic output produced up to any actual moment in human history; rather, a Turing machine could also produce any possible set of symbolic output that might have been produced up to that moment.⁴

3 Indeed, a Turing machine merely enumerating all possible character string combinations would eventually produce, e.g., the current ASO. The current ASO will be the set of all character strings produced up to the current moment; it will be a finite number of finitely long strings. One could take all of the strings in the ASO and place them one after another, forming one extremely long string. This string, call it “a”, will be the combination of a finite number of finitely long strings, and so will itself be finite. So, a will have length n, where n is some finite number. Given that all of the alphabets that form the symbolic languages used to produce the strings in ASO are finite, the number of different characters in a will be finite. But given a finite alphabet, a Turing machine can enumerate all possible strings of length n in a finite amount of time. That is, a Turing machine, just simply enumerating possible character strings, could produce a, and so the ASO, in a finite amount of time; though of course it might take the machine an absurdly long time produce a. Note that below, we suggest that there are more efficient ways for Turing machines to produce at least particular sets (or subsets of the ASO) that humans have (or could have) produced. 4 One might object that there could be a logically possible world in which there are an infinite number of people able to produce an infinite number of character strings, or perhaps there is a logically possible world that contains one person who creates strings for an infinite amount of time. If so, then not all 3ASOs are finite, and these possible infinite sets might be computable or

On Computable Metaphysics ... |

97

Now consider the set PSO, for “possible symbolic output”. This set consists of all character strings that humans could possibly produce. Of course, the ASO for any moment will be a subset of PSO; if we have actually produced a set of strings, then it is possible for us to produce the set. But since, presumably, there are strings that we have not yet produced even though we could produce them, PSO will contain strings not in the ASO. PSO – unlike the ASO – is probably infinite; e.g., we could produce the string “1”, and the string “11”, and “111”, and so on ad infinitum. Note that one 3ASO is one possible path that our symbolic output could take. But then all 3ASOs would be all possible paths that our symbolic output could take. But then PSO can be thought of as being the union of all 3ASOs. Since any 3ASO is finite, and since PSO is infinite, PSO must be the union of an infinite number of finite sets. And since any 3ASO is computable, PSO is the union of an infinite number of computable sets.⁵ Finally, consider any random character string s that we could possibly produce. That is, consider any given string in PSO. Note that PSO is the union of all 3ASOs, so s must be in at least one 3ASO. But all 3ASO could be the output of a Turing machine. But then s could be the output of a Turing machine. In other words, any character string that we could possibly produce could be the output of a Turing machine.⁶

might not be; after all, some infinite sets are computable, and some are not. But note that we are not concerned here with other logically possible worlds; rather, we are concerned with what is physically possible in the actual world. In the actual world, at least, any 3ASO will be produced by a finite number of people capable of producing only a finite number of strings, and so will be finite. So, e.g., if Turing had lived to produce the missing Turing papers, in which case the ASO for the present moment would be different, this new ASO would still be the product of a finite number of people producing a finite number of strings, and so would still be finite. 5 One could take the strings in any given possible ASO and place one string after another, thereby generating a single long character string s of length l. The argument above assumes that different possible ASO sets can contain more output than others; so, the length of one s might be l, the length of a different s might be l+1, or l+2, and so on. Each length will be finite, but for any length, we could always add one more character; this implies that the PSO contains an infinite number of possible strings. One could argue that there will be an upper bound to the lengths that these strings could have; so, for example, there would be some maximum length l that an s could have. But this l would be some finite number, and so it would be even easier to show that a Turing machine could produce any possible ASO. That is, even if we grant that strings have some length limit, our key claims (e.g., that a Turing machine could produce any possible ASO) are still true, so this worry is ultimately not an issue. 6 This has various implications; some of these implications are discussed below. Some other implications, e.g., the Lucas-Penrose argument (which claims that human mathematical reasoning is not computable) cannot be sound, are discussed in Megill et al. (2013) and Megill and Melvin

98 | Jason Megill and Daniel Linford In sum, we’ve established the following claims: (i) the sum total of human symbolic output produced up to any given moment in human history could be the output of a Turing machine, i.e., actual human symbolic output is computable; (ii) even if our symbolic output had been different, that output still would have been Turing computable, i.e., any individual possible set of human symbolic output is computable; and (iii) a Turing machine can produce any particular string of characters that we could possibly produce.

2 Consider some of the great metaphysical or theological systems. Think, e.g., of Spinoza’s (1994) metaphysical system as developed in The Ethics, which consists of a number of definitions (e.g., of substance, mode and attribute) and axioms (e.g., everything that exists either exists in itself or in something else) that are then used to deduce various propositions (e.g., God exists, everything is in God, etc) in a manner reminiscent of Euclidean geometry. Or think of Leibniz’s metaphysical system; a handful of basic principles such as the Principle of Sufficient Reason and the Principle of Non-Contradiction form the foundation of his system, in which eventually the existence of God and his theory of monads (among other things) are asserted.⁷ Or one can consider older systems than these, e.g., Aquinas’s, or more recent examples of metaphysical systems, e.g., David Lewis’s metaphysical views. Indeed, one can consider whatever metaphysical or theological system one wishes; our claims in this section apply to any of them. However, when considering various metaphysical or theological systems, a question arises. When considering, e.g., Spinoza’s system, what should be included? Should we take a sparse view and claim that Spinoza’s system consists of and only of the metaphysical claims in The Ethics? Or should we, as seems plausible, include statements concerning Spinoza’s system that Spinoza himself made elsewhere, for instance, in conversation with Leibniz? Or should we also include statements about Spinoza’s system made by scholars of it; e.g., if some scholar correctly pointed out that Spinoza’s axioms entail some theorem not included in The Ethics, does Spinoza’s system also include that novel theorem? It will become

(2014). For more on the Lucas-Penrose argument, see Lucas (1961), Lucas (1996), Penrose (1989), Penrose (1994), and Megill (2012). 7 See, e.g., Leibniz (1998).

On Computable Metaphysics ... |

99

apparent that this issue will not affect our argument; indeed, our argument succeeds no matter what one decides to include in Spinoza’s system. For note that one can form a set that consists of the strings of characters that comprise Spinoza’s system, at least as it has been developed up to a certain moment in human history. Call this set S. Again, S will contain all strings of characters that are (i) a part of Spinoza’s system that (ii) have been produced as output up to some moment in human history (presumably, before Spinoza, S was the empty set). Crucially, this set might contain just the strings of characters in The Ethics, or it might contain other statements that Spinoza made elsewhere (e.g., in other works), or it might even contain statements made by people aside from Spinoza (e.g., Spinoza scholars). But whatever strings of characters we think should be included in S, S will consists of strings of characters produced by humans up to a certain moment in time; that is, S will consist of human symbolic output produced up to a certain moment in time. But then, since the ASO is the set of all symbolic output produced up to a particular moment, and S is the set of symbolic output that comprises Spinoza’s metaphysical system produced up to a particular moment, S will be a subset of the ASO.⁸ The ASO is finite, and a subset of a finite set is finite (this is basic set theory); therefore, S will be finite. But then, given fundamental results in recursion theory discussed above, S will also be recursive, recursively enumerable, axiomatizable and Turing computable. That is, regardless of what we think Spinoza’s system includes, that system – at least as it has been developed by us up to a certain moment in time – could be the output of a Turing machine. At least in principle, a Turing machine could have produced The Ethics, or any metaphysical or theological system, be it Leibniz’s, Aquinas’s, Lewis’s, or some other system. One might object: you are discussing Spinoza’s system as it exists to us at a particular moment in time. In other words, you are discussing Spinoza’s system only insofar as you are considering the claims in the system that have actually been produced by us up to a given moment. But perhaps Spinoza’s system is, in a sense, more than this? Perhaps Spinoza’s system also contains all of the entailments of his claims, even though many of these entailments remain unproduced? In other words, S, like the ASO, is not closed under logical entailment. But perhaps Spinoza’s system is closed under logical entailment? Perhaps any claim that is logically entailed by Spinoza’s axioms is a part of Spinoza’s system, even if that

8 Of course, humans have produced output that is a part of Spinoza’s metaphysical system; this output will be included in the set of all human symbolic output. That is, the current ASO, e.g., will contain S. But of course, not all human symbolic output is a part of Spinoza’s metaphysical system; there will be claims in the ASO that are not in S. So S will be a subset of ASO.

100 | Jason Megill and Daniel Linford statement has not been produced by anyone? So, posit a new set: S+UE. S+UE will be the union of two sets, S, which contains those strings of characters in Spinoza’s system that we have produced up to a given moment in history, and UE (for “unproduced entailments), which contains the statements in Spinoza’s system that are logically entailed by S but have not yet been produced as output by anyone. The objection claims that Spinoza’s system is not merely S, but is S+UE; and furthermore, while S could be the output of a Turing machine, this does not imply that S+UE, which is a larger set than S, could be. Our response is the following: first, note that Spinoza’s system, as it exists to us or from our perspective at any given moment just is the set S for that moment. At any particular moment, all we know about Spinoza’s system will be those character strings that we have produced in that system up to that moment; i.e., at any moment, Spinoza’s system – from our perspective – will be a particular S. We cannot step outside of our epistemic perspective to magically know claims about Spinoza’s system that will be produced, say, in 100 years, or even perhaps will never be produced. The point is this: even if we suppose that Spinoza’s system really is S+UE, and even if we suppose that S+UE is – unlike S – non-computable, it is still the case that Spinoza’s system as it exists from our epistemic perspective, and as it will always exist from our epistemic perspective, will be equivalent to an S, and so computable. Second, it appears that S+UE might be computable in any event. Note that S+UE is composed of S plus those unproduced statements in UE that are logically entailed by S. So consider some putative character string that might or might not be in the S+UE. If the string is not in S+UE, then it is not logically entailed by S; but then the string is not a part of S+UE and so is not a part of Spinoza’s system, even when that system is construed broadly to include UE. But if the string is in S+UE, then it can be logically derived from the sentences in S, presumably by using the logical inference rules of, say, first-order logic (again, the claims in UE are, by definition, entailed by those in S). But as we saw above, S could be produced by a Turing machine; and a Turing machine is perfectly capable of producing logical entailments of a set of sentences using the rules of first-order logic, and so perfectly capable of also producing UE if given S, at least in principle.⁹ But if a Turing machine can produce S, and then could further produce UE from S, then a Turing machine can produce S+UE as well. Third, note that a Turing machine can produce any particular character string that we can produce (section I). But then, a Turing machine can produce any particular string that we have pro-

9 Of course, if we are waiting for a machine to produce a particular string, the machine might never halt, though this is also true for humans.

On Computable Metaphysics ... |

101

duced, and so any string in S, and could produce, and so any humanly producible string in UE. So a Turing machine could produce S+UE. Finally, it is plausible that Spinoza’s system could have developed differently than it has. For example, perhaps it is possible that Leibniz, after personally discussing Spinoza’s system with Spinoza, produced a very detailed account of that system (one more detailed than any he did in fact produce), one in which Leibniz – with his logical acumen – is able to derive theorems from Spinoza’s definitions and axioms that no one has in fact derived? In this alternative timeline, Spinoza’s system would have developed differently; so, e.g., instead of S as it exists to us currently, Spinoza’s system would consist of a different set S′ . But again, all 3ASOs are finite. And any given possible S′ would be a subset of at least one 3ASO, and since all 3ASOs are finite, S′ would be finite too. But then any possible S′ will be computable. That is, even had Spinoza’s system been different, or been comprised of different character strings, it still could be the output of a Turing machine. In sum, any metaphysical or theological system – as it exists to us at a given moment throughout history – can be thought of as consisting of a set of strings of characters that have been produced up to that moment. This set will be computable; for example, it will be recursively enumerable, and so there will be a computer program capable of producing that set as output. Any metaphysical system, as it exists from our perspective at any given moment, could be the output of a Turing machine. Moreover, even if we think of a metaphysical system as a set of sentences consisting of both (i) the strings of characters in that system that have been produced and (ii) have not been produced, this set will still be Turing computable. Finally, note that even if a metaphysical system could have developed differently than it in fact has, and so would be a set of sentences whose members are different than the actual set, these merely possible sets could still be the output of a Turing machine.

3 In this section, we briefly discuss (a) automated theorem proving and (b) computational metaphysics. In the following section, we draw on this discussion, as well as the results of the previous two sections, to draw various conclusions. Automated theorem proving is one method in which computers have been used to aid mathematical research; in automated theorem proving, generally some mathematical system is translated into a logical language (e.g., first-order logic), and then computers mechanically perform various logical operations (e.g., they search for a proof of some particular claim) in this system. To elaborate, there

102 | Jason Megill and Daniel Linford are a variety of methods and approaches in automated theorem proving. Some automated theorem provers seek to fully automate mathematical discovery while other systems require interaction with human mathematicians. Some systems use the propositional calculus, some use first-order logic, and some use more powerful logics. Some systems use the “resolution method”. Some use the “tableaux method”. Some rely on brute search. Other systems use still some other method.¹⁰ Many automated theorem provers attempt to derive a particular, predetermined mathematical theorem from a set of axioms; examples include McCune’s proof (McCune (1997)) of the Robbins conjecture and Appel and Haken’s proof (Appel and Haken (1977)) of the four color theorem. But sometimes a system is able to generate novel theorems on its own, and then search for a proof of those theorems; an example is Bagai etal’s work (Bagai et al (1993)) in the automation of geometrical reasoning.¹¹ Automated theorem provers have also been used to check humanly generated mathematical proofs. One early and philosophically interesting use of automated theorem proving was Wang’s system (Wang (1960)) that derived Russell and Whitehead’s theorems (Russell and Whitehead (1910/1912/1913)) in the Principia in minutes. Broadly construed, computational metaphysics has, thus far, involved the following activities (as summarized in Beavers (2011)): 1. 2. 3.

implementing metaphysical systems in a computational environment; proving new theorems from well-defined collections of axioms; identifying new conceptual connections between existing ideas.¹²

We describe each of these in turn. Although various authors have contributed to all four efforts, E. Zalta and others at Stanford’s Metaphysics Research Laboratory

10 Davis (2001) and MacKenzie (1995) discuss the history of automated theorem proving. Duffy (1991) and Bibel and Schmitt (1998) discuss particular methods used in automated theorem proving (e.g., the resolution method, the tableaux method, and so on). For an overview of automated theorem proving techniques, see Portoraro (2014). 11 Bagai et al (1993), p. 415), ... introduce a conceptual framework for discovery of theorems in geometry and a mechanism which systematically discovers such theorems. [The] mechanism incrementally generates geometrical situations, makes conjectures about them, uses a geometry theorem prover to determine the consistency of situations, and keeps valid conjectures as theorems. 12 While there are other existing applications of computer technology in philosophy (for simulating social systems or for pedagogy), they are beyond the scope of this paper (but see Beavers (2011) for a summary).

On Computable Metaphysics ... |

103

(see https://mally.stanford.edu/) have delivered the most sustained activity thus far. The bulk of their activity has consisted in using automated theorem provers (such as PROVER9) to implement existing metaphysical systems (Zalta and Fitelson (2007); Beavers (2011)) and to prove new theorems within those implementations (Zalta and Oppenheimer (2011); Beavers (2011)). Zalta and Fitelson (2007) have implemented various theories of objects (Forms, abstracta, etc) and of concepts (e.g., Leibniz’s) in a computational environment, thereby demonstrating that at least some metaphysical systems can be represented on a computer. Subsequently, members of the Stanford Lab have used their implementation of existing metaphysical systems for proving novel results, including a shorter version of the ontological argument for the existence of God (Zalta and Oppenheimer (2011)). Others have used previously existing software and techniques to implement metaethical (Lokhorst (2010)) systems in computational environments. Also, there are on-going efforts to map and catalogue existing philosophical ideas and theories. These efforts have demonstrated novel connections between ideas; for example, the Indiana Philosophy Ontology Project demonstrated a connection between divine illumination and mental representation through the medieval problem of universals, a connection not commonly known by philosophers of mind but discoverable by algorithms searching the Stanford Encyclopedia of Philosophy (see Beavers (2011), pp. 6–8)).

4 We now draw various conclusions involving computational metaphysics. These conclusions are: (i) computational metaphysics can completely succeed at least in principle; (ii) computational metaphysics – and indeed human metaphysics – suffer from the same limitations as Turing machines; and (iii) we discuss some other possible methods that could be used in computational metaphysics, including the possibility that machines could generate novel metaphysical systems. First, note that: (a) any particular metaphysical claim (e.g., the Principle of Sufficient Reason, or some as yet unproduced metaphysical principle) that we could produce, a Turing machine could produce as well (section II); (b) any particular metaphysical system (e.g., Spinoza’s system, or some as yet unproduced metaphysical system) that we could produce, a Turing machine could produce as well (section II); and (c) the sum total of possible and actual human metaphysical

104 | Jason Megill and Daniel Linford output could be produced by a Turing machine (section II).¹³ So there is nothing, in theory at least, that would prevent Turing machines from being the equals of human metaphysicians at least in terms of output. Consider the following claims, none of which seem obviously or a priori false: (A) Human metaphysics depends at least partly upon human intuition. It isn’t clear that a Turing machine could have this intuition. So, perhaps humans are capable of producing metaphysical claims (or systems or output) that a Turing machine could not? (B) Doing human metaphysics requires some sort of creativity. It isn’t clear that a Turing machine, which simply computes in accordance with predetermined instructions, could be creative in the necessary way. So, perhaps humans are capable of producing metaphysical claims (or systems or output) that a Turing machine could not?¹⁴ (C) It might be that some human cognitive processes are non-computable; indeed, this possibility has been the subject of vigorous debate for decades. So, perhaps human metaphysics is the product of a non-computable cognitive process, in which case there could be metaphysical claims (or systems or output) that we could produce but no Turing machine could?¹⁵

13 Consider, for instance, a particular metaphysical system; e.g., Spinoza’s. Think of this system as a set of sentences S. There are various methods through which a Turing machine might produce this system as output; these methods differ in terms of their interest and feasibility. One could, for example, construct a simple program along the following lines: output the first letter of The Ethics, then output the second, and so on until the entirety of The Ethics is the output of the program. Of course, this would be uninteresting. A second possibility is: The Ethics is a work composed of a finite number of claims that are in turn composed of a finite number of different symbols. One could form a large string composed of all of the strings in The Ethics. This string will have a finite length n. A program that would halt in a finite amount of time and that could produce all of the possible strings of length n could be constructed; presumably, most of these strings would be gibberish, some (likely a very small percentage) will correspond to interesting alternative versions of The Ethics, while one will correspond to The Ethics itself. Such brute force methods might have their uses in computational metaphysics, and indeed brute force methods have sometimes been used with success in automated theorem proving in mathematics. But a possible problem with this method is its computational feasibility; while the program would halt in a finite amount of time, the amount of time needed might be extremely large. A third possibility, however, is that there are algorithms that could produce The Ethics (or any other actual or even possible metaphysical system) in a more guided, intelligent manner. We return to this possibility below. 14 The objection that computers can never capture human cognition because they will lack creativity goes back to the early days of Artificial Intelligence; for example, Turing (1950), who calls it the “Lady Lovelace” objection, discusses it. 15 One example of the debate over the computability of human cognition is the Lucas-Penrose argument; again, Lucas-Penrose concludes that human mathematical reasoning is not computable.

On Computable Metaphysics ... |

105

These possibilities are not exhaustive; there might be other worries that would lead one to believe that humans can produce metaphysical claims (or systems or output) that a Turing machine could not. But again, these worries are not decisive; a Turing machine could produce any metaphysical claim (system, or output) that humans can. This implies that, in theory at least, computational metaphysics can be a successful research program; indeed, in principle all actual and possible metaphysical reasoning could be automated. And this provides even more motivation, aside from the successes already achieved in computational metaphysics, to pursue it. Of course, designing machines that could, say, independently produce novel metaphysical systems on its own might be difficult in practice (and we return to this issue below), but we do know that it is at least possible in principle, which is a start. One might object that there is at least one limitation machines have that human metaphysicians do not: Searle’s (Searle (1980)) Chinese room argument shows that while humans have semantics, i.e., an understanding of what the strings of symbols they produce mean, a machine will merely have syntax, and so will not know what the output it is producing means. However, various objections have been raised to the Chinese room argument.¹⁶ And more importantly, the objection is irrelevant. The goal of computational metaphysics is to aid and perhaps even accelerate the development of metaphysics. A machine could do this by producing theorems of systems etc. without needing to understand its output; indeed, a machine need not even have any understanding of what task it is performing. Computers are now ubiquitous in scientific research, and no one complains that the machines do not understand their output or what tasks they are performing; indeed, that is beside the point.¹⁷ Second, given that all extant or even possible metaphysical systems could be the output of a Turing machine (e.g., there is a Turing machine that could produce Spinoza’s system, and one that could produce Leibniz’s, etc), a new possibility arises. Since the early modern period, a central concern of philosophy has been trying to determine the limits of metaphysics; this was, e.g., an important part of Hume’s and Kant’s respective projects. But if any metaphysical system could be the output of a Turing machine, then one can treat the limits of metaphysics at least in part as a mathematical problem as opposed to an epistemological one. That is, if a Turing machine could produce any metaphysical system, then the mathe-

16 See Cole (2014) for an overview of the debate surrounding Searle’s argument. 17 Another way to put the point is that Searle’s argument targeted “Strong Artificial Intelligence” (among other things). But computational metaphysics can succeed even assuming that Strong A. I. cannot.

106 | Jason Megill and Daniel Linford matical results that apply to the limitations of Turing machines will also affect particular metaphysical systems and indeed, human metaphysics as a whole. So, e.g., the Halting problem (Turing (1939)), and any other mathematical result that demonstrates a limitation of Turing machines, might also be a limitation of particular metaphysical systems and of human metaphysics in general. Third, thus far, computational metaphysics has focused on finding new and better (i.e., shorter) versions of certain metaphysical arguments (e.g., the ontological argument) in a given axiomatic system and, at times, finding new connections between what were apparently disparate ideas (e.g., between divine illumination and mental representation). But it appears that a number of different – and previously unused – techniques could be used. We now discuss a few examples. This discussion also allows us to address a concern that many readers will have: it might very well be that, e.g., a Turing machine could produce an entirely novel metaphysical system. But what would the method be? There are many methods that a Turing machine could use to produce strings of symbols, for example; but, say, simply randomly producing strings of symbols would be a hopelessly inefficient way to produce a new metaphysical system. Like the proverbial monkeys at a keyboard typing until they produce Hamlet, the vast percentage of what this computer would produce would be gibberish, and such a machine might run for a million years, or even forever, before producing a system that has any interest. We agree; we think that there are other methods – or effective procedures – that a Turing machine could use to develop new metaphysical systems that would make the mechanical creation of new metaphysical systems much more efficient and computationally feasible. But first, one method that could be used in computational metaphysics is: computers could be used to check the internal consistency of a metaphysical system, at least in a limited sense. Consider a set of sentences, and consider two different ways that the set might be inconsistent. The set might contain an “explicit” contradiction, which means that the set contains both a claim and the claim’s negation. That is, a set S contains an explicit contradiction iff for some x in S, ¬x is also in S. Or the set might contain an “implicit” contradiction, which means that even though the set does not contain a claim and the claim’s negation, a contradiction could be inferred from the claims in the set. So, while {P, Q, ¬P, ¬Q} contains two explicit contradictions, {P, Q, P → ¬Q} contains an implicit contradiction, because one can infer ¬Q from the set, which contradicts Q, which is in the set. One can devise a simple algorithmic and so Turing computable procedure for checking a metaphysical system for explicit contradictions. Consider the set of sentences that comprise a metaphysical system as it has been developed by us up to a certain moment. Take the first sentence in the set. Check to see if the second sentence is the negation of the first; if it is, the system contains an explicit contra-

On Computable Metaphysics ... |

107

diction. If it is not, then check the third sentence to see if it is the negation of the first sentence; continue until one either obtains a contradiction or else checks all strings. If no string contradicts the first sentence, then repeat the procedure for the second sentence, and so on for all sentences, until the system is either found to be explicitly contradictory or else is proven to be free from explicit contradictions. One might object that it would be better if we could show – with a simple method and without needing to derive all the theorems of a metaphysical system – that the system entails no contradictions at all. That is, it would be preferable if we could show that a system contains an implicit contradiction without having to show that it contains an explicit contradiction. This would be preferable; however, it appears to be impossible. In order to show that a system contains an implicit contradiction, one would need to derive the contradiction; but once one derives the contradiction, the contradiction instantly becomes an explicit contradiction. Moreover, computational methods could be used to study the relationships between two (or more) metaphysical systems. Consider two metaphysical systems A and B. A Turing machine could, e.g., mechanically check to see if the two systems are consistent with each other; or rather, a Turing machine could at least check for any explicit inconsistencies between the two systems. Brute force could be used; consider the following effective (and so computable) procedure. A and B will both be sets of sentences. Consider the first sentence s in A. Check to see if the first sentence in B is the negation of s. If it is, then halt: the two systems are inconsistent with one another. If it is not, then check the second string, then the third, and so on for all strings in B until we either reach a contradiction or else determine that s does not contradict B. Then, repeat the procedure for the second string in A, and then the third and so on, until either a contradiction is reached or the two sets are demonstrated to be consistent, at least in their current incarnations. Or perhaps more subtle methods could be used? For example, one could take a selection of claims from A and then check to see if their negations are derivable in B. Indeed, if one could show with traditional automated theorem proving techniques that it is possible to derive in B the negation of any of the axioms of A, we could conclude that the sets are consistent. Furthermore, again, at least in principle, there is no reason why the creation of entirely new metaphysical systems could not be at least partially automated; and indeed, automated in such a way that these machines will be able to produce new and interesting systems in a reasonably efficient (e.g., timely) manner. But first, note that the mechanical search for new metaphysical systems need not be completely mechanical; humans could be a part of the process. Recall that different automated proving procedures are such that some are fully automated whereas others require human interaction and involvement. So, we can make the mechan-

108 | Jason Megill and Daniel Linford ical search for the new metaphysics intelligent simply by a being a part of the process itself. Nevertheless, there appears to be no reason why the search for new metaphysical systems could not be largely (or even completely) automated. That is, there are various effective – and so computable – procedures that machines could follow that would allow them to produce novel metaphysical systems in an intelligent or efficient manner. Just to sketch one example: we could form a set A that consists of some of the prominent “axioms” or principles from the history of metaphysics all translated into, say, a suitable first-order language; e.g., A could contain the Principle of Sufficient Reason, the Principle of Causal Adequacy, and so on. Each principle could be associated with a letter: P, Q, etc. Then, the principles and their negations could be partitioned into all possible combinations, where each individual combination corresponds to a possible metaphysical system. For example, given just P and Q as axioms, there are four prima facie consistent combinations: (i) both are true, (ii) both are false, (iii) P is true and Q is false, and (iv) P is false and Q is true.¹⁸ Note that the number of possible systems will be 2n , where n is the number of principles in our initial set. So the number of possible systems will become quite unmanageable fairly quickly for a human metaphysician; yet computers could generate a great number of systems (though admittedly, there will be a point at which even our best computers would be overwhelmed). To explain this procedure more systematically: form a set A that consists of a number of basic principles from metaphysics, all translated into a logical language. Then, assign letters to these principles: call the first principle in the set “P”, the second “Q”, the third “R”, and so on until each principle is assigned a unique letter. Next, mechanically generate all possible combinations of these principles and their negations: so one combination might contain P, ¬Q, and R, while another might contain ¬P, Q, and R, and so on. One way that a computer could generate all possible combinations would be to string all of the letters together into a conjuction (e.g., P ∧ Q ∧ R) and then generate a truth-table for that conjunction. Each row would correspond to one possible combination of the principles and their negations, and every possible combination would be found in the truth-table. Once the possible combinations are determined, form a set that corresponds to each possibility: for example, one row might contain the Principle of Sufficient Reason and the negation of the Principle of Causal Adequacy; so, form a

18 We assume that, at least so far as we know, the various principles are logically independent and that, e.g., one does not entail the falsity of the other. It might turn out that there are various logical relationships between the principles, e.g., two of them entail a third, but some of these relationships will hopefully be discovered later in the process.

On Computable Metaphysics ... |

109

set that consists of these claims. Each set will correspond to a collection of axioms that could form the basis of a possible metaphysical system. Then, once we have these collections of axioms, these sets of sentences, logical inferences – and only logical inferences – could be used to generate additional theorems of each system; we could have the machine infer as many claims as is feasible. Rather than simply mechanically producing random strings, which would result largely in gibberish, this process would generate a number of potentially interesting metaphysical systems, systems that contain axioms that have been of interest to metaphysicians along with theorems that logically follow from these axioms. Finally, once these systems are suitably developed, i.e., once a number of theorems have been deduced from the axioms, additional effective procedures could be performed on these systems. Some of these procedures would be performed on a single, particular set. For example, above, an effective procedure was discussed for determining whether a given set is internally explicitly inconsistent. This procedure could be carried out on the various metaphysical systems. For instance, it might be that the 37th theorem deduced in a given system contradicts the 814th . And this contraction will tell us something; we could then infer that at least two of the axioms in that system are contradictory. But then, any system that contains those two axioms could be discarded. And some of these procedures could be used to compare two (or more) different metaphysical systems. For example, some of these sets might be explicitly inconsistent; above, we discussed an effective procedure for checking whether two sets are explicitly inconsistent, and this procedure could be used here. It might be that, say, the 374th theorem derived in one system contradicts the 745th theorem derived in a different system. And this contradiction will teach us something; e.g., we’ll be able to infer that one principle (or axiom of a system) is inconsistent with another principle (or axiom of the other system), even though these principles were not obviously or prima facie inconsistent. And there appears to be no reason why additional effective procedures could not be developed, ones that do not simply address the consistency of the various metaphysical systems, thereby making this process more intelligent and efficient. In sum, it does appear possible that at least in principle – and quite possibly in practice – machines could be used to develop novel metaphysics.

Bibliography Appel, K. and Haken, W. (1977), “The Solution of the Four-Color Map Problem”, in Sci. Amer.: 237, 108–121. Bagai, R. et al. (1993), “Automatic theorem generation in plane geometry”, in Methodologies for Intelligent Systems, Berlin/Heidelberg: Springer, 415–424.

110 | Jason Megill and Daniel Linford Beavers, A. (2011), “Recent Developments in Computing and Philosophy”, in Journal for the General Philosophy of Science: 42, 385–397. Bibel, W. and Schmitt, P. H. (1998), Automated Deduction – A Basis for Applications (Foundations – Calculi and Methods, Vol. I; Systems and Implementation Techniques, Vol. II; and Applications, Vol. III), Dordrecht: Springer. Boolos, G. S. and Jeffrey, R. C. (1999), Computability and Logic, 3rd Edition, Cambridge: Cambridge University Press. Cole, D. (2014), “The Chinese Room Argument”, in The Stanford Encyclopedia of Philosophy (Summer 2014 Edition), edited by E. N. Zalta. URL = . Craig, W. (1953), “On Axiomatizability Within a System”, in The Journal of Symbolic Logic: 18, 30–32. Davis, M. (2001), “The Early History of Automated Deduction”, in Handbook of Automated Reasoning: 1, 3–15. Duffy, D. A. (1991), Principles of Automated Theorem Proving, John Wiley & Sons, Inc. Gödel, K. (1931), “Über formal unentscheidbare Sätze der Principia Mathematica und verwandter Systeme I”, in Monash. Math. Phys.: 38, 173–198. Leibniz, G. W. (1998), Philosophical Texts, translated by R. Francks and R. S. Woolhouse, Oxford: Oxford University Press. Lokhorst, G. J. C. (2011), “Computational Meta-Ethics”, in Minds and Machines: 21(2), 261–274. Lucas, J. R. (1961), “Minds, Machines and Gödel”, in Philosophy: 36, 112–127. Lucas, J. R. (1996), “The Godelian Argument: Turn Over the Page”, BSPS Conference, Oxford. http://users.ox.ac.uk/ jrlucas/Godel/turn.html. MacKenzie, D. (1995), “The Automation of Proof: A Historical and Sociological Exploration”, in Annals of the History of Computing, IEEE: 17(3), 7–29. McCune, W. (1997), “Solution of the Robbins Problem”, in J. Automated Reasoning: 19(3), 263–276. Megill, J. (2012), “The Lucas-Penrose Argument”, in The Internet Encyclopedia of Philosophy, edited by J. Feiser and B. Dowdon. http://www.iep.utm.edu/lp-argue/. Megill, J., Melvin, T. and A. Beal (2013), “On Some Properties of Humanly Known and Humanly Knowable Mathematicsm”, in Axiomathes: 24(1), 81–88. Megill, J. and Melvin, T. (2014), “Computability and Human Symbolic Output”, in Logic and Logical Philosophy: 23(4), 391–401. Penrose, R. (1989), The Emperor’s New Mind, Oxford: Oxford University Press. Penrose, R. (1994), Shadows of the Mind, Oxford: Oxford University Press. Portoraro F. (2014), “Automated Reasoning”, in The Stanford Encyclopedia of Philosophy (Winter 2014 Edition), edited by E. N. Zalta. URL = . Russell, B. and Whitehead, A. N. (1910/1912/1913), Principia Mathematica, 3 vols, Cambridge: Cambridge University Press. Searle, J. R. (1980), “Minds, Brains, and Programs”, in Behavioral and Brain Sciences: 3(3), 417–424. Spinoza, B. (1994), A Spinoza Reader: The Ethics and Other Works, edited and translated by E. Curley, Princeton: Princeton University Press. Turing, A. M. (1936), “On Computable Numbers, with an Application to the Entscheidungsproblem”, in Proc. London Maths. Soc.: 2.42, 230–265.

On Computable Metaphysics ... |

111

Turing, A. M. (1939), “Systems of Logic Defined by Ordinals”, in Proc. London Maths. Soc.: 2.45, 161–228. Turing, A. M. (1950), “Computing Machinery and Intelligence”, in Mind: 49, 433–460. Wang, H. (1960), “Toward Mechanical Mathematics”, in IBM Journal of Research and Development: 4(1), 2–22. Wang, H. (1981), Popular Lectures on Mathematical Logic, Mineolam NY: Dover. Whitehead, A. N. and Russell, B. (1984), Principia Mathematica, Medusa-Verlag. Zalta, E. and Fitelson, B. (2007), “Steps Towards a Computational Metaphysics”, in Journal of Philosophical Logic: 36(2), 227–247. Zalta, E. and Oppenheimer, P. (2011), “A Computationally-Discovered Simplification of the Ontological Argument”, in Australasian Journal of Philosophy: 89(2), 333–350.

Uwe Meixner

What Evil Must Be in Order to Exist If God created the world, then what we call “the evil in the world” is made possible, and practically unavoidable, by the conjunction of four facts: (1) the fact that God created a finite world and determined the laws of nature to be precisely thus as he in fact determined them to be (and not otherwise); (2) the fact that the created world — in spite of finite resources — is full of beings which are conscious and sensitive to pain, which want to live and to live well; (3) the fact that many creatures, in order to live and live well, use and consume other creatures against the will those creatures have to live and to live well, being more or less forced to do so by the constitution of the world and by their own will to live and to live well; (4) the fact that God gave to the created world a certain amount of independence, not with respect to its sheer existence, but with respect to its development over time: an independence that stems from absolute (ontic, non-epistemic) chance and from free will. However, is what we call “the evil in the world” really evil? Does the evil in the world really exist? Evil poses a problem for God’s perfection — and for his existence if God is, qua God, taken to be perfectly good, omnipotent, and omniscient — only if evil does itself exist. But the existence of evil has again and again been denied. It has been alleged that evil does not exist, more precisely: that it does not really exist and is therefore properly speaking non-existent, although it might in a sense be called “existent”; that evil is only the substanceless, in itself beingless, shadow of being, which will “pop up” quite unavoidably if beings distinct from God are present on all the countless levels of (greater or lesser) similitude to God in the manner of existence. It is alleged that the plenitude of creation, on all levels of being, is a consequence of God’s perfection; and so is, therefore, the shadow of negativity that necessarily accompanies that plenitude. This shadow is the evil in the world, and although God does not specifically will it, he does accept it and permit it. Indeed, he must accept it and permit it as a sort of “collateral damage” of his creating the world in God-imaging plenitude. Nevertheless, no opprobrium falls on God — for evil, it is alleged, does not exist, because it is “without substance”. Against this classical way of denying the existence of evil, it can, with justice, be objected that it makes use of an overly demanding concept of existence: According to the classical denial, to exist (in the primary, proper sense) means to be substantially real, that is: to be actual and integrated into a substance, or in other words: to be something actual that is a substance or belongs to the “contents” of a substance; in short, to exist means to be a res (cf. Summa Theologiae I, qu. 48, https://doi.org/9783110566512-008

114 | Uwe Meixner a. 2, ad 2; p. 239, column 1).¹ But this concept of existence cannot serve as an adequate general concept of existence (which is, for now, conceived of as predicative existence: existence which, in propositions, takes singular terms and not predicates as arguments, in striking contrast to quantificative existence). For according to that concept, evil is not the only item that is non-existent; according to it, many other plausibly existing items are non-existent, too: in the first place, all privative accidentia, whether individual or universal, and in the second place, all events, states of affairs, and external relations. Moreover, it can be objected that the argumentative goal — the well-justified denial of the existence of evil — is not quite attained. For if there were an actual personal substance which is essentially evil, then one could not help admitting — in spite of using the demanding concept of existence under consideration — that something evil exists. Note that it cannot be with certainty excluded that there is an essentially evil personal substance; even common human experience affords prima facie candidates which are far from being obviously inadequate. Defining (predicative) existence as substantial reality is, structurally, the same mistake as defining existence as necessary being, which definition would entail that, except for God and perhaps some ideal entities (numbers and the like), nothing at all exists (that is, exists properly speaking). It is, structurally, the same mistake as the mistake Plato made, and before him Parmenides, when he defined being or existence as changeless actuality, with the consequence that the entire empirically given world, and especially the material world as we know it from experience, does not exist, that is: does not properly speaking exist, and is therefore properly speaking non-existent, although it may in a sense be called “existent”; for the empirical world is certainly not changeless. The identical structural mistake in each of these three cases is the following: existence is conceived of in too narrow a sense. The motivation for such narrowness is not far to seek: it allows one to disregard, or at least to relegate to the darker corners of the background of one’s attention, areas of actuality that one feels disturbed or oppressed by in one’s intellectual dealings with the world — a manner of coping with actuality that is quite typical for philosophers, contemporary philosophers not excluded (in consideration of the fact that, like the majority of their predecessors, contemporary philosophers are fascinated by monistic ontologies). If one has an intellectual dislike of the contingent — well, if existence is taken to be necessary existence,

1 The second objection to the thesis that “malum invenitur in rebus” — in articulus 2 of quaestio 48 of the Summa theologiae — starts with the assertion “ens et res convertuntur”. Thomas is responding to this objection in qu. 48, a. 2, ad 2. Note that he does not reject “ens et res convertuntur”.

