Dialectic Journal Autumn 2021 Edition by Dialectic Journal

ISSN 2754-8856

Volume XV | Issue 3 | Autumn 2021

The Undergraduate Journal of Philosophy at The University of York

The Contributors Editor Thomasina Cass

Graphics and Design Thomasina Cass

Co-Editor Alice Letts

Reviewers Chiara Bassini William Coutts Dylan Elliot Jack Lyons Joe Newton Thresh Sarah Quinn Mihaela Sotirova

Contributors Romain Marchant Armand Babakhanian Nicholas Alan Barrow Dr John Shand Interviewee Dr Fiora Salis

Copy Editors Chiara Bassini Jack Lyons Joe Newton Thresh Mihaela Sotirova

Contents Editor’s Preface

1. Do not mistake the indecisive businesswoman

for a resolute assassin: A Critique of Robert Kane’s libertarianism

2. Dismissing doubts about Descartes: Cartesian meditation for epistemological crises

3. Is it the case we may as well be Nihilists?

4. On Never Knowing the Definition of Knowledge

Interview with Dr Fiora Salis

Welcome to Dialectic! Despite not having a theme for this autumn edition of Dialectic, the four essays featured foster a similar tone to one another while remaining a true dialectic of ideas and philosophies. As always, this edition follows its ancestors in intending to promote varied discourse and measured critique and debate. “Life is an unfoldment, and the further we travel the more truth we can comprehend. To understand the things are that are at our door is the best preparation for understanding those that lie beyond” Hypatia of Alexandria

In the first paper, Romain Marchant discusses and critiques Robert Kane’s account of indeterministic processes of decision making, opening up the weaknesses of Kane’s argument and ultimately rejecting Kane’s account on the basis of Kane’s failure to acknowledge bias and address the issues of indeterministic agency without altering the description of the Businesswoman case to fit. In the second paper, Armand Babakhanian investigates Alasdair MacIntyre’s account of an epistemological crisis, formulated by addressing Descartes’ account of scepticism. Babakhanian argues that MacIntyre misunderstood Descartes’ project in writing Meditations by not accounting for Cartesian Context. In the third paper, Nicholas Alan Barrow identifies and rejects the argument that we may as well be Nihilists through investigating the effect of parsimony and the use of the paraphrasing technique used to achieve Nihilism’s trademark simplicity. In the fourth paper, Dr John Shand aims to show that knowing that one has knowledge of the correct definition of knowledge is not possible due to the circularity created, even if a nondefinitional count of knowledge is offered.

Do not mistake the indecisive businesswoman for a resolute assassin: a critique of Robert Kane’s libertarianism Romain Marchant is a Philosophy Masters student at the Katholieke Universiteit Leuven, and previously graduated with a Bachelors from the Université libre de Bruxelles. romain.marchant@ulb.be Introduction In this essay, I will dispute Robert Kane’s claim that a decision resulting from an indeterministic deliberation process is unrelated to luck. My opposition to Kane is based on his paper “Responsibility, luck and chance: reflections on free will and indeterminism.” Firstly, I will give a summary of his article. Secondly, I will discuss what I consider to be weaknesses in his argument. I will ultimately reject Kane’s description of a free indeterministic decision because his argumentation suffers from an implicit bias he fails to acknowledge. Kane’s libertarianism Before discussing Kane’s article, I will summarize where he stands in the free will debate. After “do humans have it?”, the second most important question about free will is “is it compatible with determinism?”. Determinism claims that the complete set of laws and the exhaustive historical description of the world (the 5

minute description of every prior state of events) entails every future event. Compatibilists claim that determinism may be true when free will is. In other words, they see free will as compatible with the necessitation of every event by the complete history and laws of nature. Kane denies that claim. He is an incompatibilist: he sees determinism as undermining free will. Kane also believes that humans have free will, meaning they have enough control over their actions to be morally responsible. The conjunction of incompatibilism and faith in free will is called the libertarian view (or libertarianism). Following this, Kane must give an account of free will together with the negation of determinism, that is, indeterminism. Kane is not only keen to show that indeterminism and free will are compatible but also that there is, in fact, an indeterministic aspect of human decision-making that makes humans free. In his article, Kane (1999) argues against the Luck Principle (or LP). This is how he formulates it: (LP) If an action is undetermined at a time t, then its happening rather than not happening at t would be a matter of chance or luck, so it could not be a free and responsible action. This principle articulates the familiar intuition that if some event is not deterministically caused, if it could not have happened given the otherwise same circumstances, then its happening is random. It also contains the second claim that if something is a matter of chance, there cannot be an agent responsible for bringing it about. These two claims put together, if correct, would refute libertarianism because indeterminism and free will could not both be true. Kane restricts the need for indeterminism to some particular actions. 6

He acknowledges that some actions occur under deterministic circumstances, but he argues that some particularly important actions only occur when brought about under undetermined agency. He calls them ‘self-forming actions’ or SFAs1. Other determined decisions and actions follow from these instances of freedom. SFAs occur ‘under conditions of uncertainty,’2 ‘at times in life when we are torn between competing visions of what we should do or become.’3 The period preceding an SFA is a time of doubt. Both choices (assuming that the agent has only two choices) compete in the agent’s mind when deliberating, and neither is determined to happen before her choosing. Deciding consists in making ‘one set of competing reasons or motives prevail over the others.’4 Kane illustrates his argument with the Businesswoman case: a businesswoman witnesses a street mugging on her way to work. She can either help the stranger by calling the police and scaring the muggers away or continue her walk to the office. The choice is tough for her to make because she cannot afford to be late to the office as she might lose her job. The two choices Kane presents correspond respectively to a moral reason and a selfinterested motive. In an ‘effort of will,’ the agent may overcome her egotistical tendencies and do the right thing.5 The case of the assassin exemplifies Kane’s claim that indeterminism is compatible with responsibility6: Consider an assassin who is trying to kill the prime minister, 1 Robert Kane, Responsibility, Luck, and Chance: Reflections on Free Will and Indeterminism. p. 224. 2 Kane p. 225. 3 Kane p. 224. 4 Kane p. 225. 5 Kane p. 232. 6 Kane p. 227.

but might miss because of some undetermined events in his nervous system which might lead to a jerking or wavering of this arm. If he does hit his target, can he be held responsible? The answer is “yes,” because he intentionally and voluntarily succeeded in doing what he was trying to do—kill the prime minister. Yet his killing the prime minister was undetermined. If the assassin succeeds, he is responsible even though his success is undetermined. Therefore, one can be morally responsible for an undetermined event, and LP is refuted. From these results, Kane imagines a stronger Luck Principle, LP*: (…) if it is undetermined at t whether an agent voluntarily and intentionally does A at t or voluntarily and intentionally does otherwise, then the agent’s doing one of these rather than the other at t would be a matter of luck or chance, and so could not be a free and responsible action. The Assassin example cannot refute LP* because the assassin would not be intuitively blameworthy to have killed the prime minister if he did enough not to kill him, that is, if he could not have killed him with the same laws and the same past up to the moment of murder. Nonetheless, Kane also refutes LP*. Kane uses the Businesswoman case against LP* with two moves. Firstly, he ‘put[s] the indeterminacy involved in the efferent neural pathways of the […] assassin into the central neural processes of the businesswoman […].’7 Secondly, he describes the businesswoman’s deliberation as ‘in the presence of background indeterministic noise,’ as if she is trying to solve ‘two competing (deliberative) tasks at once.’ The businesswoman is not determined to do either because she desires to do both things. He accounts for indeterminism by identifying Kane p. 231.

it to the balanced nature of the competing reasons such that none deterministically emerges victorious: ‘the effort is indeterminate and the indeterminism is a property of the effort.’8 In other words, indeterminism is the uncertainty in deliberating. Kane also claims that the ‘whole process is her effort of will.’9 He continues10: She may therefore fail to do what she is trying to do, just like the assassin […]. But I argue that, if she nevertheless succeeds, then she can be held responsible because, like [him], she will have succeeded in doing what she was trying to do. And the interesting thing is that this will be true of her, whichever choice is made, because she was trying to make both choices and one is going to succeed. Kane sensibly restricts the indeterministic free agency to one kind of action (the SFA) and puts indeterminism inside the agent’s deliberative ability. He acknowledges that the decisions enactment has to deterministically follow from the decision for the agent to be in control of the action. Furthermore, as Franklin (2011) remarks, if indeterminism occurs before deliberation, it does not enhance responsibility. Kane calls the kind of control the businesswoman has over her decision a ‘plural voluntary control’ because she can voluntarily make either one of the two decisions with the amount of control required for responsibility.11 She controls which option to choose, when to choose it and why, without deterministic internal or external influence. Her plural voluntary control over both choices shows, according to Kane, that indeterminism is the condition for a free SFA. Kane p. 232. Kane p. 232. 10 Kane p. 231. 11 Kane p. 237. 8 9