What Evil Must Be in Order to Exist |

115

then nothing contingent exists, or in any case, nothing contingent exists properly speaking (as one certainly prefers to put it), everything contingent does properly speaking not exist (though in a sense it may be taken to exist). If one has an intellectual dislike of change, and in particular of transience — well, if existence is taken to be changeless actuality, then all changing and transient items do not exist, or in any case, they do not properly speaking exist, they are properly speaking non-existent. And if evil is a thorn in one’s side (and for which person of good will is it not a thorn in her side?) — well, if existence and goodness are interchangeable (ens et bonum convertuntur), then nothing evil exists, or in any case, everything evil does not properly speaking exist, does properly speaking not exist. Two classical attempts to deny the existence of evil — or at least to belittle it (as not being properly speaking existence), for absolute denial seems counterintuitive — by means of narrowing the concept of existence have now been presented: (a) allegedly, existence conceptually requires substantial reality; (b) allegedly, existence conceptually requires goodness. Both these attempts originate in the philosophy of antiquity, primarily in Platonism, and were used by Christian thinkers — Augustine and Thomas Aquinas primarily — to relieve the considerable pressure that evil exerts (in human thought) on God, on his perfect goodness, on his omnipotence and omniscience, and, yes, on his existence itself. But there is yet another classical attempt of malum-elimination by purely conceptual means, an attempt, however, that has nothing to do with classical theistic, Christian philosophy. This other attempt can be extracted from the powerful appendix to the first part of Spinoza’s Ethica (ordine geometrico demonstrata). If to exist means to be actual and wholly objective (which, for Spinoza, is to be actual in the objectivity of the deus sive natura, we would say today: actual in the objectivity of natural science), then nothing evil exists; for according to Spinoza, everything evil is, qua evil, not purely objective. According to Spinoza, good and evil are merely a matter of parochial egocentric interests, first, of the species — of Man — in competition with the egocentric interests of other species, and second, of the human individual — this man, this woman — in competition with the egocentric interests of other human beings. Whether something is evil or not, is, according to Spinoza, always and necessarily relative to certain egocentric interests. The predication of “evil” varies with those egocentric interests: the same event — for example, the killing of a human being — which is evil relative to the egocentric interests of X, for example, the victim, is not at all evil relative to the egocentric interests of Y, for example, the killer. But now, what is wholly objective is ipso facto not relative to any egocentric interests whatsoever, and therefore: everything evil is, qua evil, not wholly objective (since what is evil is, qua evil, relative to certain egocentric interests). And therefore — now the specifically Spinozistic manner of malum-elimination by means of narrowing the concept of existence is put into operation — every-

116 | Uwe Meixner thing evil does not exist, or in any case, it does not properly speaking exist, that is, it does properly speaking not exist. Consider an event which, in common parlance, is said to be an “evil act”. If one insists that the act is evil, then it is, qua evil, not wholly objective, and hence it does not exist (not properly speaking); and if one insists that the act exists (properly speaking), then it is, qua existent, wholly objective, hence not evil. In any case, existence and evil are not to be had together in one and the same object.² Doubtless, this is counterintuitive; yet it is the inescapable result of conceptual decisions which have seemed plausible not only to Spinoza. And, note, the two Spinozistic suppositions that value-concepts are relative to egocentric interests (and therefore do not have a wholly objective content) and that existence is actuality in pure objectivity together entail the following: Everything good does not exist. For if value-concepts are relative to egocentric interests and if existence is actuality in pure objectivity, then also everything good is, qua good, not wholly objective, and hence (properly speaking) non-existent. In the atmosphere of true existence, of actuality in pure objectivity, sub specie aeternitatis, both good and evil disappear (in a sense). This is Spinoza’s vision and the basis of his peculiar theodicy, in which a very icy wind is blowing: the wind of superhuman indifference, called “perfection” by Spinoza: For many are wont to argue as follows: If everything follows from the necessity of the most perfect nature of God, how is it, then, that there are so many imperfections in Nature, the decay of things till they stink, the nauseating deformity of things, confusion, evil, sin, and so forth? But [. . . ] they are easy to refute. For the perfection of things is only to be appraised on the basis of their nature and power; and things are not more perfect or less perfect because they delight or offend the senses of human beings, or are amenable to human nature, or repugnant to it. To those, however, who query why God has not created all human beings in such a manner as to be ruled solely by the guidance of reason, I answer nothing else but this: Because he did not lack matter for creating everything, from the highest degree of perfection, doubtless, down to the lowest. Or speaking more properly: Because the laws of his nature were so vast that they sufficed to bring forth everything that can be conceived by an infinite intellect³ (the translation into English: U.M.).

2 According to this manner of thinking, what is not wholly objective, and therefore non-existent, may (and will as a rule) have an actual kernel which is wholly objective and therefore existent. What in an evil act can be described by using merely the vocabulary of physics constitutes such an actual and wholly objective kernel; Spinoza would not have doubted the existence of the purely physical goings-on that enter into the constitution of a murder. Spinozistically, one can, therefore, say that the evil act, the murder, does exist if taken in abstraction from its evilness, but does not exist if taken in non-abstraction from its evilness, or in other words: the murder, qua evil, does not exist; the murder, qua physical event, exists. 3 “Solent enim multi sic argumentari. Si omnia ex necessitate perfectissimae Dei naturae sunt consecuta, unde ergo tot imperfectiones in natura ortae? Videlicet, rerum corruptio ad faetorem

What Evil Must Be in Order to Exist |

117

Spinoza’s concept of perfection (that is, what he calls “perfection”) is a perfectly naturalistic and therefore perfectly objective concept, but it is also without axiological content that could be even remotely comprehensible to, and acceptable to, normal human beings. In this, Spinoza is very different from the Christian practitioners of the narrowing of the concept of existence with God-justifying intent; note also that they made only evil disappear (in a sense), not also good along with evil. Nevertheless, many people have seen a deep, a truly liberating wisdom precisely in Spinoza’s wholly objective — or rather: value-rejecting — conception of existence.⁴ It was not a new conception even in Spinoza’s time. Shakespeare makes Hamlet say en passant and playfully: “[T]here is nothing either good or bad, but thinking makes it so” (Hamlet, act 2, scene 2; p. 932, column 1); and Shakespeare, too, already had this from somewhere. Yet it is a conception that impresses us as decidedly modern. Wittgenstein, for example, shares it with Shakespeare (or Hamlet) and Spinoza: “In the world, everything is as it is, and everything happens as it happens; there is in it no value”⁵ (the translation into English: U.M.). When Wittgenstein is speaking of “value” here, negative value is certainly intended along with positive value. Moreover, the tautological formulations “is as it is” and “happens as it happens” are figures of speech: in front of the second “is” and the second “happens”, the expression “wholly objectively” is implicit. Explicitly formulated, Wittgenstein’s apothegm must, therefore, be phrased as follows: “In the world, everything is as it wholly objectively is, and everything happens as it wholly objectively happens; there is in it no positive or negative value.” If one is able to accept this, make it one’s own, live it, then one is with oneself, with the world, and with God — if one still believes in him (but Spinoza, in his own way, certainly did so) — at peace.

usque, rerum deformitas, quae nauseam moveat, confusio, malum, peccatum &c. Sed [. . . ] facile confutantur. Nam rerum perfectio ex sola earum natura, & potentia est aestimanda, nec ideo res magis, aut minus perfectae sunt, propterea quod hominum sensum delectant, vel offendunt, quod humanae naturae conducunt, vel quod eidem repugnant. Iis autem, qui quaerunt, cur Deus omnes homines non ita creavit, ut solo rationis ductu gubernarentur? nihil aliud respondeo, quam quia ei non defuit materia ad omnia, ex summo nimirum ad infimum perfectionis gradum, creanda; vel magis proprie loquendo, quia ipsius naturae leges adeo amplae fuerunt, ut sufficerent ad omnia, quae ab aliquo infinito intellectu concipi possunt, producenda” (Ethica, First Part, Appendix; p. 106 and 108). 4 Wholly objective does not in itself and uncontroversially entail value-rejecting; but in the eyes of many — and certainly in Spinoza’s eyes — it does. 5 “In der Welt ist alles, wie es ist, und geschieht alles, wie es geschieht; es gibt in ihr keinen Wert” (Tractatus, 6.41; p. 82).

118 | Uwe Meixner The attempts to deny the existence of evil that have been considered so far in this essay were based on a predicative, not a quantificative conception of : each attempt operated with its specific, rather exclusive concept of predicative existence. Predicative existence is characterized by the fact that “exists” connects with a singular term to form a proposition; for example, in the proposition “Pegasus exists”, “exists” connects with the singular term “Pegasus”. If the predicative conception of existence is embraced, then denying the existence of evil is bound to be logically equivalent with asserting the one or the other of the following two propositions: “Everything evil does not exist”, or in other words, “Nothing evil exists”. But, note, neither of these two (logically equivalent) propositions entails “Nothing is evil”. That nothing evil exists — that everything evil does not exist — does not logically exclude that something is evil: If no F is G, if every F is not G (or: if every F is non-G), it simply does not follow that nothing is F. Philosophers who believe that everything evil does not exist may, from the logical point of view, also believe that something is evil, indeed, that many things are evil; it is only that all this evil they acknowledge must nevertheless have no (proper) existence for them — say, because it is, qua evil, not substantially real, or not wholly objective, or, simply, not good. Now, a far more radical attempt to deny the existence of evil than the attempts considered so far can be launched if one proceeds on the basis of a quantificative, not a predicative conception of existence. The relevant concept of existence can be simply expressed by the word “something” — i.e., by a quantifier. Quantificative existence is characterized by the fact that “something” connects with a complete predicate to form a proposition; for example, in the proposition “Something is a winged horse”, “something” connects with the complete predicate “is a winged horse”. If the quantificative conception of existence is embraced, then the denial of the existence of evil is bound to amount to asserting the following proposition: “Nothing is evil”. In contrast to the predicative conception, the quantificative conception of existence does not invite the interpretation and re-interpretation of existenceexpressions. For when one is employing the quantificative concept of existence, one is simply formulating a proposition of cardinality: “Something is a winged horse” means nothing else than that the number of winged horses is either 1 or a larger number. The negation of “Something is a winged horse” is “Nothing is a winged horse”, and this, too, is simply a proposition of cardinality: it means nothing else than that the number of winged horses is 0. Therefore, if one wishes to justify that nothing is evil — in other words, if one wishes to deny the existence of evil while embracing the quantificative conception of existence — then this means that one must justify that the number of (what is) evil is 0. And in the proposition “The number of evil is 0” there is only one element which can be

What Evil Must Be in Order to Exist |

119

manipulated, only one switch, so to speak, that can be turned this way or that way, and this switch is the interpretation of the word “evil” itself. But has it ever been really attempted, in the course of the history of philosophy, to justify not only the assertion that nothing evil exists, but also the logically stronger assertion that nothing is evil? The first thing to be said in response to this question is that, right up to the present, many philosophers — including the founder of modern logic, Gottlob Frege — have confessed themselves entirely unable to distinguish the logical content of the one assertion from that of the other (whereas they would all be perfectly able, I presume, to distinguish, with respect to logical content, the assertion that nothing evil is good, or that nothing evil is substantially real, or that nothing evil is wholly objective, from the logically stronger assertion that nothing is evil). To those philosophers, the two sentences “Nothing evil exists” and “Nothing is evil” (or their counterparts in other languages) have seemed to have the very same logical content, have seemed to express that content merely in different ways. And those philosophers are not entirely wrong. The reason why they are not entirely wrong is this: If one takes “to exist” in its widest possible sense under the predicative conception, according to which sense “to exist” means as much as “to be identical with something”, then, indeed, “Nothing evil exists” and “Nothing is evil” have the same logical content, because “Nothing evil is identical with something” and “Nothing is evil” do have the same logical content.⁶ Unfortunately, the widest possible sense of “to exist” was not the intended sense of “to exist” when philosophers asserted that nothing evil exists; as we have seen, they intended senses of “to exist” that were much narrower (or stronger), logically speaking, than the sense of “to be identical with something”. Whether philosophers believe or don’t believe that they are able to distinguish logically between “Nothing evil exists” and “Nothing is evil”, it will usually be apparent whether they are inclined, or not inclined, to assert that nothing is evil. Leibniz, doubtless, was inclined to assert that nothing is evil, considering his notorious thesis that the actual world, this world (“the world”, as one also says) is the best of all possible worlds; for Leibniz, the best of all possible worlds can only be one in which, in a certain (Leibnizian) sense, nothing is evil. In what sense? Consider here that Leibniz, if Voltaire’s Candide had come into his hands, would not have needed to read the book in order to be aware that this, allegedly, best of all possible worlds is full of calamities; this fact was known to Leibniz just as

6 Consider that “Something evil is identical with something” and “Something is evil” are logically equivalent — and so are, therefore, the negations of these propositions, “Nothing evil is identical with something” and “Nothing is evil”.

120 | Uwe Meixner much as it was known to Voltaire. However, for Leibniz, not one of the countless calamities of this world is evil (or an evil). Why is this so? The sense in which nothing is evil in this best of all possible worlds must be taken to be determined by Leibniz’s choice of the sense of the word “evil” (alternative senses of the quantifier “nothing” are not on offer for his choosing). According to Leibniz, only that is evil that stays evil even if the totality of the world is taken into account. In other words, Leibniz defined “evil” as “evil in relation to the whole (world)”. Now, Leibniz was convinced (on the basis of a priori proof, he thought) that the actual world, this world, is the best of all possible worlds; among other things, this meant for him that the actual world is a world in which nothing is evil in relation to the whole. Hence, according to Leibniz’s holistic conception of evil, the actual world is a world in which nothing is evil. Nothing, therefore, is evil — that is, simpliciter evil, evil in the (for Leibniz) primary sense; for of course many things are, though not in relation to the whole, yet in relation to a part of the whole (world) evil — evil not in global, but in particular respect. Leibniz was far from denying this. The strategy Leibniz employed is, in principle, well-known to us. If a calamity happens to us, then the following comforting thought may not be far from us: Who knows what it is good for? However, this thought hardly comes to us if we are conscious of failings on our side which contributed to bringing about the calamity, or if the calamity simply surpasses a certain size. In any case, that thought proposes to us to consider the larger context, the kind of context which, if taken into account, often makes the evaluation of something turn out to be rather different from what it was when the matter was more or less myopically considered — the larger context which we automatically, and quite reasonably, take to be the more appropriate context for evaluation. And if we insightfully consider the largest context (no doubt, the proper context for ultimate evaluation), that is, the totality of everything existing, the actual world as a whole — if we manage to do this, then, Leibniz firmly believed, everything which seemed evil at first turns out to be good in the end. But we are not really able to effect in us the insightful consideration of the largest context. Already for that reason alone we have no idea — Leibniz, too, had no idea — what good with respect to the grand totality could be in seemingly meaningless raging pain, in the violent death of millions, in the utter destruction of cities and regions. Nevertheless, what is radically evil in relation to a (comparatively) small part might be good, or at least not evil, in relation to the grand totality. How this might be is certainly incomprehensible for our limited intellects, but it would not be for the intellect of God. Thus, one is drawing some comfort — only some — from human ignorance, especially in this case, where God, his very existence, is taken to be in jeopardy as an object of belief for the believer.

What Evil Must Be in Order to Exist |

121

The idea to nullify (in cognition) what is prima facie evil by considering it in the context of the totality of the actual world, the idea to (cognitively) “bonify” it in the grand totality, cannot be carried out by any human being; it can, at best, only be firmly believed that this idea is realizable in principle (by a being with superhuman cognitive abilities) — for which realizability the idea must, of course, reflect the truth. Leibniz was not the first to have that idea and not the first to believe that it reflects the truth and is in principle realizable; the idea is recognizably employed, for God-exculpating purposes, already in the thought of Thomas Aquinas (and likely even earlier). It must be emphasized that Thomas, just like Leibniz, was far from denying that some things in the world are evil (malum) in relation to a part of the world. Thomas even affirms that something evil exists, provided “exists” (Thomas says “est ens” or “dicitur ens”) is understood in such a way as to be synonymous with the non-copulative “is [est]”⁷ taken in a very wide signification: “[H]oc modo etiam malum dicitur ens” (S. Th. I, qu. 48, a. 2, ad 2; p. 239, column 1). If, however, “exists” is taken in the much narrower sense according to which “ens [. . . ] convertitur cum re” (ibid.), that is, in the sense according to which “exists” means as much as “is substantially real” (this latter sense is certainly the primary one for Thomas), then Thomas affirms: “[H]oc modo, nulla privatio est ens: unde nec malum” (ibid.). And Thomas even believes, like Leibniz, that the actual world — “the world”, in short — is a world in which nothing is malum in relation to the totality (of the world), which belief amounts for him — as for Leibniz — to the belief that nothing is (simpliciter, in the primary sense) evil. Thomas, therefore, does not only deny the existence of evil by means of interpreting existence predicatively in some sense that makes predicative denial — the assertion that every evil is non-existent — rationally possible, even necessary (which is the form of evil-denial that emerges, regarding Thomas, at the end of the previous paragraph). He also denies the existence of evil by means of interpreting existence not predicatively but quantificatively and proceeding, like Leibniz, on the basis of a concept of evil (simpliciter, in the primary sense) that makes quantificative denial — the assertion that nothing is evil — at least rationally possible; it is, in fact, the same concept of evil that Leibniz uses, the holistic one, mentioned (again) just at the beginning of this paragraph: God, and Nature, and any agent, does what is better in relation to the whole, but not what is better in relation to an arbitrary part, unless it is considered in its function for the whole

7 The non-copulative “is” is the “is” which already by itself constitutes an entire predicate, as in “God is [Deus est]”.

122 | Uwe Meixner [. . . ]. This whole, however, which is the universe of creatures, is better and more perfect if it contains some things which can fall short of the good, and which occasionally do so, without God preventing it. [. . . ] [M]uch that is good would be eliminated if God did not permit anything evil [evil in parte!] to be [to be in a very wide sense!]. For fire would not be generated if air were not destroyed; nor would the life of the lion be conserved if the ass were not killed; nor would vindicating justice and suffering patience be praised if there were no iniquity⁸ (the translation into English: U.M.).

Taken as a whole, therefore, the world is good for Thomas, and indubitably so, in consideration of the authorities Thomas acknowledged; indeed, it is very good for him, as we can say in view of Genesis 1, 31. The many evils in it (Thomas did not close his eyes to them) are evil only in parte (compare footnote 8) — in relation to a part. In relation to the whole — in toto (compare footnote 8) — they are not evil; for they, by their evilness in parte, serve the greater goodness in toto. If they were lacking, the degree of goodness of the world would be lower, though the degree of goodness of parts of the world would be higher. The world, according to this way of thinking, is very good, and the many evils in it are evil only in relation to this or that part of it. Is there anything that is evil in relation to the whole (world)? The world, certainly, (being good) is not evil in relation to the whole (i.e., to itself), and — it is alleged — nothing else in the world is evil in relation to the whole (i.e., the world). This exhausts the universe of quantification of the actual world. Identifying the universe of quantification with the universe of quantification of the actual world, it is therefore true (i.e., true in the actual world): nothing (in the world) is evil in relation to the whole, that is, nothing is (simpliciter, in the primary sense) evil. This is asserted by Thomas as much as it is asserted by Leibniz. But these two great metaphysical optimists had no proof whatsoever for their thesis (no matter what Leibniz thought); it is for them a thesis of fundamental belief, and as such it seems to have the inestimable advantage of being irrefutable. Denied, however, it has been: other people, other experiences, other metaphysical attitudes.

8 “Deus et natura, et quodcumque agens, facit quod melius est in toto; sed non quod melius est in unaquaque parte, nisi per ordinem ad totum [. . . ]. Ipsum autem totum quod est universitas creaturarum, melius et perfectius est, si in eo sint quaedam quae a bono deficere possunt, quae interdum deficiunt, Deo hoc non impediente. [. . . ] [M]ulta bona tollerentur, si Deus nullum malum permitteret esse. Non enim generaretur ignis, nisi corrumperetur aer; neque conservaretur vita leonis, nisi occideretur asinus; neque etiam laudaretur iustitia vindicans, et patientia sufferens, si non esset iniquitas” (S. Th. I, qu. 48, a. 2, ad 3; p. 239, column 1).

What Evil Must Be in Order to Exist |

123

In the first place, one must point here to the movement of Gnosticism,⁹ which flowered in later antiquity. For the Gnostics, the world — the whole, the totality — was not at all good; for them, it was thoroughly evil; almost everything in it was unconditionally evil. For this reason, God could not be the creator of the world for the Gnostics. And the god told about in the Old Testament could not be God; for they did regard the god of the Old Testament as the creator of the world. For the Gnostics, Yahwe is an inferior impostor, certainly not the father of Jesus Christ. For them, the father of Jesus Christ is the unknowable, true god, who not only transcends the world but is also, as it were, at an infinite distance from it — aside from the fact that he sent mankind a redeemer. Secondly, one must point to the philosophy of Arthur Schopenhauer, whose deep metaphysical pessimism is a rather rare occurrence in western philosophy. According to Schopenhauer, too, the world is thoroughly evil; almost everything in it is unconditionally evil. The world is the manifestation of a will-to-exist who is divided against itself. This will — at first unconsciously, then with sensation and perception, then guided by reason — surges into being in uncountable particularizations that are pitted against each other in a destructive and utterly painful struggle for existence. On a very high level of its evolution, namely, on the level of humanness, the will-to-exist finally realizes its own nature, after infinite pain, and is enabled by this realization to negate itself at least in part (in the person of a human being), thus reaching a state of liberation from itself which is not dissimilar to the state that the Buddhists call “nirvana”. For Schopenhauer, God is no longer a factor to be reckoned with. Is this world a hell? Who has to live in this world under hell-like conditions will tend to agree with the Gnostics and with Schopenhauer: this world is indeed a hell. Who is more fortunate will tend to disagree, as is only to be expected in view of the fact that the ability of true and deep empathy with those who suffer is fairly underdeveloped among human beings. But the question whether this world is hell-like or not need not be answered; at least it is not a question of primary interest. The question whether the world is (on the whole) evil is of primary interest. If the world is (on the whole) evil, a thing that had rather not been, as the Gnostics and Schopenhauer believed, then — whether this world is a hell or not (in places it is not, otherwise there would be no philosophizing here and now) — . . . then the strategy of Thomas and Leibniz fails, the strategy of making the many local evils of the world vanish, as it were, in the goodness or even perfection of

9 It must be noted that aside from its more or less pronounced metaphysical dualism, gnosticism has no unitary form. By “Gnosticism”, I here mean Christian gnosticism, the paradigmatic form of which is Marcionism.

124 | Uwe Meixner the world as a whole: as being necessary for the global goodness. If the world itself is evil, then it cannot be denied that something is simpliciter, in the primary sense evil; this is so because the world, if it is evil, is trivially evil in relation to the whole, the totality of the world. Then it cannot be denied, moreover, that something evil exists, no matter how demanding is the concept of existence one uses; for the world certainly exists in a stronger sense than all of its (proper) parts, even all of its substantially real parts.¹⁰ Then also the double strategy of Spinoza fails, which consisted, on the one hand, in defining existence as a concept that is independent from the perspective of the egocentric interests of any particular being: as actuality in pure objectivity; and on the other hand, in binding good and evil with conceptual necessity to precisely some such perspective or other, thus cutting good and evil off from pure objectivity, and hence from existence. Spinoza’s strategy fails; for the world is certainly actual in pure objectivity, actual independently from any egocentric interests, and therefore exists in Spinoza’s sense. And if it is evil — under this hypothesis my considerations have stood just now, and still stand — then this, contrary to Spinoza, must be so wholly objectively, that is, be so, but not in the perspective of the egocentric interests of a particular being; for the perspective of the egocentric interests of a particular being does not refer to the world as a whole; it invariably refers only to a local, very small part of the world. (This is also the case for human beings; as “global players” they persistently confuse the world with the earth, and as “non-global players” they often confuse it — in practice if not in theory — with their wider environment, which is delimited by their radius of action.) But is it true that the world is (on the whole) evil? Thomas, Leibniz, and Spinoza, whatever their philosophical differences, agree in holding that it is not evil. But there are certain indications that it is evil after all. The world appears to be a closed system. Therefore, according to the Second Law of Thermodynamics, the temporal progress of the world is directed at an equal distribution in space of all the energy in it (the quantity of which is constant, according to the Law of the Conservation of Energy), at a levelling of all energy-differences. Globally, the degree of energy-levelling in the world is always increasing. In other words, the world is fixedly directed towards destruction and decay; the cosmic desert is constantly growing. Of course, the world is not everywhere and always going in this direction; it is only doing so on the whole, or as a whole. Consider this dramatic reversal: Whereas Thomas and Leibniz believed that locally there is malum in the world, but that the world as a whole is a bonum, even perfectum

10 One might still hold that existence requires goodness, and then conclude that the world does not exist from the (presumed) fact that the world is evil. But how plausible is that?

What Evil Must Be in Order to Exist |

125

(in relation to which every local evil in the world is a global good), natural science strongly suggests that locally there is bonum in the world, something that, for the moment, is not heading towards its ruin, but that the world as a whole is a malum: something that heads towards its complete ruin. The rare islands in space-time that oppose the cosmic decay, the living beings, are spatially and temporally extremely local, extremely confined: they are transitory, ephemeral phenomena. Global decay will sooner or later — usually very quickly — break their resistance. The most horrible thing is this: The global process of decay itself brings forth these phenomena, the living beings, apparently automatically (given that the conditions in this or that place, at this or that time, have the right quality and constellation) – only to destroy them later on. Apparently automatically, the global process of decay itself provides living beings with many means that might conceivably serve the purpose of resisting destruction — only to smash them in the end, as children will smash the castles in the sand which they build precisely for the purpose of smashing them. In particular, the global process of decay provides action-relevant consciousness to many living beings — which, if the animal is in an optimal state, is certainly its strongest weapon in the struggle for survival, but which, if the animal is not in an optimal state, turns out to be the locus of its pain and suffering. Another evil is this: higher living beings can maintain their life-processes only by destroying other living beings; thus, life itself serves annihilation: a living being destroys many others before it is itself destroyed. A further evil is this: consciousness does not only protect a living being; on the highest evolutionary level it makes the demands of the animal — and thus the chances for (at least) unhappiness — rise immoderately; on that level, consciousness finds reasons and occasions for annihilation — be it of one’s own life or the life of others — everywhere, even in the biologically irrelevant, and, to boot, produces for annihilation, with never failing inventiveness, the most efficient means. All of this is to be observed, even in a form that endangers all terrestrial life, in human beings. Thus, consciousness, too, serves annihilation. The world is directed towards decay and destruction (and is not, as certain appearances suggest, in a perpetual cycle of coming into being and passing away — what would be some comfort indeed). The world is heading towards the state of a universal desert in which everywhere in the universe the same temperature reigns forever — not far from absolute zero. Long before this, there will have been an end to mankind and to all life on earth. Not unlikely, mankind itself will have brought about its own demise; if not, something else will have taken care of that. The great end will come; it is as certain as the “little end” — our own death — is certain to come. Everything is under the rule of death. For this reason, one can only succeed by a special form of obtuseness in living up to what Spinoza recommended per im-

126 | Uwe Meixner plicationem: to be a free human being and, as such, to think of nothing less than of (one’s own) death (Ethica, Fourth Part, Prop. 67; p. 580). Why is the world like this? To this question, there is no scientific answer. If there is an answer, it must be a metaphysical one. Indeed, as a metaphysician, one cannot reject that question, and especially one cannot do this if, besides being a metaphysician, one professes: “I believe in God, the father almighty, creator of heaven and earth” — if, in other words, one has already assumed a certain basic metaphysical attitude, an attitude which obviously is especially hard to reconcile with the fact that the world is heading towards universal death. One does not come near an answer to the question of why the world is heading towards universal death by simply advancing the traditional doctrine of the Fall (of Man); for the world had begun to run its course of death long before Man appeared on the stage of evolution and was able to sin. All considerations on evil that refer to the free will of (some) creatures are irrelevant here.¹¹ Although the free will of human beings, and perhaps also of other rational creatures, can very well accelerate the death-run of the world (in any case, here on earth, as the past and the present amply show), the free will of rational creatures cannot stop it or reverse it — not even if each rational creature were, out of free will, perfectly good. To speak with the Bible: The form of this world passes away (1 Corinthians 7, 31), and we have here no continuing city (Hebrews 13, 14). As Rilke says, “We arrange it. It falls apart. We arrange it again, and fall apart ourselves” (from the Eighth Duinesian Elegy; translation into English: U.M.).¹² Man is not responsible for the fact of global decay. The global decay cannot be a deserved punishment for superhuman (immaterial) rational creatures; it is also not a deserved punishment for us human beings. On the contrary, from the beginning human beings have been victims of the global decay. Their status as victims is in considerable part also to be diagnosed with regard to the peculiar human proneness to do evil (which proneness is traditionally called “original sin”); for universal transitoriness, as an inexorable constant of nature, brings forth selfishness, immoderate craving, ruthlessness, envy, hardheartedness, avarice; it does so phylogenetically, by dint of biological evolution, hence hereditarily. The vices just named are at the root of the greater part of human evil doings, and with worst

11 In general, creaturely free will is accorded a far too great weight in the debate on theodicy. The so-called free will defense (advanced by Alvin Plantinga and many others in the course of the history of philosophy) is really no defense at all. 12 “Wir ordnens. Es zerfällt. Wir ordnens wieder und zerfallen selbst”; Rilke, Die Gedichte, p. 660.

What Evil Must Be in Order to Exist |

127

effect if they are combined with a sense of superiority — arrogance — and have forced reason and free will into their service.¹³ No, if God created the world, then he himself, not reacting to anything, has from the start doomed it to destruction — certainly in such a way that it is going down with a great amount of sparkling radiance, like fireworks at night, but also in such a way that it is going down with immense creaturely pain, like a sinking ship. But why is this so if God is perfectly good, almighty, and all-knowing? Or has something or someone other than God produced the world, something or someone whose malum of cosmic dimensions has been transferred, in the act of creation, onto the so-called “cosmos” and made manifest in it? This is the position of the Gnosis, and Schopenhauer’s, but completely unacceptable for every orthodox Jew, Christian, or Muslim. Or is the world — after its accidental or necessary coming into being — doomed to darkness by inner, absolute necessity, hence inexorably moved to its death entirely without God and also entirely without “the prince of this world”? Truly, this is the bleakest position, since it excludes all consolation, confronting us with the existence of evil not only undeniably but also without any contingent explanation and, consequently, without any hope of salvation. If the world is eo ipso — with intrinsic necessity — evil, if there is a First Evil in this sense, then the question of what evil must be in order to exist inexplicably and irredeemably has found its answer. For everyone, however, who continues to believe in God’s existence and who continues to believe that God created the world and is a loving father, there remains, on the one hand, the agonizing riddle that such a god (the Father Almighty) created this world, and, on the other hand, the more or less desperate hope of a change of this world for the better and best, a change brought about precisely by God (the Father Almighty). For those who believe in Christ, there is, moreover, consolation in the fact that God already acted similarly before, within the history of this world, within the human sphere. For them, there is a unique historical prototype of the fulfilment of a hope beyond hope and of continuing metaphysical incomprehension. There is, on the one hand, the riddle that this god (the Father Almighty) had such a man, Jesus, die on the cross (certainly not for our sins), and there is, on the other hand, the miracle not one of Jesus’s companions would have dared to hope for: that just this god resurrected just this man from the dead.

13 Free will which is forced into the service of evil stays free; it is only that its range of choices is reduced.

128 | Uwe Meixner

Bibliography Rilke, R. M. (1990), Die Gedichte [The Poems], Frankfurt a. M.: Inselverlag. Shakespeare, W. (1972), “Hamlet, Prince of Denmark”, in The Complete Signet Classic Shakespeare, edited by S. Barnet, New York: Harcourt-Brace-Jovanovich, 917–961. Spinoza, B. (1997), Die Ethik [Ethica], Latin and German, Stuttgart: Reclam. Thomas Aquinas (1988), Summa Theologiae, Milan: Editiones Paulinae. Wittgenstein, L. (1984), “Tractatus logico-philosophicus”, in Werkausgabe, Vol. 1, Frankfurt a. M.: Suhrkamp, 7–85.

Elisa Paganini

Normative Rules for Indeterminacy 1 Introduction J. Robert G. Williams (2012) proposed a single solution to two puzzles. The first puzzle is the plurality puzzle and its challenge is to characterize a single phenomenon running through cases normally classified as indeterminate. The second puzzle is the consensus puzzle and it invites us to find a consensus on the correct attitude to be assumed towards borderline cases of paradigmatically vague predicates. M. Eklund (2013) has recently questioned the adequacy of the first puzzle, as it cannot be taken for granted that there is a similarity between the phenomena grouped under the heading ‘indeterminacy’; moreover, not primarily intending to evaluate Williams’s solution, his main concern is the comparison between Williams’s solution to the puzzles and the Ambiguity view Williams rejects. My main concern, instead, is to evaluate Williams’s solution to the puzzles. After presenting his solution to the puzzles, I will argue that it is inadequate: while Williams proposes that there is no general normative rule governing God’s and humans’ belief attitudes towards indeterminacies, I claim that human rationality and philosophical inquiry require general normative rules leading our belief attitudes towards indeterminacies and that God’s belief attitudes are more difficult to define than Williams assumes.

2 Williams’s puzzles and his solution to them Williams groups together many philosophical challenges characterized by indeterminacy. Among them, let us just mention future contingents, semantic paradoxes (like the Liar), borderline cases of paradigmatically vague predicates, vague composition, counterfactuals and sentences with unsatisfied definite descriptions. He recognizes that people do not have the same epistemic attitudes towards such different philosophical puzzles related to indeterminacy; he therefore denies that all these indeterminacy-related problems prompt a single epistemic reaction shared by everyone. But he also denies that ignorance can explain what they have in common, because ignorance cannot explain all aspects of indeterminacy (for example, the inquiry-endingness often going with indeterminacy https://doi.org/9783110566512-009

130 | Elisa Paganini would remain unexplained if ignorance were considered the only explanation of indeterminacy). He therefore looks for a common characterization of all indeterminacies different from a shared, positive epistemic attitude and different from shared ignorance. He then narrows the scope of indeterminacy, reflecting on vagueness narrowly considered, i.e. borderline cases of paradigmatically vague predicates, and points out that there is a considerable lack of agreement among philosophers about the correct epistemic attitude to adopt towards them. He is once again challenged to find a characterization reaching consensus among philosophers on this matter. Williams finds the common characteristic of indeterminacy in general, and of vagueness narrowly considered in particular, in an alleged absence of general alethic normative rules governing people’s and God’s beliefs towards indeterminate sentences. The idea is that God has beliefs as people do and when God and humans come to consider an indeterminate sentence, they are not constrained by any general normative rule as to what to believe. The difference between God and humans concerning attitudes towards indeterminate sentences is that while humans are constrained by conventions in their epistemic attitudes towards indeterminate sentences, God is not constrained at all and he can indifferently believe or refuse to believe any indeterminate sentence as he pleases. Let us consider Williams’s words: The Normative Silence model of indeterminacy says there is no doxastic attitude it is right or wrong to take to p, when p is indeterminate. God would believe all truths and disbelieve all falsehoods. But when it comes to indeterminacies, God is unconstrained – he can believe or disbelieve what he likes. (Williams (2012), p. 222, emphasis in the original) The ‘local alethic norms’ should be thought of as having the status of conventions. They are regularities in our attitudinal reactions to indeterminate cases, regularities we play a role in sustaining, and that we have an interest in maintaining. There is no need to have the same convention for all cases of indeterminacy [...] Nor do the conventions need to cover all instances [...] The picture is from the start Pluralism-friendly. (Williams (2012), p. 223, my emphasis)

Let me sum up Williams’s proposal. Williams recommends characterizing indeterminacy in general and vagueness narrowly considered in particular as an absence of general normative rules governing both humans’ and God’s beliefs towards indeterminate sentences. Any constraint we humans experience when considering an indeterminate sentence is a local norm with the status of convention and God is not constrained by it. I will claim instead that humans’ belief attitudes towards indeterminacies are regulated by general normative rules and therefore that not all constraints we hu-

Normative Rules for Indeterminacy |

131

mans recognize when adopting a belief attitude are local or conventional, contrary to Williams’s proposal. As long as Williams’s claim explicitly concerns also God, I will try to consider God’s belief attitudes and I will explain why they are not so easy to handle as Williams assumes. In order to present my point of view, I consider two main types of situation in which humans or God decide to believe something about indeterminate matters: the situations in which we believe something in order to act and the situations in which we believe something about philosophical matters.

3 Believing something in order to act In many everyday situations, even though we are aware that we are confronted with indeterminate matters, we decide to have a belief attitude for pragmatic reasons: we have to decide what to believe in order to act. Let us consider some examples. Suppose that I believe that I will deliver a lecture tomorrow. I believe it even though I believe that future contingents are indeterminate and I believe it because it is important in order to plan my day. In particular, my belief that I will deliver a lecture tomorrow induces my preparing it today. Even though I may be prevented from delivering it and therefore my belief may turn out to be false, my belief is what explains my present behaviour (i.e. preparing the lecture). Or consider that I have to decide how to evaluate students’ examinations and I decide to establish a threshold for the number of mistakes acceptable in order to pass the examination and what is, instead, unacceptable. Once I have made this decision, I believe that a single, small mistake makes a difference between a sufficiently well prepared student and one who is not sufficiently well prepared. While I am aware I have a belief about a very indeterminate matter, I have to believe it in order to evaluate my students. The relevant question to consider now is the following: is there any general norm governing doxastic attitudes when intended for acting? I suppose that coherence is a general norm to which we humans try to conform when adopting doxastic attitudes in order to act and when we are confronted with indeterminate matters we try to have beliefs coherent with the others we already have. I am not claiming that human beings have coherent beliefs, I am claiming that - as rational beings - they consider the attempt to be coherent a general normative rule. And such a general normative rule is incompatible with the Normative Silence of Indeterminacy claimed by Williams. Let us now consider God. It may be tempting to suppose that, mutatis mutandis, God has beliefs about indeterminate matters when he has to make choices in

132 | Elisa Paganini order to act. Suppose that God has beliefs as to whether human beings are good or not in order to decide their destiny in the afterlife. It is evidently true that certain humans are good and therefore God believes that they are good; and it is true that certain others are not and God believes that they are not. But there are also indeterminate cases: human beings such that they are not really good and not really not good. If God wants to decide what to believe in those cases, he has to make a choice about indeterminate matters and believe, for example, that a very trivial sin makes the difference between good and not good humans. Suppose moreover that God is infallible, i.e., whenever he believes p, p is true. And imagine that he takes into consideration certain future contingents and decides to believe something about them. In such a case, his decision is an act: it transforms an indeterminate matter into something true. For example, if he believes that I will not deliver a lecture tomorrow, then – of course – I will not. In order to evaluate the role of God for establishing general normative rules governing indeterminacy, it may be useful to reflect on the following question: is God constrained to coherence as a general norm for his doxastic attitudes? It is difficult to say. If he is infallible, then he should have coherent beliefs. In order to see this, suppose (for reductio) that he is infallible and he believes both that I will deliver a lecture tomorrow and that I will not. A contradiction would follow. Therefore, if God is infallible, then coherence is a general norm of his doxastic attitudes, a norm he actually adopts. But it is coherent to believe that God is neither infallible nor coherent. And if he were not coherent (or at least did not consider coherence a constraint), then God would be radically unlike human beings: while human beings consider coherence a general norm at least ideally guiding their belief attitudes, God would not. And God’s attitudes would be irrelevant in order to point out the absence of general norms governing humans’ belief attitudes.

4 Believing philosophical matters Let us now abstract from our everyday life, forget about the many actions we have to carry out and the pragmatic decisions we are forced to make. Let us step back from lectures and the evaluation of students. Let us just think about philosophical matters. For a moment, a sense of relief seems to emerge. We do not have to believe anything in particular about what will go on in the future, nor have we to decide arbitrary thresholds of vague predicates. And we can even consider philosophical matters not at all pressing for our everyday life, like the correct attitude to have towards the Liar or the Truthteller (neither of which we have to believe

Normative Rules for Indeterminacy |

133

or disbelieve in order to act), or towards the continuation of completed fictional stories. What is the role of belief when we consider indeterminate matters from a philosophical perspective? Before giving an answer to this question, let me say what I would like to answer. I would like to say that when we consider indeterminacy from a philosophical perspective, we are mainly interested in the metaphysical grounds of indeterminacy, i.e., the origins of indeterminacy dictated by semantics on the one side and by the world on the other. And once we have come up with a satisfying metaphysical picture of the origins of different indeterminacies, we try to adopt a belief attitude towards indeterminacies adequate to the metaphysical pictures we have adopted. An ideal like this one dictates a general norm for belief attitudes to which philosophers are generally attracted and which contradicts the Normative Silence proposed by Williams. According to this ideal, it is not the case that philosophers believe or disbelieve indeterminate matters without any general constraint, they consider themselves instead obliged to adapt their belief attitudes to the metaphysical pictures they have adopted; and, as long as different philosophers assume different metaphysical views for each indeterminacy, they take different attitudes towards indeterminate matters, as Williams himself acknowledges in his paper. I have presented what I consider an ideal general norm for belief attitudes towards indeterminate philosophical matters at least for most philosophers, but – as I wrote – it is an ideal. Philosophy is not the consolation we would like it to be when we approach it and we may discover that it originates more doubts than beliefs. At least when considering certain indeterminacies, we may suspect there is no complete metaphysical picture of indeterminacy. Let me say something about it and then consider the complication it raises. Maudlin (2004) – the inspiration of Williams’s proposal – considers the Liar Paradox together with other semantic paradoxes and constructs ‘a semantics for a language with a single univocal truth predicate’ (Maudlin (2004), p. 28). One of the consequences of this semantics is that certain sentences, like for example the Liar and the Truthteller, do not have simply a truth-value different from truth and falsity, i.e., it is not true that they are ungrounded (where ‘ungrounded’ is a third truth-value different from truth and falsity); it is not even true that it is ungrounded that they are ungrounded and in general there is nothing true about the truth-value of such sentences. It is evident that if there is nothing true about the truth-value of such sentences, this creates a fracture in the metaphysical picture of the semantics of sentences. Another example is more pressing from my point of view. I argued (in Paganini (2012)) that if there is no boundary to the extension of paradigmatically vague predicates, then there are sentences allegedly describing borderline cases

134 | Elisa Paganini such that there is nothing true about their truth-value at least in a supervaluationist framework. Once again, if there is nothing true about the truth-value of such sentences, this prevents a complete metaphysical picture of vagueness from being given. These difficulties call into question the ideal normative rule I presented at the beginning of this section. According to the ideal presented, the general normative rule says that our beliefs should conform to the metaphysical pictures we adopt of indeterminacies. But if we have reason to believe that the metaphysical pictures of certain indeterminacies are not complete and that many sentences we would like to use in order to describe the metaphysical pictures themselves are without truthvalue, how should we regulate our beliefs on these matters? This question raises a difficult topic and the Normative Silence model may initially seem tempting. But I believe that we should resist the temptation if we want to take philosophical practice seriously. First of all, it is important to note that not all indeterminacies call into question the possibility of offering a complete metaphysical picture of them. It is therefore a general normative rule to conform to the ideal presented at the beginning of this section at least for as long as it is viable. When the difficulties presented above show up, other general, normative rules are advisable. One of them is to suspend judgment as far as possible on sentences of which we doubt that they have a truth-value. This norm by itself contradicts the Normative Silence proposed by Williams: if we suspect that certain sentences do not have a truth-value, we are not free to adopt any belief attitude we like towards them, but we consider ourselves constrained to suspend judgement and think more carefully about them. If we want to endorse a theory and have to explain why the metaphysical picture of certain indeterminacies is not complete according to it, we may be forced to use and to believe sentences not having a truth-value according to the same theory: in such a case, the general norm to be adopted is to try to be coherent. The proposal presented here is intended to provide a general norm governing beliefs about philosophically indeterminate matters. To see that it is general, we have to realize that it does not depend on any specific indeterminacies, it regulates belief attitudes concerning all indeterminacies. It maintains that we should conform to the ideal presented at the beginning if we can obtain a complete metaphysical picture of the indeterminacy under consideration; when we cannot have a complete metaphysical picture of it, either we suspend judgement or, if we endorse an incomplete theory, we should try to be coherent. I presented what I consider a general norm regulating belief attitudes towards indeterminate philosophical topics taking into consideration only human beings. But Williams explicitly mentions God and so I suppose I should consider the following question: does God have belief attitudes towards philosophical topics in

Normative Rules for Indeterminacy |

135

general and philosophically indeterminate matters in particular? It is very difficult to give an answer to such a question. It may be claimed that philosophy is a human activity and that God’s epistemic attitudes towards such matters are so different from ours to prevent a comparison. But if God had propositional beliefs towards philosophically indeterminate matters, I tend to believe that he would be constrained as we are and, in such a case, he would be an ideal model to emulate.¹

Bibliography Eklund, M. (2013), “Williams on the Normative Silence of Indeterminacy”, in Analysis: 73(2), 264–271. Maudlin, T. (2004), Truth and Paradox. Solving the Riddles, Oxford: Clarendon Press. Paganini E. (2012), “God’s Silence”, in Philosophical Studies: 157(2), 287–298. Williams, J. R. G. (2012), “Indeterminacy and Normative Silence”, in Analysis: 72(2), 217–225.

1 I am indebted to Alfredo Tomasetta who read previous versions of this work and gave some useful advice.

Eleonore Stump

The Openness of God: Eternity and Free Will 1 Introduction The understanding of God’s mode of existence as eternal is foundational for very many other views of God in the history of philosophy of religion. The concept of eternity also makes a significant difference to a variety of issues in contemporary philosophy of religion, including, for instance, the apparent incompatibility of divine omniscience with human freedom and of divine immutability with the efficacy of petitionary prayer. But the concept has come under attack in current philosophical discussion as inefficacious to solve the philosophical puzzles for which it seems so promising. Although Boethius in the early 6th century thought that the concept could resolve the apparent incompatibility between divine foreknowledge and human free will, some contemporary philosophers, such as Alan Plantinga, have argued that eternity gives no help with this problem. Other philosophers, such as William Hasker, have argued that whatever help the concept of eternity may give with that puzzle is more than vitiated by the religiously pernicious implications of the concept for notions of God’s providence and action in time. In this paper, I want to examine these arguments against the doctrine of God’s eternity. I’m going to focus especially on Hasker’s position, but I’ll look briefly at Plantinga’s as well. In various publications, William Hasker has argued for what he calls ‘the openness of God’. It is part of the openness of God, in Hasker’s view, that God does not have comprehensive knowledge of the future; in particular, the God of open theism lacks knowledge of the future free choices of human beings. Hasker sees his position as an alternative to classical theism, as represented, for example, by standard Thomism, which Hasker rejects. Hasker thinks that the Thomistic account of God as eternal or timeless solves the problem of foreknowledge and free will only at the cost of making God’s timeless knowledge useless to God in interaction with the temporal world.¹ Hasker says,

1 There are others who make similar claims. See, for example, Plantinga (1986) and Zagzebski (2011). https://doi.org/9783110566512-010

138 | Eleonore Stump I ... regard the doctrine of timelessness as coherent and intelligible. . . . But divine timelessness. . . does not help. . . in enabling us to understand God’s actions in providence and prophecy.²

Here I will first examine Hasker’s argument for thinking free will and timeless knowledge are compatible, and I will give reasons for thinking that Hasker’s argument is itself incompatible with the doctrine of God’s eternity. Then I will try to show that considerations derived from the doctrine of eternity yield a more effective way to argue for the same conclusion. Finally, I will use those same considerations to undercut Hasker’s conclusion that timeless knowledge could be of no use to God in guiding his actions in time.

2 Eternity Boethius, who gives the classical definition of eternity, says that eternity is “the complete possession all at once of illimitable life”.³ Eternity is a timeless mode of duration, but nothing in the concept of eternity denies the reality of time or implies that temporal duration or temporal events are illusory.⁴ Boethius and others who accept the concept of eternity suppose that reality includes both time and eternity as two distinct modes of duration, neither of which is reducible to the other or to any third thing. Nonetheless, it is possible for inhabitants of the differing modes of duration to interact. To understand the nature of the interactions, it is important to see that, as Boethius and many others in the traditions of the major monotheisms understand it, eternity is a mode of existence characterized by both the absence of succession and also limitless duration. Temporal events are ordered in terms of the A-series – past, present and future – and the B-series – earlier than, simultaneous with, later than. Because an eternal God cannot be characterized by succession, nothing in God’s life can be ordered in either of those series. Moreover, no temporal entity or event can be past or future with respect to, or earlier or later than, the whole life of an eternal God, because otherwise God would himself be part of a temporal series.

2 Hasker (2004), p. 100. 3 The translation of Boethius’s definition is one Norman Kretzmann and I constructed; see Stump and Kretzmann (1981). 4 See our Stump and Kretzmann (1981), Stump and Kretzmann (1991) and Stump and Kretzmann (1992).

The Openness of God: Eternity and Free Will |

139

On the other hand, eternity is also characterized by limitless duration, that is, the duration of a present that is not limited by either future or past. Because the mode of existence of an eternal God is characterized by a limitless and atemporal kind of presentness, the relation between an eternal God and anything in time has to be one of simultaneity. Of course, the presentness and simultaneity associated with an eternal God cannot be temporal presentness or temporal simultaneity. Taking the concept of eternity seriously involves recognizing that it introduces technical senses for several familiar words, including ‘now’, ‘present’, and ‘simultaneous with’, as well as for the present-tense forms of many verbs. The relations between eternity and time therefore require a special sense of ‘simultaneity’. In earlier work, Norman Kretzmann and I called this special sort of simultaneity ‘ET-simultaneity’, for ‘simultaneity between what is eternal and what is temporal’. A relationship that can be recognized as a kind of simultaneity will of course be symmetric. But, since its relata have relevantly distinct modes of existence, ET-simultaneity will be neither reflexive nor transitive. In particular, each of two temporal events can be ET-simultaneous with one and the same eternal event without being ET-simultaneous with each other. Given the doctrine of eternity, God does not have foreknowledge. He knows any given thing or state of affairs that is a future contingent with respect to us only as it itself is temporally present, and not as it is future. For the same reasons, God cannot change the past or act on the future. Such actions require a temporal location, without which there can be neither past nor future. Nonetheless, the proponents of the doctrine of eternity thought that, in the eternal present, God can directly know and affect events that are past or future with respect to us in time. For example, God can will in the eternal present that something occur or that something come into existence at any particular point in time, including those points that are past or future with respect to us. With this much review of the doctrine of eternity, we can now turn to a sketch of Hasker’s position as regards God’s eternal knowledge of future free choices.

3 Hasker’s position Hasker begins the development of his position as regards God’s timeless knowledge by examining a much-discussed argument of Alvin Plantinga’s which attempts to show that taking God’s knowledge to be timeless does not solve the

140 | Eleonore Stump problem of foreknowledge and free will.⁵ In this argument, Plantinga is making use of a common intuition, namely, that divine eternity is somehow now as fixed and determinate as the past is. Linda Zagzebski puts that intuition this way: [W]e have no more reason to think we can do anything about God’s timeless knowing than about God’s past knowledge. The timeless realm is as much out of our reach as the past.⁶

And so, she says, the timelessness move does not avoid the problem of theological fatalism since an argument structurally parallel to the basic argument [for the incompatibility of foreknowledge and free will] can be formulated for timeless knowledge.⁷

Here is Plantinga’s version of such an argument (with dates changed for the sake of the discussion here): Suppose in fact Paul will mow his lawn in 2095. Then the proposition God (eternally) knows that Paul mows in 2095 is now true. That proposition, furthermore, was true eighty years ago; the proposition God knows (eternally) that Paul mows in 2095 not only is true now, but was true then. Since what is past is necessary, it is now necessary that this proposition was true eighty years ago. But it is logically necessary that if this proposition was true eighty years ago, then Paul mows in 2095. Hence his mowing then is necessary in just the way the past is. But, then it neither now is nor in future will be within Paul’s power to refrain from mowing. ⁸

Plantinga thinks that since this argument makes use of the notion of God’s eternal knowledge and nevertheless leads to the conclusion that Paul’s “mowing [in 2095] is necessary in just the way the past is ..., the claim that God is outside of time is essentially irrelevant”⁹ to any solution to the problem of foreknowledge and free will. Plantinga’s argument depends on taking the past truth of the proposition God eternally knows that Paul mows in 2095 as a hard fact about the past, to which the fixity of the past applies. But Hasker argues that whether or not this is a hard fact about the past depends on whether the proposition God eternally knows that

5 Hasker (1989). Hasker takes himself to have given conclusive arguments for his view of eternity and free will in this book, and I will here concentrate on his arguments in that book. 6 Zagzebski (2011). 7 Zagzebski (2011). 8 Plantinga (1986), p. 239. 9 Plantinga (1986), p. 240.

The Openness of God: Eternity and Free Will |

141

Paul mows in 2095 is itself a hard fact. From Hasker’s point of view, the success of Plantinga’s argument depends on whether or not “propositions about the eternal acts of God [are] ‘necessary’ in the same way in which the past is necessary”.¹⁰ On the one hand, in the spirit of the intuition expressed by Zagzebski, Hasker claims that it certainly seems as if they are. He says, as of the present moment, it is in many respects not yet determined how the future shall be ... . God’s timeless eternity. . . certainly cannot be open in this way; every fact is determined to be as it is, and not in any other way.¹¹

On the other hand, however, Hasker says that when an eternal God looks at time: “God distinguishes necessities and contingencies [in time] even though there is no contingency left in the latter in the form in which they reach His gaze.”¹² And so God looks at all of time as a temporal being would look at the temporal past. Hasker thinks that it follows from this that we are related to God’s eternal present as we are related to the future: if God in his eternity looks upon our time as one would look back on the past, it follows that in a certain respect we can view, or rather conceive of, eternity as we conceive of the future!¹³ And from this claim, Hasker goes on to infer that “eternity is like the future, and unlike the past, in that it is still open to our influence.¹⁴

Consequently, Hasker says, facts about God’s eternal knowledge ... are not hard facts. ... [P] There are things that God timelessly believes which are such that it is in my power, now, to bring it about that God does not timelessly believe these things.¹⁵

And he concludes this way: If, and only if, this proposition [P] is possible, is the doctrine of divine timelessness consistent with libertarian free will.¹⁶

10 Hasker (1989), p. 174. 11 Hasker (1989), p. 174. 12 Hasker (1989), p. 175. Hasker is here quoting Arthur Prior, who himself attributes the thought to Anselm. 13 Hasker (1989), p. 175. 14 Hasker (1989), pp. 175–176. 15 Hasker (1989), p. 176. 16 Hasker (1989), p. 176.

142 | Eleonore Stump So, Hasker thinks Plantinga’s argument is unsuccessful. If it is not a hard fact that God eternally knows that Paul mows in 2095, then the necessity of the past does not apply to it; and so the inference in Plantinga’s argument to the conclusion that it is necessary that Paul mow is invalid. Nonetheless, Hasker thinks, this rebuttal of Plantinga’s argument should give no joy to the proponent of God’s timelessness, because it comes at a considerable cost. That is because it is impossible that God should use a knowledge “derived from the actual occurrence of future events to determine his own prior actions in the providential governance of the world.”¹⁷ Even if God’s timeless knowledge of the future is not incompatible with human free will, on Hasker’s view, God cannot use that knowledge in interacting with human beings.

4 Plantinga’s argument and Hasker’s objection Although Hasker is trying to defend the doctrine of eternity against Plantinga’s argument, his objections to Plantinga’s argument are themselves hard to square with the doctrine of eternity. In particular, the premises of Hasker’s argument for his crucial claim that facts about God’s eternal knowledge are not hard facts seem incompatible with the doctrine of eternity. Consider, to begin with, Hasker’s statement that when God looks at time, he looks at it as if it were the temporal past, in which no contingency is left in anything that was once contingent. On the doctrine of eternity, it is not possible for God to be related to anything as past. On the contrary, everything in time is ETsimultaneous with the whole of God’s life. For the same reason, it is not true that for an eternal God all contingency has gone out of contingencies in time. God is related to contingent things as they are present, but nothing about this relation renders the contingent things past or non-contingent with respect to God. Someone might worry here that even if contingent things are present with respect to God, there is still the necessity of the present. But however exactly we are to understand the necessity of the present, it does not take away contingency. If Paula in the temporal present sees Jerome smile at her, it does not follow that Jerome’s smiling at her loses its contingency because it is present. What makes Jerome’s smiling contingent is the fact that he might not have smiled; her seeing his smile does not entail that any state of affairs prior to his smiling made his smiling necessary. And that fact about the contingency of his smiling stays the same

17 Hasker (1989), p. 176.

The Openness of God: Eternity and Free Will |

143

even though his smiling is present. Analogously, for God in the eternal present, the contingencies of time remain contingent even when in the eternal present God is related to them as ET-simultaneous with the whole of eternity. By parity of reasoning, the doctrine of eternity also rules out Hasker’s claim that, with respect to things in the temporal present, eternity is future or relevantly like the future. On the doctrine of eternity, it is not possible for anything in eternity to be future with respect to time. The only relation that holds between the eternal present of God and any events in time is ET-simultaneity. At any point of time, the whole of eternity is present to that time with ET-simultaneity; nothing about the eternal present is future with respect to any time. Finally, consider the conclusion Hasker draws from these premises: “There are things that God timelessly believes which are such that it is in my power, now, to bring it about that God does not timelessly believe these things.” In other words, on Hasker’s view, in the eternal present God believes p; but I have it in my power in the temporal present to bring it about that in the eternal present God does not believe p. But, on the doctrine of eternity, this is also impossible. There is no succession in eternity. And so it is not possible for an eternal God first to believe p and then to believe not-p. If there are things that God believes in the eternal present, those are the things that God believes; and it is not possible for him to believe things different from those. A fortiori, it is not in anyone’s power in the temporal present to bring it about that in the eternal present God believes things different from those that he believes [had believed?] in the eternal present. So, it seems that the premises of Hasker’s argument against Plantinga are not compatible with the doctrine of eternity, and the conclusion he draws from them seems incompatible with the doctrine as well. Consequently, Hasker’s attempt to rebut Plantinga’s argument and defend the compatibility of free will and God’s eternal knowledge is not successful.

5 Plantinga’s argument and the doctrine of eternity Nonetheless, in my view, Hasker’s evaluation of Plantinga’s argument is right: Plantinga’s argument does not succeed in demonstrating that there is an incompatibility between free will and God’s eternal knowledge. The compatibility of free will and God’s eternal knowledge can be defended against Plantinga’s argument in a way different from Hasker’s.

144 | Eleonore Stump From the past truth of a proposition about God’s eternal knowledge of a future event, Plantinga’s argument tries to show that the future event is somehow fixed or inevitable now, before the event occurs. In my view, the doctrine of eternity renders this move problematic. To see what difference the doctrine of eternity makes to this move, consider the same move on the supposition that God is temporal. On this supposition, if (a) In 1932 (g) God knows that in 2095 Paul mows is true,

then in 1932 there is a state of affairs that corresponds to (g). And that state of affairs is God’s knowing in 1932 that in 2095 Paul mows. Furthermore, in 1932 God knows that in 2095 Paul mows only if in 2095 Paul mows. So since in 1932 God does know this, then in 1932 the world must be the way God knows it to be. If in 1932 there were no mowing in 2095, then in 1932 the world would not be the way it must be for God in 1932 to know that in 2095 Paul mows; and so it would not be knowledge that God had in 1932. But since God does have this knowledge, then in 1932 it is the case that in 2095 Paul mows. Consequently, it is now (where, for purposes of this discussion, now is after 1932 and before 2095) the case that in 2095 Paul mows. If God were temporal, then, these inferences would be valid: (a) In 1932 (g) God knows that in 2095 Paul mows is true.

Therefore, (b) in 1932 God knows that in 2095 Paul mows.

Therefore, (c) in 2095 Paul mows.

Therefore, (d) it is now the case that in 2095 Paul mows.

But once we add in the doctrine of eternity, the inference from a suitably reformulated version of (a) to (b) is invalid, and it no longer supports (d) either.

The Openness of God: Eternity and Free Will |

145

On the doctrine of eternity, the state of affairs of God’s knowing that in 2095 Paul mows obtains in the eternal present. God’s eternal knowledge does not obtain in 1932, because it does not obtain at any temporal location whatsoever. In 1932, (g) is true only because in the eternal present God has the relevant knowledge, and the eternal present is ET-simultaneous with 1932. So, from (a′ ) In 1932 (g′ ) God in the eternal present knows that in 2095 Paul mows is true.

it does not follow that (b) in 1932 God knows that in 2095 Paul mows,

because God’s knowledge cannot be temporally located in 1932. So much is relatively uncontroversial. It is also the case, however, that if (a) is suitably reformulated as (a’), it no longer supports (d). When the object of the knowledge God has in the eternal present is something temporal, then what is known by an eternal God has a temporal location; but it does not share that temporal location with God’s knowing of it. Instead, God’s knowing is ET-simultaneous with the temporal location of what is known. God’s knowing in the eternal present that in 2095 Paul will mow is ET-simultaneous with the time in 2095 when Paul mows. Certainly, God’s knowing in the eternal present that in 2095 Paul mows requires that in 2095 Paul mows. If there were no mowing on Paul’s part in 2095, then it would not be knowledge that God has in the eternal present. But it is not the case that if in 1932 there were no mowing in 2095 to correspond to God’s knowing, then it would not be knowledge that God has in the eternal present. In order for it to be knowledge about Paul’s mowing that God has in the eternal present, it is sufficient that there be a relation of ET-simultaneity between God’s eternal present and the temporal location in which Paul mows. And there is, since God is ET-simultaneous with every time, including the time in 2095 when Paul mows. But it does not follow that it is the case now, in the temporal present, that in 2095 Paul mows. In order to ground God’s knowledge of Paul’s mowing in 2095, it is not necessary that Paul’s mowing in 2095 somehow obtains or is fixed already in the temporal present. What grounds God’s knowledge obtains in 2095; and, unlike God, the temporal present is not simultaneous in any sense with respect to 2095. In other words, from

146 | Eleonore Stump (a′ ) In 1932 (g′ ) God in the eternal present knows that in 2095 Paul mows is true.

it follows that (c) in 2095 Paul mows.

But it does not follow and is not true that (d) it is now the case that in 2095 Paul mows.

Of course, from the denial of (d) it does not follow that (e) it is now the case that in 2095 Paul does not mow.

Because in the eternal present God can be ET-simultaneous with future events that do not yet obtain in the temporal present, God’s knowledge can have a grounding in something future with respect to us without its being the case that the future event is already fixed in the temporal present. Now, in the temporal present, neither Paul’s mowing nor his not mowing is fixed. Nonetheless, in the eternal present God can know that in 2095 Paul mows, since God is ET-simultaneous with the time in 2095 at which Paul’s mowing occurs. Claim (a’) is true because there is a relation of ET-simultaneity between the eternal present and 1932, a time past with respect to us. And claim (g’) is true because there is a relation of ET-simultaneity between the eternal present and Paul’s mowing in 2095, a time future with respect to us. But ET-simultaneity is not a transitive relation. From the fact that 1932 is ET-simultaneous with the eternal present and the eternal present is ET-simultaneous with 2095, it does not follow that 1932 is simultaneous with 2095. And so Paul’s mowing in 2095 is not something that is the case in 1932. It is therefore also not the case that, because of God’s timeless knowledge of it, it is necessary with the necessity of the past. The intransitivity of ET-simultaneity invalidates all inferences of the form ‘It was true that God knows p; therefore, it is now the case that p’, where ‘p’ ranges over future contingents. So the crucial claim of Plantinga’s argument can be true: Necessarily, if God eternally knows that Paul mows in 2095 was true eighty years ago, then Paul mows in 2095;

and yet the conclusions Plantinga derives from this claim can be false. It follows from this claim that in 2095 Paul mows, but it does not follow that it is now nec-

The Openness of God: Eternity and Free Will |

147

essary that in 2095 Paul mows or that Paul has no power over whether or not he mows in the future. God’s knowledge in the eternal present of events that are present to him but future with respect to us does not imply that those future events are the case in the temporal present, fixed somehow before they actually occur in time. And so God’s knowledge in the eternal present of events future with respect to us is compatible with human free will in those future events.

6 Hasker on the uselessness of eternal knowledge Similar reflections also undercut Hasker’s reasons for rejecting the usefulness of God’s eternal knowledge to God’s ability to act in time. What Hasker actually says is, it is impossible that God should use a knowledge derived from the actual occurrence of future events to determine his own prior actions in the providential governance of the world.

Here Hasker is presupposing that God’s actions are prior to the occurrence of future events, but this presupposition is impossible on the doctrine of eternity. Nothing in the eternal life of God is prior with respect to anything in time, and nothing in time is future with respect to anything in the eternal life of God. We can, however, reformulate Hasker’s reasons so that they do not inadmissibly attribute temporal succession to an eternal God. Hasker’s thought seems to be or to depend on the other side of the coin of the intuition expressed in the quotation from Zagzebski above: the present is fixed and determinate, the eternal present as much as the temporal present. In either mode, once something is present, it seems that nothing can be done to alter it, not even by God. On Hasker’s way of thinking about it, when God knows future events, they are already there for him to know; and so Hasker is attributing a temporal ordering to the relation between future events and God’s knowledge. But even if there is no temporal succession as between future events and God’s knowledge, there is a logical order; an event’s obtaining is logically prior to God’s knowing it. So even if the future events are not already there for God to know, it still seems as if the future events must be there in order for God to know them. And, in that case, Hasker’s point still seems to apply: since a future event must be there for God to know it, it seems that God cannot use his knowledge of that future event to act on it. And for that reason, God cannot act on a future event in light of his knowledge of it.

148 | Eleonore Stump And so we have the conclusion Hasker wants, without attributing succession to an eternal God. Even if it is eternal, God’s knowledge of things future with respect to us seems useless for any action of God’s on future events. In fact, we can make Hasker’s point stronger. On Hasker’s way of thinking about the matter, it seems that an eternal God cannot act in time at all. Every temporal event is ET-simultaneous with the whole of an eternal God’s life. So any act of God’s intended to have a causal effect at a particular time is ET-simultaneous with the things at that time. But then the things at that time are present to God. And if they are present, they are there, too. So in what way could God do anything about them? On Hasker’s approach, then, not only is an eternal God’s knowledge useless for guiding his interactions with things in time, but in fact God cannot act in time at all, with or without the guidance of knowledge. But here we might stop to consider how anything in time acts on anything else in time. Consider a relatively simple case, drawn from neurophysiology, of causal interaction between two neurons: neuron 1 causes neuron 2 to fire. Here is how the causal interaction works. In the axon of neuron 1, there are seminal vesicles, small membrane-enclosed sacs. Each seminal vesicle contains molecules of a neurotransmitter – ay, serotonin, for the sake of the example. When neuron 1 fires, the membranes of some of the seminal vesicles in the axon of neuron 1 fuse with the membrane of neuron 1’s axon at the axon terminal, the end of the axon. When a seminal vesicle’s membrane fuses with the membrane of the axon terminal, the seminal vesicle is opened; and its contents, the serotonin molecules, are spilled into the synaptic cleft, the small space between neuron 1 and neuron 2. Once in the synaptic cleft, a serotonin molecule moves to dock into a receptor on the cell membrane of a dendrite of neuron 2. When it does, the receptor opens up and allows ions to enter into neuron 2, thereby changing the transmembrane potential and contributing to the firing of neuron 2. Suppose that we think just about three temporally ordered events in the causal sequence in this example. Event 1 at t1 : causal interaction between the membrane of a seminal vesicle in neuron 1 and the cell membrane at the axon terminal of neuron 1 causes the membranes to fuse and the seminal vesicle to open. Event 2 at t2 : causal interaction between the serotonin molecules in an opened seminal vesicle and molecules in the synaptic cleft cause the serotonin molecules in that seminal vesicle to move across the synaptic cleft between neuron 1 and neuron 2. Event 3 at t3 : causal interaction between a serotonin molecule in the synaptic cleft and a receptor on the membrane of a dendrite of neuron 2 causes that receptor to open.

Two things about the exercise of causal power in these events are worth noting.

The Openness of God: Eternity and Free Will |

149

First, in each event, the thing exercising causal power co-exists with the thing on which its causal power is exercised. In event 1, the membrane of the seminal vesicle and the membrane of the axon terminal both exist at t1 . And the same point holds about the serotonin molecule and the molecules in the synaptic cleft in event 2 and about the serotonin molecule and the receptor in event 3. In these ordinary kinds of cases, the thing that exercises causal power is simultaneous with the thing its causal power is exercised on.¹⁸ Although God himself is not located at a time, God can meet this condition for causal influence on things in time in virtue of being ET-simultaneous with any thing in time. In the eternal present, God can will that there be a causal influence on things at a time; and the things at that time, whatever that time is, will be ETsimultaneous with God’s willing in the eternal present. Secondly, event 3 at t3 happens at least in part because of event 2 at t2 , and event 2 at t2 happens at least in part because of event 1 at t1 . But the because of relation here should not be confused with a temporal relation. As things are in the temporal world, the because of relation obtaining between one of these events and another takes place in a temporally ordered series. But it is the because of relation that is doing the work. By way of a help to intuition here, consider a petitionary prayer for healing made at t1 . Someone might suppose that an eternal God could not respond to this prayer because a response to prayer has to come after the prayer, but an eternal God cannot do anything after anything else. This supposition is mistaken, however. For something to be a response to a prayer, it has to occur because of the prayer. But this is not the same as occurring after the prayer, even if in the temporal world a response that occurs because of a prayer typically occurs after the prayer. In one and the same the eternal present, God can be aware of the prayer for healing at t1 and will that there be healing at t2 . In this case, although God’s willing of healing is not later than the prayer, it is nonetheless because of the prayer. And being because of the prayer is sufficient for God’s willing to count as a response to the prayer. Analogously, the movement at t2 of serotonin molecules across the synaptic cleft happens because of the fusing of the membrane of the vesicle with the membrane of the axon of neuron 1 at t1 . It is true that the exercise of causal power at t1 is temporally located prior to the effects of the exercise of that causal power at t2 . But the effects at t2 happen because of the causal influences operating at t1 , and not in virtue of the temporal location of the things exercising the causal influence.

18 My point is not that causal power is always exercised in this way, only that it can be and ordinarily is exercised in this way.

150 | Eleonore Stump Consequently, an event 2 at a time t2 could happen at least in part because of what God wills, even if God’s willing is not prior to the event at t2 . God in the eternal present could will to ward off some cause that (but for God’s causal intervention) would have destroyed neuron 1 and all its contents right after the fusing of membranes at t1 and right before the release of the serotonin into the synaptic cleft. Then what happens in event 2 at t2 would happen at least in part because of what in the eternal present God wills, even though God’s willing is not temporally ordered with respect to event 2. A similar point can be made about event 3. In the eternal present, God could will to ward off a cause that (but for God’s causal intervention) would have destroyed all the serotonin in the synaptic cleft in the period between t2 and t3 . Then what happens at event 3 would happen at least in part because of what God wills in the eternal present. The fact that God is not temporally ordered with respect to event 3 in no way hinders this from being the case. So the fact that event 2 is ET-simultaneous with God’s eternal present does not mean that God gets to event 2 too late to act on it, as it were. It is a mistake to suppose that God is unable to exercise causal influence on event 2 on the grounds that, for God, event 2 is there and fixed with the necessity of present. Because God is ET-simultaneous with what is prior to event 2, event 2 is what it is at least in part because of what God in the eternal present wills to happen at times prior to t2 . Since God is ET-simultaneous with every moment of time as that moment is present, God can exercise causal influence in the same manner at any time. What happens at t n+m happens at least in part because of the casual effects which God in the eternal present wills to happen at t n . In this way, without being himself in time, in one and the same eternal present, God can will in such a way that he exercises causal influence over the whole temporally ordered causal sequence of events in time. This interpretation of an eternal God’s actions in time can be applied also to God’s knowledge of things in time. In the example above, it is true that there is a logical dependence between event 2 and God’s knowledge of event 2. God knows event 2 because event 2 obtains, and not the other way around.¹⁹ But, in the eternal present, which is ETsimultaneous with t1 , God wills to exercise causal influence at t1 in such a way that event 2 happens at least in part because of what God wills to happen before t2 . God’s knowledge of event 2, then, depends on event 2; but event 2 itself depends on God’s causal influence at t1 . God’s knowledge of event 2 therefore includes knowledge of his own causal influence helping to bring about event 2.

19 Cf., Stump and Kretzmann (1998).

The Openness of God: Eternity and Free Will |

151

And, clearly, this conclusion generalizes. Suppose that time has an end, as well as a beginning, and that there is a last time, tn, as well as a first time, t1 . And suppose that at tn there is a last event E n . Although God knows E n because it is there, E n is there for God to know at least in part because of what God in one and the same eternal present wills to happen in the whole of time before t n . And, in the eternal present, God wills what he does with respect to all those causal influences on E n in light of everything God knows in the eternal present about every other event in time, which has also occurred at least in part because of what God in the eternal present wills to occur in the period prior to t n . Since God’s knowledge of E n takes place in the one eternal present that is the whole of God’s life, the knowledge that God has when he acts with respect to En is the same knowledge that God has with respect to events taking place at t1 . Consequently, for any events, from t1 to t n , God can use his knowledge of an event at one time to shape the events of a later time. With respect to any event, although God knows that event because it is there, the event is there at least in part because of the causal influence that in the eternal present God exercises in time. Since God in the eternal present knows everything that he wills to occur as a result of his causal influence in time, what happens happens at least because of God’s knowledge of it. So when God knows E n , E n is there for God to know at least in part because of the causal influence God in the eternal present exerts at times prior to t n . The flaw in Hasker’s argument for the uselessness of God’s eternal knowledge is the supposition that the logical dependence of God’s knowledge on the events known obviates God’s ability to use his knowledge to shape his actions. On the doctrine of eternity, the logical dependence of God’s knowledge on the events known does not rule out the causal dependence of those events on God’s acts, and those acts are included in God’s knowledge. And so, in this sense, the events are dependent on God’s knowledge. Because God is not temporally ordered with respect to events in time, God’s act of will with respect to any event at a time tm will be made in light of God’s knowledge of all the events in time, including those future with respect to us.

152 | Eleonore Stump

7 Conclusion In various other places, I have argued against Hasker’s view that the God of classical theism is religiously inadequate or disappointing.²⁰ In those places, I have tried to show that a simple, eternal, immutable, impassible God can be as intimate with human beings and responsive to them as any open theist could desire. For a classical theist such as Aquinas, God is a risk-taker, too.²¹ In this paper, I have not recapitulated those arguments for classical theism. Instead, I have focused on the second of Hasker’s reasons for rejecting classical theism, namely, that even if it could reconcile God’s timeless knowledge of the future with human free will, it has to do so in a way that makes God’s knowledge of the future useless for God’s governance of the world. As I have tried to show, the doctrine of eternity can resolve the problem of divine foreknowledge and free will without the cost Hasker supposes it to have.

20 See, most recently, Stump and Kretzmann (2011). See also the chapters on simplicity and on grace and free will in Stump (2003), and the papers listed in footnote 3 above. 21 The Thomistic God may, however, not be a risk-taker in precisely Hasker’s sense. Hasker defines divine risk-taking this way: God takes risks if he makes decisions that depend for their outcomes on the responses of free creatures in which the decisions themselves are not informed by knowledge of the outcomes. (Hasker (2004), p. 125) The Thomistic God is a risk-taker in the sense that he makes decisions that depend for their outcomes on the responses of free creatures, when those responses are not themselves determined by God. So Aquinas accepts both the biblical claim that God wants all human beings to be saved and the doctrine that some human beings are not saved. That this is so explains why Aquinas needs to distinguish between God’s antecedent and his consequent will. His antecedent will is what God would have willed if things had been up to him alone; his consequent will is what God in fact does will given what creatures freely will. From my point of view, the version of risk-taking engaged in by the Thomistic God is sufficient for real risk. Hasker says that God is a risk-taker in the sense that “creatures’ decisions may be contrary to God’s wishes, and in this case God’s intentions in making those decisions may be at least partly frustrated.” (Hasker 2004, p. 125). If we substitute ‘God’s antecedent will’ for ‘God’s wishes’ in Hasker’s claim, then Aquinas’s views commit him to the same claim, without the implication of frustration, since God’s consequent will is in harmony with the way the world is.

The Openness of God: Eternity and Free Will |

153

Bibliography Hasker, W. (1989), God, Time, and Knowledge, Ithaca, New York: Cornell University Press. Hasker, W. (2004), “Providence, Evil and the Opennes of God”, in Routledge Studies in the Philosophy pf Religion, vol. 3, London and New York: Routledge. Plantinga, A. (1986), “On Ockham’s Way Out”, in Faith and Philosophy: 3, 235–269. Reprinted in The Concept of God, edited by T. V. Morris, Oxford: Oxford University Press, 1987, 171– 200; and in The Dilemma of Freedom and Foreknowledge, edited L. Zagzebski, New York and Oxford: Oxford University Press, 1991. Stump, E. (2003), Aquinas, London and New York: Routledge. Stump, E. and Kretzmann, N. (1981), “Eternity”, in Journal of Philosophy: 78, 429–458. Stump, E. and Kretzmann, N. (1991), “Prophecy, Past Truth, and Eternity”, in Philosophical Perspectives: 5, 395–424. Stump, E. and Kretzmann, N. (1992), “Eternity, Awareness, and Action”, in Faith and Philosophy: 9, 463–482. Stump, E. and Kretzmann, N. (1998), “Eternity and God’s Knowledge: A Reply to Shanley”, in The American Catholic Philosophical Quarterly: 72, 439–445. Stump, E. and Kretzmann, N. (2011), “Eternity, Simplicity, and Presence”, in The Science of Being as Being: Metaphysical Investigations, edited by G. T. Doolan, Washington, DC: Catholic University of America Press, 243–263. Reprinted in God, Eternity, and Time, edited by Ch. Tapp and E. Runggaldier, Aldershot: Ashgate, 2011, 29–45. Zagzebski, L. (2011), “Foreknowledge and Free Will”, in The Stanford Encyclopedia of Philosophy (2011 Edition), URL = http://plato.stanford.edu/archives/fall2008/entries/free-willforeknowledge/.