My objection to Kane Kane’s refutation of the weaker Luck Principle LP is sound. His presentation of the Assassin case brings into light valuable and subtle differences between luck and indeterminacy. However, I believe that his argumentation against the stronger Luck Principle LP* is unconvincing and fails to justifies his libertarian view. My objection to Kane is that his way of presenting the agent’s dilemma between two moral choices is inessential to the situation of an SFA. To portray the agent as trying to overcome selfishness in an effort towards virtuous conduct is not essentially what makes an action ‘self-forming.’12 If we replace the Businesswoman with another situation leading to an SFA, we can observe, by contrast, some notable features in Kane’s picture. Consider the following case: a businessman got promoted in his job, but his promotion requires him to be transferred abroad. Accepting or refusing the promotion is morally indifferent. However, his personality may be impacted by his decision (if he leaves, he is adventurous, if he stays, he is prudent), and it is a self-forming action. Running Kane’s argument with the Businessman case has highlighted some peculiarities of his description of the Businesswoman case that I will elaborate on below. Kane explicitly claims that the businesswoman is trying to do the moral thing. If she fails to do the moral thing, it ‘will not be a mistake or accident, but a voluntary and intentional doing of the other [which is the self-interested deed].’13 So that, if she were to exit the indeterministic effort of choice by choosing to go to her meeting, she would be said to have freely chosen that outcome. But Kane also says that ‘if [she] overcome[s] temptation, it will be Kane p. 224. Kane p. 234.

12 13

the result of [her] effort; and if [she] fail[s], it will be because [she] did not allow [her] effort to succeed.’14 So there seem to be two distinct meanings to the word “effort.” One effort aims at doing the moral thing; another effort is the general attempt to choose whichever. Not only is this very confusing, but it also seems incorrect. It is hard to see in what sense the businesswoman can make the effort to help the stranger before choosing to help him if she can very well freely choose not to do so. I do not understand how she could have made any effort towards doing some action before deciding, even less so if she is responsible for making the decision corresponding to the opposite to her effort’s end. To speak of an “effort” to do the moral thing before deciding to do it might come from the consideration that moral reasons are stronger than self-interested reasons, that humans have selfish inclinations or impulses that do not deserve the title “effort of the will” and that only moral considerations do. I do not see any argument from Kane supporting this hierarchy. Furthermore, I think it is incorrect to give this kind of priority to moral reasons precisely because they can be beaten in the deliberative fight against self-interested reasons. They have no more weight than amoral reasons, and Kane’s indeterminism depends on that. Thus, I disagree with Kane’s use of the word “effort” as the agent’s pre-choice orientation to do the moral thing. In addition, it is no clearer what the second notion of effort, the effort to decide, means. Can someone make an effort to do two incompatible things? Or is this effort aiming at a disjunction of decisions rather than a conjunction? If one cannot but succeed in her effort towards a disjunction of exhaustive possible outcomes, Kane p. 225.

is it not incompatible with the possibility of failure as a necessary condition for an effort? Although I will not answer these questions, I challenge Kane’s use of the word by raising them. The notion of “effort of will” is intuitively problematic because the word “effort” commonly refers to the deliberate pursuit of something which assumes that an aim is set to orient the effort. A compelling argumentation is needed to show that the businesswoman is indeed deciding by her “effort of will”. I suspect that these weak points about Kane’s argument are not peripheral mistakes that require further clarifications. Instead, I believe that the ambiguous meaning of “effort” is essential to Kane’s thesis. Whether deliberately or not, Kane mimics the Assassin case when building the Businesswoman. To say that the businesswoman’s effort is to do the praiseworthy action and that ‘if [she] fails, it will be because [she] did not allow [her] effort to succeed’ is to alter the case according to the framework of the assassin’s example.15 The assassin tries to do something and might fail because of some indeterministic part of his action. It is evident in this situation that there is some action the assassin is trying to do for which he will be blameworthy if it happens, even though he cannot ensure that it happens. However, this is assuming much more than what the Businesswoman case tells us. The assassin has already decided to murder, whereas the businesswoman is in the process of deciding what to do (with no premade decision). In short, to say that the businesswoman desires to do the moral thing, even though she might fail in her effort and end up freely not doing it, makes the situation seem as simple as the Assassin’s. Both cases use the framework: 1) the agent wishes to do A, but 2) something might happen that makes her not do A. The Kane p. 225.

Businesswoman case adds: 3) that thing that precludes her doing A is also her decision. If 3 seems odd at first glance, it can be successfully argued for on its own. The idea of sabotaging one’s own will, or doing something else than initially thought, seems intuitively sound. Moreover, as the assassin case shows, 1 and 2 are not problematic. Hence, to affirm 1, 2, and 3 together seems right. But this is the wrong picture. The first two steps are not an accurate depiction of an indeterministic SFA such as the Businesswoman case. Instead of 1, from the very first step, we must see that the agent has no will or desire to choose neither option over the other. She is precisely in the process of deciding which one to choose and has no already made decision that can be hindered by indeterminism. She indeed has reasons to do A, but she also has reasons to do B. I presume that the ‘wish to do A’ (from 1) shares the same semantic role as Kane’s phrasing ‘the effort to do A.’ To make the Businesswoman case as if 1, 2, and 3 were correct is to imitate the Assassin case. Blurring the differences between the two cases helps Kane’s argument against LP* because the assassin case successfully refutes LP. Although Kane acknowledges the differences between the two cases, I believe he overlooks the singularity of the Businesswoman case. Conclusion This essay has been a detailed opposition to Robert Kane’s article. The critique I have made here is not grounded on some opposing positive view. I am agnostic concerning the issues of (in)compatibilism and libertarianism. I have expressed in the preceding pages my skepticism to Kane’s defence of libertarianism without taking a stance in the debate. In conclusion, I find Kane’s argument against LP* exemplified by the Businesswoman case unsatisfactory because Kane fails to address the issue of 13

indeterministic (deliberative) agency without altering the case description to fit his thesis. Bibliography • Kane, Robert. (1999). Responsibility, Luck, and Chance: Reflections on Free Will and Indeterminism. The Journal of Philosophy, 96 (5): 217-40. • Franklin, Christopher Evan. (2011). Farewell to the Luck (and Mind) Argument. Philosophical Studies: An International Journal for Philosophy in the Analytic Tradition, 156 (2): 199-230.

Dismissing doubts about Descartes: Cartesian meditation for epistemological crises Armand Babakhanian is a third-year undergraduate student at Biola University armand.babakhanian@biola.edu. In his essay Epistemological Crises, Dramatic Narrative, and the Philosophy of Science, Alasdair MacIntyre defines an epistemological crisis as when a schema for understanding the world suddenly fails to adequately do so in irremediable ways (MacIntyre 2006: 8). MacIntyre critiques the work of Rene Descartes in his Meditations on First Philosophy as a disastrous attempt at addressing an epistemological crisis, due to Descartes’s impossible self-assigned task of doubting all beliefs until he uncovers a contextless first principle. This essay will argue that MacIntyre misunderstood Descartes’s project in the Meditations and was not able to fully appreciate his proposed solution to the epistemological crisis he faced. Understanding the Meditations through the interpretative lens of Descartes’s theological perspective, his rejection of Aristotelian scholasticism and his commitment to the new sciences of his time, provides a richer account of his project that MacIntyre fails to capture. The strength of MacIntyre’s critique of Descartes leans on a couple of claims. First, his claim that all philosophical inquiry is dependent on a context of prior background beliefs. Second, his claim that Descartes in his Meditations had attempted to conduct philosophical inquiry independent of a context of prior 15

background beliefs by deploying his method of global doubt in the first meditation. MacIntyre argues that Descartes’s way of resolving the epistemological crisis he faced was doomed from the beginning because it consisted of the impossible task of conducting philosophical inquiry in a theoretical vacuum (MacIntyre 2006: 17). These two claims are necessary for explaining Descartes’s critical error in the Meditations. This essay will focus on MacIntyre’s second claim, which appears to be a common misinterpretation of the Meditations that prevents a fuller appreciation of Descartes’s proposed solution to an epistemological crisis. The common misinterpretation that MacIntyre advances of Descartes’s Meditations begins by casting Descartes’s project as a primarily epistemological one of rebuilding all knowledge from an indubitable starting point. This is accompanied by an extreme focus on the first two meditations of the work where the method of doubt, the evil genius skeptical scenario and the indubitable epistemic solution of the cogito ergo sum are discussed. Descartes’s later discussions in the Meditations on the real distinction between the soul and the body and the arguments for God’s existence, are interpreted as ad hoc attempts to secure the epistemological foundation of the cogito. John Carriero in his seminal work, Between Two Worlds: A Reading of Descartes’s Meditations, refers to this misinterpretation as the ‘familiar view’ (Carriero 2009: 21). The product of the familiar view is a caricature of Descartes as a secular, detached and analytic epistemologist who is miserably failing to adequately answer skeptical objections to knowledge claims. The familiar view advanced by MacIntyre does manage to successfully capture the essence of the skeptical arguments put forth in the first meditation and the nature of the cogito. However, the familiar view fails to adequately integrate the content of the first meditation with the content of the Meditations as a whole. 16