Mirosław Szatkowski

The Recovery of St. Thomas Aquinas. Part I: The Doctrine of St. Thomas Aquinas and its Non-Analytical Versions* 1 A preliminary definition of Thomism and its varieties Thomas Aquinas (ca.1225-1274) was the most prominent philosopher and theologian of the thirteenth century. During his 50-year-long life he produced an enormous body of work showing the broad scope of his interests in both speculative and practical fields, moving from metaphysics and theology to ethics and politics. Aristotelianism as he understood it served as the foundation of his philosophy, even though Aristotle’s views were considered incompatible with Christian teaching by many in his time. Therefore, to understand Thomas’ thought, it is necessary to go back to Aristotle. But doing so raises some doubts as to whether Thomism should be defined on the basis of the views of Thomas Aquinas or Aristotle, and, consequently, if it was more theoretically appropriate to refer to Thomas’ form of Aristotelianism. Resolving this question is important because before recognising analytically oriented Thomism as a version of Thomism, there should be clarity as to the actual meaning of Thomism. However, it would be well beyond the scope of this paper to determine to what degree Thomas Aquinas is a faithful interpreter of Aristotle, and to what degree his doctrine incorporates new, original elements and refinements. Going forward, we will assume that although Aquinas adopted many ideas of Aristotle, his thought is entirely autonomous and not derivative. J. Haldane has given us a good conventional definition of the term ‘Thomism’: Thomism, conceived of as the set of broad doctrines and style of thought expressed in the works of St. Thomas Aquinas and of those who follow him, first emerged in the thirteenth century. (Haldane (1997a), p. 485)

* The second part of this paper, namely, “The Recovery of St. Thomas Aquinas. Part II: What is Analytically Oriented Thomism” was published in Analytically Oriented Thomism, edited by M. Szatkowski, Editiones Scholasticae, 1–53. This work (Part I and II) has been supported by the National Science Center under the project 2012/07/B/HS1/01978. https://doi.org/9783110566512-011

156 | Mirosław Szatkowski This definition comprises three significant aspects: the first embodies the corpus of the thinking of Thomas Aquinas, the second embodies the style of his thinking, and the third the thinking of his followers in the thirteenth and following centuries. It seems to me that the second element of the definition is rather hard to assess. On the one hand, it is hard to specify the style of philosophy characteristic of Thomas Aquinas. On the other hand, provided that the corpus of any philosophical doctrine is accepted, its followers should have some degree of latitude in interpreting and applying different research methods to it. What is the corpus of thought of Thomas Aquinas? Only its crucial tenents will concern us in this paper. The answer to this question is neither simple nor uncontroversial. Those who treat Thomas Aquinas as an interpreter of Aristotle claim that Aquinas expanded and applied the Aristotelian doctrines of actus and potentia in a consistent and logical manner, and this is the key to his doctrinal innovations (cf., Swartz (2009), pp. 120–121). Pope Pius X’s Postquam Sanctissimus of 27 July 1914 – extended this very narrow characterization of Thomas Aquinas’ thought. The 24 theses advanced in the document provide believers with the canon of Thomas’ doctrine. And the acceptance or rejection of all these claims is an indicator of being a Thomist or not. In Section 2, we quote these 24 theses. Clearly, if the term ‘Thomism’ refers not only to the original Thomas’ doctrine but also to its variants, then – consistently – the term ‘Analytically Oriented’ in the phrase ‘Analytically Oriented Thomism’ may also refer to these variants. In Section 3, we present an overview of the different variants that Thomas Aquinas’ doctrine has given rise to, which are other than analytic in the philosophical sense. In general, the difficulty here is to determine when a philosophical doctrine is analytically oriented and when it is not. An analytically oriented doctrine may be said to impose itself when it takes its method or approach from analytic philosophy. Again, it is not easy to determine all the elements that would signify that one is working in the analytic mode, as there is no agreed upon canon of characteristics that definitively determine if an approach is analytic or not, or whether it can be translated into other philosophical idioms. Despite this ambiguity, in Section 4 we will try to answer the question: What is analytic philosophy? We characterize analytic philosophy, on the one hand, from the perspective of philosophers classified as analytic and, on the other hand, from a theoretical perspective (the nature and program of a school of philosophy qualify it as analytic). After clarifying these issues an attempt to answer the question: What is analytical Thomism? is made in Section 5. Finally, in Section 6, we present opinions critical of analytical philosophy, and, consequently, of analytical Thomism, and take a critical stance towards these opinions.

Part I: The Doctrine of St. Thomas Aquinas and its Non-Analytical Versions |

157

2 The twenty-four fundamental thomistic theses In the decree Postquam Sanctissimus, Pope St. Pius X declared that the following 24 theses contain the principles and more important aspects of Thomas’ doctrine. Though the list does not give the exact wording of St. Thomas, the ideas are certainly those of St. Thomas. Consequently, a philosopher who intellectually accepts all these theses is a Thomist.

1.

2.

3.

4.

5.

6.

7.

8.

Ontology Potency and Act divide being in such a way that whatever is, is either pure act, or of necessity it is composed of potency and act as primary and intrinsic principles. (St Th. Ia. Q.77, a.1; Metaph. VII, 1 and IX, 1 and 9) Since act is perfection, it is not limited except through a potency which itself is a capacity for perfection. Hence in any order in which an act is pure act, it will only exist, in that order, as a unique and unlimited act. But whenever it is finite and manifold, it has entered into a true composition with potency. (St Th. Ia. Q.7, a.1-2; Cont. Gent. I, c.43; I Sent. Dist. 43, Q.2 Consequently, the one God, unique and simple, alone subsists in absolute being. All other things that participate in being have a nature whereby their being is restricted; they are constituted of essence and being, as really distinct principles. (St Th. Ia. Q.50, a.2, ad 3; Cont. Gent. I, c.38,52,53,54; I Sent. Dist.19, Q.2, a.2; De Ent. et Ess. c.5; De Spir. Creat. a.1 ; De Verit. Q.27, a.1, ad 8) A thing is called a being because of being (“esse”). God and creature are not called beings univocally, nor wholly equivocally, but analogically, by an analogy both of attribution and of proportionality. (St Th. Ia. Q.13, a.5; Cont. Gent. I, c.32,33,34; De Pot. Q.7, a.7) In every creature there is also a real composition of the subsisting subject and of added secondary forms, i.e., accidental forms. Such composition cannot be understood unless being is really received in an essence distinct from it. (St Th. Ia. Q.3, a.6 ; Cont. Gent. I, c.23; Cont. Gent. II, c.52; De Ent. et Ess. c.5) Besides the absolute accidents there is also the relative accident, relation. Although by reason of its own character relation does not signify anything inhering in another, it nevertheless often has a cause in things, and hence a real entity distinct from the subject. (St Th. Ia. Q.28, mainly a.1) A spiritual creature is wholly simple in its essence. Yet there is still a twofold composition in the spiritual creature, namely, that of the essence with being, and that of the substance with accidents. (St Th. Ia. Q.50 and ff; De Spirit. Creat. a.1) However, the corporeal creature is composed of act and potency even in its very essence. These act and potency in the order of essence are designated by the names form and matter respectively. (St Th. De Spirit. Creat. a.1)

158 | Mirosław Szatkowski Cosmology 9.

Neither the matter nor the form have being of themselves, nor are they produced or corrupted of themselves, nor are they included in any category otherwise than reductively, as substantial principles. (St Th. Ia. Q.45, a.4; De Pot. Q.3, a.5, ad 3) 10. Although extension in quantitative parts follows upon a corporeal nature, nevertheless it is not the same for a body to be a substance and for it to be quantified. For of itself substance is indivisible, not indeed as a point is indivisible, but as that which falls outside the order of dimensions is indivisible. But quantity, which gives the substance extension, really differs from the substance and is truly an accident. (St Th. Cont. Gent. IV, c.65; I Sent. Dist. 37, Q.2, a.1, ad 3; II Sent. Dist. 30, Q.2, a.1) 11. The principle of individuation, i.e., of numerical distinction of one individual from another with the same specific nature, is matter designated by quantity. Thus in pure spirits there cannot be more than individual in the same specific nature. (St Th. Cont. Gent. II, c.92-93 ; Ia. Q.50, a.4; De Ent. et Ess. c.2) 12. By virtue of a body’s quantity itself, the body is circumscriptively in a place, and in one place alone circumscriptively, no matter what power might be brought to bear. (St Th. IIIa. Q.75; IV Sent. Dist. 10, a.3) 13. Bodies are divided into two groups; for some are living and others are devoid of life. In the case of the living things, in order that there be in the same subject an essentially moving part and an essentially moved part, the substantial form, which is designated by the name soul, requires an organic disposition, i.e., heterogeneous parts. (St Th. Ia. Q.18, a.1-2 and Q.75, a.1; Cont. Gent. I, c.97; De Anima everywhere) Psychology 14. Souls in the vegetative and sensitive orders cannot subsist of themselves, nor are they produced of themselves. Rather, they are no more than principles whereby the living thing exists and lives; and since they are wholly dependent upon matter, they are incidentally corrupted through the corruption of the composite. (St Th. Ia. Q.75, a.3 and Q.90, a.2; Cont. Gent. II, c.80 and 82) 15. On the other hand, the human soul subsists of itself. When it can be infused into a sufficiently disposed subject, it is created by God. By its very nature, it is incorruptible and immortal. (St Th. Ia. Q.75, a.2 and Q.90 and 118; Cont. Gent. II, c.83 and ff.; De Pot. Q.3, a.2 ; De Anim. a.14) 16. This rational soul is united to the body in such a manner that it is the only substantial form of the body. By virtue of his soul a man is a man, an animal, a living thing, a body, a substance and a being. Therefore the soul gives man every essential degree of perfection; moreover, it gives the body a share in the act of being whereby it itself exists. (St Th. Ia. Q.76; Cont. Gent. II, c.56, 68-71; De Anim. a.1 ; De Spirit. Creat. a.3) 17. From the human soul there naturally issue forth powers pertaining to two orders, the organic and the non-organic. The organic powers, among which are the senses, have the composite as their subject. The non-organic powers have the soul alone as their subject. Hence, the intellect is a power intrinsically independent of any bodily organ. (St Th. Ia. Q.77-79; Cont. Gent. II, c.72; De Anim. a.12 and ff.; De Spirit. Creat. a.11) 18. Intellectuality necessarily follows upon immateriality, and furthermore, in such manner that the father the distance from matter, the higher the degree of intellectuality.

Part I: The Doctrine of St. Thomas Aquinas and its Non-Analytical Versions |

159

Any being is the adequate object of understanding in general. But in the present state of union of soul and body, quiddities abstracted from the material conditions of individuality are the proper object of the human intellect. (St Th. Ia. Q.14, a.1 and Q.74, a.7 and Q.89, a.1-2; Cont. Gent. I, c.59 and 72, and IV, c.2) 19. Therefore, we receive knowledge from sensible things. But since sensible things are not actually intelligible, in addition to the intellect, which formally understands, an active power must be acknowledged in the soul, which power abstracts intelligible likeness or species from sense images in the imagination. (St Th. Ia. Q.79, a.3-4 and Q.85, a.6-7; Cont. Gent. II, c.76 and ff. ; De Spirit. Creat. a.10) 20. Through these intelligible likenesses or species we directly know universals, i.e., the natures of things. We attain to singulars by our senses, and also by our intellect, when it beholds the sense images. But we ascend to knowledge of spiritual things by analogy. (St Th. Ia. Q.85-88) 21. The will does not precede the intellect but follows upon it. The will necessarily desires that which is presented to it as a good in every respect satisfying the appetite. But it freely chooses among the many goods that are presented to it as desirable according to a changeable judgment or evaluation. Consequently, the choice follows the final practical judgment. But the will is the cause of it being the final one. (St Th. Ia. Q.82-83 ; Cont. Gent. II, c.72 and ff.; De Verit. Q.22, a.5; De Malo Q.11) Theodicy 22. We do not perceive by an immediate intuition that God exists, nor do we prove it a priori. But we do prove it a posteriori, i.e., from the things that have been created, following an argument from the effects to the cause: namely, from things which are moved and cannot be the adequate source of their motion, to a first unmoved mover; from the production of the things in this world by causes subordinated to one another, to a first uncaused cause; from corruptible things which equally might be or not be, to an absolutely necessary being; from things which more or less are, live, and understand, according to degrees of being, living and understanding, to that which is maximally understanding, maximally living and maximally a being; finally, from the order of all things, to a separated intellect which has ordered and organized things, and directs them to their end. (St Th. Ia. Q.2; Cont. Gent. I, c.12 and 31 and III c.10-11; De Verit. Q.1 and 10; De Pot. Q.4 and 7) 23. The metaphysical motion of the Divine Essence is correctly expressed by saying that it is identified with the exercised actuality of its won being, or that it is subsistent being itself. And this is the reason for its infinite and unlimited perfection. (St Th. Ia. Q.4, a.2 and Q.13, a.11; I Sent. Dist. 8, Q.1) 24. By reason of the very purity of His being, God is distinguished from all finite beings. Hence it follows, in the first place, that the world could only have come from God by creation; secondly, that not even by way of a miracle can any finite nature be given creative power, which of itself directly attains the very being of any being; and finally, that no created agent can in any way influence the being of any effect unless it has itself been moved by the first Cause. (St Th. Ia. Q.44-45 and 105; Cont. Gent. II, c.6-15 and III c.66-69 and IV c.44; De Pot. mainly Q.3, a.7)

160 | Mirosław Szatkowski

3 Philosophical non-analytic orientations that describe themselves as ‘Thomistic’ Four approaches to Thomism are characterized in this section, namely, Aristotelian Thomism, Existential Thomism, Transcendental Thomism, and Lublin Philosophical School’s Thomism. The first two can be grouped under the name ‘Neo-Thomism’. We now turn to the question of what distinguishes these approaches from each other.

3.1 Aristotelian Thomism It has been observed that the phrase ‘Aristotelian Thomism’ unfortunatly assumes the existence of two autonomous philosophies: of Aristotle and of Thomas Aquinas; and that furthermore, Aristotelian Thomism should be considered as some interpretation of the philosophy of Thomas Aquinas, which overlaps with the philosophy of Aristotle. Meanwhile, many philosophers maintain that the philosophy of Aquinas is simply a new, revised Stagyrite’s philosophy – it is an old philosophy – Aristotle’s – in the new given to it by Thomas Aquinas. The philosophy of Thomas Aquinas is also called the Christian variety of Aristotelianism. In this view, Aristotelian Thomism should be regarded as some Thomistic interpretation of Aristotle’s philosophy. It is, of course, true that Thomas Aquinas wrote several commentaries on Aristotle’s works – among which there are On the Soul, Nicomachean Ethics and Metaphysics – and took many ideas from Aristotle – including the analysis of physical objects, the concept of perception and intelectual knowledge, the views on space, time and movement, as well as the Aristotle’s cosmology and moral philosophy¹ –, but Thomism cannot be considered as a sort of the Stagyrite’s philosophy. Thomas was also familiar with the achievements of other authors, from whom he took a lot of valuable ideas, for example, from Augustine and from Platonists: Boethius, Pseudo-Dionysius, and Proclus. This does not mean that Aquinas’ doctrine was a purely eclectic collection of ideas from other authors. The philosophy of Thomas Aquinas is primarily an original philosophy which cannot be totally characterized by any views of earlier thinkers. To point out some differences between Thomas Aquinas and Aristotle, consider their conceptions of God. For Aristotle, only what is finite and limited can be perfect. Aquinas rejects this and argues that God Himself is infinite and perfect. Aristotle

1 Cf., O’Callaghan and McInerny (2014).

Part I: The Doctrine of St. Thomas Aquinas and its Non-Analytical Versions |

161

rejects the concept of a personal God, while Aquinas embraces it. Aquinas also embraces the doctrine of the limitation of act by potency as well as the doctrine of participation, whereas Aristotle rejects both of them.² After these caveats, the phrase ‘Aristotelian Thomism’ seems less problematic. An account of Aristotelian Thomism should therefore point to the component (or, the components) of Aristotle’s philosophy, which was not at all or only partially accepted by Thomas Aquinas, in the light of which Aquinas’s doctrine is interpreted. Is there such a component (or, such components)? It is pertinent to remark that the alternative labels under which ‘Aristotelian Thomism’ has continued to attract contemporary support are ‘Laval Thomism’ and ‘River Forest Thomism’, the first coming from the name of the University of Laval in Quebec, where a representative of this approach, Charles De Koninck, was a professor, and the second is derived from a suburb of Chicago, the location of the Albertus Magnus Lyceum for Natural Science, whose members Benedict Ashley, James A. Weisheipl, and William A. Wallace mong the proponents of this approach.³ Starting with River Forest Thomism, we rely here on the works of its representatives Ashley (1991, 2006), Weisheipl (1961), and Reese (2016). Ashley formulated eight principal theses of River Forest Thomism⁴, which we quote here in a condensed form, as summarized by P. N. Reese. 1) 2) 3) 4) 5) 6) 7) 8)

Aquinas’s philosophy is best drawn from his Aristotelian commentaries. Aquinas should be interpreted as a convinced Aristotelian. Aquinas’s order for learning the sciences is also necessary for establishing those sciences; thus, metaphysics cannot be established without natural philosophy. Natural philosophy and natural science are united in object, scope, and method. Aquinas’s natural philosophy is best understood in light of Aristotle’s logical works, especially the Posterior Analytics. Seeming differences between contemporary natural sciences and natural philosophy are either only apparent differences, or are non-essential. Aristotelian natural philosophy provides the tools for resolving present-day scientific paradoxes. Neither metaphysics nor theology can supply the necessary theoretical basis for the modern sciences. (Reese (2016))

For Reese, “River Forest Thomism was a unique, North American school of NeoThomism that was speculatively focused on the importance of natural philosophy,

2 See, for more detailed information, Clarke (1952), Hart (1951), and Owens (1963, 1993). 3 Cf., Feser (2009). J. Knasas (2003) considers William Kane, Benedict Ashley, James A. Weisheipl, William A. Wallace and Vincent Smith as representatives of River Forest Thomism. K. F. Klenk (2015) expands this list to include Raymond J. Nogar. 4 Ashley (1991).

162 | Mirosław Szatkowski both for engaging with contemporary science and for establishing the proper basis of metaphysics” (ibid, p. 12).⁵ And Ashley adds: 1.

2.

There can be no valid metaphysics formally distinct from natural science unless its subject, Being as Being (esse), as it analogically includes both material and immaterial being, has first been validated in a manner proper to the foundations integral to natural science by a demonstration of the existence of immaterial being as the cause of material beings. Modern natural science can achieve such a demonstration, but only if its own foundations are rendered unequivocally consistent with sense observation by an analysis such as is exemplified by Aristotle’s Physics as interpreted by Aquinas. (Ashley (2006), p. 53)

Ashley claims that “a return to the natural science of Aquinas as the foundation for a revision of the world-view of modern science opens the way also to a metaphysics which is not merely subjective but open to public dialogue” (Ashley (1991), p. 14). Therefore, as he states, “interpretations of Aquinas which ignore the strongly empirical bent of Aristotle’s thought, or attempt to isolate metaphysics from nature science, fall under grave suspicion” (ibid, p. 3).⁶ The position of metaphysics among other sciences is unique because it compares, coordinates, and unifies the results of the special sciences. Natural science cannot fulfill such a leading role, as the First Cause of the existence and action of ens mobile, which is not itself a physical object, cannot be studied by the principles of this science. For example, physicists operate with several diverse models which are mathematically complete, but are unable to provide a unified physical explanation of the universe. Besides, the first principles of metaphysics and natural philosophy are totally different. “Terms and principles usually labeled “metaphysical” such as substantia, essentia, and esse, as well as the first principles (judgments) com-

5 The three main theses of Fernandez-Alonso’s natural philosophy (Fernandez-Alonso (1935)) are related to River Forest Thomism’s 8 theses. These three theses are: (1) Natural philosophy is distinct, both formally and materially, from metaphysics. (2) natural philosophy is distinct formally, but not materially, from mathematized versions of the natural sciences (like mathematical physics). (3) Natural philosophy is not distinct, either formally or materially, from the (nonmathematized) natural sciences (like biology, chemistry, physics, or medicine). (Reese (2016), p. 8, and see also Weisheipl (1961), pp. xxviii–xxix) 6 J. Knasas adds: “The Aristotelian Thomists claim that the most fundamental understanding of the real things confronting us in sensation is Aristotle’s hylomorphic understanding of radically changeable substance” (Knasas (2003), p. 10).

Part I: The Doctrine of St. Thomas Aquinas and its Non-Analytical Versions |

163

posed from them would be restricted simply to ens mobile and fall under the subject of natural philosophy” (ibid, p. 4). By contrast, the first principles of natural philosophy concern, among other things, the definitions of formal subjects, all the genera, sub-genera, and species of ens mobile. As far as the relation of natural philosophy to experimental science is concerned, Weisheipl writes: In practice, courses in natural philosophy rarely get beyond general considerations, and courses in experimental science rarely get beyond particular considerations and experiments. However, the Lyceum considers that in both the general and particular parts of this unique discipline there are to be found diverse types of certainty: demonstrative, most probable, tentative, hypothetical, factual and even historical. Finally, the Lyceum maintains that the single science of nature is autonomous in its own field, and in the order of learning prior to and independent of metaphysics. There are many advantages to this view. First, it recognizes the dignity of a scientific study of the natural world which includes man, animals, plants and inanimate realities. Second, it recognizes the importance of this science for moral, metaphysical and theological concepts. Third, it offers a real possibility of cooperation between the professional philosopher and the experimental scientist. Fourth, it is consistent with the teaching of St. Thomas and St. Albert, for whom natural science is incomplete unless after studying the general theory found in the Physics, one proceeds to more and more particular species and varieties of living and non-living natures. Fifth, it is consistent with the actual practice of modern scientists, who begin with very particular varieties and gradually ascend to a more embracing unity, usually in old age. (Weisheipl (1961), p. xxxi)

Laval Thomism is almost entirely limited to the work of Charles De Koninck, which was mainly concerned with the philosophy of science, philosophy of religion and metaphysics. Koninck’s philosophy is largely rooted in his interpretation of Aristotle and Thomas Aquinas. For him, Aristotelian and Thomistic philosophy were a source of inspiration and a fundamental model for formulating and solving contemporary philosophical problems. According to L. Armour, Koninck’s whole philosophical career was focused on answering three questions: 1. 2. 3.

How can we understand the growing chasm between our scientific world pictures and the world as it appears to common sense? How can we understand the power of modern science and accept its insights while maintaining our most central and traditional religious beliefs? And how can we maintain the responsibility and dignity of the individual without undermining the communities in which we live and without denying the scientific accounts of human nature? (Armour (2008), p. 1)

More specifically, in De Koninck (1936/2008, 1960a) and in many of his other articles, the central problem he addresses is the relationship of science to our everyday experience of nature. Repeatedly, he takes modern mathematical physics as the paragon of science. For him, the really existing world – cosmos, as a unity

164 | Mirosław Szatkowski of order – is a collection of objects which can be sensed in a variety of ways. But objects available to us in experience are much richer than those described in modern mathematical physics. “We know about the world because we make use of a ‘common sense’, i.e., we have a facility for putting together these elements in a way which assures us of the existence of the ordinary, solid world around us” (Armour (2008), p. 7). And “mathematical physics deals, literally, with abstractions and there is a tendency to take these abstractions for the whole of reality. The result is what De Koninck meant by the expression ‘hollow universe’ ” (ibid, p. 7). What De Koninck understands by “the really existing world” is everything that is revealed by the normal functioning human intellect within its proper context of rationality. And the aim of the philosophy of nature is to investigate if this analysis is correct. It is primarily about reflecting on the nature of experience itself. But it is also about a multi-faceted analysis of the fundamental concepts that are used to grasp the original situation, about the way in which the human intellect works together with the senses – especially, about the role of “common sense”, and about substantiating the thesis that the objects of immediate experience are quite different in kind from the sorts of things which can be designated by scientific symbols (cf., Armour (2008), p. 15). To do this, De Koninck uses the Aristotle’s and Aquinas’ philosophy. For De Koninck, each of the three disciplines: science, philosophy of nature, and philosophy (which he limits to metaphysics), is radically separable from the others.⁷ Yet, he is aware that identifying the formal distinction between science, philosophy of nature and philosophy is very difficult.⁸ Specifically, Kon-

7 See, his numerous studies on this subject, De Koninck (1946, 1947, 1956, 1957, 1959, 1960a,b, 1961, 2008). 8 One of the reasons for this difficulty is the fact that experimental science is not a single research discipline, but a multitude of disciplines. Hence, the formal differences between these fields would have to be established already within the area of experimental science. Identifying different methods among them does not solve the problem. As remarked by De Koninck: Different sciences have different subjects and different methods, no doubt. However, even when the formal subject (taken from the mode of defining) is tha same, one and the same science can still use different methods. Not only each science, but each branch of a single science may stand in need of diverse methods. Physics (in Aristotle’s sense) and psychology are parts of one and tha same sciencs: but they use quite different methods ... the first being mainly based upon external experience, the latter upon the internal experience of beinf alive. And when, in this same science of physics, we apply mathematics to nature, the method again becomes widely different: for we will now have to do with movement qua measurable (In Boethium de Trin., q.5, a. 3, ad. 5). But none of this has anything to do with a distinction between philosophy and science. ... The various departments of knowledge about nature all arise from wonder and aim to dispel ignorance, while on the other hand

Part I: The Doctrine of St. Thomas Aquinas and its Non-Analytical Versions |

165

inck’s attention is directed to the relationship between science and theology.⁹ On this topic, L. Armour writes: Basically, his argument would go like this: science shows us a world of abstractions which must be dealt with by means of symbols, usually numbers. These symbols relate to universal principles. Experience, however, confirms our view that the objects from which these abstractions proceed – and on which they throw such great light as to give us our modern and massive power over nature – are actually unique particulars. These unique particulars do not, per se, have scientific explanations even though they have properties which do have such explanations. The explanation of a genuine irreducible and unique particular must always be a unique action. Such unique actions encompassing a whole universe and rendering it a single unique particular in and of itself must come from God, the only being who would have such powers. (Armour (2008), p. 64)

Koninck claims that the inner being of objects given in experience is determined by their relation to the ultimate reality, which is God. In his view, scientific laws which are exploratory in nature recognize causal relations among things, but they are not sufficient to capture all dimensions of reality. To overcome these difficulties, he uses the notion of ‘universal cause’. This notion is, for instance, a key to solve problems which evolutionary theory poses to Christian philosophers. Koninck strived for the unification of knowledge. The method to accomplish this aim is to use ‘common sense’, i.e., the capacity to understand that some entities can only be recognized after putting together information from various senses (cf., Armour (2008), p. 42).

3.2 Existential Thomism ‘Existential Thomism’ focuses on the primacy of the act of existence in the metaphysics of Thomas Aquinas. Jacques Maritain and Étienne Henry Gilson are among the leading figures of Existential Thomism. In this section we devote our attention to these two thinkers.

their mode of defining is the same, namely cum materia sensibili. (De Koninck (1959), pp. 252–253) In determining the nature of experimental science, De Konicka uses the notions: subalternation and dialectic. “Subalternation provides a way of explaining the use of mathematics to cast light on the natural world; dialectics as a method which does not attain the term of certitude seems applicable to experimental sciences” (McInerny (1965), p. 498). 9 See, for example, De Koninck (1960a).

166 | Mirosław Szatkowski Starting with Maritain, we limit ourselves to a brief characterization of his metaphysics¹⁰, epistemology¹¹, philosophy of nature¹², as well as natural theology and philosophy of religion¹³. The key concept of Maritain’s metaphysics – or of the whole his philosophy and, according to Maritain, also of Thomas Aquinas’ philosophy – is the ‘intuition of being’ (I’intuition de I’ ^etre).¹⁴ It is difficult to explicitly determine what Maritain meant by this concept – he already used the constituent concept ‘being’ (^etre) in two senses.¹⁵ On the one hand, he understood ‘being’ (^etre) in the sense of Thomas Aquinas’ subject of metaphysics – ens. In his words: A philosopher is not a philosopher if he is not a metaphysician. And it is the intuition of being [I’ intuition de l’ ^etre] ... that makes the metaphysician. I mean the intuition of being in its pure and all-pervasive properties, in its typical and primordial intelligible density; the intuition of being secundum quod est ens [I’ intuition de l’ ^etre secundum quod est ens]. (Maritain (1947/1948), p. 19)

Ens, according to Maritain, is an intelligible object which is universally communicable¹⁶, an analogous commonality¹⁷, and tran-sensible¹⁸. On the other hand,

10 J. Maritain did not built a comprehensive metaphisical doctrine, but many of his metaphysical conclusions can be found in: Maritain (1932/1959, 1934/1939, 1935/1940, 1941, 1944, 1947/1948, 1948/1952). 11 See, Maritain (1932/1959, 1939, 1948/1952). 12 See, Maritain (1921/1944, 1932/1959, 1935/1940, 1936/1951). 13 See, Maritain (1921/1944, 1953/1954). 14 Maritain (1947/1948), p. 42: “The intuition of being is not only, like the reality of the world and of things, the absolutely primary foundation of philosophy. It is the absolutely primary principle of philosophy.” See also, Maritain (1934/1939), p. 44: “It is this intuition that makes the metaphysician.” 15 Cf., Knasas (1988), pp. 83–84. 16 Maritain (1932/1959), p. 210: “[It] is not the privilege of one of the classes of things that the Logician calls species, genus, or category. It is universally communicable.” And Knasas (1988), p. 83: “The scholastics called such objects transcendentals.” 17 Knasas (1988), p. 83: “By an analogous commonality, Maritain understands a unum in multis, a one in many, that implicitly but actually contains the differences of the instances of the multa.” Cf., Maritain (1932/1959), pp. 212–213. 18 Maritain writes: Such objects [e.g., ens] are trans-sensible. For, though t:rey are realized in the sensible in which we first grasp them, they are offered to the mind as transcending every genus and every category, and as able to be realized in subjects of a wholly other essence than those in which they are apprehended. It is extremely remarkable that being, the first object attained by our mind in ttlings ... bears within itself the sign that beings of another order than the sensible are thinkable and possible. ( Maritain (1932/1959), p. 214)

Part I: The Doctrine of St. Thomas Aquinas and its Non-Analytical Versions |

167

Maritain used the ‘being’ (^etre) in the sense of what Thomas Aquinas calls esse. He writes: This is why, at the root of metaphysical knowledge, St. Thomas places the intellectual intuition of that mysterious reality disguised under the most commonplace and commonly used word in the language, the word to be [^etre]; a reality revealed to us as the uncircumscribable subject of a science which the gods begrudge us when we release, in the values that appertain to it, the act of existing [cet acte d’ exister] which is exercised by the humblest thing – that victorious thrust by which it triumphs over nothingness. (Maritain (1947/1948), p. 19)

However, there is an interdependence between these two senses. Knasas states: “The second sense is the basis for the first. ...[And] the concept of existence (esse) cannot be cut off from the concept of being (ens, that-which-is, that-which-exists, that whose act is to exist), ... To have the intuition of ens is to have the intuition of esse. The insight into the immateriality of ens is rooted in an insight into the intelligibility of esse.” (Knasas (1988), pp. 84–85) For these reasons, it seems unproblematic to claim that Maritain used the term ‘intuition of being’ in the sense of ‘intuition of ens’. For an explanation of what was going on in the intuition of ens, Maritain refers to the role of judgment in the original Thomistic doctrine. And he says that the function of judgment is “existential, [judgment] transposes the mind from the plane of simple essence of the simple object of thought, to the plane of the thing, of the subject possessing existence” (Maritain (1947/1948), p. 17). The following passage is worth quoting here: But, since existence is of another order than essence, existence is not an intelligible or objectlike essence. It cannot be grasped conceptually, but only in judgment. The concept of existence cannot be visualized apart from being. Being contains within itself the two-fold valence of essence and existence, a notion of “what something is” and the judgment “that it is”. In the intuition of being, the mind surges beyond the grasp of essence to the existence of things. But the existence of the being draws the mind to consider something more than brute facticity or a dark surd. The intuition of being grasps the formality of existence, a “superintelligible” datum for the mind. (Hittinger (1988), pp. 72–73)

Next, Maritain distinguishes, just as Thomas Aquinas did, between the form and the matter of sensible things, and adds that what determines the nature of sensible beings is the form, whereas what reflets their individuality is the matter. For Maritain, the metaphysics is the study of being as being (ens inquantum ens), which means that “[it] investigates the first principles of things and their highest

168 | Mirosław Szatkowski causes”, which are: the principle of identity, the principle of sufficient reason, the principle of efficient causality, and the principle of finality.¹⁹ Turning now to Maritain’s epistemology, his main works in this area are: The Degrees of Knowledge (Maritain (1932/1959)), Science and Wisdom (Maritain (1935/1940)), Four Essays on Spirit (Maritain (1939)), and The Range of Reason (Maritain (1948/1952)). Maritain distinguishes many ways of knowing and, consequently, different kinds of knowledge, and tries to sort them into a hierarchy. According to him, our ‘degrees of knowledge’ sort naturally according to their object, it being the case that “[t]hose objects which are highest in intelligibility, immateriality, and potential to be known are the objects of the highest degree of knowledge” (Sweet (2013)). Distinguishing non-rational knowledge²⁰ (nonconceptual knowledge²¹) from the rational knowledge, he argues that, as far as the latter is concerned, the knowledge of the sensible world (the knowledge of objects of experimental science) is lower than the mathematical knowledge (the knowledge of objects that do not have a direct relation to the actual), and that it, in turn, is lower than metaphysical knowledge. Maritain writes: The metaphysician considers an object of knowing of a specifically higher nature and intelligibility, and from it he acquires a proper knowledge, a scientific knowledge, by means that absolutely transcend those of the physicist or the mathematician. (Maritain (1932/1959), p. 37)

Clearly, for Maritain, the structure and method of the various kinds of knowledge is determined by the nature of objects to be known. Thus, the purpose of knowledge of sensible nature is to discover the laws governing perceived objects. This process involves perinoetical knowledge and dianoetical knowledge, which are not independent of one another. On the level of perinoetical knowledge, those regular patterns existing in nature are discovered with empirio-logical methods (observation, formulating and testing hypotheses), whereas on the level of dianoetical knowledge, laws about essences or natures of perceived objects and essential connections between them are formulated. A feature of mathematical objects (such as quantities, numbers, and so on) is that they cannot exist unless they are material things, but they can be conceived without any reference to material things (cf., Sweet (2013)). In turn, objects of the metaphysical knowledge (such as substances, qualities, and so on) exist independently of matter and they can be conceived only indirectly, by analogy with material things. As to the epis-

19 Maritain (1934/1939), p. 27, and cf., Maritain (1934/1939), p. 90. 20 Maritain (1948/1952), p. 23. 21 Maritain (1948/1952), p. 25.

Part I: The Doctrine of St. Thomas Aquinas and its Non-Analytical Versions |

169

temological question posed by knowledge, mataphysical or otherwise, Maritain makes the non-Kantian claim that it is not a preliminary discipline in relation to metaphysics, but an integral part of metaphysics itself.²² Expressing the view of Maritain in condensed form, G. A. McCool writes: It is no more than the metaphysician’s reflection on the principles of his own science and a reflex justification of the natural metaphysics of the mind manifested to the philosopher of nature in his judgments about sensible reality. (McCool (1992), p. 118)²³

Maritain’s epistemology, which he called critical realism, maintains that what the mind knows and what really exists is identical. More precisely, getting to know reality is a conceptual process, which is recognizable by reflection. In the course of getting to know what reality is, the mind behaves both passively (it receives sensory impressions) and actively (it constructs knowledge on the basis of these sensory impressions). The knower’s intellect affirms directly sensible reality and its own being in the judgment of reflex consciousness. Maritain called the grasp of being at this level the eidetic intuition of being. Then, in the process of abstraction, the knower’s mind finds both units of the primitive concept of being: an essence and an existence. And, as McCool writes: ... Maritain claimed, the same essence that exists outside the mind through its natural act of existence also exists in the act of knowledge through its intentional existence. From the metaphysical point of view, knowledge should be looked upon as a special sort of existence enjoyed by a essence intentionally united to a faculty of knowledge through a species or intentional form. (McCool (1992), p. 122)

Maritain thinks that “What is, is”, i.e., the principle of identity, holds as the norm of every judgment. This principle cannot be grounded on contingent being, but is grounded on the absolute ratio entis, i.e., it is an expresion of the necessary intelligibility of being (cf., Maritain (1932/1959) , p. 77–80). Maritain’s works in the philosophy of nature are: Introduction to Philosophy, vol. 1 (Maritain (1921/1944)), The Degrees of Knowledge (Maritain (1932/1959)), Science and Wisdom (Maritain (1935/1940)), and The Philosophy of Nature (Maritain (1936/1951)). These works deal with the consequences of his metaphysics. In Maritain’s words:

22 See, Maritain (1948/1952), p. 25: “...the critique of knowledge is part of metaphysics.” 23 Cf., Maritain (1932/1959), pp. 21–22 and 83–87.

170 | Mirosław Szatkowski [The philosophy of nature is] a knowledge whose object, present in all things of corporeal nature, is mobile being as such and the ontological principles which account for its mutability. (Maritain (1932/1959), p. 197)

In other words, the philosophy of nature is concerned with sensible, changing things, seeks the first principles that govern them, and refers to experience to justify its own conclusion.²⁴ As an intermediate discipline between science and metaphysics, it differs both from the former, which concerns sensible being qua the observable or measurable, and the latter, which deals with all being qua being. “The philosophy of nature is specifically concerned with the nature of movement, of corporeal substance (i.e., matter and form), of life, and of the constitutive principles of organisms” (Sweet (2013)). Maritain writes: It belongs to the philosophy of nature to instruct us about the nature of the continuum and of number, of quantity, of space, of motion, of time, of corporeal substance, of transitive action, of vegetative and sensitive life, of the soul and its operative powers, etc. (Maritain (1932/1959), p. 186)

What distinguishes the philosophy of nature from metaphysics is the fact that its general principles are formulated on the basis of knowledge derived from the natural sciences. And finally, Maritain’s work in natural theology and the philosophy of religion includes Approaches to God (Maritain (1953/1954)), although these topics were also considered Range of Reason (Maritain (1948/1952)) and Degrees of Knowledge (Maritain (1932/1959)). For Maritain, God – i.e., His existence and attributes – is knowable in different ways, and there is no conflict between faith and rational knowledge. The knowledge of God can be natural (conscious knowledge that is potentially available to every person) or connatural (non-conscious knowledge, belonging only to some people).²⁵ This first of them can, in turn, be spontaneous (pre-scientific) or systemous (scientific).²⁶ The spontaneous knowledge of God is an intuition which cannot be expressed in words. This kind of human experience has two versions: extra-subjective (cosmological) and intra-subjective (anthropological). Two stages can be distinguished in the cosmological version of the spontaneous knowledge of God: (1). primordial intuition of existence, and (2). reflection upon it. In each of these two stages, Maritain distinguishes different elements, namely, in (1): (a). recognition of many things existing beyond me, (b). feeling

24 See, Maritain (1932/1959), pp. 136–137 and 175–178. 25 See, Maritain (1948/1952), pp. 69–71. 26 See, Maritain (1953/1954), pp. 13–24.

Part I: The Doctrine of St. Thomas Aquinas and its Non-Analytical Versions |

171

one’s own fragile existence, (c). realization that the impermanent existence of things beyond me, as well as my own being, indicate the existence of something more permanent; and in (2): (a). vague reasoning concerning my own contingency and dependence on nature, (b). vague reasoning concerning the contingency of the whole universe, (c). vague reasoning searching for the cause of contingent beings (beings with nothingness), and (d). recognition of existence of an Absolute Being (Being without nothingness). Human cognitive activity regarding the spontaneous knowledge of God in the anthropological version is on the one hand grounded on spatiotemporal sensual cognition, and on the other hand subject to laws of the spirit, which reach beyond space and time. This extra-spatiotemporal thought points to the necessity of the existence of an Absolute Spirit (God) as its real source. The so-called “Maritain’s Sixth Way” of proving the existence of God can be explained as follows (as presented by M. A. Krąpiec and Z. J. Zdybicka): 1.

2.

3.

4.

Intellectus supra tempus. Intellect is spiritual in nature. Thought is linked to time only externally, not substantially. The real “location” of a spiritual act is extratemporal existence. Actiones sunt suppositorum. Actions emanate from a subject or person. There is no action more personal than thought. Thought emanates from the subject from body and soul – through the “I”. The “I” exists in time and is born in time. But as the subject of spiritual actions – the centre of spiritual activity, capable of existing by virtue of immaterial existence of the intellect – is timeless like thought itself. This “I” starts in time. But nothing starts absolutely. Everything somehow preexists in its adequate causes. The human is a material-spiritual being. Although “I” comes into being in time, it somehow preexists. And spirit can come only from the Spirit. Hence, the spiritual “I” preexisted in the first being, different from any other being existing in time. At this point there comes a conclusion about the existence of a personal God. (Krąpiec and Zdybicka (1983), p. 11)²⁷

Systemous (scientific) knowledge of God can be theological (acquired through faith and human reasoning) or philosophical (acquired only through faith). Philosophical knowledge of God can be acquired through the analysis of the activity of practical intellect (for example, all striving after a contingent good is an acceptance of the Absolute Good (God), and all creative activity aimed at concrete beauty is an acceptance of the Absolute Beauty (God)) or a purely theoretical way (Aquinas’ five ways for the existence of God are a good example). Finally, the mystical experience of God is a connatural knowledge of God. This experience consists of cognitive and volitional elements. It is characterized by a certain revelatory level of certainty.