The familiar view is mistaken because of its overemphasis on the epistemological aspects of the Meditations in addition to a misunderstanding of what those epistemological aspects truly are. Descartes’s genuine intentions in the Meditations are to defend the existence of God and the real distinction between mind and body while working under the new scientific paradigm of his time. Descartes writes to the faculty of sacred theology of Paris concerning his Meditations that he has, ’always thought that two issues - namely, God and the soul - are chief among those that ought to be demonstrated with the aid of philosophy rather than theology,’ in order to rationally convince unbelievers of its vitally important truth (Descartes 1641/1993:1). The uniqueness of Descartes’s project is in his manner of demonstrating the existence of God and the soul. Descartes uses the support of Holy Scripture to claim that, ’everything that can be known about God can be shown by reasons drawn exclusively from our own mind [emphasis added].’ (Descartes 1641/1993:1). What is revolutionary about Descartes’s approach is its stark contrast with his immediate Aristotelian scholastic ancestors, represented best by the medieval theologian and philosopher St. Thomas Aquinas. Contrary to Descartes, Aquinas believed that knowledge of God can only be known a posteriori through experience of created effects (Carriero 2009: 176). For Aquinas, the mind is not able to comprehend God’s existence but can only infer that something, which one calls God, must exist behind the created effects of God which one experiences. What Descartes attempts to do in the Meditations is demonstrate that one can know God and have an intellectual grasp of His existence without the use of the senses and experience of created effects. Descartes’s motivation for doing so is complex. It is tied to his commitment to post-Aristotelian natural science and his desire 17

of preserving his Catholic theological beliefs. Descartes was a confident proponent of the new sciences and advanced his own interpretation of them. Descartes believed that the most general laws of motion which governed the material world are pure intellectual ideas, not empirical principles derived from observations. In order to ground genuine knowledge of these pure mathematical laws and of the creator of these laws (God), Descartes had to provide a new account of the mind and cognition that was independent of the senses. This account of cognition that exclusively uses the powers of the mind and not the senses, was a clean break from the Aristotelian-Thomism that exercised its influence on the scholastic mind of Descartes’s predecessors (Carriero 2009: 16-17). The Thomist account of cognition known as “abstractionism” makes extensive use of the body, sense organs and inputs from the external world. Thomistic abstractionism holds that knowledge is of universal incorporeal forms that are arrived at by the immaterial intellect’s operation of abstracting them from phantasms or “sense images”. These phantasms are acquired by the sense organs receiving impressions of incorporeal forms or sensible species from the external world and storing them in the memory and imagination in a spiritual mode. Hence the famous dictum by Aquinas, nihil est in intellectu, nisi prius fuerit in sensu or ’nothing is in the mind that has not previously been perceived by the senses.’ (Carriero 2009: 12). Sense organs, inputs from the external world and the imagination are all involved in the Thomist account of cognition. This renders abstractionism incompatible with Descartes’s scientific views about nature and how one comes to know its fundamental laws. As a result, Descartes was forced to find a new way of philosophically demonstrating the truth of his religious beliefs in God and the soul that was congenial to his beliefs about cognition. It had to also be convincing to those who were inculcated in the scholastic tradition and would be reading 18

his revolutionary work. Disillusioning the meditator of the belief in the necessity of the senses for all knowledge, is part of the setup for Descartes to present his novel rationalist foundation with its emphasis in the ability of pure reason to acquire knowledge and its belief in the existence of innate ideas as objects of knowledge (Cottingham 2012: 30-31). To prepare this project in his Meditations, he has to rid the hypothetical scholastic meditator of belief in the necessity of the senses for knowledge. This was the true purpose to which the method of doubt and various skeptical scenarios about the external world was put towards; to help suspend the grip of AristotelianThomist epistemology on the meditator’s mind. Descartes states in his synopsis of the Meditations that, ’all the arguments on the basis of which we may infer the existence of material things are presented... because, through a consideration of these arguments, one realizes that they are neither so firm nor so evident as the arguments leading us to the knowledge of our mind and of God,’ (Descartes 1641/1993: 10). The epistemic primacy of the mind with its innate, immaterial ideas about the laws of nature, God and the soul, is the anti-Aristotelian attitude he is seeking to bolster. He is not addressing skeptical objections to general knowledge claims as an end in itself, but as helpful devices for achieving his larger goal. In fact, Descartes believes that ’no one of sound mind has ever seriously doubted’ the existence of material things (Descartes 1641/1993:10)! Clearly, Descartes did not intend to employ a method of global doubt in order to see what would be left standing in the epistemological wasteland. He is not guilty of the mistake of trying to conduct philosophical inquiry independent of a context of prior beliefs. Descartes was steeped within the theological meditative tradition of Augustine, Anselm and Bonaventure which combines 19

philosophical inquiry with religious devotion (Cottingham 2012: 25). The meditative tradition is characterized by both an intellectual yet faithful ascent to God from the mutable world of the senses. This is exemplified in Anselm’s Proslogion, where Anselm actually addresses God in prayer and adoration while simultaneously trying to demonstrate His existence. This mode of philosophical inquiry does not occur in a conceptual vacuum, but in a rich theological web of beliefs. It is within the conceptual network of the meditative theological tradition, the collapse of the Aristotelian-Thomism and the new sciences, that Descartes attempted to formulate his solution to his epistemological crisis. Rather than being a secular and dispassionate treatise as is the case in most of analytic philosophy today, the Meditations can be more fruitfully understood as a genuinely meditative and spiritual exercise analogous to works like Augustine’s Confessions and Bonaventure’s Journey of the Mind to God. Understanding Descartes’s project through the interpretative lens of this context can help to flesh out how one can appreciate the Meditations as a helpful guide for resolving epistemological crises. During an epistemological crisis the fallible nature of conceptual schemas, one’s own reasoning and human authorities on intellectual matters, is exposed. What the Meditations offer is a way of transcending the fickle nature of all-too-human schemas in order to contemplate and apprehend the Truth itself in God. It is difficult to place one’s trust in the dependability of another conceptual schema after enduring a disillusioning epistemological crisis as described by MacIntyre. The anxious mind who seeks stability in the midst of competing rival interpretations of reality, requires a firm foundation in something which is beyond the products of imperfect human engineering. The Meditations can serve as a guide for those who choose to take up this quest and 20