27 Translated by K. Jodłowiec.

172 | Mirosław Szatkowski It is sometimes difficult to know if we should classify É. H. Gilson as a historian of philosophy or as a philosopher in his own right. For Gilson himself, the relationship between the history of philosophy and philosophy is two-directional: pursuing history of philosophy is not possible without pursuing one’s own vision of philosophy and vice versa – pursuing one’s own vision of philosophy is not possible without knowing the history of philosophy.²⁸ Both disciplines are auxi-liary to each other. Gilson contributed greatly to our understanding of medieval philosophy, but also to the advancement of philosophy itself – particularly in metaphysics and epistemology (philosophy of knowledge), within the strand of Thomism known as “Existential Thomism”, which is less interested in defending literal adherence to Thomas Aquinas’ teachings than with building a conceptual structure in communication with its spirit. To emphasize his distance from idealistic metaphysics, Gilson named his own metaphysical principles “metaphysical realism”. Here, in our brief precis of Gilson metaphysical views, we rely on his following three works: Le thomisme: Introduction a 1il Philosophie de Saint Thomas d’Aquin, Réalisme thomiste et critique de la connaissance, and L’être et l’essence (Gilson (1922, 1939, 1948)). Gilson analyzes being, which is the subject of metaphysics, in two different, but mutually complementary, ways: in the aspect of its essential structure and of its reality as an act of being (esse)that is created and actualizes “it, what is to be”. The essential analysis consists in grasping being as a substance, it is about answering the question: “Why is being what it is?” Substance is an ontological unit that constitutes a distinct whole and has its internal structure. Analyzing the internal structure of substance, it is possible to discover determinants which are essential for it and complementing the substance in its entirety. The typical determinants of the sub-

28 For example, in the Preface to God and Philosophy (Gilson (1969)), Gilson claims that the history of philosophy is a handmaid to philosophy. At the same time, he does not consider his book Being and Some Philosophers (Gilson (1952), pp. ix–x) as a work in the history of philosophy, but as a presentation of his own philosophy. To quote B. Garcia: Gilson offers a historically-based defense of Thomistic metaphysics in Being and Some Philosophers. Throughout his analysis he provides critiques of alternative metaphysical positions which have occupied places of prominence throughout the history of philosophy; he situates metaphysics of esse within a historical context and puts it into dialogue with rival metaphysical conceptions. The historical context constructed, however, is neither entirely chronological nor exhaustive; he selects major representatives from a given metaphysical tradition to demonstrate the way in which the fact of existence has been repeatedly overlooked. As already stated, Gilson is clear that he is not presenting a historical account of philosophy, but rather a philosophic argument. (Garcia (2010), p. 82)

Part I: The Doctrine of St. Thomas Aquinas and its Non-Analytical Versions |

173

stance determine its essence and make the substance a recognizable whole, which can be given a definition. These determinants constitute a substance. The determinants complementing the entire substance – and hence non-constitutive – are called attributes. For Gilson, substances are always actually existing entities. All elements constituting a substance exist by virtue of the act of existence. This act of existence is an act of a substance due to which every substance is an existing substance. In turn, attributes exist if a substance exists – their esse (existence) is inesse (existence in). Looking at a substance on the one hand as a member of a certain species, and on the other hand as a distinct unit of existence, Gilson distinguishes two real factors in a substance: form and matter. Form is what allows a substance to be classified as a member of a certain species. Matter, on the other hand, functions as an individuator. However, neither the former nor the latter factor can exist independently. This brief outline obviously does not reflect the complexity and depth of Gilson’s essential analysis of being. Yet what is important is that the essential analysis alone is not enough to understand being. In this kind of analysis, the intellect perceives only the essence of what exists, i.e., perceives only what is general, but is incapable of grasping the act of existence, and does not perceive what is specific and individual. Here, the existential analysis of being comes to our aid. In this kind of analysis being is approached from the perspective of existence, which, however, is not being itself. Although the composition of essence and existence is real, it is not part of physical, but only of the metaphysical order, as a relationship between the actual and the possible. Existence is not included in the essence. Its ultimate explanation is the efficient cause, which is external in relation to any specific being. Only God, as a pure act of existence can be an ultimate explanation. The act of existence of being cannot be conceptualized, so how can we understand it? Unlike Maritain, Gilson is not a supporter of the intuition of being. Following Knasas (1990), let us quote Gilson’s criticisms of the intuition of being: What is the existence (l′ ^e tre meaning esse) of the existent (l′ e´ tant). It is not itself a being (un ^e tre). As such the existence of the existent does not exist. It does not have some proper existence apart from that of the substance which it makes an existent. The substance exists only by the existence, but the existence exists only in the substance and as the existence of this existent. This is even why one could not have the intuitive intellection of the existence of an existent (d′ intelledion intuitive de l′ ^e tre d′ un e´ tant), because the existence is perceptible to us only in the sensible perception of the substance which it actualizes. From the act of perceiving such or such an existent, we are able to abstract the abstract notion of existence, this common and universal existence attributable to all that which exists; but the existence proper to each existent is known to us only as a cause imminentto that which it makes exist. The only esse perceptible in itself and as such is God, because “God is esse itself” (C. G. 1,22; 1,33); “the esse of God is his substance” (C.G. II, 52, 7); “the quiddity of God

174 | Mirosław Szatkowski is his being itself” (C. G. I, 25, 5). No existent is such that its quiddity is its existence; it is not then necessary to take the sensible intuition of the existent for an intellectual intuition of its existence. (Gilson (1974), p. 10)

And, Consequently, we apprehend existence only as the existence-of-such-an- existent, which is for us an object of sensible intuition; we never apprehend existence in itself and apart in its proper quality of existence. It is necessary to return to this text: “it is not properly said that esse exists but that through esse something exists” (De div. nom. Pera, 751). One has the intuition of things that exist in virtue of their esse, one could not have an intuition of an act of existence which itself does not exist. One is able to distinguish as many degrees of abstraction as one wishes; nothing will make our apprehension of existence not be an abstraction of the intellect taken from the sensible ... . We see the actual existence only in the effect in which it manifests itself, which is the existent sensibly perceived and intellectually known. If the existence were perceptible in itself, as it is in the case of God and only thus, it would indeed be an object of intellectual intuition. This is not a question of degrees of abstraction if it is not that. The very nature of the human intellect is the cause: the human intellect “does not think without an image”, and since there is not some image of existence insofar as existence, which is a pure intelligible, the intellectual intuition is refused here below to minds that are most skilled in metaphysical meditation. (ibid, p. 11)²⁹

The way of recognizing the esse – which Gilson prefers – begins with the formulation of simple existential judgments. One can say that the existential judgment is an operation in which the real existence of a particular subject is affirmed or denied. For example, the existential judgment “Hillary Clinton is” means that Hillary Clinton really exists. The statement “Hillary Clinton is” may occur in a dual role. In the statement “Hillary Clinton is the Democratic Party’s candidate for US president in the 2016 election”, the word ‘is’ says nothing about the real existence of Hillary Clinton, but it informs us about the identity of Hillary Clinton, because she is the only Democratic Party’s candidate for US president in the 2016 election. On the other hand, the short statement “Hillary Clinton is”, which is a sentence with no predicate, stresses the existence of Hillary Clinton. To quote Gilson: An existential judgment is an act affirming an act; an act of thought that affirms an act of existence. What makes this act of thought judgment per se, although it contains no predicative expression about the subject, is the fact that it remains an act consisting of a concept

29 An in-depth analysis of the Gilson’s criticism of the intuition of being can be found in Knasas (1990). In this article Knasas states that “Gilson admits that there is an intuition of being if that means the judgmental grasp of the esse of things.” What Gilson criticizes is only one of the three senses in which Maritain’s intuition of being can be understood.

Part I: The Doctrine of St. Thomas Aquinas and its Non-Analytical Versions |

175

and something else. An existential judgment confirms a subject and its act of existence; it connects the subject with its existence in thought in the same way as they are connected in reality. (Gilson (1963), pp. 249–250)³⁰

Existential judgments point to the difference between the essence and actual existence, but this actual existence is something other than esse. Only God really has esse, which means that the distinction between the essence and esse in things other than God cannot be demonstrated. Although “Gilson sees Thomas’s esse as a revolutionary factor in the history of metaphysics” (Dewan (1999), p. 71), he has a different conception of esse. L. Dewan explains this difference between Thomas’ and Gilson’s notion of esse as follows: This brings us already face to face with a problem. Thomas tells us that, while God’s essence and his esse are identical, they are both beyond our minds. Thus, it does not seem possible to find the properly Thomistic meaning of “esse” through a consideration of divine esse. Rather, the very texts which Gilson is following show that we are supposed to know about esse first, and through it we reach some conclusions about God. For example, if we consider the arguments used by Thomas in SCG 1.22, we see how sure he is that we understand what is meant by “esse” in things. Consider the following: ... Each thing is [est] through its esse. Therefore, what is not its own esse is not “through itself necessary being” [per se necesse esse]. But God is “through itself necessary being.” Therefore, God is his own esse.³¹ We are certainly supposed to be able to grasp the proposition: “each thing is through its own esse.” We are even supposed to be able to entertain the idea of something not being its own esse. Obviously, for Thomas, esse is already known to us, as found in material beings. Gilson’s attempt to equate Thomas’s doctrine concerning God with the view that God has no essence reveals a conception of essence which is other than that of St. Thomas. And the very fact that Gilson has a different conception of essence entails his having a different conception of esse, the act of the essence. (Dewan (1999), pp. 71–72)

To further reflecting what Gilson understands by esse, which for Gilson himself is tantamount to the question: “What is God?”, we can turn to other passages from his writings: [Thomas’s] own formulation of the conclusion is significant. Before Thomas Aquinas, Avicenna had bluntly said that the First has no quiddity: Primus igitur non habet quidditatem. Thomas himself seems to have avoided this uncompromising language. Not that he had any objection to the truth of what it says, for if essence is understood as something in any way different from God’s act of being, then it must be conceded that God has no essence. But Thomas Aquinas does not want us, or himself, to lose contact with the quiddity of sensible

30 Translated by K. Jodłowiec 31 SCG 1.22 (ed. Pera, 206; Pegis, 5).

176 | Mirosław Szatkowski things, our necessary starting point for investigating the nature of God. To know something is for us to know what it is. If God has no essence, He has no “whatness”, so that to the question: What is God? the correct answer should be, nothing. Many mystics have not hesitated to say so, in the definite sense that God is no-thing, but they certainly were not doubting God’s existence. To say that God has no essence would be to render Him completely unthinkable. ... More important still, it would be to betray the true meaning of the negative method in theology. A negation necessarily requires an affirmation; namely, the very affirmation it denies. To say that God has no essence really means that God is a beyond-essence. This is best expressed by saying that God is the being whose essence is to be beyond essence or, in other words, God is the being whose essence it is to be. (Gilson (1960), pp. 133–134, and cf., Dewan (1999), pp. 75–76)

And he continues, When we reach the question, what is God? the time has come for our intellect to cast off its moorings and to set sail on the infinite ocean of pure esse, or act, whereby that which is actually is. ... What is the very last thing a concrete substance would have to give up in order to achieve utter simplicity? Its essence, of course. In our attempt to describe God by removing from Him what is proper to the being of creatures, we must give up essence in order to reach the open sea of pure actual existence, but we must also keep the notion of essence present to the mind so as not to leave it without any object. This we do when, to the question, where do we find God? we simply answer, beyond essence. By establishing himself in the definite negation of posited essence, the theologian realizes that he is placing God above that which is deepest in the only kind of reality he knows. At that moment, the theologian is not beyond being; on the contrary, he is, beyond essence, at the very core of being. (Gilson (1960), p. 134, and cf., Dewan (1999), pp. 77–78) This is to point out, in a negative way, an object of thought more positive than all the definable ones. Were we to say that God is this, be it essence, our proposition would entail the consequence that God is not that. On the contrary, in saying that God is neither this nor that, we implicitly affirm that there is nothing that, in His own transcendent way, God is not. To affirm that God is only being is to deny of Him all that which, because it is a determination of being, is a negation of it. (Gilson (1960), p. 135, and cf., Dewan (1999), p. 79)

An extended discussion of Gilson’s thought on esse and, more generally, on metaphysics is beyond the scope of this paper. Instead, let’s conclude our portrait of Gilson’s Thomism with a discussion of Gilson’s epistemology. We must first note that the central problem, here, is that of the objectivity of our knowledge of the actually existing world. Gilson does not put ontology before epistemology and vice versa, but rather, he views them as parallel to each other in their theoretical development, intertwined and complementary as objects of study. Gilson rejects not only idealism (Gilson (1936)), but also the critical realism originated by the members of the Louvain School: D. Mercier and L. Noël (Gilson (1936, 1939)). Idealism, according to Gilson, is at variance with experience, destructive and epistemologically futile. It is based on two unobvious and unfounded principles: of

Part I: The Doctrine of St. Thomas Aquinas and its Non-Analytical Versions |

177

immanence and of precedence. To be exact, it is worth stressing that Gilson uses two statements of the immanence principle: 1. Everything we get to know is part of the subject (consciousness) and 2. Everything we get to know depends on the subject (consciousness). The precedence principle gives precedence to thought over being in the epistemological order. For Gilson, there is a significant difference between a “a being given in thought” and “a being given by thought”, one which is not recognized by idealists. As far as Gilson’s critique of critical realism is concerned, his critique of immediate realism, represented by D. Mercier, should be distinguished from his critique of direct realism, represented by L. Noël. The immediate realism – rejecting the immanence principle – claims that the cognizing subject becomes convinced of the existence of a real world independent of him or her only through discourse, on the basis of the analysis of consciousness (cogito). For Mercier, sensory experience alone is not sufficient to recognize the existence of the world independent of the cognizing subject. Only a certain reasoning that refers to the metaphysical principle of efficient causality can give certainty to empirical facts. Mercier’s argument is based on three premises: 1) 2) 3)

If I am passive in receiving sensations, I am not the cause of the received sensations; I am the cause of the received sensations, or a transcendent real object is the cause of the received sensations; I am passive in receiving sensations. (cf., Stępień (1962), p. 139)³²

According to Gilson, premise 3) is of a purely empirical nature and does not provoke any objections, whereas premises 1) and 2), which refer to relations between facts, require a metaphysical principle of efficient causality. Gilson attacks Mercier’s Realism in two ways: directly, when he claims that the thesis about the world being real does not follow from the assumed premises, and indirectly, when he attacks the very formulation of the so-called “bridge problem”, which Mercier sought to solve. In the first case, Gilson claims that Mercier mixes up the intensional and the extensional. There is no passage from the former, i.e., the order of thought, to the latter, i.e., the order of being – the metaphysical principle of efficient causality is of no avail in this context. In the second case, the so-called “bridge problem” does not exist, because in our cognition there is no such division between, on the one hand, consciousness (or the subject), and on the other hand, the object. Direct realism, which also rejects the immanence principle, claims that the cognizing subject grasps the surrounding reality, which is independent of him/her directly, without any intermediary. For Noël, the direct data of consciousness, accepted as given, are the starting point for exploring this

32 Translated by K. Jodłowiec.

178 | Mirosław Szatkowski reality: what is cognized is real. The subjective aspect of cognition plays the crucial role here. In other words, the starting point for learning the existence of the real world is the “open cogito”, which means that by analyzing what is cognized we find the transcendental. According to Gilson, the “open cogito” can be either a thought or a direct representation of things. In the first case, it does not differ from the Cartesian “closed cogito” and does not lead to accepting the existence of the outside world, whereas in the second case, it makes the petitio principii error in already assuming what is still to be proved. Gilson not only criticizes Mercier’s and Noël’s views, but also formulates his own stance, trying to answer the question about how we make cognitive contact with external things. Over time, his views evolved, but all theses he advanced at that time were named “methodical realism” by him. At this point, P. Chojnacki’s characterization of Gilson’s views is worth quoting: At first, Gilson claimed that ... the cognizing subject assimilates the substitute form of subject, instead of the natural form. It would be absurd to think that natural forms would move to the cognizing subject. At that time (circa 1927 - 1932) Gilson maintained that the existence of things should be accepted as the condition for the creation of concepts, since it is the simplest hypothesis if one wants to explain the objectivity of concepts and the truthfulness of judgments. In the later period, Gilson changed his views inasmuch as he considered the existence of the outside world obvious, meaning the specific obviousness of sensory perception, which is abstractly expressed in an observational judgment. Since, according to Thomists, the intellect cognizes what is general, whereas senses cognize what is individual and specific, and what is real is specific, there is no other option than to assume that sensory cognition identifies what is real. Before the adequation of judgment with what is real occurs, there is a pre-reflexive adaptation of cognition to reality, that is experienced adaptation. Only then the intellect can understand the nature of this adaptation. Therefore, the entire realistic metaphysics of cognition is subsequent to the primordial sensory obviousness, to the things – in a sense – getting to us through sensations. Hence, at the second stage of his understanding of Thomistic realism, Gilson thinks that it is mainly sensory perception that convinces us of the existence of things, since sensations – in terms of content – are different from images, from illusionary perceptions. No reasoning can be more reliable than this perceptual certainty. Yet, Gilson gradually moves away from that view, under the influence of the ideas of the proponents of actual intellectual intuition. The departure starts from Gilson’s realization that the concept of the real belongs to metaphysical concepts and refers to what is conceived of as important as well as the existence which determines actual reality, in contrast to possible reality. At this point, he faced new difficulties. It is impossible to identify existence by means of sensations. On the other hand, intellectual cognition is directed towards “quidditas”, as its proper object. Hence, the specific, real existence lies beyond its scope. Gilson solved this difficulty with Aristotelian-Thomistic hylomorphism, according to which neither senses nor the intellect cognize separately, but a human being cognizes with senses

Part I: The Doctrine of St. Thomas Aquinas and its Non-Analytical Versions |

179

and the intellect. And this is not a conceptual, abstract human, but an actually existing one. This stage of development of Gilson’s thought is dated from circa 1939 to circa 1942. The fourth stage does not reject the views of the third one, but rather complements it. Gilson states that only judgments reach the reality. Only a judgment through the essence of beings reachable via concepts is capable of establishing existence, about which we know that it is the source of entire reality. Judgment, in turn, is created by the reflection of the intellect on itself, through which it discovers that it can legitimately establish the existence of things, because it refers to the initial moment of cognition, during which it perceives an actual, existing thing with the support of sensory perception. (Chojnacki (1969), pp. 200–202)

A. Stępień lists the differences between methodical realism and critical realism as follows: 1)

2)

3)

Methodical realism has its roots in being and from its perspective analyses and evaluates cognition. Critical realism has its roots in thought and seeks conditions for the cognizability of things in thought. Methodical realism has its roots in establishing a certain fact, whereas critical realism has its roots in an a priori determination of conditions that must be met by the cognition of reality, thus giving priority to possibility over fact. Methodical realism justifies its stance negatively, by demonstrating the incorrectness of the idealistic stance, and positively, by showing the obviousness and fruitfulness of its basic principles. Critical realism is justified through criticism. By analyzing thoughts it aims to reach existing things. (cf., Stępień (1962), pp. 150–151)³³

3.3 Transcendental Thomism Transcendental Thomism is a variety of Thomism which seeks to reconcile Kant’s transcendental method with Thomas Aquinas’ doctrine. Therefore, we will start with a brief explanation of Kant’s method, and then discuss the ways certain Thomists have either absorbed or rejected Kant’s insights.

3.3.1 Kant’s transcendental method Kant introduced the term ‘transcendental’ in the first (A version) and then modified its meaning in the second (B version) edition, of the Critique of Pure Reason (Kant (1929)). “I call transcendental all knowledge which is concerned, not so much with objects, as with our concepts a priori about objects in general." (A 11-12), and “I call transcendental all knowledge which is concerned, not so much

33 Translated by K. Jodłowiec.

180 | Mirosław Szatkowski with objects, as with our mode of knowledge of objects in general, in so far as this [knowledge] is to be possible a priori.” (B 25). The term ‘transcendental’ is further explained in the definition of ‘transcendental argument’, viz. ‘transcendental deduction’: “The explanation of the manner in which concepts can thus relate a priori to objects I entitle their transcendental deduction” ( A 85/B 117). In A 87/B 117 Kant states that a transcendental deduction is an explanation and a justification of certain kinds of a priori knowledge. Also, a transcendental deduction deals with the activities through which we put things in objects (in his words): "adopting as our new method of thought ... that we can know a priori of things what we put into them” (B xviii)) and thus produce the objects of our knowlwdge (in his words: “... reason that insight only into that which it produces after a plan of its own” (B xiii)). This new method of thought is based on this insight, which reason “must adopt as its guide ... that which it has itself put into nature” (B xiv). And the ‘mode of knowledge’ is thought of as an activity on the rational human being’s part through which one creates or produces a priori knowledge. Thus, as J. Hintikka puts it, “... in transcendental knowledge we are dealing not so much with our knowledge as a static system or with our own concepts thought of as inert tools, as with our mode of knowledge acquisition.” (Hintikka (1984), p. 100). Kant contrasts transcendental knowledge to empirical knowledge. He originally thought that transcendental knowledge is purely analytic. Yet, he later corrected this view: “We are not here concerned with analytic propositions, which can be produced by mere analysis of concepts” (A 718/B 746), and “A transcendental proposition is therefore synthetic knowledge though reason, in accordance with mere concepts” (A 722/B 750). That means that transcendental propositions are synthetic a priori. Therefore, the “general problem of transcendental philosophy”, as Kant notes, is to answer the question “how are synthetic a priori judgements possible?” (B 73). The question can be posed differently: How can one and the same knowledge be about objects and about our mode of knowledge? Unfortunately, Kant’s explanations of this issue are very obscure and sometimes inconsistent, which makes multiple interpretations possible. From the extensive literature on that subject, we will here focus on two works: of S. Stapleford and of J. Hintikka, which can be considered representative of two different interpretations. S. Stapleford points to passages in the Critique of Pure Reason that stress the difference between the transcendental (philosophical ) and the mathematical method – the former constructs its concepts, whereas the latter derives all of its principles from concepts. “Philosophical knowledge is the knowledge gained by reason from concepts, mathematical knowledge is the knowledge gained by reason from the construction of concepts” (A 713/B 741). Alternatively, philosophy is “the discursive employment of reason in accordance with concepts”, and mathematics “its

Part I: The Doctrine of St. Thomas Aquinas and its Non-Analytical Versions |

181

intuitive employment by means of the construction of concepts” (A 719/B 747), where “to construct a concept means to exhibit a priori the intuition corresponding to it” (A 713/B 741). Further, “mathematics can achive nothing by concepts alone but hastens at once to intuition, in which it considers the concept in concreto.” (A 714/B 743). This intuition must be non-empirical, because otherwise the propositions derived from its concepts would lack the universality and necessity characteristic of mathematical truth (cf., A 718/ B 746). Concepts that can be exhibited in a priori intuition are given originally (cf., A 729-30/B 757-8). A synthesis, as Kant writes, requires intuition: “If we are to judge synthetically in regard to a concept, we must go beyond this concept and appeal to the intuition in which it is given” (A 721/B 749). S. Stapleford writes: Now mathematics has no difficulty in getting on synthetically, for, as we have just seen, its concepts are constructed in pure intuitions (space and time), which impose their own (necessary) structure on the objects thus constructed, thereby giving them something that belongs to them synthetically and yet necessarily. This means that there is a vast supply of intuitive content (in the imagination) from which synthetic principles are easily derived. (Stapleford (2008), p. 38)

Philosophy, in contrast, seems to possess no intuitive resources: “While philosophical knowledge must do without this advantage [of intuition], inasmuch as it has always to consider the universal in abstracto durch Begriffe (by means of concepts)” (A 734/B 762). This text can be supplemented by others: “Philosophy confines itself to universal concepts” (A 715/B 743); “... philosophy is simply what reason knows by means of concepts ...” (A 732/B 760); and “the Axiom of Intuition ... is itself no more than a principle derived from concepts” (A 733/B 762). In addition: “By mere concepts we can produce no synthetic a priori propositions” (Kant (2002): 416—20:340); “Thus no one can acquire insight into the proposition that everything which happens has its cause, merely from the concepts involved” (A 737/B 765); “I cannot obtain knowledge of such a principle directly and immediately from the concepts alone” (A 733/B 761); and “No one, therefore, has ever yet succeeded in proving a synthetic proposition merely from pure concepts of the understanding – as, for instance, that everything which exists contingently has a cause” (B 289). S. Stapleford argues that two ideas can prove helpful in solving this incompatibility: the idea of the possibility of experience and the recognition that categories are subject to special conditions of instantiation. And he starts the clarification of this first idea from demonstrating the contrast that Kant makes between the genuine transcendental proofs of critical philosophy and ‘dogmatic’ proofs of traditional philosophy, or of dogma, which is defined by Kant as follows: “A synthetic proposition directly derived from concepts is a dogma” (A 736/B 764), and ‘a dogmatic proof of the Analogies’ is characterized as follows

182 | Mirosław Szatkowski Had we attempted to prove these analogies dogmatically; had we, that is to say, attempted to show from concepts that everything which exists is to be met with only in that which is permanent, that every event presupposes something in the preceding state upon whih it follows in conformity with a rule ... all our labour would have been wasted. For through mere concepts of these things, analyse them as we may, we can never advance from one objest and its existence to the existence of another or to its mode of existence. (A 216—217/B 263–264)

In sum, the conclusions of both dogmatic proofs and of transcendental arguments are derived from concepts, but the first are derived directly from them, while the second are not. As far as the ‘criterion of the possibility of transcendental proofs’ is concerned, Kant says: “This criterion consists in the requirement that proof should not proceed directly to the desired predicate but only by means of a principle that will demonstrate the possibility of extending our given concept in an a priori manner.” (A 785/B 813). Stapleford adds that “there has to be some mediating link between the concepts involved in a philosophical proof, and ... it couldn’t be a merely conceptual link, for that would not be synthetic” (Stapleford (2008), p. 42); and he quotes Kant: “Now one concept cannot be combined with anoher synthetically and also at the same time immediately, since, to be able to pass beyond either concept, a third something is required to mediate our knowledge” (A 732/B 760). Further, according to Stapleford, That leaves intuition as the only available medium for connecting concepts in philosophical proofs. Kant thinks there are two kinds of intuitions, namely, a priori and empirical. But since philosophical concepts like ‘substance’ and ‘cause’ are not constructed in a priori intuitions, their objects are not given a priori either. And if their objects are not given a priori, then there is no question of simply reading off their properties a priori. What I mean is this. If the concept ‘event’ could be constructed a priori, that is, if we could produce an actual event in the imagination independently of all experience, then we might simply be able to see that events are never non-causal sequences in the way we can see that triangles never have more than one 90-degree angle – this would be revealed to us in the act of constructing the object. But things are a little more complicated with philosophical concepts: ‘I cannot represent in intuition the concept of a cause in general except in an example supplied by experience; and similarly with other concepts’ (A 715/B 743). The categories may be a priori, but their objects, and thus their sensible contents, are given a posteriori. (Stapleford (2008), p. 42)

What binds concepts in philosophical proofs is a possible experience, but this relationship is indirect. There is an ambiguity in Kant about what possible experience is and how concepts get connected indirectly through possible experience. And Stapleford, seeking clarification on these issues, first asks: How do a priori concepts relate to intuition? Here Kant’s text proves helpful: All our knowledge relates, finally, to possible intuitions, for it is through them alone that an object is given. Now an a priori concept, that is, a concept which is not empirical, either

Part I: The Doctrine of St. Thomas Aquinas and its Non-Analytical Versions |

183

already includes in itself a pure intuition (and if so, it can be constructed), or it includes nothing but the synthesis of possible intuitions which are not given a priori. In this latter case we can indeed make use of it in forming synthetic a priori judgements, but only discursively in accordance with concepts, never intuitively through the construction of the concept. (A 719-720/B 747-748)

An a priori knowledge from concepts, which is referred to in the above passage, can be of two types: the first relies on a concept which ‘includes in itself a pure intuition’, and the second begins from a concept that ‘includes nothing but the synthesis of possible intuitions which are not given a priori’. It is this second type that, according to Stapleford, is an answer to his question about how the a priori concepts relate to intuition. And he adds: Philosophical concepts and their objects do not coincide a priori, for the intuitions that satisfy them are procured a posteriori through sensation. They relate to intuition only as rules for connecting that which is given independently in empirical intuition – this is what Kant means by calling them a ‘synthesis of possible intuitions’. (Stapleford (2008), pp. 47–48)

Then Stapleford argues that the ‘possibility of experience’ – which is needed to connect the concepts in a philosophical proof – is “a sensible image of an object, an image of something that could be given in sense perception” (ibid, p. 52). It is about the possibility of experiencing an object corresponding to some concept. It is worth stressing that: The presentation of a concept in empirical intuition is just the presentation of it. The presentation of a concept in a priori intuition is the construction of it. And the presentation of a concept in a possible intuition, I suggest, is the reproduction of its object in a sensible image whose content was originally acquired in experience. (ibid, p. 57)

The basic difference between the mathematical and philosophical method is that the first creates its own objects, whereas the second does nothing more than bring appearances under concepts, according to their actual content, which is acquired epistemically (cf., A 723/B 751 and Stapleford (2008), p. 54). In contrast to Stapleford, J. Hintikka does not differentiate between the transcendental (philosophical) and the mathematical method. In his opinion “Kant’s theory of mathematics offers an excellent example of the applications of his [Kant’s] transcendental method” (Hintikka (1984), p. 99). In his opinion “Kant’s theory of mathematics offers an excellent example of the applications of his [Kant’s] transcendental method” (Hintikka (1984), p. 99). Hintikka denies that Kant tried to explain how the synthetic a priori knowledge in mathematics is obtained by means of intuition. According to him:

184 | Mirosław Szatkowski Kant’s real problem is to explain why we can obtain synthetic knowledge a priori in mathematics, not by means of “intuition”, but by means of what twentieth-century logicians would call instantiations, that is, by means of considering “arbitrarily chosen” representatives of general concepts in mathematical arguments. ... Kant’s problem is not how instantiations are used in logical or mathematycal arguments. He is accepting the traditional concepts of a mathematical (for us: logical) argumentation in which appeals to geometrical “intuition” in our sense play no role. Kant’s problem is: how can such argumentation, prominently including apparent anticipations of absent particulars in instantations, yield knowledge which is applicable to all experience a priori. (Hintikka (1984), p. 101)

In support of his view, he cites Kant: But with this step the difficulty seems rather to grow than to decrease; For now the question runs: How is it possible to intuit anything a priori? Intuition is a representation, such as would depend directly on the presence of the object. Hence it seems impossible to intuit anything a priori originally, because the intuition would then have to take place without any object being present, either previously or now, to which it could refer .... (Kant (1783), Part I, Section 8; and cf., Hintikka (1984), p. 102)

Hence, an important question raised in this passage, is: How can intuitions yield knowledge a priori even when they are used in the absence of their objects? Here, according to Hintikka, Kant’s transcendental method comes into play. He says: Since “reason has insight only into that which it produces after a plan of its own”, the explanation of the universal applicability of knowledge obtained by using instantiation methods, i. e., by anticipating certain properties and relations of particulars, can only lie in the fact that we have ourselves put those properties and relations into objects in the processes through which we come to know individuals (particulars) (Hintikka (1984), p. 102).

Kant himself explains that the properties and relations are put into objects in sense-perception. His transcendental assumption is: If our intuition had to be of such a nature that it represented things as they are in themselves, no intuition a priori could ever take place and intuition would be empirical every time. ... There is thus only one way in which it is possible for my intuition to precede the reality of the object and take place as knowledge a priori, namely, if it contains nothing else than the form of sensibility which in me precedes all real impressions through which I am affected by objects. (Kant (1783), Part I, Section 9; and cf., Hintikka (1984), p. 102)

Importantly, Hintikka rejects Kant’s assumption that the particular objects to which mathematics applies are always given to us by sense-perception. In Hintikka’s opinion: The most general description of the ways in which we do reach the information which we actually have about individuals (especially their existence) is not passive perception, but ac-

Part I: The Doctrine of St. Thomas Aquinas and its Non-Analytical Versions |

185

tive seeking and finding. ... Indeed, we can speak of seeking and finding even in areas where sense-perception is not involved at all, e.g., in dealing with numbers and other abstract entities. (Hintikka (1984), p. 103)

If the fault lies in Kant’s too emphatic notion of the passivity of the senses, Hintikka’s solution is to go back to the subject as the locus of action – or, as he puts it, seeking and finding. Obviously, the ramifications of such a change for Kant’s philosophy must be meticulously studied. But our interest here is in the intersection, in the modern era, of a Kantian influenced philosophy and Thomism.

3.3.2 Thomistic transcendental method Transcendental Thomism is assiciated with such names as, for example, Joseph Maréchal, Emerich Coreth, Karl Rahner, Joseph Donceel and Bernard Lonergan.³⁴ Each of these thinkers brought his individual contribution to the understanding of Kant’s transcendental method and its application to St.Thomas’ doctrine. Below, we would like to outline these different individual projections. The starting point for the application of Kant’s transcendental method to Thomism is J. Maréchal’s major five-volume work Le Point de depart de la metaphysique (Maréchal (1927-49)). In this comparative study of St. Thomas and Kant, Maréchal argues that St. Thomas’ realism and Kant’s idealism can be compatible with one another. The methodology of the metaphysics of St. Thomas need not be changed, but instead, it can coexist with Kant’s transcendental method, if we modify the latter in such a way that, starting with it, one could arrive at the metaphysics of St. Thomas (cf., Maréchal (1927-49), vol. V, pp. 66–71, and Henle (1981), p. 93). For Maréchal, metaphysics is the science of possible essence whose judgments are an absolute universal and apodictic necessity, in the sense of Kant and the great Post-Kantian systems.³⁵ Its starting point is a particular and contingent subject, for which it seeks consolidation. Like St. Thomas Aquinas, Maréchal thinks that only an Infinite Being (God) can be an ultimate guarantor of finite objects – Kant failed to observe that. Maréchal’s main aim, according to C. A. McCool, “was the epistemological grounding of the realistic metaphysics of

34 Among the leading figures of Transcendental Thomism, O. Muck counts: J. Marechal, J. de Vries, A. Marc, G. Isaye, E. Coreth, K. Rahner, B. J. F. Lonergan, A. Gregoire, and J. B. Lotz (Muck (1964), pp. 12–13). V. B. Brezik also adds: J. Donceel and G. McCool (Henle (1981), p. 110), whereas others consider as the main representatives of this intellectual movement: J. de Finance, H. de Lubac, B. O’Brien and R. De Smet. 35 McCool (1992), p. 90, and cf., A Maréchal Reader, pp. 235—236.

186 | Mirosław Szatkowski being (McCool (1994), p. 117). When it comes to the question whether and how it is possible to achieve the absolute universality and apodictic necessity of metaphysical judgments, Maréchal responds that the metaphysics of knowledge which he extracted from Thomas Aquinas’ writings meets this standard. He calls this metaphysics “as the definitive solution to the metaphysical problems of the ancients as well as of the earlier medieval thinkers” (Henle (1981), p. 96, and cf., Maréchal (1927-49), vol. I, p. 121). As Maréchal observed, Thomas Aquinas introduces the operation of the Aristotelian active intellect in place of the Augustinian intuition of divine ideas as the ground of the universality and necessity of knowledge (cf., McCool (1994), p. 121, and Maréchal (1927-49), vol. 5, pp. 276–279). Consequently, and this is also in the spirit of Kant, ... no real object could be known until the abstract content of the mind’s universal concepts has been reunited through the cooperation of the imagination with the concrete singulars of the spatiotemporal world. For both philosophers the whole representative content of man’s conceptual knowledge was restricted to the world of space and time. (McCool (1994), pp. 125–126, and Maréchal (1927-49), vol. 5, pp. 298–304)

Moreover, In essence, its [the Maréchal’s fifth volume of Le Point de depart de la metaphysique] aim was to find the answer to two questions: What was there in Thomas’s metaphysics of knowledge which justified the Angelic Doctor’s claim that the discursive mind could legitimately place the objects of its judgments in reality under the universal law of being, even though, as Thomas himself admitted, it could have no intellectual intuition of any essence, even of its own? Why was it that, although both Thomas and Kant claimed that the content of the discursive reason’s conceptual objects must be unified by the human mind from the disparate data of sense intuition, Thomas’s discursive intellect could unify these objects under the allembracing, transcendental and analogous unity of being whereas their unification by Kant’s discursive reason was limited to the univocal and categorical unity of his phenomenal world of space and time? (McCool (1994), pp. 122–123)

Maréchal’s dialogue with Kant, and consequently – his correction of Kant, according to McCool, led to two important breaks with the other schools of NeoThomism: Maréchal’s metaphysical critique of the object justified the mind’s hold on extra-mental reality through the dynamic reference of the mind’s conceptual objects to God’s necessary being which took place in the metaphisical affirmation of the judgment. In his transcendental critique of the object, Maréchal had given his approval to Kant’s subjective starting point for philosophy. A Thomistic critique of knowledge, he argued, could begin with a “precisive” approach to realism and idealism. It could then work its way from the world of phenomena into the world of real being through its consistent use of Kant’s transcendental method. None of the other Neo-Thomists would sanction the validity of this

Part I: The Doctrine of St. Thomas Aquinas and its Non-Analytical Versions |

187

approach to Thomistic epistemology.³⁶ Thus Thomists in the tradition of Maréchal and Rousselot were distinguished from other Neo-Thomists by the significant role which they assigned to the dynamism of the mind in grounding the validity of metaphysics. For Maréchal, as well as for Rousselot, the intellect was “the sense of the real” precisely because it was “the sense of the divine”.³⁷ McCool (1994), pp. 132–133)

Another application of the transcendental method to Thomism was E. Coreth’s excellent textbook Metaphysics (Coreth (1961)). Coreth clams that for any scholarly discipline there is only one appropriate method.³⁸ There is bidirectional interdependence: on the one hand, the nature of the investigation in a given scholarly discipline determines the methodology, and on the other hand, the methodology determines the nature of the discipline, and especially the nature of the investigation conducted within the discipline.³⁹ R. J. Henle, commenting on Coreth’s view, adds: The methodological principles as well as the starting points of any discipline determine its charakter and its conclusions, and its character and its conclusuions must be understood and interpreted in terms of the argumentation, both with regard to argumentation and the methodology itself. (Henle (1981), pp. 93–94)

Consequently, according to Coreth, the only legitimate method for metaphysics is the transcendental method. Furthermore, the metaphysics established by the transcendental method is substantially different from the metaphysical realism of St. Thomas Aquinas. Hence the accusation that Coreth’s metaphysics“is not a Thomism but a kind of Christianized transcendentalism, or an idealism corrected by Christian realism” (Henle (1981), p. 94). The transcendental method, according to Coreth, has its roots in Plato and Aristotle, and was practiced not only by Kant, but also by Fichte, Schelling, Hegel, Husserl and Heidegger. Coreth stresses the great contribution of Kant and the other thinkers in the development of the transcendental method, but he also identifies some defects in how they implemented this method. The difference between Coreth’s and Kant’s approach to the transcendental method has two main aspects: (i). the evaluation of this method’s application, and (ii). the evaluation of the course of reflection that accompanies

36 For an excellent account of Maréchal’s epistemology see Van Riet, L’Epistémologie Thomiste, pp. 263–300. See also, the more recent article by Johannes B. Lotz, S. J., “Joseph Maréchal” in E. Coreth, Christliche Philosophie im katholischen Denken, v. 2, pp. 453–69. In addition to its expository value Lotz’s article contains a very useful and well chosen bibliogaphy. 37 L’Intellectualisme de Saint Thomas by Par Pierre Rousselot, S. J.. 38 See, Coreth (1961), pp. 55 and 94. 39 See, Coreth (1961), p. 94.