’have both the ability and the desire to meditate seriously with me,’ (Descartes 1641/1993: 6). The journey begins by deciding to turn away from the mutable world of the senses that have led one to experience epistemological crises. Instead of looking to the world of the senses, one peers into one’s own soul to uncover the infallible light of truth. The meditator is then eventually led beyond themselves to contemplate the source of this light, who is God, the source and guarantee of indefeasible truth. It is humble submission to this infallible light of truth, rather than trusting human authorities, the senses and fallible schemas, that Descartes prescribes as a solution to epistemological crises (Cottingham 2012: 28). Although MacIntyre believes this sort of project to be impossible, he does successfully interpret Descartes’s Meditations as attempting to ground epistemic confidence in a transcendent foundation. However, MacIntyre’s adherence to the familiar view of Descartes leads him to misconstrue the manner in which Descartes conducted this project. In conclusion, contrary to MacIntyre’s claim, Descartes is shown to not be attempting to conduct philosophical inquiry independent of a context of background beliefs due to his radical method of global doubt. Although MacIntyre does successfully identify epistemological components of the first meditation, he fails to accurately depict their role within the Meditations as a whole. Descartes intended to focus the method of doubt upon the Aristotelian epistemology of the scholastics which exercised a powerful influence on the meditator’s mind. In the place of Aristotelian epistemology, Descartes presented his new foundation of cognition to support his arguments for the existence of God and the immaterial soul while under the purview of the emergent scientific worldview. As embedded in the Christian meditative tradition, the Meditations is a partly spiritual exercise 21

to help guide the meditator to a firm epistemic footing in the infallible light of truth emanating from God Himself. Interpreting Descartes’s project in the Meditations with these facts in mind provides an opportunity to more fully appreciate the solution which Descartes offers for resolving epistemological crises than MacIntyre had thought possible. Bibliography • Carriero, J. (2009). Between Two Worlds: A Reading of Descartes’s Meditations. Princeton: Princeton University Press. • Cottingham, J. (2012). The Desecularization of Descartes. In: C. Firestone & N. Jacobs, ed. The Persistance of the Sacred in Modern Thought. Notre Dame: University of Notre Dame Press, 2012. pp. 15-37. • Descartes, R. (1641). Meditations on First Philosophy. 3rd ed. Indiana: Hackett Publishing Company (1993). • MacIntyre, A. (2006). Epistemological Crises, Dramatic Narrative, and the Philosophy of Science. In: A. MacIntyre, ed. The Tasks of Philosophy: Selected Essays, Volume I. Cambridge: Cambridge University Press, 2006. pp. 3-23.

Is it the case we may as well be Nihilists? Nicholas Alan Barrow is a Philosophy of AI Masters student at the University of York and previously graduated with a Bachelors from the University of Kent nick.barrow@hotmail.co.uk Twitter: @thatssonick

Abstract In this short essay I identify a seemingly underappreciated line of argument available to the mereological nihilist; that we may as well be nihilists. I derive this from the fact that there is, arguably, little practical difference between the nihilist explanation of compositional objects and that of the more costly common-sense view. However, despite introducing a new way of appreciating the simplicity of the nihilist view, it is ultimately shown that the way in which the nihilist achieves it remains flawed; that is, through the vehicle of paraphrase. This essay will evaluate the eliminativist view that there is no truthmaker for the claim <there is a house>. That is, the Mereological Nihilist’s claim that there is no point where a plurality of parts shift to a singularity of unity. In order to exemplify Mereological Nihilism1, it will be contrasted with what will be referred to as ‘common-sense’ views of composition i.e., views that accept the From here on, I shall refer to this theory and its proponents as ‘Nihilism’ and ‘Nihilist’, respectively. 1

transition of ‘parts’ to ‘whole’2. It will identify that, for the Nihilist, there is very little practical difference between the two opposing views. There is, however, a substantial ontological difference. Given this merit of quantitative parsimony, the Nihilist can claim that we may as well be Nihilists.However, as will be shown, the Nihilist cannot rely on parsimony independently. Fortunately, she does not. What the Nihilist does rely on, though, is paraphrasing. This reliance is what will ultimately diminish the Nihilist’s position. Through Tallant’s (2014) analysis of this paraphrasing technique, this essay will conclude that due to its reliance on counterfactuals, Nihilism is unable to secure the referent of certain terms. Thus, by highlighting that there are significant practical differences after all, it is not the case that we may as well be Nihilists. 1. Mereological Nihilism At what point does a house become a house? Is it when the final brick is placed? The final light fitted? Vis-à-vis, is there a point at which the plurality of parts i.e., the bricks, lights etc. compose together to become a unity, a whole, a house? Can we ever have a truthmaker for the claim <there exists a house>? Our intuitive, common sense ‘naive’ view is that we can: ‘small things sometimes come together to form larger things’ (Rosen&Dorr 2002: 152). There is a collection of parts (bricks, cement, windows etc.) that, when combined in a certain way, become a house. Moreover, it is also intuitive for us to be able to identify these parts within the whole after they are combined. We are, for instance, able to point out individual parts such as windows, rooms, doors etc. ‘Folktalk’ like this is also useful when describing objects. To analyse a table, we appeal to its parts by explaining that it has X number of No specific theory will be analysed - only the common, shared notion of most realist theories that there are composite objects. 2

legs and a top – these parts make up the whole. However, Mereological Nihilists such as Gideon Rosen; Cian Dorr (2002); Trenton Merricks (2003); and Jonathan Schaffer (2007)3 etc., deny this ‘common sense’ view. There is, strictly speaking, no truthmaker for the claim <there exists a house>: there is no such thing as a house. This is not to say, however, that they are not there. The Nihilist ‘accepts that these are all as real as rocks’ (Schaffer 2007: 176), just not as composite objects. All concrete objects for the Nihilist are simple4, ‘none are mereologically composite’ (Ibid). Instead, the Nihilist, whilst denying the composite existence of a ‘house’, will agree that there are somethings arranged “house-wise” (Rosen&Dorr 2002). Thus, instead of talking of “houses”, the Nihilist will talk of “an arrangement of simples arranged house-wise.” This paraphrasing technique is crucial for the Nihilist. Simultaneously though, ‘folk-ontological beliefs for being (are)... nearly as good as true’ (Merricks 2003: 9). Despite it being a false proposition according to the Nihilist, it is acceptable for us to talk of ‘an F being composed of the Xs’ since propositions of the form ‘<there is a collection of simples – the Xs – arranged F-wise>’ are true. The fact that this latter statement is true ‘suffices for the assertability of the former’ (Tallant 2014: 1512). It is fair to say then that, although the Nihilist denies the truthfulness of composition, she has no qualms retaining talk of composition insofar that it acts in language as a summary; it ‘is useful shorthand’ (Schaffer 2005: 176). Schaffer is not a Nihilist; he is a Monist. Nevertheless, he also rejects this common-sense view. 4 It does not matter what these ‘simples’ are – one of the benefits of Nihilism is that it can fit in with whatever the true working of the universe is; whether everything is made up of an arrangement of quarks, or quantum energy, it does not matter for the Nihilist as long as it is the smallest unit. 3

2. Parsimony as a virtue Why then, should we accept this counter-intuitive view? As with most anti-realist theories, Nihilism appears to benefit from the ‘argument from quantitative parsimony’ (Smid 2015: 3254). The argument runs as follows: i. ‘Theory T1 entails that there are fewer...(objects)... in the world than theory T2 does. ii. Positive ontological parsimony: We have some at least prima-facie epistemic reason to prefer theories that require fewer ontological commitments to theories that require more. iii. Therefore, we have some at least prima-facie epistemic reason to prefer T1 to T2’ (Metcalf 2016: 703). Nihilists (T1) only posit one type of object in the world – simples. Anything and everything else are simply arrangements of these simples. The Nihilist posits an extremely conservative ontology. In contrast, our common-sense view (T2) of composition is ontologically committed to a whole host of further objects – those that have been composed. It was previously noted that, although it is not true to claim an ‘F is made up of some Xs,’ it is acceptable to use shorthand. When such a claim is made, what is really meant is that, according to the Nihilist, <there are an arrangement of simples arranged F-Wise.> If this is the case, we should abandon our common-sense view in light of Nihilism as, practically, what would we change? There is very little tangible difference: we may as well accept Nihilism as it is less ontologically taxing and ‘undetectably different from the preferred alternative’ (Rosen&Dorr 2002: 158). As Schaffer states, ‘if particles are the only concrete objects needed to explain things...then there is nothing left for the composites to explain’ (2015: 177). 26