188 | Mirosław Szatkowski it (cf., Herbut (1991/1992), pp. 123 and 125 – 127). As far as (i) is concerned, Kant studied only the scientific experience articulated in the laws of mathematics and physics of his age, whereas Coreth thinks that scholarly knowledge cannot be explained without reflection on pre-scientific knowledge. The main propositions of metaphysics, e.g., the law of noncontradiction and the principle of sufficient reason, can only be reached by means of transcendental reflection on prescientific cognitive acts, expressed in non-necessary judgements about things. For Coreth, in order to reach these two principles it is necessary to start from the indubitable experience of one’s own cognitive activity that can be expressed in the sentence: I realize that I am asking about something.⁴⁰ Then, reflecting upon the act of asking, the questioner should notice that the necessary condition for asking any question is the occurrence of certain conditions both in the one who asks and in the object to which the question refers to – in other words, the question refers to something. At this stage, it is the realization that the question refers to some indefinite being. Then comes the awareness that this indefinite being has some definite fundamental and derivative properties. This, in turn, leads to the statement that it is impossible for an object to have some properties and not to have them at the same time – known as the law of noncontradiction: nothing can be and not be in the same respect. Furthermore, the question about the properties of an object arises: why is an object such and such. The transcendental argumentation – in the reflection upon this question – leads to the principle of sufficient reason: each object must have a reason for being such and such. As far as (ii) is concerned, Coreth argues that transcendental reflection differs from narrowly understood logical analysis, that is, an analysis determining different relations between clearly formulated concepts or propositions under the aspect of non-contradiction and the relation of entailment. In contrast, transcendental reflection is about reconstructing the “logical genesis” of a given concept, proposition etc., which goes in the direction that is opposite to narrowly understood logical dependencies. Obviously, transcendental reflection also concerns logical relations, but these are relations between clearly specified components of knowledge and the components that are implicitly co-adopted. The emergent chain of such relations concludes with the most basic rules of thinking, such as the law of identity, the law of noncontradiction, the law of excluded middle, etc. K. Rahner also tried to combine Aquinas and Kant. His metaphysical doctrine was best explained in Spirit in the World (Rahner (1939)) and Hearers of the Word (Rahner (1941)). IFor our purposes here, the first of these works is sufficient to understand Rahner’s philosophy, whereas the second addresses the same issues

40 See Coreth (1969) and cf., Herbut (1991/1992), pp. 130 – 132.

Part I: The Doctrine of St. Thomas Aquinas and its Non-Analytical Versions |

189

in greater detail. He is particularly interested in the question asked by Thomas Aquinas in Summa Theologica (I, Q84, a7): Can the intellect actually know anything through the intelligible species which it possesses, without making itself vulnerable to phantasms? This issue is closely related to the following one: How can the human intellect know any non-sensible thing or God?⁴¹ Rahner attempts to answer these questions “under the general influence of Maréchal and with a few particular borrowings from Heidegger, a rereading of Aquinas through the lens of Kant and the post-Kantians” (Kilby (2004), p. 4). According to Rahner, the absolute beginning for philosophy is that “Man questions – necessarily” (Dych’s translation of Rahner (1939), p. 58, and cf., also Cullen (1999), p. 74). He adds that the starting point of philosophy is a metaphysical question, and explains: For in fact, to put it first of all quite formally, the metaphysical question is that question which in a final and radical sharpening of man’s questioning turns upon itself as such and thereby turns upon the presuppositions which are operative in itself; it is the question turned consciously upon itself, the transcendental question, which does not merely place something asked about in question, but the one questioning and his question itself, and thereby absolutely everything. (The Dych’s translation of Rahner (1939), p. 74)

He further argues: This gives the starting point of metaphysics a peculiar duality and a unity at once: the starting point is questioning man, who as such is already with being in its totality ... this starting point is a question and no answer reaches out beyond the horizon which the question has already set as a limit beforehand. (The Dych’s translation of Rahner (1939), p. 75)

For Rahner, being is being-present-to-self : Knowing does not come about ‘through a contact of the intellect with the intelligible thing’, but being and knowing are the same. Knowing is the being-present-to-self of being, and the being-present-to-self is the being of the existent. (The Dych’s translation of Rahner (1939), p. 69)⁴²

41 Cf., “Karl Rahner (1904 - 1984)” in the Boston Collaborative Encyclopedia of Western Theology. 42 Ch. M. Cullen thinks (Cullen (1999), p. 76) that the first half of this statement: “Knowing is the being present-to-self of being” may be interpreted as a Rahnerian version of the Thomistic teaching that knowing is the union of the knower and the object, whereas the second half of this statement: “The being present-to-self is the being of the existent” expresses the requirement that being is subjectivity.

190 | Mirosław Szatkowski An important term in Rahner’s argument is the ‘pre-apprehension of Being’ – which he uses interchangeably with the ‘pre-apprehension of Infinite (Absolute) Esse’ or ‘pre-apprehension of God’. He contends: Pre-apprehension as such does not attain to an object. By its very essence, it is one of the conditions of the possibility of an objective knowledge. Every represented object of human knowledge is able to be apprehended itself only in a pre-apprehension. (The Dych’s translation of Rahner (1939), p. 143)⁴³

God is the ontological ground for all human judgments. In addition, he says that “the absolute esse (that is God) is implicitly and simultaneously affirmed in every act of the agent intellect in every judgment.” (The Dych’s translation of Rahner (1939), p. 226). Which Cullen explains as follows: In this way the agent intellect is a participation in the light of the Absolute Spirit. The reason for this is because knowledge of finite objects is always a knowledge of their limits and this in turn implies that they can transcend their limits against the horizon of infinite esse. (Cullen (1999), p. 77)

Rahner considers the first principles of Thomas Aquinas as a priori ontological conditions of knowledge. Speaking for Cullen (1999), p. 79: “For Thomas the first principles are not just any product of the intellect, but the fundamental product”; these first principles are “the instrumental principle for abstraction” and “stand on the side of the agent intellect.”⁴⁴ The light of the agent intellect is the a priori and formal condition for the objectivity of the world. But it is also the source for the first principles: “Insofar as it [the agent intellect] apprehends this material of sensibility within its anticipatory (vorwegnehmenden) dynamism to esse, it ‘illuminates’ this material, gives it those metaphysical structures of being which were expressed in the first principles.”⁴⁵

One of the philosophers whom K. Rahner influenced was the Belgian Jesuit J. Donceel, “... gave Rahner’s ideas a much wider audience than they might have otherwise enjoyed by explaining them in more accessible language in his text books” (Cullen (1999), p. 73). More specifically, he defends Rahner’s view that the starting point of metaphysics is pragmatic objects (Donceel (1974), pp. 74–75), by which is meant such objects as trees, houses, money, dogs, other people and so on. Pragmatic objects are commonplace, every day objects, as opposed to those which re-

43 According to Cullen (1999), p. 83: “Rahner’s doctrine of the pre-apprehension of absolute esse seems to beremarkably close to this fundamental Hegelian notion of identity-in-difference.” 44 The Dych’s translation of Rahner (1939), p. 204. 45 The Dych’s translation of Rahner (1939), p. 225.

Part I: The Doctrine of St. Thomas Aquinas and its Non-Analytical Versions |

191

main mostly implicit in these day-to-day activities, for example, space, time, and God. These pragmatic objects can all be reduced to the following formula: ‘Something which is X’, where X is the information which comes from the senses while ‘something which’ is the contribution of our intellect. Donceel says that “of all that which we get to know in our daily activity we may say a priori that it is ‘something which’ and for further information about it we shall have to recur to the senses” (p. 75). And he adds that ‘something which’ is equivalent to ‘a being that’, containing implicitly the whole of metaphysics. ‘Something which’ or ‘a being that’, as “an a priori contribution of our intellect,”(p. 76)... “is the most universal of our concepts” (p. 75). Donceel stresses that: For Transcendental Thomism ... ‘being’ is contributed a priori by the intellect itself. Not in the idealistic sense that we make that which we know into ‘beings’, but in the sense that we become aware that the objects which we know are ‘beings’ because and inasmuch as we know them through our intellect. (p. 76)

Furthermore, ‘Being’ comes to us through the senses, but not from the senses. Its knowledge is inborn in us, but only virtually inborn. It may be known explicitly only through and in sense knowledge. (p. 77)

For Donceel, “ metaphysics is the set of all our basic affirmations about reality” (p.79); it is “perfectly one” (p. 80) and “ontologically prior to all really human knowledge and a condition of its possibility” (p. 77). He states that the agent intellect is the formal cause of all our knowledge, and explains, in Donceel (1957), the key role of this element in the epistemology of Transcendental Thomism. The agent intellect always contributes something to sense experience – the agent intellect is “the formal element of the cause of our intellectual knowledge”, while sense experience provides the material element of the cause.⁴⁶ Cullen explains this position as follows: It is the agent intellect then that makes the a priori contribution of our intellect to our knowledge. ... Donceel appeals to the Thomistic position that that which is known is known after the manner of the knower. Hence, that which I know is the tree, but that in which I know this tree is the affirmed concept and is in part a construction of my intellect. (Cullen (1999), p. 78)

46 Cf., Donceel (1957), p. 190, and Cullen (1999), p. 78.

192 | Mirosław Szatkowski And Donceel contends that “... the light of the agent intellect consists precisely in the truth of the first principles”, which are “the a priori contribution of our intellect to every object we know” (Donceel (1957), p. 193). Thus, in Donceel’s conception, metaphysical knowledge is absolutely certain because it shares in the certitude of divine knowledge (ibid., p. 196). When it comes to knowledge of the existence of God, Donceel thinks that all human beings affirm His existence necessarily, though unconsciously, in every existential affirmation. In a later work, Donceel (1979) uses transcendental method to validate the ontological proofs and the First Way of Thomas Aquinas. According to him, Maréchal, Rahner and other Transcendental Tomists use neither a priori nor “a posteriori arguments for the existence of God but proffer a simultaneo vindication of God’s existence” (Ogden (2007), p. 215). Donceel recognizes difficulties with explaining this: A basic principle is one that is a condition of the possibility of every affirmation, one which we affirm in the very act of denying it. It is easy to show that the principle of identity fulfills these conditions. I do not see how the principle of universal correlativity does. (Donceel (1979), p. 186)

One of the big issues for Denceel was the nature of God. In his opinion the perennial philosophical system of Aristotle and Aquinas, even modified by Maréchal, Rahner or others, faces serious problems. He formulated his own position as intermediate between traditional Thomism and pantheism.⁴⁷ B. Lonergan uses the transcendental method in a different manner than Transcendental Thomists, but like Rahner emphasizes the dynamism of the intellect, “he works with empirically observable psychological function, and attempts to chart the operations of the intelligence as it comes to an objective judgment of fact. Thus, for Lonergan, objective truth is reached by empirically observable subjective operations.” (Streeter (2016), p. 2) In short, Lonergan works empirically, in contrast to other Transcendental Thomists, who work theoretically. In Lonergan’s main philosophical work: Insight: A Study of Human Understanding (Lonergan (1957/1992)) the author struggles with two questions: What is happening when we know? and What is known when that is happening? The answer to the first question yields a cognitional theory and an epistemology, while to the second question yields a metaphysics. The primary aim of his General Empirical Method is to examine the human mind in the process of knowing, to observe what people do when they are having, formulating and verifying ideas. Knowing, according to Lonergan, is constituted by four fundamental kinds of acts and levels: (i). the level of presentations (experience, by which we aware of all reality, physical, social, and

47 See, the second part of Donceel (1979).

Part I: The Doctrine of St. Thomas Aquinas and its Non-Analytical Versions |

193

spiritual); (ii). the level of intelligence (understanding, by which we ask questions to reach an understanding this data); (iii). the level of reflection (judging, by which we assess the corrections of our understanding); and (iv). the level of decision (deciding, by which we evaluate what is worth acting upon from what we know). Indeed, Lonergan does not concentrate on theoretical results of Thomas Aquinas, but he tried to understand “how Thomas’ magnificet mind worked” (cf., Streeter (2016), p. 3).⁴⁸ The aim of knowing, according to Lonergan, is to know what really is. And he defines reality, not in terms of what one knows, but in terms of the objective to be known when one will know correctly all that is to be know.⁴⁹ Lonergan distinguishes the principal and the partial notion of objectivity, taking the latter to be decomposable into experiential, normative and absolute objectivity. The principal notion of objectivity occurs on the level of judgment. This objectivity is presupposed when the judgment ‘I am a knower’ is distinguished from other judgments, for example, ‘This is a dog’ and ‘I am not this dog’. Consequently, the validity of this kind objectivity depends on the validity of the particular judgments. The experiential notion of objectivity is simply the givenness of the data; the normative notion of objectivity manifests on the level of understanding (one of the elements in normative objectivity is the absence of contradiction); and the absolute notion of objectivity appears on the level of judgment (true judgments are always absolutely true, and in this sense they are absolutely objective).⁵⁰ The above cog-

48 Lonergan (1957/1992), pp. 768–770 (see also, Streeter (2016), p. 3): To penetrate to the mind of a medieval thinker is to go beyond his words and phrases. It is to grasp questions as they were grasped. It is to take the omnia opera of such a writer as St. Thomas Aquinas and to follow through successive works the variations and developments of his views. . . . to arrive at a grasp of their motives and causes. It is to discover for oneself that the intellect of Aquinas. . . reached a position of dynamic equilibrium without ever ceasing to drive towards a fuller and more nuanced synthesis, without ever halting complacently in some finished mental edifice, as though his mind had become dull, or his brain exhausted, or his judgment had lapsed into the error of those that forget man to be a potency in the realm of intelligence. Nor is this labor of penetration enough... After spending years reaching up to the mind of Aquinas, I came to a twofold conclusion. . . that reaching had changed me profoundly ... (and) that change was the essential benefit. . . my detailed investigations of the thought of Aquinas. . . have been followed by the present essay in aid of a personal appropriation of one’s own rational self-consciousness. ... In the Introduction I stated a program. Thoroughly understand what it is to understand, and not only will you understand the broad lines of all there is to be understood but also you will possess a fixed base, an invariant pattern, opening upon all further developments of understanding. 49 Cf., Lonergan (1957/1992), pp. 372–373 and 659–662. 50 Cf., Lonergan (1957/1992), pp. 402–404.

194 | Mirosław Szatkowski nitional and epistemological notions define categories Lonergan’s metaphysics, which is simply the metaphysics of the knowing subject. In Lonergan’s words: “[Metaphysics] underlies all other departments, for its principles are neither terms nor propositions, neither concepts nor judgments, but the detached and disinterested drive of the pure desire to know and its unfolding in the empirical, intellectual and rational consciousness of the self-affirming subject.” (Morelli and Morelli (1997), p. 228) And, “Now let us say that explicit metaphysics is the conception, affirmation, and implimentation of the integral heuristic structure of proportionate being.” (p. 230) And now, the metaphisical answer, in Lonergan’s sense, to the question of what the knowing subject knows is: He knows that he does not know everything, he knows that he has the capacity to know what he does not yet know, and he knows that what he wants to know is the being of what he does not know. Finally, let us mention that Lonergan’s second important work is Method in Theology (Lonergan (1971)), in which he applies his general empirical method, as developed in Lonergan (1957/1992), to theology. He distinguishes two phases of theological activity: (1). the phase of learning from the past, and (2). the phase of solving contemporary problems. He also distinguishes eight functional specialties in theology: research, interpretation, history, dialectics, science of foundations, science of doctrines, science of systems, and science of communication. In summary, what distinguishes Transcendental Thomism from Neo-Thomism is that they derive from different epistemological theories. Human knowing is, according to the former, a projection of the knower upon the real, whereas according to the latter, it is a reception from the real. This projection is a special kind of intellectual dynamism of the knower to the unconceptualizable term of Infinite Being. Otherwise, the contact of the intellect with reality is not through concepts abstracted from things, but through its own dynamism as it turns to Infinite Being. Transcendental Thomism rejects the Neo-Thomist’s view that the objectification of things (i.e., finite beings) takes place through an automatic and natural abstraction of the ratio entis. Neo-Thomists claim that there is only one, true philosophy, as there is only one universal concept of being, which, however, remains not fully explored. In contrast, Transcendental Thomists claim that one, true philosophy cannot exist, since all concepts are formulated in the course of intellectual dynamism and no concept reaches the end of this dynamism.⁵¹ Critics of Transcendental Thomism argue that the idealistic and subjective metaphysics of any version of the transcendental method is incompatible with the realistic metaphysics of St. Thomas. This idealism and subjectivism of Tran-

51 Cf., Knasas (2005).

Part I: The Doctrine of St. Thomas Aquinas and its Non-Analytical Versions |

195

scendental Thomism, according to Ch. M. Cullen, manifests itself in the following notions: (1) man as questioning is the certain starting point for metaphysics; (2) man is already with being in its totality; (3) being is subjectivity; (4) the intellect pre-apprehends Infinite Esse; (5) the agent intellect is the power of forming the first principles of transcendental validity; (6) the first principles function as a priori conditions for knowledge. (Cullen (1999), p. 73)

In the opinion of adherents of more traditional Thomism, Transcendental Thomism is internally inconsistent and metaphysically unsound, and there is no philosophical reason to call it ‘Thomism’ at all.⁵²

3.4 Lublin Philosophical School The term ‘Lublin Philosophical School’ denotes a specific approach to realistic (classical) philosophy developed in the 1950s by a group of philosophers at the Catholic University of Lublin (KUL), Poland. The initiators of the School were: M. A. Krąpiec, J. Kalinowski and S. Świeżawski. Yet, the main contribution to the advancement of its philosophical programme was made by M. A. Krąpiec (Krąpiec (1959a,b, 1962, 1963, 1966), Krąpiec and Kamiński (1962)⁵³) and S. Kamiński (Kamiński (1959a,b, 1960a,b,c, 1961a,b, 1962, 1963, 1964a,b, 1965a,b), Kamiński and Krąpiec (1961)⁵⁴), who – each in his own way – also had a major impact on its philosophical character. The School was opposed to both Suarezian Neoscholasticism, i.e., Ch. Wolff’s and J. Kleutgen’s essentialism, which – according to the Lublin circle – had deformed the conceptions of classical philosophy, as well as to the Lvov-Warsaw School’s approach to philosophy. According to M. A. Krąpiec and A. Maryniarczyk (see, Krąpiec and Maryniarczyk (2005)), the approach to philosophy by the Lublin Philosophical School can be characterized by the following aspects. (1). Cognitive realism: The subject of philosophy is a really existing being. The existential conception of being, i.e., understanding being as something that exists, which – according to the creators of the School – is the only correct interpretation of St. Thomas Aquinas’ works, is the first and basic distinctive feature of the

52 For example, in Henle (1981), pp. 92 and 110, R. J. Henle notes that Transcendental Thomism is a “Christianized version of German idealism ... and has no philosophical right to be called ‘Thomism’.” 53 The full bibliography of Krąpiec’s works can be found in Szymaniak (2007). 54 The full bibliography of Kamiński’s works can be found in Buczek and Szubka (1987).

196 | Mirosław Szatkowski Lublin School. The proper way to identify such a being is by means of separation, as developed by St. Thomas Aquinas. In acts of direct cognition, before the perception of subject-object duality emerges, the being itself is experienced. These acts of direct cognition are expressed in existential judgements. They clearly state the existence of something which is not well known: “there exists something”, “it exists”, “somebody exists”. At the second stage existential judgements referring to single existing objects are formulated, e.g., “John exists” or “soul exists”. Analyzing these results leads to the conclusion that existing beings are characterized by nonidentity between essence and existence, i.e., that each specific being consists of the content factor and the factor which actualizes this content to existence. During the next stage, a transition occurs from the categorial approach of essence and existence to transcendental approaches, pertaining to such elements in a specific John which are necessary not only for the existence of John to exist as something concrete, but also for John to exist as a real being. In this way, the understanding of being as something that has a specific content and exists proportionally to this content is explicated. (2). Maximalism: Metaphysics is not limited to the study of a certain fragment of reality, but addresses all existentially significant issues – it investigates laws governing everything that exists, explores the inner structure of everything that exists, and asks about its causes. It formulates the principle that the ultimate reason for the existence of complex, changeable and unnecessary beings is the existence of an Absolute Being. (3). Transcendentalist orientation: The inherent transcendentalism of the Lublin approach is reflected in the language it uses to describe metaphysics and the universal nature of metaphysical statements. More precisely, the language of metaphysics is treated as a first-order language, i.e., an object language, which is aimed at representing things, instead of defining concepts. In addition, it is an analogical-transcendentalizing language, in which the formulated statements represent transcendental properties, i.e., properties of all beings.⁵⁵

55 S. Kamiński puts it as follows: The language of the theory of being differs in its character from the language of other types of knowledge, and moreover, it is difficult to make a full semiotic determination of it. Although with respect to the analytical aspect it is close to the language of the formal sciences, at the same time it is marked by an integral and almost extreme realism. It concerns the qualitative aspect of reality, but at the same time it gives ontological and cognitive primacy to the general-existential aspect. In terms of its genesis it is derived from ordinary language and is chiefly based on it, but at the same time it uses terminology that has more specialized semiotic functions. Finally, it uses names with wider scope, and at the same time it ascribes

Part I: The Doctrine of St. Thomas Aquinas and its Non-Analytical Versions |

197

(4). Methodological-epistemological unity: The School put great emphasis on methodological reflection. On the one hand, the indispensability of using historical-reconstructive method in philosophy was stressed (taking historical experience, especially St. Thomas Aquinas’ thought, into account), and on the other, creating independent methodology for philosophy – and especially metaphysics as its central discipline – was recognized as an important goal. The conviction is that philosophy, in spite of being one of the oldest cognitive disciplines, had never developed its own adequate methodology (see, Krąpiec (1998), p. 249). The specific features of metaphysical cognition are reflected in its analogical-transcendental language and the procedures of explanation, proof and justification, which often occur simultaneously. The basis of metaphysical cognition is formed by the common-sense cognition, which ensures connection to the real world. Then the common-sense cognition data are analyzed in relation to the question “why” – such decontradictifying factors are involved whose negation results in the rejection of the fact that is being explained. This gives us a method of object-explanation that must be distinguished from deduction in the Aristotelian sense (syllogistic deduction) or its mathematical equivalent. In fact, metaphysical concepts are indefinable per genus proximum et differentiam specificam, but, as well, they extend beyond genus and their scope is unlimited. (5). Coherence: The School is concerned to show the unity of the philosophical approach created by the general metaphysics and specific branches of metaphysics (cognition theory, philosophy of nature, philosophy of God, philosophy of morality, philosophy of culture, etc.). This unity is guaranteed by the inner identity of the object (that is, everything that exists) and the unity of cognition of this object.⁵⁶

to these names content that is not at all impoverished. To reconcile these oppositions and to develop the difficulties connected with them, a doctrine of analogy, participation, the transcendentals, and necessary truths was developed that was adequate to the language of theory of being. (Kamiński (1989), p. 81, as translated by Hugh McDonald) 56 S. Kamiński writes: ... the theory of being so conceived takes in all the disciplines of realistically conceived metaphysics and constitutes uniform philosophical cognition under the epistemologicalmethodological aspect. This means that the theory of being covers the entire fundamental problematic of so-called classical philosophy and develops it basically in the same way in all its disciplines. Here the theory of cognition is not distinguished from metaphysics as two domains of philosophy (in view of the method of ultimate explanation). The theory of cognition as a separate philosophical discipline simply loses its reason for existence, since many of its chief questions sprang up on erroneous ways of metaphysics, hence it has a

198 | Mirosław Szatkowski (6). Objectivity: The School holds that it is only attainable by the self-reflexive verifiability of statements, which is achieved by relating them every time to object evidence. The Lublin Philosophical School has not gained much attention in philosophical literature worldwide. Its members did not manage – and probably even did not try – to promote their ideas in international specialist journals. Another rather inexplicable omission by the School is that it failed to address such major developments in the foundations of mathematics as those concerning infinity, truth, and omniscience. These are issues crucial for Thomism and cannot be avoided.

Bibliography Armour, L. (2008), “The Philosophy of Charles De Koninck”, in McInerny (ed. and tr.), 1–68. Ashley, B. (1991), “The River Forest School and the Philosophy of Nature Today”, in Philosophy and the God of Abraham: Essays in Memory of James A. Weisheipl, OP, edited by R. J. Long, Toronto: Pontifical Institute of Mediaeval Studies, 1–15.

metaphilosophical character. Discussion on the various kinds of idealism can happen on the occasion of a meta-philosophical rational justification of the way the concept of being is formed. Controversies on the value of cognition can be examined in the history of philosophy (in the context of establishing errors and distortations of metaphysical thought). The history of philosophy in particular should provide the theory of being with factual experience for the choice of the proper way to cultivate metaphysics.(Kamiński (1989), p. 76, as translated by Hugh McDonald) And further, This unity of cognition in the theory of being is achieved by accepted objective philosophical thought and the ultimate explanation exclusively on the basis of the internal structure of being. However, he who assumes that non-dogmatic philosophical explanation must be meta-objective (of a reflective or interpretative type), or that we can appeal ultimately to exclusively qualitative structures of reality, establishes the theory of cognition as the ultimate (first) philosophical discipline, and he breaks metaphysics apart into methodologically different disciplines. Meanwhile, metaphysical cognition in the theory of being is broken into particular disciplines only in view of the different starting point (a separate type of object of the data of experience), but not in the way of ultimate explanation (and the formal object of the most theoretical theses). (Kamiński (1989), p. 76, as translated by Hugh McDonald).

Part I: The Doctrine of St. Thomas Aquinas and its Non-Analytical Versions |

199

Ashley, B. (2006), The Way toward Wisdom: An Interdisciplinary and Contextual Introduction to Metaphysics, Houston: University of Notre Dame Press for the Center of Thomistic Studies. Buczek, A. I. and Szubka, T. (1987), “Bibliografia prac Stanisława Kamińskiego” [“A Bibliography of Stanislaw Kaminski’s works”], in Roczniki Filozoficzne: 35, 5–47. Chojnacki, P. (1969), “Realism metodyczny Etienne Gilsona” [“A methodical realism of Etienne Gilson”], in Studia Philosophiae Christianae: 5(2), 199–202. Clarke, W. N. (1952), “The Limitation of Act by Potency”, in New Scholasticism: 26(2), 167–194. Coreth, E. (1961), Metaphysik: Eine methodisch-systematische Grundlegung [Metaphysics], Innsbruck/Wien/München: Tyrolia. English edition by J. Donceel (New York: Herder & Herder, 1968). Coreth, E. (1969), “Zum Begründungsproblem der Metaphysik”, in Akten des XIV Internationalen Kongresses für Philosophie, 2–9 September 1968, Bd. 3, Logik, Erkenntnis- und Wissenschaftstheorie. Sprachphilosophie. Ontologie und Mrtaphysik, Wien: Universität Wien, 596–602. Cullen, Ch. M. (1999), “Transcendental Thomism: Realism Rejected”, in The Failure of Modernism, edited by B. Sweetman, CUA,72–86. De Koninck, Ch. (1936/2008), “The Cosmos”, in McInerny (ed. and tr.), 235–354. De Koninck, Ch. (1946), “Concept, Process and Reality”, in Laval théologique et philosophique: 2(2), 141–146. De Koninck, Ch. (1947), “Introduction a l’etude de l’âme”, in Laval théologique et philosophique: 3(1), 9–65. De Koninck, Ch. (1956), “Random Reections on Science and Calculation”, in Laval théologique et philosophique: 12(1), 84–119. De Koninck, Ch. (1957), “Abstraction from Matter, I”, in Laval théologique et philosophique: 13(2), 133–196. De Koninck, Ch. (1959), “Natural Science as Philosophy”, in Culture: 20(3), 245–267. De Koninck, Ch. (1960a), The Hollow Universe, New York: Oxford University Press. De Koninck, Ch. (1960b), “Abstraction from Matter, II”, in Laval théologique et philosophique: 16(1), 53–69. De Koninck, Ch. (1961), “The Unity and Diversity of Natural Science”, in The Philosophy of Physics, edited by V. E. Smith, New York: St. John’s University Press, 5–24. De Koninck, Ch. (2008), “Are the Experimental Sciences Distinct from the Philosophy of Nature?”, in McInerny (ed. and tr.), 441–455. De Koninck, T. (2008), “Charles De Koninck: A Biographical Sketch”, in McInerny (ed. and tr.), 69–97. Dewan, L. (1999), “Etienne Gilson and the Actus Essendi”, in Maritain Studies/Etudes Maritainiennes: 15, 71–96. Donceel, J. (1957), “A Thomistic Misapprehension”, in Thought: A Journal of Philosophy: 32, 189–198. Donceel, J. (1974), “Transcendental Thomism”, in The Monist: 58, 67–85. Donceel, J. (1979), The Searching Mind: An Introduction to a Philosophy of God, Notre Dame: University of Notre Dame Press. Fernandez-Alonso, A. (1935), “Scientiae et philosophia secundum S. Albertum Magnum”, in Angelicum: 12, 24–59. Feser, E. (2009), The Thomistic tradition, Part I. URL = .

200 | Mirosław Szatkowski Garcia, B. (2010), “Philosophy and Its History: An Analysis of Gilsin’s Historical Method and Treatment of Neoplatonism”, in RAMIFY: The Journal of the Braniff Graduate School of Liberal Arts: 1, 82–93. Gilson, É. H. (1922), Le Thomisme: Introduction a 1il Philosophie de Saint Thomas d’Aquin, Paris: Librairie Philosophique J. Vrin. Also: Thomism: The Philosophy of Thomas Aquinas, 2002, translated by L. K. Shook and A. Maurer, Toronto: Pontifical Institute of Mediaevel Studies. Gilson, É. H. (1936), Le Réalisme Méthodique, Paris: Téqui. Gilson, É. H. (1937), The Unity of Philosophical Experience, San Francisco: Ignatius Press. Gilson, É. H. (1939), Réalisme thomiste et critique de la connaissance, Paris: Librairie Philosophique J. Vrin. Also: Thomist Realism and the Critique of Knowledge, 1986, translated by M. A. Wauck, San Francisco: Ignatius Press. Gilson, É. H. (1948), L’être et l’essence [The Essence of Being], Paris: Librairie Philosophique J. Vrin. Gilson, É. H. (1952), Being and Some Philosophers, 2nd ed., Toronto: Pontifical Institute of Mediaeval Studies. Gilson, É. H. (1960), Elements of Christian Philosophy, New York: Doubleday & Company, Inc.. Gilson, É. H. (1963), Byt i Istota, tłumaczył P. Lubacz i J. Nowak, Warszawa: PAX. Gilson, É. H. (1969), God and Philosophy, 2nd ed., with foreword by J. Pelikan, New Haven: Yale University Press. Gilson, É. H. (1974), “Propos sur l’^etre et sa notion”, in San Tommaso e il pensiero moderno, edited by A. Piolanti, Citta Nuova: Pontificia Accademia Romana de S. Tommaso d’Aquino. Haldane, J. (ed.) (1997a), Analytical Thomism, The Monist: 80(4). Hart, D. B. (1951), “Twenty-Five Years of Thomism”, in New Scholasticism: 25(1), 3–45. Henle, R. J. (1981), “Transcendental Thomism. A Critical Assessment”, in One Hundred Years of Thomism, edited by V. B. Brezik, Houston: University of St. Thomas - Center for Thomistic Studies, 90–116. Herbut, J. (1991/1992), “Aktualność Transcendentalnej Metody Filozofowania” [“Actuality of the Transcendental Method of Philosophizing”], in Roczniki Filozoficzne [Annals of Philosophy]: 39/40, 119–134. Hintikka, J. (1984), “Kant’s Transcendental Method and His Theory of Mathematics”, in Topoi: 3, 99–108. Hittinger, J. P. (1987), “The Intuition of Being: Metaphysics or Poetry?”, in Jacques Maritain, the Man and His Metaphysics, edited by J. Knasas, Mishawaka: American Maritain Association, 71–82. Kamiński, S. (1959a), “O logicznych związkach zachodzących między tezami metafizyki ogólnej” [“The Logical Relationships Occurring Between Theses of General Metaphysics”], in Sprawozdania z Czynności Wydawniczej i Posiedzeń Naukowych oraz Kronika TN KUL: 10, 180–184. Kamiński, S. (1959b), “O ostatecznych przesłankach w filozofii bytu” [“About the Final Premises in the Philosophy of Being”], in Roczniki Filozoficzne: 7, 41–72. Kamiński, S. (1960a), “Rola dedukcji w metafizyce tomistycznej” [“The Role of the Deduction in Thomistic Metaphysics”], in Sprawozdania z Czynności Wydawniczej i Posiedzeń Naukowych oraz Kronika TN KUL: 11, 64–72. Kamiński, S. (1960b), “O definicjach w systemu metafizyki ogólnej” [“About Definitions in the System of General Metaphysics”], in Roczniki Filozoficzne: 8, 37–54.

Part I: The Doctrine of St. Thomas Aquinas and its Non-Analytical Versions |

201

Kamiński, S. (1960c), “O niejednostronną metodykę metafizyki” [“Oh no Unilateral Methodology of Metaphysics”], in Znak: 12, 1423–1428. Kamiński, S. (1961a), “Logika współczesna a filozofia” [“The Contemporary Logic and Philosophy”], in Roczniki Filozoficzne: 9, 49–84. Kamiński, S. (1961b), Pojęcie nauki i klasyfikacja nauk [The Concept of Science and Classification of Sciences], Lublin: KUL. Kamiński, S. (1962), “O uzasadnianiu tez filozoficznych” [“The Justification of Philosophical Theses”], in Roczniki Filozoficzne: 10, 37–65. Kamiński, S. (1963), “Czym są w filozofii i w logice tzw. pierwsze zasady?” [“What are So Called First Principles in the Philosophy and Logic?”], in Roczniki Filozoficzne: 11, 5–23. Kamiński, S. (1964a), “Co daje stosowanie logiki formalnej do metafizyki klasycznej?” [“What Does the Use of the Formal Logic to Classical Metaphysics?”], in Roczniki Filozoficzne: 12, 107–112. Kamiński, S. (1964b), “Koncepcja analityczności a konieczność tez metafizyki” [“The Concept of Analyticity and the Necessity of Thesis of the Metaphysics”], in STNKUL: 14, 65–70. Kamiński, S. (1965a), “Aksjomatyzowalność klasycznej metafizyki ogólnej” [“Axiomatizability of the Classical General Metaphysics”], in Studia Philosophiae Christianae: 1, 103–116. Kamiński, S. (1965b), “O podziale filozofii klasycznej” [“The Breakdown of Classical Philosophy”], in (STNKUL: 15, 55–57. Kamiński, S. (1989), “Osobliwość metodologiczna teorii bytu” [“The Methodological Peculiarity of the Theory of Being”], in Jak filozofować? Studia z metodologii filozofii klasycznej [How to Philosophize? Studies in the Methodology of Classical Philosophy], Printing material has been prepared by Tadeusz Szubka, Lublin: Towarzystwo Naukowe KUL, 71–87. http://www.ptta.pl/lsf/history.pdf Kamiński, S. and Krąpiec, M. A. (1961), “Specyficzność poznania metafizycznego” [“A Specificity of the Metaphysical Knowledge”], in Znak: 13, 629–637. Kant, I. (1929), Critique of Pure Reason, translated by Norman Kemp Smith, New York: St. Martin’s Press. Kant, I. (1764), “Untersuchung über die Deutlichkeit der Grundsätze der natürlichen Theologie und der Moral” [“Inquiry concerning the Distinctness of the Principles of Natural Theology and Morality”], in Theoretocal Philosophy, 1775 - 1770, edited and translated by D. Walford and R. Meerbote (1992), 247–275. Kant, I. (1783), Prolegomena zu einer jeden künftigen Metaphysik die als Wissenschaft wird auftreten können [Prologomena to any Future Metaphysic that can be Present itself as a Science], see, www.earlymoderntexts.com/assets/.../kant1783.pdf. Kant, I. (2002), Theoretical Philosophy after 1781, translated by H. Allison, M. Friedman, G. Hatfield, P, Heath, New York: Cambridge University Press. Kilby, K. (2004), Karl Rahner: Theology and Philosophy, London & New York: Routledge. Klenk, K. F. (2015), The Albertus Magnus Lyceum – An Attempt at Integrating the Modern Sciences with a Realist Philosophy of Nature. URL = . Knasas, J. (1988), “How Thomistic Is the Intuition of Being?”, in Jacques Maritain, the Man and His Metaphysics, edited by J. Knasas, Mishawaka: American Maritain Association, 83–91. Knasas, J. (1990), “Gilson vs. Maritain: The Start of Thomistic Metaphysics”, in Doctor Communis: 43, 250–265. Knasas, J. (2003), Being and Some Twentieth-Century Thomists, New York: Fordham University Press.

202 | Mirosław Szatkowski Knasas, J. (2005), “John Knasas on Thomistic Metaphysics Past, Present and Future. Interview”, in Inner Explorations, http://www.innerexplorations.com/philtext/john.htm. Krąpiec, M. A. (1959a), Realizm ludzkiego poznania [Realism of Human Cognition], Poznań: Pallottinum. Krąpiec, M. A. (1959b), Teoria analogii bytu [Theory of the Analogy of Being], Lublin: RW KUL. Krąpiec, M. A. (1962), Dlaczego zło? Rozważania filozoficzne [Why Evil? Philosophical Analysis], Kraków: Znak. Krąpiec, M. A. (1963), Struktura bytu. Charakterystyczne elementy systemu Arystotelesa i Tomasza z Akwinu [Structure of Being: Characteristic Elements of Aristotle’s and Thomas Aquinas’s System], Lublin: TN KUL. Krąpiec, M. A. (1966), Metafizyka. Zarys podstawowych zagadnień, Poznań: Pallottinum; Metaphysics. An Outline of the Theory of Being, translared by M. Lescoe, A. Woznicki and Th. Sandok, New York: Mariel Publications, (1991). Krąpiec, M. A. (1998), “O filozoficznej szkole lubelskiej” [“The Lublin Philosophical School”], in M. A. Krąpiec, Dzieła. Człowiek, Kultura, Uniwersytet [Works. Man, Culture, University], vol. XII, Lublin: TN KUL, 249–255. Krąpiec, M. A. and Kamiński, S. (1962), Z teorii i metodologii metafizyki [On the Theory and Methodology of Metaphysics], Lublin: TN KUL. Krąpiec, M. A. and Maryniarczyk, A. (2005), “Lubelska Szkoła Filozoficzna” [“The Lublin Philosophical School”], in Powszechna Encyklopedia Filozofii [Universal Encyclopedia of Philosophy], Lublin: KUL: 6, 532–550. Krąpiec, M. A. and Zdybicka, Z. J. (1983), “Poznanie Boga w ujęciu J. Maritaina” [“Knowledge of God according to J. Maritain”], in Roczniki Filozoficzne [....]: 31(2), 9–17. Lonergan, B. (1957/1992), Insight: A Study of Human Understanding, Toronto, Buffalo, London: University of Toronto Press. Lonergan, B. (1971), Method in Theology, London: Darton, Longman & Todd. Maréchal, J. (1927-49), Le Point de depart de la metaphysique [The Starting Point of Metaphysics], 5 vols., Paris: Desclee de Brouwer, and Brussels: L’Edition Universelle. A Maréchal Reader, 1970, edited and translated by J. Doncell, New York: Herder & Herder (This is an anthology of Maréchal’s writings taken almost entirely from his major work, Le Point de depart de la metaphysique). Maritain, J. (1921/1944), Elements de philosophie: introduction generale a la philosophie, Vol. I, (Paris: Librairie Pierre Tequi, 1921) [Introduction to Philosophy, Vol. I, (1944)], translated by E. I. Watkin, London: Sheed and Ward. Maritain, J. (1932/1959), Les degrès du savoir (Paris: Desclée de Brouwer, 1932) [The Degrees of Knowledge (1959)], New York: Scribner’s. Maritain, J. (1934/1939), Sept leçons sur l’être et les premiers principes de la raison spéculative (Téqui, 1934) [A Preface to Metaphysics: Seven Lectures on Being (1939)], New York and London: Sheed and Ward. Maritain, J. (1935/1940), Science et sagesse (Paris: Labergerie, 1935) [Science and Wisdom (1940)], translated by B. J. Wall, London: Geoffrey Blas. Maritain, J. (1936/1951), La philosophie de la nature, essai critique sur ses frontières et son objet (Téqui, 1936) [Philosophy of Nature (1951)], translated by I. C. Byrne, New York: Philosophical Library. Maritain, J. (1939), Quatre essais sur l’esprit dans sa condition charnelle [Four Essays on Spirit, Paris: Desclée de Brouwer. Maritain, J. (1941), Ransoming the Time, New York: Scribner’s.