2.1 The limits of parsimony What this implies, however, is that ontological parsimony has some prima-facie weight for accepting a theory – but does it? (Metcalf 2016) ‘After all’ Schaffer muses, ‘the main motivation for Nihilism...is that it provides a simple yet sufficient ontology’ (2015: 187). In scientific domains, empiricist accounts of parsimony can be extremely useful and are widely accepted. Take causation for example; if one were to hear someone coming from downstairs, it is ontologically parsimonious to conclude it is someone you live with, rather than a burglar that has forced their way in with a hammer. Science’s ‘excellent track record...is evidence that its methodology, including the use of Ockham’s razor, is reliable’ (Huemer 2009: 228). We cannot, however, make a similar case for philosophy. Parsimonious philosophy has seldom produced an impressive track record – it is ‘a leap of faith’ to assume we can use parsimonious reasoning with two very different disciplines (Ibid). In order to illustrate this, it is useful to consider the lack of value parsimony has on its own, without the accompanying epistemic evidence science provides. Metcalf (2016) asks us to consider three ontologies: (O1) O (O2) O U {x} (O3) (O-e1) U {x} (O1) entails our original ontology. (O2) entails our original ontology with the addition of some entity x. (O3) entails our original ontology, not including some part, e1, and with the addition of x. According to ontological parsimony, (O3) is equally as attractive as (O1) – by removing a part of our ontology, we can replace it and retain the same level of commitment. Ontological 27

parsimony, therefore, has no real prima facie value in the sense that it ‘does not tell us (on its own) which of (O1) and (O3) to accept’ (Ibid: 707). Therefore, it is unjustified to rely on it, in virtue of itself, in the way that the antirealist arguably does. Does it matter that parsimonious arguments aren’t enough on their own in this particular case? The Nihilist can explain; there is no equal ontology. We are not left with a disjunct between Nihilism and some other theory, simply because there is no other theory. Perhaps, however, the Nihilist is mistaken that realist theories of compositionality are ontologically extravagant. Perhaps the ontological commitment of claiming that some composite object, A exists, is no more of a commitment than claiming its parts, x y and z exist (Smid 2015). Parsimonious strategies are therefore ineffective when discussing compositionality as they ‘assume that counting objects is the same as counting ontological commitments’ (Ibid: 3256). In order to make <A exists> true, we do not need any further truth conditions than what makes the fact its parts x y and z exist true. A requires the existence of its parts, and so the demand put on the world for A to exist is just for its parts to exist (Ibid)5. This principle of innocent compositionality entails that ontological commitment goes no further than the commitment of the most fundamental parts – let’s also call them simples6. Thus, we can commit to composite objects and make statements like <there exists a house> true, whilst retaining an ontologically parsimonious position. At the very least, innocence of compositionality illustrates that any Nihilistic strategy of positive parsimony does not give them an edge over realist theories. At This claim is contentious. Parts cannot just simply be ‘arranged’ for them to form a whole – there must be some binding process. I do not have the scope to discuss this here, however. 6 Because, what is the difference? 5

the most, it shows that Nihilism is false in its claim that composite objects do not exist. 3. Mereological Nihilism: more than just a cheap ontology However, if we were to ask why we should accept mereological nihilism, it is not the case that the answer we receive will be that of just parsimony. The Nihilist can accept both Metcalf’s and Smid’s arguments and be unphased – parsimony on its own is not enough to establish one theory over another. Even if it was, the Nihilist could further accept that realist mereological theories are equally parsimonious. Parsimony is not the sole reason to accept her theory: despite being a strong motivation for adopting Nihilism, parsimony is merely a consequential merit. It is a mistake to claim that Nihilists argue for their theory by simply referring to its parsimonious nature, and, on account of accepting both Smid’s and Metcalf’s view, it is a mistake to argue for a theory in this way simpliciter. The Nihilist, and other anti-realists are mistaken if they present parsimony as anything more than an attractive side effect of their theory. It should not, on its own, be used as an argument to accept it. The Nihilist needn’t appeal to quantitative ontological parsimony in the way Metcalf presents it – the Nihilist can provide arguments independent of parsimony. Although it is not necessary here to go into such arguments, it is pertinent to mention two of the most prominent. Firstly, Nihilism solves puzzles such as the Ship of Theseus due to them being able to claim, simply, that there are just no such things as ships. Secondly, Merricks (2003) claims that composite objects are causally redundant insofar that we can explain causal interaction by appealing only to the individual simples that would make up a composite object, thus avoiding the need to appeal to the composite object itself – rendering them redundant. If we accept 29

such arguments in favour of Nihilism, the final motivation we need to accept it over common-sense mereology is that Nihilism is more parsimonious7, so we may as well be Nihilists. Although Metcalf (2016) may be correct in arguing parsimony is not itself a reason to choose one theory over another, it does not apply in the case of Mereological Nihilism as that is not how the Nihilist uses it – there are other arguments to consider when choosing between (01) and (03). 4. The special arrangement question Even if we assume that such additional arguments are to be accepted, the Nihilist is not out of the woods yet. We can further ask the Nihilist what Tallant (2014) coins, the ‘Special arrangement question’: What do we mean when we say, ‘the simples are arranged F-Wise?’ For the purposes of this essay, it will be focused on two examples of counterfactual responses to this question. Rosen and Dorr (2002: 158) answer in terms of behaviour – an arrangement of simples house-wise, for example, must behave in such a ‘way that they would, were they to compose a house.’ Merricks (2003: 3) focuses on supervenience whereby ‘...if (x) existed, those atoms composing (x) would not trivially supervene’ that is, there’d be a minimal difference. Both answers rely on counterfactuals; they reference ‘how the world would be were there to exist composite objects’ (Tallant 2014: 1515, emphasis added). We are, therefore, unable to fix referents of certain terms. For Nihilists, cats and dogs do not exist. When we talk of them, we are failing to pick out anything specific in the world. At most, we are picking out ‘some particles arranged cat-wise’ (Ibid). ‘Cat’ would have no intension. We’d struggle to identify what conditions provided us with On the surface – it may well be the case that we accept the innocence of compositionality, but that is not necessarily a part of common-sense mereology. 7

cats. Without the ability to pick out what it’d be to be arranged cat-wise, we are unable to paraphrase. If actual cats don’t exist, how could we envisage what it would be like for something to be arranged like a cat? (Ibid) It seems that there is quite a heavy practical difference; it is not the case we may as well be Nihilists. Conclusion Once we move on from our naive, common-sense views on compositionality, Mereological Nihilism seems to be an attractive position. The Nihilist appears to benefit from a parsimonious and therefore superior ontology, whilst retaining the common-sense, intuitive, practical nature of realist theories via paraphrasing. However, it was shown that arguments from parsimony are, in themselves, not a suitable reason (in philosophy) to prefer one theory over another. Moreover, even if parsimony was a valid argument, Nihilism would have no benefit over realist views. This objection was ultimately refuted as the Nihilist does not rely on parsimony in this way; it is simply a consequential benefit of the overall theory, supported by other independent arguments. Despite this, the Nihilist’s paraphrasing method was critically analysed to determine whether it was true that there was little practical difference, concluding that this was not the case – it is not as simple as Nihilists try to claim. Bibliography • Horgan, T. and Potr, M. (2000). Blobjectivism and indirect correspondence. Facta Philosophica, 2, 249-270. • Huemer, M. (2009). When is parsimony a virtue. Philosophical Quarterly, 59(235), 216-236. doi:10.1111/j.1467-9213.2008.569.x 31

• Merricks, T. (2003). Objects and persons. New York: Oxford University Press. • Metcalf, T. (2016). Ontological parsimony, erosion, and conservatism. Metaphilosophy. 47(4-5), 700-718. https://doi:10.1111/meta.12210 • Rosen, G. and Dorr, C. (2002). Composition as a Fiction. In R.M. Gale (Ed.) The Blackwell Guide to Metaphysics. New Jersey: Blackwell. pp. 151-174 https://doi.org/10.1002/9780470998984.ch8 • Schaffer, J. (2007). From nihilism to monism. Australasian Journal of Philosophy. 85(2), 175-191. https://doi. org/10.1080/00048400701343150 • Smid, J. (2015). The ontological parsimony of mereology. Philosophical Studies. 172(12), 3253-3271. https://doi:10.1007/s11098-015-0468-3 • Tallant, J. (2014). Against mereological nihilism. Synthese. 191(7), 1511-1527. https://doi.org/10.1007/s11229-013-0343-8

On Never Knowing the Definition of Knowledge Dr John Shand is a Visiting Fellow in Philosophy at the Open University john.shand@open.ac.uk Introduction The aim of this paper is to show that although there may be a correct definition of knowledge – that there is a correct definition of knowledge is, and need not be, asserted or denied for the purpose of the argument here – we could never know that we had the correct definition of knowledge. This is because in order to know that a definition of knowledge is correct we would have to already be in possession of a correct understanding of what knowledge is, otherwise we could not be in position to say that we know a definition of knowledge is correct. Clearly any proposed definition cannot be used to affirm its own correctness on pain of vicious justificatory circularity. In that case we would have to call upon a non-definitional account of what knowledge is. The problem then is that there is no agreed non-definitional account of knowledge and such accounts may be incompatible with each other. Indeed to undercut the vagueness and variety of nondefinitional accounts of knowledge, to settle what is knowledge and what is not knowledge, was the point of aiming at a onceand-for-all definition. But that just takes us back to the problem of using a definition to affirm its own correctness. What remains the case is that we do not have an agreed understanding of what knowledge is and therefore we cannot be in a position to know that we have ever have a correct definition of knowledge even if 33

we have it. Consider K K is a proposed definition of knowledge. By this is meant a real definition, not a stipulative or lexical definition. If K is correct, all and only things that fall under the definition are knowledge. If something fits the definition it is knowledge, and if something does not fit the definition it is not knowledge. Such definitions are usually formed up as a set of necessary and sufficient conditions. There is no commitment here, nor need there be, as to whether knowledge is or is not definable. The question of whether knowledge is ‘luminous’ is irrelevant.1 Rather it is supposed for the sake of argument that knowledge is definable. Testing definitions for correctness As the argument here generally refers to all possible definitions of knowledge, therefore giving a particular example of a definition of knowledge would be inappropriate and beside the point. But we can illustrate the general mechanism of how one might determine whether a definition K is correct or not as a definition of knowledge. Definitions, such as that for knowledge, are usually couched in a set of necessary and sufficient conditions. Thus K=abc, where each condition a, b, and c is necessary and they are jointly sufficient. This means respectively that there can be no instance of knowledge that does not have the feature a or b or c, and anything that has together the features abc is knowledge. This gives a way of testing whether a definition is correct. If we find an instance of knowledge that does not have any one or more of the features a See on this (Williamson, 2000). And in a different way (Cook Wilson, 1926, p.39-40). Also (Zagzebski, 1994). 1

or b or c, then the definition has failed to provide the necessary conditions for knowledge. If we find an instance of something that is not knowledge but has all the features abc, then the definition has failed to provide the sufficient conditions for knowledge. But here you can see the problem immediately: how are we to know that instances we pick out to test the definition are respectively instances of knowledge and not knowledge. Without knowing that, we cannot know whether the definition of knowledge K in question is correct or has failed. Being in a position to know whether a definition of knowledge is correct But what is it for us to be in the position where we may say that the definition K is correct or incorrect? Surely the only position where this may be the case is when we know the definition is correct or incorrect. This is not to say the definition cannot be correct or incorrect without our knowing it to be so. But what use could that be to us? Without knowing that the definition is correct it cannot justifiable and reasonable count for us as a definition of knowledge regardless of whether it is correct or not. To say that we may nevertheless treat it as the correct definition of knowledge would be an arbitrary gamble. How is it possible for us to know whether the definition of knowledge is the correct definition of knowledge? Using the definition One way would be to see whether we know it to be the correct definition of knowledge according to the definition of knowledge K. But this would be hopelessly justificatorally circular whereby we say we know K is correct by checking that knowledge of K’s 35

correctness accords with knowing as defined by K. Using another definition If we cannot know that K is correct by checking the definition according to K, then we must, in order to do so, know what knowledge is independently of K. This could be according to another definition of knowledge K1. But this would get us nowhere as far as knowing whether K is correct as it would start an infinite regress of definitions and our knowing that K is correct according to K1, and K1 is correct according to K2, and so on. It also supposes that there may be more than one definition of knowledge, and then if they were not logically identical, there would have to be some way of overarchingly deciding what knowledge really was out of all the possible definitions, which would return us to a singular definition. So we have to suppose that there is some K that is the unique definition of knowledge.2 Non-definitional understanding of knowledge If our knowing that K is correct cannot be derived from any definition of knowledge, including that given by K, it must be the case, if we are to know K is correct at all, we would have to know what knowledge is in some non-definitional way. But what could this amount to? What would knowing what knowledge is in a way that does not appeal to a definition amount to? On a purely practical utility level of pointlessness, it is hard to understand what the value of trying to define knowledge is if we already In a general paper on definitions, I generalize problems with knowing if definitions are correct or not. But the case of the definition of knowledge itself presents additional difficulties because of logical self-reference. See (Shand, 2008). 2

know all and only the things that count as knowledge, that is what knowledge is, by some means other than consulting a definition. It certainly seems that people do seem to suppose that they know what knowledge is without consulting a definition of it, otherwise how would the definition ever be assessed or tested? When a definition is proposed it is usually assessed on the basis of examples and counterexamples compared to the necessary and sufficient conditions comprising the definition. But in order to be used validly as examples and counterexamples, by which means we might know that the definition of knowledge K is correct or not, we would have to know that they were examples of knowledge and counterexamples of not-knowledge. We have already seen on pain of circularity that the definition cannot help us with that. Yet the common parlance is: here is a proposed definition of knowledge K, but it cannot be a correct definition of knowledge as something K counts as knowledge is ‘obviously’ or ‘clearly’ not knowledge, or alternatively something that ‘obviously’ or ‘clearly’ is knowledge is not counted as such by K. It is hard on this basis to see how any assessment of a proposed definition of knowledge may be made, for what is the basis for saying that the counterexample is, or is not, obviously not knowledge? What would a nondefinitional knowing what knowledge is amount to? An intuition? A particular feeling that you know something? Knowing it (that is knowledge) when you see it? What people agree to calling knowing something or call knowledge? None of these seem a very sound way of judging the correctness or the incorrectness of K, which can only be founded on something that puts us in a position to know that K is the correct definition of knowledge. As the various nondefinitional accounts of knowledge may not be all logically identical, and may indeed be contradictory, as what falls under ‘know’, one would have 37

to find a way of deciding between them and have some kind of overarching understanding of what knowledge is. What features are required or not required for saying someone knows something is highly contentious. It is highly unsettled what features must and must not be a feature of knowledge.3 In that case we would be no better off than before in trying to assess whether K is correct, as we would instead be in a position of having to assess which of the nondefinitional accounts of knowing or knowledge is correct, which in the case of determining whether K is correct was the whole point of appealing to a nondefinitional account of knowing or knowledge. One thing is for certain, that there are no agreed criteria or conditions for knowing whether something counts as a counterexample to K, therefore through no counterexample may it be shown that K is or is not, as example suggests that it is, the correct or incorrect definition of knowledge. One would have to be sure one knows that any putative counterexample is an instance of knowledge or not knowledge for it to count as an example or counterexample to test the validity or correctness of K. It should be made clear that the issue and the concern of the

A good example is ‘The Rocking-horse Winner’, a short story by D. H. Lawrence. This is nicely explicated and explored in (Hamlyn, 1977, pp. 81-82). Whenever a boy sits on a rocking-horse he consistently gets the winner of a horse race right. At this level of rightness we would quite likely say that he knows the winner, without there being any idea of how he knows, any rationale for how he might come to know, despite that often being regarded as a condition for knowledge. Another example is that even truth, often thought uncontentious, has been disputed as a necessary condition for knowledge. (Popovic, 2015) The point here is not whether the argument for that is correct or not, but that even that is respectably disputable and leaves what knowledge is unsettled in such a way that we are in no position to assess a definition of knowledge as correct. The necessary connection between knowledge and belief may also be broken, or at least challenged, with a distinction between subjective and objective knowledge. (Popper, 1979). 3

argument here are quite different from the well-trodden matter of whether if you know something you must know that you know it.4 For the subject here is not whether you can be a in a state of mind, or stand in some relation to the object of knowledge, such that one knows without knowing that one knows one is in that state of mind or knows one is stand in that relation also being so, but rather whether we can know that any definition of knowledge is correct or not. The issue here is not whether if one knows a definition of knowledge is correct whether one needs to know that one knows it is correct, but only whether you may know a definition of knowledge to be correct. As has been said, typical of the fatally flawed method in assessing the correctness of the definition of knowledge is that it casually stated that from a counterexample it is ‘clear’ – sometimes the presumptive and one supposes rhetorically encouraging epithet ‘obvious’ or some other synonym is applied – that we do not know, and yet the case fits some proposed definition of knowledge, thus invalidating it as a correct definition of knowledge.5 To say something is ‘clear’ is weak, thin, and ironically vague, stuff indeed, and nowhere near enough to form a ground for assessing the correctness of a definition. ‘Clear’ to whom? And how is That is that Kp→KKp. Here of course K is the knowledge operator ‘knows that’, and not the designator used for the definition of knowledge K in this paper. 5 See (Gettier, 1963). In his paper Gettier uses phrases such as ‘…it is equally clear that Smith does not KNOW…’, and ‘…then Smith does not KNOW that…’ and ‘….does not state a sufficient condition for someone’s knowing a given proposition.’ But from whence is the clarity derived and justified. If it is said that we know we have such clarity, we are back with circularity. This argument completely undermines the validity of Gettier’s argument, not because of whether he is right or wrong about the correctness of the tripartite definition of knowledge he happens to examine, but because he has no way on such weak and vague grounds of determining whether the definition, or any definition, is correct or not, for he has not shown he knows what knowledge is and thus that the counterexamples are counterexamples to the definition. 4