Part I: The Doctrine of St. Thomas Aquinas and its Non-Analytical Versions |

203

Maritain, J. (1944), De Bergson à Thomas d’Aquin, essais de métaphysique et de morale, New York: Edité par Maison Française. Maritain, J. (1947/1948), Court traité de l’existence et de l’existant (Paris: Hartman, 1947 [Existence and the Existent (1948)], New York: Pantheon. Maritain, J. (1948/1952), Raison et raisons (Paris: Egloff, 1948) [The Range of Reason (1952)], New York: Scribner’s. Maritain, J. (1953/1954), Approches de Dieu (Paris: Alsatia, 1953) [Approaches to God (1954)], New York: Harper and Brothers. McCool, G. A. (1992), From Unity to Pluralism. The Internal Evolution of Thomism, New York: Fordham University Press. McCool, G. A. (1994), The Neo-Thomists, Marquette: Marquette University Press. McInerny, R. M. (1965), “Charles De Konick: A Philosopher of Order”, in The New Scholasticism: 39(4), 491-516. McInerny, R. M. (2008), The Writings of Charles De Koninck, Vol. 1, Notre Dame, Indiana: University of Notre Dame Press. Morelli M. D. and Morelli E. A. (eds.) (1997), The Lonergan Reader, Toronto/Buffalo/London: University of Toronto Press. Muck, O. (1964), Die transzendentale Methode in der scholastischen Philosophie der Gegenwart, Innsbruck: Rauch. The Transcendental Method, translated by W. D. Seidensticker, New York: Herder and Herder (1968). O’Callaghan, J. and McInerny, R. (2014), “Saint Thomas Aquinas”, in The Stanford Encyclopedia of Philosophy. URL = . Ogden, S. G. (2007), The Presence of God in the World, Bern/Berlin/Bruxelles/Frankfurt am Main/New York/Oxford/Wien: Peter Lang. Owens, J. (1963), The Doctrine of Being in the Aristotelian Metaphysics, Toronto: Pontifical Institute of Mediaeval Studies. Owens, J. (1993), “Aristotle and Aquinas”, in The Cambridge Companion to Aquinas, edited by N. Kretzmann and E. Stump, Cambridge: Cambridge University Press, 38–59. Rahner, K. (1939), Geist im Welt: Zur Metaphisik der endlichen Erkenntnis bei Thomas von Aquin, Innsbruck: Verlag Felizian Rauch. Spirit in the World, translated by William V. Dych and revised by J. B. Metz, New York: Herder and Herder (1968). Rahner, K. (1941), Hörer des Wortes: Zur Grundlegung einer Religionsphilosophie, München: Verlag Kösel-Pustet. Hearers of the Word, translated by Michael Richards and revised by J. B. Metz, New York: Herder and Herder (1969). Reese, P. N. (2016), A Brief Note on how to Understand the River Forest School of Thomism, Draft. URL = . Stapleford, S. (2008), Kants’s Transcendental Arguments, New York: Continuum International Publishing Group. Stępień, A. (1962), “Stanowisko Gilsona w Sprawie Metody Teorii Poznania” [“Gilson’s Position on the Method of Theory of Knowledge”], in Roczniki Filozoficzne: 10(1), 135–159. Streeter, C. M. (2016), Lonergan as Aquinas Scholar: General Empirical Method and Vatican II’s Interfaith Dialogue Challenges. https://www.aquinas.edu/sites/default/files/Streeter Swartz, N. P. (2009), “Rosmini: Thomist or Neo-Thomist? The Debate Continues”, in Journal of Politics and Law: 2, 120–131. Sweet, W. (2013), “Jacques Maritain”, in The Stanford Encyclopedia of Philosophy, edited by E. N. Zalta. URL = .

204 | Mirosław Szatkowski Szymaniak, A. (2007), “Bibliografia prac Mieczysława A. Krąpca OP oraz opracowania dotyczące jego osoby, poglądów i twórczości” [“Publications by and about prof. Mieczysław Albert Krąpiec”], in Człowiek w Kulturze [Man in Culture]: 19, 11–64. Weisheipl, J. A. (1961), “Introduction: The Dignity of Science”, in The Dignity of Science, edited by J. A. Weisheipl, Washington: Thomist Press, xvii–xxxiii.

William F. Vallicella

Does God Exist Because He Ought To Exist? Anselm of Canterbury (1033 – 1109) had a profound insight: he realized that God, understood as “that than which no greater can be conceived”, must exist of metaphysical necessity if he exists at all. To appreciate it properly we must distinguish between Anselm’s Insight and Anselm’s Argument, where the latter is the modal argument of Proslogion III. The Insight may be put as follows. God, by definition, is an ens perfectissimum, a maximally perfect being. A maximally perfect being, however, cannot be modally contingent, but must be modally noncontingent: it must be necessary (existent in every metaphysically possible world), or else impossible (existent in no metaphysically possible world). For if God were modally contingent (existent in some but not all metaphysically possible worlds), then a greater could be conceived, namely, one that exists in all worlds. I will take it for granted that a being worthy of worship, one than which no greater can be conceived, must have the modal status of necessity. God is after all a candidate for the office of Absolute, and surely no such candidate could merely happen to exist. A contingent Absolute would be no Absolute at all. The Insight, then, consists in the realization that the alternative that God faces is not contingent existence versus contingent nonexistence, but necessity versus impossibility. It is clear, however, that the Insight, by itself, is not a compelling reason to accept the existence of God. For the Insight is not that God necessarily exists, but that God either necessarily exists or is impossible. Equivalently, Anselm’s Insight is that if God is so much as possible, then God actually exists. Obviously, the truth of this conditional is consistent with God’s not being possible. The Insight is an insight into the divine modal status, not into the divine existence, and so leaves open the question whether God exists. The Argument, building on the Insight, proceeds: It is possible that there be a maximally perfect being. (There is at least one possible world in which God exists.) Therefore, God exists in every possible world, whence it follows that he exists in the actual world. Briefly, if God is possible, then God is actual. God is possible, therefore God is actual. The Insight is unexceptionable, or so I would maintain, but the same scarcely holds for the Argument. The main problem is to give a good reason for thinking that God is possible. The fact that one can conceive of a maximally perfect being without contradiction does not establish that such a being is possible in reality. Conceivability (thinkability without contradiction) is no sure guide to real, extramental, possibility. Given the finitude of our minds, what is conceivable to us may be impossible in reality. The following modal ontological argument, then, though https://doi.org/9783110566512-012

206 | William F. Vallicella valid, and perhaps sound, is not probative unless we can supply a good reason for supposing that a maximally perfect being is really or extramentally possible, as opposed to being merely thinkable by us without apparent contradiction: (1) If a maximally perfect being is possible, then it is actual. (2) A maximally perfect being is possible. Therefore (3) A maximally perfect being is actual.

What we need is an argument for (2). Here is one: (4) A maximally perfect being ought to exist. (5) Whatever ought to exist, is possible. Therefore (2) A maximally perfect being is possible.

This second argument, like the first, is valid in point of logical form, and the premises are plausible. If a being is maximally perfect, then it is presumably deontically perfect and so ought to exist. To deny this is to say that a being can be both perfect in every respect but also either such that it ought not exist (which would be absurd) or such that it neither ought to exist, nor ought not exist. Either way, a greater can be conceived, namely a being that ought to exist. The other premise, (5), is also plausible. If you were to deny it you would be saying that there are things or states of affairs that both ought to exist and are impossible.Now putting the deontic subargument and the ontological subargument together we get a deontically supercharged modal ontological argument that is not only valid but appears probative: (4) A maximally perfect being ought to exist. (5) Whatever ought to exist, is possible. (1) If a maximally perfect being is possible, then it is actual. Therefore (3) A maximally perfect being is actual.

So a maximally perfect being exists. Whether this is “what all men call God” (to borrow the phrase with which Aquinas ends each of his quinque viae) is a further question well beyond the scope of this article. Some, like Tertullian, question what Athens has to do with Jerusalem, while others, like Pascal, question whether the God of the philosophers is the God of Abraham, Isaac, and Jacob. Be this as it may. If nothing else, the above argument seems at least to prove the existence of a maximally perfect being. Whether this is merely a God of the philosophers cannot be discussed here. (The contemporary sources of the above argument are in Ewing

Does God Exist Because He Ought To Exist? |

207

(1973), ch. 7; Findlay (1970), pp. 98 ff.; Kordig (1981), pp. 207–208; Leslie (1989), ch. 8; Mackie (1982), ch. 13.) Premise (1) is about as solid as anything in philosophy. Premises (4) and (5), however, are somewhat less luminous to the intellect. The following two sections will say something in defense of these premises.

1 Can we speak of ‘oughts’ in non-agential contexts? One way of resisting the argument is by questioning the propriety of talk about non-epistemic ‘oughts’ in contexts in which agency and moral obligation are not relevant. For example, what could it mean to say that God, classically defined as a being who realizes all perfections, ought to exist? The ‘ought’ here is not to be taken epistemically: The idea is not that the existence of God is rendered certain or probable by anything we know. The ‘ought’ is to be read ontically despite the fact that God is under no moral obligation to bring himself into existence, or keep himself in existence: if God is a necessary being, then he cannot come into existence or pass out of existence. And surely no non-divine agent could be under any obligation to bring God into existence, maintain him in existence, or refrain from killing him. So a quick way of countering the above argument is by claiming that locutions of the form ‘X ought to exist’ and ‘X ought not exist’ are meaningless apart from contexts in which the oughts-to-exist supervene on oughts-to-do. The critic will concede that there are states of affairs that ought to be. But he will insist that for each ought-to-be there is an ought-to-do that underpins it. He will insist that every state of affairs that ought to be or ought not to be necessarily involves an agent with power sufficient to either bring about or prevent the state of affairs in question. Thus it ought to be that one feeds one’s children, but this ought-to-be supervenes upon an ought-to-do. It may not be possible to prove definitively that there are non-agential oughts, but their postulation is in line with ordinary ways of thinking and talking and there seem to be no decisive arguments against their postulation. Consider a possible world W in which there are no moral agents, but there are sentient beings who are in a constant state of pain from which they cannot free themselves. It seems both meaningful and reasonable to say that W ought not exist, that its nonexistence is an axiological requirement. And this quite apart from the power of any agent to actualize or prevent such a world. One simply intuits the disvalue of such a world. One might express the intuition in the words, ‘Such a world ought not be’.

208 | William F. Vallicella Non-agential oughts are axiologically required, while non-agential oughts-not are axiologically prohibited. Or consider our world, the actual world, with its nature red in tooth and claw, a world in which life lives at the expense of life. It is filled with vast quantities of natural and moral evil. Assume that naturalism is true, that there is no God or afterlife, and that the evils of this world will forever go unredeemed. It may be false, but it seems meaningful to say it would be better if this world did not exist, that it ought never to have existed. The metaphysical pessimist may be wrong, but he is not talking nonsense when he exclaims, “Better some other world or even nothing at all rather than this sorry state of things!” On the other hand, there are those who are struck by the sheer existence of things and are moved to exclaim, “It is good that there is something rather than nothing!” Such optimists are not talking nonsense when they say that things are as they ought to be even in the absence of any agent or agents who are responsible for things being as they are. The sense of these exclamations does not seem to depend on the existence of moral agents with power sufficient to bring about or prevent the mentioned states of affairs. That something rather than nothing exists could be good even if it is no one’s duty to bring it about and no one’s responsibility if it obtains. That a world of uncompensated and unalleviated misery is bad does not depend on some free agent’s moral failure.

2 Does ‘ought’ imply ‘can’ in non-agential contexts? But even if there are non-agential oughts-to-be, so that we can speak meaningfully of God’s oughtness-to-be, how does this secure the real possibility of the divine existence? It is usually admitted that ‘ought’ implies ‘can’ in agential contexts. Thus, if an agent ought to do X, then he can do X, where ‘can’ is interpreted in terms of ability. The idea seems correct. If I ought to feed my cat, if that is one of the things I am morally obliged to do, then it must be not only logically and nomologically possible for me to feed her, but I must also have the ability to feed her. If on a given day I am prevented from feeding her by no fault of my own, and also prevented from calling for assistance, then I cannot be held responsible for her not being fed on that day. My inability to perform an action, either in general or in some specific circumstances, absolves me of moral responsibility for failure to perform the action. Of course, qualifications would have to be added to make the preceding sentence logically ‘air-tight’. For example, if I can’t do my duty because I

Does God Exist Because He Ought To Exist? |

209

have allowed myself to become a heroin addict, this self-induced inability doesn’t absolve me from my duty. The question arises whether we can extend the ‘ought’ implies ‘can’ principle to what ought to exist in non-agential contexts. I argued earlier that ‘ought to exist’ and ‘ought not exist’ are predicates that can be applied meaningfully to states of affairs whose obtaining/nonobtaining is not traceable to any agent. Thus the existence of God (classically defined as ens perfectissimum) is not up to God or any agent, and yet it seems to be meaningful to claim that the state of affairs of God’s existence ought to exist or obtain, while the nonexistence of God ought not exist or obtain. If one insists on reserving ‘ought’ for the agential cases, then we could speak of the existence of God being axiologically required. Suppose you agree that the existence of God non-agentially ought to be, or is axiologically required. Does it follow that God’s existence is possible? In the agential case, possibility is interpreted as ability. If I ought to do A, then I can, am able, to do A. But abilities are the abilities of agents, and in the non-agential case there are no agents. So if God ought to exist, and “Whatever ought to exist can exist”, then ‘can’ in this formula cannot be cashed out in terms of ability. But this is not a problem since ‘can’ can be read in terms of metaphysical (broadly logical) possibility. Accordingly, whatever ought to exist is metaphysically-possibly such that it exists. What we have, then, are two analogically related ‘ought’ implies ‘can’ principles. Agential Principle: What an agent ought to do, an agent must be able to do. Non-agential Principle: What ought to exist, must be metaphysically possible.

Both can be classified as ‘ought’ implies ‘can’ principles, but while in the agential case the ‘ought’ is an ‘ought to do’ and the ‘can’ signifies an agent’s ability, in the non-agential case the ‘ought’ is an ‘ought to be’ and the ‘can’ signifies metaphysical possibility. By my lights, both principles are true, and indeed analytically true. If you tell me that I am under a moral obligation to do X even in circumstances in which it impossible for me to do X, then I will respond that your view is incoherent and that you do not understand the relevant concepts. Similarly, if you agree that there are non-agential contexts in which some state of affairs S ought to exist even though S is impossible, then I will respond that your view is incoherent and that you do not understand the relevant concepts. There is no sense in which what cannot exist ought to exist or is axiologically required. I note in passing that Nicolai Hartmann seems to demur. “Because something is in itself a value, it does not follow that someone ought to do it; it does mean, however, that it Ought to ‘Be’, and unconditionally – irrespective of its actuality or even its possibility” (Hartmann (1932), pp.

210 | William F. Vallicella 247–248, emphasis added). To explain why Hartmann thinks that what he calls the pure or ideal ought-to-be is what it is regardless of its possibility would require a lengthy excursus into his curious modal doctrine.

3 Is the argument circular? God must exist of modal necessity if he exists at all. But this Anselmian Insight leaves open the possibility that God does not exist. For the Insight is an insight into the divine modal status, not into the divine existence. The Insight reveals the divine noncontingency, which is compatible with both the existence and the nonexistence of God. To show that God exists, one must show that God is possible in reality. At this point the broadly deontic considerations lately mentioned come into play. It seems undeniable that whatever ought to exist is metaphysically (broadly logically) possible. So whether or not the argument is probative comes down to the question whether it is self-evidently true that a maximally perfect being ought to exist. Suppose one concedes that the concept of a maximally perfect being is the concept of a being that ought to exist. This concession, however, leaves us with the question whether the concept is instantiated. It is this question that our argument cannot answer, or cannot answer compellingly. A maximally perfect being ought to exist – but only if it exists. For if it does not exist, then it is impossible and therefore (by the contrapositive of the non-agential ‘ought’ implies ‘can’ principle) not such that it ought to exist. Thus the suspicion arises that the deontically supercharged modal ontological argument is no more able to prove the existence of God that the plain old modal ontological argument. Both arguments appear circular or question-begging. Since a maximally perfect being is possible if and only if it is actual, to know that the possibility premise of the modal ontological argument is true one must already know that the conclusion is true. But then the argument begs the question. The same goes for the composite argument under consideration in this paper. Since a maximally perfect being ought to exist if and only if it is possible, and since it is possible if and only if it is actual, it follows that a maximally perfect being ought to exist if and only if it is actual. But this is just to say that the argument begs the question.

Does God Exist Because He Ought To Exist? |

211

4 The price of avoiding the circle I can imagine an objector who accuses me of failing to grasp the thrust of the argument: You are missing the whole Neoplatonic point of the argument and your circularity objection is wide of the mark: the oughtness-to-exist of a maximally perfect being is what conjures it into existence. On Neoplatonism, “ethical needs for the existence of things are in some cases creatively effective” (Leslie (1989), p. 165). This oughtness-to-exist is independent of the existence of anything, including a maximally perfect being. God’s oughtness-to-exist is not a property of God that presupposes the existence of God. Furthermore, it itself does not exist, being epekeina tes ousias, “on the far side of being” (Mackie (1982), p. 231). And so the circularity objection you bring against the modal ontological argument cannot be brought against the deontically supplemented modal ontological argument. The first argument is circular because we have no reason to accept the possibility premise apart from a prior acceptation of its conclusion. But the central premise of the second argument – A maximally perfect being ought to exist – can be known to be true apart from knowledge of the truth of the conclusion.

This objection that I have concocted brings out the strength of the composite argument. It is an argument that cannot be dismissed out of hand. Being a theist, I should like it to be a compelling ‘knock-down’ proof. But I am afraid that it isn’t, even if one accepts that (i) there are objectively prescriptive values, and thus that both ethical naturalism and non-cognitivism are false; that (ii) there are nonagential values and oughts; (iii) that there is a non-agential ‘ought’ implies ‘can’ principle. The main difficulty, as I see it, is the notion that there are non-agential oughts or axiological requirements that are “creatively effective” as Leslie puts it, but do not exist. It seems reasonable to protest that what does not exist is just nothing and so cannot be creatively effective or anything else. The price of avoiding circularity is to accept that there are creative axiological requirements that are beyond Being like Plato’s Form of the Good. Thoughts that enter this dimension taper off into mysticism and leave the discursive precincts of philosophy behind. So although the argument under examination elevates our thoughts into the region of the transdiscursive, and in so doing raises a number of fascinating issues, it cannot count as a proof in any reasonably strict sense of the term.

212 | William F. Vallicella

Bibliography Ewing, A. C. (1973), Value and Reality, London: Allen and Unwin. Findlay, J. N. (1970), “Some Reflections on Necessary Existence”, Ascent to the Absolute, London: Allen and Unwin. Hartmann, N. (1932), Ethics, vol. I, translated by S. C. Coit, London: Allen and Unwin. Kordig, C. R. (1981), “A Deontic Argument for God’s Existence”, in Nous: 15(2), 207–208. Leslie, J. (1989), Universes, London: Routedge. Mackie, J. L. (1982), The Miracle of Theism, Oxford: Oxford University Press.

Peter van Inwagen

God’s Being and Ours It is a popular view that the being of God and the being of creatures are in some important way different. I want to discuss this idea. I write as an analytical philosopher whose special study is ontology and who considers himself an orthodox Christian. I will begin with a discussion of ontology. Ontology is the part of philosophy – the part of metaphysics – that investigates the nature of being or existence. (In this essay, I will assume that ‘being’ and ‘existence’ are essentially synonyms. This is a controversial assumption, but if I tried to say something about every disputed question that was relevant to my topic I should not get very far.) One of the central questions of ontology is whether there are “modes” of being or of existence: whether there could be two or more objects all of which exist but (somehow) exist in different ways. I will give an example that illustrates the considerations that have led some philosophers to embrace the idea of modes of being. Imagine, first, an archaeologist whose attempts to discover the site and physical remains of the lost city of Imram-hut have so far been an abject failure. His wife, who is not an archaeologist, asks him whether he is sure what he is seeking exists; how, she asks him, can he be sure that Imram-hut ever existed? He replies, “Oh, there’s no doubt that Imram-hut was a real city. There are lots of documents and inscriptions that give detailed and more or less consistent descriptions of its layout; and linguistic analysis proves that many of them are wholly independent of one another. That would be vastly improbable if Imram-hut had never existed. And if it existed, there will certainly be some sort of physical traces of it at its site.” Imagine, secondly, a mathematician who is attempting to solve a recalcitrant differential equation called Rostov’s Equation. Her husband, who gratefully abandoned the study of mathematics as soon as he had scraped through highschool algebra, asks her why she is so sure that the equation has any solutions. She replies, “Oh, I know that solutions to Rostov’s Equation exist. That’s an immediate consequence of the Peano existence theorem, which says that if a differential equation satisfies certain conditions, solutions to that equation exist; and it’s easy to show that Rostov’s Equation satisfies those conditions.” We may imagine that both these savants are right: Imram-hut exists, or once did, and solutions to Rostov’s Equation exist: what each of the two is trying to find is, in some sense, “there” – there to be found. But can the existence of a physical object like the ruins of an ancient city and the existence a mathematical object like a solution to an equation be the same sort of thing? Many philosophers (and many mathematicians) have contended that the “existence” of the city and the https://doi.org/9783110566512-013

214 | Peter van Inwagen “existence” of the solution must be vastly different kinds of existence – in the jargon of metaphysics, that the “mode” of existence of the city cannot be the “mode” of existence of the solution. Some philosophers who take this line make the distinction of “modes” a distinction of meaning. We may, for example, imagine a philosopher, Dr Sinn, who tells us, “When we apply the word ‘existence’ to a physical object like a ruined city, it means ‘occupies some region of space-time’. But when we apply it to a mathematical object like a solution to an equation, it means ‘has a description that is free from contradiction’.” Dr Sinn’s colleague, Professor Bedeutung, however, defends a similar but subtly different position. “The word ‘existence’ means the same thing no matter what kind of object one applies it to,” she says, “but it nevertheless refers to different things when one applies it to objects of radically different kinds. The existence of Imram-hut is that city’s occupation of a certain region of space-time, and the existence of a solution to Rostov’s equation is the freedom from contradiction of the description ‘solution to Rostov’s equation’. The phrase ‘existence of’ can be usefully compared to certain mathematical operators – ‘square root of’, for example. The words ‘square root of’ mean the same thing when you apply them to positive numbers and to negative numbers, but the values of the square-root function for positive and negative arguments are different sorts of thing: real numbers in the one case, and imaginary numbers in the other. These different kinds of thing that ‘the existence of the city’ and ‘the existence of the solution’ refer to are illustrations of what we metaphysicians call ‘different modes of being’.” For my part, I must confess that I don’t find Dr Sinn’s position at all plausible. It seems obvious to me that the word ‘exists’ means the same thing when you apply it to physical and to and to mathematical objects – and to anything else. As for Professor Bedeutung, I think she’s talking nonsense. At any rate, I don’t understand the words I’ve put into her mouth. (But I didn’t invent this nonsense: the words I’ve foisted on my helpless creature of fiction are perfectly representative of the kind of language some philosophers use.) Neverthless, the theses I’ve ascribed to my imaginary philosophers are relevant to the question of the relation between ontology and the existence of God, because it has been a popular opinion among metaphysicians and metaphysically minded theologians that God and creatures enjoy different modes of being. God and St Peter both “exist,” say these authorities; they both “have being,” but the being of God and the being of St Peter are very different kinds of being – although they are, some of them would say, united by analogy. (Perhaps some examples involving other concepts than being will help to explain what philosophers mean when they say that applications of the same word to radically different kinds of things are “united by analogy”: The average global temperature is rising and the

God’s Being and Ours |

215

hot-air balloon is rising, but the kind of rising done by the temperature and the kind of rising done by the balloon are clean different kinds of thing – although “united by analogy”; My liver is a part of me, and the third movement of the Eroica is a part of that symphony, but the parthood relations in the two cases are distinct relations – albeit united by analogy.) As one might expect from what I said about the philosophies of Dr Sinn and Professor Bedeutung, I can boast no real understanding of the thesis that God’s being is a different kind of being from the being of creatures, despite the fact that it has won the assent of great doctors of the Church and great philosophers outside the Church. It cannot be that the words ‘exist’ and ‘be’ mean one thing when they are applied to God and another thing when they apply to creatures. I will defend this idea by fashioning an “intutition pump” – that is a story, a sort of parable, that is intended to make this the thesis that ‘exists’ applies “univocally” to God and creatures intuitively plausible. A band of nineteenth-century revolutionaries ride into a little Mexican village with the intention of stripping its church of its gold and silver images and vessels. The find the church empty of precious metals. Their leader, Captain Gonzalez, asks the priest what has become of them. The priest tells him, “The people knew you were robbing churches, and they have taken the holy things away and hidden them in the hills. Naturally, they did not tell me where.” Gonzalez replies, “Well, they had better take me to them.” “They will not do that,” the priest tells him, “because they fear God more than you.” Gonzalez answers, “They will soon learn to fear me more than God, for I exist and God does not.” If ‘exist’ did not mean the same thing when applied to the Deity and to Gonzalez, the latter’s last statement would be a syllepsis. A syllepsis is a rhetorical figure, generally employed for humorous effect, in which a single occurrence of a word, in virtue of its syntactical relations with other elements of the sentence in which it occurs, must be understood in two senses simultaneously. The classic example is Dickens’s “Miss Bolo went home ... in a flood of tears and a sedan chair.” One may indeed go home in a flood of tears, and one may indeed go home in a sedan chair, but not in the same sense of ‘go home in’. If God exists (or does not exist) and Gonzalez exists, but in different senses of ‘exist’, then ‘I exist and God does not’ ought to produce in those who encounter it the same sense of – let us call it – semantical dislocation that is produced by ‘She went home in a flood of tears and a sedan chair’ or ‘Last summer, I lost my passport and my love of foreign

216 | Peter van Inwagen travel.” But – in my own case at any rate – Gonzalez’s statement produces no such sense of dislocation.¹ Might it then be that, although ‘exists’ and ‘there is’ and other such existential idioms apply to God and creatures univocally, God and creatures enjoy or participate in different modes of being? If I am told that I should ascribe different modes of being to two objects (a city and a solution to an equation; God and St Peter) that statement seems to require that I be able to treat the existence or being of a thing (on the one hand) and its essence or nature (on the other) as in some sense different aspects of that thing, and to go on to maintain that some of the – what word shall I use? – features of the thing are grounded in the kind of existence it has and others in the kind of nature it has. Now I think we can all agree that God’s nature (divinity) and Peter’s nature (humanity: what is common to all members of Homo sapiens) are vastly different – perhaps even as different as two rational natures could possibly be. What I can’t make any sense of is the idea that there’s something more than this to be said about how God and Peter differ – that there’s some further thesis about how they differ, a thesis that is properly expressed by the words, ‘The kind of being God has is different from the kind of being Peter has’. I deny that God and Peter enjoy different modes of being because I think that there is only one mode of being for anything to have. And if that is true, the word “mode” has no point: one should speak simply of being, of the one thing called being. (In this sentence, I use the word ‘thing’ as a mere grammatical convenience: I do not mean to imply that ‘being’ is the name of a thing in the way in which ‘Catherine the Great’ or ‘the Taj Mahal’ or the ‘the cube root of 2’ are names of things. I do not mean to deny the reality of what Heidegger calls the Ontologische Differenz: “Das Sein des Seienden ‘ist’ nicht selbst ein Seiendes” – “The being of the being ‘is’ not itself a being.” (Treat the second ‘the being’ as you would treat ‘the lion’ in ‘The ferociousness of the lion is proverbial’.) On that point Heidegger is certainly right.²

1 My intuition pump presupposes that when, for example, the Fool says, “Deus non est” and St Anselm says, “Deus est”, each is expressing a proposition that is the contradictory of the proposition that the other expresses. I concede that this presupposition could be disputed. 2 At least on my understanding of these words. I take Heidegger’s neologism ‘ein Seiendes’ to mean what an analytical philosopher like myself would call ‘a concrete being’ – like Catherine and the Taj and unlike the cube root of 2. (In my view, ‘being’ is the name of an abstract thing – the name of (if Kant and Frege will forgive me) an attribute. I do not mean to imply that Heidegger would accept that thesis. I would therefore say that the being of any Seinedes is identical with the being of any other Seinedes: the being of Catherine is identical with the being of the Taj, a position that is radically inconsistent with Heidegger’s.) I would also dispute Heidegger’s

God’s Being and Ours |

217

To my mind, the concept of being or existence has been best explicated in a simple but profound remark of Frege’s: “Es ist ja Bejahung der Existenz nichts Anderes als Verneinung der Nullzahl” – “Affirmation of existence is indeed nothing other than denial of the number zero.” (I concede, however, that when I say that affirmation of existence is nothing other than denial of the number zero, what I mean by that statement is subtly different from what Frege meant by those words.) If my account of the concept of being is correct, to say that God exists is to say that the number of Gods – the number of things that are God – is not zero. And to say that God does not exist – to affirm atheism – is to say simply that the number of things that are God is zero. And of course, to say that St Peter never existed is to say that the number of things that are or have been St Peter is zero – and to dispute that historical contention is to say that the number of things that are or have been St Peter is not zero. There is, therefore, only one kind of being – since to say ‘x is’ or ‘x exists’ is simply to say that the number of things that are x is not zero. If you like, therefore, you can say that God and Peter have only one kind of being – for there is only one “kind” of being for anything to have. A parenthetical remark. In saying that God and Peter have the same kind of being, I do not mean to deny that God’s being is necessary or is a consequence of his nature, and I don’t mean to deny that Peter’s being is contingent and is not a consequence of his nature.³ But it does not follow from the necessity of God’s being and the contingency of Peter’s being that ‘being’ refers to one thing when we apply the word to Peter and another thing when we apply it to his Creator. God, I would say, has “being” of necessity, and Peter has “being” only contingently – but it’s one and the same thing that God has of necessity and Peter has only contingently. A familiar proposition of Roman Catholic theology may serve as an

implicit contention that there is a reason (that there is any possible reason) to place ‘ist’ inside “scare quotes”. He does this because he rejects the philosophical commonplace that “the ‘is’ of predication” is not “the ‘is’ of being or existence”. 3 Catholic philosophers have often said not that God’s existence is a consequence of his nature but that his existence and his nature are identical. This doctrine is one of the many implications of the more general “doctrine of Divine Simplicity”, according to which phrases like ‘God’s power’, ‘God’s wisdom’, ‘God’s love’, ‘God’s nature’ and ‘God’s existence’ all denote one and the same thing, namely the Divine Substance – that is, God, God himself, God full stop. The doctrine of Divine Simplicity, however, presupposes an Aristotelian ontology of substance and attribute (for present purposes, “Aristotelianism”). From the point of view of a Platonist like myself, the doctrine of Divine Simplicity is wrong simply because it presupposes Aristotelianism, and Aristotelianism is false. Here I can only note the existence of (and my Platonistic rejection of) the doctrine of Divine Simplicity. A discussion of the doctrine cannot be carried out within the scope of this essay – if only because such a discussion would presuppose a prior discussion of Platonism and Aristotelianism.

218 | Peter van Inwagen analogy: Jesus and his Mother were both without sin, but Jesus had the property “commits no sins” of necessity, by his very nature, whereas Mary had that property but had it only contingently: “commits no sins” was (and is) a consequence of Jesus’ nature and merely co-present with Mary’s; there is one thing, the property or attribute “commits no sins,” that one of the two had of necessity and the other had contingently. Although I think that the doctrine that God and creatures enjoy different modes of being rests on a metaphysical mistake, I do think that it was not inexplicable that this mistake should have been made. For the natures of God and creatures are vastly different – different to such a degree that I can find within myself some sympathy for those thinkers who have been so overwhelmed by the vastness of the gulf that separates the divine nature from any creaturely nature that they attempted to conceptualize that gulf in terms of distinct divine and creaturely modes of being. Let me try to give my own account of that gulf, an account that does not involve the idea of modes of being. I don’t see why I shouldn’t begin with a joke. I once saw a cartoon in which a backwoods preacher was shouting the following words from the pulpit: “His socks are as big as New Jersey and his tee-shirts are the size of Texas!” Now there is no doubt more than one theological point that one might want to make in response to that preacher’s doctrine, but perhaps the most fundamental point is this: the preacher is treating God as one of our fellow inhabitants of the physical universe or cosmos – as a being that is in that respect like Zeus or some vastly powerful and intelligent science – fictional being from the galactic core. And, of course, any such being would have its origin within the cosmos, and thus could not be its creator. Zeus has an “origin story” in exactly the same sense as any comic-book superhero, and – according to that story – there are mountains that are older than he is. Any rational extra-terrestrial being, however much more ancient and powerful and intelligent it may be than we, must be a product of some set of physical processes that went on somewhere in the cosmos and must thus be younger than that cosmos: it could no more be responsible for the existence of the physical universe than an elephant could be responsible for the existence of terrestrial life. If God is the creator of the universe, he must not be one of its inhabitants. He must be in some sense outside the universe. (And, I hasten to add, there must be another sense in which he is not outside the universe: God is not remote from us; he is, in the words of the Koran, closer to you than the pulse in your throat.) But in what sense is he outside the universe? He could not be spatially outside the universe, for nothing can be spatially outside the universe: the universe includes all the space there is – and always has, at every time, included all the space there was at that time. When we try to give a clear sense to the idea that God is outside

God’s Being and Ours |

219

the universe, we run up against the inherent limitations of our language, which is (if I may use the word) designed to enable us to talk about and deal with things that – like us – are parts of the physical universe. That fact doesn’t mean that we can’t possibly talk about things that are not parts of the physical universe, for the fact that a tool wasn’t designed for a certain purpose doesn’t mean that it’s impossible to use it for that purpose. It is possible, after all, use an axe to drive nails or a screwdriver to pry open a can of paint. But it does mean that when we try to speak of the way in which God is “outside the universe”, we must be sensitive to the immense difficulties that necessarily confront any such attempt. When we try to talk about God’s relation to Creation or to the physical universe (which may or may not be the same thing: everything physical is created, but there may be created things – souls, angels – that are not physical), we have only two resources: abstraction and analogy. I choose the way of analogy. One analogy many writers have used is along these lines: God is to the universe as the author of a work of fiction is to the “fictional world” of that work – the imaginary world inhabited by the characters the author has created. One who makes use of this analogy will find that there are some intuitive advantages to taking the “work of fiction” in question to be an animated cartoon rather than a novel or play. If we make that choice, we may frame the analogy this way: God is to the physical world as Walt Disney is to the “world” that is displayed on the screen in Snow White and the Seven Dwarfs (let’s pretend that Walt was the only person involved in the production of Snow White – that he drew and colored every “cel” himself). If we watch Snow White, we shall notice that Walt never appears on the screen: he is not a part of the story. And, in that sense, he is “outside” the world of Snow White. And yet he is intimately related to that world: he is its creator; everything in that world is the way it is because it was his will that it should be that way. That is the useful part of the analogy. But the analogy, however useful it may be in part, has many defects. (As I said: we must be sensitive to the immense difficulties that necessarily confront any attempt to speak of God’s relation to the created world.) And one of these defects, surely, is more important and more obvious than any other: the world of a work of fiction, whether it is a novel or an animated cartoon is, well, fictional: Snow White and the Wicked Queen and the dwarfs do not exist and never have. The screen that is before one as one watches Snow White is not a window into a world of real things, but simply a surface that displays rapidly changing static pictures none of which corresponds to anything in reality. But let us try to pretend – per impossibile – that the world of Snow White, like the physical universe that God has created, is a real world, that it contains real, conscious persons, persons who experience things like dwarfs and cottages and poisoned apples, things that are spread out in three-dimensional space and are in continual causal interaction with one another.

220 | Peter van Inwagen The inhabitants of this world – if we grant that they could know anything at all – could know about Walt. (If he chose, he could confer knowledge of his own existence on some or all of the characters in his story – as easily as he could confer any other property on them. Although Walt was not a part of the story of Snow White, he could have been if he had chosen. He could have written himself into the story.) And what would they conclude about the way in which he was related to them and to the objects in their environment? Well, they would be wrong if they concluded that they could visit him if they chose a direction and walked far enough in that direction. In one sense, he is – from their point of view – nowhere. In another sense, he is everywhere. He is not to be found anywhere in what they call ‘space’, but everything that is anywhere in what they call ‘space’ is where it is and has the properties it does because Walt has made things so by an exercise of his creative power. This idea – in application to God and the real world – was a commonplace among the medieval philosophers. In their jargon, Walt is – in relation to the world inhabited by the characters in Snow White – totally present everywhere and locally present nowhere. Although the occupants of every region of space in the world of Snow White reflect the totality of his being – as Snow White’s creator, he is closer to her than the pulse in her throat – he occupies no region of space (not even all space: he is not spread out through all space as the luminiferous ether was supposed to be). Now let me improve the picture I am painting a bit. Let’s first suppose that the story-line of the animated cartoon I am imagining really is the story of a whole world, that this world has not got any narrative center (that is, the events of a vast region of space are represented not on separate two-dimensional cels intended to produce the illusion of looking through a window at the events occurring at one place among all the places in a much larger world but in the successive states of some vast three-dimensional arena). Let’s call the animator we are imagining Hyper-Walt. And let us suppose that Hyper-Walt proposes to represent billions of years of “story” the successive states of the arena. And let us suppose that the representations are to be very detailed: that part of the story of the world Hyper-Walt is creating is a physics, and that he proposes to be so scrupulous as to represent the successive states of every constituent elementary particle of the world he is “creating”. He has invented a set of physical laws and proposes to draw the mutual movements of the “particles” in his world just as they would be if they were entirely due to those laws. (Of course, he’s a lot better at this sort of thing than any human being could ever hope to be. We human beings may not have a complete grasp of the laws of physics, but – so I understand – we’re pretty sure that in principle there’s nothing relevant to how a protein folds but Maxwell’s equations and the “external” electrical properties of its constituent amino acid molecules. But if we try to use a computer to simulate the folding of a protein of any complexity

God’s Being and Ours |

221

from “first principles”, we fail: the needed computing power vastly exceeds any we or our descendants could possibly have access to.) In one sense this is all Hyper-Walt does: all he does is to trace the trajectories of the elementary particles as those trajectories evolve in accordance with the laws he has formulated. It is, however, true that in whatever sense there are elementary particles in Hyper-Walt’s world, there are also larger things, things composed of elementary particles; there are also – let’s suppose – atoms and molecules and bacteria and grains of sand and elephants and mountains and planets and stars and galaxies. All these “composite objects” are generated automatically, as it were, by what Hyper-Walt is doing with the many individual elementary particles – since, as present-day analytical philosophers like to say, everything that happens in the physical world “supervenes on” the properties of, arrangement of, and mutual interactions among, elementary particles. That is to say: if some factor settles the properties of, arrangement of, and mutual interactions among, all the elementary particles, that factor will thereby, without having to do anything further, settle the properties of, arrangement of, and mutual interactions among all the atoms and molecules and bacteria and grains of sand and elephants and mountains and planets and stars and galaxies. This is a lot of supposing, but I am afraid that we have more supposing to do. Let us next suppose that Hyper-Walt’s world contains few if any miracles: that is, although Hyper-Walt can “draw” elementary particles whose behavior does not conform to the laws he has formulated as easily as he can draw elementary particles whose behavior does conform to those laws, he has not done this or has done so very rarely – so rarely that if one were to examine a large number of small regions in the space-time of Hyper-Walt’s world at random, it would be vastly unlikely that one would observe a departure from the laws in any of them. Let us suppose, moreover, that it is a part of the story-line of Hyper-Walt’s world that it contains scientists – physicists, cosmologists, chemists, biologists, paleontologists, geologists, ... . The physicists, if they are good at their job (and if Hyper-Walt, like Einstein’s God, is not malicious, however subtle he may be) will eventually discover the laws Hyper-Walt has “built into” his world – or if not those laws in all their subtlety, at any rate a good model of some significant portion of them. And the cosmologists can hope to discover something of the history of their cosmos. (Perhaps they will discover that it evolved out of an initial singularity. Hyper-Walt may have “drawn” the trajectories of the elementary particles in such a way that they emerge from a space-time singularity, and if he has, the cosmologists may be able to discover this.) The same thing may be true, mutatis mutandis, of the evolution of life. It may be, for example, that the biologists and paleontologists are able to give a good account of the evolution of life by postu-

222 | Peter van Inwagen lating that a gradual accumulation of genetic advantages has taken place within the constraints of selection pressure. Let us suppose, finally, that a lot of the people in Hyper-Walt’s world believe that that world is the creation of an intelligent being, an omnipresent being who did not come to be within it and who created it out of nothing and responsible for the existence and properties of everything it contains. The sources of this belief – creationism, let us call it – are ancient and traditional and long antedate that world’s analogues of Copernicus and Galileo and Newton. Now in Hyper-Walt’s world, creationism is of course true, but that doesn’t mean that the creationists of that world are to be praised for believing it. Perhaps they’re just right by accident, and a truly rational inhabitant of their world would suspend judgment about creationism or even affirm its falsity. (The physicist and cosmologist Sean Carroll has told me that in his view any rational inhabitant of Hyper-Walt’s world should believe that that world had no creator and existed on its own.) The question I want to raise is this: Does their science have anything to tell the inhabitants of HyperWalt’s world about the truth or falsity of creationism? We are supposing that the discoveries of science in Hyper-Walt’s world are comparable in scope and importance to those of our world, and science may therefore be presumed to have both proved both the existence and the non-existence of many things. Might the scientists of our imaginary world discover something proves that creationism is true or at least significantly raises its probability? Or might they discover something that seems to prove that creationism is false or significantly lowers its probability? I think that the answer to these questions has to be No. Everything the scientists can discover pertains to things that are locally present in their world; their discoveries have nothing to tell their fellow inhabitants of Hyper-Walt’s world about either the existence or non-existence of a being that, if it exists, is omnipresent. But this little argument is very abstract – it makes no reference to the actual, concrete discoveries of those scientists – and arguments that that level of abstraction are never very convincing. If they have any value, it is generally as a kind of summing up of the salient features of a large set of concrete cases. And I have no such concrete cases to discuss, since Hyper-Walt’s World is a fiction, a world constructed to provide an analogical explanation of the concepts “local presence” and “omnipresence”. Being a fiction, it is incomplete; just as the Sherlock Holmes stories include only such information about Holmes as Conan Doyle chose to put into them,⁴ the story of Hyper-Walt’s world, as I have told it, contains almost nothing about the history of science in that world. To supplement

4 What was Holmes’s mother’s maiden name? Conan Doyle never said, so the question has no answer.

God’s Being and Ours |

223

my abstract argument, therefore, I should have to turn to the real world, a world in which a lot is known about the history of science. And there is no scope for a discussion of the history of science in an essay of this sort – nor am I qualified to make any statements other than the most obvious truisms about the history of science. Well, I hope that Hyper-Walt’s world has done the work I wanted it to do, and has provided some sort of analogical illumination of local presence and omnipresence. To sum up: the God of our real world, I maintain, is no more to be found within it than Hyper-Walt is to be found within his world; and yet he is not remote from its inhabitants. If the picture I have tried to paint of the Divine Nature is correct, then the Divine Nature is vastly – unimaginably – different from the nature of any possible creature. But that is only an abstract way of saying that God is vastly different from any possible creature. The fact that God is vastly different from any creature, I believe, explains the ontological mistake that many philosophers and theologians have made. Impressed, overwhelmed, by vastness of the gulf that separates God and creatures, they have tried to describe this overwhelming vastness by ascribing different modes of being to God and creatures. But the difference, the vast difference, between God and creatures needs no such explanation – which, in my view is fortunate, for (as I see matters) the idea of “modes of being” makes no sense. The difference is to be described as the difference between any two beings is to be described: in term of the different properties (qualities, attributes, features, characteristics, call them what you will) of those beings. In short, God’s properties are vastly different from ours. What God is like is vastly different from what we are like. When we have said that, we have not only said all that we need to say to express the difference between God and creatures, we have said all that can be said.

Authors of Contributed Papers Christopher Daly is Professor of Philosophy at the University of Manchester. His principal research interests are in metaphysics, philosophical methodology, and philosophy of language. He is the author of Introduction to Philosophical Methods (Broadview, 2010) and Philosophy of Language: An Introduction (Bloomsbury, 2012). He has published papers on natural kinds, properties, truthmaker theory, grounding, philosophy of mathematics, and fictionalism. He is an Associate Editor of the Australasian Journal of Philosophy. He and David Liggins are currently directing a research project entitled The Foundations of Ontology, funded by the UK Arts and Humanities Research Council. Gabriele De Anna is currently juniorprofessor of Philosophy at Bamberg University, in Germany. He taught Political Philosophy, Philosophy of Mind, and Philosophy of Science at the University of Udine, in Italy. He received two PhDs in Philosophy, one from the University of Padua (Italy) and one from the University of St Andrews (Scoland), after completing a Laurea degree (Padua) and Master of literature (St Andrews). He was Visiting Student at the University of Santa Barbara (USA), Visiting Fellow at the Centre for Philosophy of Science at the University of Pittsburgh (USA), Marie Curie Fellow at the Centre for Research in the Arts, Social Sciences and Humanities at the University of Cambridge (England), and First Chair of Philosophy at the University of Bamberg (Germany). He wrote over thirty articles in professional journals and collective volumes, and six monographs, including Scienza, Normatività, politica. La natura umana tra l’immagine scientifica e quella manifesta, Milan: Franco Angeli, 2012, Azione e Rappresentanza. Un problema “metafisico” del liberalismo contemporaneo, Naples: Edizioni Scientifiche Italiane, 2012; Causa, Forma, Rappresentazione. Una trattazione a partire da Tommaso d’Aquino, Milan: Franco Angeli, 2010. He has edited seven volumes, including Willing the Good. Empirical Challenges to the Explanation of Human Behaviour, Necastle: Cambridge Scholars Publishing, and Evolutionary Ethics and Contemporary Biology, Cambridge, Cambridge University Press, 2006 (with G. Boniolo). Michał Głowala is an assistant professor at the Institute of Philosophy, University of Wrocław. His research concerns mainly scholastic metaphysics (especially the metaphysics of powers and the metaphysics of action in late scholasticism) in the context of contemporary analytical metaphysics (in particular the metaphysics of powers). His recent publications include Singleness. Self-Individuation and Its Rejection in the Scholastic Debate on Principles of Individuation (De Gruyter 2016), https://doi.org/9783110566512-014

226 | Authors of Contributed Papers “Power Individuation: A New Version of the Single-Tracking View” (in Metaphysica, 16 (2015), issue 2), “Unity and Plurality in Joint Manifestations of Powers: A Scholastic Approach” (in Revista Portuguesa de Filosofia, 71 (2015), issue 4), “How do Powers Tend toward Their Manifestations? Three Thomistic Theories of the Prevention of Powers” (in Analytically Oriented Thomism, ed. M. Szatkowski (2016)). Christian Kanzian is a Professor of Philosophy at the University of Innsbruck, Austria. He is, since 2006, the president of Austrian Ludwig Wittgenstein Society (Österreichische Ludwig Wittgenstein Gesellschaft). Kanzian’s main interests are the analytic philosophy, the history of philosophy, and ontology. He is the author of four books: Originalität und Krise – Zur systematischen Rekonstruktion der Frühschriften Kants (1994), Grundproblerne der Analytischen Ontologie (with E. Runggaldier, 1998), Ereignisse und andere Partikularien (2001), Ding – Substanz – Person. Eine Alltags-ontologie (2009), and an editor or co-editor of eleven books. He has also published over seventy articles. Daniel Linford is a visiting instructor in the Department of Philosophy and Religious Studies at Christopher Newport University and an adjunct professor at Thomas Nelson Community College. He earned degrees in both physics (University of Rochester) and philosophy (Virginia Tech). His research concerns metaphysics, value theory, and philosophy of religion. Jason Megill has research interests in philosophy of religion, philosophy of mind and early modern philosophy. He has taught at the University of Colorado at Boulder and Old Dominion University. He currently teaches at Bentley University in Waltham, Massachusetts. Uwe Meixner is a professor of philosophy at the University of Augsburg, Germany. He previously was bound up with the University of Regensburg (till 2010) and earned his PhD from that University in 1986, under the direction of Franz von Kutschera. His main fields of research are Theoretical Philosophy (especially, Logic, Metaphysics, and Philosophy of Mind) and the History of Philosophy. The places of his visiting professorships or research projects are: Innsbruck, Mainz, München, Münster in Westfalen, Notre Dame (USA), Osnabrück, Regensburg, Saarbrücken, and Salzburg. Uwe Meixner is the author of eleven books: Handlung, Zeit, Notwendigkeit: Eine ontologisch-semantische Untersuchung (1987), Ereignis und Substanz: Die Metaphysik von Realität und Realisation (1997), Axiomatic Formal Ontology (1997), Theorie der Kausalität: Ein Leitfaden zum Kausalbegriff in zwei Teilen (2001), The Two Sides of Being: A Reassessment of Psycho-Physical Dualism (2004), Einführung in die Ontologie (2004), David Lewis (2006), The Theory

Authors of Contributed Papers |

227

of Ontic Modalities (2006), Modalität: Möglichkeit, Notwendigkeit, Essenzialismus (2008), Philosophische Anfangsgründe der Quantenphysik (2009), Modelling Metaphysics: The Metaphysics of a Model (2010). He has published over 100 articles, and is a co-editor of two journals: Logical Analysis and History of Philosophy and Metaphysica. Elisa Paganini is Lecturer in Philosophy at the University of Milan (Università degli Studi di Milano). Her principal research interests are in philosophy of language and metaphysics. She is the author of two books: La realtà del tempo [The Reality of Time] (2000) and La vaghezza [Vagueness] (2008). Her published papers have addressed philosophy of time and vagueness; they have appeared in Philosophical Studies, Erkenntnis, Dialectica, and other journals. Eleonore Stump is the Robert J. Henle Professor of Philosophy at Saint Louis University, where she has taught since 1992. She received a B.A. in classical languages from Grinnell College (1969), an M.A. in Biblical Studies (New Testament) from Harvard University (1971), and an M.A. and Ph.D in Medieval Studies (Medieval Philosophy) from Cornell University (1975). Before coming to Saint Louis University, she taught at Oberlin College, Virginia Polytechnic Institute and State University, and University of Notre Dame. She has published extensively in medieval philosophy, philosophy of religion, and contemporary metaphysics. Her books include her major study Aquinas (Routledge, 2003) and her extensive treatment of the problem of evil, Wandering in Darkness: Narrative and the Problem of Suffering (Oxford, 2010). Her articles have been published, among other things, in such professional journals as: Aletheia, American Philosophical Quarterly, Faith and Philosophy, History and Philosophy of Logic, International Journal for the Philosophy of Religion, Journal of Analytic Theology, Journal of the History of Philosophy, Journal of Philosophy, Midwest Studies, Philosophical Perspectives, Philosophical Topics, Philosophy of Religion, Philosophy and Phenomenological Research, Proceedings and Addresses of The American Philosophical Association, Proceedings of the American Catholic Philosophical Association, Synthese, The American Catholic Philosophical Quarterly, The Cambridge History of Later Medieval Philosophy, The Canadian Journal of Philosophy, The Journal of Ethics, The Monist. E. Stump is a member of the Editorial Board or consultant many professional journals, and a member of many scientific societies. Mirosław Szatkowski is a professor of philosophy at the Warsaw University of Technology, Poland, and the Director of the International Center for Formal Ontology. He earned his PhD in philosophy from Jagiellonian University in Cracow, Poland, and was habilitated at the Ludwig-Maximilians University in Munich,

228 | Authors of Contributed Papers Germany. Szatkowski’s main fields of research are: logic, the foundations of mathematics, and formal ontology. In these areas, he has published papers in the following professional journals: Studia Logica, Zeitschrift für mathematische Logik und Grundlagen der Mathematik (Mathematical Logic Quarterly), Archiv für Mathematische Logik und Grundlagenforschung (Archive for Mathematical Logic), Notre Dame Journal of Formal Logic, Journal of Applied Non-Classical Logics, Journal of Logic, Language and Information, and Metaphysica; and in several collective volumes. He has edited five volumes: Ontological Proofs Today (Frankfurt: Ontos Verlag, 2012), Dualistic Ontology of the Human Person (München: Philosophia, 2013), Substantiality ana Causality (Boston/Berlin/Munich: Walter de Gruyter, 2014), God, Truth, and other Enigmas (Boston/Berlin/Munich: Walter de Gruyter, 2015), and Analytically Oriented Thomism (Editiones Scholasticae, 2016). William F. Vallicella is an American philosopher. He has a Ph.D. (Boston College; 1978), taught philosophy for a number of years at University of Dayton (where he was a tenured Associate Professor of Philosophy; 1978–91) and Case Western Reserve University (Visiting Associate Professor of Philosophy; 1989–91), and retired to Gold Canyon, Arizona from where he now contributes to philosophy mainly online. His main fields of research are metaphysics and philosophy of religion. W. F. Vallicella is the author of two books: Kant, subjectivity and facticity (Boston College, 1978), and A Paradigm Theory of Existence: Onto-Theology Vindicated (Kluwer Academic Publishers 2002); and 60+ articles in such professional journals as Analysis, Dialectica, Faith and Philosophy, International Philosophical Quarterly, Journal of Philosophical Research, Nous, Philosophia, Ratio, and The Monist. Peter van Inwagen is an American analytic philosopher and the John Cardinal O’Hara Professor of Philosophy at the University of Notre Dame. He previously taught at Syracuse University and earned his PhD from the University of Rochester under the direction of Richard Taylor and Keith Lehrer. Van Inwagen is one of the leading figures in contemporary metaphysics, philosophy of religion, and philosophy of action. Peter van Inwagen has published seven books: An Essay on Free Will (1983), Material Beings (1990), Metaphysics (1993, and rev. ed. 2002), God, Knowledge and Mystery: Essays in Philosophical Theology (1995), The Possibility of Resurrection and Other Essays in Christian Apologetics (1998), Ontology, Identity, and Modality: Essays in Metaphysics (2002), The Problem of Evil (2006). He also has published over 150 articles on general metaphysics, free will, material objects and human persons, philosophy of religion, logic and language. In 2005 Peter van Inwagen was elected to the American Academy of Arts and Sciences, and in 2011

Authors of Contributed Papers |

229

he received the degree Doctor Divinitatis (honoris causa) from the University of St Andrews.

Abstracts Gabriele De Anna “Theism and the Ontological Ground of Moral Realism” This essay suggests that moral realism implies a restricted, purely explanatory reading of the claim that God is the ontological ground of morality, a reading which is consistent with the obvious fact that atheists can be moral people just as religious believers. The argument moves from considerations suggesting that moral realism implies normative fitness, the view that there is a correspondence or an adequacy between values and practical rationality. Such fitness calls for an explanation and teleology is the best explanation of it. The essay further argues that the kind of teleology needed to account for the fitness between values and our practical rationality is of an intentional kind, and this will lead to conclusions concerning the ontological ground of morality, and the interpretation of it which is supported by moral realism.

Chris Daly “Agnosticism and the Balance of Evidence” The agnostic suspends judgement as to whether there is a God: she thinks that there is no good evidence that there is God and no good evidence that there isn’t. The agnostic faces the challenge of Russell’s teapot: that, just as the hypothesis that there is a cosmic teapot is not supported by any evidence and therefore there is evidence that it is false, so too the hypothesis that there is a God is not supported by evidence and therefore there is evidence that it is false. Does this challenge show that the agnostic is irrational? What relevant difference is there between the cosmic teapot hypothesis and the God hypothesis? This paper defends the agnostic’s position.

Michał Głowala “Polygeny, Pleiotropy, and Two Kinds of Concurrentist Ontology” In the paper I consider, from a general metaontological point of view, some aspects of polygeny (one and the same effect having various causes) and pleiotropy (one and the same causal power having various effects in concert with various other causes) that are important for the issue of concursus divinus in general. I https://doi.org/9783110566512-015

232 | Abstracts focus on the thesis that for each of the concurring causes there is something that this cause contributes to the effect, and that the effect is a composition or combination of these contributions (Partial Causation Principle, PCP). Although PCP has a very strong intuitive appeal I argue that it is false in some important cases of polygeny, which do not belong to philosophical theology, but are traditionally used as models for divine concurrence. Then I show the role of PCP in two rival versions of concurrentism: the Jesuit and the Dominican one. My conclusion is that Molinism, as opposed to Thomism, accepts some weak version of PCP: that I hope sheds some important light on both standpoints and the relationship between them.

Christian Kanzian “ “Bottom-up” versus “top-down” ” Important currents in the mainstream of ontology seem to be guided by a specific world-view which posits that there are different levels of reality. The supreme example that seems to legitimate this insight is taken from the seeming divide between micro-physics and the macro-level of our world. Basically, it is assumed that atom-like entities (however they are called) constitute the bottom level of reality, and that, moving from the bottom up, we come upon higher levels, including the level of the macro-things of our everyday life-world. What I intend to do in my article is to show how this bottom-up world-view influences and determines contemporary ontological theories. My work in this paper will be firstly to point out some typical, and usually hidden, bottom-up premises in ontology today, and secondly, to critique these premises and those ontological theories like supervenience-, emergence-, or constitution-theories which depend on them. I want to argue that a top-down world-view is a worthy alternative to the bottom-up view, and that we can get to one by putting together an appropriate ontological frame. This, I argue, will bring us to a theistic top-down perspective on our world.

Daniel Linford and Jason Megill “Cognitive Bias, The Axiological Question and the Epistemic Probability of the Theistic Belief” Some recent work in philosophy of religion addresses what can be called the “axiological question,” i.e., regardless of whether God exists, would it be good or bad if God exists? Would the existence of God make the world a better or a worse place?

Abstracts |

233

Call the view that the existence of God would make the world a better place “ProTheism.” We argue that Pro-Theism is not implausible, and moreover, many Theists, at least, (often implicitly) think that it is true. That is, many Theists think that various good outcomes would arise if Theism is true. We then discuss work in cognitive science concerning human cognitive bias, before discussing two noteworthy attempts to show that at least some religious beliefs arise because of cognitive bias: Hume’s, and Draper’s and Nichols’s. We then argue that, as a result of certain cognitive biases that result when good outcomes might be at stake, Pro-Theism causes many Theists to inflate the epistemic probability that God exists, and as a result, Theists should lower the probability they assign to God’s existence. Finally, based our arguments, we develop a novel objection to Pascal’s wager.

Jason Megill and Daniel Linford “On Computable Metaphysics: On the Uses and Limitations of Computational Metaphysics” Humans constantly produce strings of characters in symbolic languages, e.g., sentences in natural languages. We show that for any given moment in human history, the set of character strings that have been produced up to that moment, i.e., the sum total of human symbolic output up to that moment, is finite and so Turing computable. We then prove a much stronger result: a Turing machine can produce any particular set of symbolic output that we could possibly have produced. We then discuss metaphysical and/or theological systems, e.g., Spinoza’s, Leibniz’s, or Aquinas’s. We argue that any particular metaphysical system could be the output of a Turing machine. We then briefly discuss (i) automated theorem proving, the attempt to generate and/or prove theorems using computers, and (ii) computational metaphysics, the attempt to apply methods from computer science such as automated theorem proving to metaphysics. We conclude by arguing that: (i) computational metaphysics can succeed, at least in theory; indeed, all human metaphysical reasoning could be automated; (ii) computational metaphysics – and indeed metaphysical systems in general – will have the same limitations as Turing machines; and (iii) a number of different methods – not currently used in computational metaphysics – could be used to advance metaphysical research.

234 | Abstracts Uwe Meixner “What Evil Must Be in Order to Exist” The paper presents various ways in which philosophers (Thomas Aquinas, Spinoza, Wittgenstein, Leibniz) have attempted to deny the existence of evil by making use of suitable concepts of existence and/or evil. The paper shows that there is evil whose existence cannot be denied.

Elisa Paganini “Normative Rules for Indeterminacy” Williams (2012)* recently proposed the Normative Silence model of Indeterminacy in order to account for a single phenomenon running through all cases of indeterminacy and to reach consensus on the correct epistemic attitude to adopt towards borderline cases of paradigmatically vague predicates. Williams’s Normative Silence model says there is no general normative rule governing God’s and humans’ belief attitudes towards indeterminacies. I claim instead that human rationality and philosophical inquiry require general normative rules leading our belief attitudes towards indeterminacies and that God’s belief attitudes are more difficult to define than Williams assumes. * Williams, J. R. G. (2012), “Indeterminacy and Normative Silence”, Analysis: 72(2), 217–225.

Eleonore Stump “The Openness of God: Eternity and Free Will” The understanding of God’s mode of existence as eternal makes a significant difference to a variety of issues in contemporary philosophy of religion, including, for instance, the apparent incompatibility of divine omniscience with human freedom. But the concept has come under attack in current philosophical discussion as inefficacious to solve the philosophical puzzles for which it seems so promising. Although Boethius in the early 6th century thought that the concept could resolve the apparent incompatibility between divine foreknowledge and human free will, some contemporary philosophers, such as Al Plantinga, have argued that eternity gives no help with this problem. Other philosophers, such as William Hasker, have argued that whatever help the concept of eternity may give with that puzzle is more than vitiated by the religiously pernicious implications of the concept for

Abstracts |

235

notions of God’s providence and action in time. In this paper, I will examine and respond to these arguments against the doctrine of God’s eternity.

William F. Vallicella “Does God Exist Because He Ought To Exist?” Modal ontological arguments for the existence of God require a possibility premise to the effect that a maximally perfect being is possible. Admitting the possibility of such a being may appear to be a minimal concession, but it is not given that the admission, together with the uncontroversial premise that a necessary being is one whose possibility entails its actuality, straightaway entails the actual existence of a maximally perfect being. The suspicion thus arises that the modal ontological argument begs the question at its possibility premise. So various philosophers, including J. N. Findlay, A.C. Ewing, John Leslie, and Carl Kordig have attempted to support the possibility premise by broadly deontic considerations concerning what ought-to-be, where this ought-to-be subsists independently of the powers of any agent. The basic idea is that God, conceived as a maximally perfect being, is possible because (i) he ought to exist, and (ii) whatever ought to exist is possible. The basic idea is that the non-agential oughtness or axiological requiredness of the divine existence certifies the possibility and in turn the actuality of the divine existence. The overall argument could be described as a broadly deontic God proof along modal ontological lines. This article sets forth and defends the argument before explaining why it is not ultimately compelling.

Peter van Inwagen “God’s Being and Ours” This essay is an examination of the idea that God and creatures enjoy different “modes of being”. It begins with an attempt at an explanation of the concept of modes of being, an attempt that the author concedes is a failure. He goes on to contend that any such attempt must be a failure, owing to the fact that the thesis that there are distinct modes of being is unintelligible. The remainder of the essay concerns the vast gulf that separates the divine nature from all creaturely natures. It is suggested that it is precisely the incomprehensible vastness of this gulf that has led some philosophers and theologians mistakenly to suppose that the existence of God and the existence of a creature like St Peter must be different kinds of existence. It is specifically asserted that the proposition, “The word ‘existence’ neither differs in meaning nor denotes different kinds of existence when it is ap-

236 | Abstracts plied to God and to Peter”, is consistent with the proposition (which the author affirms), “God exists necessarily and Peter exists only contingently”.

Person Index Adams R. 19, 36 Alvarez D. 53, 56–58, 60 Anjum R. L. 41, 43, 61 Anselm of Canterbury 205 Appel K. 102, 109 Aquinas Thomas XVIII, 46–48, 51, 52, 55, 60, 99, 114, 115, 121–124, 128, 152, 155, 156, 160, 161, 163, 165, 167, 175, 179, 196, 197 Aristotle 49, 155, 156, 160, 161, 163 Armour L. 163–165, 198 Arriaga R. 45, 60 Ashley B. 161, 162, 198 Augustine 115 Bagai R. 102, 109 Bain A. XIII, XV, XXI Baker L. R. 69–71, 73, 75 Barackman F. H. XVII, XXI Bastardi A. 85, 90 Beal A. 110 Beavers A. 103, 110 Bedau M. 26, 36 Benjamins R. V. XXI Beothius A. M. S. 137 Bibel W. 102, 110 Bigelow J. 67, 75 Billuart C. 50, 60 Boethius A. M. S. 138, 160 Boolos G. S. 95, 110 Braithwaite R. B. XIII, XV, XVI, XXI Britton T. 75 Buczek A. I. 195, 199 Campbell K. 64, 67, 68, 76 Camus A. 77, 90 Carroll W. E. 40, 46, 60 Cartwright N. 43, 60 Casebeer W. 28, 36 Chandrasekaran B. X, XXI Chojnacki P. 178, 179, 199 Clarke S. 24 Clarke W. N. 199 https://doi.org/9783110566512-016

Cole D. 110 Coreth E. 199 Craig E. XI, XII, XXI Craig W. 19, 36, 79, 81, 84, 90, 110 Creary L. 43, 60 Cullen Ch. M. 199 Daly Ch. 1 Davidson D. XIII, XIV, XXI, 67, 76 Davis M. 102, 110 Dawkins R. XVII, XXI De Anna G. 19, 30, 34, 36 De Koninck Ch. 199 De Koninck Ch. 161, 163, 165, 166, 199 De Koninck T. 199 Dewan L. 175, 176, 199 Donceel J. 199 Dore C. 5, 6, 16 Dragos C. 77–79, 90 Draper P. 2, 16, 81, 84, 90 Duffy D. A. 102, 110 Dunning D. 82, 90 Duns Scotus J. 45, 60 Dvořák P. 40, 44, 46, 55, 57, 60 Eklund M. 129, 135 Elders L. J. 50, 60 Ewing A. C. 206, 212 Feldman R. 4, 16 Fernandez-Alonso A. 162, 199 Feser E. 161, 199 Findlay J. N. 207, 212 Finnis J. 23, 36 Fitelson B. 111 Freddoso A. J. 40, 44–49, 56, 60 Frege G. 217 Gödel K. 110 Głowala M. 39, 48, 60 Garbacz P. XI, XXI Garcia B. 172, 200 Genesereth M. R. IX, XXI

238 | Person Index Giaretta P. IX, XXI Gilson É. H. 54, 165, 172, 173, 175–177, 200 Gordon J. 79, 90 Gruber T. R. IX, X, XXI Guarino N. XXI Guatino N. IX Haber J. 79, 90 Hacker P. M. S. XIII, XVI, XVII, XXI Haken W. 102, 109 Haldane J. 155, 200 Hart D. B. 200 Hartmann N. 209, 210, 212 Hasker W. 137–139, 141–143, 147, 148, 151–153 Heil J. 68, 76 Henle R. J. 200 Herbut J. 200 Hintikka J. 183, 184, 200 Hitchens C. 77, 88, 90 Hittinger J. P. 167, 200 Hume D. XIII, XIV, XXI, 8, 16, 82–84, 90 Ikeda M. X, XXII James W. XIII, XIV, XXI Jeffrey R. C. 95, 110 John of St. Thomas 45, 60 Josephson J. R. XXI Joyce R. 28, 36 Kahane G. 77, 81, 88, 90 Kahneman D. 81, 82, 90, 91 Kalinowski J. 195 Kamiński S. 195–198, 202 Kane W. 161 Kant I. 3, 16, 105, 179–182, 184, 201, 216 Kanzian Ch. 63, 72, 76 Kilby K. 201 Kim J. 68, 76 Klagges B. 73, 76 Klenk K. F. 161, 201 Kleutgen 195 Knasas J. 161, 162, 166, 173, 175, 201 Kordig C. R. 207, 212 Krąpiec M. A. 171, 195, 197, 202

Kraay K. 77–79, 90 Kretzmann N. 138, 139, 150, 152, 153 Kruger J. 82, 90 Le Poidevin R. 2, 6, 16 Leibniz G. W. 98, 99, 110, 119–124 Leslie J. 207, 211, 212 Levi I. XIX, XXI Lewis D. 64, 76, 98, 99 Linford D. 77, 93 Lokhorst G. J. C. 103, 110 Lonergan B. 42, 60, 202 Lowe E. J. XI, XXI, 71, 76 Lucas J. R. 98, 110 MacKenzie D. 102, 110 Mackie J. L. 8, 16, 207, 211, 212 Maréchal J. 202 Maritain J. 165–170, 173, 202, 203 Marmodoro A. 42, 47, 60 Maryniarczyk A. 195, 202 Massin O. 43, 61 Maudlin T. 133, 135 Mawson T. 77, 90 McCann H. J. 40, 46, 54, 61 McCool G. A. 169, 203 McCune W. 102, 110 McDonald H. 198 McGrath A. E. XVII, XXII McInerny R. M. 160, 203 McKitrick J. 48, 61 Megill J. 77, 88, 90, 93, 94, 97, 110 Meixner U. XXII, 113 Melvin T. 94, 97, 110 Mercier D. 176, 177 Messick S. 85, 91 Metz T. 77, 91 Mill J. S. 44, 61 Mitchell J. 88, 90 Mizoguchi R. X, XXII Molina L. 46, 50, 52–56, 58, 61 Molnar G. 39, 41–43, 45, 49, 61 Montesinos L. 45, 61 Morelli E. A. 203 Morelli M. D. 203 Morgenbesser S. XIX, XXI Morris T. V. 40, 61

Person Index |

Muck O. 203 Mumford S. 41–43, 61 Nagar R. J. 161 Nagel T. 21, 22, 25, 29–33, 36, 77, 78, 88, 91 Newton I. 63, 76 Nichols R. 81, 84, 90 Nilsson N. J. IX, XXI Noël L. 176, 177 O’Callaghan J. 160, 203 Oberle D. IX, XXI Obitko M. IX, XI, XXII Ogden S. G. 203 Oppenheimer P. 103, 111 Oppy G. 4, 11, 16 Owens J. 203 Paganini E. 129, 133, 135 Pascal B. 89 Penafiel L. 51, 61 Penrose R. 98, 110 Plantinga A. 4, 11, 13, 16, 126, 137, 139, 140, 142, 143, 146, 153 Pohl R. F. 82, 91 Pope St. Pius X 157 Portoraro F. 102, 110 Price H. H. XXII Proclus 160 Pseudo-Dionysius 160 Putnam H. 12, 16 Quine W. V. O. XIII, XVI, XXII Rahner K. 203 Ramsey F. P. XIII, XIV, XXII Reagor A. 88, 91 Reese P. N. 161, 203 Rilke R. M. 126, 128 Rosen G. 4, 16 Rosenhan D. 85, 91 Ross L. 85, 90 Rowe W. L. 12, 16 Russell B. XIII, XIV, XXII, 1, 5, 6, 8, 11, 16, 77, 91, 110 Ryle G. XIII, XVI, XXII

239

Schaffer J. 63, 76 Schlesinger G. 10, 16 Schmitt P. H. 102, 110 Schopenhauer A. 123, 127 Scriven M. 7, 11, 12, 16 Searle J. R. XIII–XV, XXII, 105, 110 Shakespeare W. 117, 128 Shalkowski S. A. 3, 6, 16 Shanley B. J. 40, 61 Sharot T. 85, 91 Simons P. 64, 65, 72, 76 Smith B. XI, XXII, 73, 76 Smith V. 161 Sober E. 13, 16 Spinoza B. 98–100, 103, 110, 115–117, 124, 125, 128 Stępień A. 177, 179, 203 Staab S. IX, XXI Stapleford S. 180–183, 203 Street S. 28, 36 Streeter C. M. 203 Stump E. 137, 138, 150, 152, 153 Suárez F. 40, 46, 50, 61 Swartz N. P. 203 Sweet W. 168, 170, 203 Swieżawski S. 195 Swinburne R. 8–11, 16 Szatkowski M. IX, XVIII, XXII, 155 Szubka T. 195, 199 Szymaniak A. 195, 203 Toledo F. 50, 61 Tooley M. 77, 91 Trypuz R. XI, XXI Turing A. M. 95–97, 104, 105, 107, 110 Tversky A. 81, 82, 90, 91 Uhlmann E. L. 85, 90 Vázquez G. 45, 61 Vallicella W. F. 205 Van Inwagen P. 6–9, 13, 15, 16, 213 Vogler C. 36, 37 Wallace W. A. 161 Wang H. 95, 111 Weisheipl J. A. 161, 163, 204 White R. 37

240 | Person Index Whitehead A. N. 110, 111 Whitten D. XI, XXII Williams J. R. G. 129–131, 133–135 Williams N. E. 39, 61 Williamson T. XIII, XV, XXII Wippel J. F. 54, 61

Wittgenstein L. 117, 128 Wolff Ch. 195 Wolterstorff N. 11, 17 Zagzebski L. 137, 140, 141, 147, 153 Zalta E. 102, 111 Zdybicka Z. J. 171, 202

Subject Index accident 157 act 157 actual symbolic output 93–96 agnosticism 2, 11 – global form 2 – Kantian 3, 11 – local form 2 argument – modal ontological 205 – transcendental 180 argument from justice 79 atheism XVII – intellectual 87 being 166, 168, 173, 213, 216, 217 – maximally perfect 205, 206, 210 belief XIII, 21 – non-theistic IX – theistic IX, 93 body 158 bottom-up 63, 65, 69, 72, 73 bridge problem 177 cause – partial 46, 50 cognitive bias 81, 82, 84 coherence 197 conceptualization IX, X concurrentism 40, 44, 46 concursus causarum 39 concursus divinus 39, 57, 59 concursus generalis 39, 44 concursus indifferens 53–55 concursus praevious 56–59 concursus simultaneus 52, 56 conservationism 40 constitution 69 deduction – transcendental 180 degrees of knowledge 168 doctrine of actus 156 doctrine of potentia 156 https://doi.org/9783110566512-017

esse 157, 167, 174 essence 159, 175 ET-simultaneity 139, 145, 146, 148, 150 eternity 137–139, 143–145 evidence 2, 4, 5 – agreed 3 – dieputed 3 Existence 113–116, 123, 124 – as actuality in pure objectivity 116 – as changeless actuality 114, 115 – as goodness 115, 124 – as necessary being 114 – as substantial reality 114 – concepts of existence 113–115, 117 – contingent 115 – necessary 114 – of evil 113–115, 118, 121, 127 – of God 113, 120, 127 – predicative 114, 118 – quantificative 114, 118 – true 116 existence XII, 169, 173, 175, 213–216 – ontological XII – semantical XII feeling XIX fitness 24–26 – normative 24 Gnostics 123 God 34, 52, 53, 113–117, 120–123, 126, 127, 157, 159, 171, 175, 209, 217, 218 – God’s existence 127 – God’s perfection 113 ground – ontological 20, 35 hypothesis – cosmic teapot 1, 6 – Fitzgerald and Lorentz’s contraction 14 – Russell’s teapot 1, 5 illumination – divine 103

242 | Subject Index immanence principle 177 indeterminacy 129, 130, 133 justify 21 knowledge – a priori 183 – connatural 170 – dianoetical 168 – eternal 140, 141, 147 – natural 170 – non-rational 168 – perinoetical 168 – rational 168 – transcendental 180 Law of the Conservation of Energy 124 Lublin Philosophical School 195 Maritain’s Sisth Way 171 maximalism 196 method – resolution 102 – transcendental 179, 184, 187 – thomistic 185 monotheism XVII morality 19 naturalism – evolutionary 29 Neo-Thomism 186, 195 object – pragmatic 190 objectivity 198 occasionalism 40 omnibenevolence XVIII omnipotence XVIII omniscience XVIII ontology IX, XI openness of God 137 optimism bias 85 paradox – Liar 133 Pascal’s wager 89 perfection 157

philosophy – analytical 156 polygeny of effects 39 polygeny of powers 39 polygeny pluralism 59 polytheism XVII possible symbolic output 97 potency 157 potentiality 23, 24, 32 principle – partial causation 40, 41, 45, 46, 48 – relatively strong version 43 – strong version 44 – weak version 44 – precedence 177 Principle of Causal Adequacy 108 Principle of Non-Contradiction 98 Principle of Suflcient Reason 98, 103 pro-theism 78–81 puzzle – William’s 129 question – axiological 77, 78 rationality – practical 26 razor – Ockham’s 12 realism 23 – cognitive 195 – critical 169 – metaphysical 172 – methodical 179 – moral 21–24 representation – mental 103 Rostov’s Equation 213, 214 Second Law of Thermodynamics 124 set – Turing computable 95, 96 specification IX, X standarization X state – mental XIV, XV substance 173

Subject Index |

supervenience 67–69 teapot 7, 8, 11 teleology – degree-one 27 – degree-three 30, 31, 33–35 – degree-two 27 theism XVII, 9 – de facto XVII – intellectual 87 – strong XVII thesis – composition 43, 51 – contribution 42, 49 – monogeny 42, 49, 51, 55, 56 – monotropy 42, 49, 53 Thomism 155, 156 – analytically oriented 155 – Aristotelian 160 – existential 165 – Laval 161, 163 – River Forest 161, 162

243

– transcendental 179, 185, 194 top-down 64, 72–74 transcendentalist 196 unity – methodological-epistemological 197 value 20, 22, 26 – moral 22 world 113, 123–127 – the (actual) world 124 – actual 120 – as a hell 123 – as a whole is a bonum 124 – as a whole is a malum 120, 123, 125, 127 – as a whole is a perfectum 125 – best possible 120 – created 113, 127 – is fixedly directed towards destruction and decay 124 – substantially real parts of the world 124