‘clear’ to be distinguished from the mere appearance of clarity. In bringing up something being clear we are no better off than when we say we know. Worse we would have to say we know what is clear and not clear for clarity to be of any use in judging via a counterexample the definition of knowledge K, so we are back to knowing again. If we are to be sure enough of the counterexamples to use them as counterexamples against a definition of knowledge, then we would surely have to be able to present something like a definition of knowledge or some other sure rigorous account which identifies them as our knowing that they were indeed such counterexamples. But then, we would be back to square one in our attempt to find out whether a definition of knowledge is correct by needing a definition, or some like enough to a definition, in order to assess the correctness of any definition. So, we cannot tell whether the counterexamples undermine the definition or not, as we do not know if they are counterexamples or not and would have to know it in order for them to carry the force of argument to either affirm or contradict the putative definition of knowledge K. It should be added, that K not an empirical contingent theory that might be supported in degrees by stronger or weaker evidence, but a necessary logical definition whose validity or correctness is either absolutely contradicted or not contradicted at all, according to its conditions, by counterexamples. One cannot affirm the correctness or incorrectness of something with a particular level of certainty by appealing to something whose certainty is less that what one is affirming. Conclusion There is no way to know if any definition of knowledge is correct or incorrect. This is because it would either involve the circularity of referring to the definition of knowledge, or referring to some nondefinitional way of knowing what knowledge is, but there is 40

not one whereby we may say we know the definition is correct or not. In conclusion, therefore, we can never know that we have the definition of knowledge K, even if we do have it. For that we would need to know what knowledge is before it is defined. From which it follows that the attempt to define knowledge by all the usual criteria might be regarded as pointless. We may never know the definition of knowledge. Bibliography • Cook Wilson, John. (1926) Statement and Inference. Oxford: Oxford University Press, Vol. 1, pp.39-40. • Gettier, Edmund. (1963) ‘Is True Justified Belief Knowledge’, Analysis, Vol. 23 (6), pp.121-123. • Hamlyn, D. W. (1977) The Theory of Knowledge. London: Macmillan. • Popovic, Nenad. (2015), Why Truth is Not a Necessary Condition for Knowledge’, Philosophical Forum Volume 46, Issue 4, pp 397-401. • Popper, Karl. (1979) Objective Knowledge. Oxford: Oxford University Press. • Shand, John. (2008) ‘Futile Definitions’, Think 18. • Williamson, Timothy. (2000) Knowledge and Its Limits. Oxford University Press. • Zagzebski, Linda (1994) ‘The Inescapability of Gettier Problems’. The Philosophical Quarterly, Vol. 44, No. 174, pp. 65-73. 41

Interview with Dr Fiora Salis Fiora Salis is Associate Lecturer at the Department of Philosophy at the University of York and Research Associate at the Centre for Philosophy of Natural and Social Science at the London School of Economics. Before joining the Department in York, she was a Marie SkłodowskaCurie Fellow at the London School of Economics (2015-2017), and a Postdoctoral Fellow at the Centre for Philosophy at the University of Lisbon (2012-2015). She received a PhD in Philosophy from the University of Barcelona (2011), an MA in Cognitive Science from the same University (2007), and an MA in Philosophy and History of Ideas from the University of Turin (2006). Her main research focus is on the nature and varieties of imagination, and on its distinct uses in artistic fictions, scientific models, and scientific thought experiments. She also works on fiction and the semantics and pragmatics of fictional discourse, fictional names, mental files, episodic memory and mental imagery. She has presented work in a number of international events in the UK, Europe, the United States, South America and Japan. She has published a number of singled authored and co-authored papers, including, more recently: “Of Rabbits and Men: Fiction and Scientific Modelling” (in Philosophical Fictionalism, edited by Bradley Armour-Garb and Fred Kroon for OUP, 2017, co-authored with Roman Frigg) “Models and Denotation” (in Abstract Objects, edited by José Luís Falguera and Concha Martínez-Vidal for Springer Nature Switzerland, 2020, co-authored with Roman Frigg and James Nguyen) “Capturing the Scientific Imagination” (in The Scientific Imagination, edited by Peter Godfrey-Smith and Arnon Levy for OUP, 2020, co43

authored with Roman Frigg) “Learning Through the Scientific Imagination” (Argumenta 6/1, 2020), “The New Fiction View of Models” (British Journal for the Philosophy of Science 72/3, 2021) “The Meanings of Fictional Names” (Organon F 21/1, 2021) “Bridging the Gap: The Artifactual View Meets the Fiction View of Models” (Models and Idealisations in Science, edited by Alejandro Cassini and Juan Redmond for Springer, 2021) fiora.salis@york.ac.uk Could you briefly introduce some of your main theories and ideas on Imagination to introduce our readers to your work within this branch of philosophy? My original area of expertise is the philosophy of fiction, where I made a number of contributions on the intentionality of thought and discourse about fictional characters, the nature of fictional entities, the meanings of fictional names and the sort of mental attitudes that are involved in our conative and emotional engagement with fiction. My work in this area has been characterised by a robust sense of reality and a strong commitment to the principle of parsimony, according to which other things being equal, a theory that is ontologically more parsimonious than another is to be preferred. For example, in one of my earliest papers, I addressed the problem of intentionality and of the intersubjective identification of fictional characters, which arises from the difficulties of explaining how our thoughts and discourse about, say, Oliver Twist or Hedda Gabler can be directed towards the same (or different) characters. Some theorists solve this problem by postulating exotic fictional entities such as non-existent objects, created objects (or fictional 44

artefacts) and possible objects. Alternatively, I offered a solution compatible with the idea that there are no such entities. According to my proposal, our ability to think and talk about the same object is to be explained that in terms of participation in a practice of thinking and talking about the same thing, whether or not there is such a thing. Developing this theory required the provision of an account of the identification of name-using practices, of names, of the acquisition of information through name-using practices, and of the storage of such information in cognitive structures called mental files. With these tools in hand, the theory I developed could explain a number of important phenomena, including, for example, individual and intersubjective identification and misidentification of fictional characters and disagreement about their features. Coherent with the principle of parsimony, this theory is preferable to realist theories of fictional characters because it explains the same cognitive and linguistic phenomena without committing to the existence of implausible (and obviously unnecessary) fictional entities. In another, more recent paper, I addressed the paradox of fiction, which emerges from considerations about the rationality of our emotional engagement with fiction. When watching a performance of Shakespeare’s Othello, I pity Desdemona, loathe Iago and fear Othello. Or do I? It seems that the sort of emotions we seem to have when we imaginatively engage with fiction have certain features that are different from those of ordinary, genuine emotions. For example, emotions are reactions to something and they always seem to be about something. Iago, Desdemona and Othello, however, do not exist. So, what causes these emotions and what are they about? Emotions are subject to epistemological standards of correctness by which we assess whether they are justified or not. In this respect, emotions are similar to many other 45

cognitive states, like beliefs and perceptual states. They have the mind-to-world direction of fit. I know, however, that Desdemona, Iago and Othello do not exist. So, how can I really pity, loathe, and fear them? Emotions have also an important motivational role, which relates to their behavioural adequacy. Fear can enter practical reasoning mechanisms that lead to certain types of action (e.g., run or freeze). But my fear of Othello’s actions does not motivate me to do anything. I keep my fear quarantined. I do not act upon it. This discrepancy must be explained. Upholders of a view known as narrow cognitivism explain these features by postulating a new type of mental state, or quasi-emotions, which are imaginative counterparts of genuine emotions, or emotion-like imaginings. Upholders of the alternative view known as broad cognitivism argue that our emotional engagement with fiction involves genuine emotions that are triggered by certain thoughts or imaginings and that are relevant for our correct appreciation of fiction. I agree with the main insight of broad cognitivism, but I also recognise that narrow cognitivism can explain more features of our emotional engagement with fiction. So, I developed a new version of broad cognitivism that could deal with old and new challenges without committing to any new mental states. Since 2015 I started working on a new area in philosophy of science concerning the nature of scientific models, the nature and varieties of scientific imagination, how imagination contributes to the construction and the development of models, and how knowledge of models is converted into knowledge of reality. Modern science crucially relies on idealized descriptions of hypothetical systems – or model systems – that scientists use to represent certain parts or aspects of the world – or target systems. Paradigmatic examples of scientific models include Newton’s model of the solar system, Fibonacci’s model of population growth and Lotka-Volterra model 46

of predator-prey interaction. A hypothetical system is chosen as the object of study because it is less complex and easier to study than its target system and because by using the hypothetical system to represent its target we can learn about the latter. The hypothetical system is supposed to reveal the causal relations between phenomena, to enable predictions about such phenomena, or to explore alternative possible ways things could be. Thus, rather than working directly on reality, scientists construct, develop, manipulate and test models in order to discover truths about their targets. This prompts a number of questions, including about the nature of scientific models, the nature and roles of the scientific imagination, and the ways in which models and imagination can enable knowledge of reality. In this context, I developed the new fiction view of models and the integrated fiction view of models which put imagination at the core of the modelling practice. One caveat is important here. In his book, Mimesis as Make-Believe, Kendal Walton developed a paradigmatic theory of fiction as a game of make-believe that has been influential among contemporary developments of the fiction view of models. Construing models as akin to works of fiction and games of make-believe, however, does not amount to portraying modelling as something unserious or even frivolous. Walton submits that one of the reasons why children engage in games of make-believe is to cope with their environment. He mentions an extreme case of a game played in Auschwitz called ‘going to the gas chamber’ and argues that the children playing the game were ‘facing the reality of genocide with the utmost seriousness’. The ability to engage in games of make-believe continues in our adult life when we interact with works of art. We imaginatively engage with literary fictions, dramas and other works of art that can in no way be dismissed as unserious (think of Anna Karenina 47

or Othello). Scientists imaginatively engage with models for many serious purposes, including learning about reality. It is my idea that scientific models are human-made artefacts created through a scientist’s activity of modelling. Modellers use model descriptions such as linguistic and mathematical representations to specify model systems having properties that only concrete objects could have. Yet there are no such objects. Scientists know this. They use their imagination to build and manipulate surrogate systems for different purposes in different contexts. A view that puts imagination at the core of the modelling practice needs to say something about what sort of imagination that is and how that contributes to the generation of new knowledge of reality. For this purpose, together with Roman Frigg at the London School of Economics, I developed the first account of the scientific imagination as propositional imagination of the make-believe variety, where make-believe is a social imaginative activity that is constrained by the use of props and by certain principles of generation. With this notion in place, I have recently started to develop a new account of knowledge through the scientific imagination. Key to this account is the idea that the sort of imagination involved in modelling is constrained imagination. Make-believe is a highly constrained variety of imagination. But not all uses of make-believe produce knowledge of reality. Understanding how certain uses of make-believe in scientific modelling generate knowledge of reality requires an investigation into the sort of constraints operating on make-believe in this particular context. After defining the first overarching taxonomy of such constraints, I conducted a number of empirical case studies on economic modelling that have helped me further develop and refine a number of hypotheses. This work has been recently presented in a co-authored paper with our own Prof Mary Leng at York. 48

As a current student of Philosophy, my peers and myself talk about what branches of philosophy we are drawn to. What influenced you to pursue this particular branch of philosophy? How have your ideas evolved over time? I think I partially answered the second part of this question above, where I drew a sort of conceptual map of the evolution of my research expertise from key issues in philosophy of fiction to key issues about scientific models, imagination and scientific knowledge in philosophy of science. I have a 10 years educational background in music and musicology that I took in parallel with my normal studies at school before continuing my education at university. Through those studies I became interested in the aesthetics of music, which I explored during a BA at the University of Turin, where I wrote a short dissertation under the supervision of philosopher Stefano Kobau and the great historian of music, Giorgio Pestelli. This naturally led me to philosophical investigations into the broader area of aesthetics and the ontology of music during an MA at the same University. I obtained a second MA in cognitive science and language at the University of Barcelona, where I had the great fortune of working as a member of the Logos group, an international research group that was mainly focused on philosophy of mind and language, and through that I got interested in philosophical reflections on fiction. I read Walton’s Mimesis as Make-Believe as part of the work I conducted to write a PhD dissertation on the semantics of fictional names and fictional discourse. I suppose what influenced me to move through these different subjects was genuine curiosity for everything that seem relevant and important to better understand the world we live in. Every question opens up new questions that are worth considering and exploring given the right opportunities.

Philosophy often isn't seen as an accessible subject, and many potential philosophers may get overwhelmed with a quick general google search. For our readers interested in exploring this field, where do you suggest they start? I started in high school. In Italy, philosophy is part of the national curriculum in schools that are called Licei. These are state schools that were originally designed for the education of those who (independently of social or economic background) intended to continue their studies at university. Like in Germany, Spain and France (among other European countries) there are almost no private schools that offer the same knowledge rich curriculum. I started reading ancient philosophy with a wonderful teacher in my first year of high school. She was obviously very intelligent and knowledgeable, but also compassionate. She wasn’t afraid of disagreement. That was the fun part of it. I looked up to her and while I didn’t always share her opinions, I loved the fact that we could reason together through the most difficult and interesting ideas. I wouldn’t recommend going through a google search. It’s so impersonal, and there are too many irrelevant sources. Philosophy is about thinking together with others. You notice this immediately when you read ancient philosophy. Anyone interested in philosophy should probably start there. Plato’s Symposium and Aristotle’s Organon are very different but (I’d say) they are a pleasure to read, and surprisingly accessible. Reading these texts, one realises that – accounting for historical context – these were people very similar to us, minus the cell phones and atomic power. And there’s so many others, like Mary Wollstonecraft (who wrote a seminal work in feminism called A Vindication of the Rights of Woman), Jean Jacque Rousseau (you can read his Discourse on the Origin and Basis of Inequality Among Men, which is a must-read for those interested in theories of social justice) and Hobbes (especially his Leviathan). Read Descartes, Kant, and 50

John Stuart Mill. Read contemporary philosophers like Hannah Arendt (The Human Condition and The Origins of Totalitarianism), Martha Nussbaum (The Fragility of Goodness, Sex and Social Justice and her more recent From Disgust to Humanity: Sexual Orientation and Constitutional Law). You may not fully appreciate everything at once. You may not agree with everything they say. It does not matter. It is really just the start. Be curious, be charitable, and read the original literature with an open mind. Where do philosophical inquiries on Imagination cross over, intersect and interact with other disciplines? What effect does having philosophical discussion have on other disciplines? Philosophers have worked with different conceptions of imagination to address different problems, including modal knowledge, mindreading, art, morality, politics, emotions, dreams, hallucinations, scientific modelling, scientific theories, scientific and philosophical thought experimenting, and more. Philosophical reflections on imagination thus cross over different areas of philosophy. But they can also cross over different disciplines such as philosophy, cognitive science, sociology and political theories among others. In my case, my own reflections on the nature of the scientific imagination may help scientists better understand their own practices and methodologies. How do you see philosophical discourse of Imagination evolving? What are the new dilemmas and questions we face and debate? This is a difficult question, and I cannot give an exhaustive answer here. I can see at least three areas in which philosophical discourse on imagination is going to evolve. The first area 51

concerns philosophical reflections on the nature of imagination. Philosophers such as Peter Strawson and Kendall Walton worry that the uses and applications of the term ‘imagination’ make up a family that is too diverse to fit with any coherent and systematic definition. Recent attempts have been made to develop a number of different taxonomies that fit well with this diversity of uses and characterisations (by myself, but also by others, including for example our own Gregory Currie, Amy Kind, and Margherita Arcangeli). I believe this is really just the beginning. The second area concerns the relation between imagination and mental imagery. Interestingly, some cognitive scientists (including, for example, Shaun Nichols and Stephen Stich) tend to think about imagination as a propositional attitude distinct from mental imagery. Some philosophers concur (including myself). Mental imagery is involved in a number of different types of mental states and mental activities, including for example episodic memory, that do not seem to satisfy certain core features of imagination. Some philosophers, on the other hand, have argued for a richer and more robust notion of imagination that includes mental imagery. The debate is still ongoing. Finally, the third area concerns epistemic uses of imagination in distinct knowledge practices. Recent work by Amy Kind, Timothy Williamson, Magdalena Balcerak-Jackson, Neil Van Leeuwen, myself and others goes exactly in this direction.